Training in progress, step 7820, checkpoint

checkpoint-7820/config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "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": 2,
+  "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,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 30000
+}

checkpoint-7820/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "pad_token_id": 2,
+  "transformers_version": "4.56.1"
+}

checkpoint-7820/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a991faefdda2378ed478afdab7fce08a241e2cf24d6fef54c36e63ec9750b0c5
+size 435544704

checkpoint-7820/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5452579c323fc29fb687208e0bddee11396041036966b7f02a8d78f850635468
+size 871183627

checkpoint-7820/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf7288b0bfacb7db7f090259fbbcd70963ef252b090421737f0a7bde4ce6c8a8
+size 14645

checkpoint-7820/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f85d4bdba50d945e851aeab4d6c35670010d85c51eadcea7ad7fc231827fd2d
+size 1383

checkpoint-7820/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2228ca91d60f4d5fbaa26237c5737b472ba097cd903e9a7edb7eca80a503d063
+size 1465

checkpoint-7820/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

checkpoint-7820/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "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": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-7820/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 6256,
+  "best_metric": 4.335485935211182,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 7820,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025575447570332483,
+      "grad_norm": 15.986227989196777,
+      "learning_rate": 0.0,
+      "loss": 10.3853,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.020483136177063,
+      "learning_rate": 7.800000000000001e-05,
+      "loss": 6.8232,
+      "step": 391
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.762936115264893,
+      "eval_runtime": 22.5806,
+      "eval_samples_per_second": 1113.346,
+      "eval_steps_per_second": 4.384,
+      "step": 391
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8947324156761169,
+      "learning_rate": 9.615068493150685e-05,
+      "loss": 5.406,
+      "step": 782
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.187472820281982,
+      "eval_runtime": 22.1471,
+      "eval_samples_per_second": 1135.138,
+      "eval_steps_per_second": 4.47,
+      "step": 782
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.8840883374214172,
+      "learning_rate": 9.07945205479452e-05,
+      "loss": 5.0087,
+      "step": 1173
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.933140277862549,
+      "eval_runtime": 22.0467,
+      "eval_samples_per_second": 1140.306,
+      "eval_steps_per_second": 4.49,
+      "step": 1173
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8751250505447388,
+      "learning_rate": 8.543835616438357e-05,
+      "loss": 4.7829,
+      "step": 1564
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.771390438079834,
+      "eval_runtime": 22.5391,
+      "eval_samples_per_second": 1115.396,
+      "eval_steps_per_second": 4.392,
+      "step": 1564
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.0416064262390137,
+      "learning_rate": 8.008219178082192e-05,
+      "loss": 4.6167,
+      "step": 1955
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.658230781555176,
+      "eval_runtime": 20.7477,
+      "eval_samples_per_second": 1211.702,
+      "eval_steps_per_second": 4.772,
+      "step": 1955
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.9706395268440247,
+      "learning_rate": 7.472602739726028e-05,
+      "loss": 4.4815,
+      "step": 2346
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.569734573364258,
+      "eval_runtime": 19.6449,
+      "eval_samples_per_second": 1279.722,
+      "eval_steps_per_second": 5.039,
+      "step": 2346
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.0626953840255737,
+      "learning_rate": 6.936986301369864e-05,
+      "loss": 4.3666,
+      "step": 2737
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.503659725189209,
+      "eval_runtime": 19.5162,
+      "eval_samples_per_second": 1288.158,
+      "eval_steps_per_second": 5.073,
+      "step": 2737
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.1014233827590942,
+      "learning_rate": 6.401369863013699e-05,
+      "loss": 4.2653,
+      "step": 3128
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.452770233154297,
+      "eval_runtime": 19.4491,
+      "eval_samples_per_second": 1292.607,
+      "eval_steps_per_second": 5.09,
+      "step": 3128
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.123853087425232,
+      "learning_rate": 5.865753424657534e-05,
+      "loss": 4.1745,
+      "step": 3519
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.417596340179443,
+      "eval_runtime": 19.4015,
+      "eval_samples_per_second": 1295.779,
+      "eval_steps_per_second": 5.103,
+      "step": 3519
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.1906471252441406,
+      "learning_rate": 5.330136986301371e-05,
+      "loss": 4.0912,
+      "step": 3910
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.3872833251953125,
+      "eval_runtime": 19.4313,
+      "eval_samples_per_second": 1293.79,
+      "eval_steps_per_second": 5.095,
+      "step": 3910
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.283703327178955,
+      "learning_rate": 4.794520547945205e-05,
+      "loss": 4.017,
+      "step": 4301
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.368010997772217,
+      "eval_runtime": 19.5442,
+      "eval_samples_per_second": 1286.314,
+      "eval_steps_per_second": 5.065,
+      "step": 4301
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.327539324760437,
+      "learning_rate": 4.258904109589041e-05,
+      "loss": 3.9506,
+      "step": 4692
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.350579261779785,
+      "eval_runtime": 19.8262,
+      "eval_samples_per_second": 1268.018,
+      "eval_steps_per_second": 4.993,
+      "step": 4692
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.3899896144866943,
+      "learning_rate": 3.723287671232877e-05,
+      "loss": 3.8905,
+      "step": 5083
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.3434319496154785,
+      "eval_runtime": 19.6128,
+      "eval_samples_per_second": 1281.818,
+      "eval_steps_per_second": 5.048,
+      "step": 5083
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4696071147918701,
+      "learning_rate": 3.187671232876712e-05,
+      "loss": 3.8364,
+      "step": 5474
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.3391242027282715,
+      "eval_runtime": 19.5052,
+      "eval_samples_per_second": 1288.889,
+      "eval_steps_per_second": 5.076,
+      "step": 5474
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.527008295059204,
+      "learning_rate": 2.6520547945205482e-05,
+      "loss": 3.7881,
+      "step": 5865
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.339064121246338,
+      "eval_runtime": 19.4952,
+      "eval_samples_per_second": 1289.547,
+      "eval_steps_per_second": 5.078,
+      "step": 5865
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.549364686012268,
+      "learning_rate": 2.1164383561643834e-05,
+      "loss": 3.745,
+      "step": 6256
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.335485935211182,
+      "eval_runtime": 19.4207,
+      "eval_samples_per_second": 1294.496,
+      "eval_steps_per_second": 5.098,
+      "step": 6256
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.6290392875671387,
+      "learning_rate": 1.5808219178082193e-05,
+      "loss": 3.707,
+      "step": 6647
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.338380813598633,
+      "eval_runtime": 19.5085,
+      "eval_samples_per_second": 1288.666,
+      "eval_steps_per_second": 5.075,
+      "step": 6647
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.6309077739715576,
+      "learning_rate": 1.0452054794520549e-05,
+      "loss": 3.6743,
+      "step": 7038
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 4.3386549949646,
+      "eval_runtime": 19.523,
+      "eval_samples_per_second": 1287.712,
+      "eval_steps_per_second": 5.071,
+      "step": 7038
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.6526262760162354,
+      "learning_rate": 5.095890410958904e-06,
+      "loss": 3.6471,
+      "step": 7429
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 4.340756416320801,
+      "eval_runtime": 19.5229,
+      "eval_samples_per_second": 1287.716,
+      "eval_steps_per_second": 5.071,
+      "step": 7429
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7820,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3068782272512e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7820/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de1bc7d0325fc96682c56c891c931ecd7a72ba6e79ac54977acc7bfd389282b4
+size 5905