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 sha256:8c5957f240e9b8cfece9448a6907d5a8435774572d071771548ced9d875c6c42
+size 435544704

checkpoint-8500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80e76bace2df4ed2728b0a74c6930b89df9329a223f3be285306213334d80ea1
+size 871183627

checkpoint-8500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf13b9ff62aacbaf92b823b6abeea5b8582c04edff7c93331ed141915ef860db
+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": 5445,
+  "best_metric": 3.0573878288269043,
+  "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": 27.369495391845703,
+      "learning_rate": 0.0,
+      "loss": 10.3689,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.4256974458694458,
+      "learning_rate": 9.88e-05,
+      "loss": 4.8295,
+      "step": 495
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.7844455242156982,
+      "eval_runtime": 19.6614,
+      "eval_samples_per_second": 1616.87,
+      "eval_steps_per_second": 6.358,
+      "step": 495
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 1.164453148841858,
+      "learning_rate": 9.478678038379531e-05,
+      "loss": 3.5371,
+      "step": 990
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 3.4359753131866455,
+      "eval_runtime": 19.6774,
+      "eval_samples_per_second": 1615.558,
+      "eval_steps_per_second": 6.352,
+      "step": 990
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.0340933799743652,
+      "learning_rate": 8.950959488272922e-05,
+      "loss": 3.2797,
+      "step": 1485
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 3.2933497428894043,
+      "eval_runtime": 19.6096,
+      "eval_samples_per_second": 1621.145,
+      "eval_steps_per_second": 6.374,
+      "step": 1485
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.068572998046875,
+      "learning_rate": 8.42324093816631e-05,
+      "loss": 3.1349,
+      "step": 1980
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 3.207876443862915,
+      "eval_runtime": 19.9041,
+      "eval_samples_per_second": 1597.16,
+      "eval_steps_per_second": 6.28,
+      "step": 1980
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.0857131481170654,
+      "learning_rate": 7.895522388059702e-05,
+      "loss": 3.0332,
+      "step": 2475
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 3.1527912616729736,
+      "eval_runtime": 19.6429,
+      "eval_samples_per_second": 1618.398,
+      "eval_steps_per_second": 6.364,
+      "step": 2475
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.0869224071502686,
+      "learning_rate": 7.367803837953093e-05,
+      "loss": 2.9527,
+      "step": 2970
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 3.1138641834259033,
+      "eval_runtime": 19.7113,
+      "eval_samples_per_second": 1612.783,
+      "eval_steps_per_second": 6.342,
+      "step": 2970
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.11025869846344,
+      "learning_rate": 6.840085287846483e-05,
+      "loss": 2.8841,
+      "step": 3465
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 3.091607093811035,
+      "eval_runtime": 19.6526,
+      "eval_samples_per_second": 1617.6,
+      "eval_steps_per_second": 6.36,
+      "step": 3465
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.156492829322815,
+      "learning_rate": 6.312366737739872e-05,
+      "loss": 2.824,
+      "step": 3960
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 3.074742555618286,
+      "eval_runtime": 19.9556,
+      "eval_samples_per_second": 1593.036,
+      "eval_steps_per_second": 6.264,
+      "step": 3960
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.2233030796051025,
+      "learning_rate": 5.784648187633263e-05,
+      "loss": 2.7682,
+      "step": 4455
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 3.062962293624878,
+      "eval_runtime": 19.6864,
+      "eval_samples_per_second": 1614.821,
+      "eval_steps_per_second": 6.35,
+      "step": 4455
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.3041876554489136,
+      "learning_rate": 5.256929637526653e-05,
+      "loss": 2.7163,
+      "step": 4950
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 3.059563636779785,
+      "eval_runtime": 19.6977,
+      "eval_samples_per_second": 1613.894,
+      "eval_steps_per_second": 6.346,
+      "step": 4950
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.3490029573440552,
+      "learning_rate": 4.7292110874200426e-05,
+      "loss": 2.6677,
+      "step": 5445
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 3.0573878288269043,
+      "eval_runtime": 19.6953,
+      "eval_samples_per_second": 1614.092,
+      "eval_steps_per_second": 6.347,
+      "step": 5445
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.4894585609436035,
+      "learning_rate": 4.201492537313433e-05,
+      "loss": 2.6216,
+      "step": 5940
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 3.062603712081909,
+      "eval_runtime": 19.9746,
+      "eval_samples_per_second": 1591.524,
+      "eval_steps_per_second": 6.258,
+      "step": 5940
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.5706915855407715,
+      "learning_rate": 3.673773987206823e-05,
+      "loss": 2.576,
+      "step": 6435
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 3.07055401802063,
+      "eval_runtime": 19.7413,
+      "eval_samples_per_second": 1610.326,
+      "eval_steps_per_second": 6.332,
+      "step": 6435
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.7185755968093872,
+      "learning_rate": 3.1460554371002134e-05,
+      "loss": 2.5352,
+      "step": 6930
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 3.074070453643799,
+      "eval_runtime": 19.7238,
+      "eval_samples_per_second": 1611.759,
+      "eval_steps_per_second": 6.338,
+      "step": 6930
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.6921051740646362,
+      "learning_rate": 2.6183368869936037e-05,
+      "loss": 2.4975,
+      "step": 7425
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 3.085761785507202,
+      "eval_runtime": 19.7187,
+      "eval_samples_per_second": 1612.175,
+      "eval_steps_per_second": 6.339,
+      "step": 7425
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.8211185932159424,
+      "learning_rate": 2.0906183368869936e-05,
+      "loss": 2.4642,
+      "step": 7920
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 3.093783140182495,
+      "eval_runtime": 19.6893,
+      "eval_samples_per_second": 1614.581,
+      "eval_steps_per_second": 6.349,
+      "step": 7920
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.937579870223999,
+      "learning_rate": 1.562899786780384e-05,
+      "loss": 2.435,
+      "step": 8415
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 3.103745698928833,
+      "eval_runtime": 19.6907,
+      "eval_samples_per_second": 1614.467,
+      "eval_steps_per_second": 6.348,
+      "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 sha256:3dc4060dd795aa12df36c6902992de15c9a1e2a387af1415d8a06271402cf562
+size 5905