d3-edge v3.0.2

MODEL_MANIFEST.json

@@ -7,11 +7,11 @@
  "attention_mode": "noncausal_full_attention",
  "max_input_tokens": null,
  "identity": {
-  "model_sha256": "d7310c2f01e4ae4a2afe17b551bd44817a6a1c19baf61edf8f53b2f56a379c28",
+  "model_sha256": "5afbc0251ed54b60fa9ca704c4fe3141f3af5b8b3a0d0d1810785ecf13a3fdc2",
   "files_sha256": {
    "chat_template.jinja": "273d8e0e683b885071fb17e08d71e5f2a5ddfb5309756181681de4f5a1822d80",
-   "model-00001-of-00001.safetensors": "bbb17dc36c9c0f3b76384b090f49340a83df50f51a31cc8e80816a1865a393f2",
-   "readout.safetensors": "784aace7e722ffde162f902bb7f1cabba094935fdb18ceb4d8e37a90660a12ad",
+   "model-00001-of-00001.safetensors": "602da669692760b1bb8d36ef6a0ba928cbf2cc1d3277a11663498726c6602358",
+   "readout.safetensors": "27d73d80a6890e4ed854a0933112239517f6ae079c24c8cc295acf77004fff18",
    "tokenizer.json": "5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42",
    "tokenizer_config.json": "49e2b6e395f959f077f1e992b338919c0d4a9732fc6e613995e06557f843500c"
   },
@@ -543,10 +543,10 @@
   }
  },
  "parameters": {
-  "text": 623060864,
+  "text": 558394784,
   "vision": 100592896,
   "readout": 261120,
-  "loaded": 723914880
+  "loaded": 659248800
  },
  "base_model": {
   "repo_id": "Qwen/Qwen3.5-0.8B",
@@ -578,17 +578,17 @@
    }
   ]
  },
- "built_utc": "2026-10-11T00:31:41+00:00",
+ "built_utc": "2026-10-11T03:28:29+00:00",
  "files_sha256": {
   "LICENSE": "c71d239df91726fc519c6eb72d318ec65820627232b2f796219e87dcf35d0ab4",
   "NOTICE": "0e7694dee1927d83b3637be7a8f989b941ebd884dd40767b995c7adee484789b",
-  "README.md": "e419fdaa9158c20fb9dede9c486cc65f11ac602ceb5689cd3480547c2cb6d042",
-  "assets/banner.png": "3730b93b461b3d8e9683854873d54bbbcb2cc070c9353af9ce511e677f897336",
+  "README.md": "9b59a17514081b499fa8243404bfdc92be9e8641043392d2f63e50c227752a23",
+  "assets/banner.png": "352353b5ac097e31e6447ac0db8185439894e9e6c66376565e30f33c39d27913",
   "assets/example-receipt.png": "b9dbb8b103bacd45f1de824d844a3c07b81d57cb1a7466712ef060ad7dcb3926",
-  "assets/index-areas.png": "d729a56ce2f7a59ea11480b9a0f304530120f3fd379b3562458275172b27f3a2",
-  "assets/index-pareto.png": "2ae1dccb3e1851c29a6f18b9af1af354f118dfb2d7a06e797397d0cc375bd3d0",
+  "assets/index-areas.png": "9e8471ba167fe866d286b9873dfa1f48d0a509adb507d11bf8d4242f9a35397c",
+  "assets/index-pareto.png": "90a4f22032d1f03caf8890357c7f182a92ffa44e33cc2b445b372bc129bed754",
   "chat_template.jinja": "273d8e0e683b885071fb17e08d71e5f2a5ddfb5309756181681de4f5a1822d80",
-  "config.json": "daae2593536a0cfc5257361f99afe22c90ad0bd445b34774c795048cf987ae0f",
+  "config.json": "c06c8fab4bbbd151934e82d8c835e85974d360d0782905a184b940461595ddee",
   "d3_engine.py": "77628954a52cdd03048125899504f1ef55770a768ff8d2c34971a513bfb45a62",
   "d3_fast.py": "68a1fb9a2c3bde9ac2763bf504decae592ca12722698fbde1b39a31262955535",
   "d3_format.py": "187a0f359f6f89a844e723377769556f78133b118851e702e4b0fe5123ccb0f1",
@@ -597,12 +597,12 @@
   "d3_server.py": "11211faa83fee50b957903f0d596c153b0921cb5414f18edee31d8dac76db9fc",
   "decision_config.json": "30b6095b945d5a93697868afdf9be9123ea892085fbaff3078112c48da20dc6b",
   "merges.txt": "a9d356d7bdf1ef4949e3e748e95b8e10ad9d4e2e838eddc38a0a7b6b94d1db8d",
-  "model-00001-of-00001.safetensors": "bbb17dc36c9c0f3b76384b090f49340a83df50f51a31cc8e80816a1865a393f2",
-  "model.safetensors.index.json": "4e85e982af1fb788c98004e1f2e0d5893529d563104757fd2b4715a4f90130fe",
+  "model-00001-of-00001.safetensors": "602da669692760b1bb8d36ef6a0ba928cbf2cc1d3277a11663498726c6602358",
+  "model.safetensors.index.json": "1b835c61450c3dde31d1ab6e8c187ee435dd7c6b152c571e9efe0d87a5f16aed",
   "modeling_d3.py": "c87cb2c608ce078c8cb2ea8070624ab5a1c4eca02790492bd6e33c2fdc279acb",
   "pipeline_d3.py": "c0975a91316bcda9c09c5f14bf634ceec94e6980ed7e76a29b718fb451fe4091",
   "preprocessor_config.json": "27225450ac9c6529872ee1924fcb0962ff5634834f817040f444118116f4e516",
-  "readout.safetensors": "784aace7e722ffde162f902bb7f1cabba094935fdb18ceb4d8e37a90660a12ad",
+  "readout.safetensors": "27d73d80a6890e4ed854a0933112239517f6ae079c24c8cc295acf77004fff18",
   "requirements.txt": "3b0f92cb48584015ec0c9adf1559445806bd2ba9f6a41a92be6b2dd07fa48a21",
   "tokenizer.json": "5f9e4d4901a92b997e463c1f46055088b6cca5ca61a6522d1b9f64c4bb81cb42",
   "tokenizer_config.json": "49e2b6e395f959f077f1e992b338919c0d4a9732fc6e613995e06557f843500c",
@@ -612,13 +612,13 @@
  "files_bytes": {
   "LICENSE": 11357,
   "NOTICE": 106,
-  "README.md": 6230,
-  "assets/banner.png": 2234508,
+  "README.md": 6228,
+  "assets/banner.png": 2235551,
   "assets/example-receipt.png": 65918,
-  "assets/index-areas.png": 125826,
-  "assets/index-pareto.png": 186596,
+  "assets/index-areas.png": 125963,
+  "assets/index-pareto.png": 186784,
   "chat_template.jinja": 7755,
-  "config.json": 2728,
+  "config.json": 2650,
   "d3_engine.py": 4227,
   "d3_fast.py": 22331,
   "d3_format.py": 4881,
@@ -627,8 +627,8 @@
   "d3_server.py": 8333,
   "decision_config.json": 5504,
   "merges.txt": 3353259,
-  "model-00001-of-00001.safetensors": 1447352840,
-  "model.safetensors.index.json": 33257,
+  "model-00001-of-00001.safetensors": 1318015400,
+  "model.safetensors.index.json": 29351,
   "modeling_d3.py": 11673,
   "pipeline_d3.py": 2601,
   "preprocessor_config.json": 390,

README.md

@@ -17,21 +17,21 @@ tags:
 
 # d3-edge
 
-**d3-edge** is the 0.7B multimodal foundation decision model of Decision 3.0, the decision models of [vLLM Semantic Router](https://github.com/vllm-project/semantic-router). Give it an input (text or JSON, optionally with images) and the questions you need answered: pick one of several options, say yes or no, or rate on a scale. It answers them all in one call and returns a probability for every answer, without generating text.
+**d3-edge** is the 0.6B multimodal foundation decision model of Decision 3.0, the decision models of [vLLM Semantic Router](https://github.com/vllm-project/semantic-router). Give it an input (text or JSON, optionally with images) and the questions you need answered: pick one of several options, say yes or no, or rate on a scale. It answers them all in one call and returns a probability for every answer, without generating text.
 
 | | |
 | --- | --- |
-| **Parameters** | 0.72B, including the 0.10B vision encoder |
+| **Parameters** | 0.66B, including the 0.10B vision encoder |
 | **Inputs** | Text or JSON, plus images (several per request) |
 | **Decision types** | Choice · Yes / No · Score |
 | **License** | Apache-2.0 |
 
 ## Highlights
 
-- **Jev Decision Index 0.3, public suite: 30.06**, measured with the official 0.3 kit on the released weights: all 140,178 public requests answered, none unsupported.
-- **+13.4 on the public suite over Decision 2.0** (its 0.6B model: 16.68 on the board), ahead in all five areas.
+- **Jev Decision Index 0.3, public suite: 30.55**, measured with the official 0.3 kit on the released weights: all 140,178 public requests answered, none unsupported.
+- **+13.9 on the public suite over Decision 2.0** (its 0.6B model: 16.68 on the board), ahead in all five areas.
 - **Reads images:** multiple images per request (PNG, JPEG or WebP), given as paths, URLs, PIL images or base64 data URLs; every question of the request sees all of them.
-- **Speed:** a median of 10.3 ms for a text request and 44.7 ms for a request with an image, on one AMD Instinct MI325X GPU, one request at a time.
+- **Speed:** a median of 8.6 ms for a text request and 42.5 ms for a request with an image, on one AMD Instinct MI325X GPU, one request at a time.
 - **Many questions, one call:** Choice, Yes / No and Score questions about the same input are answered together, each from its own forward pass over the input, with a probability for every option.
 
 ## Quickstart
@@ -123,7 +123,7 @@ Images go before the text of the request, each read at up to 1.6 megapixels; eve
 
 | Model | Jev Decision Index ↑ | Public ↑ | Same-skill tests ↑ | New-domain tasks ↑ |
 | --- | ---: | ---: | ---: | ---: |
-| **d3-edge** | **19.1** | **30.1** | **24.5** | **9.4** |
+| **d3-edge** | **19.7** | **30.6** | **25.4** | **9.4** |
 | Bekko System One v0 400M | 15.2 | 22.5 | 21.6 | 7.2 |
 | Dinah-0 | 13.9 | 26.1 | 18.8 | 5.5 |
 | Decision 2.0 (0.6B) | 13.6 | 16.7 | 16.9 | 11.8 |
@@ -138,22 +138,22 @@ Images go before the text of the request, each read at up to 1.6 megapixels; eve
 
 | Model | Vision Index ↑ | Public ↑ | Private ↑ |
 | --- | ---: | ---: | ---: |
-| **d3-edge** | **42.1** | **44.4** | **39.8** |
+| **d3-edge** | **38.8** | **41.0** | **36.6** |
 
 d3-edge on the public vision benchmarks († approximate rebuild):
 
 | Benchmark | d3-edge |
 | --- | ---: |
-| CV-Bench | 59.9 |
-| BLINK | 35.7 |
-| RealWorldQA | 30.1 |
-| CharXiv † | 50.4 |
-| InfographicVQA † | 73.2 |
-| Mind2Web † | 44.8 |
-| Winoground | 48.0 |
-| KIE (CORD+FUNSD) † | 90.7 |
-| Moderation (Hateful Memes) | 13.4 |
-| R-Bench-M | 11.2 |
+| CV-Bench | 58.6 |
+| BLINK | 34.1 |
+| RealWorldQA | 27.6 |
+| CharXiv † | 46.9 |
+| InfographicVQA † | 70.0 |
+| Mind2Web † | 36.8 |
+| Winoground | 38.8 |
+| KIE (CORD+FUNSD) † | 88.6 |
+| Moderation (Hateful Memes) | 12.5 |
+| R-Bench-M | 9.0 |
 | MMMU-Pro vision | 5.3 |
 
 <sub>d3-edge: internal evaluation. Others: live board data, text 2026-10-10, vision 2026-10-09.</sub>

config.json

@@ -12,7 +12,7 @@
     "bos_token_id": null,
     "dtype": "bfloat16",
     "eos_token_id": 248044,
-    "full_attention_interval": 3,
+    "full_attention_interval": 4,
     "head_dim": 256,
     "hidden_act": "silu",
     "hidden_size": 1024,
@@ -29,13 +29,10 @@
       "linear_attention",
       "full_attention",
       "linear_attention",
-      "linear_attention",
       "full_attention",
       "linear_attention",
-      "linear_attention",
       "full_attention",
       "linear_attention",
-      "linear_attention",
       "full_attention"
     ],
     "linear_conv_kernel_dim": 4,
@@ -50,7 +47,7 @@
     "mtp_num_hidden_layers": 1,
     "mtp_use_dedicated_embeddings": false,
     "num_attention_heads": 8,
-    "num_hidden_layers": 18,
+    "num_hidden_layers": 15,
     "num_key_value_heads": 2,
     "pad_token_id": null,
     "partial_rotary_factor": 0.25,

model-00001-of-00001.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbb17dc36c9c0f3b76384b090f49340a83df50f51a31cc8e80816a1865a393f2
-size 1447352840
+oid sha256:602da669692760b1bb8d36ef6a0ba928cbf2cc1d3277a11663498726c6602358
+size 1318015400

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_size": 1447307520
+    "total_size": 1317975360
   },
   "weight_map": {
     "language_model.embed_tokens.weight": "model-00001-of-00001.safetensors",
@@ -33,44 +33,41 @@
     "language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.10.input_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.A_log": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.conv1d.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.dt_bias": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.in_proj_a.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.in_proj_b.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.in_proj_z.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.10.linear_attn.out_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.10.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.10.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.10.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.10.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.10.self_attn.k_norm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.10.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.10.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.10.self_attn.q_norm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.10.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.10.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.11.input_layernorm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.A_log": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.conv1d.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.dt_bias": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.in_proj_a.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.in_proj_b.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.in_proj_qkv.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.in_proj_z.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.norm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.11.linear_attn.out_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.11.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.11.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.11.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.11.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.11.self_attn.k_norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.11.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.11.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.11.self_attn.q_norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.11.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.11.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.12.input_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.A_log": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.conv1d.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.dt_bias": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.in_proj_a.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.in_proj_b.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.in_proj_z.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.12.linear_attn.out_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.12.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.12.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.12.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.12.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.12.self_attn.k_norm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.12.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.12.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.12.self_attn.q_norm.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.12.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
+    "language_model.layers.12.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.13.input_layernorm.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.13.linear_attn.A_log": "model-00001-of-00001.safetensors",
     "language_model.layers.13.linear_attn.conv1d.weight": "model-00001-of-00001.safetensors",
@@ -96,45 +93,6 @@
     "language_model.layers.14.self_attn.q_norm.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.14.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.14.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.input_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.A_log": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.conv1d.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.dt_bias": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.in_proj_a.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.in_proj_b.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.in_proj_qkv.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.in_proj_z.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.linear_attn.out_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.15.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.input_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.A_log": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.conv1d.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.dt_bias": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.in_proj_a.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.in_proj_b.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.in_proj_z.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.linear_attn.out_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.16.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.input_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.mlp.up_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.post_attention_layernorm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.self_attn.k_norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.self_attn.k_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.self_attn.o_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.self_attn.q_norm.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.self_attn.q_proj.weight": "model-00001-of-00001.safetensors",
-    "language_model.layers.17.self_attn.v_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.2.input_layernorm.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.2.mlp.down_proj.weight": "model-00001-of-00001.safetensors",
     "language_model.layers.2.mlp.gate_proj.weight": "model-00001-of-00001.safetensors",

readout.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:784aace7e722ffde162f902bb7f1cabba094935fdb18ceb4d8e37a90660a12ad
+oid sha256:27d73d80a6890e4ed854a0933112239517f6ae079c24c8cc295acf77004fff18
 size 1044560