Training in progress, step 7000, checkpoint

checkpoint-7000/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-7000/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-7000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b24806537222767ce524260e8e2c4fdbbe459ffd3855a1f42a703cd6712d82d3
+size 435544704

checkpoint-7000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f60e1b92bbc3ece469ada9b62015a893a623e111e0d219857988dbc6edcd2789
+size 871183627

checkpoint-7000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9473e60df214cc39c01d5bccff09d91b3dfeec0cc9f3cd2e2ead8d4da6378ed5
+size 14709

checkpoint-7000/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be40a03f70bb2a35bb567114fb5ceb05ce1222c8df24b1e1100746c617c5be10
+size 1383

checkpoint-7000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28f64db2c1980ed19749438d6000d120a310f2ece6f00f36cb92c8be840e0ae9
+size 1465

checkpoint-7000/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-7000/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-7000/trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_global_step": 6783,
+  "best_metric": 4.362748146057129,
+  "best_model_checkpoint": null,
+  "epoch": 17.54385964912281,
+  "eval_steps": 500,
+  "global_step": 7000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002506265664160401,
+      "grad_norm": 16.157943725585938,
+      "learning_rate": 0.0,
+      "loss": 10.3293,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.1738367080688477,
+      "learning_rate": 7.960000000000001e-05,
+      "loss": 6.9636,
+      "step": 399
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.99122428894043,
+      "eval_runtime": 15.8506,
+      "eval_samples_per_second": 1614.959,
+      "eval_steps_per_second": 6.309,
+      "step": 399
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.1836960315704346,
+      "learning_rate": 9.601876675603218e-05,
+      "loss": 5.6174,
+      "step": 798
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.353178977966309,
+      "eval_runtime": 19.6219,
+      "eval_samples_per_second": 1304.56,
+      "eval_steps_per_second": 5.096,
+      "step": 798
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.1592552661895752,
+      "learning_rate": 9.067024128686328e-05,
+      "loss": 5.1611,
+      "step": 1197
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 5.057892799377441,
+      "eval_runtime": 19.8214,
+      "eval_samples_per_second": 1291.434,
+      "eval_steps_per_second": 5.045,
+      "step": 1197
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.1559314727783203,
+      "learning_rate": 8.532171581769438e-05,
+      "loss": 4.8977,
+      "step": 1596
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.866283893585205,
+      "eval_runtime": 19.8295,
+      "eval_samples_per_second": 1290.907,
+      "eval_steps_per_second": 5.043,
+      "step": 1596
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.2687958478927612,
+      "learning_rate": 7.997319034852548e-05,
+      "loss": 4.7052,
+      "step": 1995
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.730337619781494,
+      "eval_runtime": 19.7815,
+      "eval_samples_per_second": 1294.039,
+      "eval_steps_per_second": 5.055,
+      "step": 1995
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.2619954347610474,
+      "learning_rate": 7.462466487935658e-05,
+      "loss": 4.5511,
+      "step": 2394
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.629268646240234,
+      "eval_runtime": 20.0233,
+      "eval_samples_per_second": 1278.411,
+      "eval_steps_per_second": 4.994,
+      "step": 2394
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.3349852561950684,
+      "learning_rate": 6.927613941018766e-05,
+      "loss": 4.4221,
+      "step": 2793
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.555833339691162,
+      "eval_runtime": 19.796,
+      "eval_samples_per_second": 1293.09,
+      "eval_steps_per_second": 5.052,
+      "step": 2793
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.3761085271835327,
+      "learning_rate": 6.392761394101878e-05,
+      "loss": 4.3109,
+      "step": 3192
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.497971534729004,
+      "eval_runtime": 19.8544,
+      "eval_samples_per_second": 1289.283,
+      "eval_steps_per_second": 5.037,
+      "step": 3192
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.4724619388580322,
+      "learning_rate": 5.8579088471849864e-05,
+      "loss": 4.2124,
+      "step": 3591
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.4558491706848145,
+      "eval_runtime": 19.7669,
+      "eval_samples_per_second": 1294.992,
+      "eval_steps_per_second": 5.059,
+      "step": 3591
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.5505802631378174,
+      "learning_rate": 5.3230563002680965e-05,
+      "loss": 4.1255,
+      "step": 3990
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.42624044418335,
+      "eval_runtime": 19.7583,
+      "eval_samples_per_second": 1295.56,
+      "eval_steps_per_second": 5.061,
+      "step": 3990
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.6021140813827515,
+      "learning_rate": 4.7882037533512065e-05,
+      "loss": 4.0484,
+      "step": 4389
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.400725364685059,
+      "eval_runtime": 19.86,
+      "eval_samples_per_second": 1288.919,
+      "eval_steps_per_second": 5.035,
+      "step": 4389
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.6689815521240234,
+      "learning_rate": 4.2533512064343165e-05,
+      "loss": 3.9801,
+      "step": 4788
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.383282661437988,
+      "eval_runtime": 19.9494,
+      "eval_samples_per_second": 1283.149,
+      "eval_steps_per_second": 5.013,
+      "step": 4788
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.7844934463500977,
+      "learning_rate": 3.7184986595174266e-05,
+      "loss": 3.9184,
+      "step": 5187
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.376080513000488,
+      "eval_runtime": 19.8076,
+      "eval_samples_per_second": 1292.335,
+      "eval_steps_per_second": 5.049,
+      "step": 5187
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.8226245641708374,
+      "learning_rate": 3.1836461126005366e-05,
+      "loss": 3.863,
+      "step": 5586
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.368453502655029,
+      "eval_runtime": 19.9905,
+      "eval_samples_per_second": 1280.511,
+      "eval_steps_per_second": 5.002,
+      "step": 5586
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.9303264617919922,
+      "learning_rate": 2.648793565683646e-05,
+      "loss": 3.8137,
+      "step": 5985
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.363076686859131,
+      "eval_runtime": 19.8022,
+      "eval_samples_per_second": 1292.685,
+      "eval_steps_per_second": 5.05,
+      "step": 5985
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.9857256412506104,
+      "learning_rate": 2.113941018766756e-05,
+      "loss": 3.7698,
+      "step": 6384
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.362977981567383,
+      "eval_runtime": 19.7778,
+      "eval_samples_per_second": 1294.282,
+      "eval_steps_per_second": 5.056,
+      "step": 6384
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 2.023864269256592,
+      "learning_rate": 1.579088471849866e-05,
+      "loss": 3.7316,
+      "step": 6783
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.362748146057129,
+      "eval_runtime": 19.8055,
+      "eval_samples_per_second": 1292.47,
+      "eval_steps_per_second": 5.049,
+      "step": 6783
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7980,
+  "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": 1.16904472068096e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22e02ae7be784e713ec5e825ef87f76e1a8b1722870cdd3a53841653946ba86a
+size 5905