Training in progress, step 7500, checkpoint

checkpoint-7500/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-7500/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-7500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:42c68156bd07ad82028f8bf8c679ad4dcfdaad836002ea9ce39047013faf4674
+size 435544704

checkpoint-7500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7407cabbe358f660c9a26e8b9ed61ab928e2b77cdca14b338f8b368298f34062
+size 871183627

checkpoint-7500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47535e1f1708aaa81d9209600f3e8c3549732594773ab9a6d32d32cbaf505831
+size 14645

checkpoint-7500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:642827d515e7c3d88d20c04224822701abbbc9250deb1d366045f974d08b8258
+size 1383

checkpoint-7500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b6cdb05b9e32f2aac9e9dbd5d0d169bf677303349b5289e444ebd238bf4865e
+size 1465

checkpoint-7500/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-7500/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-7500/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 5096,
+  "best_metric": 5.1381754875183105,
+  "best_model_checkpoint": null,
+  "epoch": 19.132653061224488,
+  "eval_steps": 500,
+  "global_step": 7500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002551020408163265,
+      "grad_norm": 15.997297286987305,
+      "learning_rate": 0.0,
+      "loss": 10.3668,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.2358328104019165,
+      "learning_rate": 7.82e-05,
+      "loss": 6.9042,
+      "step": 392
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.92939567565918,
+      "eval_runtime": 19.2319,
+      "eval_samples_per_second": 1310.896,
+      "eval_steps_per_second": 5.148,
+      "step": 392
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.759972333908081,
+      "learning_rate": 9.613387978142077e-05,
+      "loss": 5.6847,
+      "step": 784
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.581369876861572,
+      "eval_runtime": 19.3238,
+      "eval_samples_per_second": 1304.658,
+      "eval_steps_per_second": 5.123,
+      "step": 784
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7742834091186523,
+      "learning_rate": 9.077868852459017e-05,
+      "loss": 5.4562,
+      "step": 1176
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.447832107543945,
+      "eval_runtime": 19.2983,
+      "eval_samples_per_second": 1306.384,
+      "eval_steps_per_second": 5.13,
+      "step": 1176
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.7993216514587402,
+      "learning_rate": 8.542349726775956e-05,
+      "loss": 5.3286,
+      "step": 1568
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 5.363584995269775,
+      "eval_runtime": 19.3964,
+      "eval_samples_per_second": 1299.779,
+      "eval_steps_per_second": 5.104,
+      "step": 1568
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7784366011619568,
+      "learning_rate": 8.006830601092896e-05,
+      "loss": 5.2299,
+      "step": 1960
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 5.298907279968262,
+      "eval_runtime": 19.4976,
+      "eval_samples_per_second": 1293.03,
+      "eval_steps_per_second": 5.078,
+      "step": 1960
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.8116441965103149,
+      "learning_rate": 7.471311475409837e-05,
+      "loss": 5.1442,
+      "step": 2352
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.24989652633667,
+      "eval_runtime": 19.5437,
+      "eval_samples_per_second": 1289.979,
+      "eval_steps_per_second": 5.066,
+      "step": 2352
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.8277950286865234,
+      "learning_rate": 6.935792349726776e-05,
+      "loss": 5.067,
+      "step": 2744
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.2103986740112305,
+      "eval_runtime": 19.3025,
+      "eval_samples_per_second": 1306.1,
+      "eval_steps_per_second": 5.129,
+      "step": 2744
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.9170593023300171,
+      "learning_rate": 6.400273224043716e-05,
+      "loss": 4.9956,
+      "step": 3136
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.180742263793945,
+      "eval_runtime": 19.4029,
+      "eval_samples_per_second": 1299.342,
+      "eval_steps_per_second": 5.102,
+      "step": 3136
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.9232907295227051,
+      "learning_rate": 5.864754098360656e-05,
+      "loss": 4.9288,
+      "step": 3528
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.161396026611328,
+      "eval_runtime": 19.3213,
+      "eval_samples_per_second": 1304.827,
+      "eval_steps_per_second": 5.124,
+      "step": 3528
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.020440697669983,
+      "learning_rate": 5.329234972677596e-05,
+      "loss": 4.8655,
+      "step": 3920
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.148547649383545,
+      "eval_runtime": 19.3239,
+      "eval_samples_per_second": 1304.654,
+      "eval_steps_per_second": 5.123,
+      "step": 3920
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.0839093923568726,
+      "learning_rate": 4.793715846994536e-05,
+      "loss": 4.8056,
+      "step": 4312
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.141533374786377,
+      "eval_runtime": 19.3136,
+      "eval_samples_per_second": 1305.348,
+      "eval_steps_per_second": 5.126,
+      "step": 4312
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.1907027959823608,
+      "learning_rate": 4.258196721311475e-05,
+      "loss": 4.7484,
+      "step": 4704
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.139172077178955,
+      "eval_runtime": 19.5421,
+      "eval_samples_per_second": 1290.087,
+      "eval_steps_per_second": 5.066,
+      "step": 4704
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.2344415187835693,
+      "learning_rate": 3.722677595628416e-05,
+      "loss": 4.6944,
+      "step": 5096
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.1381754875183105,
+      "eval_runtime": 19.2465,
+      "eval_samples_per_second": 1309.899,
+      "eval_steps_per_second": 5.144,
+      "step": 5096
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.3157490491867065,
+      "learning_rate": 3.1871584699453554e-05,
+      "loss": 4.6433,
+      "step": 5488
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.140122890472412,
+      "eval_runtime": 19.3693,
+      "eval_samples_per_second": 1301.593,
+      "eval_steps_per_second": 5.111,
+      "step": 5488
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.3936015367507935,
+      "learning_rate": 2.651639344262295e-05,
+      "loss": 4.5983,
+      "step": 5880
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.146300315856934,
+      "eval_runtime": 19.3881,
+      "eval_samples_per_second": 1300.333,
+      "eval_steps_per_second": 5.106,
+      "step": 5880
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.4432587623596191,
+      "learning_rate": 2.116120218579235e-05,
+      "loss": 4.5582,
+      "step": 6272
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.153017520904541,
+      "eval_runtime": 19.2624,
+      "eval_samples_per_second": 1308.822,
+      "eval_steps_per_second": 5.14,
+      "step": 6272
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.5094021558761597,
+      "learning_rate": 1.5806010928961746e-05,
+      "loss": 4.5233,
+      "step": 6664
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 5.158046245574951,
+      "eval_runtime": 19.2685,
+      "eval_samples_per_second": 1308.408,
+      "eval_steps_per_second": 5.138,
+      "step": 6664
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.5764825344085693,
+      "learning_rate": 1.0450819672131149e-05,
+      "loss": 4.4937,
+      "step": 7056
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 5.163742542266846,
+      "eval_runtime": 19.2425,
+      "eval_samples_per_second": 1310.171,
+      "eval_steps_per_second": 5.145,
+      "step": 7056
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.6419003009796143,
+      "learning_rate": 5.095628415300547e-06,
+      "loss": 4.4692,
+      "step": 7448
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 5.169681072235107,
+      "eval_runtime": 19.3185,
+      "eval_samples_per_second": 1305.021,
+      "eval_steps_per_second": 5.125,
+      "step": 7448
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7840,
+  "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": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.25309714153472e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68a529a8498a9d965819bf81859a3b9e43c9e3cfb93ce4bac7a5d224b0837459
+size 5905