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 sha256:588e7ba052d700d68ea553c13117c44409ef3020bda822af0708c2159df15c26
+size 435544704

checkpoint-7500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:670461202ec29e46c96e03fb3f1a1ec9429082bba1d4f1cb9e8889d293d01ac1
+size 871183627

checkpoint-7500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7ea6f80e8d19e7b21fefc9deb386a852eb6a8416479cde15bffb5147ac0a86d
+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 sha256:7c2718f4c6456312c6300989eb47f9ee8f37785e33d2baff5b09508b4dddac30
+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 +1,311 @@
+{
+  "best_global_step": 5586,
+  "best_metric": 5.5568108558654785,
+  "best_model_checkpoint": null,
+  "epoch": 18.796992481203006,
+  "eval_steps": 500,
+  "global_step": 7500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002506265664160401,
+      "grad_norm": 15.486847877502441,
+      "learning_rate": 0.0,
+      "loss": 10.3817,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.988777756690979,
+      "learning_rate": 7.960000000000001e-05,
+      "loss": 7.2659,
+      "step": 399
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.563127040863037,
+      "eval_runtime": 19.6799,
+      "eval_samples_per_second": 1300.365,
+      "eval_steps_per_second": 5.081,
+      "step": 399
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8962027430534363,
+      "learning_rate": 9.601876675603218e-05,
+      "loss": 6.3067,
+      "step": 798
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 6.204437255859375,
+      "eval_runtime": 19.5845,
+      "eval_samples_per_second": 1306.694,
+      "eval_steps_per_second": 5.106,
+      "step": 798
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.8679091334342957,
+      "learning_rate": 9.067024128686328e-05,
+      "loss": 6.0406,
+      "step": 1197
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 6.018397808074951,
+      "eval_runtime": 19.6084,
+      "eval_samples_per_second": 1305.103,
+      "eval_steps_per_second": 5.1,
+      "step": 1197
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.9358043074607849,
+      "learning_rate": 8.532171581769438e-05,
+      "loss": 5.8724,
+      "step": 1596
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 5.89630651473999,
+      "eval_runtime": 19.7935,
+      "eval_samples_per_second": 1292.902,
+      "eval_steps_per_second": 5.052,
+      "step": 1596
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.971953809261322,
+      "learning_rate": 7.997319034852548e-05,
+      "loss": 5.7384,
+      "step": 1995
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 5.800810813903809,
+      "eval_runtime": 19.6145,
+      "eval_samples_per_second": 1304.696,
+      "eval_steps_per_second": 5.098,
+      "step": 1995
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.0337553024291992,
+      "learning_rate": 7.462466487935658e-05,
+      "loss": 5.6236,
+      "step": 2394
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.728259086608887,
+      "eval_runtime": 19.8046,
+      "eval_samples_per_second": 1292.172,
+      "eval_steps_per_second": 5.049,
+      "step": 2394
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.1045554876327515,
+      "learning_rate": 6.927613941018766e-05,
+      "loss": 5.5224,
+      "step": 2793
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.671586990356445,
+      "eval_runtime": 19.7125,
+      "eval_samples_per_second": 1298.212,
+      "eval_steps_per_second": 5.073,
+      "step": 2793
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.176742672920227,
+      "learning_rate": 6.392761394101878e-05,
+      "loss": 5.4326,
+      "step": 3192
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.6318511962890625,
+      "eval_runtime": 19.571,
+      "eval_samples_per_second": 1307.599,
+      "eval_steps_per_second": 5.11,
+      "step": 3192
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.2789745330810547,
+      "learning_rate": 5.8579088471849864e-05,
+      "loss": 5.3503,
+      "step": 3591
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.602179527282715,
+      "eval_runtime": 19.6127,
+      "eval_samples_per_second": 1304.82,
+      "eval_steps_per_second": 5.099,
+      "step": 3591
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.3048105239868164,
+      "learning_rate": 5.3230563002680965e-05,
+      "loss": 5.2747,
+      "step": 3990
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.580410480499268,
+      "eval_runtime": 19.6991,
+      "eval_samples_per_second": 1299.094,
+      "eval_steps_per_second": 5.076,
+      "step": 3990
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.42238187789917,
+      "learning_rate": 4.7882037533512065e-05,
+      "loss": 5.2042,
+      "step": 4389
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.565695762634277,
+      "eval_runtime": 19.6414,
+      "eval_samples_per_second": 1302.91,
+      "eval_steps_per_second": 5.091,
+      "step": 4389
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.5199205875396729,
+      "learning_rate": 4.2533512064343165e-05,
+      "loss": 5.1385,
+      "step": 4788
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.559201240539551,
+      "eval_runtime": 19.5112,
+      "eval_samples_per_second": 1311.605,
+      "eval_steps_per_second": 5.125,
+      "step": 4788
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.6148369312286377,
+      "learning_rate": 3.7184986595174266e-05,
+      "loss": 5.0767,
+      "step": 5187
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.556906700134277,
+      "eval_runtime": 19.5998,
+      "eval_samples_per_second": 1305.678,
+      "eval_steps_per_second": 5.102,
+      "step": 5187
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.7083661556243896,
+      "learning_rate": 3.1836461126005366e-05,
+      "loss": 5.0205,
+      "step": 5586
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.5568108558654785,
+      "eval_runtime": 19.7252,
+      "eval_samples_per_second": 1297.374,
+      "eval_steps_per_second": 5.07,
+      "step": 5586
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.8180469274520874,
+      "learning_rate": 2.648793565683646e-05,
+      "loss": 4.9708,
+      "step": 5985
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.560354232788086,
+      "eval_runtime": 19.6256,
+      "eval_samples_per_second": 1303.962,
+      "eval_steps_per_second": 5.095,
+      "step": 5985
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.8938281536102295,
+      "learning_rate": 2.113941018766756e-05,
+      "loss": 4.9268,
+      "step": 6384
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.565486431121826,
+      "eval_runtime": 19.674,
+      "eval_samples_per_second": 1300.75,
+      "eval_steps_per_second": 5.083,
+      "step": 6384
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.9624561071395874,
+      "learning_rate": 1.579088471849866e-05,
+      "loss": 4.8895,
+      "step": 6783
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 5.56983757019043,
+      "eval_runtime": 19.7647,
+      "eval_samples_per_second": 1294.786,
+      "eval_steps_per_second": 5.06,
+      "step": 6783
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 2.052680492401123,
+      "learning_rate": 1.044235924932976e-05,
+      "loss": 4.8576,
+      "step": 7182
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 5.574769496917725,
+      "eval_runtime": 19.6267,
+      "eval_samples_per_second": 1303.889,
+      "eval_steps_per_second": 5.095,
+      "step": 7182
+    }
+  ],
+  "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": 1.25230869282816e+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:ed5469620139c1b9e18cc193389d8694ad90a34f9c1417f0546e2dad424a3d38
+size 5905