Training in progress, step 8500, checkpoint

checkpoint-8500/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-8500/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-8500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fad941ba1d215884fafa92dfff4f12c844f73fa84be47a8d1c855b54c5fe605
+size 435544704

checkpoint-8500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9677a3ceb53e9406e94e1f464a79b0d8b1660e591f83d07c65b3c48cc7793082
+size 871183627

checkpoint-8500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ef39f6b61843a254271a2b8689cd1276b392107b782930a283c64ad0bc591fe
+size 14709

checkpoint-8500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:793190053d1293f4ec72cd72bea61df6e622ffbc2a94ef18fbadb35cfca4423b
+size 1383

checkpoint-8500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69e8c999fea234f39a74455cd191610d153e365247bf00ee894b77163d73608d
+size 1465

checkpoint-8500/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-8500/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-8500/trainer_state.json

@@ -0,0 +1,296 @@
+{
+  "best_global_step": 5940,
+  "best_metric": 3.0583221912384033,
+  "best_model_checkpoint": null,
+  "epoch": 17.171717171717173,
+  "eval_steps": 500,
+  "global_step": 8500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.00202020202020202,
+      "grad_norm": 30.443870544433594,
+      "learning_rate": 0.0,
+      "loss": 10.3392,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.3942521810531616,
+      "learning_rate": 9.88e-05,
+      "loss": 4.8272,
+      "step": 495
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.784661054611206,
+      "eval_runtime": 19.6467,
+      "eval_samples_per_second": 1618.086,
+      "eval_steps_per_second": 6.362,
+      "step": 495
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.0911511182785034,
+      "learning_rate": 9.478678038379531e-05,
+      "loss": 3.5402,
+      "step": 990
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.438631534576416,
+      "eval_runtime": 19.6741,
+      "eval_samples_per_second": 1615.83,
+      "eval_steps_per_second": 6.354,
+      "step": 990
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0547674894332886,
+      "learning_rate": 8.950959488272922e-05,
+      "loss": 3.2821,
+      "step": 1485
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.2927286624908447,
+      "eval_runtime": 19.6268,
+      "eval_samples_per_second": 1619.722,
+      "eval_steps_per_second": 6.369,
+      "step": 1485
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.0422005653381348,
+      "learning_rate": 8.42324093816631e-05,
+      "loss": 3.137,
+      "step": 1980
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.211667537689209,
+      "eval_runtime": 19.8925,
+      "eval_samples_per_second": 1598.086,
+      "eval_steps_per_second": 6.284,
+      "step": 1980
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.0484884977340698,
+      "learning_rate": 7.895522388059702e-05,
+      "loss": 3.0348,
+      "step": 2475
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.154648780822754,
+      "eval_runtime": 19.6658,
+      "eval_samples_per_second": 1616.511,
+      "eval_steps_per_second": 6.356,
+      "step": 2475
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.0860364437103271,
+      "learning_rate": 7.367803837953093e-05,
+      "loss": 2.954,
+      "step": 2970
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.115788221359253,
+      "eval_runtime": 19.7596,
+      "eval_samples_per_second": 1608.836,
+      "eval_steps_per_second": 6.326,
+      "step": 2970
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.168921947479248,
+      "learning_rate": 6.840085287846483e-05,
+      "loss": 2.8855,
+      "step": 3465
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.09100341796875,
+      "eval_runtime": 19.6923,
+      "eval_samples_per_second": 1614.338,
+      "eval_steps_per_second": 6.348,
+      "step": 3465
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.189679741859436,
+      "learning_rate": 6.312366737739872e-05,
+      "loss": 2.825,
+      "step": 3960
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.0727946758270264,
+      "eval_runtime": 19.8838,
+      "eval_samples_per_second": 1598.79,
+      "eval_steps_per_second": 6.287,
+      "step": 3960
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.2425495386123657,
+      "learning_rate": 5.784648187633263e-05,
+      "loss": 2.7694,
+      "step": 4455
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.0644381046295166,
+      "eval_runtime": 19.6997,
+      "eval_samples_per_second": 1613.729,
+      "eval_steps_per_second": 6.345,
+      "step": 4455
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.2954844236373901,
+      "learning_rate": 5.256929637526653e-05,
+      "loss": 2.7177,
+      "step": 4950
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.0592901706695557,
+      "eval_runtime": 19.6912,
+      "eval_samples_per_second": 1614.424,
+      "eval_steps_per_second": 6.348,
+      "step": 4950
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.3890143632888794,
+      "learning_rate": 4.7292110874200426e-05,
+      "loss": 2.669,
+      "step": 5445
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.0597317218780518,
+      "eval_runtime": 19.6861,
+      "eval_samples_per_second": 1614.841,
+      "eval_steps_per_second": 6.35,
+      "step": 5445
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.4902243614196777,
+      "learning_rate": 4.201492537313433e-05,
+      "loss": 2.6224,
+      "step": 5940
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.0583221912384033,
+      "eval_runtime": 19.7225,
+      "eval_samples_per_second": 1611.868,
+      "eval_steps_per_second": 6.338,
+      "step": 5940
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.599095106124878,
+      "learning_rate": 3.673773987206823e-05,
+      "loss": 2.5777,
+      "step": 6435
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.0667269229888916,
+      "eval_runtime": 19.7076,
+      "eval_samples_per_second": 1613.084,
+      "eval_steps_per_second": 6.343,
+      "step": 6435
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.6431478261947632,
+      "learning_rate": 3.1460554371002134e-05,
+      "loss": 2.5359,
+      "step": 6930
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.073500156402588,
+      "eval_runtime": 19.7048,
+      "eval_samples_per_second": 1613.316,
+      "eval_steps_per_second": 6.344,
+      "step": 6930
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.743394136428833,
+      "learning_rate": 2.6183368869936037e-05,
+      "loss": 2.4989,
+      "step": 7425
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.0842723846435547,
+      "eval_runtime": 19.7143,
+      "eval_samples_per_second": 1612.538,
+      "eval_steps_per_second": 6.341,
+      "step": 7425
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.7900463342666626,
+      "learning_rate": 2.0906183368869936e-05,
+      "loss": 2.4655,
+      "step": 7920
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.091942310333252,
+      "eval_runtime": 19.7086,
+      "eval_samples_per_second": 1613.0,
+      "eval_steps_per_second": 6.342,
+      "step": 7920
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.864115595817566,
+      "learning_rate": 1.562899786780384e-05,
+      "loss": 2.4359,
+      "step": 8415
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.1009953022003174,
+      "eval_runtime": 19.7098,
+      "eval_samples_per_second": 1612.901,
+      "eval_steps_per_second": 6.342,
+      "step": 8415
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9900,
+  "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": 1.41979623211008e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-8500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1f4f8a984dc0bf87ec580dd1f6ed4858b247c41ee9e9a4e9c55a429eb143894
+size 5905