初始化项目，由ModelHub XC社区提供模型

Model: asinha08/slm-125m-sft-pilot-2k
Source: Original Platform

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,91 @@
+---
+license: apache-2.0
+base_model: thesreedath/slm-125m-base
+library_name: transformers
+pipeline_tag: text-generation
+tags:
+- llama
+- causal-lm
+- sft
+- legal
+- financial
+- small-language-model
+---
+
+# SLM 125M SFT Pilot 2K
+
+This is a supervised fine-tuned 125M-parameter causal language model based on
+[`thesreedath/slm-125m-base`](https://huggingface.co/thesreedath/slm-125m-base).
+
+## Training Summary
+
+| Field | Value |
+|---|---:|
+| SFT dataset run | `pilot-2k-v2` |
+| Train examples | 1800 |
+| Validation examples | 100 |
+| Epochs | 3 |
+| Optimizer steps | 339 |
+| Train tokens per epoch | 1262981 |
+| Assistant-label tokens per epoch | 62190 |
+| Total train-token exposures | 3788943 |
+| Total assistant-label token exposures | 186570 |
+| Initial validation loss | 3.6077 |
+| Initial validation perplexity | 36.88 |
+| Final validation loss | 1.5028 |
+| Final validation perplexity | 4.49 |
+| Training elapsed seconds | 160.1 |
+| Estimated GPU-only training cost | $0.0355 |
+
+## Dataset
+
+The SFT dataset contains 2,000 generated and judged instruction examples:
+
+| Split | Examples |
+|---|---:|
+| Train | 1,800 |
+| Validation | 100 |
+| Test | 100 |
+
+Source mix:
+
+| Source | Examples |
+|---|---:|
+| Case law | 703 |
+| SEC filings | 890 |
+| FineWeb-Edu | 407 |
+
+Task types include grounded question answering, unanswerable/refusal cases,
+multi-step QA, JSON extraction, summarization, plain-English rewrites, and
+comparative QA.
+
+## Intended Use
+
+This is a small research/experimentation model for legal and financial
+instruction-following behavior. It is not a substitute for professional legal,
+financial, or compliance advice.
+
+## Prompt Format
+
+The model was trained with the tokenizer/chat markers from the base model:
+
+```text
+<|system|>
+You are a careful legal and financial assistant...
+<|user|>
+Context:
+...
+
+Question:
+...
+<|assistant|>
+...
+<|eos|>
+```
+
+## Limitations
+
+- The SFT dataset is small, so behavior should be evaluated carefully.
+- Answers should be grounded in supplied context.
+- The model may hallucinate if used without retrieval/context.
+- Perplexity is measured on the generated validation split, not a broad external benchmark.

config.json

@@ -0,0 +1,30 @@
+{
+  "_name_or_path": "thesreedath/slm-125m-base",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "max_position_embeddings": 1024,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "num_key_value_heads": 12,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.46.3",
+  "use_cache": false,
+  "vocab_size": 16384
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.46.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a8eeb86e2bbdcf93c2a0b1c894c4be89f809ba638cbf8f53e5262180bd9dfe45
+size 251708904

sft_summary.json

@@ -0,0 +1,308 @@
+{
+  "run_name": "pilot-2k-v2",
+  "finetune_name": "slm-125m-sft-pilot-2k-e3",
+  "base_model_repo": "thesreedath/slm-125m-base",
+  "output_dir": "/data/sft/thesreedath-slm-125m/finetunes/slm-125m-sft-pilot-2k-e3",
+  "epochs": 3,
+  "optimizer_steps": 339,
+  "train_examples": 1800,
+  "val_examples": 100,
+  "train_tokens_per_epoch": 1262981,
+  "train_assistant_tokens_per_epoch": 62190,
+  "tokens_seen": 3788943,
+  "assistant_tokens_seen": 186570,
+  "initial_val_loss": 3.6076530987264204,
+  "initial_val_ppl": 36.87939886230172,
+  "final_val_loss": 1.5027814324426207,
+  "final_val_ppl": 4.494171937794634,
+  "elapsed_seconds": 160.08841013908386,
+  "estimated_gpu_usd": 0.03553962705087662,
+  "history": [
+    {
+      "step": 0,
+      "event": "initial_eval",
+      "val_loss": 3.6076530987264204,
+      "val_ppl": 36.87939886230172
+    },
+    {
+      "step": 1,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 3.3445167541503906,
+      "lr": 3.125e-06
+    },
+    {
+      "step": 10,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 3.2654700875282288,
+      "lr": 3.125e-05
+    },
+    {
+      "step": 20,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 2.046467971801758,
+      "lr": 4.9981082231081334e-05
+    },
+    {
+      "step": 30,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.6243530839681626,
+      "lr": 4.976858599798299e-05
+    },
+    {
+      "step": 40,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.5121346324682237,
+      "lr": 4.932196167813559e-05
+    },
+    {
+      "step": 50,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.4252592638134955,
+      "lr": 4.864543104251587e-05
+    },
+    {
+      "step": 50,
+      "event": "eval",
+      "epoch": 1,
+      "val_loss": 1.6063852179899305,
+      "val_ppl": 4.9847598014490835
+    },
+    {
+      "step": 60,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.4405100345611572,
+      "lr": 4.774538907965318e-05
+    },
+    {
+      "step": 70,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.3378228411078452,
+      "lr": 4.663034354621863e-05
+    },
+    {
+      "step": 80,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.404534536600113,
+      "lr": 4.531083454641531e-05
+    },
+    {
+      "step": 90,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.3485690742731093,
+      "lr": 4.3799334900354815e-05
+    },
+    {
+      "step": 100,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.5101487711071968,
+      "lr": 4.211013224320062e-05
+    },
+    {
+      "step": 100,
+      "event": "eval",
+      "epoch": 1,
+      "val_loss": 1.5369550711230229,
+      "val_ppl": 4.650408527530095
+    },
+    {
+      "step": 110,
+      "event": "train_log",
+      "epoch": 1,
+      "train_loss": 1.3192693248391152,
+      "lr": 4.025919396955019e-05
+    },
+    {
+      "step": 120,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.2987151977262998,
+      "lr": 3.826401629968575e-05
+    },
+    {
+      "step": 130,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.1192938029766082,
+      "lr": 3.6143458894413465e-05
+    },
+    {
+      "step": 140,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.1743364408612251,
+      "lr": 3.3917566581814276e-05
+    },
+    {
+      "step": 150,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.1988286420702934,
+      "lr": 3.160737988105633e-05
+    },
+    {
+      "step": 150,
+      "event": "eval",
+      "epoch": 2,
+      "val_loss": 1.5214578169787263,
+      "val_ppl": 4.578895523539861
+    },
+    {
+      "step": 160,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.192124105989933,
+      "lr": 2.9234736114315616e-05
+    },
+    {
+      "step": 170,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.1844297125935555,
+      "lr": 2.6822062986818796e-05
+    },
+    {
+      "step": 180,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.2872302904725075,
+      "lr": 2.4392166586218103e-05
+    },
+    {
+      "step": 190,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.18301559984684,
+      "lr": 2.1968015805260062e-05
+    },
+    {
+      "step": 200,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.217595275491476,
+      "lr": 1.9572525225519227e-05
+    },
+    {
+      "step": 200,
+      "event": "eval",
+      "epoch": 2,
+      "val_loss": 1.5065097094689361,
+      "val_ppl": 4.51095872924231
+    },
+    {
+      "step": 210,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.2071400985121727,
+      "lr": 1.7228338514514846e-05
+    },
+    {
+      "step": 220,
+      "event": "train_log",
+      "epoch": 2,
+      "train_loss": 1.2201956912875176,
+      "lr": 1.495761438367577e-05
+    },
+    {
+      "step": 230,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.162185350531026,
+      "lr": 1.2781817130412088e-05
+    },
+    {
+      "step": 240,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.2239526972174644,
+      "lr": 1.0721513744220169e-05
+    },
+    {
+      "step": 250,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.0888006776571273,
+      "lr": 8.796179494700441e-06
+    },
+    {
+      "step": 250,
+      "event": "eval",
+      "epoch": 3,
+      "val_loss": 1.5030882143383795,
+      "val_ppl": 4.495550879887996
+    },
+    {
+      "step": 260,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.166143298894167,
+      "lr": 7.02401383919088e-06
+    },
+    {
+      "step": 270,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.1675694316625596,
+      "lr": 5.421768390171933e-06
+    },
+    {
+      "step": 280,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.1555261366069316,
+      "lr": 4.004588568601897e-06
+    },
+    {
+      "step": 290,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.0684210263192653,
+      "lr": 2.7858704399739895e-06
+    },
+    {
+      "step": 300,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.1304051853716373,
+      "lr": 1.7771340863698306e-06
+    },
+    {
+      "step": 300,
+      "event": "eval",
+      "epoch": 3,
+      "val_loss": 1.5036853406266901,
+      "val_ppl": 4.498236093124735
+    },
+    {
+      "step": 310,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.17097322717309,
+      "lr": 9.879147114753545e-07
+    },
+    {
+      "step": 320,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.0470945492386818,
+      "lr": 4.256725079024554e-07
+    },
+    {
+      "step": 330,
+      "event": "train_log",
+      "epoch": 3,
+      "train_loss": 1.1463549926877021,
+      "lr": 9.572213880643111e-08
+    }
+  ]
+}

special_tokens_map.json

@@ -0,0 +1,35 @@
+{
+  "additional_special_tokens": [
+    "<|user|>",
+    "<|assistant|>",
+    "<|system|>"
+  ],
+  "bos_token": {
+    "content": "<|bos|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eos|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|pad|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|unk|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,72 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<|bos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|eos|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<|unk|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "6": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|user|>",
+    "<|assistant|>",
+    "<|system|>"
+  ],
+  "bos_token": "<|bos|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|eos|>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|pad|>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<|unk|>"
+}