Upload folder using huggingface_hub

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/config.json

@@ -0,0 +1,39 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "pad_token_id": 50256,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "transformers_version": "4.56.0",
+  "use_cache": true,
+  "vocab_size": 50257
+}

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.56.0"
+}

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/lr_scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c35da3e5bbde9e797a5c5553120ff59acd38a9f4e960ebb6476bd15a03d1fb22
+size 1483

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:162a58da14b121f4491e1ea2d1c28087c3567295adc65936ff5a899e69a2a2e6
+size 497774208

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84960f14491685b331e700ed0df0b85ac091809ab265b536600a1d0d04cbda6b
+size 995642763

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

experiments/sft_gpt2-120m/20251209_070038/checkpoints/epoch_16/tokenizer_config.json

@@ -0,0 +1,21 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

experiments/sft_gpt2-120m/20251209_070038/sft_gpt2-120m.log

@@ -854,3 +854,46 @@
 2025-12-09 07:55:56,448 - absl - INFO - Using default tokenizer.
 2025-12-09 07:55:57,754 - root - INFO - Step 23169/28600 train rougeL: 0.979368932038835
 2025-12-09 07:55:57,810 - root - INFO - Step 23169/28600 loss: 0.017524462193250656, nll_loss: 0.017524462193250656
+2025-12-09 07:56:08,854 - absl - INFO - Using default tokenizer.
+2025-12-09 07:56:11,919 - root - INFO - Step 23297/28600 train rougeL: 0.9888107126313195
+2025-12-09 07:56:11,982 - root - INFO - Step 23297/28600 loss: 0.030149059370160103, nll_loss: 0.030149059370160103
+2025-12-09 07:56:23,052 - absl - INFO - Using default tokenizer.
+2025-12-09 07:56:25,326 - root - INFO - Step 23425/28600 train rougeL: 0.9094240541000018
+2025-12-09 07:56:25,389 - root - INFO - Step 23425/28600 loss: 0.031148415058851242, nll_loss: 0.031148415058851242
+2025-12-09 07:56:36,454 - absl - INFO - Using default tokenizer.
+2025-12-09 07:56:37,184 - root - INFO - Step 23553/28600 train rougeL: 1.0
+2025-12-09 07:56:37,240 - root - INFO - Step 23553/28600 loss: 0.04822136089205742, nll_loss: 0.04822136089205742
+2025-12-09 07:56:48,299 - absl - INFO - Using default tokenizer.
+2025-12-09 07:56:51,400 - root - INFO - Step 23681/28600 train rougeL: 0.8622660301159286
+2025-12-09 07:56:51,464 - root - INFO - Step 23681/28600 loss: 0.034108687192201614, nll_loss: 0.034108687192201614
+2025-12-09 07:57:02,536 - absl - INFO - Using default tokenizer.
+2025-12-09 07:57:04,043 - root - INFO - Step 23809/28600 train rougeL: 1.0
+2025-12-09 07:57:04,100 - root - INFO - Step 23809/28600 loss: 0.007645273581147194, nll_loss: 0.007645273581147194
+2025-12-09 07:57:15,162 - absl - INFO - Using default tokenizer.
+2025-12-09 07:57:18,286 - root - INFO - Step 23937/28600 train rougeL: 0.9782634446813552
+2025-12-09 07:57:18,349 - root - INFO - Step 23937/28600 loss: 0.021661849692463875, nll_loss: 0.021661849692463875
+2025-12-09 07:57:29,412 - absl - INFO - Using default tokenizer.
+2025-12-09 07:57:31,099 - root - INFO - Step 24065/28600 train rougeL: 0.9977272727272727
+2025-12-09 07:57:31,155 - root - INFO - Step 24065/28600 loss: 0.022607143968343735, nll_loss: 0.022607143968343735
+2025-12-09 07:57:42,219 - absl - INFO - Using default tokenizer.
+2025-12-09 07:57:45,239 - root - INFO - Step 24193/28600 train rougeL: 0.9979804421768708
+2025-12-09 07:57:45,302 - root - INFO - Step 24193/28600 loss: 0.03091508336365223, nll_loss: 0.03091508336365223
+2025-12-09 07:57:55,424 - root - INFO - Epoch 17/20 finished
+2025-12-09 07:57:55,453 - absl - INFO - Using default tokenizer.
+2025-12-09 07:57:58,824 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:01,735 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:05,059 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:07,981 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:11,667 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:14,930 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:18,235 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:21,610 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:24,956 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:28,280 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:31,603 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:34,971 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:38,325 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:41,470 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:44,248 - absl - INFO - Using default tokenizer.
+2025-12-09 07:58:46,987 - root - INFO - Epoch 17/20 eval rougeL: 0.23964140298317876
+2025-12-09 07:58:49,024 - root - INFO - Early stopping triggered at epoch 17