Training in progress, step 6500, checkpoint

checkpoint-6500/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-6500/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-6500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63756435ee1e8f2b977d0bc2903d324c6e96b6cac60ab85d776ced233ba005c1
+size 435544704

checkpoint-6500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02fc2975b04a10d10e2786b0589f269d913e9d320a57b8674bd9a9d27ccafd94
+size 871183627

checkpoint-6500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7030f710e3b7b84f967b02ac46413c914e5e594614d2632a6c069d0d556473b8
+size 14709

checkpoint-6500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72ec040f7dbfafb8a0d9c346b901cc53a9760e25574232230ec90bf3c4a5f84d
+size 1383

checkpoint-6500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79e26752c1e062e4d19fd70ec3d008761fd48f642309df5a3f0689a3be39b590
+size 1465

checkpoint-6500/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-6500/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-6500/trainer_state.json

@@ -0,0 +1,281 @@
+{
+  "best_global_step": 5586,
+  "best_metric": 5.558029651641846,
+  "best_model_checkpoint": null,
+  "epoch": 16.290726817042607,
+  "eval_steps": 500,
+  "global_step": 6500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002506265664160401,
+      "grad_norm": 16.206899642944336,
+      "learning_rate": 0.0,
+      "loss": 10.3203,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.7419695258140564,
+      "learning_rate": 7.960000000000001e-05,
+      "loss": 7.2424,
+      "step": 399
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.557571887969971,
+      "eval_runtime": 15.7818,
+      "eval_samples_per_second": 1621.551,
+      "eval_steps_per_second": 6.336,
+      "step": 399
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8143869042396545,
+      "learning_rate": 9.601876675603218e-05,
+      "loss": 6.3077,
+      "step": 798
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 6.204631328582764,
+      "eval_runtime": 15.7933,
+      "eval_samples_per_second": 1620.371,
+      "eval_steps_per_second": 6.332,
+      "step": 798
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.8700923323631287,
+      "learning_rate": 9.067024128686328e-05,
+      "loss": 6.0417,
+      "step": 1197
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 6.021187782287598,
+      "eval_runtime": 15.8286,
+      "eval_samples_per_second": 1616.753,
+      "eval_steps_per_second": 6.318,
+      "step": 1197
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.9790167212486267,
+      "learning_rate": 8.532171581769438e-05,
+      "loss": 5.8739,
+      "step": 1596
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 5.898504257202148,
+      "eval_runtime": 15.7952,
+      "eval_samples_per_second": 1620.176,
+      "eval_steps_per_second": 6.331,
+      "step": 1596
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.0000659227371216,
+      "learning_rate": 7.997319034852548e-05,
+      "loss": 5.7396,
+      "step": 1995
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 5.8024163246154785,
+      "eval_runtime": 15.8237,
+      "eval_samples_per_second": 1617.262,
+      "eval_steps_per_second": 6.32,
+      "step": 1995
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.9891586303710938,
+      "learning_rate": 7.462466487935658e-05,
+      "loss": 5.6246,
+      "step": 2394
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.729940414428711,
+      "eval_runtime": 15.8227,
+      "eval_samples_per_second": 1617.36,
+      "eval_steps_per_second": 6.32,
+      "step": 2394
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.0765498876571655,
+      "learning_rate": 6.927613941018766e-05,
+      "loss": 5.5237,
+      "step": 2793
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.674533367156982,
+      "eval_runtime": 15.8362,
+      "eval_samples_per_second": 1615.977,
+      "eval_steps_per_second": 6.315,
+      "step": 2793
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.1928595304489136,
+      "learning_rate": 6.392761394101878e-05,
+      "loss": 5.4334,
+      "step": 3192
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.6343817710876465,
+      "eval_runtime": 15.8433,
+      "eval_samples_per_second": 1615.256,
+      "eval_steps_per_second": 6.312,
+      "step": 3192
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.2139155864715576,
+      "learning_rate": 5.8579088471849864e-05,
+      "loss": 5.3507,
+      "step": 3591
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.601651191711426,
+      "eval_runtime": 15.8447,
+      "eval_samples_per_second": 1615.111,
+      "eval_steps_per_second": 6.311,
+      "step": 3591
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.3513448238372803,
+      "learning_rate": 5.3230563002680965e-05,
+      "loss": 5.2753,
+      "step": 3990
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.58331298828125,
+      "eval_runtime": 15.8245,
+      "eval_samples_per_second": 1617.173,
+      "eval_steps_per_second": 6.319,
+      "step": 3990
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.4445401430130005,
+      "learning_rate": 4.7882037533512065e-05,
+      "loss": 5.2045,
+      "step": 4389
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.569336891174316,
+      "eval_runtime": 15.8642,
+      "eval_samples_per_second": 1613.124,
+      "eval_steps_per_second": 6.303,
+      "step": 4389
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.5078763961791992,
+      "learning_rate": 4.2533512064343165e-05,
+      "loss": 5.1388,
+      "step": 4788
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.561720848083496,
+      "eval_runtime": 15.8463,
+      "eval_samples_per_second": 1614.949,
+      "eval_steps_per_second": 6.311,
+      "step": 4788
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.625733733177185,
+      "learning_rate": 3.7184986595174266e-05,
+      "loss": 5.076,
+      "step": 5187
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.558526992797852,
+      "eval_runtime": 15.8791,
+      "eval_samples_per_second": 1611.617,
+      "eval_steps_per_second": 6.298,
+      "step": 5187
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.7136132717132568,
+      "learning_rate": 3.1836461126005366e-05,
+      "loss": 5.0198,
+      "step": 5586
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.558029651641846,
+      "eval_runtime": 15.7954,
+      "eval_samples_per_second": 1620.155,
+      "eval_steps_per_second": 6.331,
+      "step": 5586
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.830065369606018,
+      "learning_rate": 2.648793565683646e-05,
+      "loss": 4.9702,
+      "step": 5985
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.562778472900391,
+      "eval_runtime": 15.8517,
+      "eval_samples_per_second": 1614.401,
+      "eval_steps_per_second": 6.308,
+      "step": 5985
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.8890538215637207,
+      "learning_rate": 2.113941018766756e-05,
+      "loss": 4.9266,
+      "step": 6384
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.567871570587158,
+      "eval_runtime": 15.8713,
+      "eval_samples_per_second": 1612.41,
+      "eval_steps_per_second": 6.301,
+      "step": 6384
+    }
+  ],
+  "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": 1.08529213243392e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-6500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82b3cd9d93a757f178ab42bdab8791470771cd9dfdfc8c9e6b64381633c1ff10
+size 5905