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 sha256:74d8b6cefdb133b1d2331821233e6e801550d64339aeafc1f9a236c557b5ebf3
+size 435544704

checkpoint-7500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30af5ad33d0b4956b00c373c2b40a49337a5d93f2a44b126f2b5d19f83a1e1f2
+size 871183627

checkpoint-7500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5625a83f0ae8bcbd20ba38a254e1890d2fafe71134f189506d34029300b4c97d
+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:cee978c8158f406fbe140a66d4bee8ebecfcd1364c44cd3d30e6b18c1fd33cac
+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 +1,266 @@
+{
+  "best_global_step": 4464,
+  "best_metric": 3.094877004623413,
+  "best_model_checkpoint": null,
+  "epoch": 15.120967741935484,
+  "eval_steps": 500,
+  "global_step": 7500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0020161290322580645,
+      "grad_norm": 28.535022735595703,
+      "learning_rate": 0.0,
+      "loss": 10.3231,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.756303787231445,
+      "learning_rate": 9.900000000000001e-05,
+      "loss": 4.5921,
+      "step": 496
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.5128679275512695,
+      "eval_runtime": 19.6704,
+      "eval_samples_per_second": 1618.525,
+      "eval_steps_per_second": 6.355,
+      "step": 496
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 3.945728063583374,
+      "learning_rate": 9.477659574468086e-05,
+      "loss": 3.3197,
+      "step": 992
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.296919584274292,
+      "eval_runtime": 19.7075,
+      "eval_samples_per_second": 1615.48,
+      "eval_steps_per_second": 6.343,
+      "step": 992
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 3.4638240337371826,
+      "learning_rate": 8.950000000000001e-05,
+      "loss": 3.1648,
+      "step": 1488
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.2211971282958984,
+      "eval_runtime": 19.6423,
+      "eval_samples_per_second": 1620.836,
+      "eval_steps_per_second": 6.364,
+      "step": 1488
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 3.9391467571258545,
+      "learning_rate": 8.422340425531914e-05,
+      "loss": 3.0772,
+      "step": 1984
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.1721932888031006,
+      "eval_runtime": 19.9493,
+      "eval_samples_per_second": 1595.896,
+      "eval_steps_per_second": 6.266,
+      "step": 1984
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 4.159926414489746,
+      "learning_rate": 7.894680851063831e-05,
+      "loss": 3.0122,
+      "step": 2480
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.142347574234009,
+      "eval_runtime": 19.6986,
+      "eval_samples_per_second": 1616.208,
+      "eval_steps_per_second": 6.346,
+      "step": 2480
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 4.257205009460449,
+      "learning_rate": 7.367021276595744e-05,
+      "loss": 2.9591,
+      "step": 2976
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.121657371520996,
+      "eval_runtime": 19.7215,
+      "eval_samples_per_second": 1614.326,
+      "eval_steps_per_second": 6.338,
+      "step": 2976
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 4.3956170082092285,
+      "learning_rate": 6.83936170212766e-05,
+      "loss": 2.9117,
+      "step": 3472
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.1076769828796387,
+      "eval_runtime": 19.7076,
+      "eval_samples_per_second": 1615.472,
+      "eval_steps_per_second": 6.343,
+      "step": 3472
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 4.139720916748047,
+      "learning_rate": 6.311702127659575e-05,
+      "loss": 2.8674,
+      "step": 3968
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.097724437713623,
+      "eval_runtime": 19.9221,
+      "eval_samples_per_second": 1598.075,
+      "eval_steps_per_second": 6.274,
+      "step": 3968
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 4.987766742706299,
+      "learning_rate": 5.7840425531914896e-05,
+      "loss": 2.8258,
+      "step": 4464
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.094877004623413,
+      "eval_runtime": 19.7658,
+      "eval_samples_per_second": 1610.713,
+      "eval_steps_per_second": 6.324,
+      "step": 4464
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 4.529844284057617,
+      "learning_rate": 5.256382978723404e-05,
+      "loss": 2.785,
+      "step": 4960
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.0965349674224854,
+      "eval_runtime": 19.7943,
+      "eval_samples_per_second": 1608.392,
+      "eval_steps_per_second": 6.315,
+      "step": 4960
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 5.920090198516846,
+      "learning_rate": 4.728723404255319e-05,
+      "loss": 2.7468,
+      "step": 5456
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.095909833908081,
+      "eval_runtime": 19.6979,
+      "eval_samples_per_second": 1616.266,
+      "eval_steps_per_second": 6.346,
+      "step": 5456
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 6.821107864379883,
+      "learning_rate": 4.201063829787234e-05,
+      "loss": 2.7092,
+      "step": 5952
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.1043176651000977,
+      "eval_runtime": 19.7048,
+      "eval_samples_per_second": 1615.7,
+      "eval_steps_per_second": 6.344,
+      "step": 5952
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 6.687826156616211,
+      "learning_rate": 3.673404255319149e-05,
+      "loss": 2.6716,
+      "step": 6448
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.114293336868286,
+      "eval_runtime": 23.2491,
+      "eval_samples_per_second": 1369.389,
+      "eval_steps_per_second": 5.377,
+      "step": 6448
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 7.3623199462890625,
+      "learning_rate": 3.1457446808510635e-05,
+      "loss": 2.6365,
+      "step": 6944
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.1229028701782227,
+      "eval_runtime": 25.584,
+      "eval_samples_per_second": 1244.41,
+      "eval_steps_per_second": 4.886,
+      "step": 6944
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 7.889801025390625,
+      "learning_rate": 2.6180851063829785e-05,
+      "loss": 2.6025,
+      "step": 7440
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.1356382369995117,
+      "eval_runtime": 25.7301,
+      "eval_samples_per_second": 1237.345,
+      "eval_steps_per_second": 4.858,
+      "step": 7440
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 9920,
+  "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": 1.2517325438976e+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 sha256:17261dbab88026aca109ed6d3879c861b8bf0dc7b3278e1586b9e4c234807c2b
+size 5905