AXQ 6-bit manual agent-coding recipe (no 4-bit, attention 8); Tier 1 pass
Browse files- README.md +25 -154
- axquant_manifest.json +85 -79
- axquant_plan.json +0 -0
- axquant_quantizer_execution.json +193 -193
- axquant_runtime.json +2 -2
- config.json +386 -386
- model-00001-of-00019.safetensors +3 -0
- model-00002-of-00019.safetensors +3 -0
- model-00003-of-00019.safetensors +3 -0
- model-00004-of-00019.safetensors +3 -0
- model-00005-of-00019.safetensors +3 -0
- model-00006-of-00019.safetensors +3 -0
- model-00007-of-00019.safetensors +3 -0
- model-00008-of-00019.safetensors +3 -0
- model-00009-of-00019.safetensors +3 -0
- model-00010-of-00019.safetensors +3 -0
- model-00011-of-00019.safetensors +3 -0
- model-00012-of-00019.safetensors +3 -0
- model-00013-of-00019.safetensors +3 -0
- model-00014-of-00019.safetensors +3 -0
- model-00015-of-00019.safetensors +3 -0
- model-00016-of-00019.safetensors +3 -0
- model-00017-of-00019.safetensors +3 -0
- model-00018-of-00019.safetensors +3 -0
- model-00019-of-00019.safetensors +3 -0
- model.safetensors.index.json +0 -0
README.md
CHANGED
|
@@ -1,174 +1,45 @@
|
|
| 1 |
---
|
| 2 |
-
|
| 3 |
library_name: mlx
|
| 4 |
-
base_model: openai/gpt-oss-120b
|
| 5 |
-
base_model_relation: quantized
|
| 6 |
pipeline_tag: text-generation
|
| 7 |
tags:
|
| 8 |
- mlx
|
| 9 |
-
- apple-silicon
|
| 10 |
-
- quantized
|
| 11 |
-
- mixed-precision
|
| 12 |
- axquant
|
| 13 |
-
- axq
|
| 14 |
-
- development
|
| 15 |
- gpt-oss
|
| 16 |
-
-
|
| 17 |
-
|
|
|
|
| 18 |
---
|
| 19 |
|
| 20 |
# AX-gpt-oss-120b-MLX-AXQ-6bit
|
| 21 |
|
| 22 |
-
|
| 23 |
-
the BF16 source model. The language path is quantized under AXQuant protection floors (embeddings, norms, and other protected tensors remain higher precision).
|
| 24 |
-
|
| 25 |
-
> **Development evidence — not a certified AXQuant release.** This package has conversion and
|
| 26 |
-
> artifact-integrity records, but it does not publish measured quality, long-context, kernel-speed,
|
| 27 |
-
> or MTP-speed evidence. Do not interpret the AXQ product label as a benchmark claim.
|
| 28 |
-
|
| 29 |
-
|
| 30 |
-
## Model details
|
| 31 |
|
| 32 |
| Property | Value |
|
| 33 |
| --- | --- |
|
| 34 |
-
|
|
| 35 |
-
|
|
| 36 |
-
|
|
| 37 |
-
| Source
|
| 38 |
-
|
|
| 39 |
-
|
|
| 40 |
-
|
|
| 41 |
-
| AXQuant base precision class | `6bit` |
|
| 42 |
-
| Planned storage-adjusted BPW | 6.0000 |
|
| 43 |
-
| Measured main-model BPW | 6.0000 |
|
| 44 |
-
| Measured total BPW | **6.0000** |
|
| 45 |
-
| Safetensors weight size | 87.62 GB |
|
| 46 |
-
| Approximate complete download | 87.65 GB |
|
| 47 |
-
| Configured maximum context | 131,072 tokens; practical limits depend on unified memory |
|
| 48 |
-
| Primary MLX runtime | MLX-LM |
|
| 49 |
-
| AX Engine native execution | Not established; no validated native manifest is included |
|
| 50 |
-
| MTP present | `False` |
|
| 51 |
-
| Vision present | `False` |
|
| 52 |
-
| Audio present | `False` |
|
| 53 |
-
|
| 54 |
-
This repository contains MLX Safetensors. It does **not** contain PyTorch or GGUF weights.
|
| 55 |
-
|
| 56 |
-
## Choosing an AXQ pack
|
| 57 |
-
|
| 58 |
-
AXQ names describe a **storage-budget product class**, not one uniform precision applied to every
|
| 59 |
-
tensor. Protected tensors remain at higher precision, so the exact measured BPW is authoritative.
|
| 60 |
-
In particular, a `6bit`-named mixed plan may retain `4bit` as its base precision while selecting
|
| 61 |
-
6-bit, 8-bit, or BF16 for other tensors to meet an approximately 6-BPW total budget. Protection
|
| 62 |
-
floors can also raise a `4bit`-named pack close to (or above) a `6bit` budget on small or heavily
|
| 63 |
-
protected models. When that collapse happens, AutomatosX does **not** publish a separate
|
| 64 |
-
misleading `4bit` sibling for that base.
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
| Sibling | Intended trade-off |
|
| 68 |
-
| --- | --- |
|
| 69 |
-
| [4bit sibling](https://huggingface.co/AutomatosX/AX-gpt-oss-120b-MLX-AXQ-4bit) | Lower-storage AXQ budget; check its exact BPW |
|
| 70 |
-
| [6bit sibling](https://huggingface.co/AutomatosX/AX-gpt-oss-120b-MLX-AXQ-6bit) | Higher average precision near the 6-BPW budget |
|
| 71 |
-
|
| 72 |
-
See the [AutomatosX MLX model catalog](https://huggingface.co/collections/AutomatosX/automatosx-mlx-model-catalog)
|
| 73 |
-
for related MLX and OptiQ alternatives.
|
| 74 |
-
|
| 75 |
-
## Download
|
| 76 |
-
|
| 77 |
-
```bash
|
| 78 |
-
python -m pip install -U huggingface_hub
|
| 79 |
-
hf download AutomatosX/AX-gpt-oss-120b-MLX-AXQ-6bit --local-dir ./AX-gpt-oss-120b-MLX-AXQ-6bit
|
| 80 |
-
```
|
| 81 |
-
|
| 82 |
-
Allow at least 87.65 GB of free disk space. Pin the resulting Hub commit in reproducible
|
| 83 |
-
deployments rather than relying indefinitely on `main`.
|
| 84 |
-
|
| 85 |
-
## Run with MLX-LM
|
| 86 |
-
|
| 87 |
-
```bash
|
| 88 |
-
python -m pip install -U mlx-lm
|
| 89 |
-
mlx_lm.generate \
|
| 90 |
-
--model AutomatosX/AX-gpt-oss-120b-MLX-AXQ-6bit \
|
| 91 |
-
--prompt "Explain mixed-precision quantization in three sentences." \
|
| 92 |
-
--max-tokens 128 \
|
| 93 |
-
--temp 0.0
|
| 94 |
-
```
|
| 95 |
-
|
| 96 |
-
MLX-LM compatibility covers standard **text/backbone inference**. It may ignore AXQuant runtime
|
| 97 |
-
metadata and optional sidecars (`vision.safetensors`, `mtp.safetensors`); this command therefore
|
| 98 |
-
does not establish MTP acceleration or vision-language quality. The artifact records MLX
|
| 99 |
-
`0.32.0` and MLX-LM `0.31.3` from conversion.
|
| 100 |
-
|
| 101 |
-
## AX Engine status
|
| 102 |
-
|
| 103 |
-
This package does **not** include a validated native `model-manifest.json`, so AX Engine execution
|
| 104 |
-
is not established by this release. The AX Engine fields in `axquant_runtime.json` describe the
|
| 105 |
-
intended compatibility contract, not observed runtime evidence. Use the architecture-specific MLX
|
| 106 |
-
runtime path above. The artifact records AX Engine version
|
| 107 |
-
`not recorded`, but version discovery alone is not a runtime check.
|
| 108 |
-
|
| 109 |
-
## Quantization layout
|
| 110 |
-
|
| 111 |
-
| Main-weight precision | Parameters | Share |
|
| 112 |
-
| --- | ---: | ---: |
|
| 113 |
-
| `4bit` | 44.59B | 38.17% |
|
| 114 |
-
| `6bit` | 70.41B | 60.27% |
|
| 115 |
-
| `8bit` | 1.21B | 1.03% |
|
| 116 |
-
| `bf16` | 619.45M | 0.53% |
|
| 117 |
-
|
| 118 |
-
- Quantization methods: `affine, bf16`.
|
| 119 |
-
- Group sizes used by quantized assignments: `32, 64`.
|
| 120 |
-
- MTP sidecar: not included.
|
| 121 |
-
- Vision sidecar: not included.
|
| 122 |
-
- Optimization scope: `text-path`.
|
| 123 |
-
- Support tier: `convertible`.
|
| 124 |
-
|
| 125 |
-
BF16 sidecars, when present, are included in total download size. Their presence does not by itself
|
| 126 |
-
establish MTP acceleration or vision-language quality.
|
| 127 |
-
|
| 128 |
-
## Evidence and validation status
|
| 129 |
-
|
| 130 |
-
| Check | Status |
|
| 131 |
-
| --- | --- |
|
| 132 |
-
| Planning evidence | `architecture_prior` |
|
| 133 |
-
| Calibration | none; the allocation is based on architecture priors |
|
| 134 |
-
| Quantizer execution | 289/289 recorded module conversions succeeded; 0 fallbacks |
|
| 135 |
-
| AX Engine native manifest | not included |
|
| 136 |
-
| Quality versus BF16 or uniform baselines | Not published; no quality-retention claim |
|
| 137 |
-
| MTP acceptance and speed | not measured; no MTP speedup claim |
|
| 138 |
-
| AX Engine kernel evidence | `unmeasured` |
|
| 139 |
-
| Vision-language quality | Not applicable (no vision tower in this package) |
|
| 140 |
-
| Speech-recognition quality | Not applicable |
|
| 141 |
-
| Long-context quality | 131,072-token capacity is config metadata, not a validated claim |
|
| 142 |
-
| Release certification | **Not certified**; formal AXQuant M0-M8 gates are not closed |
|
| 143 |
-
|
| 144 |
-
## Intended use and limitations
|
| 145 |
-
|
| 146 |
-
- Intended for local development and evaluation on Apple Silicon with MLX-compatible runtimes.
|
| 147 |
-
- No minimum unified-memory figure is claimed; loadability depends on model size, context length,
|
| 148 |
-
KV-cache policy, runtime buffers, and other processes using unified memory.
|
| 149 |
-
- Architecture-prior allocation is not measured sensitivity. It must not be presented as measured
|
| 150 |
-
model quality.
|
| 151 |
-
- The configured context window can require substantially more memory as the KV cache grows.
|
| 152 |
-
- AX Engine execution is not established because this package has no validated native manifest.
|
| 153 |
|
| 154 |
-
|
| 155 |
|
| 156 |
-
|
|
|
|
|
|
|
| 157 |
|
| 158 |
-
|
| 159 |
-
contract, software versions, and file checksums.
|
| 160 |
-
- [`axquant_plan.json`](axquant_plan.json): per-tensor precision decisions and planning evidence.
|
| 161 |
-
- [`axquant_quantizer_execution.json`](axquant_quantizer_execution.json): conversion coverage and
|
| 162 |
-
fallback records.
|
| 163 |
-
- [`axquant_runtime.json`](axquant_runtime.json): declared AX Engine and MLX compatibility metadata; runtime checks remain separate evidence.
|
| 164 |
|
| 165 |
-
|
| 166 |
-
|
| 167 |
-
|
| 168 |
-
|
| 169 |
|
| 170 |
-
##
|
| 171 |
|
| 172 |
-
|
| 173 |
-
|
| 174 |
-
|
|
|
|
| 1 |
---
|
| 2 |
+
language: en
|
| 3 |
library_name: mlx
|
|
|
|
|
|
|
| 4 |
pipeline_tag: text-generation
|
| 5 |
tags:
|
| 6 |
- mlx
|
|
|
|
|
|
|
|
|
|
| 7 |
- axquant
|
|
|
|
|
|
|
| 8 |
- gpt-oss
|
| 9 |
+
- moe
|
| 10 |
+
license: apache-2.0
|
| 11 |
+
base_model: openai/gpt-oss-120b
|
| 12 |
---
|
| 13 |
|
| 14 |
# AX-gpt-oss-120b-MLX-AXQ-6bit
|
| 15 |
|
| 16 |
+
AXQuant affine re-pack of mlx-community/gpt-oss-120b-MXFP4-Q4 for Apple Silicon MLX.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
|
| 18 |
| Property | Value |
|
| 19 |
| --- | --- |
|
| 20 |
+
| Product | AXQ 6-bit (agent-coding manual recipe, no 4-bit trunk) |
|
| 21 |
+
| Measured total BPW | 6.577 |
|
| 22 |
+
| Architecture | GptOssForCausalLM (MoE, no MTP) |
|
| 23 |
+
| Source | mlx-community/gpt-oss-120b-MXFP4-Q4@bce781bef0f2fc85ed4e575af74054f5aad73ddd |
|
| 24 |
+
| Upstream | openai/gpt-oss-120b |
|
| 25 |
+
| Plan | plan-manual agent-coding: experts 6-bit, attention 8-bit, no 4-bit |
|
| 26 |
+
| Runtime | MLX-LM |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 27 |
|
| 28 |
+
## Checkpoint Tier 1
|
| 29 |
|
| 30 |
+
Certified on host df-macbookpro-m5 with AXQuant 1.6.1 development suites
|
| 31 |
+
(agent-coding + general, seed 20260728, max tokens 64) vs the matched MXFP4-Q4
|
| 32 |
+
reference. MTP Tier 2 is not applicable (no MTP).
|
| 33 |
|
| 34 |
+
## Load
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
|
| 36 |
+
\`\`\`bash
|
| 37 |
+
pip install mlx-lm
|
| 38 |
+
python -m mlx_lm.generate --model AutomatosX/AX-gpt-oss-120b-MLX-AXQ-6bit --prompt Hello
|
| 39 |
+
\`\`\`
|
| 40 |
|
| 41 |
+
## Notes
|
| 42 |
|
| 43 |
+
- Converted with AXQUANT_FORCE_CPU=1 after Metal GPU timeouts on large re-pack.
|
| 44 |
+
- Size ratio vs MXFP4-Q4 is ~1.54 (within the 6-bit max 1.55 gate).
|
| 45 |
+
- Higher-fidelity 6-bit product layout (storage-adjusted BPW ~6.58), not uniform 6.0.
|
axquant_manifest.json
CHANGED
|
@@ -1,27 +1,27 @@
|
|
| 1 |
{
|
| 2 |
"axquant_version": "1.6.1",
|
| 3 |
"calibration": null,
|
| 4 |
-
"created_at": "2026-08-
|
| 5 |
-
"effective_bpw":
|
| 6 |
"files": [
|
| 7 |
{
|
| 8 |
"path": "README.md",
|
| 9 |
-
"sha256": "
|
| 10 |
-
"size_bytes":
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"path": "axquant_plan.json",
|
| 14 |
-
"sha256": "
|
| 15 |
-
"size_bytes":
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"path": "axquant_quantizer_execution.json",
|
| 19 |
-
"sha256": "
|
| 20 |
"size_bytes": 71171
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"path": "axquant_runtime.json",
|
| 24 |
-
"sha256": "
|
| 25 |
"size_bytes": 1628
|
| 26 |
},
|
| 27 |
{
|
|
@@ -31,7 +31,7 @@
|
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"path": "config.json",
|
| 34 |
-
"sha256": "
|
| 35 |
"size_bytes": 83027
|
| 36 |
},
|
| 37 |
{
|
|
@@ -40,93 +40,103 @@
|
|
| 40 |
"size_bytes": 177
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"path": "model-00001-of-
|
| 44 |
-
"sha256": "
|
| 45 |
-
"size_bytes":
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"path": "model-00002-of-
|
| 49 |
-
"sha256": "
|
| 50 |
-
"size_bytes":
|
| 51 |
},
|
| 52 |
{
|
| 53 |
-
"path": "model-00003-of-
|
| 54 |
-
"sha256": "
|
| 55 |
-
"size_bytes":
|
| 56 |
},
|
| 57 |
{
|
| 58 |
-
"path": "model-00004-of-
|
| 59 |
-
"sha256": "
|
| 60 |
-
"size_bytes":
|
| 61 |
},
|
| 62 |
{
|
| 63 |
-
"path": "model-00005-of-
|
| 64 |
-
"sha256": "
|
| 65 |
-
"size_bytes":
|
| 66 |
},
|
| 67 |
{
|
| 68 |
-
"path": "model-00006-of-
|
| 69 |
-
"sha256": "
|
| 70 |
-
"size_bytes":
|
| 71 |
},
|
| 72 |
{
|
| 73 |
-
"path": "model-00007-of-
|
| 74 |
-
"sha256": "
|
| 75 |
-
"size_bytes":
|
| 76 |
},
|
| 77 |
{
|
| 78 |
-
"path": "model-00008-of-
|
| 79 |
-
"sha256": "
|
| 80 |
-
"size_bytes":
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"path": "model-00009-of-
|
| 84 |
-
"sha256": "
|
| 85 |
-
"size_bytes":
|
| 86 |
},
|
| 87 |
{
|
| 88 |
-
"path": "model-00010-of-
|
| 89 |
-
"sha256": "
|
| 90 |
-
"size_bytes":
|
| 91 |
},
|
| 92 |
{
|
| 93 |
-
"path": "model-00011-of-
|
| 94 |
-
"sha256": "
|
| 95 |
-
"size_bytes":
|
| 96 |
},
|
| 97 |
{
|
| 98 |
-
"path": "model-00012-of-
|
| 99 |
-
"sha256": "
|
| 100 |
-
"size_bytes":
|
| 101 |
},
|
| 102 |
{
|
| 103 |
-
"path": "model-00013-of-
|
| 104 |
-
"sha256": "
|
| 105 |
-
"size_bytes":
|
| 106 |
},
|
| 107 |
{
|
| 108 |
-
"path": "model-00014-of-
|
| 109 |
-
"sha256": "
|
| 110 |
-
"size_bytes":
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"path": "model-00015-of-
|
| 114 |
-
"sha256": "
|
| 115 |
-
"size_bytes":
|
| 116 |
},
|
| 117 |
{
|
| 118 |
-
"path": "model-00016-of-
|
| 119 |
-
"sha256": "
|
| 120 |
-
"size_bytes":
|
| 121 |
},
|
| 122 |
{
|
| 123 |
-
"path": "model-00017-of-
|
| 124 |
-
"sha256": "
|
| 125 |
-
"size_bytes":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 126 |
},
|
| 127 |
{
|
| 128 |
"path": "model.safetensors.index.json",
|
| 129 |
-
"sha256": "
|
| 130 |
"size_bytes": 110141
|
| 131 |
},
|
| 132 |
{
|
|
@@ -143,9 +153,9 @@
|
|
| 143 |
"format": "mlx",
|
| 144 |
"logical_parameters": 116829156672,
|
| 145 |
"main_logical_parameters": 116829156672,
|
| 146 |
-
"main_weight_file_size_bytes":
|
| 147 |
-
"measured_main_bpw": 6.
|
| 148 |
-
"measured_total_bpw": 6.
|
| 149 |
"mtp_acceptance_retention": null,
|
| 150 |
"mtp_distribution": {},
|
| 151 |
"mtp_measured_speedup": null,
|
|
@@ -163,8 +173,8 @@
|
|
| 163 |
},
|
| 164 |
"mtp_present": false,
|
| 165 |
"mtp_weight_file_size_bytes": 0,
|
| 166 |
-
"plan_sha256": "
|
| 167 |
-
"profile": "
|
| 168 |
"protected_weight_file_size_bytes": 0,
|
| 169 |
"quantizer": "axquant",
|
| 170 |
"runtime": {
|
|
@@ -173,7 +183,7 @@
|
|
| 173 |
"fused_mtp": null,
|
| 174 |
"kernel_evidence": "unmeasured",
|
| 175 |
"model_manifest": "model-manifest.json",
|
| 176 |
-
"preferred_group_size":
|
| 177 |
},
|
| 178 |
"compatible_runtimes": [
|
| 179 |
{
|
|
@@ -190,7 +200,7 @@
|
|
| 190 |
"support_level": "standard-inference"
|
| 191 |
}
|
| 192 |
],
|
| 193 |
-
"created_at": "2026-08-
|
| 194 |
"kv_cache": null,
|
| 195 |
"memory_policy": {
|
| 196 |
"kv_cache_precision": "runtime-default",
|
|
@@ -238,28 +248,24 @@
|
|
| 238 |
"source_model": {
|
| 239 |
"architecture": "GptOssForCausalLM",
|
| 240 |
"format": "mlx",
|
| 241 |
-
"local_path":
|
| 242 |
"model_id": "openai/gpt-oss-120b",
|
| 243 |
"revision": "bce781bef0f2fc85ed4e575af74054f5aad73ddd"
|
| 244 |
},
|
| 245 |
-
"target_class": "
|
| 246 |
"weight_distribution": {
|
| 247 |
-
"4bit": {
|
| 248 |
-
"fraction": 0.3816743668294139,
|
| 249 |
-
"parameters": 44590694400
|
| 250 |
-
},
|
| 251 |
"6bit": {
|
| 252 |
-
"fraction": 0.
|
| 253 |
-
"parameters":
|
| 254 |
},
|
| 255 |
"8bit": {
|
| 256 |
-
"fraction": 0.
|
| 257 |
-
"parameters":
|
| 258 |
},
|
| 259 |
"bf16": {
|
| 260 |
"fraction": 0.005302201348068633,
|
| 261 |
"parameters": 619451712
|
| 262 |
}
|
| 263 |
},
|
| 264 |
-
"weight_file_size_bytes":
|
| 265 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"axquant_version": "1.6.1",
|
| 3 |
"calibration": null,
|
| 4 |
+
"created_at": "2026-08-11T02:36:33.788678Z",
|
| 5 |
+
"effective_bpw": 6.576869766416386,
|
| 6 |
"files": [
|
| 7 |
{
|
| 8 |
"path": "README.md",
|
| 9 |
+
"sha256": "748c964f9b7e5f2c3770ce013bbc0153be7c54311d9692c343b6188eefe77ac6",
|
| 10 |
+
"size_bytes": 81
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"path": "axquant_plan.json",
|
| 14 |
+
"sha256": "83ea85356db58bfb1811d32a15fd7d9bfe02165e0eba63bf6b3bb777bb6e0e5a",
|
| 15 |
+
"size_bytes": 1022081
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"path": "axquant_quantizer_execution.json",
|
| 19 |
+
"sha256": "d11f70a755221daa07fde8ba1fd85fd619420238f0b52d52c37b9ca41a1145c1",
|
| 20 |
"size_bytes": 71171
|
| 21 |
},
|
| 22 |
{
|
| 23 |
"path": "axquant_runtime.json",
|
| 24 |
+
"sha256": "bcb11a41c0f74dd932cd17cb1af325ffc2adf0ee2eca24ac99e7620551579362",
|
| 25 |
"size_bytes": 1628
|
| 26 |
},
|
| 27 |
{
|
|
|
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"path": "config.json",
|
| 34 |
+
"sha256": "43ec137f5726216cc222022f89a4009fb8fa7aef2726db2c16c9eb61463af4a6",
|
| 35 |
"size_bytes": 83027
|
| 36 |
},
|
| 37 |
{
|
|
|
|
| 40 |
"size_bytes": 177
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"path": "model-00001-of-00019.safetensors",
|
| 44 |
+
"sha256": "1186e99a101cfbc2fb7ddbc4343b4daa9bd2e8d5f06203f86db797b37f6a0b95",
|
| 45 |
+
"size_bytes": 4988954394
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"path": "model-00002-of-00019.safetensors",
|
| 49 |
+
"sha256": "baf5106a0043aae7788a5c2a46e01e15581d96a7c2d3b7bdac6d1dc39105e172",
|
| 50 |
+
"size_bytes": 5237378386
|
| 51 |
},
|
| 52 |
{
|
| 53 |
+
"path": "model-00003-of-00019.safetensors",
|
| 54 |
+
"sha256": "ff4e87775512d8f36f6fb7d355bb2b33c4798c65be9516ba7ad3c4c9dc7fef7a",
|
| 55 |
+
"size_bytes": 5237378426
|
| 56 |
},
|
| 57 |
{
|
| 58 |
+
"path": "model-00004-of-00019.safetensors",
|
| 59 |
+
"sha256": "7b42f472293c0793843bc8daa79d263507049bc31b9704cfee2802112f91340e",
|
| 60 |
+
"size_bytes": 5237378464
|
| 61 |
},
|
| 62 |
{
|
| 63 |
+
"path": "model-00005-of-00019.safetensors",
|
| 64 |
+
"sha256": "21d2e0666b8156cf9d887bb93da3438f89fd02072cdc334320268fb25f57e6fa",
|
| 65 |
+
"size_bytes": 5237378426
|
| 66 |
},
|
| 67 |
{
|
| 68 |
+
"path": "model-00006-of-00019.safetensors",
|
| 69 |
+
"sha256": "3bb4f5ad61f197a33cfe154ea704c9c8b12e68b0cfe9f8b796677e726b5d86ea",
|
| 70 |
+
"size_bytes": 5237378534
|
| 71 |
},
|
| 72 |
{
|
| 73 |
+
"path": "model-00007-of-00019.safetensors",
|
| 74 |
+
"sha256": "54b5439d12ab36b58ab659fd258e8792d7319e1befd06aeeb48f92e61129ada2",
|
| 75 |
+
"size_bytes": 5237378486
|
| 76 |
},
|
| 77 |
{
|
| 78 |
+
"path": "model-00008-of-00019.safetensors",
|
| 79 |
+
"sha256": "37b717aae69360982abda9165fd13b965187b28d6031f04e9906bc59cc4bbf45",
|
| 80 |
+
"size_bytes": 5237378478
|
| 81 |
},
|
| 82 |
{
|
| 83 |
+
"path": "model-00009-of-00019.safetensors",
|
| 84 |
+
"sha256": "64d3981b098d5ce152f94a44d061c2e285e0900f403a3f4a731e3eb04c0e0723",
|
| 85 |
+
"size_bytes": 5237378526
|
| 86 |
},
|
| 87 |
{
|
| 88 |
+
"path": "model-00010-of-00019.safetensors",
|
| 89 |
+
"sha256": "b5a7b1492d3c61d9542fbca6ee655b2e16596f21bcd407f981b35ec6d8bbfc39",
|
| 90 |
+
"size_bytes": 5237378464
|
| 91 |
},
|
| 92 |
{
|
| 93 |
+
"path": "model-00011-of-00019.safetensors",
|
| 94 |
+
"sha256": "8f83b7530039fad1ac2e514abd1f800b13b1e9e5dc97e150402fce62bace4670",
|
| 95 |
+
"size_bytes": 5237378494
|
| 96 |
},
|
| 97 |
{
|
| 98 |
+
"path": "model-00012-of-00019.safetensors",
|
| 99 |
+
"sha256": "d352dda8c10083b530549c33c44798d639609f17337eeb057745fcc3b2340122",
|
| 100 |
+
"size_bytes": 5237378486
|
| 101 |
},
|
| 102 |
{
|
| 103 |
+
"path": "model-00013-of-00019.safetensors",
|
| 104 |
+
"sha256": "f07fbaf05c6cefeb5d896c49c751355b867f3609d0ac9bdc0badd4590d45eb45",
|
| 105 |
+
"size_bytes": 5237378490
|
| 106 |
},
|
| 107 |
{
|
| 108 |
+
"path": "model-00014-of-00019.safetensors",
|
| 109 |
+
"sha256": "1ea0c6fa4285cb7e658f7d9720483c2bb8398f7d1d5e9e607b2243a30fe2c791",
|
| 110 |
+
"size_bytes": 5237378536
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"path": "model-00015-of-00019.safetensors",
|
| 114 |
+
"sha256": "9124aa1feb8abff1c72bdc0411fc82d32a2cf007a3c27e3516a44d17973f07c9",
|
| 115 |
+
"size_bytes": 5237378510
|
| 116 |
},
|
| 117 |
{
|
| 118 |
+
"path": "model-00016-of-00019.safetensors",
|
| 119 |
+
"sha256": "33d4fa556a17200392b588923642bcbe6af1a7bf9bf1b3ab8178ddb8ffdf568d",
|
| 120 |
+
"size_bytes": 5237378510
|
| 121 |
},
|
| 122 |
{
|
| 123 |
+
"path": "model-00017-of-00019.safetensors",
|
| 124 |
+
"sha256": "2dd8eff0b3bccc1c113c34d22d401ccb73cb9e544d4169fa80bad93f0480eb4e",
|
| 125 |
+
"size_bytes": 5237378506
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"path": "model-00018-of-00019.safetensors",
|
| 129 |
+
"sha256": "172d816c062df1087c0fa447015bfc7cddcac32fea8e535f40f10c8991aead51",
|
| 130 |
+
"size_bytes": 5237378510
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"path": "model-00019-of-00019.safetensors",
|
| 134 |
+
"sha256": "ec18218e771c835577fd99a07783d698d275d05f07d1974af311da71bc92407d",
|
| 135 |
+
"size_bytes": 2022026460
|
| 136 |
},
|
| 137 |
{
|
| 138 |
"path": "model.safetensors.index.json",
|
| 139 |
+
"sha256": "79420b8a10f3f47638273aa9299be8fa70698ce86eadcd99ee779cbf202d10b4",
|
| 140 |
"size_bytes": 110141
|
| 141 |
},
|
| 142 |
{
|
|
|
|
| 153 |
"format": "mlx",
|
| 154 |
"logical_parameters": 116829156672,
|
| 155 |
"main_logical_parameters": 116829156672,
|
| 156 |
+
"main_weight_file_size_bytes": 96046415086,
|
| 157 |
+
"measured_main_bpw": 6.576879801034742,
|
| 158 |
+
"measured_total_bpw": 6.576879801034742,
|
| 159 |
"mtp_acceptance_retention": null,
|
| 160 |
"mtp_distribution": {},
|
| 161 |
"mtp_measured_speedup": null,
|
|
|
|
| 173 |
},
|
| 174 |
"mtp_present": false,
|
| 175 |
"mtp_weight_file_size_bytes": 0,
|
| 176 |
+
"plan_sha256": "4c680c40943529cae51fa0adc80420aaf5ada6e00bb5cca9a85ed2e4b88ffde1",
|
| 177 |
+
"profile": "agent-coding",
|
| 178 |
"protected_weight_file_size_bytes": 0,
|
| 179 |
"quantizer": "axquant",
|
| 180 |
"runtime": {
|
|
|
|
| 183 |
"fused_mtp": null,
|
| 184 |
"kernel_evidence": "unmeasured",
|
| 185 |
"model_manifest": "model-manifest.json",
|
| 186 |
+
"preferred_group_size": 64
|
| 187 |
},
|
| 188 |
"compatible_runtimes": [
|
| 189 |
{
|
|
|
|
| 200 |
"support_level": "standard-inference"
|
| 201 |
}
|
| 202 |
],
|
| 203 |
+
"created_at": "2026-08-11T02:33:57.174300Z",
|
| 204 |
"kv_cache": null,
|
| 205 |
"memory_policy": {
|
| 206 |
"kv_cache_precision": "runtime-default",
|
|
|
|
| 248 |
"source_model": {
|
| 249 |
"architecture": "GptOssForCausalLM",
|
| 250 |
"format": "mlx",
|
| 251 |
+
"local_path": "/Volumes/Ext4T/huggingface/hub/models--mlx-community--gpt-oss-120b-MXFP4-Q4/snapshots/bce781bef0f2fc85ed4e575af74054f5aad73ddd",
|
| 252 |
"model_id": "openai/gpt-oss-120b",
|
| 253 |
"revision": "bce781bef0f2fc85ed4e575af74054f5aad73ddd"
|
| 254 |
},
|
| 255 |
+
"target_class": "6p6bpw",
|
| 256 |
"weight_distribution": {
|
|
|
|
|
|
|
|
|
|
|
|
|
| 257 |
"6bit": {
|
| 258 |
+
"fraction": 0.9814483718470643,
|
| 259 |
+
"parameters": 114661785600
|
| 260 |
},
|
| 261 |
"8bit": {
|
| 262 |
+
"fraction": 0.013249426804866973,
|
| 263 |
+
"parameters": 1547919360
|
| 264 |
},
|
| 265 |
"bf16": {
|
| 266 |
"fraction": 0.005302201348068633,
|
| 267 |
"parameters": 619451712
|
| 268 |
}
|
| 269 |
},
|
| 270 |
+
"weight_file_size_bytes": 96046415086
|
| 271 |
}
|
axquant_plan.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|
axquant_quantizer_execution.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
-
"created_at": "2026-08-
|
| 3 |
-
"plan_sha256": "
|
| 4 |
"records": [
|
| 5 |
{
|
| 6 |
"bits": 8,
|
|
@@ -13,9 +13,9 @@
|
|
| 13 |
"success": true
|
| 14 |
},
|
| 15 |
{
|
| 16 |
-
"bits":
|
| 17 |
"fallback": false,
|
| 18 |
-
"group_size":
|
| 19 |
"metadata": {},
|
| 20 |
"method": "affine",
|
| 21 |
"module_path": "model.layers.0.mlp.experts.down_proj",
|
|
@@ -23,9 +23,9 @@
|
|
| 23 |
"success": true
|
| 24 |
},
|
| 25 |
{
|
| 26 |
-
"bits":
|
| 27 |
"fallback": false,
|
| 28 |
-
"group_size":
|
| 29 |
"metadata": {},
|
| 30 |
"method": "affine",
|
| 31 |
"module_path": "model.layers.0.mlp.experts.gate_proj",
|
|
@@ -33,9 +33,9 @@
|
|
| 33 |
"success": true
|
| 34 |
},
|
| 35 |
{
|
| 36 |
-
"bits":
|
| 37 |
"fallback": false,
|
| 38 |
-
"group_size":
|
| 39 |
"metadata": {},
|
| 40 |
"method": "affine",
|
| 41 |
"module_path": "model.layers.0.mlp.experts.up_proj",
|
|
@@ -53,9 +53,9 @@
|
|
| 53 |
"success": true
|
| 54 |
},
|
| 55 |
{
|
| 56 |
-
"bits":
|
| 57 |
"fallback": false,
|
| 58 |
-
"group_size":
|
| 59 |
"metadata": {},
|
| 60 |
"method": "affine",
|
| 61 |
"module_path": "model.layers.0.self_attn.k_proj",
|
|
@@ -63,9 +63,9 @@
|
|
| 63 |
"success": true
|
| 64 |
},
|
| 65 |
{
|
| 66 |
-
"bits":
|
| 67 |
"fallback": false,
|
| 68 |
-
"group_size":
|
| 69 |
"metadata": {},
|
| 70 |
"method": "affine",
|
| 71 |
"module_path": "model.layers.0.self_attn.o_proj",
|
|
@@ -73,9 +73,9 @@
|
|
| 73 |
"success": true
|
| 74 |
},
|
| 75 |
{
|
| 76 |
-
"bits":
|
| 77 |
"fallback": false,
|
| 78 |
-
"group_size":
|
| 79 |
"metadata": {},
|
| 80 |
"method": "affine",
|
| 81 |
"module_path": "model.layers.0.self_attn.q_proj",
|
|
@@ -83,9 +83,9 @@
|
|
| 83 |
"success": true
|
| 84 |
},
|
| 85 |
{
|
| 86 |
-
"bits":
|
| 87 |
"fallback": false,
|
| 88 |
-
"group_size":
|
| 89 |
"metadata": {},
|
| 90 |
"method": "affine",
|
| 91 |
"module_path": "model.layers.0.self_attn.v_proj",
|
|
@@ -93,9 +93,9 @@
|
|
| 93 |
"success": true
|
| 94 |
},
|
| 95 |
{
|
| 96 |
-
"bits":
|
| 97 |
"fallback": false,
|
| 98 |
-
"group_size":
|
| 99 |
"metadata": {},
|
| 100 |
"method": "affine",
|
| 101 |
"module_path": "model.layers.1.mlp.experts.down_proj",
|
|
@@ -103,9 +103,9 @@
|
|
| 103 |
"success": true
|
| 104 |
},
|
| 105 |
{
|
| 106 |
-
"bits":
|
| 107 |
"fallback": false,
|
| 108 |
-
"group_size":
|
| 109 |
"metadata": {},
|
| 110 |
"method": "affine",
|
| 111 |
"module_path": "model.layers.1.mlp.experts.gate_proj",
|
|
@@ -113,9 +113,9 @@
|
|
| 113 |
"success": true
|
| 114 |
},
|
| 115 |
{
|
| 116 |
-
"bits":
|
| 117 |
"fallback": false,
|
| 118 |
-
"group_size":
|
| 119 |
"metadata": {},
|
| 120 |
"method": "affine",
|
| 121 |
"module_path": "model.layers.1.mlp.experts.up_proj",
|
|
@@ -133,9 +133,9 @@
|
|
| 133 |
"success": true
|
| 134 |
},
|
| 135 |
{
|
| 136 |
-
"bits":
|
| 137 |
"fallback": false,
|
| 138 |
-
"group_size":
|
| 139 |
"metadata": {},
|
| 140 |
"method": "affine",
|
| 141 |
"module_path": "model.layers.1.self_attn.k_proj",
|
|
@@ -143,9 +143,9 @@
|
|
| 143 |
"success": true
|
| 144 |
},
|
| 145 |
{
|
| 146 |
-
"bits":
|
| 147 |
"fallback": false,
|
| 148 |
-
"group_size":
|
| 149 |
"metadata": {},
|
| 150 |
"method": "affine",
|
| 151 |
"module_path": "model.layers.1.self_attn.o_proj",
|
|
@@ -153,9 +153,9 @@
|
|
| 153 |
"success": true
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"bits":
|
| 157 |
"fallback": false,
|
| 158 |
-
"group_size":
|
| 159 |
"metadata": {},
|
| 160 |
"method": "affine",
|
| 161 |
"module_path": "model.layers.1.self_attn.q_proj",
|
|
@@ -163,9 +163,9 @@
|
|
| 163 |
"success": true
|
| 164 |
},
|
| 165 |
{
|
| 166 |
-
"bits":
|
| 167 |
"fallback": false,
|
| 168 |
-
"group_size":
|
| 169 |
"metadata": {},
|
| 170 |
"method": "affine",
|
| 171 |
"module_path": "model.layers.1.self_attn.v_proj",
|
|
@@ -173,9 +173,9 @@
|
|
| 173 |
"success": true
|
| 174 |
},
|
| 175 |
{
|
| 176 |
-
"bits":
|
| 177 |
"fallback": false,
|
| 178 |
-
"group_size":
|
| 179 |
"metadata": {},
|
| 180 |
"method": "affine",
|
| 181 |
"module_path": "model.layers.2.mlp.experts.gate_proj",
|
|
@@ -183,9 +183,9 @@
|
|
| 183 |
"success": true
|
| 184 |
},
|
| 185 |
{
|
| 186 |
-
"bits":
|
| 187 |
"fallback": false,
|
| 188 |
-
"group_size":
|
| 189 |
"metadata": {},
|
| 190 |
"method": "affine",
|
| 191 |
"module_path": "model.layers.2.mlp.experts.up_proj",
|
|
@@ -193,9 +193,9 @@
|
|
| 193 |
"success": true
|
| 194 |
},
|
| 195 |
{
|
| 196 |
-
"bits":
|
| 197 |
"fallback": false,
|
| 198 |
-
"group_size":
|
| 199 |
"metadata": {},
|
| 200 |
"method": "affine",
|
| 201 |
"module_path": "model.layers.2.self_attn.k_proj",
|
|
@@ -203,9 +203,9 @@
|
|
| 203 |
"success": true
|
| 204 |
},
|
| 205 |
{
|
| 206 |
-
"bits":
|
| 207 |
"fallback": false,
|
| 208 |
-
"group_size":
|
| 209 |
"metadata": {},
|
| 210 |
"method": "affine",
|
| 211 |
"module_path": "model.layers.2.self_attn.o_proj",
|
|
@@ -213,9 +213,9 @@
|
|
| 213 |
"success": true
|
| 214 |
},
|
| 215 |
{
|
| 216 |
-
"bits":
|
| 217 |
"fallback": false,
|
| 218 |
-
"group_size":
|
| 219 |
"metadata": {},
|
| 220 |
"method": "affine",
|
| 221 |
"module_path": "model.layers.2.self_attn.q_proj",
|
|
@@ -223,9 +223,9 @@
|
|
| 223 |
"success": true
|
| 224 |
},
|
| 225 |
{
|
| 226 |
-
"bits":
|
| 227 |
"fallback": false,
|
| 228 |
-
"group_size":
|
| 229 |
"metadata": {},
|
| 230 |
"method": "affine",
|
| 231 |
"module_path": "model.layers.2.self_attn.v_proj",
|
|
@@ -233,9 +233,9 @@
|
|
| 233 |
"success": true
|
| 234 |
},
|
| 235 |
{
|
| 236 |
-
"bits":
|
| 237 |
"fallback": false,
|
| 238 |
-
"group_size":
|
| 239 |
"metadata": {},
|
| 240 |
"method": "affine",
|
| 241 |
"module_path": "model.layers.2.mlp.experts.down_proj",
|
|
@@ -253,9 +253,9 @@
|
|
| 253 |
"success": true
|
| 254 |
},
|
| 255 |
{
|
| 256 |
-
"bits":
|
| 257 |
"fallback": false,
|
| 258 |
-
"group_size":
|
| 259 |
"metadata": {},
|
| 260 |
"method": "affine",
|
| 261 |
"module_path": "model.layers.3.mlp.experts.down_proj",
|
|
@@ -263,9 +263,9 @@
|
|
| 263 |
"success": true
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"bits":
|
| 267 |
"fallback": false,
|
| 268 |
-
"group_size":
|
| 269 |
"metadata": {},
|
| 270 |
"method": "affine",
|
| 271 |
"module_path": "model.layers.3.mlp.experts.gate_proj",
|
|
@@ -273,9 +273,9 @@
|
|
| 273 |
"success": true
|
| 274 |
},
|
| 275 |
{
|
| 276 |
-
"bits":
|
| 277 |
"fallback": false,
|
| 278 |
-
"group_size":
|
| 279 |
"metadata": {},
|
| 280 |
"method": "affine",
|
| 281 |
"module_path": "model.layers.3.mlp.experts.up_proj",
|
|
@@ -293,9 +293,9 @@
|
|
| 293 |
"success": true
|
| 294 |
},
|
| 295 |
{
|
| 296 |
-
"bits":
|
| 297 |
"fallback": false,
|
| 298 |
-
"group_size":
|
| 299 |
"metadata": {},
|
| 300 |
"method": "affine",
|
| 301 |
"module_path": "model.layers.3.self_attn.k_proj",
|
|
@@ -303,9 +303,9 @@
|
|
| 303 |
"success": true
|
| 304 |
},
|
| 305 |
{
|
| 306 |
-
"bits":
|
| 307 |
"fallback": false,
|
| 308 |
-
"group_size":
|
| 309 |
"metadata": {},
|
| 310 |
"method": "affine",
|
| 311 |
"module_path": "model.layers.3.self_attn.o_proj",
|
|
@@ -313,9 +313,9 @@
|
|
| 313 |
"success": true
|
| 314 |
},
|
| 315 |
{
|
| 316 |
-
"bits":
|
| 317 |
"fallback": false,
|
| 318 |
-
"group_size":
|
| 319 |
"metadata": {},
|
| 320 |
"method": "affine",
|
| 321 |
"module_path": "model.layers.3.self_attn.q_proj",
|
|
@@ -323,9 +323,9 @@
|
|
| 323 |
"success": true
|
| 324 |
},
|
| 325 |
{
|
| 326 |
-
"bits":
|
| 327 |
"fallback": false,
|
| 328 |
-
"group_size":
|
| 329 |
"metadata": {},
|
| 330 |
"method": "affine",
|
| 331 |
"module_path": "model.layers.3.self_attn.v_proj",
|
|
@@ -333,9 +333,9 @@
|
|
| 333 |
"success": true
|
| 334 |
},
|
| 335 |
{
|
| 336 |
-
"bits":
|
| 337 |
"fallback": false,
|
| 338 |
-
"group_size":
|
| 339 |
"metadata": {},
|
| 340 |
"method": "affine",
|
| 341 |
"module_path": "model.layers.4.mlp.experts.down_proj",
|
|
@@ -343,9 +343,9 @@
|
|
| 343 |
"success": true
|
| 344 |
},
|
| 345 |
{
|
| 346 |
-
"bits":
|
| 347 |
"fallback": false,
|
| 348 |
-
"group_size":
|
| 349 |
"metadata": {},
|
| 350 |
"method": "affine",
|
| 351 |
"module_path": "model.layers.4.mlp.experts.gate_proj",
|
|
@@ -353,9 +353,9 @@
|
|
| 353 |
"success": true
|
| 354 |
},
|
| 355 |
{
|
| 356 |
-
"bits":
|
| 357 |
"fallback": false,
|
| 358 |
-
"group_size":
|
| 359 |
"metadata": {},
|
| 360 |
"method": "affine",
|
| 361 |
"module_path": "model.layers.4.mlp.experts.up_proj",
|
|
@@ -373,9 +373,9 @@
|
|
| 373 |
"success": true
|
| 374 |
},
|
| 375 |
{
|
| 376 |
-
"bits":
|
| 377 |
"fallback": false,
|
| 378 |
-
"group_size":
|
| 379 |
"metadata": {},
|
| 380 |
"method": "affine",
|
| 381 |
"module_path": "model.layers.4.self_attn.k_proj",
|
|
@@ -383,9 +383,9 @@
|
|
| 383 |
"success": true
|
| 384 |
},
|
| 385 |
{
|
| 386 |
-
"bits":
|
| 387 |
"fallback": false,
|
| 388 |
-
"group_size":
|
| 389 |
"metadata": {},
|
| 390 |
"method": "affine",
|
| 391 |
"module_path": "model.layers.4.self_attn.o_proj",
|
|
@@ -393,9 +393,9 @@
|
|
| 393 |
"success": true
|
| 394 |
},
|
| 395 |
{
|
| 396 |
-
"bits":
|
| 397 |
"fallback": false,
|
| 398 |
-
"group_size":
|
| 399 |
"metadata": {},
|
| 400 |
"method": "affine",
|
| 401 |
"module_path": "model.layers.4.self_attn.q_proj",
|
|
@@ -403,9 +403,9 @@
|
|
| 403 |
"success": true
|
| 404 |
},
|
| 405 |
{
|
| 406 |
-
"bits":
|
| 407 |
"fallback": false,
|
| 408 |
-
"group_size":
|
| 409 |
"metadata": {},
|
| 410 |
"method": "affine",
|
| 411 |
"module_path": "model.layers.4.self_attn.v_proj",
|
|
@@ -413,9 +413,9 @@
|
|
| 413 |
"success": true
|
| 414 |
},
|
| 415 |
{
|
| 416 |
-
"bits":
|
| 417 |
"fallback": false,
|
| 418 |
-
"group_size":
|
| 419 |
"metadata": {},
|
| 420 |
"method": "affine",
|
| 421 |
"module_path": "model.layers.5.mlp.experts.gate_proj",
|
|
@@ -423,9 +423,9 @@
|
|
| 423 |
"success": true
|
| 424 |
},
|
| 425 |
{
|
| 426 |
-
"bits":
|
| 427 |
"fallback": false,
|
| 428 |
-
"group_size":
|
| 429 |
"metadata": {},
|
| 430 |
"method": "affine",
|
| 431 |
"module_path": "model.layers.5.mlp.experts.up_proj",
|
|
@@ -433,9 +433,9 @@
|
|
| 433 |
"success": true
|
| 434 |
},
|
| 435 |
{
|
| 436 |
-
"bits":
|
| 437 |
"fallback": false,
|
| 438 |
-
"group_size":
|
| 439 |
"metadata": {},
|
| 440 |
"method": "affine",
|
| 441 |
"module_path": "model.layers.5.self_attn.k_proj",
|
|
@@ -443,9 +443,9 @@
|
|
| 443 |
"success": true
|
| 444 |
},
|
| 445 |
{
|
| 446 |
-
"bits":
|
| 447 |
"fallback": false,
|
| 448 |
-
"group_size":
|
| 449 |
"metadata": {},
|
| 450 |
"method": "affine",
|
| 451 |
"module_path": "model.layers.5.self_attn.o_proj",
|
|
@@ -453,9 +453,9 @@
|
|
| 453 |
"success": true
|
| 454 |
},
|
| 455 |
{
|
| 456 |
-
"bits":
|
| 457 |
"fallback": false,
|
| 458 |
-
"group_size":
|
| 459 |
"metadata": {},
|
| 460 |
"method": "affine",
|
| 461 |
"module_path": "model.layers.5.self_attn.q_proj",
|
|
@@ -463,9 +463,9 @@
|
|
| 463 |
"success": true
|
| 464 |
},
|
| 465 |
{
|
| 466 |
-
"bits":
|
| 467 |
"fallback": false,
|
| 468 |
-
"group_size":
|
| 469 |
"metadata": {},
|
| 470 |
"method": "affine",
|
| 471 |
"module_path": "model.layers.5.self_attn.v_proj",
|
|
@@ -473,9 +473,9 @@
|
|
| 473 |
"success": true
|
| 474 |
},
|
| 475 |
{
|
| 476 |
-
"bits":
|
| 477 |
"fallback": false,
|
| 478 |
-
"group_size":
|
| 479 |
"metadata": {},
|
| 480 |
"method": "affine",
|
| 481 |
"module_path": "model.layers.5.mlp.experts.down_proj",
|
|
@@ -493,9 +493,9 @@
|
|
| 493 |
"success": true
|
| 494 |
},
|
| 495 |
{
|
| 496 |
-
"bits":
|
| 497 |
"fallback": false,
|
| 498 |
-
"group_size":
|
| 499 |
"metadata": {},
|
| 500 |
"method": "affine",
|
| 501 |
"module_path": "model.layers.6.mlp.experts.down_proj",
|
|
@@ -503,9 +503,9 @@
|
|
| 503 |
"success": true
|
| 504 |
},
|
| 505 |
{
|
| 506 |
-
"bits":
|
| 507 |
"fallback": false,
|
| 508 |
-
"group_size":
|
| 509 |
"metadata": {},
|
| 510 |
"method": "affine",
|
| 511 |
"module_path": "model.layers.6.mlp.experts.gate_proj",
|
|
@@ -513,9 +513,9 @@
|
|
| 513 |
"success": true
|
| 514 |
},
|
| 515 |
{
|
| 516 |
-
"bits":
|
| 517 |
"fallback": false,
|
| 518 |
-
"group_size":
|
| 519 |
"metadata": {},
|
| 520 |
"method": "affine",
|
| 521 |
"module_path": "model.layers.6.mlp.experts.up_proj",
|
|
@@ -533,9 +533,9 @@
|
|
| 533 |
"success": true
|
| 534 |
},
|
| 535 |
{
|
| 536 |
-
"bits":
|
| 537 |
"fallback": false,
|
| 538 |
-
"group_size":
|
| 539 |
"metadata": {},
|
| 540 |
"method": "affine",
|
| 541 |
"module_path": "model.layers.6.self_attn.k_proj",
|
|
@@ -543,9 +543,9 @@
|
|
| 543 |
"success": true
|
| 544 |
},
|
| 545 |
{
|
| 546 |
-
"bits":
|
| 547 |
"fallback": false,
|
| 548 |
-
"group_size":
|
| 549 |
"metadata": {},
|
| 550 |
"method": "affine",
|
| 551 |
"module_path": "model.layers.6.self_attn.o_proj",
|
|
@@ -553,9 +553,9 @@
|
|
| 553 |
"success": true
|
| 554 |
},
|
| 555 |
{
|
| 556 |
-
"bits":
|
| 557 |
"fallback": false,
|
| 558 |
-
"group_size":
|
| 559 |
"metadata": {},
|
| 560 |
"method": "affine",
|
| 561 |
"module_path": "model.layers.6.self_attn.q_proj",
|
|
@@ -563,9 +563,9 @@
|
|
| 563 |
"success": true
|
| 564 |
},
|
| 565 |
{
|
| 566 |
-
"bits":
|
| 567 |
"fallback": false,
|
| 568 |
-
"group_size":
|
| 569 |
"metadata": {},
|
| 570 |
"method": "affine",
|
| 571 |
"module_path": "model.layers.6.self_attn.v_proj",
|
|
@@ -573,9 +573,9 @@
|
|
| 573 |
"success": true
|
| 574 |
},
|
| 575 |
{
|
| 576 |
-
"bits":
|
| 577 |
"fallback": false,
|
| 578 |
-
"group_size":
|
| 579 |
"metadata": {},
|
| 580 |
"method": "affine",
|
| 581 |
"module_path": "model.layers.7.mlp.experts.down_proj",
|
|
@@ -583,9 +583,9 @@
|
|
| 583 |
"success": true
|
| 584 |
},
|
| 585 |
{
|
| 586 |
-
"bits":
|
| 587 |
"fallback": false,
|
| 588 |
-
"group_size":
|
| 589 |
"metadata": {},
|
| 590 |
"method": "affine",
|
| 591 |
"module_path": "model.layers.7.mlp.experts.gate_proj",
|
|
@@ -593,9 +593,9 @@
|
|
| 593 |
"success": true
|
| 594 |
},
|
| 595 |
{
|
| 596 |
-
"bits":
|
| 597 |
"fallback": false,
|
| 598 |
-
"group_size":
|
| 599 |
"metadata": {},
|
| 600 |
"method": "affine",
|
| 601 |
"module_path": "model.layers.7.mlp.experts.up_proj",
|
|
@@ -613,9 +613,9 @@
|
|
| 613 |
"success": true
|
| 614 |
},
|
| 615 |
{
|
| 616 |
-
"bits":
|
| 617 |
"fallback": false,
|
| 618 |
-
"group_size":
|
| 619 |
"metadata": {},
|
| 620 |
"method": "affine",
|
| 621 |
"module_path": "model.layers.7.self_attn.k_proj",
|
|
@@ -623,9 +623,9 @@
|
|
| 623 |
"success": true
|
| 624 |
},
|
| 625 |
{
|
| 626 |
-
"bits":
|
| 627 |
"fallback": false,
|
| 628 |
-
"group_size":
|
| 629 |
"metadata": {},
|
| 630 |
"method": "affine",
|
| 631 |
"module_path": "model.layers.7.self_attn.o_proj",
|
|
@@ -633,9 +633,9 @@
|
|
| 633 |
"success": true
|
| 634 |
},
|
| 635 |
{
|
| 636 |
-
"bits":
|
| 637 |
"fallback": false,
|
| 638 |
-
"group_size":
|
| 639 |
"metadata": {},
|
| 640 |
"method": "affine",
|
| 641 |
"module_path": "model.layers.7.self_attn.q_proj",
|
|
@@ -643,9 +643,9 @@
|
|
| 643 |
"success": true
|
| 644 |
},
|
| 645 |
{
|
| 646 |
-
"bits":
|
| 647 |
"fallback": false,
|
| 648 |
-
"group_size":
|
| 649 |
"metadata": {},
|
| 650 |
"method": "affine",
|
| 651 |
"module_path": "model.layers.7.self_attn.v_proj",
|
|
@@ -653,9 +653,9 @@
|
|
| 653 |
"success": true
|
| 654 |
},
|
| 655 |
{
|
| 656 |
-
"bits":
|
| 657 |
"fallback": false,
|
| 658 |
-
"group_size":
|
| 659 |
"metadata": {},
|
| 660 |
"method": "affine",
|
| 661 |
"module_path": "model.layers.8.mlp.experts.gate_proj",
|
|
@@ -663,9 +663,9 @@
|
|
| 663 |
"success": true
|
| 664 |
},
|
| 665 |
{
|
| 666 |
-
"bits":
|
| 667 |
"fallback": false,
|
| 668 |
-
"group_size":
|
| 669 |
"metadata": {},
|
| 670 |
"method": "affine",
|
| 671 |
"module_path": "model.layers.8.mlp.experts.up_proj",
|
|
@@ -673,9 +673,9 @@
|
|
| 673 |
"success": true
|
| 674 |
},
|
| 675 |
{
|
| 676 |
-
"bits":
|
| 677 |
"fallback": false,
|
| 678 |
-
"group_size":
|
| 679 |
"metadata": {},
|
| 680 |
"method": "affine",
|
| 681 |
"module_path": "model.layers.8.self_attn.k_proj",
|
|
@@ -683,9 +683,9 @@
|
|
| 683 |
"success": true
|
| 684 |
},
|
| 685 |
{
|
| 686 |
-
"bits":
|
| 687 |
"fallback": false,
|
| 688 |
-
"group_size":
|
| 689 |
"metadata": {},
|
| 690 |
"method": "affine",
|
| 691 |
"module_path": "model.layers.8.self_attn.o_proj",
|
|
@@ -693,9 +693,9 @@
|
|
| 693 |
"success": true
|
| 694 |
},
|
| 695 |
{
|
| 696 |
-
"bits":
|
| 697 |
"fallback": false,
|
| 698 |
-
"group_size":
|
| 699 |
"metadata": {},
|
| 700 |
"method": "affine",
|
| 701 |
"module_path": "model.layers.8.self_attn.q_proj",
|
|
@@ -703,9 +703,9 @@
|
|
| 703 |
"success": true
|
| 704 |
},
|
| 705 |
{
|
| 706 |
-
"bits":
|
| 707 |
"fallback": false,
|
| 708 |
-
"group_size":
|
| 709 |
"metadata": {},
|
| 710 |
"method": "affine",
|
| 711 |
"module_path": "model.layers.8.self_attn.v_proj",
|
|
@@ -713,9 +713,9 @@
|
|
| 713 |
"success": true
|
| 714 |
},
|
| 715 |
{
|
| 716 |
-
"bits":
|
| 717 |
"fallback": false,
|
| 718 |
-
"group_size":
|
| 719 |
"metadata": {},
|
| 720 |
"method": "affine",
|
| 721 |
"module_path": "model.layers.10.mlp.experts.down_proj",
|
|
@@ -723,9 +723,9 @@
|
|
| 723 |
"success": true
|
| 724 |
},
|
| 725 |
{
|
| 726 |
-
"bits":
|
| 727 |
"fallback": false,
|
| 728 |
-
"group_size":
|
| 729 |
"metadata": {},
|
| 730 |
"method": "affine",
|
| 731 |
"module_path": "model.layers.10.mlp.experts.gate_proj",
|
|
@@ -733,9 +733,9 @@
|
|
| 733 |
"success": true
|
| 734 |
},
|
| 735 |
{
|
| 736 |
-
"bits":
|
| 737 |
"fallback": false,
|
| 738 |
-
"group_size":
|
| 739 |
"metadata": {},
|
| 740 |
"method": "affine",
|
| 741 |
"module_path": "model.layers.10.mlp.experts.up_proj",
|
|
@@ -753,9 +753,9 @@
|
|
| 753 |
"success": true
|
| 754 |
},
|
| 755 |
{
|
| 756 |
-
"bits":
|
| 757 |
"fallback": false,
|
| 758 |
-
"group_size":
|
| 759 |
"metadata": {},
|
| 760 |
"method": "affine",
|
| 761 |
"module_path": "model.layers.10.self_attn.k_proj",
|
|
@@ -763,9 +763,9 @@
|
|
| 763 |
"success": true
|
| 764 |
},
|
| 765 |
{
|
| 766 |
-
"bits":
|
| 767 |
"fallback": false,
|
| 768 |
-
"group_size":
|
| 769 |
"metadata": {},
|
| 770 |
"method": "affine",
|
| 771 |
"module_path": "model.layers.10.self_attn.o_proj",
|
|
@@ -773,9 +773,9 @@
|
|
| 773 |
"success": true
|
| 774 |
},
|
| 775 |
{
|
| 776 |
-
"bits":
|
| 777 |
"fallback": false,
|
| 778 |
-
"group_size":
|
| 779 |
"metadata": {},
|
| 780 |
"method": "affine",
|
| 781 |
"module_path": "model.layers.10.self_attn.q_proj",
|
|
@@ -783,9 +783,9 @@
|
|
| 783 |
"success": true
|
| 784 |
},
|
| 785 |
{
|
| 786 |
-
"bits":
|
| 787 |
"fallback": false,
|
| 788 |
-
"group_size":
|
| 789 |
"metadata": {},
|
| 790 |
"method": "affine",
|
| 791 |
"module_path": "model.layers.10.self_attn.v_proj",
|
|
@@ -793,9 +793,9 @@
|
|
| 793 |
"success": true
|
| 794 |
},
|
| 795 |
{
|
| 796 |
-
"bits":
|
| 797 |
"fallback": false,
|
| 798 |
-
"group_size":
|
| 799 |
"metadata": {},
|
| 800 |
"method": "affine",
|
| 801 |
"module_path": "model.layers.11.mlp.experts.gate_proj",
|
|
@@ -803,9 +803,9 @@
|
|
| 803 |
"success": true
|
| 804 |
},
|
| 805 |
{
|
| 806 |
-
"bits":
|
| 807 |
"fallback": false,
|
| 808 |
-
"group_size":
|
| 809 |
"metadata": {},
|
| 810 |
"method": "affine",
|
| 811 |
"module_path": "model.layers.11.mlp.experts.up_proj",
|
|
@@ -813,9 +813,9 @@
|
|
| 813 |
"success": true
|
| 814 |
},
|
| 815 |
{
|
| 816 |
-
"bits":
|
| 817 |
"fallback": false,
|
| 818 |
-
"group_size":
|
| 819 |
"metadata": {},
|
| 820 |
"method": "affine",
|
| 821 |
"module_path": "model.layers.11.self_attn.k_proj",
|
|
@@ -823,9 +823,9 @@
|
|
| 823 |
"success": true
|
| 824 |
},
|
| 825 |
{
|
| 826 |
-
"bits":
|
| 827 |
"fallback": false,
|
| 828 |
-
"group_size":
|
| 829 |
"metadata": {},
|
| 830 |
"method": "affine",
|
| 831 |
"module_path": "model.layers.11.self_attn.o_proj",
|
|
@@ -833,9 +833,9 @@
|
|
| 833 |
"success": true
|
| 834 |
},
|
| 835 |
{
|
| 836 |
-
"bits":
|
| 837 |
"fallback": false,
|
| 838 |
-
"group_size":
|
| 839 |
"metadata": {},
|
| 840 |
"method": "affine",
|
| 841 |
"module_path": "model.layers.11.self_attn.q_proj",
|
|
@@ -843,9 +843,9 @@
|
|
| 843 |
"success": true
|
| 844 |
},
|
| 845 |
{
|
| 846 |
-
"bits":
|
| 847 |
"fallback": false,
|
| 848 |
-
"group_size":
|
| 849 |
"metadata": {},
|
| 850 |
"method": "affine",
|
| 851 |
"module_path": "model.layers.11.self_attn.v_proj",
|
|
@@ -853,9 +853,9 @@
|
|
| 853 |
"success": true
|
| 854 |
},
|
| 855 |
{
|
| 856 |
-
"bits":
|
| 857 |
"fallback": false,
|
| 858 |
-
"group_size":
|
| 859 |
"metadata": {},
|
| 860 |
"method": "affine",
|
| 861 |
"module_path": "model.layers.8.mlp.experts.down_proj",
|
|
@@ -873,9 +873,9 @@
|
|
| 873 |
"success": true
|
| 874 |
},
|
| 875 |
{
|
| 876 |
-
"bits":
|
| 877 |
"fallback": false,
|
| 878 |
-
"group_size":
|
| 879 |
"metadata": {},
|
| 880 |
"method": "affine",
|
| 881 |
"module_path": "model.layers.9.mlp.experts.down_proj",
|
|
@@ -883,9 +883,9 @@
|
|
| 883 |
"success": true
|
| 884 |
},
|
| 885 |
{
|
| 886 |
-
"bits":
|
| 887 |
"fallback": false,
|
| 888 |
-
"group_size":
|
| 889 |
"metadata": {},
|
| 890 |
"method": "affine",
|
| 891 |
"module_path": "model.layers.9.mlp.experts.gate_proj",
|
|
@@ -893,9 +893,9 @@
|
|
| 893 |
"success": true
|
| 894 |
},
|
| 895 |
{
|
| 896 |
-
"bits":
|
| 897 |
"fallback": false,
|
| 898 |
-
"group_size":
|
| 899 |
"metadata": {},
|
| 900 |
"method": "affine",
|
| 901 |
"module_path": "model.layers.9.mlp.experts.up_proj",
|
|
@@ -913,9 +913,9 @@
|
|
| 913 |
"success": true
|
| 914 |
},
|
| 915 |
{
|
| 916 |
-
"bits":
|
| 917 |
"fallback": false,
|
| 918 |
-
"group_size":
|
| 919 |
"metadata": {},
|
| 920 |
"method": "affine",
|
| 921 |
"module_path": "model.layers.9.self_attn.k_proj",
|
|
@@ -923,9 +923,9 @@
|
|
| 923 |
"success": true
|
| 924 |
},
|
| 925 |
{
|
| 926 |
-
"bits":
|
| 927 |
"fallback": false,
|
| 928 |
-
"group_size":
|
| 929 |
"metadata": {},
|
| 930 |
"method": "affine",
|
| 931 |
"module_path": "model.layers.9.self_attn.o_proj",
|
|
@@ -933,9 +933,9 @@
|
|
| 933 |
"success": true
|
| 934 |
},
|
| 935 |
{
|
| 936 |
-
"bits":
|
| 937 |
"fallback": false,
|
| 938 |
-
"group_size":
|
| 939 |
"metadata": {},
|
| 940 |
"method": "affine",
|
| 941 |
"module_path": "model.layers.9.self_attn.q_proj",
|
|
@@ -953,9 +953,9 @@
|
|
| 953 |
"success": true
|
| 954 |
},
|
| 955 |
{
|
| 956 |
-
"bits":
|
| 957 |
"fallback": false,
|
| 958 |
-
"group_size":
|
| 959 |
"metadata": {},
|
| 960 |
"method": "affine",
|
| 961 |
"module_path": "model.layers.11.mlp.experts.down_proj",
|
|
@@ -973,9 +973,9 @@
|
|
| 973 |
"success": true
|
| 974 |
},
|
| 975 |
{
|
| 976 |
-
"bits":
|
| 977 |
"fallback": false,
|
| 978 |
-
"group_size":
|
| 979 |
"metadata": {},
|
| 980 |
"method": "affine",
|
| 981 |
"module_path": "model.layers.12.mlp.experts.down_proj",
|
|
@@ -983,9 +983,9 @@
|
|
| 983 |
"success": true
|
| 984 |
},
|
| 985 |
{
|
| 986 |
-
"bits":
|
| 987 |
"fallback": false,
|
| 988 |
-
"group_size":
|
| 989 |
"metadata": {},
|
| 990 |
"method": "affine",
|
| 991 |
"module_path": "model.layers.12.mlp.experts.gate_proj",
|
|
@@ -993,9 +993,9 @@
|
|
| 993 |
"success": true
|
| 994 |
},
|
| 995 |
{
|
| 996 |
-
"bits":
|
| 997 |
"fallback": false,
|
| 998 |
-
"group_size":
|
| 999 |
"metadata": {},
|
| 1000 |
"method": "affine",
|
| 1001 |
"module_path": "model.layers.12.mlp.experts.up_proj",
|
|
@@ -1023,9 +1023,9 @@
|
|
| 1023 |
"success": true
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
-
"bits":
|
| 1027 |
"fallback": false,
|
| 1028 |
-
"group_size":
|
| 1029 |
"metadata": {},
|
| 1030 |
"method": "affine",
|
| 1031 |
"module_path": "model.layers.12.self_attn.o_proj",
|
|
@@ -1033,9 +1033,9 @@
|
|
| 1033 |
"success": true
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
-
"bits":
|
| 1037 |
"fallback": false,
|
| 1038 |
-
"group_size":
|
| 1039 |
"metadata": {},
|
| 1040 |
"method": "affine",
|
| 1041 |
"module_path": "model.layers.12.self_attn.q_proj",
|
|
@@ -1053,9 +1053,9 @@
|
|
| 1053 |
"success": true
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
-
"bits":
|
| 1057 |
"fallback": false,
|
| 1058 |
-
"group_size":
|
| 1059 |
"metadata": {},
|
| 1060 |
"method": "affine",
|
| 1061 |
"module_path": "model.layers.13.mlp.experts.down_proj",
|
|
@@ -1063,9 +1063,9 @@
|
|
| 1063 |
"success": true
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
-
"bits":
|
| 1067 |
"fallback": false,
|
| 1068 |
-
"group_size":
|
| 1069 |
"metadata": {},
|
| 1070 |
"method": "affine",
|
| 1071 |
"module_path": "model.layers.13.mlp.experts.gate_proj",
|
|
@@ -1073,9 +1073,9 @@
|
|
| 1073 |
"success": true
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
-
"bits":
|
| 1077 |
"fallback": false,
|
| 1078 |
-
"group_size":
|
| 1079 |
"metadata": {},
|
| 1080 |
"method": "affine",
|
| 1081 |
"module_path": "model.layers.13.mlp.experts.up_proj",
|
|
@@ -2205,7 +2205,7 @@
|
|
| 2205 |
{
|
| 2206 |
"bits": 8,
|
| 2207 |
"fallback": false,
|
| 2208 |
-
"group_size":
|
| 2209 |
"metadata": {},
|
| 2210 |
"method": "affine",
|
| 2211 |
"module_path": "model.layers.27.mlp.router",
|
|
@@ -2285,7 +2285,7 @@
|
|
| 2285 |
{
|
| 2286 |
"bits": 8,
|
| 2287 |
"fallback": false,
|
| 2288 |
-
"group_size":
|
| 2289 |
"metadata": {},
|
| 2290 |
"method": "affine",
|
| 2291 |
"module_path": "model.layers.28.mlp.router",
|
|
@@ -2405,7 +2405,7 @@
|
|
| 2405 |
{
|
| 2406 |
"bits": 8,
|
| 2407 |
"fallback": false,
|
| 2408 |
-
"group_size":
|
| 2409 |
"metadata": {},
|
| 2410 |
"method": "affine",
|
| 2411 |
"module_path": "model.layers.29.mlp.router",
|
|
@@ -2445,7 +2445,7 @@
|
|
| 2445 |
{
|
| 2446 |
"bits": 8,
|
| 2447 |
"fallback": false,
|
| 2448 |
-
"group_size":
|
| 2449 |
"metadata": {},
|
| 2450 |
"method": "affine",
|
| 2451 |
"module_path": "model.layers.30.mlp.router",
|
|
@@ -2525,7 +2525,7 @@
|
|
| 2525 |
{
|
| 2526 |
"bits": 8,
|
| 2527 |
"fallback": false,
|
| 2528 |
-
"group_size":
|
| 2529 |
"metadata": {},
|
| 2530 |
"method": "affine",
|
| 2531 |
"module_path": "model.layers.31.mlp.router",
|
|
@@ -2645,7 +2645,7 @@
|
|
| 2645 |
{
|
| 2646 |
"bits": 8,
|
| 2647 |
"fallback": false,
|
| 2648 |
-
"group_size":
|
| 2649 |
"metadata": {},
|
| 2650 |
"method": "affine",
|
| 2651 |
"module_path": "model.layers.32.mlp.router",
|
|
@@ -2685,7 +2685,7 @@
|
|
| 2685 |
{
|
| 2686 |
"bits": 8,
|
| 2687 |
"fallback": false,
|
| 2688 |
-
"group_size":
|
| 2689 |
"metadata": {},
|
| 2690 |
"method": "affine",
|
| 2691 |
"module_path": "model.layers.33.mlp.router",
|
|
@@ -2765,7 +2765,7 @@
|
|
| 2765 |
{
|
| 2766 |
"bits": 8,
|
| 2767 |
"fallback": false,
|
| 2768 |
-
"group_size":
|
| 2769 |
"metadata": {},
|
| 2770 |
"method": "affine",
|
| 2771 |
"module_path": "model.layers.34.mlp.router",
|
|
@@ -2885,7 +2885,7 @@
|
|
| 2885 |
{
|
| 2886 |
"bits": 8,
|
| 2887 |
"fallback": false,
|
| 2888 |
-
"group_size":
|
| 2889 |
"metadata": {},
|
| 2890 |
"method": "affine",
|
| 2891 |
"module_path": "model.layers.35.mlp.router",
|
|
|
|
| 1 |
{
|
| 2 |
+
"created_at": "2026-08-11T02:33:57.164554Z",
|
| 3 |
+
"plan_sha256": "4c680c40943529cae51fa0adc80420aaf5ada6e00bb5cca9a85ed2e4b88ffde1",
|
| 4 |
"records": [
|
| 5 |
{
|
| 6 |
"bits": 8,
|
|
|
|
| 13 |
"success": true
|
| 14 |
},
|
| 15 |
{
|
| 16 |
+
"bits": 6,
|
| 17 |
"fallback": false,
|
| 18 |
+
"group_size": 64,
|
| 19 |
"metadata": {},
|
| 20 |
"method": "affine",
|
| 21 |
"module_path": "model.layers.0.mlp.experts.down_proj",
|
|
|
|
| 23 |
"success": true
|
| 24 |
},
|
| 25 |
{
|
| 26 |
+
"bits": 6,
|
| 27 |
"fallback": false,
|
| 28 |
+
"group_size": 64,
|
| 29 |
"metadata": {},
|
| 30 |
"method": "affine",
|
| 31 |
"module_path": "model.layers.0.mlp.experts.gate_proj",
|
|
|
|
| 33 |
"success": true
|
| 34 |
},
|
| 35 |
{
|
| 36 |
+
"bits": 6,
|
| 37 |
"fallback": false,
|
| 38 |
+
"group_size": 64,
|
| 39 |
"metadata": {},
|
| 40 |
"method": "affine",
|
| 41 |
"module_path": "model.layers.0.mlp.experts.up_proj",
|
|
|
|
| 53 |
"success": true
|
| 54 |
},
|
| 55 |
{
|
| 56 |
+
"bits": 8,
|
| 57 |
"fallback": false,
|
| 58 |
+
"group_size": 64,
|
| 59 |
"metadata": {},
|
| 60 |
"method": "affine",
|
| 61 |
"module_path": "model.layers.0.self_attn.k_proj",
|
|
|
|
| 63 |
"success": true
|
| 64 |
},
|
| 65 |
{
|
| 66 |
+
"bits": 8,
|
| 67 |
"fallback": false,
|
| 68 |
+
"group_size": 64,
|
| 69 |
"metadata": {},
|
| 70 |
"method": "affine",
|
| 71 |
"module_path": "model.layers.0.self_attn.o_proj",
|
|
|
|
| 73 |
"success": true
|
| 74 |
},
|
| 75 |
{
|
| 76 |
+
"bits": 8,
|
| 77 |
"fallback": false,
|
| 78 |
+
"group_size": 64,
|
| 79 |
"metadata": {},
|
| 80 |
"method": "affine",
|
| 81 |
"module_path": "model.layers.0.self_attn.q_proj",
|
|
|
|
| 83 |
"success": true
|
| 84 |
},
|
| 85 |
{
|
| 86 |
+
"bits": 8,
|
| 87 |
"fallback": false,
|
| 88 |
+
"group_size": 64,
|
| 89 |
"metadata": {},
|
| 90 |
"method": "affine",
|
| 91 |
"module_path": "model.layers.0.self_attn.v_proj",
|
|
|
|
| 93 |
"success": true
|
| 94 |
},
|
| 95 |
{
|
| 96 |
+
"bits": 6,
|
| 97 |
"fallback": false,
|
| 98 |
+
"group_size": 64,
|
| 99 |
"metadata": {},
|
| 100 |
"method": "affine",
|
| 101 |
"module_path": "model.layers.1.mlp.experts.down_proj",
|
|
|
|
| 103 |
"success": true
|
| 104 |
},
|
| 105 |
{
|
| 106 |
+
"bits": 6,
|
| 107 |
"fallback": false,
|
| 108 |
+
"group_size": 64,
|
| 109 |
"metadata": {},
|
| 110 |
"method": "affine",
|
| 111 |
"module_path": "model.layers.1.mlp.experts.gate_proj",
|
|
|
|
| 113 |
"success": true
|
| 114 |
},
|
| 115 |
{
|
| 116 |
+
"bits": 6,
|
| 117 |
"fallback": false,
|
| 118 |
+
"group_size": 64,
|
| 119 |
"metadata": {},
|
| 120 |
"method": "affine",
|
| 121 |
"module_path": "model.layers.1.mlp.experts.up_proj",
|
|
|
|
| 133 |
"success": true
|
| 134 |
},
|
| 135 |
{
|
| 136 |
+
"bits": 8,
|
| 137 |
"fallback": false,
|
| 138 |
+
"group_size": 64,
|
| 139 |
"metadata": {},
|
| 140 |
"method": "affine",
|
| 141 |
"module_path": "model.layers.1.self_attn.k_proj",
|
|
|
|
| 143 |
"success": true
|
| 144 |
},
|
| 145 |
{
|
| 146 |
+
"bits": 8,
|
| 147 |
"fallback": false,
|
| 148 |
+
"group_size": 64,
|
| 149 |
"metadata": {},
|
| 150 |
"method": "affine",
|
| 151 |
"module_path": "model.layers.1.self_attn.o_proj",
|
|
|
|
| 153 |
"success": true
|
| 154 |
},
|
| 155 |
{
|
| 156 |
+
"bits": 8,
|
| 157 |
"fallback": false,
|
| 158 |
+
"group_size": 64,
|
| 159 |
"metadata": {},
|
| 160 |
"method": "affine",
|
| 161 |
"module_path": "model.layers.1.self_attn.q_proj",
|
|
|
|
| 163 |
"success": true
|
| 164 |
},
|
| 165 |
{
|
| 166 |
+
"bits": 8,
|
| 167 |
"fallback": false,
|
| 168 |
+
"group_size": 64,
|
| 169 |
"metadata": {},
|
| 170 |
"method": "affine",
|
| 171 |
"module_path": "model.layers.1.self_attn.v_proj",
|
|
|
|
| 173 |
"success": true
|
| 174 |
},
|
| 175 |
{
|
| 176 |
+
"bits": 6,
|
| 177 |
"fallback": false,
|
| 178 |
+
"group_size": 64,
|
| 179 |
"metadata": {},
|
| 180 |
"method": "affine",
|
| 181 |
"module_path": "model.layers.2.mlp.experts.gate_proj",
|
|
|
|
| 183 |
"success": true
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"bits": 6,
|
| 187 |
"fallback": false,
|
| 188 |
+
"group_size": 64,
|
| 189 |
"metadata": {},
|
| 190 |
"method": "affine",
|
| 191 |
"module_path": "model.layers.2.mlp.experts.up_proj",
|
|
|
|
| 193 |
"success": true
|
| 194 |
},
|
| 195 |
{
|
| 196 |
+
"bits": 8,
|
| 197 |
"fallback": false,
|
| 198 |
+
"group_size": 64,
|
| 199 |
"metadata": {},
|
| 200 |
"method": "affine",
|
| 201 |
"module_path": "model.layers.2.self_attn.k_proj",
|
|
|
|
| 203 |
"success": true
|
| 204 |
},
|
| 205 |
{
|
| 206 |
+
"bits": 8,
|
| 207 |
"fallback": false,
|
| 208 |
+
"group_size": 64,
|
| 209 |
"metadata": {},
|
| 210 |
"method": "affine",
|
| 211 |
"module_path": "model.layers.2.self_attn.o_proj",
|
|
|
|
| 213 |
"success": true
|
| 214 |
},
|
| 215 |
{
|
| 216 |
+
"bits": 8,
|
| 217 |
"fallback": false,
|
| 218 |
+
"group_size": 64,
|
| 219 |
"metadata": {},
|
| 220 |
"method": "affine",
|
| 221 |
"module_path": "model.layers.2.self_attn.q_proj",
|
|
|
|
| 223 |
"success": true
|
| 224 |
},
|
| 225 |
{
|
| 226 |
+
"bits": 8,
|
| 227 |
"fallback": false,
|
| 228 |
+
"group_size": 64,
|
| 229 |
"metadata": {},
|
| 230 |
"method": "affine",
|
| 231 |
"module_path": "model.layers.2.self_attn.v_proj",
|
|
|
|
| 233 |
"success": true
|
| 234 |
},
|
| 235 |
{
|
| 236 |
+
"bits": 6,
|
| 237 |
"fallback": false,
|
| 238 |
+
"group_size": 64,
|
| 239 |
"metadata": {},
|
| 240 |
"method": "affine",
|
| 241 |
"module_path": "model.layers.2.mlp.experts.down_proj",
|
|
|
|
| 253 |
"success": true
|
| 254 |
},
|
| 255 |
{
|
| 256 |
+
"bits": 6,
|
| 257 |
"fallback": false,
|
| 258 |
+
"group_size": 64,
|
| 259 |
"metadata": {},
|
| 260 |
"method": "affine",
|
| 261 |
"module_path": "model.layers.3.mlp.experts.down_proj",
|
|
|
|
| 263 |
"success": true
|
| 264 |
},
|
| 265 |
{
|
| 266 |
+
"bits": 6,
|
| 267 |
"fallback": false,
|
| 268 |
+
"group_size": 64,
|
| 269 |
"metadata": {},
|
| 270 |
"method": "affine",
|
| 271 |
"module_path": "model.layers.3.mlp.experts.gate_proj",
|
|
|
|
| 273 |
"success": true
|
| 274 |
},
|
| 275 |
{
|
| 276 |
+
"bits": 6,
|
| 277 |
"fallback": false,
|
| 278 |
+
"group_size": 64,
|
| 279 |
"metadata": {},
|
| 280 |
"method": "affine",
|
| 281 |
"module_path": "model.layers.3.mlp.experts.up_proj",
|
|
|
|
| 293 |
"success": true
|
| 294 |
},
|
| 295 |
{
|
| 296 |
+
"bits": 8,
|
| 297 |
"fallback": false,
|
| 298 |
+
"group_size": 64,
|
| 299 |
"metadata": {},
|
| 300 |
"method": "affine",
|
| 301 |
"module_path": "model.layers.3.self_attn.k_proj",
|
|
|
|
| 303 |
"success": true
|
| 304 |
},
|
| 305 |
{
|
| 306 |
+
"bits": 8,
|
| 307 |
"fallback": false,
|
| 308 |
+
"group_size": 64,
|
| 309 |
"metadata": {},
|
| 310 |
"method": "affine",
|
| 311 |
"module_path": "model.layers.3.self_attn.o_proj",
|
|
|
|
| 313 |
"success": true
|
| 314 |
},
|
| 315 |
{
|
| 316 |
+
"bits": 8,
|
| 317 |
"fallback": false,
|
| 318 |
+
"group_size": 64,
|
| 319 |
"metadata": {},
|
| 320 |
"method": "affine",
|
| 321 |
"module_path": "model.layers.3.self_attn.q_proj",
|
|
|
|
| 323 |
"success": true
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"bits": 8,
|
| 327 |
"fallback": false,
|
| 328 |
+
"group_size": 64,
|
| 329 |
"metadata": {},
|
| 330 |
"method": "affine",
|
| 331 |
"module_path": "model.layers.3.self_attn.v_proj",
|
|
|
|
| 333 |
"success": true
|
| 334 |
},
|
| 335 |
{
|
| 336 |
+
"bits": 6,
|
| 337 |
"fallback": false,
|
| 338 |
+
"group_size": 64,
|
| 339 |
"metadata": {},
|
| 340 |
"method": "affine",
|
| 341 |
"module_path": "model.layers.4.mlp.experts.down_proj",
|
|
|
|
| 343 |
"success": true
|
| 344 |
},
|
| 345 |
{
|
| 346 |
+
"bits": 6,
|
| 347 |
"fallback": false,
|
| 348 |
+
"group_size": 64,
|
| 349 |
"metadata": {},
|
| 350 |
"method": "affine",
|
| 351 |
"module_path": "model.layers.4.mlp.experts.gate_proj",
|
|
|
|
| 353 |
"success": true
|
| 354 |
},
|
| 355 |
{
|
| 356 |
+
"bits": 6,
|
| 357 |
"fallback": false,
|
| 358 |
+
"group_size": 64,
|
| 359 |
"metadata": {},
|
| 360 |
"method": "affine",
|
| 361 |
"module_path": "model.layers.4.mlp.experts.up_proj",
|
|
|
|
| 373 |
"success": true
|
| 374 |
},
|
| 375 |
{
|
| 376 |
+
"bits": 8,
|
| 377 |
"fallback": false,
|
| 378 |
+
"group_size": 64,
|
| 379 |
"metadata": {},
|
| 380 |
"method": "affine",
|
| 381 |
"module_path": "model.layers.4.self_attn.k_proj",
|
|
|
|
| 383 |
"success": true
|
| 384 |
},
|
| 385 |
{
|
| 386 |
+
"bits": 8,
|
| 387 |
"fallback": false,
|
| 388 |
+
"group_size": 64,
|
| 389 |
"metadata": {},
|
| 390 |
"method": "affine",
|
| 391 |
"module_path": "model.layers.4.self_attn.o_proj",
|
|
|
|
| 393 |
"success": true
|
| 394 |
},
|
| 395 |
{
|
| 396 |
+
"bits": 8,
|
| 397 |
"fallback": false,
|
| 398 |
+
"group_size": 64,
|
| 399 |
"metadata": {},
|
| 400 |
"method": "affine",
|
| 401 |
"module_path": "model.layers.4.self_attn.q_proj",
|
|
|
|
| 403 |
"success": true
|
| 404 |
},
|
| 405 |
{
|
| 406 |
+
"bits": 8,
|
| 407 |
"fallback": false,
|
| 408 |
+
"group_size": 64,
|
| 409 |
"metadata": {},
|
| 410 |
"method": "affine",
|
| 411 |
"module_path": "model.layers.4.self_attn.v_proj",
|
|
|
|
| 413 |
"success": true
|
| 414 |
},
|
| 415 |
{
|
| 416 |
+
"bits": 6,
|
| 417 |
"fallback": false,
|
| 418 |
+
"group_size": 64,
|
| 419 |
"metadata": {},
|
| 420 |
"method": "affine",
|
| 421 |
"module_path": "model.layers.5.mlp.experts.gate_proj",
|
|
|
|
| 423 |
"success": true
|
| 424 |
},
|
| 425 |
{
|
| 426 |
+
"bits": 6,
|
| 427 |
"fallback": false,
|
| 428 |
+
"group_size": 64,
|
| 429 |
"metadata": {},
|
| 430 |
"method": "affine",
|
| 431 |
"module_path": "model.layers.5.mlp.experts.up_proj",
|
|
|
|
| 433 |
"success": true
|
| 434 |
},
|
| 435 |
{
|
| 436 |
+
"bits": 8,
|
| 437 |
"fallback": false,
|
| 438 |
+
"group_size": 64,
|
| 439 |
"metadata": {},
|
| 440 |
"method": "affine",
|
| 441 |
"module_path": "model.layers.5.self_attn.k_proj",
|
|
|
|
| 443 |
"success": true
|
| 444 |
},
|
| 445 |
{
|
| 446 |
+
"bits": 8,
|
| 447 |
"fallback": false,
|
| 448 |
+
"group_size": 64,
|
| 449 |
"metadata": {},
|
| 450 |
"method": "affine",
|
| 451 |
"module_path": "model.layers.5.self_attn.o_proj",
|
|
|
|
| 453 |
"success": true
|
| 454 |
},
|
| 455 |
{
|
| 456 |
+
"bits": 8,
|
| 457 |
"fallback": false,
|
| 458 |
+
"group_size": 64,
|
| 459 |
"metadata": {},
|
| 460 |
"method": "affine",
|
| 461 |
"module_path": "model.layers.5.self_attn.q_proj",
|
|
|
|
| 463 |
"success": true
|
| 464 |
},
|
| 465 |
{
|
| 466 |
+
"bits": 8,
|
| 467 |
"fallback": false,
|
| 468 |
+
"group_size": 64,
|
| 469 |
"metadata": {},
|
| 470 |
"method": "affine",
|
| 471 |
"module_path": "model.layers.5.self_attn.v_proj",
|
|
|
|
| 473 |
"success": true
|
| 474 |
},
|
| 475 |
{
|
| 476 |
+
"bits": 6,
|
| 477 |
"fallback": false,
|
| 478 |
+
"group_size": 64,
|
| 479 |
"metadata": {},
|
| 480 |
"method": "affine",
|
| 481 |
"module_path": "model.layers.5.mlp.experts.down_proj",
|
|
|
|
| 493 |
"success": true
|
| 494 |
},
|
| 495 |
{
|
| 496 |
+
"bits": 6,
|
| 497 |
"fallback": false,
|
| 498 |
+
"group_size": 64,
|
| 499 |
"metadata": {},
|
| 500 |
"method": "affine",
|
| 501 |
"module_path": "model.layers.6.mlp.experts.down_proj",
|
|
|
|
| 503 |
"success": true
|
| 504 |
},
|
| 505 |
{
|
| 506 |
+
"bits": 6,
|
| 507 |
"fallback": false,
|
| 508 |
+
"group_size": 64,
|
| 509 |
"metadata": {},
|
| 510 |
"method": "affine",
|
| 511 |
"module_path": "model.layers.6.mlp.experts.gate_proj",
|
|
|
|
| 513 |
"success": true
|
| 514 |
},
|
| 515 |
{
|
| 516 |
+
"bits": 6,
|
| 517 |
"fallback": false,
|
| 518 |
+
"group_size": 64,
|
| 519 |
"metadata": {},
|
| 520 |
"method": "affine",
|
| 521 |
"module_path": "model.layers.6.mlp.experts.up_proj",
|
|
|
|
| 533 |
"success": true
|
| 534 |
},
|
| 535 |
{
|
| 536 |
+
"bits": 8,
|
| 537 |
"fallback": false,
|
| 538 |
+
"group_size": 64,
|
| 539 |
"metadata": {},
|
| 540 |
"method": "affine",
|
| 541 |
"module_path": "model.layers.6.self_attn.k_proj",
|
|
|
|
| 543 |
"success": true
|
| 544 |
},
|
| 545 |
{
|
| 546 |
+
"bits": 8,
|
| 547 |
"fallback": false,
|
| 548 |
+
"group_size": 64,
|
| 549 |
"metadata": {},
|
| 550 |
"method": "affine",
|
| 551 |
"module_path": "model.layers.6.self_attn.o_proj",
|
|
|
|
| 553 |
"success": true
|
| 554 |
},
|
| 555 |
{
|
| 556 |
+
"bits": 8,
|
| 557 |
"fallback": false,
|
| 558 |
+
"group_size": 64,
|
| 559 |
"metadata": {},
|
| 560 |
"method": "affine",
|
| 561 |
"module_path": "model.layers.6.self_attn.q_proj",
|
|
|
|
| 563 |
"success": true
|
| 564 |
},
|
| 565 |
{
|
| 566 |
+
"bits": 8,
|
| 567 |
"fallback": false,
|
| 568 |
+
"group_size": 64,
|
| 569 |
"metadata": {},
|
| 570 |
"method": "affine",
|
| 571 |
"module_path": "model.layers.6.self_attn.v_proj",
|
|
|
|
| 573 |
"success": true
|
| 574 |
},
|
| 575 |
{
|
| 576 |
+
"bits": 6,
|
| 577 |
"fallback": false,
|
| 578 |
+
"group_size": 64,
|
| 579 |
"metadata": {},
|
| 580 |
"method": "affine",
|
| 581 |
"module_path": "model.layers.7.mlp.experts.down_proj",
|
|
|
|
| 583 |
"success": true
|
| 584 |
},
|
| 585 |
{
|
| 586 |
+
"bits": 6,
|
| 587 |
"fallback": false,
|
| 588 |
+
"group_size": 64,
|
| 589 |
"metadata": {},
|
| 590 |
"method": "affine",
|
| 591 |
"module_path": "model.layers.7.mlp.experts.gate_proj",
|
|
|
|
| 593 |
"success": true
|
| 594 |
},
|
| 595 |
{
|
| 596 |
+
"bits": 6,
|
| 597 |
"fallback": false,
|
| 598 |
+
"group_size": 64,
|
| 599 |
"metadata": {},
|
| 600 |
"method": "affine",
|
| 601 |
"module_path": "model.layers.7.mlp.experts.up_proj",
|
|
|
|
| 613 |
"success": true
|
| 614 |
},
|
| 615 |
{
|
| 616 |
+
"bits": 8,
|
| 617 |
"fallback": false,
|
| 618 |
+
"group_size": 64,
|
| 619 |
"metadata": {},
|
| 620 |
"method": "affine",
|
| 621 |
"module_path": "model.layers.7.self_attn.k_proj",
|
|
|
|
| 623 |
"success": true
|
| 624 |
},
|
| 625 |
{
|
| 626 |
+
"bits": 8,
|
| 627 |
"fallback": false,
|
| 628 |
+
"group_size": 64,
|
| 629 |
"metadata": {},
|
| 630 |
"method": "affine",
|
| 631 |
"module_path": "model.layers.7.self_attn.o_proj",
|
|
|
|
| 633 |
"success": true
|
| 634 |
},
|
| 635 |
{
|
| 636 |
+
"bits": 8,
|
| 637 |
"fallback": false,
|
| 638 |
+
"group_size": 64,
|
| 639 |
"metadata": {},
|
| 640 |
"method": "affine",
|
| 641 |
"module_path": "model.layers.7.self_attn.q_proj",
|
|
|
|
| 643 |
"success": true
|
| 644 |
},
|
| 645 |
{
|
| 646 |
+
"bits": 8,
|
| 647 |
"fallback": false,
|
| 648 |
+
"group_size": 64,
|
| 649 |
"metadata": {},
|
| 650 |
"method": "affine",
|
| 651 |
"module_path": "model.layers.7.self_attn.v_proj",
|
|
|
|
| 653 |
"success": true
|
| 654 |
},
|
| 655 |
{
|
| 656 |
+
"bits": 6,
|
| 657 |
"fallback": false,
|
| 658 |
+
"group_size": 64,
|
| 659 |
"metadata": {},
|
| 660 |
"method": "affine",
|
| 661 |
"module_path": "model.layers.8.mlp.experts.gate_proj",
|
|
|
|
| 663 |
"success": true
|
| 664 |
},
|
| 665 |
{
|
| 666 |
+
"bits": 6,
|
| 667 |
"fallback": false,
|
| 668 |
+
"group_size": 64,
|
| 669 |
"metadata": {},
|
| 670 |
"method": "affine",
|
| 671 |
"module_path": "model.layers.8.mlp.experts.up_proj",
|
|
|
|
| 673 |
"success": true
|
| 674 |
},
|
| 675 |
{
|
| 676 |
+
"bits": 8,
|
| 677 |
"fallback": false,
|
| 678 |
+
"group_size": 64,
|
| 679 |
"metadata": {},
|
| 680 |
"method": "affine",
|
| 681 |
"module_path": "model.layers.8.self_attn.k_proj",
|
|
|
|
| 683 |
"success": true
|
| 684 |
},
|
| 685 |
{
|
| 686 |
+
"bits": 8,
|
| 687 |
"fallback": false,
|
| 688 |
+
"group_size": 64,
|
| 689 |
"metadata": {},
|
| 690 |
"method": "affine",
|
| 691 |
"module_path": "model.layers.8.self_attn.o_proj",
|
|
|
|
| 693 |
"success": true
|
| 694 |
},
|
| 695 |
{
|
| 696 |
+
"bits": 8,
|
| 697 |
"fallback": false,
|
| 698 |
+
"group_size": 64,
|
| 699 |
"metadata": {},
|
| 700 |
"method": "affine",
|
| 701 |
"module_path": "model.layers.8.self_attn.q_proj",
|
|
|
|
| 703 |
"success": true
|
| 704 |
},
|
| 705 |
{
|
| 706 |
+
"bits": 8,
|
| 707 |
"fallback": false,
|
| 708 |
+
"group_size": 64,
|
| 709 |
"metadata": {},
|
| 710 |
"method": "affine",
|
| 711 |
"module_path": "model.layers.8.self_attn.v_proj",
|
|
|
|
| 713 |
"success": true
|
| 714 |
},
|
| 715 |
{
|
| 716 |
+
"bits": 6,
|
| 717 |
"fallback": false,
|
| 718 |
+
"group_size": 64,
|
| 719 |
"metadata": {},
|
| 720 |
"method": "affine",
|
| 721 |
"module_path": "model.layers.10.mlp.experts.down_proj",
|
|
|
|
| 723 |
"success": true
|
| 724 |
},
|
| 725 |
{
|
| 726 |
+
"bits": 6,
|
| 727 |
"fallback": false,
|
| 728 |
+
"group_size": 64,
|
| 729 |
"metadata": {},
|
| 730 |
"method": "affine",
|
| 731 |
"module_path": "model.layers.10.mlp.experts.gate_proj",
|
|
|
|
| 733 |
"success": true
|
| 734 |
},
|
| 735 |
{
|
| 736 |
+
"bits": 6,
|
| 737 |
"fallback": false,
|
| 738 |
+
"group_size": 64,
|
| 739 |
"metadata": {},
|
| 740 |
"method": "affine",
|
| 741 |
"module_path": "model.layers.10.mlp.experts.up_proj",
|
|
|
|
| 753 |
"success": true
|
| 754 |
},
|
| 755 |
{
|
| 756 |
+
"bits": 8,
|
| 757 |
"fallback": false,
|
| 758 |
+
"group_size": 64,
|
| 759 |
"metadata": {},
|
| 760 |
"method": "affine",
|
| 761 |
"module_path": "model.layers.10.self_attn.k_proj",
|
|
|
|
| 763 |
"success": true
|
| 764 |
},
|
| 765 |
{
|
| 766 |
+
"bits": 8,
|
| 767 |
"fallback": false,
|
| 768 |
+
"group_size": 64,
|
| 769 |
"metadata": {},
|
| 770 |
"method": "affine",
|
| 771 |
"module_path": "model.layers.10.self_attn.o_proj",
|
|
|
|
| 773 |
"success": true
|
| 774 |
},
|
| 775 |
{
|
| 776 |
+
"bits": 8,
|
| 777 |
"fallback": false,
|
| 778 |
+
"group_size": 64,
|
| 779 |
"metadata": {},
|
| 780 |
"method": "affine",
|
| 781 |
"module_path": "model.layers.10.self_attn.q_proj",
|
|
|
|
| 783 |
"success": true
|
| 784 |
},
|
| 785 |
{
|
| 786 |
+
"bits": 8,
|
| 787 |
"fallback": false,
|
| 788 |
+
"group_size": 64,
|
| 789 |
"metadata": {},
|
| 790 |
"method": "affine",
|
| 791 |
"module_path": "model.layers.10.self_attn.v_proj",
|
|
|
|
| 793 |
"success": true
|
| 794 |
},
|
| 795 |
{
|
| 796 |
+
"bits": 6,
|
| 797 |
"fallback": false,
|
| 798 |
+
"group_size": 64,
|
| 799 |
"metadata": {},
|
| 800 |
"method": "affine",
|
| 801 |
"module_path": "model.layers.11.mlp.experts.gate_proj",
|
|
|
|
| 803 |
"success": true
|
| 804 |
},
|
| 805 |
{
|
| 806 |
+
"bits": 6,
|
| 807 |
"fallback": false,
|
| 808 |
+
"group_size": 64,
|
| 809 |
"metadata": {},
|
| 810 |
"method": "affine",
|
| 811 |
"module_path": "model.layers.11.mlp.experts.up_proj",
|
|
|
|
| 813 |
"success": true
|
| 814 |
},
|
| 815 |
{
|
| 816 |
+
"bits": 8,
|
| 817 |
"fallback": false,
|
| 818 |
+
"group_size": 64,
|
| 819 |
"metadata": {},
|
| 820 |
"method": "affine",
|
| 821 |
"module_path": "model.layers.11.self_attn.k_proj",
|
|
|
|
| 823 |
"success": true
|
| 824 |
},
|
| 825 |
{
|
| 826 |
+
"bits": 8,
|
| 827 |
"fallback": false,
|
| 828 |
+
"group_size": 64,
|
| 829 |
"metadata": {},
|
| 830 |
"method": "affine",
|
| 831 |
"module_path": "model.layers.11.self_attn.o_proj",
|
|
|
|
| 833 |
"success": true
|
| 834 |
},
|
| 835 |
{
|
| 836 |
+
"bits": 8,
|
| 837 |
"fallback": false,
|
| 838 |
+
"group_size": 64,
|
| 839 |
"metadata": {},
|
| 840 |
"method": "affine",
|
| 841 |
"module_path": "model.layers.11.self_attn.q_proj",
|
|
|
|
| 843 |
"success": true
|
| 844 |
},
|
| 845 |
{
|
| 846 |
+
"bits": 8,
|
| 847 |
"fallback": false,
|
| 848 |
+
"group_size": 64,
|
| 849 |
"metadata": {},
|
| 850 |
"method": "affine",
|
| 851 |
"module_path": "model.layers.11.self_attn.v_proj",
|
|
|
|
| 853 |
"success": true
|
| 854 |
},
|
| 855 |
{
|
| 856 |
+
"bits": 6,
|
| 857 |
"fallback": false,
|
| 858 |
+
"group_size": 64,
|
| 859 |
"metadata": {},
|
| 860 |
"method": "affine",
|
| 861 |
"module_path": "model.layers.8.mlp.experts.down_proj",
|
|
|
|
| 873 |
"success": true
|
| 874 |
},
|
| 875 |
{
|
| 876 |
+
"bits": 6,
|
| 877 |
"fallback": false,
|
| 878 |
+
"group_size": 64,
|
| 879 |
"metadata": {},
|
| 880 |
"method": "affine",
|
| 881 |
"module_path": "model.layers.9.mlp.experts.down_proj",
|
|
|
|
| 883 |
"success": true
|
| 884 |
},
|
| 885 |
{
|
| 886 |
+
"bits": 6,
|
| 887 |
"fallback": false,
|
| 888 |
+
"group_size": 64,
|
| 889 |
"metadata": {},
|
| 890 |
"method": "affine",
|
| 891 |
"module_path": "model.layers.9.mlp.experts.gate_proj",
|
|
|
|
| 893 |
"success": true
|
| 894 |
},
|
| 895 |
{
|
| 896 |
+
"bits": 6,
|
| 897 |
"fallback": false,
|
| 898 |
+
"group_size": 64,
|
| 899 |
"metadata": {},
|
| 900 |
"method": "affine",
|
| 901 |
"module_path": "model.layers.9.mlp.experts.up_proj",
|
|
|
|
| 913 |
"success": true
|
| 914 |
},
|
| 915 |
{
|
| 916 |
+
"bits": 8,
|
| 917 |
"fallback": false,
|
| 918 |
+
"group_size": 64,
|
| 919 |
"metadata": {},
|
| 920 |
"method": "affine",
|
| 921 |
"module_path": "model.layers.9.self_attn.k_proj",
|
|
|
|
| 923 |
"success": true
|
| 924 |
},
|
| 925 |
{
|
| 926 |
+
"bits": 8,
|
| 927 |
"fallback": false,
|
| 928 |
+
"group_size": 64,
|
| 929 |
"metadata": {},
|
| 930 |
"method": "affine",
|
| 931 |
"module_path": "model.layers.9.self_attn.o_proj",
|
|
|
|
| 933 |
"success": true
|
| 934 |
},
|
| 935 |
{
|
| 936 |
+
"bits": 8,
|
| 937 |
"fallback": false,
|
| 938 |
+
"group_size": 64,
|
| 939 |
"metadata": {},
|
| 940 |
"method": "affine",
|
| 941 |
"module_path": "model.layers.9.self_attn.q_proj",
|
|
|
|
| 953 |
"success": true
|
| 954 |
},
|
| 955 |
{
|
| 956 |
+
"bits": 6,
|
| 957 |
"fallback": false,
|
| 958 |
+
"group_size": 64,
|
| 959 |
"metadata": {},
|
| 960 |
"method": "affine",
|
| 961 |
"module_path": "model.layers.11.mlp.experts.down_proj",
|
|
|
|
| 973 |
"success": true
|
| 974 |
},
|
| 975 |
{
|
| 976 |
+
"bits": 6,
|
| 977 |
"fallback": false,
|
| 978 |
+
"group_size": 64,
|
| 979 |
"metadata": {},
|
| 980 |
"method": "affine",
|
| 981 |
"module_path": "model.layers.12.mlp.experts.down_proj",
|
|
|
|
| 983 |
"success": true
|
| 984 |
},
|
| 985 |
{
|
| 986 |
+
"bits": 6,
|
| 987 |
"fallback": false,
|
| 988 |
+
"group_size": 64,
|
| 989 |
"metadata": {},
|
| 990 |
"method": "affine",
|
| 991 |
"module_path": "model.layers.12.mlp.experts.gate_proj",
|
|
|
|
| 993 |
"success": true
|
| 994 |
},
|
| 995 |
{
|
| 996 |
+
"bits": 6,
|
| 997 |
"fallback": false,
|
| 998 |
+
"group_size": 64,
|
| 999 |
"metadata": {},
|
| 1000 |
"method": "affine",
|
| 1001 |
"module_path": "model.layers.12.mlp.experts.up_proj",
|
|
|
|
| 1023 |
"success": true
|
| 1024 |
},
|
| 1025 |
{
|
| 1026 |
+
"bits": 8,
|
| 1027 |
"fallback": false,
|
| 1028 |
+
"group_size": 64,
|
| 1029 |
"metadata": {},
|
| 1030 |
"method": "affine",
|
| 1031 |
"module_path": "model.layers.12.self_attn.o_proj",
|
|
|
|
| 1033 |
"success": true
|
| 1034 |
},
|
| 1035 |
{
|
| 1036 |
+
"bits": 8,
|
| 1037 |
"fallback": false,
|
| 1038 |
+
"group_size": 64,
|
| 1039 |
"metadata": {},
|
| 1040 |
"method": "affine",
|
| 1041 |
"module_path": "model.layers.12.self_attn.q_proj",
|
|
|
|
| 1053 |
"success": true
|
| 1054 |
},
|
| 1055 |
{
|
| 1056 |
+
"bits": 6,
|
| 1057 |
"fallback": false,
|
| 1058 |
+
"group_size": 64,
|
| 1059 |
"metadata": {},
|
| 1060 |
"method": "affine",
|
| 1061 |
"module_path": "model.layers.13.mlp.experts.down_proj",
|
|
|
|
| 1063 |
"success": true
|
| 1064 |
},
|
| 1065 |
{
|
| 1066 |
+
"bits": 6,
|
| 1067 |
"fallback": false,
|
| 1068 |
+
"group_size": 64,
|
| 1069 |
"metadata": {},
|
| 1070 |
"method": "affine",
|
| 1071 |
"module_path": "model.layers.13.mlp.experts.gate_proj",
|
|
|
|
| 1073 |
"success": true
|
| 1074 |
},
|
| 1075 |
{
|
| 1076 |
+
"bits": 6,
|
| 1077 |
"fallback": false,
|
| 1078 |
+
"group_size": 64,
|
| 1079 |
"metadata": {},
|
| 1080 |
"method": "affine",
|
| 1081 |
"module_path": "model.layers.13.mlp.experts.up_proj",
|
|
|
|
| 2205 |
{
|
| 2206 |
"bits": 8,
|
| 2207 |
"fallback": false,
|
| 2208 |
+
"group_size": 64,
|
| 2209 |
"metadata": {},
|
| 2210 |
"method": "affine",
|
| 2211 |
"module_path": "model.layers.27.mlp.router",
|
|
|
|
| 2285 |
{
|
| 2286 |
"bits": 8,
|
| 2287 |
"fallback": false,
|
| 2288 |
+
"group_size": 64,
|
| 2289 |
"metadata": {},
|
| 2290 |
"method": "affine",
|
| 2291 |
"module_path": "model.layers.28.mlp.router",
|
|
|
|
| 2405 |
{
|
| 2406 |
"bits": 8,
|
| 2407 |
"fallback": false,
|
| 2408 |
+
"group_size": 64,
|
| 2409 |
"metadata": {},
|
| 2410 |
"method": "affine",
|
| 2411 |
"module_path": "model.layers.29.mlp.router",
|
|
|
|
| 2445 |
{
|
| 2446 |
"bits": 8,
|
| 2447 |
"fallback": false,
|
| 2448 |
+
"group_size": 64,
|
| 2449 |
"metadata": {},
|
| 2450 |
"method": "affine",
|
| 2451 |
"module_path": "model.layers.30.mlp.router",
|
|
|
|
| 2525 |
{
|
| 2526 |
"bits": 8,
|
| 2527 |
"fallback": false,
|
| 2528 |
+
"group_size": 64,
|
| 2529 |
"metadata": {},
|
| 2530 |
"method": "affine",
|
| 2531 |
"module_path": "model.layers.31.mlp.router",
|
|
|
|
| 2645 |
{
|
| 2646 |
"bits": 8,
|
| 2647 |
"fallback": false,
|
| 2648 |
+
"group_size": 64,
|
| 2649 |
"metadata": {},
|
| 2650 |
"method": "affine",
|
| 2651 |
"module_path": "model.layers.32.mlp.router",
|
|
|
|
| 2685 |
{
|
| 2686 |
"bits": 8,
|
| 2687 |
"fallback": false,
|
| 2688 |
+
"group_size": 64,
|
| 2689 |
"metadata": {},
|
| 2690 |
"method": "affine",
|
| 2691 |
"module_path": "model.layers.33.mlp.router",
|
|
|
|
| 2765 |
{
|
| 2766 |
"bits": 8,
|
| 2767 |
"fallback": false,
|
| 2768 |
+
"group_size": 64,
|
| 2769 |
"metadata": {},
|
| 2770 |
"method": "affine",
|
| 2771 |
"module_path": "model.layers.34.mlp.router",
|
|
|
|
| 2885 |
{
|
| 2886 |
"bits": 8,
|
| 2887 |
"fallback": false,
|
| 2888 |
+
"group_size": 64,
|
| 2889 |
"metadata": {},
|
| 2890 |
"method": "affine",
|
| 2891 |
"module_path": "model.layers.35.mlp.router",
|
axquant_runtime.json
CHANGED
|
@@ -4,7 +4,7 @@
|
|
| 4 |
"fused_mtp": null,
|
| 5 |
"kernel_evidence": "unmeasured",
|
| 6 |
"model_manifest": "model-manifest.json",
|
| 7 |
-
"preferred_group_size":
|
| 8 |
},
|
| 9 |
"compatible_runtimes": [
|
| 10 |
{
|
|
@@ -21,7 +21,7 @@
|
|
| 21 |
"support_level": "standard-inference"
|
| 22 |
}
|
| 23 |
],
|
| 24 |
-
"created_at": "2026-08-
|
| 25 |
"kv_cache": null,
|
| 26 |
"memory_policy": {
|
| 27 |
"kv_cache_precision": "runtime-default",
|
|
|
|
| 4 |
"fused_mtp": null,
|
| 5 |
"kernel_evidence": "unmeasured",
|
| 6 |
"model_manifest": "model-manifest.json",
|
| 7 |
+
"preferred_group_size": 64
|
| 8 |
},
|
| 9 |
"compatible_runtimes": [
|
| 10 |
{
|
|
|
|
| 21 |
"support_level": "standard-inference"
|
| 22 |
}
|
| 23 |
],
|
| 24 |
+
"created_at": "2026-08-11T02:33:57.174300Z",
|
| 25 |
"kv_cache": null,
|
| 26 |
"memory_policy": {
|
| 27 |
"kv_cache_precision": "runtime-default",
|
config.json
CHANGED
|
@@ -64,8 +64,8 @@
|
|
| 64 |
"output_router_logits": false,
|
| 65 |
"pad_token_id": 199999,
|
| 66 |
"quantization": {
|
| 67 |
-
"group_size":
|
| 68 |
-
"bits":
|
| 69 |
"mode": "affine",
|
| 70 |
"model.embed_tokens": {
|
| 71 |
"group_size": 64,
|
|
@@ -73,38 +73,38 @@
|
|
| 73 |
"mode": "affine"
|
| 74 |
},
|
| 75 |
"model.layers.0.self_attn.q_proj": {
|
| 76 |
-
"group_size":
|
| 77 |
-
"bits":
|
| 78 |
"mode": "affine"
|
| 79 |
},
|
| 80 |
"model.layers.0.self_attn.k_proj": {
|
| 81 |
-
"group_size":
|
| 82 |
-
"bits":
|
| 83 |
"mode": "affine"
|
| 84 |
},
|
| 85 |
"model.layers.0.self_attn.v_proj": {
|
| 86 |
-
"group_size":
|
| 87 |
-
"bits":
|
| 88 |
"mode": "affine"
|
| 89 |
},
|
| 90 |
"model.layers.0.self_attn.o_proj": {
|
| 91 |
-
"group_size":
|
| 92 |
-
"bits":
|
| 93 |
"mode": "affine"
|
| 94 |
},
|
| 95 |
"model.layers.0.mlp.experts.gate_proj": {
|
| 96 |
-
"group_size":
|
| 97 |
-
"bits":
|
| 98 |
"mode": "affine"
|
| 99 |
},
|
| 100 |
"model.layers.0.mlp.experts.up_proj": {
|
| 101 |
-
"group_size":
|
| 102 |
-
"bits":
|
| 103 |
"mode": "affine"
|
| 104 |
},
|
| 105 |
"model.layers.0.mlp.experts.down_proj": {
|
| 106 |
-
"group_size":
|
| 107 |
-
"bits":
|
| 108 |
"mode": "affine"
|
| 109 |
},
|
| 110 |
"model.layers.0.mlp.router": {
|
|
@@ -113,38 +113,38 @@
|
|
| 113 |
"mode": "affine"
|
| 114 |
},
|
| 115 |
"model.layers.1.self_attn.q_proj": {
|
| 116 |
-
"group_size":
|
| 117 |
-
"bits":
|
| 118 |
"mode": "affine"
|
| 119 |
},
|
| 120 |
"model.layers.1.self_attn.k_proj": {
|
| 121 |
-
"group_size":
|
| 122 |
-
"bits":
|
| 123 |
"mode": "affine"
|
| 124 |
},
|
| 125 |
"model.layers.1.self_attn.v_proj": {
|
| 126 |
-
"group_size":
|
| 127 |
-
"bits":
|
| 128 |
"mode": "affine"
|
| 129 |
},
|
| 130 |
"model.layers.1.self_attn.o_proj": {
|
| 131 |
-
"group_size":
|
| 132 |
-
"bits":
|
| 133 |
"mode": "affine"
|
| 134 |
},
|
| 135 |
"model.layers.1.mlp.experts.gate_proj": {
|
| 136 |
-
"group_size":
|
| 137 |
-
"bits":
|
| 138 |
"mode": "affine"
|
| 139 |
},
|
| 140 |
"model.layers.1.mlp.experts.up_proj": {
|
| 141 |
-
"group_size":
|
| 142 |
-
"bits":
|
| 143 |
"mode": "affine"
|
| 144 |
},
|
| 145 |
"model.layers.1.mlp.experts.down_proj": {
|
| 146 |
-
"group_size":
|
| 147 |
-
"bits":
|
| 148 |
"mode": "affine"
|
| 149 |
},
|
| 150 |
"model.layers.1.mlp.router": {
|
|
@@ -153,38 +153,38 @@
|
|
| 153 |
"mode": "affine"
|
| 154 |
},
|
| 155 |
"model.layers.2.self_attn.q_proj": {
|
| 156 |
-
"group_size":
|
| 157 |
-
"bits":
|
| 158 |
"mode": "affine"
|
| 159 |
},
|
| 160 |
"model.layers.2.self_attn.k_proj": {
|
| 161 |
-
"group_size":
|
| 162 |
-
"bits":
|
| 163 |
"mode": "affine"
|
| 164 |
},
|
| 165 |
"model.layers.2.self_attn.v_proj": {
|
| 166 |
-
"group_size":
|
| 167 |
-
"bits":
|
| 168 |
"mode": "affine"
|
| 169 |
},
|
| 170 |
"model.layers.2.self_attn.o_proj": {
|
| 171 |
-
"group_size":
|
| 172 |
-
"bits":
|
| 173 |
"mode": "affine"
|
| 174 |
},
|
| 175 |
"model.layers.2.mlp.experts.gate_proj": {
|
| 176 |
-
"group_size":
|
| 177 |
-
"bits":
|
| 178 |
"mode": "affine"
|
| 179 |
},
|
| 180 |
"model.layers.2.mlp.experts.up_proj": {
|
| 181 |
-
"group_size":
|
| 182 |
-
"bits":
|
| 183 |
"mode": "affine"
|
| 184 |
},
|
| 185 |
"model.layers.2.mlp.experts.down_proj": {
|
| 186 |
-
"group_size":
|
| 187 |
-
"bits":
|
| 188 |
"mode": "affine"
|
| 189 |
},
|
| 190 |
"model.layers.2.mlp.router": {
|
|
@@ -193,38 +193,38 @@
|
|
| 193 |
"mode": "affine"
|
| 194 |
},
|
| 195 |
"model.layers.3.self_attn.q_proj": {
|
| 196 |
-
"group_size":
|
| 197 |
-
"bits":
|
| 198 |
"mode": "affine"
|
| 199 |
},
|
| 200 |
"model.layers.3.self_attn.k_proj": {
|
| 201 |
-
"group_size":
|
| 202 |
-
"bits":
|
| 203 |
"mode": "affine"
|
| 204 |
},
|
| 205 |
"model.layers.3.self_attn.v_proj": {
|
| 206 |
-
"group_size":
|
| 207 |
-
"bits":
|
| 208 |
"mode": "affine"
|
| 209 |
},
|
| 210 |
"model.layers.3.self_attn.o_proj": {
|
| 211 |
-
"group_size":
|
| 212 |
-
"bits":
|
| 213 |
"mode": "affine"
|
| 214 |
},
|
| 215 |
"model.layers.3.mlp.experts.gate_proj": {
|
| 216 |
-
"group_size":
|
| 217 |
-
"bits":
|
| 218 |
"mode": "affine"
|
| 219 |
},
|
| 220 |
"model.layers.3.mlp.experts.up_proj": {
|
| 221 |
-
"group_size":
|
| 222 |
-
"bits":
|
| 223 |
"mode": "affine"
|
| 224 |
},
|
| 225 |
"model.layers.3.mlp.experts.down_proj": {
|
| 226 |
-
"group_size":
|
| 227 |
-
"bits":
|
| 228 |
"mode": "affine"
|
| 229 |
},
|
| 230 |
"model.layers.3.mlp.router": {
|
|
@@ -233,38 +233,38 @@
|
|
| 233 |
"mode": "affine"
|
| 234 |
},
|
| 235 |
"model.layers.4.self_attn.q_proj": {
|
| 236 |
-
"group_size":
|
| 237 |
-
"bits":
|
| 238 |
"mode": "affine"
|
| 239 |
},
|
| 240 |
"model.layers.4.self_attn.k_proj": {
|
| 241 |
-
"group_size":
|
| 242 |
-
"bits":
|
| 243 |
"mode": "affine"
|
| 244 |
},
|
| 245 |
"model.layers.4.self_attn.v_proj": {
|
| 246 |
-
"group_size":
|
| 247 |
-
"bits":
|
| 248 |
"mode": "affine"
|
| 249 |
},
|
| 250 |
"model.layers.4.self_attn.o_proj": {
|
| 251 |
-
"group_size":
|
| 252 |
-
"bits":
|
| 253 |
"mode": "affine"
|
| 254 |
},
|
| 255 |
"model.layers.4.mlp.experts.gate_proj": {
|
| 256 |
-
"group_size":
|
| 257 |
-
"bits":
|
| 258 |
"mode": "affine"
|
| 259 |
},
|
| 260 |
"model.layers.4.mlp.experts.up_proj": {
|
| 261 |
-
"group_size":
|
| 262 |
-
"bits":
|
| 263 |
"mode": "affine"
|
| 264 |
},
|
| 265 |
"model.layers.4.mlp.experts.down_proj": {
|
| 266 |
-
"group_size":
|
| 267 |
-
"bits":
|
| 268 |
"mode": "affine"
|
| 269 |
},
|
| 270 |
"model.layers.4.mlp.router": {
|
|
@@ -273,38 +273,38 @@
|
|
| 273 |
"mode": "affine"
|
| 274 |
},
|
| 275 |
"model.layers.5.self_attn.q_proj": {
|
| 276 |
-
"group_size":
|
| 277 |
-
"bits":
|
| 278 |
"mode": "affine"
|
| 279 |
},
|
| 280 |
"model.layers.5.self_attn.k_proj": {
|
| 281 |
-
"group_size":
|
| 282 |
-
"bits":
|
| 283 |
"mode": "affine"
|
| 284 |
},
|
| 285 |
"model.layers.5.self_attn.v_proj": {
|
| 286 |
-
"group_size":
|
| 287 |
-
"bits":
|
| 288 |
"mode": "affine"
|
| 289 |
},
|
| 290 |
"model.layers.5.self_attn.o_proj": {
|
| 291 |
-
"group_size":
|
| 292 |
-
"bits":
|
| 293 |
"mode": "affine"
|
| 294 |
},
|
| 295 |
"model.layers.5.mlp.experts.gate_proj": {
|
| 296 |
-
"group_size":
|
| 297 |
-
"bits":
|
| 298 |
"mode": "affine"
|
| 299 |
},
|
| 300 |
"model.layers.5.mlp.experts.up_proj": {
|
| 301 |
-
"group_size":
|
| 302 |
-
"bits":
|
| 303 |
"mode": "affine"
|
| 304 |
},
|
| 305 |
"model.layers.5.mlp.experts.down_proj": {
|
| 306 |
-
"group_size":
|
| 307 |
-
"bits":
|
| 308 |
"mode": "affine"
|
| 309 |
},
|
| 310 |
"model.layers.5.mlp.router": {
|
|
@@ -313,38 +313,38 @@
|
|
| 313 |
"mode": "affine"
|
| 314 |
},
|
| 315 |
"model.layers.6.self_attn.q_proj": {
|
| 316 |
-
"group_size":
|
| 317 |
-
"bits":
|
| 318 |
"mode": "affine"
|
| 319 |
},
|
| 320 |
"model.layers.6.self_attn.k_proj": {
|
| 321 |
-
"group_size":
|
| 322 |
-
"bits":
|
| 323 |
"mode": "affine"
|
| 324 |
},
|
| 325 |
"model.layers.6.self_attn.v_proj": {
|
| 326 |
-
"group_size":
|
| 327 |
-
"bits":
|
| 328 |
"mode": "affine"
|
| 329 |
},
|
| 330 |
"model.layers.6.self_attn.o_proj": {
|
| 331 |
-
"group_size":
|
| 332 |
-
"bits":
|
| 333 |
"mode": "affine"
|
| 334 |
},
|
| 335 |
"model.layers.6.mlp.experts.gate_proj": {
|
| 336 |
-
"group_size":
|
| 337 |
-
"bits":
|
| 338 |
"mode": "affine"
|
| 339 |
},
|
| 340 |
"model.layers.6.mlp.experts.up_proj": {
|
| 341 |
-
"group_size":
|
| 342 |
-
"bits":
|
| 343 |
"mode": "affine"
|
| 344 |
},
|
| 345 |
"model.layers.6.mlp.experts.down_proj": {
|
| 346 |
-
"group_size":
|
| 347 |
-
"bits":
|
| 348 |
"mode": "affine"
|
| 349 |
},
|
| 350 |
"model.layers.6.mlp.router": {
|
|
@@ -353,38 +353,38 @@
|
|
| 353 |
"mode": "affine"
|
| 354 |
},
|
| 355 |
"model.layers.7.self_attn.q_proj": {
|
| 356 |
-
"group_size":
|
| 357 |
-
"bits":
|
| 358 |
"mode": "affine"
|
| 359 |
},
|
| 360 |
"model.layers.7.self_attn.k_proj": {
|
| 361 |
-
"group_size":
|
| 362 |
-
"bits":
|
| 363 |
"mode": "affine"
|
| 364 |
},
|
| 365 |
"model.layers.7.self_attn.v_proj": {
|
| 366 |
-
"group_size":
|
| 367 |
-
"bits":
|
| 368 |
"mode": "affine"
|
| 369 |
},
|
| 370 |
"model.layers.7.self_attn.o_proj": {
|
| 371 |
-
"group_size":
|
| 372 |
-
"bits":
|
| 373 |
"mode": "affine"
|
| 374 |
},
|
| 375 |
"model.layers.7.mlp.experts.gate_proj": {
|
| 376 |
-
"group_size":
|
| 377 |
-
"bits":
|
| 378 |
"mode": "affine"
|
| 379 |
},
|
| 380 |
"model.layers.7.mlp.experts.up_proj": {
|
| 381 |
-
"group_size":
|
| 382 |
-
"bits":
|
| 383 |
"mode": "affine"
|
| 384 |
},
|
| 385 |
"model.layers.7.mlp.experts.down_proj": {
|
| 386 |
-
"group_size":
|
| 387 |
-
"bits":
|
| 388 |
"mode": "affine"
|
| 389 |
},
|
| 390 |
"model.layers.7.mlp.router": {
|
|
@@ -393,38 +393,38 @@
|
|
| 393 |
"mode": "affine"
|
| 394 |
},
|
| 395 |
"model.layers.8.self_attn.q_proj": {
|
| 396 |
-
"group_size":
|
| 397 |
-
"bits":
|
| 398 |
"mode": "affine"
|
| 399 |
},
|
| 400 |
"model.layers.8.self_attn.k_proj": {
|
| 401 |
-
"group_size":
|
| 402 |
-
"bits":
|
| 403 |
"mode": "affine"
|
| 404 |
},
|
| 405 |
"model.layers.8.self_attn.v_proj": {
|
| 406 |
-
"group_size":
|
| 407 |
-
"bits":
|
| 408 |
"mode": "affine"
|
| 409 |
},
|
| 410 |
"model.layers.8.self_attn.o_proj": {
|
| 411 |
-
"group_size":
|
| 412 |
-
"bits":
|
| 413 |
"mode": "affine"
|
| 414 |
},
|
| 415 |
"model.layers.8.mlp.experts.gate_proj": {
|
| 416 |
-
"group_size":
|
| 417 |
-
"bits":
|
| 418 |
"mode": "affine"
|
| 419 |
},
|
| 420 |
"model.layers.8.mlp.experts.up_proj": {
|
| 421 |
-
"group_size":
|
| 422 |
-
"bits":
|
| 423 |
"mode": "affine"
|
| 424 |
},
|
| 425 |
"model.layers.8.mlp.experts.down_proj": {
|
| 426 |
-
"group_size":
|
| 427 |
-
"bits":
|
| 428 |
"mode": "affine"
|
| 429 |
},
|
| 430 |
"model.layers.8.mlp.router": {
|
|
@@ -433,13 +433,13 @@
|
|
| 433 |
"mode": "affine"
|
| 434 |
},
|
| 435 |
"model.layers.9.self_attn.q_proj": {
|
| 436 |
-
"group_size":
|
| 437 |
-
"bits":
|
| 438 |
"mode": "affine"
|
| 439 |
},
|
| 440 |
"model.layers.9.self_attn.k_proj": {
|
| 441 |
-
"group_size":
|
| 442 |
-
"bits":
|
| 443 |
"mode": "affine"
|
| 444 |
},
|
| 445 |
"model.layers.9.self_attn.v_proj": {
|
|
@@ -448,23 +448,23 @@
|
|
| 448 |
"mode": "affine"
|
| 449 |
},
|
| 450 |
"model.layers.9.self_attn.o_proj": {
|
| 451 |
-
"group_size":
|
| 452 |
-
"bits":
|
| 453 |
"mode": "affine"
|
| 454 |
},
|
| 455 |
"model.layers.9.mlp.experts.gate_proj": {
|
| 456 |
-
"group_size":
|
| 457 |
-
"bits":
|
| 458 |
"mode": "affine"
|
| 459 |
},
|
| 460 |
"model.layers.9.mlp.experts.up_proj": {
|
| 461 |
-
"group_size":
|
| 462 |
-
"bits":
|
| 463 |
"mode": "affine"
|
| 464 |
},
|
| 465 |
"model.layers.9.mlp.experts.down_proj": {
|
| 466 |
-
"group_size":
|
| 467 |
-
"bits":
|
| 468 |
"mode": "affine"
|
| 469 |
},
|
| 470 |
"model.layers.9.mlp.router": {
|
|
@@ -473,38 +473,38 @@
|
|
| 473 |
"mode": "affine"
|
| 474 |
},
|
| 475 |
"model.layers.10.self_attn.q_proj": {
|
| 476 |
-
"group_size":
|
| 477 |
-
"bits":
|
| 478 |
"mode": "affine"
|
| 479 |
},
|
| 480 |
"model.layers.10.self_attn.k_proj": {
|
| 481 |
-
"group_size":
|
| 482 |
-
"bits":
|
| 483 |
"mode": "affine"
|
| 484 |
},
|
| 485 |
"model.layers.10.self_attn.v_proj": {
|
| 486 |
-
"group_size":
|
| 487 |
-
"bits":
|
| 488 |
"mode": "affine"
|
| 489 |
},
|
| 490 |
"model.layers.10.self_attn.o_proj": {
|
| 491 |
-
"group_size":
|
| 492 |
-
"bits":
|
| 493 |
"mode": "affine"
|
| 494 |
},
|
| 495 |
"model.layers.10.mlp.experts.gate_proj": {
|
| 496 |
-
"group_size":
|
| 497 |
-
"bits":
|
| 498 |
"mode": "affine"
|
| 499 |
},
|
| 500 |
"model.layers.10.mlp.experts.up_proj": {
|
| 501 |
-
"group_size":
|
| 502 |
-
"bits":
|
| 503 |
"mode": "affine"
|
| 504 |
},
|
| 505 |
"model.layers.10.mlp.experts.down_proj": {
|
| 506 |
-
"group_size":
|
| 507 |
-
"bits":
|
| 508 |
"mode": "affine"
|
| 509 |
},
|
| 510 |
"model.layers.10.mlp.router": {
|
|
@@ -513,38 +513,38 @@
|
|
| 513 |
"mode": "affine"
|
| 514 |
},
|
| 515 |
"model.layers.11.self_attn.q_proj": {
|
| 516 |
-
"group_size":
|
| 517 |
-
"bits":
|
| 518 |
"mode": "affine"
|
| 519 |
},
|
| 520 |
"model.layers.11.self_attn.k_proj": {
|
| 521 |
-
"group_size":
|
| 522 |
-
"bits":
|
| 523 |
"mode": "affine"
|
| 524 |
},
|
| 525 |
"model.layers.11.self_attn.v_proj": {
|
| 526 |
-
"group_size":
|
| 527 |
-
"bits":
|
| 528 |
"mode": "affine"
|
| 529 |
},
|
| 530 |
"model.layers.11.self_attn.o_proj": {
|
| 531 |
-
"group_size":
|
| 532 |
-
"bits":
|
| 533 |
"mode": "affine"
|
| 534 |
},
|
| 535 |
"model.layers.11.mlp.experts.gate_proj": {
|
| 536 |
-
"group_size":
|
| 537 |
-
"bits":
|
| 538 |
"mode": "affine"
|
| 539 |
},
|
| 540 |
"model.layers.11.mlp.experts.up_proj": {
|
| 541 |
-
"group_size":
|
| 542 |
-
"bits":
|
| 543 |
"mode": "affine"
|
| 544 |
},
|
| 545 |
"model.layers.11.mlp.experts.down_proj": {
|
| 546 |
-
"group_size":
|
| 547 |
-
"bits":
|
| 548 |
"mode": "affine"
|
| 549 |
},
|
| 550 |
"model.layers.11.mlp.router": {
|
|
@@ -553,8 +553,8 @@
|
|
| 553 |
"mode": "affine"
|
| 554 |
},
|
| 555 |
"model.layers.12.self_attn.q_proj": {
|
| 556 |
-
"group_size":
|
| 557 |
-
"bits":
|
| 558 |
"mode": "affine"
|
| 559 |
},
|
| 560 |
"model.layers.12.self_attn.k_proj": {
|
|
@@ -568,23 +568,23 @@
|
|
| 568 |
"mode": "affine"
|
| 569 |
},
|
| 570 |
"model.layers.12.self_attn.o_proj": {
|
| 571 |
-
"group_size":
|
| 572 |
-
"bits":
|
| 573 |
"mode": "affine"
|
| 574 |
},
|
| 575 |
"model.layers.12.mlp.experts.gate_proj": {
|
| 576 |
-
"group_size":
|
| 577 |
-
"bits":
|
| 578 |
"mode": "affine"
|
| 579 |
},
|
| 580 |
"model.layers.12.mlp.experts.up_proj": {
|
| 581 |
-
"group_size":
|
| 582 |
-
"bits":
|
| 583 |
"mode": "affine"
|
| 584 |
},
|
| 585 |
"model.layers.12.mlp.experts.down_proj": {
|
| 586 |
-
"group_size":
|
| 587 |
-
"bits":
|
| 588 |
"mode": "affine"
|
| 589 |
},
|
| 590 |
"model.layers.12.mlp.router": {
|
|
@@ -613,18 +613,18 @@
|
|
| 613 |
"mode": "affine"
|
| 614 |
},
|
| 615 |
"model.layers.13.mlp.experts.gate_proj": {
|
| 616 |
-
"group_size":
|
| 617 |
-
"bits":
|
| 618 |
"mode": "affine"
|
| 619 |
},
|
| 620 |
"model.layers.13.mlp.experts.up_proj": {
|
| 621 |
-
"group_size":
|
| 622 |
-
"bits":
|
| 623 |
"mode": "affine"
|
| 624 |
},
|
| 625 |
"model.layers.13.mlp.experts.down_proj": {
|
| 626 |
-
"group_size":
|
| 627 |
-
"bits":
|
| 628 |
"mode": "affine"
|
| 629 |
},
|
| 630 |
"model.layers.13.mlp.router": {
|
|
@@ -1188,7 +1188,7 @@
|
|
| 1188 |
"mode": "affine"
|
| 1189 |
},
|
| 1190 |
"model.layers.27.mlp.router": {
|
| 1191 |
-
"group_size":
|
| 1192 |
"bits": 8,
|
| 1193 |
"mode": "affine"
|
| 1194 |
},
|
|
@@ -1228,7 +1228,7 @@
|
|
| 1228 |
"mode": "affine"
|
| 1229 |
},
|
| 1230 |
"model.layers.28.mlp.router": {
|
| 1231 |
-
"group_size":
|
| 1232 |
"bits": 8,
|
| 1233 |
"mode": "affine"
|
| 1234 |
},
|
|
@@ -1268,7 +1268,7 @@
|
|
| 1268 |
"mode": "affine"
|
| 1269 |
},
|
| 1270 |
"model.layers.29.mlp.router": {
|
| 1271 |
-
"group_size":
|
| 1272 |
"bits": 8,
|
| 1273 |
"mode": "affine"
|
| 1274 |
},
|
|
@@ -1308,7 +1308,7 @@
|
|
| 1308 |
"mode": "affine"
|
| 1309 |
},
|
| 1310 |
"model.layers.30.mlp.router": {
|
| 1311 |
-
"group_size":
|
| 1312 |
"bits": 8,
|
| 1313 |
"mode": "affine"
|
| 1314 |
},
|
|
@@ -1348,7 +1348,7 @@
|
|
| 1348 |
"mode": "affine"
|
| 1349 |
},
|
| 1350 |
"model.layers.31.mlp.router": {
|
| 1351 |
-
"group_size":
|
| 1352 |
"bits": 8,
|
| 1353 |
"mode": "affine"
|
| 1354 |
},
|
|
@@ -1388,7 +1388,7 @@
|
|
| 1388 |
"mode": "affine"
|
| 1389 |
},
|
| 1390 |
"model.layers.32.mlp.router": {
|
| 1391 |
-
"group_size":
|
| 1392 |
"bits": 8,
|
| 1393 |
"mode": "affine"
|
| 1394 |
},
|
|
@@ -1428,7 +1428,7 @@
|
|
| 1428 |
"mode": "affine"
|
| 1429 |
},
|
| 1430 |
"model.layers.33.mlp.router": {
|
| 1431 |
-
"group_size":
|
| 1432 |
"bits": 8,
|
| 1433 |
"mode": "affine"
|
| 1434 |
},
|
|
@@ -1468,7 +1468,7 @@
|
|
| 1468 |
"mode": "affine"
|
| 1469 |
},
|
| 1470 |
"model.layers.34.mlp.router": {
|
| 1471 |
-
"group_size":
|
| 1472 |
"bits": 8,
|
| 1473 |
"mode": "affine"
|
| 1474 |
},
|
|
@@ -1508,14 +1508,14 @@
|
|
| 1508 |
"mode": "affine"
|
| 1509 |
},
|
| 1510 |
"model.layers.35.mlp.router": {
|
| 1511 |
-
"group_size":
|
| 1512 |
"bits": 8,
|
| 1513 |
"mode": "affine"
|
| 1514 |
}
|
| 1515 |
},
|
| 1516 |
"quantization_config": {
|
| 1517 |
-
"group_size":
|
| 1518 |
-
"bits":
|
| 1519 |
"mode": "affine",
|
| 1520 |
"model.embed_tokens": {
|
| 1521 |
"group_size": 64,
|
|
@@ -1523,38 +1523,38 @@
|
|
| 1523 |
"mode": "affine"
|
| 1524 |
},
|
| 1525 |
"model.layers.0.self_attn.q_proj": {
|
| 1526 |
-
"group_size":
|
| 1527 |
-
"bits":
|
| 1528 |
"mode": "affine"
|
| 1529 |
},
|
| 1530 |
"model.layers.0.self_attn.k_proj": {
|
| 1531 |
-
"group_size":
|
| 1532 |
-
"bits":
|
| 1533 |
"mode": "affine"
|
| 1534 |
},
|
| 1535 |
"model.layers.0.self_attn.v_proj": {
|
| 1536 |
-
"group_size":
|
| 1537 |
-
"bits":
|
| 1538 |
"mode": "affine"
|
| 1539 |
},
|
| 1540 |
"model.layers.0.self_attn.o_proj": {
|
| 1541 |
-
"group_size":
|
| 1542 |
-
"bits":
|
| 1543 |
"mode": "affine"
|
| 1544 |
},
|
| 1545 |
"model.layers.0.mlp.experts.gate_proj": {
|
| 1546 |
-
"group_size":
|
| 1547 |
-
"bits":
|
| 1548 |
"mode": "affine"
|
| 1549 |
},
|
| 1550 |
"model.layers.0.mlp.experts.up_proj": {
|
| 1551 |
-
"group_size":
|
| 1552 |
-
"bits":
|
| 1553 |
"mode": "affine"
|
| 1554 |
},
|
| 1555 |
"model.layers.0.mlp.experts.down_proj": {
|
| 1556 |
-
"group_size":
|
| 1557 |
-
"bits":
|
| 1558 |
"mode": "affine"
|
| 1559 |
},
|
| 1560 |
"model.layers.0.mlp.router": {
|
|
@@ -1563,38 +1563,38 @@
|
|
| 1563 |
"mode": "affine"
|
| 1564 |
},
|
| 1565 |
"model.layers.1.self_attn.q_proj": {
|
| 1566 |
-
"group_size":
|
| 1567 |
-
"bits":
|
| 1568 |
"mode": "affine"
|
| 1569 |
},
|
| 1570 |
"model.layers.1.self_attn.k_proj": {
|
| 1571 |
-
"group_size":
|
| 1572 |
-
"bits":
|
| 1573 |
"mode": "affine"
|
| 1574 |
},
|
| 1575 |
"model.layers.1.self_attn.v_proj": {
|
| 1576 |
-
"group_size":
|
| 1577 |
-
"bits":
|
| 1578 |
"mode": "affine"
|
| 1579 |
},
|
| 1580 |
"model.layers.1.self_attn.o_proj": {
|
| 1581 |
-
"group_size":
|
| 1582 |
-
"bits":
|
| 1583 |
"mode": "affine"
|
| 1584 |
},
|
| 1585 |
"model.layers.1.mlp.experts.gate_proj": {
|
| 1586 |
-
"group_size":
|
| 1587 |
-
"bits":
|
| 1588 |
"mode": "affine"
|
| 1589 |
},
|
| 1590 |
"model.layers.1.mlp.experts.up_proj": {
|
| 1591 |
-
"group_size":
|
| 1592 |
-
"bits":
|
| 1593 |
"mode": "affine"
|
| 1594 |
},
|
| 1595 |
"model.layers.1.mlp.experts.down_proj": {
|
| 1596 |
-
"group_size":
|
| 1597 |
-
"bits":
|
| 1598 |
"mode": "affine"
|
| 1599 |
},
|
| 1600 |
"model.layers.1.mlp.router": {
|
|
@@ -1603,38 +1603,38 @@
|
|
| 1603 |
"mode": "affine"
|
| 1604 |
},
|
| 1605 |
"model.layers.2.self_attn.q_proj": {
|
| 1606 |
-
"group_size":
|
| 1607 |
-
"bits":
|
| 1608 |
"mode": "affine"
|
| 1609 |
},
|
| 1610 |
"model.layers.2.self_attn.k_proj": {
|
| 1611 |
-
"group_size":
|
| 1612 |
-
"bits":
|
| 1613 |
"mode": "affine"
|
| 1614 |
},
|
| 1615 |
"model.layers.2.self_attn.v_proj": {
|
| 1616 |
-
"group_size":
|
| 1617 |
-
"bits":
|
| 1618 |
"mode": "affine"
|
| 1619 |
},
|
| 1620 |
"model.layers.2.self_attn.o_proj": {
|
| 1621 |
-
"group_size":
|
| 1622 |
-
"bits":
|
| 1623 |
"mode": "affine"
|
| 1624 |
},
|
| 1625 |
"model.layers.2.mlp.experts.gate_proj": {
|
| 1626 |
-
"group_size":
|
| 1627 |
-
"bits":
|
| 1628 |
"mode": "affine"
|
| 1629 |
},
|
| 1630 |
"model.layers.2.mlp.experts.up_proj": {
|
| 1631 |
-
"group_size":
|
| 1632 |
-
"bits":
|
| 1633 |
"mode": "affine"
|
| 1634 |
},
|
| 1635 |
"model.layers.2.mlp.experts.down_proj": {
|
| 1636 |
-
"group_size":
|
| 1637 |
-
"bits":
|
| 1638 |
"mode": "affine"
|
| 1639 |
},
|
| 1640 |
"model.layers.2.mlp.router": {
|
|
@@ -1643,38 +1643,38 @@
|
|
| 1643 |
"mode": "affine"
|
| 1644 |
},
|
| 1645 |
"model.layers.3.self_attn.q_proj": {
|
| 1646 |
-
"group_size":
|
| 1647 |
-
"bits":
|
| 1648 |
"mode": "affine"
|
| 1649 |
},
|
| 1650 |
"model.layers.3.self_attn.k_proj": {
|
| 1651 |
-
"group_size":
|
| 1652 |
-
"bits":
|
| 1653 |
"mode": "affine"
|
| 1654 |
},
|
| 1655 |
"model.layers.3.self_attn.v_proj": {
|
| 1656 |
-
"group_size":
|
| 1657 |
-
"bits":
|
| 1658 |
"mode": "affine"
|
| 1659 |
},
|
| 1660 |
"model.layers.3.self_attn.o_proj": {
|
| 1661 |
-
"group_size":
|
| 1662 |
-
"bits":
|
| 1663 |
"mode": "affine"
|
| 1664 |
},
|
| 1665 |
"model.layers.3.mlp.experts.gate_proj": {
|
| 1666 |
-
"group_size":
|
| 1667 |
-
"bits":
|
| 1668 |
"mode": "affine"
|
| 1669 |
},
|
| 1670 |
"model.layers.3.mlp.experts.up_proj": {
|
| 1671 |
-
"group_size":
|
| 1672 |
-
"bits":
|
| 1673 |
"mode": "affine"
|
| 1674 |
},
|
| 1675 |
"model.layers.3.mlp.experts.down_proj": {
|
| 1676 |
-
"group_size":
|
| 1677 |
-
"bits":
|
| 1678 |
"mode": "affine"
|
| 1679 |
},
|
| 1680 |
"model.layers.3.mlp.router": {
|
|
@@ -1683,38 +1683,38 @@
|
|
| 1683 |
"mode": "affine"
|
| 1684 |
},
|
| 1685 |
"model.layers.4.self_attn.q_proj": {
|
| 1686 |
-
"group_size":
|
| 1687 |
-
"bits":
|
| 1688 |
"mode": "affine"
|
| 1689 |
},
|
| 1690 |
"model.layers.4.self_attn.k_proj": {
|
| 1691 |
-
"group_size":
|
| 1692 |
-
"bits":
|
| 1693 |
"mode": "affine"
|
| 1694 |
},
|
| 1695 |
"model.layers.4.self_attn.v_proj": {
|
| 1696 |
-
"group_size":
|
| 1697 |
-
"bits":
|
| 1698 |
"mode": "affine"
|
| 1699 |
},
|
| 1700 |
"model.layers.4.self_attn.o_proj": {
|
| 1701 |
-
"group_size":
|
| 1702 |
-
"bits":
|
| 1703 |
"mode": "affine"
|
| 1704 |
},
|
| 1705 |
"model.layers.4.mlp.experts.gate_proj": {
|
| 1706 |
-
"group_size":
|
| 1707 |
-
"bits":
|
| 1708 |
"mode": "affine"
|
| 1709 |
},
|
| 1710 |
"model.layers.4.mlp.experts.up_proj": {
|
| 1711 |
-
"group_size":
|
| 1712 |
-
"bits":
|
| 1713 |
"mode": "affine"
|
| 1714 |
},
|
| 1715 |
"model.layers.4.mlp.experts.down_proj": {
|
| 1716 |
-
"group_size":
|
| 1717 |
-
"bits":
|
| 1718 |
"mode": "affine"
|
| 1719 |
},
|
| 1720 |
"model.layers.4.mlp.router": {
|
|
@@ -1723,38 +1723,38 @@
|
|
| 1723 |
"mode": "affine"
|
| 1724 |
},
|
| 1725 |
"model.layers.5.self_attn.q_proj": {
|
| 1726 |
-
"group_size":
|
| 1727 |
-
"bits":
|
| 1728 |
"mode": "affine"
|
| 1729 |
},
|
| 1730 |
"model.layers.5.self_attn.k_proj": {
|
| 1731 |
-
"group_size":
|
| 1732 |
-
"bits":
|
| 1733 |
"mode": "affine"
|
| 1734 |
},
|
| 1735 |
"model.layers.5.self_attn.v_proj": {
|
| 1736 |
-
"group_size":
|
| 1737 |
-
"bits":
|
| 1738 |
"mode": "affine"
|
| 1739 |
},
|
| 1740 |
"model.layers.5.self_attn.o_proj": {
|
| 1741 |
-
"group_size":
|
| 1742 |
-
"bits":
|
| 1743 |
"mode": "affine"
|
| 1744 |
},
|
| 1745 |
"model.layers.5.mlp.experts.gate_proj": {
|
| 1746 |
-
"group_size":
|
| 1747 |
-
"bits":
|
| 1748 |
"mode": "affine"
|
| 1749 |
},
|
| 1750 |
"model.layers.5.mlp.experts.up_proj": {
|
| 1751 |
-
"group_size":
|
| 1752 |
-
"bits":
|
| 1753 |
"mode": "affine"
|
| 1754 |
},
|
| 1755 |
"model.layers.5.mlp.experts.down_proj": {
|
| 1756 |
-
"group_size":
|
| 1757 |
-
"bits":
|
| 1758 |
"mode": "affine"
|
| 1759 |
},
|
| 1760 |
"model.layers.5.mlp.router": {
|
|
@@ -1763,38 +1763,38 @@
|
|
| 1763 |
"mode": "affine"
|
| 1764 |
},
|
| 1765 |
"model.layers.6.self_attn.q_proj": {
|
| 1766 |
-
"group_size":
|
| 1767 |
-
"bits":
|
| 1768 |
"mode": "affine"
|
| 1769 |
},
|
| 1770 |
"model.layers.6.self_attn.k_proj": {
|
| 1771 |
-
"group_size":
|
| 1772 |
-
"bits":
|
| 1773 |
"mode": "affine"
|
| 1774 |
},
|
| 1775 |
"model.layers.6.self_attn.v_proj": {
|
| 1776 |
-
"group_size":
|
| 1777 |
-
"bits":
|
| 1778 |
"mode": "affine"
|
| 1779 |
},
|
| 1780 |
"model.layers.6.self_attn.o_proj": {
|
| 1781 |
-
"group_size":
|
| 1782 |
-
"bits":
|
| 1783 |
"mode": "affine"
|
| 1784 |
},
|
| 1785 |
"model.layers.6.mlp.experts.gate_proj": {
|
| 1786 |
-
"group_size":
|
| 1787 |
-
"bits":
|
| 1788 |
"mode": "affine"
|
| 1789 |
},
|
| 1790 |
"model.layers.6.mlp.experts.up_proj": {
|
| 1791 |
-
"group_size":
|
| 1792 |
-
"bits":
|
| 1793 |
"mode": "affine"
|
| 1794 |
},
|
| 1795 |
"model.layers.6.mlp.experts.down_proj": {
|
| 1796 |
-
"group_size":
|
| 1797 |
-
"bits":
|
| 1798 |
"mode": "affine"
|
| 1799 |
},
|
| 1800 |
"model.layers.6.mlp.router": {
|
|
@@ -1803,38 +1803,38 @@
|
|
| 1803 |
"mode": "affine"
|
| 1804 |
},
|
| 1805 |
"model.layers.7.self_attn.q_proj": {
|
| 1806 |
-
"group_size":
|
| 1807 |
-
"bits":
|
| 1808 |
"mode": "affine"
|
| 1809 |
},
|
| 1810 |
"model.layers.7.self_attn.k_proj": {
|
| 1811 |
-
"group_size":
|
| 1812 |
-
"bits":
|
| 1813 |
"mode": "affine"
|
| 1814 |
},
|
| 1815 |
"model.layers.7.self_attn.v_proj": {
|
| 1816 |
-
"group_size":
|
| 1817 |
-
"bits":
|
| 1818 |
"mode": "affine"
|
| 1819 |
},
|
| 1820 |
"model.layers.7.self_attn.o_proj": {
|
| 1821 |
-
"group_size":
|
| 1822 |
-
"bits":
|
| 1823 |
"mode": "affine"
|
| 1824 |
},
|
| 1825 |
"model.layers.7.mlp.experts.gate_proj": {
|
| 1826 |
-
"group_size":
|
| 1827 |
-
"bits":
|
| 1828 |
"mode": "affine"
|
| 1829 |
},
|
| 1830 |
"model.layers.7.mlp.experts.up_proj": {
|
| 1831 |
-
"group_size":
|
| 1832 |
-
"bits":
|
| 1833 |
"mode": "affine"
|
| 1834 |
},
|
| 1835 |
"model.layers.7.mlp.experts.down_proj": {
|
| 1836 |
-
"group_size":
|
| 1837 |
-
"bits":
|
| 1838 |
"mode": "affine"
|
| 1839 |
},
|
| 1840 |
"model.layers.7.mlp.router": {
|
|
@@ -1843,38 +1843,38 @@
|
|
| 1843 |
"mode": "affine"
|
| 1844 |
},
|
| 1845 |
"model.layers.8.self_attn.q_proj": {
|
| 1846 |
-
"group_size":
|
| 1847 |
-
"bits":
|
| 1848 |
"mode": "affine"
|
| 1849 |
},
|
| 1850 |
"model.layers.8.self_attn.k_proj": {
|
| 1851 |
-
"group_size":
|
| 1852 |
-
"bits":
|
| 1853 |
"mode": "affine"
|
| 1854 |
},
|
| 1855 |
"model.layers.8.self_attn.v_proj": {
|
| 1856 |
-
"group_size":
|
| 1857 |
-
"bits":
|
| 1858 |
"mode": "affine"
|
| 1859 |
},
|
| 1860 |
"model.layers.8.self_attn.o_proj": {
|
| 1861 |
-
"group_size":
|
| 1862 |
-
"bits":
|
| 1863 |
"mode": "affine"
|
| 1864 |
},
|
| 1865 |
"model.layers.8.mlp.experts.gate_proj": {
|
| 1866 |
-
"group_size":
|
| 1867 |
-
"bits":
|
| 1868 |
"mode": "affine"
|
| 1869 |
},
|
| 1870 |
"model.layers.8.mlp.experts.up_proj": {
|
| 1871 |
-
"group_size":
|
| 1872 |
-
"bits":
|
| 1873 |
"mode": "affine"
|
| 1874 |
},
|
| 1875 |
"model.layers.8.mlp.experts.down_proj": {
|
| 1876 |
-
"group_size":
|
| 1877 |
-
"bits":
|
| 1878 |
"mode": "affine"
|
| 1879 |
},
|
| 1880 |
"model.layers.8.mlp.router": {
|
|
@@ -1883,13 +1883,13 @@
|
|
| 1883 |
"mode": "affine"
|
| 1884 |
},
|
| 1885 |
"model.layers.9.self_attn.q_proj": {
|
| 1886 |
-
"group_size":
|
| 1887 |
-
"bits":
|
| 1888 |
"mode": "affine"
|
| 1889 |
},
|
| 1890 |
"model.layers.9.self_attn.k_proj": {
|
| 1891 |
-
"group_size":
|
| 1892 |
-
"bits":
|
| 1893 |
"mode": "affine"
|
| 1894 |
},
|
| 1895 |
"model.layers.9.self_attn.v_proj": {
|
|
@@ -1898,23 +1898,23 @@
|
|
| 1898 |
"mode": "affine"
|
| 1899 |
},
|
| 1900 |
"model.layers.9.self_attn.o_proj": {
|
| 1901 |
-
"group_size":
|
| 1902 |
-
"bits":
|
| 1903 |
"mode": "affine"
|
| 1904 |
},
|
| 1905 |
"model.layers.9.mlp.experts.gate_proj": {
|
| 1906 |
-
"group_size":
|
| 1907 |
-
"bits":
|
| 1908 |
"mode": "affine"
|
| 1909 |
},
|
| 1910 |
"model.layers.9.mlp.experts.up_proj": {
|
| 1911 |
-
"group_size":
|
| 1912 |
-
"bits":
|
| 1913 |
"mode": "affine"
|
| 1914 |
},
|
| 1915 |
"model.layers.9.mlp.experts.down_proj": {
|
| 1916 |
-
"group_size":
|
| 1917 |
-
"bits":
|
| 1918 |
"mode": "affine"
|
| 1919 |
},
|
| 1920 |
"model.layers.9.mlp.router": {
|
|
@@ -1923,38 +1923,38 @@
|
|
| 1923 |
"mode": "affine"
|
| 1924 |
},
|
| 1925 |
"model.layers.10.self_attn.q_proj": {
|
| 1926 |
-
"group_size":
|
| 1927 |
-
"bits":
|
| 1928 |
"mode": "affine"
|
| 1929 |
},
|
| 1930 |
"model.layers.10.self_attn.k_proj": {
|
| 1931 |
-
"group_size":
|
| 1932 |
-
"bits":
|
| 1933 |
"mode": "affine"
|
| 1934 |
},
|
| 1935 |
"model.layers.10.self_attn.v_proj": {
|
| 1936 |
-
"group_size":
|
| 1937 |
-
"bits":
|
| 1938 |
"mode": "affine"
|
| 1939 |
},
|
| 1940 |
"model.layers.10.self_attn.o_proj": {
|
| 1941 |
-
"group_size":
|
| 1942 |
-
"bits":
|
| 1943 |
"mode": "affine"
|
| 1944 |
},
|
| 1945 |
"model.layers.10.mlp.experts.gate_proj": {
|
| 1946 |
-
"group_size":
|
| 1947 |
-
"bits":
|
| 1948 |
"mode": "affine"
|
| 1949 |
},
|
| 1950 |
"model.layers.10.mlp.experts.up_proj": {
|
| 1951 |
-
"group_size":
|
| 1952 |
-
"bits":
|
| 1953 |
"mode": "affine"
|
| 1954 |
},
|
| 1955 |
"model.layers.10.mlp.experts.down_proj": {
|
| 1956 |
-
"group_size":
|
| 1957 |
-
"bits":
|
| 1958 |
"mode": "affine"
|
| 1959 |
},
|
| 1960 |
"model.layers.10.mlp.router": {
|
|
@@ -1963,38 +1963,38 @@
|
|
| 1963 |
"mode": "affine"
|
| 1964 |
},
|
| 1965 |
"model.layers.11.self_attn.q_proj": {
|
| 1966 |
-
"group_size":
|
| 1967 |
-
"bits":
|
| 1968 |
"mode": "affine"
|
| 1969 |
},
|
| 1970 |
"model.layers.11.self_attn.k_proj": {
|
| 1971 |
-
"group_size":
|
| 1972 |
-
"bits":
|
| 1973 |
"mode": "affine"
|
| 1974 |
},
|
| 1975 |
"model.layers.11.self_attn.v_proj": {
|
| 1976 |
-
"group_size":
|
| 1977 |
-
"bits":
|
| 1978 |
"mode": "affine"
|
| 1979 |
},
|
| 1980 |
"model.layers.11.self_attn.o_proj": {
|
| 1981 |
-
"group_size":
|
| 1982 |
-
"bits":
|
| 1983 |
"mode": "affine"
|
| 1984 |
},
|
| 1985 |
"model.layers.11.mlp.experts.gate_proj": {
|
| 1986 |
-
"group_size":
|
| 1987 |
-
"bits":
|
| 1988 |
"mode": "affine"
|
| 1989 |
},
|
| 1990 |
"model.layers.11.mlp.experts.up_proj": {
|
| 1991 |
-
"group_size":
|
| 1992 |
-
"bits":
|
| 1993 |
"mode": "affine"
|
| 1994 |
},
|
| 1995 |
"model.layers.11.mlp.experts.down_proj": {
|
| 1996 |
-
"group_size":
|
| 1997 |
-
"bits":
|
| 1998 |
"mode": "affine"
|
| 1999 |
},
|
| 2000 |
"model.layers.11.mlp.router": {
|
|
@@ -2003,8 +2003,8 @@
|
|
| 2003 |
"mode": "affine"
|
| 2004 |
},
|
| 2005 |
"model.layers.12.self_attn.q_proj": {
|
| 2006 |
-
"group_size":
|
| 2007 |
-
"bits":
|
| 2008 |
"mode": "affine"
|
| 2009 |
},
|
| 2010 |
"model.layers.12.self_attn.k_proj": {
|
|
@@ -2018,23 +2018,23 @@
|
|
| 2018 |
"mode": "affine"
|
| 2019 |
},
|
| 2020 |
"model.layers.12.self_attn.o_proj": {
|
| 2021 |
-
"group_size":
|
| 2022 |
-
"bits":
|
| 2023 |
"mode": "affine"
|
| 2024 |
},
|
| 2025 |
"model.layers.12.mlp.experts.gate_proj": {
|
| 2026 |
-
"group_size":
|
| 2027 |
-
"bits":
|
| 2028 |
"mode": "affine"
|
| 2029 |
},
|
| 2030 |
"model.layers.12.mlp.experts.up_proj": {
|
| 2031 |
-
"group_size":
|
| 2032 |
-
"bits":
|
| 2033 |
"mode": "affine"
|
| 2034 |
},
|
| 2035 |
"model.layers.12.mlp.experts.down_proj": {
|
| 2036 |
-
"group_size":
|
| 2037 |
-
"bits":
|
| 2038 |
"mode": "affine"
|
| 2039 |
},
|
| 2040 |
"model.layers.12.mlp.router": {
|
|
@@ -2063,18 +2063,18 @@
|
|
| 2063 |
"mode": "affine"
|
| 2064 |
},
|
| 2065 |
"model.layers.13.mlp.experts.gate_proj": {
|
| 2066 |
-
"group_size":
|
| 2067 |
-
"bits":
|
| 2068 |
"mode": "affine"
|
| 2069 |
},
|
| 2070 |
"model.layers.13.mlp.experts.up_proj": {
|
| 2071 |
-
"group_size":
|
| 2072 |
-
"bits":
|
| 2073 |
"mode": "affine"
|
| 2074 |
},
|
| 2075 |
"model.layers.13.mlp.experts.down_proj": {
|
| 2076 |
-
"group_size":
|
| 2077 |
-
"bits":
|
| 2078 |
"mode": "affine"
|
| 2079 |
},
|
| 2080 |
"model.layers.13.mlp.router": {
|
|
@@ -2638,7 +2638,7 @@
|
|
| 2638 |
"mode": "affine"
|
| 2639 |
},
|
| 2640 |
"model.layers.27.mlp.router": {
|
| 2641 |
-
"group_size":
|
| 2642 |
"bits": 8,
|
| 2643 |
"mode": "affine"
|
| 2644 |
},
|
|
@@ -2678,7 +2678,7 @@
|
|
| 2678 |
"mode": "affine"
|
| 2679 |
},
|
| 2680 |
"model.layers.28.mlp.router": {
|
| 2681 |
-
"group_size":
|
| 2682 |
"bits": 8,
|
| 2683 |
"mode": "affine"
|
| 2684 |
},
|
|
@@ -2718,7 +2718,7 @@
|
|
| 2718 |
"mode": "affine"
|
| 2719 |
},
|
| 2720 |
"model.layers.29.mlp.router": {
|
| 2721 |
-
"group_size":
|
| 2722 |
"bits": 8,
|
| 2723 |
"mode": "affine"
|
| 2724 |
},
|
|
@@ -2758,7 +2758,7 @@
|
|
| 2758 |
"mode": "affine"
|
| 2759 |
},
|
| 2760 |
"model.layers.30.mlp.router": {
|
| 2761 |
-
"group_size":
|
| 2762 |
"bits": 8,
|
| 2763 |
"mode": "affine"
|
| 2764 |
},
|
|
@@ -2798,7 +2798,7 @@
|
|
| 2798 |
"mode": "affine"
|
| 2799 |
},
|
| 2800 |
"model.layers.31.mlp.router": {
|
| 2801 |
-
"group_size":
|
| 2802 |
"bits": 8,
|
| 2803 |
"mode": "affine"
|
| 2804 |
},
|
|
@@ -2838,7 +2838,7 @@
|
|
| 2838 |
"mode": "affine"
|
| 2839 |
},
|
| 2840 |
"model.layers.32.mlp.router": {
|
| 2841 |
-
"group_size":
|
| 2842 |
"bits": 8,
|
| 2843 |
"mode": "affine"
|
| 2844 |
},
|
|
@@ -2878,7 +2878,7 @@
|
|
| 2878 |
"mode": "affine"
|
| 2879 |
},
|
| 2880 |
"model.layers.33.mlp.router": {
|
| 2881 |
-
"group_size":
|
| 2882 |
"bits": 8,
|
| 2883 |
"mode": "affine"
|
| 2884 |
},
|
|
@@ -2918,7 +2918,7 @@
|
|
| 2918 |
"mode": "affine"
|
| 2919 |
},
|
| 2920 |
"model.layers.34.mlp.router": {
|
| 2921 |
-
"group_size":
|
| 2922 |
"bits": 8,
|
| 2923 |
"mode": "affine"
|
| 2924 |
},
|
|
@@ -2958,7 +2958,7 @@
|
|
| 2958 |
"mode": "affine"
|
| 2959 |
},
|
| 2960 |
"model.layers.35.mlp.router": {
|
| 2961 |
-
"group_size":
|
| 2962 |
"bits": 8,
|
| 2963 |
"mode": "affine"
|
| 2964 |
}
|
|
|
|
| 64 |
"output_router_logits": false,
|
| 65 |
"pad_token_id": 199999,
|
| 66 |
"quantization": {
|
| 67 |
+
"group_size": 64,
|
| 68 |
+
"bits": 6,
|
| 69 |
"mode": "affine",
|
| 70 |
"model.embed_tokens": {
|
| 71 |
"group_size": 64,
|
|
|
|
| 73 |
"mode": "affine"
|
| 74 |
},
|
| 75 |
"model.layers.0.self_attn.q_proj": {
|
| 76 |
+
"group_size": 64,
|
| 77 |
+
"bits": 8,
|
| 78 |
"mode": "affine"
|
| 79 |
},
|
| 80 |
"model.layers.0.self_attn.k_proj": {
|
| 81 |
+
"group_size": 64,
|
| 82 |
+
"bits": 8,
|
| 83 |
"mode": "affine"
|
| 84 |
},
|
| 85 |
"model.layers.0.self_attn.v_proj": {
|
| 86 |
+
"group_size": 64,
|
| 87 |
+
"bits": 8,
|
| 88 |
"mode": "affine"
|
| 89 |
},
|
| 90 |
"model.layers.0.self_attn.o_proj": {
|
| 91 |
+
"group_size": 64,
|
| 92 |
+
"bits": 8,
|
| 93 |
"mode": "affine"
|
| 94 |
},
|
| 95 |
"model.layers.0.mlp.experts.gate_proj": {
|
| 96 |
+
"group_size": 64,
|
| 97 |
+
"bits": 6,
|
| 98 |
"mode": "affine"
|
| 99 |
},
|
| 100 |
"model.layers.0.mlp.experts.up_proj": {
|
| 101 |
+
"group_size": 64,
|
| 102 |
+
"bits": 6,
|
| 103 |
"mode": "affine"
|
| 104 |
},
|
| 105 |
"model.layers.0.mlp.experts.down_proj": {
|
| 106 |
+
"group_size": 64,
|
| 107 |
+
"bits": 6,
|
| 108 |
"mode": "affine"
|
| 109 |
},
|
| 110 |
"model.layers.0.mlp.router": {
|
|
|
|
| 113 |
"mode": "affine"
|
| 114 |
},
|
| 115 |
"model.layers.1.self_attn.q_proj": {
|
| 116 |
+
"group_size": 64,
|
| 117 |
+
"bits": 8,
|
| 118 |
"mode": "affine"
|
| 119 |
},
|
| 120 |
"model.layers.1.self_attn.k_proj": {
|
| 121 |
+
"group_size": 64,
|
| 122 |
+
"bits": 8,
|
| 123 |
"mode": "affine"
|
| 124 |
},
|
| 125 |
"model.layers.1.self_attn.v_proj": {
|
| 126 |
+
"group_size": 64,
|
| 127 |
+
"bits": 8,
|
| 128 |
"mode": "affine"
|
| 129 |
},
|
| 130 |
"model.layers.1.self_attn.o_proj": {
|
| 131 |
+
"group_size": 64,
|
| 132 |
+
"bits": 8,
|
| 133 |
"mode": "affine"
|
| 134 |
},
|
| 135 |
"model.layers.1.mlp.experts.gate_proj": {
|
| 136 |
+
"group_size": 64,
|
| 137 |
+
"bits": 6,
|
| 138 |
"mode": "affine"
|
| 139 |
},
|
| 140 |
"model.layers.1.mlp.experts.up_proj": {
|
| 141 |
+
"group_size": 64,
|
| 142 |
+
"bits": 6,
|
| 143 |
"mode": "affine"
|
| 144 |
},
|
| 145 |
"model.layers.1.mlp.experts.down_proj": {
|
| 146 |
+
"group_size": 64,
|
| 147 |
+
"bits": 6,
|
| 148 |
"mode": "affine"
|
| 149 |
},
|
| 150 |
"model.layers.1.mlp.router": {
|
|
|
|
| 153 |
"mode": "affine"
|
| 154 |
},
|
| 155 |
"model.layers.2.self_attn.q_proj": {
|
| 156 |
+
"group_size": 64,
|
| 157 |
+
"bits": 8,
|
| 158 |
"mode": "affine"
|
| 159 |
},
|
| 160 |
"model.layers.2.self_attn.k_proj": {
|
| 161 |
+
"group_size": 64,
|
| 162 |
+
"bits": 8,
|
| 163 |
"mode": "affine"
|
| 164 |
},
|
| 165 |
"model.layers.2.self_attn.v_proj": {
|
| 166 |
+
"group_size": 64,
|
| 167 |
+
"bits": 8,
|
| 168 |
"mode": "affine"
|
| 169 |
},
|
| 170 |
"model.layers.2.self_attn.o_proj": {
|
| 171 |
+
"group_size": 64,
|
| 172 |
+
"bits": 8,
|
| 173 |
"mode": "affine"
|
| 174 |
},
|
| 175 |
"model.layers.2.mlp.experts.gate_proj": {
|
| 176 |
+
"group_size": 64,
|
| 177 |
+
"bits": 6,
|
| 178 |
"mode": "affine"
|
| 179 |
},
|
| 180 |
"model.layers.2.mlp.experts.up_proj": {
|
| 181 |
+
"group_size": 64,
|
| 182 |
+
"bits": 6,
|
| 183 |
"mode": "affine"
|
| 184 |
},
|
| 185 |
"model.layers.2.mlp.experts.down_proj": {
|
| 186 |
+
"group_size": 64,
|
| 187 |
+
"bits": 6,
|
| 188 |
"mode": "affine"
|
| 189 |
},
|
| 190 |
"model.layers.2.mlp.router": {
|
|
|
|
| 193 |
"mode": "affine"
|
| 194 |
},
|
| 195 |
"model.layers.3.self_attn.q_proj": {
|
| 196 |
+
"group_size": 64,
|
| 197 |
+
"bits": 8,
|
| 198 |
"mode": "affine"
|
| 199 |
},
|
| 200 |
"model.layers.3.self_attn.k_proj": {
|
| 201 |
+
"group_size": 64,
|
| 202 |
+
"bits": 8,
|
| 203 |
"mode": "affine"
|
| 204 |
},
|
| 205 |
"model.layers.3.self_attn.v_proj": {
|
| 206 |
+
"group_size": 64,
|
| 207 |
+
"bits": 8,
|
| 208 |
"mode": "affine"
|
| 209 |
},
|
| 210 |
"model.layers.3.self_attn.o_proj": {
|
| 211 |
+
"group_size": 64,
|
| 212 |
+
"bits": 8,
|
| 213 |
"mode": "affine"
|
| 214 |
},
|
| 215 |
"model.layers.3.mlp.experts.gate_proj": {
|
| 216 |
+
"group_size": 64,
|
| 217 |
+
"bits": 6,
|
| 218 |
"mode": "affine"
|
| 219 |
},
|
| 220 |
"model.layers.3.mlp.experts.up_proj": {
|
| 221 |
+
"group_size": 64,
|
| 222 |
+
"bits": 6,
|
| 223 |
"mode": "affine"
|
| 224 |
},
|
| 225 |
"model.layers.3.mlp.experts.down_proj": {
|
| 226 |
+
"group_size": 64,
|
| 227 |
+
"bits": 6,
|
| 228 |
"mode": "affine"
|
| 229 |
},
|
| 230 |
"model.layers.3.mlp.router": {
|
|
|
|
| 233 |
"mode": "affine"
|
| 234 |
},
|
| 235 |
"model.layers.4.self_attn.q_proj": {
|
| 236 |
+
"group_size": 64,
|
| 237 |
+
"bits": 8,
|
| 238 |
"mode": "affine"
|
| 239 |
},
|
| 240 |
"model.layers.4.self_attn.k_proj": {
|
| 241 |
+
"group_size": 64,
|
| 242 |
+
"bits": 8,
|
| 243 |
"mode": "affine"
|
| 244 |
},
|
| 245 |
"model.layers.4.self_attn.v_proj": {
|
| 246 |
+
"group_size": 64,
|
| 247 |
+
"bits": 8,
|
| 248 |
"mode": "affine"
|
| 249 |
},
|
| 250 |
"model.layers.4.self_attn.o_proj": {
|
| 251 |
+
"group_size": 64,
|
| 252 |
+
"bits": 8,
|
| 253 |
"mode": "affine"
|
| 254 |
},
|
| 255 |
"model.layers.4.mlp.experts.gate_proj": {
|
| 256 |
+
"group_size": 64,
|
| 257 |
+
"bits": 6,
|
| 258 |
"mode": "affine"
|
| 259 |
},
|
| 260 |
"model.layers.4.mlp.experts.up_proj": {
|
| 261 |
+
"group_size": 64,
|
| 262 |
+
"bits": 6,
|
| 263 |
"mode": "affine"
|
| 264 |
},
|
| 265 |
"model.layers.4.mlp.experts.down_proj": {
|
| 266 |
+
"group_size": 64,
|
| 267 |
+
"bits": 6,
|
| 268 |
"mode": "affine"
|
| 269 |
},
|
| 270 |
"model.layers.4.mlp.router": {
|
|
|
|
| 273 |
"mode": "affine"
|
| 274 |
},
|
| 275 |
"model.layers.5.self_attn.q_proj": {
|
| 276 |
+
"group_size": 64,
|
| 277 |
+
"bits": 8,
|
| 278 |
"mode": "affine"
|
| 279 |
},
|
| 280 |
"model.layers.5.self_attn.k_proj": {
|
| 281 |
+
"group_size": 64,
|
| 282 |
+
"bits": 8,
|
| 283 |
"mode": "affine"
|
| 284 |
},
|
| 285 |
"model.layers.5.self_attn.v_proj": {
|
| 286 |
+
"group_size": 64,
|
| 287 |
+
"bits": 8,
|
| 288 |
"mode": "affine"
|
| 289 |
},
|
| 290 |
"model.layers.5.self_attn.o_proj": {
|
| 291 |
+
"group_size": 64,
|
| 292 |
+
"bits": 8,
|
| 293 |
"mode": "affine"
|
| 294 |
},
|
| 295 |
"model.layers.5.mlp.experts.gate_proj": {
|
| 296 |
+
"group_size": 64,
|
| 297 |
+
"bits": 6,
|
| 298 |
"mode": "affine"
|
| 299 |
},
|
| 300 |
"model.layers.5.mlp.experts.up_proj": {
|
| 301 |
+
"group_size": 64,
|
| 302 |
+
"bits": 6,
|
| 303 |
"mode": "affine"
|
| 304 |
},
|
| 305 |
"model.layers.5.mlp.experts.down_proj": {
|
| 306 |
+
"group_size": 64,
|
| 307 |
+
"bits": 6,
|
| 308 |
"mode": "affine"
|
| 309 |
},
|
| 310 |
"model.layers.5.mlp.router": {
|
|
|
|
| 313 |
"mode": "affine"
|
| 314 |
},
|
| 315 |
"model.layers.6.self_attn.q_proj": {
|
| 316 |
+
"group_size": 64,
|
| 317 |
+
"bits": 8,
|
| 318 |
"mode": "affine"
|
| 319 |
},
|
| 320 |
"model.layers.6.self_attn.k_proj": {
|
| 321 |
+
"group_size": 64,
|
| 322 |
+
"bits": 8,
|
| 323 |
"mode": "affine"
|
| 324 |
},
|
| 325 |
"model.layers.6.self_attn.v_proj": {
|
| 326 |
+
"group_size": 64,
|
| 327 |
+
"bits": 8,
|
| 328 |
"mode": "affine"
|
| 329 |
},
|
| 330 |
"model.layers.6.self_attn.o_proj": {
|
| 331 |
+
"group_size": 64,
|
| 332 |
+
"bits": 8,
|
| 333 |
"mode": "affine"
|
| 334 |
},
|
| 335 |
"model.layers.6.mlp.experts.gate_proj": {
|
| 336 |
+
"group_size": 64,
|
| 337 |
+
"bits": 6,
|
| 338 |
"mode": "affine"
|
| 339 |
},
|
| 340 |
"model.layers.6.mlp.experts.up_proj": {
|
| 341 |
+
"group_size": 64,
|
| 342 |
+
"bits": 6,
|
| 343 |
"mode": "affine"
|
| 344 |
},
|
| 345 |
"model.layers.6.mlp.experts.down_proj": {
|
| 346 |
+
"group_size": 64,
|
| 347 |
+
"bits": 6,
|
| 348 |
"mode": "affine"
|
| 349 |
},
|
| 350 |
"model.layers.6.mlp.router": {
|
|
|
|
| 353 |
"mode": "affine"
|
| 354 |
},
|
| 355 |
"model.layers.7.self_attn.q_proj": {
|
| 356 |
+
"group_size": 64,
|
| 357 |
+
"bits": 8,
|
| 358 |
"mode": "affine"
|
| 359 |
},
|
| 360 |
"model.layers.7.self_attn.k_proj": {
|
| 361 |
+
"group_size": 64,
|
| 362 |
+
"bits": 8,
|
| 363 |
"mode": "affine"
|
| 364 |
},
|
| 365 |
"model.layers.7.self_attn.v_proj": {
|
| 366 |
+
"group_size": 64,
|
| 367 |
+
"bits": 8,
|
| 368 |
"mode": "affine"
|
| 369 |
},
|
| 370 |
"model.layers.7.self_attn.o_proj": {
|
| 371 |
+
"group_size": 64,
|
| 372 |
+
"bits": 8,
|
| 373 |
"mode": "affine"
|
| 374 |
},
|
| 375 |
"model.layers.7.mlp.experts.gate_proj": {
|
| 376 |
+
"group_size": 64,
|
| 377 |
+
"bits": 6,
|
| 378 |
"mode": "affine"
|
| 379 |
},
|
| 380 |
"model.layers.7.mlp.experts.up_proj": {
|
| 381 |
+
"group_size": 64,
|
| 382 |
+
"bits": 6,
|
| 383 |
"mode": "affine"
|
| 384 |
},
|
| 385 |
"model.layers.7.mlp.experts.down_proj": {
|
| 386 |
+
"group_size": 64,
|
| 387 |
+
"bits": 6,
|
| 388 |
"mode": "affine"
|
| 389 |
},
|
| 390 |
"model.layers.7.mlp.router": {
|
|
|
|
| 393 |
"mode": "affine"
|
| 394 |
},
|
| 395 |
"model.layers.8.self_attn.q_proj": {
|
| 396 |
+
"group_size": 64,
|
| 397 |
+
"bits": 8,
|
| 398 |
"mode": "affine"
|
| 399 |
},
|
| 400 |
"model.layers.8.self_attn.k_proj": {
|
| 401 |
+
"group_size": 64,
|
| 402 |
+
"bits": 8,
|
| 403 |
"mode": "affine"
|
| 404 |
},
|
| 405 |
"model.layers.8.self_attn.v_proj": {
|
| 406 |
+
"group_size": 64,
|
| 407 |
+
"bits": 8,
|
| 408 |
"mode": "affine"
|
| 409 |
},
|
| 410 |
"model.layers.8.self_attn.o_proj": {
|
| 411 |
+
"group_size": 64,
|
| 412 |
+
"bits": 8,
|
| 413 |
"mode": "affine"
|
| 414 |
},
|
| 415 |
"model.layers.8.mlp.experts.gate_proj": {
|
| 416 |
+
"group_size": 64,
|
| 417 |
+
"bits": 6,
|
| 418 |
"mode": "affine"
|
| 419 |
},
|
| 420 |
"model.layers.8.mlp.experts.up_proj": {
|
| 421 |
+
"group_size": 64,
|
| 422 |
+
"bits": 6,
|
| 423 |
"mode": "affine"
|
| 424 |
},
|
| 425 |
"model.layers.8.mlp.experts.down_proj": {
|
| 426 |
+
"group_size": 64,
|
| 427 |
+
"bits": 6,
|
| 428 |
"mode": "affine"
|
| 429 |
},
|
| 430 |
"model.layers.8.mlp.router": {
|
|
|
|
| 433 |
"mode": "affine"
|
| 434 |
},
|
| 435 |
"model.layers.9.self_attn.q_proj": {
|
| 436 |
+
"group_size": 64,
|
| 437 |
+
"bits": 8,
|
| 438 |
"mode": "affine"
|
| 439 |
},
|
| 440 |
"model.layers.9.self_attn.k_proj": {
|
| 441 |
+
"group_size": 64,
|
| 442 |
+
"bits": 8,
|
| 443 |
"mode": "affine"
|
| 444 |
},
|
| 445 |
"model.layers.9.self_attn.v_proj": {
|
|
|
|
| 448 |
"mode": "affine"
|
| 449 |
},
|
| 450 |
"model.layers.9.self_attn.o_proj": {
|
| 451 |
+
"group_size": 64,
|
| 452 |
+
"bits": 8,
|
| 453 |
"mode": "affine"
|
| 454 |
},
|
| 455 |
"model.layers.9.mlp.experts.gate_proj": {
|
| 456 |
+
"group_size": 64,
|
| 457 |
+
"bits": 6,
|
| 458 |
"mode": "affine"
|
| 459 |
},
|
| 460 |
"model.layers.9.mlp.experts.up_proj": {
|
| 461 |
+
"group_size": 64,
|
| 462 |
+
"bits": 6,
|
| 463 |
"mode": "affine"
|
| 464 |
},
|
| 465 |
"model.layers.9.mlp.experts.down_proj": {
|
| 466 |
+
"group_size": 64,
|
| 467 |
+
"bits": 6,
|
| 468 |
"mode": "affine"
|
| 469 |
},
|
| 470 |
"model.layers.9.mlp.router": {
|
|
|
|
| 473 |
"mode": "affine"
|
| 474 |
},
|
| 475 |
"model.layers.10.self_attn.q_proj": {
|
| 476 |
+
"group_size": 64,
|
| 477 |
+
"bits": 8,
|
| 478 |
"mode": "affine"
|
| 479 |
},
|
| 480 |
"model.layers.10.self_attn.k_proj": {
|
| 481 |
+
"group_size": 64,
|
| 482 |
+
"bits": 8,
|
| 483 |
"mode": "affine"
|
| 484 |
},
|
| 485 |
"model.layers.10.self_attn.v_proj": {
|
| 486 |
+
"group_size": 64,
|
| 487 |
+
"bits": 8,
|
| 488 |
"mode": "affine"
|
| 489 |
},
|
| 490 |
"model.layers.10.self_attn.o_proj": {
|
| 491 |
+
"group_size": 64,
|
| 492 |
+
"bits": 8,
|
| 493 |
"mode": "affine"
|
| 494 |
},
|
| 495 |
"model.layers.10.mlp.experts.gate_proj": {
|
| 496 |
+
"group_size": 64,
|
| 497 |
+
"bits": 6,
|
| 498 |
"mode": "affine"
|
| 499 |
},
|
| 500 |
"model.layers.10.mlp.experts.up_proj": {
|
| 501 |
+
"group_size": 64,
|
| 502 |
+
"bits": 6,
|
| 503 |
"mode": "affine"
|
| 504 |
},
|
| 505 |
"model.layers.10.mlp.experts.down_proj": {
|
| 506 |
+
"group_size": 64,
|
| 507 |
+
"bits": 6,
|
| 508 |
"mode": "affine"
|
| 509 |
},
|
| 510 |
"model.layers.10.mlp.router": {
|
|
|
|
| 513 |
"mode": "affine"
|
| 514 |
},
|
| 515 |
"model.layers.11.self_attn.q_proj": {
|
| 516 |
+
"group_size": 64,
|
| 517 |
+
"bits": 8,
|
| 518 |
"mode": "affine"
|
| 519 |
},
|
| 520 |
"model.layers.11.self_attn.k_proj": {
|
| 521 |
+
"group_size": 64,
|
| 522 |
+
"bits": 8,
|
| 523 |
"mode": "affine"
|
| 524 |
},
|
| 525 |
"model.layers.11.self_attn.v_proj": {
|
| 526 |
+
"group_size": 64,
|
| 527 |
+
"bits": 8,
|
| 528 |
"mode": "affine"
|
| 529 |
},
|
| 530 |
"model.layers.11.self_attn.o_proj": {
|
| 531 |
+
"group_size": 64,
|
| 532 |
+
"bits": 8,
|
| 533 |
"mode": "affine"
|
| 534 |
},
|
| 535 |
"model.layers.11.mlp.experts.gate_proj": {
|
| 536 |
+
"group_size": 64,
|
| 537 |
+
"bits": 6,
|
| 538 |
"mode": "affine"
|
| 539 |
},
|
| 540 |
"model.layers.11.mlp.experts.up_proj": {
|
| 541 |
+
"group_size": 64,
|
| 542 |
+
"bits": 6,
|
| 543 |
"mode": "affine"
|
| 544 |
},
|
| 545 |
"model.layers.11.mlp.experts.down_proj": {
|
| 546 |
+
"group_size": 64,
|
| 547 |
+
"bits": 6,
|
| 548 |
"mode": "affine"
|
| 549 |
},
|
| 550 |
"model.layers.11.mlp.router": {
|
|
|
|
| 553 |
"mode": "affine"
|
| 554 |
},
|
| 555 |
"model.layers.12.self_attn.q_proj": {
|
| 556 |
+
"group_size": 64,
|
| 557 |
+
"bits": 8,
|
| 558 |
"mode": "affine"
|
| 559 |
},
|
| 560 |
"model.layers.12.self_attn.k_proj": {
|
|
|
|
| 568 |
"mode": "affine"
|
| 569 |
},
|
| 570 |
"model.layers.12.self_attn.o_proj": {
|
| 571 |
+
"group_size": 64,
|
| 572 |
+
"bits": 8,
|
| 573 |
"mode": "affine"
|
| 574 |
},
|
| 575 |
"model.layers.12.mlp.experts.gate_proj": {
|
| 576 |
+
"group_size": 64,
|
| 577 |
+
"bits": 6,
|
| 578 |
"mode": "affine"
|
| 579 |
},
|
| 580 |
"model.layers.12.mlp.experts.up_proj": {
|
| 581 |
+
"group_size": 64,
|
| 582 |
+
"bits": 6,
|
| 583 |
"mode": "affine"
|
| 584 |
},
|
| 585 |
"model.layers.12.mlp.experts.down_proj": {
|
| 586 |
+
"group_size": 64,
|
| 587 |
+
"bits": 6,
|
| 588 |
"mode": "affine"
|
| 589 |
},
|
| 590 |
"model.layers.12.mlp.router": {
|
|
|
|
| 613 |
"mode": "affine"
|
| 614 |
},
|
| 615 |
"model.layers.13.mlp.experts.gate_proj": {
|
| 616 |
+
"group_size": 64,
|
| 617 |
+
"bits": 6,
|
| 618 |
"mode": "affine"
|
| 619 |
},
|
| 620 |
"model.layers.13.mlp.experts.up_proj": {
|
| 621 |
+
"group_size": 64,
|
| 622 |
+
"bits": 6,
|
| 623 |
"mode": "affine"
|
| 624 |
},
|
| 625 |
"model.layers.13.mlp.experts.down_proj": {
|
| 626 |
+
"group_size": 64,
|
| 627 |
+
"bits": 6,
|
| 628 |
"mode": "affine"
|
| 629 |
},
|
| 630 |
"model.layers.13.mlp.router": {
|
|
|
|
| 1188 |
"mode": "affine"
|
| 1189 |
},
|
| 1190 |
"model.layers.27.mlp.router": {
|
| 1191 |
+
"group_size": 64,
|
| 1192 |
"bits": 8,
|
| 1193 |
"mode": "affine"
|
| 1194 |
},
|
|
|
|
| 1228 |
"mode": "affine"
|
| 1229 |
},
|
| 1230 |
"model.layers.28.mlp.router": {
|
| 1231 |
+
"group_size": 64,
|
| 1232 |
"bits": 8,
|
| 1233 |
"mode": "affine"
|
| 1234 |
},
|
|
|
|
| 1268 |
"mode": "affine"
|
| 1269 |
},
|
| 1270 |
"model.layers.29.mlp.router": {
|
| 1271 |
+
"group_size": 64,
|
| 1272 |
"bits": 8,
|
| 1273 |
"mode": "affine"
|
| 1274 |
},
|
|
|
|
| 1308 |
"mode": "affine"
|
| 1309 |
},
|
| 1310 |
"model.layers.30.mlp.router": {
|
| 1311 |
+
"group_size": 64,
|
| 1312 |
"bits": 8,
|
| 1313 |
"mode": "affine"
|
| 1314 |
},
|
|
|
|
| 1348 |
"mode": "affine"
|
| 1349 |
},
|
| 1350 |
"model.layers.31.mlp.router": {
|
| 1351 |
+
"group_size": 64,
|
| 1352 |
"bits": 8,
|
| 1353 |
"mode": "affine"
|
| 1354 |
},
|
|
|
|
| 1388 |
"mode": "affine"
|
| 1389 |
},
|
| 1390 |
"model.layers.32.mlp.router": {
|
| 1391 |
+
"group_size": 64,
|
| 1392 |
"bits": 8,
|
| 1393 |
"mode": "affine"
|
| 1394 |
},
|
|
|
|
| 1428 |
"mode": "affine"
|
| 1429 |
},
|
| 1430 |
"model.layers.33.mlp.router": {
|
| 1431 |
+
"group_size": 64,
|
| 1432 |
"bits": 8,
|
| 1433 |
"mode": "affine"
|
| 1434 |
},
|
|
|
|
| 1468 |
"mode": "affine"
|
| 1469 |
},
|
| 1470 |
"model.layers.34.mlp.router": {
|
| 1471 |
+
"group_size": 64,
|
| 1472 |
"bits": 8,
|
| 1473 |
"mode": "affine"
|
| 1474 |
},
|
|
|
|
| 1508 |
"mode": "affine"
|
| 1509 |
},
|
| 1510 |
"model.layers.35.mlp.router": {
|
| 1511 |
+
"group_size": 64,
|
| 1512 |
"bits": 8,
|
| 1513 |
"mode": "affine"
|
| 1514 |
}
|
| 1515 |
},
|
| 1516 |
"quantization_config": {
|
| 1517 |
+
"group_size": 64,
|
| 1518 |
+
"bits": 6,
|
| 1519 |
"mode": "affine",
|
| 1520 |
"model.embed_tokens": {
|
| 1521 |
"group_size": 64,
|
|
|
|
| 1523 |
"mode": "affine"
|
| 1524 |
},
|
| 1525 |
"model.layers.0.self_attn.q_proj": {
|
| 1526 |
+
"group_size": 64,
|
| 1527 |
+
"bits": 8,
|
| 1528 |
"mode": "affine"
|
| 1529 |
},
|
| 1530 |
"model.layers.0.self_attn.k_proj": {
|
| 1531 |
+
"group_size": 64,
|
| 1532 |
+
"bits": 8,
|
| 1533 |
"mode": "affine"
|
| 1534 |
},
|
| 1535 |
"model.layers.0.self_attn.v_proj": {
|
| 1536 |
+
"group_size": 64,
|
| 1537 |
+
"bits": 8,
|
| 1538 |
"mode": "affine"
|
| 1539 |
},
|
| 1540 |
"model.layers.0.self_attn.o_proj": {
|
| 1541 |
+
"group_size": 64,
|
| 1542 |
+
"bits": 8,
|
| 1543 |
"mode": "affine"
|
| 1544 |
},
|
| 1545 |
"model.layers.0.mlp.experts.gate_proj": {
|
| 1546 |
+
"group_size": 64,
|
| 1547 |
+
"bits": 6,
|
| 1548 |
"mode": "affine"
|
| 1549 |
},
|
| 1550 |
"model.layers.0.mlp.experts.up_proj": {
|
| 1551 |
+
"group_size": 64,
|
| 1552 |
+
"bits": 6,
|
| 1553 |
"mode": "affine"
|
| 1554 |
},
|
| 1555 |
"model.layers.0.mlp.experts.down_proj": {
|
| 1556 |
+
"group_size": 64,
|
| 1557 |
+
"bits": 6,
|
| 1558 |
"mode": "affine"
|
| 1559 |
},
|
| 1560 |
"model.layers.0.mlp.router": {
|
|
|
|
| 1563 |
"mode": "affine"
|
| 1564 |
},
|
| 1565 |
"model.layers.1.self_attn.q_proj": {
|
| 1566 |
+
"group_size": 64,
|
| 1567 |
+
"bits": 8,
|
| 1568 |
"mode": "affine"
|
| 1569 |
},
|
| 1570 |
"model.layers.1.self_attn.k_proj": {
|
| 1571 |
+
"group_size": 64,
|
| 1572 |
+
"bits": 8,
|
| 1573 |
"mode": "affine"
|
| 1574 |
},
|
| 1575 |
"model.layers.1.self_attn.v_proj": {
|
| 1576 |
+
"group_size": 64,
|
| 1577 |
+
"bits": 8,
|
| 1578 |
"mode": "affine"
|
| 1579 |
},
|
| 1580 |
"model.layers.1.self_attn.o_proj": {
|
| 1581 |
+
"group_size": 64,
|
| 1582 |
+
"bits": 8,
|
| 1583 |
"mode": "affine"
|
| 1584 |
},
|
| 1585 |
"model.layers.1.mlp.experts.gate_proj": {
|
| 1586 |
+
"group_size": 64,
|
| 1587 |
+
"bits": 6,
|
| 1588 |
"mode": "affine"
|
| 1589 |
},
|
| 1590 |
"model.layers.1.mlp.experts.up_proj": {
|
| 1591 |
+
"group_size": 64,
|
| 1592 |
+
"bits": 6,
|
| 1593 |
"mode": "affine"
|
| 1594 |
},
|
| 1595 |
"model.layers.1.mlp.experts.down_proj": {
|
| 1596 |
+
"group_size": 64,
|
| 1597 |
+
"bits": 6,
|
| 1598 |
"mode": "affine"
|
| 1599 |
},
|
| 1600 |
"model.layers.1.mlp.router": {
|
|
|
|
| 1603 |
"mode": "affine"
|
| 1604 |
},
|
| 1605 |
"model.layers.2.self_attn.q_proj": {
|
| 1606 |
+
"group_size": 64,
|
| 1607 |
+
"bits": 8,
|
| 1608 |
"mode": "affine"
|
| 1609 |
},
|
| 1610 |
"model.layers.2.self_attn.k_proj": {
|
| 1611 |
+
"group_size": 64,
|
| 1612 |
+
"bits": 8,
|
| 1613 |
"mode": "affine"
|
| 1614 |
},
|
| 1615 |
"model.layers.2.self_attn.v_proj": {
|
| 1616 |
+
"group_size": 64,
|
| 1617 |
+
"bits": 8,
|
| 1618 |
"mode": "affine"
|
| 1619 |
},
|
| 1620 |
"model.layers.2.self_attn.o_proj": {
|
| 1621 |
+
"group_size": 64,
|
| 1622 |
+
"bits": 8,
|
| 1623 |
"mode": "affine"
|
| 1624 |
},
|
| 1625 |
"model.layers.2.mlp.experts.gate_proj": {
|
| 1626 |
+
"group_size": 64,
|
| 1627 |
+
"bits": 6,
|
| 1628 |
"mode": "affine"
|
| 1629 |
},
|
| 1630 |
"model.layers.2.mlp.experts.up_proj": {
|
| 1631 |
+
"group_size": 64,
|
| 1632 |
+
"bits": 6,
|
| 1633 |
"mode": "affine"
|
| 1634 |
},
|
| 1635 |
"model.layers.2.mlp.experts.down_proj": {
|
| 1636 |
+
"group_size": 64,
|
| 1637 |
+
"bits": 6,
|
| 1638 |
"mode": "affine"
|
| 1639 |
},
|
| 1640 |
"model.layers.2.mlp.router": {
|
|
|
|
| 1643 |
"mode": "affine"
|
| 1644 |
},
|
| 1645 |
"model.layers.3.self_attn.q_proj": {
|
| 1646 |
+
"group_size": 64,
|
| 1647 |
+
"bits": 8,
|
| 1648 |
"mode": "affine"
|
| 1649 |
},
|
| 1650 |
"model.layers.3.self_attn.k_proj": {
|
| 1651 |
+
"group_size": 64,
|
| 1652 |
+
"bits": 8,
|
| 1653 |
"mode": "affine"
|
| 1654 |
},
|
| 1655 |
"model.layers.3.self_attn.v_proj": {
|
| 1656 |
+
"group_size": 64,
|
| 1657 |
+
"bits": 8,
|
| 1658 |
"mode": "affine"
|
| 1659 |
},
|
| 1660 |
"model.layers.3.self_attn.o_proj": {
|
| 1661 |
+
"group_size": 64,
|
| 1662 |
+
"bits": 8,
|
| 1663 |
"mode": "affine"
|
| 1664 |
},
|
| 1665 |
"model.layers.3.mlp.experts.gate_proj": {
|
| 1666 |
+
"group_size": 64,
|
| 1667 |
+
"bits": 6,
|
| 1668 |
"mode": "affine"
|
| 1669 |
},
|
| 1670 |
"model.layers.3.mlp.experts.up_proj": {
|
| 1671 |
+
"group_size": 64,
|
| 1672 |
+
"bits": 6,
|
| 1673 |
"mode": "affine"
|
| 1674 |
},
|
| 1675 |
"model.layers.3.mlp.experts.down_proj": {
|
| 1676 |
+
"group_size": 64,
|
| 1677 |
+
"bits": 6,
|
| 1678 |
"mode": "affine"
|
| 1679 |
},
|
| 1680 |
"model.layers.3.mlp.router": {
|
|
|
|
| 1683 |
"mode": "affine"
|
| 1684 |
},
|
| 1685 |
"model.layers.4.self_attn.q_proj": {
|
| 1686 |
+
"group_size": 64,
|
| 1687 |
+
"bits": 8,
|
| 1688 |
"mode": "affine"
|
| 1689 |
},
|
| 1690 |
"model.layers.4.self_attn.k_proj": {
|
| 1691 |
+
"group_size": 64,
|
| 1692 |
+
"bits": 8,
|
| 1693 |
"mode": "affine"
|
| 1694 |
},
|
| 1695 |
"model.layers.4.self_attn.v_proj": {
|
| 1696 |
+
"group_size": 64,
|
| 1697 |
+
"bits": 8,
|
| 1698 |
"mode": "affine"
|
| 1699 |
},
|
| 1700 |
"model.layers.4.self_attn.o_proj": {
|
| 1701 |
+
"group_size": 64,
|
| 1702 |
+
"bits": 8,
|
| 1703 |
"mode": "affine"
|
| 1704 |
},
|
| 1705 |
"model.layers.4.mlp.experts.gate_proj": {
|
| 1706 |
+
"group_size": 64,
|
| 1707 |
+
"bits": 6,
|
| 1708 |
"mode": "affine"
|
| 1709 |
},
|
| 1710 |
"model.layers.4.mlp.experts.up_proj": {
|
| 1711 |
+
"group_size": 64,
|
| 1712 |
+
"bits": 6,
|
| 1713 |
"mode": "affine"
|
| 1714 |
},
|
| 1715 |
"model.layers.4.mlp.experts.down_proj": {
|
| 1716 |
+
"group_size": 64,
|
| 1717 |
+
"bits": 6,
|
| 1718 |
"mode": "affine"
|
| 1719 |
},
|
| 1720 |
"model.layers.4.mlp.router": {
|
|
|
|
| 1723 |
"mode": "affine"
|
| 1724 |
},
|
| 1725 |
"model.layers.5.self_attn.q_proj": {
|
| 1726 |
+
"group_size": 64,
|
| 1727 |
+
"bits": 8,
|
| 1728 |
"mode": "affine"
|
| 1729 |
},
|
| 1730 |
"model.layers.5.self_attn.k_proj": {
|
| 1731 |
+
"group_size": 64,
|
| 1732 |
+
"bits": 8,
|
| 1733 |
"mode": "affine"
|
| 1734 |
},
|
| 1735 |
"model.layers.5.self_attn.v_proj": {
|
| 1736 |
+
"group_size": 64,
|
| 1737 |
+
"bits": 8,
|
| 1738 |
"mode": "affine"
|
| 1739 |
},
|
| 1740 |
"model.layers.5.self_attn.o_proj": {
|
| 1741 |
+
"group_size": 64,
|
| 1742 |
+
"bits": 8,
|
| 1743 |
"mode": "affine"
|
| 1744 |
},
|
| 1745 |
"model.layers.5.mlp.experts.gate_proj": {
|
| 1746 |
+
"group_size": 64,
|
| 1747 |
+
"bits": 6,
|
| 1748 |
"mode": "affine"
|
| 1749 |
},
|
| 1750 |
"model.layers.5.mlp.experts.up_proj": {
|
| 1751 |
+
"group_size": 64,
|
| 1752 |
+
"bits": 6,
|
| 1753 |
"mode": "affine"
|
| 1754 |
},
|
| 1755 |
"model.layers.5.mlp.experts.down_proj": {
|
| 1756 |
+
"group_size": 64,
|
| 1757 |
+
"bits": 6,
|
| 1758 |
"mode": "affine"
|
| 1759 |
},
|
| 1760 |
"model.layers.5.mlp.router": {
|
|
|
|
| 1763 |
"mode": "affine"
|
| 1764 |
},
|
| 1765 |
"model.layers.6.self_attn.q_proj": {
|
| 1766 |
+
"group_size": 64,
|
| 1767 |
+
"bits": 8,
|
| 1768 |
"mode": "affine"
|
| 1769 |
},
|
| 1770 |
"model.layers.6.self_attn.k_proj": {
|
| 1771 |
+
"group_size": 64,
|
| 1772 |
+
"bits": 8,
|
| 1773 |
"mode": "affine"
|
| 1774 |
},
|
| 1775 |
"model.layers.6.self_attn.v_proj": {
|
| 1776 |
+
"group_size": 64,
|
| 1777 |
+
"bits": 8,
|
| 1778 |
"mode": "affine"
|
| 1779 |
},
|
| 1780 |
"model.layers.6.self_attn.o_proj": {
|
| 1781 |
+
"group_size": 64,
|
| 1782 |
+
"bits": 8,
|
| 1783 |
"mode": "affine"
|
| 1784 |
},
|
| 1785 |
"model.layers.6.mlp.experts.gate_proj": {
|
| 1786 |
+
"group_size": 64,
|
| 1787 |
+
"bits": 6,
|
| 1788 |
"mode": "affine"
|
| 1789 |
},
|
| 1790 |
"model.layers.6.mlp.experts.up_proj": {
|
| 1791 |
+
"group_size": 64,
|
| 1792 |
+
"bits": 6,
|
| 1793 |
"mode": "affine"
|
| 1794 |
},
|
| 1795 |
"model.layers.6.mlp.experts.down_proj": {
|
| 1796 |
+
"group_size": 64,
|
| 1797 |
+
"bits": 6,
|
| 1798 |
"mode": "affine"
|
| 1799 |
},
|
| 1800 |
"model.layers.6.mlp.router": {
|
|
|
|
| 1803 |
"mode": "affine"
|
| 1804 |
},
|
| 1805 |
"model.layers.7.self_attn.q_proj": {
|
| 1806 |
+
"group_size": 64,
|
| 1807 |
+
"bits": 8,
|
| 1808 |
"mode": "affine"
|
| 1809 |
},
|
| 1810 |
"model.layers.7.self_attn.k_proj": {
|
| 1811 |
+
"group_size": 64,
|
| 1812 |
+
"bits": 8,
|
| 1813 |
"mode": "affine"
|
| 1814 |
},
|
| 1815 |
"model.layers.7.self_attn.v_proj": {
|
| 1816 |
+
"group_size": 64,
|
| 1817 |
+
"bits": 8,
|
| 1818 |
"mode": "affine"
|
| 1819 |
},
|
| 1820 |
"model.layers.7.self_attn.o_proj": {
|
| 1821 |
+
"group_size": 64,
|
| 1822 |
+
"bits": 8,
|
| 1823 |
"mode": "affine"
|
| 1824 |
},
|
| 1825 |
"model.layers.7.mlp.experts.gate_proj": {
|
| 1826 |
+
"group_size": 64,
|
| 1827 |
+
"bits": 6,
|
| 1828 |
"mode": "affine"
|
| 1829 |
},
|
| 1830 |
"model.layers.7.mlp.experts.up_proj": {
|
| 1831 |
+
"group_size": 64,
|
| 1832 |
+
"bits": 6,
|
| 1833 |
"mode": "affine"
|
| 1834 |
},
|
| 1835 |
"model.layers.7.mlp.experts.down_proj": {
|
| 1836 |
+
"group_size": 64,
|
| 1837 |
+
"bits": 6,
|
| 1838 |
"mode": "affine"
|
| 1839 |
},
|
| 1840 |
"model.layers.7.mlp.router": {
|
|
|
|
| 1843 |
"mode": "affine"
|
| 1844 |
},
|
| 1845 |
"model.layers.8.self_attn.q_proj": {
|
| 1846 |
+
"group_size": 64,
|
| 1847 |
+
"bits": 8,
|
| 1848 |
"mode": "affine"
|
| 1849 |
},
|
| 1850 |
"model.layers.8.self_attn.k_proj": {
|
| 1851 |
+
"group_size": 64,
|
| 1852 |
+
"bits": 8,
|
| 1853 |
"mode": "affine"
|
| 1854 |
},
|
| 1855 |
"model.layers.8.self_attn.v_proj": {
|
| 1856 |
+
"group_size": 64,
|
| 1857 |
+
"bits": 8,
|
| 1858 |
"mode": "affine"
|
| 1859 |
},
|
| 1860 |
"model.layers.8.self_attn.o_proj": {
|
| 1861 |
+
"group_size": 64,
|
| 1862 |
+
"bits": 8,
|
| 1863 |
"mode": "affine"
|
| 1864 |
},
|
| 1865 |
"model.layers.8.mlp.experts.gate_proj": {
|
| 1866 |
+
"group_size": 64,
|
| 1867 |
+
"bits": 6,
|
| 1868 |
"mode": "affine"
|
| 1869 |
},
|
| 1870 |
"model.layers.8.mlp.experts.up_proj": {
|
| 1871 |
+
"group_size": 64,
|
| 1872 |
+
"bits": 6,
|
| 1873 |
"mode": "affine"
|
| 1874 |
},
|
| 1875 |
"model.layers.8.mlp.experts.down_proj": {
|
| 1876 |
+
"group_size": 64,
|
| 1877 |
+
"bits": 6,
|
| 1878 |
"mode": "affine"
|
| 1879 |
},
|
| 1880 |
"model.layers.8.mlp.router": {
|
|
|
|
| 1883 |
"mode": "affine"
|
| 1884 |
},
|
| 1885 |
"model.layers.9.self_attn.q_proj": {
|
| 1886 |
+
"group_size": 64,
|
| 1887 |
+
"bits": 8,
|
| 1888 |
"mode": "affine"
|
| 1889 |
},
|
| 1890 |
"model.layers.9.self_attn.k_proj": {
|
| 1891 |
+
"group_size": 64,
|
| 1892 |
+
"bits": 8,
|
| 1893 |
"mode": "affine"
|
| 1894 |
},
|
| 1895 |
"model.layers.9.self_attn.v_proj": {
|
|
|
|
| 1898 |
"mode": "affine"
|
| 1899 |
},
|
| 1900 |
"model.layers.9.self_attn.o_proj": {
|
| 1901 |
+
"group_size": 64,
|
| 1902 |
+
"bits": 8,
|
| 1903 |
"mode": "affine"
|
| 1904 |
},
|
| 1905 |
"model.layers.9.mlp.experts.gate_proj": {
|
| 1906 |
+
"group_size": 64,
|
| 1907 |
+
"bits": 6,
|
| 1908 |
"mode": "affine"
|
| 1909 |
},
|
| 1910 |
"model.layers.9.mlp.experts.up_proj": {
|
| 1911 |
+
"group_size": 64,
|
| 1912 |
+
"bits": 6,
|
| 1913 |
"mode": "affine"
|
| 1914 |
},
|
| 1915 |
"model.layers.9.mlp.experts.down_proj": {
|
| 1916 |
+
"group_size": 64,
|
| 1917 |
+
"bits": 6,
|
| 1918 |
"mode": "affine"
|
| 1919 |
},
|
| 1920 |
"model.layers.9.mlp.router": {
|
|
|
|
| 1923 |
"mode": "affine"
|
| 1924 |
},
|
| 1925 |
"model.layers.10.self_attn.q_proj": {
|
| 1926 |
+
"group_size": 64,
|
| 1927 |
+
"bits": 8,
|
| 1928 |
"mode": "affine"
|
| 1929 |
},
|
| 1930 |
"model.layers.10.self_attn.k_proj": {
|
| 1931 |
+
"group_size": 64,
|
| 1932 |
+
"bits": 8,
|
| 1933 |
"mode": "affine"
|
| 1934 |
},
|
| 1935 |
"model.layers.10.self_attn.v_proj": {
|
| 1936 |
+
"group_size": 64,
|
| 1937 |
+
"bits": 8,
|
| 1938 |
"mode": "affine"
|
| 1939 |
},
|
| 1940 |
"model.layers.10.self_attn.o_proj": {
|
| 1941 |
+
"group_size": 64,
|
| 1942 |
+
"bits": 8,
|
| 1943 |
"mode": "affine"
|
| 1944 |
},
|
| 1945 |
"model.layers.10.mlp.experts.gate_proj": {
|
| 1946 |
+
"group_size": 64,
|
| 1947 |
+
"bits": 6,
|
| 1948 |
"mode": "affine"
|
| 1949 |
},
|
| 1950 |
"model.layers.10.mlp.experts.up_proj": {
|
| 1951 |
+
"group_size": 64,
|
| 1952 |
+
"bits": 6,
|
| 1953 |
"mode": "affine"
|
| 1954 |
},
|
| 1955 |
"model.layers.10.mlp.experts.down_proj": {
|
| 1956 |
+
"group_size": 64,
|
| 1957 |
+
"bits": 6,
|
| 1958 |
"mode": "affine"
|
| 1959 |
},
|
| 1960 |
"model.layers.10.mlp.router": {
|
|
|
|
| 1963 |
"mode": "affine"
|
| 1964 |
},
|
| 1965 |
"model.layers.11.self_attn.q_proj": {
|
| 1966 |
+
"group_size": 64,
|
| 1967 |
+
"bits": 8,
|
| 1968 |
"mode": "affine"
|
| 1969 |
},
|
| 1970 |
"model.layers.11.self_attn.k_proj": {
|
| 1971 |
+
"group_size": 64,
|
| 1972 |
+
"bits": 8,
|
| 1973 |
"mode": "affine"
|
| 1974 |
},
|
| 1975 |
"model.layers.11.self_attn.v_proj": {
|
| 1976 |
+
"group_size": 64,
|
| 1977 |
+
"bits": 8,
|
| 1978 |
"mode": "affine"
|
| 1979 |
},
|
| 1980 |
"model.layers.11.self_attn.o_proj": {
|
| 1981 |
+
"group_size": 64,
|
| 1982 |
+
"bits": 8,
|
| 1983 |
"mode": "affine"
|
| 1984 |
},
|
| 1985 |
"model.layers.11.mlp.experts.gate_proj": {
|
| 1986 |
+
"group_size": 64,
|
| 1987 |
+
"bits": 6,
|
| 1988 |
"mode": "affine"
|
| 1989 |
},
|
| 1990 |
"model.layers.11.mlp.experts.up_proj": {
|
| 1991 |
+
"group_size": 64,
|
| 1992 |
+
"bits": 6,
|
| 1993 |
"mode": "affine"
|
| 1994 |
},
|
| 1995 |
"model.layers.11.mlp.experts.down_proj": {
|
| 1996 |
+
"group_size": 64,
|
| 1997 |
+
"bits": 6,
|
| 1998 |
"mode": "affine"
|
| 1999 |
},
|
| 2000 |
"model.layers.11.mlp.router": {
|
|
|
|
| 2003 |
"mode": "affine"
|
| 2004 |
},
|
| 2005 |
"model.layers.12.self_attn.q_proj": {
|
| 2006 |
+
"group_size": 64,
|
| 2007 |
+
"bits": 8,
|
| 2008 |
"mode": "affine"
|
| 2009 |
},
|
| 2010 |
"model.layers.12.self_attn.k_proj": {
|
|
|
|
| 2018 |
"mode": "affine"
|
| 2019 |
},
|
| 2020 |
"model.layers.12.self_attn.o_proj": {
|
| 2021 |
+
"group_size": 64,
|
| 2022 |
+
"bits": 8,
|
| 2023 |
"mode": "affine"
|
| 2024 |
},
|
| 2025 |
"model.layers.12.mlp.experts.gate_proj": {
|
| 2026 |
+
"group_size": 64,
|
| 2027 |
+
"bits": 6,
|
| 2028 |
"mode": "affine"
|
| 2029 |
},
|
| 2030 |
"model.layers.12.mlp.experts.up_proj": {
|
| 2031 |
+
"group_size": 64,
|
| 2032 |
+
"bits": 6,
|
| 2033 |
"mode": "affine"
|
| 2034 |
},
|
| 2035 |
"model.layers.12.mlp.experts.down_proj": {
|
| 2036 |
+
"group_size": 64,
|
| 2037 |
+
"bits": 6,
|
| 2038 |
"mode": "affine"
|
| 2039 |
},
|
| 2040 |
"model.layers.12.mlp.router": {
|
|
|
|
| 2063 |
"mode": "affine"
|
| 2064 |
},
|
| 2065 |
"model.layers.13.mlp.experts.gate_proj": {
|
| 2066 |
+
"group_size": 64,
|
| 2067 |
+
"bits": 6,
|
| 2068 |
"mode": "affine"
|
| 2069 |
},
|
| 2070 |
"model.layers.13.mlp.experts.up_proj": {
|
| 2071 |
+
"group_size": 64,
|
| 2072 |
+
"bits": 6,
|
| 2073 |
"mode": "affine"
|
| 2074 |
},
|
| 2075 |
"model.layers.13.mlp.experts.down_proj": {
|
| 2076 |
+
"group_size": 64,
|
| 2077 |
+
"bits": 6,
|
| 2078 |
"mode": "affine"
|
| 2079 |
},
|
| 2080 |
"model.layers.13.mlp.router": {
|
|
|
|
| 2638 |
"mode": "affine"
|
| 2639 |
},
|
| 2640 |
"model.layers.27.mlp.router": {
|
| 2641 |
+
"group_size": 64,
|
| 2642 |
"bits": 8,
|
| 2643 |
"mode": "affine"
|
| 2644 |
},
|
|
|
|
| 2678 |
"mode": "affine"
|
| 2679 |
},
|
| 2680 |
"model.layers.28.mlp.router": {
|
| 2681 |
+
"group_size": 64,
|
| 2682 |
"bits": 8,
|
| 2683 |
"mode": "affine"
|
| 2684 |
},
|
|
|
|
| 2718 |
"mode": "affine"
|
| 2719 |
},
|
| 2720 |
"model.layers.29.mlp.router": {
|
| 2721 |
+
"group_size": 64,
|
| 2722 |
"bits": 8,
|
| 2723 |
"mode": "affine"
|
| 2724 |
},
|
|
|
|
| 2758 |
"mode": "affine"
|
| 2759 |
},
|
| 2760 |
"model.layers.30.mlp.router": {
|
| 2761 |
+
"group_size": 64,
|
| 2762 |
"bits": 8,
|
| 2763 |
"mode": "affine"
|
| 2764 |
},
|
|
|
|
| 2798 |
"mode": "affine"
|
| 2799 |
},
|
| 2800 |
"model.layers.31.mlp.router": {
|
| 2801 |
+
"group_size": 64,
|
| 2802 |
"bits": 8,
|
| 2803 |
"mode": "affine"
|
| 2804 |
},
|
|
|
|
| 2838 |
"mode": "affine"
|
| 2839 |
},
|
| 2840 |
"model.layers.32.mlp.router": {
|
| 2841 |
+
"group_size": 64,
|
| 2842 |
"bits": 8,
|
| 2843 |
"mode": "affine"
|
| 2844 |
},
|
|
|
|
| 2878 |
"mode": "affine"
|
| 2879 |
},
|
| 2880 |
"model.layers.33.mlp.router": {
|
| 2881 |
+
"group_size": 64,
|
| 2882 |
"bits": 8,
|
| 2883 |
"mode": "affine"
|
| 2884 |
},
|
|
|
|
| 2918 |
"mode": "affine"
|
| 2919 |
},
|
| 2920 |
"model.layers.34.mlp.router": {
|
| 2921 |
+
"group_size": 64,
|
| 2922 |
"bits": 8,
|
| 2923 |
"mode": "affine"
|
| 2924 |
},
|
|
|
|
| 2958 |
"mode": "affine"
|
| 2959 |
},
|
| 2960 |
"model.layers.35.mlp.router": {
|
| 2961 |
+
"group_size": 64,
|
| 2962 |
"bits": 8,
|
| 2963 |
"mode": "affine"
|
| 2964 |
}
|
model-00001-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1186e99a101cfbc2fb7ddbc4343b4daa9bd2e8d5f06203f86db797b37f6a0b95
|
| 3 |
+
size 4988954394
|
model-00002-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:baf5106a0043aae7788a5c2a46e01e15581d96a7c2d3b7bdac6d1dc39105e172
|
| 3 |
+
size 5237378386
|
model-00003-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ff4e87775512d8f36f6fb7d355bb2b33c4798c65be9516ba7ad3c4c9dc7fef7a
|
| 3 |
+
size 5237378426
|
model-00004-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7b42f472293c0793843bc8daa79d263507049bc31b9704cfee2802112f91340e
|
| 3 |
+
size 5237378464
|
model-00005-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21d2e0666b8156cf9d887bb93da3438f89fd02072cdc334320268fb25f57e6fa
|
| 3 |
+
size 5237378426
|
model-00006-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3bb4f5ad61f197a33cfe154ea704c9c8b12e68b0cfe9f8b796677e726b5d86ea
|
| 3 |
+
size 5237378534
|
model-00007-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54b5439d12ab36b58ab659fd258e8792d7319e1befd06aeeb48f92e61129ada2
|
| 3 |
+
size 5237378486
|
model-00008-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:37b717aae69360982abda9165fd13b965187b28d6031f04e9906bc59cc4bbf45
|
| 3 |
+
size 5237378478
|
model-00009-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:64d3981b098d5ce152f94a44d061c2e285e0900f403a3f4a731e3eb04c0e0723
|
| 3 |
+
size 5237378526
|
model-00010-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b5a7b1492d3c61d9542fbca6ee655b2e16596f21bcd407f981b35ec6d8bbfc39
|
| 3 |
+
size 5237378464
|
model-00011-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8f83b7530039fad1ac2e514abd1f800b13b1e9e5dc97e150402fce62bace4670
|
| 3 |
+
size 5237378494
|
model-00012-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d352dda8c10083b530549c33c44798d639609f17337eeb057745fcc3b2340122
|
| 3 |
+
size 5237378486
|
model-00013-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f07fbaf05c6cefeb5d896c49c751355b867f3609d0ac9bdc0badd4590d45eb45
|
| 3 |
+
size 5237378490
|
model-00014-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ea0c6fa4285cb7e658f7d9720483c2bb8398f7d1d5e9e607b2243a30fe2c791
|
| 3 |
+
size 5237378536
|
model-00015-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9124aa1feb8abff1c72bdc0411fc82d32a2cf007a3c27e3516a44d17973f07c9
|
| 3 |
+
size 5237378510
|
model-00016-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:33d4fa556a17200392b588923642bcbe6af1a7bf9bf1b3ab8178ddb8ffdf568d
|
| 3 |
+
size 5237378510
|
model-00017-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2dd8eff0b3bccc1c113c34d22d401ccb73cb9e544d4169fa80bad93f0480eb4e
|
| 3 |
+
size 5237378506
|
model-00018-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:172d816c062df1087c0fa447015bfc7cddcac32fea8e535f40f10c8991aead51
|
| 3 |
+
size 5237378510
|
model-00019-of-00019.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec18218e771c835577fd99a07783d698d275d05f07d1974af311da71bc92407d
|
| 3 |
+
size 2022026460
|
model.safetensors.index.json
CHANGED
|
The diff for this file is too large to render.
See raw diff
|
|
|