Delete GLM-5.2 Vocabulary Efficiency Improvements.md

GLM-5.2 Vocabulary Efficiency Improvements.md

@@ -1,82 +0,0 @@
-# GLM-5.2 Vocabulary Pruning Candidates (`glm_5.2_prune_candidates.json`)
-
-Identified by Martial Terran, LLM Vocabulary Reduction Specialist, this repository provides a curated list of Token IDs that are candidates for pruning or deactivation from the GLM-5.2 model's active inference weights, tokenizer configuration, and output projection layers.
-
-The pruning configuration in `glm_5.2_prune_candidates.json` maps directly to rows that can be safely modified or removed within:
-* `config.json`
-* `tokenizer_config.json`
-* `tokenizer.json`
-* `chat_template.jinja`
-* **Input Embeddings:** `model.embed_tokens.weight` `{Shape: torch.Size([154880, 6144])}` in shard `model-00001-of-00282.safetensors`
-* **LM Head Projection:** `lm_head.weight` `{Shape: torch.Size([154880, 6144])}` in shard `model-00001-of-00282.safetensors`
-
-**Target Repository ID:** `zai-org/GLM-5.2`
-
-By safely deactivating or pruning these identified Token IDs from the active `lm_head` and `embed_tokens` matrices, you can reduce the parameter footprint and improve inference throughput.
-
----
-
-## Pruning Efficiency Gains
-
-Pruning these 331 inactive tokens translates to a reduction of **~20.34 Billion MACs** per 10,000 tokens of processed context.
-
-### Resource Reduction Report
-* **Number of Dead Tokens Identified:** 331
-* **Vocabulary Reduction:** 0.21% of the total vocabulary
-* **Memory Saved (Raw Weights):** 7.76 MB (at FP16/BF16 precision)
-* **Total Parameters Removed:** 4,067,328 parameters
-
-### Inference Efficiency (per 10,000 tokens)
-* **MACs Saved:** 20.34 Billion MACs
-* **FLOPs Saved:** 40.67 GFLOPs
-
----
-
-## Hardware-Aware Performance Profile
-
-To evaluate the real-world performance impact of pruning these tokens, we look beyond raw FLOP counts. During LLM deployment, execution is split into two fundamentally different hardware regimes:
-
-1. **Prefill Phase (Prompt Processing):** Compute-bound (GEMM). FLOP/MAC savings scale linearly with context length and batch size.
-2. **Decode Phase (Autoregressive Generation):** Memory-bandwidth bound (GEMV). The bottleneck is streaming the weight matrices from GPU High Bandwidth Memory (HBM) to SRAM/Registers. FLOP savings are secondary to **HBM read throughput savings**.
-
-### 1. Bandwidth Savings vs. FLOP Savings
-During the **Decode Phase**, GPU execution is heavily bound by memory bandwidth. By removing the elements identified in `glm_5.2_prune_candidates.json` from the `lm_head`, we save **3.14 MB of HBM reads per token generation step** (independent of batch size, as the weight matrix is streamed to SRAM once per step).
-
-### 2. Latency Profile Generation
-The table below translates these memory bandwidth savings into theoretical clock-time savings across NVIDIA A100, H100, and H200 accelerators based on raw hardware throughput limits.
-
----
-
-## GLM-5.2 Hardware-Aligned Pruning Report
-
-```text
-================================================================================
-      GLM-5.2 HARDWARE-ALIGNED PRUNING REPORT (d_model=6144)
-================================================================================
---- 1. VOCABULARY ALIGNMENT ANALYSIS (CUDA Warp / Tensor Core Optimization) ---
-Original Vocabulary Size:          154,880 (Multiple of 128)
-Raw Dead Tokens Identified:        331 (0.214% of vocab)
-Target Alignment Block Size:       128 tokens
-Optimized Aligned Vocab Size:      154,624
-Actual Tokens Pruned:              256 (Safely deleted rows)
-Padding Tokens Kept for Performance:75 (Preserves warp speed & coalescing)
-Effective Vocabulary Reduction:    0.165%
-
---- 2. STATIC FOOTPRINT REDUCTIONS (FP16/BF16 Precision) ---
-Weights Memory Saved:              6.000 MB (Aligned) / 7.760 MB (Raw)
-Total Parameters Trimmed:          3,145,728 parameters
-Input Embedding Matrix Shrinkage:  256 x 6144
-Output LM Head Matrix Shrinkage:   256 x 6144
-
---- 3. PREFILL PHASE PROFILE (Compute-Bound: GEMM) ---
-Execution Target Context:          8,192 tokens (Batch Size: 8)
-Computational MACs Saved:          103.0792 Billion MACs
-Computational FLOPs Saved:         206.1584 GFLOPs
-
---- 4. DECODE PHASE PROFILE (Bandwidth-Bound: GEMV) ---
-HBM Data Stream Saved per Step:    3072.00 KB / token step
-Theoretical Hardware Latency Saved per Autoregressive Step:
-  * NVIDIA A100 (SXM4)   | Bandwidth: 2.039 TB/s | Saved: 1542.78 ns
-  * NVIDIA H100 (SXM5)   | Bandwidth: 3.350 TB/s | Saved: 939.02 ns
-  * NVIDIA H200 (SXM)    | Bandwidth: 4.800 TB/s | Saved: 655.36 ns
-================================================================================