Training in progress, step 8460, checkpoint

checkpoint-8460/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-8460/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-8460/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ee2929903e33aa724cfdbb2d69461c7bd25e79a2fa8becda26addeb646aa202
+size 435544704

checkpoint-8460/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97fce5c8d250fe562a69268e000d7105f125965c1f47ef7d89d8a085a1227dab
+size 871183627

checkpoint-8460/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ddba4b69dc83515fa1ad79535bff65c7517450c3d986b0a827a176c3d52dc8f4
+size 14709

checkpoint-8460/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:172741411778ba6f6cf9de0b8baebac0cb61c8b638698f266a498bb3d188ddbc
+size 1383

checkpoint-8460/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3b3c988fa271b0af9e23af1f12ebee61be32b9a4ae4f0e903937e330ad9552e
+size 1465

checkpoint-8460/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-8460/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-8460/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 3807,
+  "best_metric": 3.8857924938201904,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 8460,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002364066193853428,
+      "grad_norm": 20.532825469970703,
+      "learning_rate": 0.0,
+      "loss": 10.34,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2134920358657837,
+      "learning_rate": 8.44e-05,
+      "loss": 5.4142,
+      "step": 423
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 4.277193546295166,
+      "eval_runtime": 25.0006,
+      "eval_samples_per_second": 1102.575,
+      "eval_steps_per_second": 4.32,
+      "step": 423
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8720648288726807,
+      "learning_rate": 9.565491183879095e-05,
+      "loss": 4.0569,
+      "step": 846
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 4.067389965057373,
+      "eval_runtime": 24.8999,
+      "eval_samples_per_second": 1107.033,
+      "eval_steps_per_second": 4.337,
+      "step": 846
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.8065767288208008,
+      "learning_rate": 9.032745591939546e-05,
+      "loss": 3.9211,
+      "step": 1269
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.998242139816284,
+      "eval_runtime": 24.8652,
+      "eval_samples_per_second": 1108.575,
+      "eval_steps_per_second": 4.343,
+      "step": 1269
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.8496889472007751,
+      "learning_rate": 8.5e-05,
+      "loss": 3.8512,
+      "step": 1692
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.9567720890045166,
+      "eval_runtime": 24.8916,
+      "eval_samples_per_second": 1107.401,
+      "eval_steps_per_second": 4.339,
+      "step": 1692
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.8519867658615112,
+      "learning_rate": 7.967254408060454e-05,
+      "loss": 3.7997,
+      "step": 2115
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.9298720359802246,
+      "eval_runtime": 25.1278,
+      "eval_samples_per_second": 1096.993,
+      "eval_steps_per_second": 4.298,
+      "step": 2115
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.8216642141342163,
+      "learning_rate": 7.434508816120908e-05,
+      "loss": 3.7557,
+      "step": 2538
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.9152233600616455,
+      "eval_runtime": 24.8512,
+      "eval_samples_per_second": 1109.204,
+      "eval_steps_per_second": 4.346,
+      "step": 2538
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.8654234409332275,
+      "learning_rate": 6.90176322418136e-05,
+      "loss": 3.7157,
+      "step": 2961
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.8974955081939697,
+      "eval_runtime": 24.8309,
+      "eval_samples_per_second": 1110.109,
+      "eval_steps_per_second": 4.349,
+      "step": 2961
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.9256306290626526,
+      "learning_rate": 6.369017632241814e-05,
+      "loss": 3.6774,
+      "step": 3384
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.890566349029541,
+      "eval_runtime": 24.7927,
+      "eval_samples_per_second": 1111.819,
+      "eval_steps_per_second": 4.356,
+      "step": 3384
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.9523813724517822,
+      "learning_rate": 5.836272040302267e-05,
+      "loss": 3.6403,
+      "step": 3807
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.8857924938201904,
+      "eval_runtime": 24.8742,
+      "eval_samples_per_second": 1108.175,
+      "eval_steps_per_second": 4.342,
+      "step": 3807
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.09074866771698,
+      "learning_rate": 5.303526448362721e-05,
+      "loss": 3.6033,
+      "step": 4230
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.8870036602020264,
+      "eval_runtime": 24.8723,
+      "eval_samples_per_second": 1108.261,
+      "eval_steps_per_second": 4.342,
+      "step": 4230
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.140649437904358,
+      "learning_rate": 4.7707808564231746e-05,
+      "loss": 3.5665,
+      "step": 4653
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.89017391204834,
+      "eval_runtime": 24.7985,
+      "eval_samples_per_second": 1111.559,
+      "eval_steps_per_second": 4.355,
+      "step": 4653
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.2508670091629028,
+      "learning_rate": 4.2380352644836276e-05,
+      "loss": 3.5296,
+      "step": 5076
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.8938369750976562,
+      "eval_runtime": 24.8236,
+      "eval_samples_per_second": 1110.436,
+      "eval_steps_per_second": 4.351,
+      "step": 5076
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.3351261615753174,
+      "learning_rate": 3.7052896725440807e-05,
+      "loss": 3.4931,
+      "step": 5499
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.9042439460754395,
+      "eval_runtime": 24.8948,
+      "eval_samples_per_second": 1107.26,
+      "eval_steps_per_second": 4.338,
+      "step": 5499
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4390782117843628,
+      "learning_rate": 3.1725440806045344e-05,
+      "loss": 3.4577,
+      "step": 5922
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.9153685569763184,
+      "eval_runtime": 24.8858,
+      "eval_samples_per_second": 1107.659,
+      "eval_steps_per_second": 4.34,
+      "step": 5922
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.5242595672607422,
+      "learning_rate": 2.6397984886649874e-05,
+      "loss": 3.4232,
+      "step": 6345
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.92732834815979,
+      "eval_runtime": 24.8852,
+      "eval_samples_per_second": 1107.686,
+      "eval_steps_per_second": 4.34,
+      "step": 6345
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.6407968997955322,
+      "learning_rate": 2.107052896725441e-05,
+      "loss": 3.3913,
+      "step": 6768
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.9370839595794678,
+      "eval_runtime": 24.924,
+      "eval_samples_per_second": 1105.96,
+      "eval_steps_per_second": 4.333,
+      "step": 6768
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.757678747177124,
+      "learning_rate": 1.574307304785894e-05,
+      "loss": 3.3608,
+      "step": 7191
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.9483115673065186,
+      "eval_runtime": 24.9083,
+      "eval_samples_per_second": 1106.659,
+      "eval_steps_per_second": 4.336,
+      "step": 7191
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.8213372230529785,
+      "learning_rate": 1.0415617128463477e-05,
+      "loss": 3.3353,
+      "step": 7614
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 3.9591968059539795,
+      "eval_runtime": 24.7704,
+      "eval_samples_per_second": 1112.821,
+      "eval_steps_per_second": 4.36,
+      "step": 7614
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.8878958225250244,
+      "learning_rate": 5.088161209068011e-06,
+      "loss": 3.3142,
+      "step": 8037
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 3.9671552181243896,
+      "eval_runtime": 24.8702,
+      "eval_samples_per_second": 1108.354,
+      "eval_steps_per_second": 4.343,
+      "step": 8037
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8460,
+  "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": 1.4125577895936e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-8460/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a7d88f70941198d2b9bc96e0e79bcdca72aeea9fe10e1170c712bd49f37c2ef
+size 5969