Training in progress, step 7820, checkpoint

checkpoint-7820/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-7820/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-7820/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e24c644a7b50863105338236f38faf2c1d9af0d978edd38c0778b51f8446cf94
+size 435544704

checkpoint-7820/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2a4a17663328dd6b134cf866d30b0edaa924adb27174c2cf8a7183fe222f796
+size 871183627

checkpoint-7820/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2a06448450d16d1680b97bcbef74839fe0a7c2396f743a30117d733721349e5
+size 14709

checkpoint-7820/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f85d4bdba50d945e851aeab4d6c35670010d85c51eadcea7ad7fc231827fd2d
+size 1383

checkpoint-7820/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2228ca91d60f4d5fbaa26237c5737b472ba097cd903e9a7edb7eca80a503d063
+size 1465

checkpoint-7820/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-7820/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-7820/trainer_state.json

@@ -0,0 +1,326 @@
+{
+  "best_global_step": 5865,
+  "best_metric": 4.335202217102051,
+  "best_model_checkpoint": null,
+  "epoch": 20.0,
+  "eval_steps": 500,
+  "global_step": 7820,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0025575447570332483,
+      "grad_norm": 15.271883964538574,
+      "learning_rate": 0.0,
+      "loss": 10.3934,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.1213458776474,
+      "learning_rate": 7.800000000000001e-05,
+      "loss": 6.8393,
+      "step": 391
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 5.744714736938477,
+      "eval_runtime": 20.5001,
+      "eval_samples_per_second": 1226.336,
+      "eval_steps_per_second": 4.829,
+      "step": 391
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.9541754722595215,
+      "learning_rate": 9.615068493150685e-05,
+      "loss": 5.3976,
+      "step": 782
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 5.181271553039551,
+      "eval_runtime": 20.857,
+      "eval_samples_per_second": 1205.353,
+      "eval_steps_per_second": 4.747,
+      "step": 782
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.9359272122383118,
+      "learning_rate": 9.07945205479452e-05,
+      "loss": 5.0032,
+      "step": 1173
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.929495811462402,
+      "eval_runtime": 20.6986,
+      "eval_samples_per_second": 1214.573,
+      "eval_steps_per_second": 4.783,
+      "step": 1173
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.9098637700080872,
+      "learning_rate": 8.543835616438357e-05,
+      "loss": 4.7792,
+      "step": 1564
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.771158218383789,
+      "eval_runtime": 20.4596,
+      "eval_samples_per_second": 1228.764,
+      "eval_steps_per_second": 4.839,
+      "step": 1564
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 1.0112477540969849,
+      "learning_rate": 8.008219178082192e-05,
+      "loss": 4.6136,
+      "step": 1955
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.65746545791626,
+      "eval_runtime": 20.5977,
+      "eval_samples_per_second": 1220.527,
+      "eval_steps_per_second": 4.806,
+      "step": 1955
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.0180248022079468,
+      "learning_rate": 7.472602739726028e-05,
+      "loss": 4.4794,
+      "step": 2346
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.568690776824951,
+      "eval_runtime": 20.7635,
+      "eval_samples_per_second": 1210.777,
+      "eval_steps_per_second": 4.768,
+      "step": 2346
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 1.044142723083496,
+      "learning_rate": 6.936986301369864e-05,
+      "loss": 4.3645,
+      "step": 2737
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.503849506378174,
+      "eval_runtime": 21.1424,
+      "eval_samples_per_second": 1189.08,
+      "eval_steps_per_second": 4.683,
+      "step": 2737
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.0976617336273193,
+      "learning_rate": 6.401369863013699e-05,
+      "loss": 4.2636,
+      "step": 3128
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.4509758949279785,
+      "eval_runtime": 20.228,
+      "eval_samples_per_second": 1242.831,
+      "eval_steps_per_second": 4.894,
+      "step": 3128
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 1.1272411346435547,
+      "learning_rate": 5.865753424657534e-05,
+      "loss": 4.173,
+      "step": 3519
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.41462516784668,
+      "eval_runtime": 20.362,
+      "eval_samples_per_second": 1234.653,
+      "eval_steps_per_second": 4.862,
+      "step": 3519
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.204750418663025,
+      "learning_rate": 5.330136986301371e-05,
+      "loss": 4.09,
+      "step": 3910
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.385906219482422,
+      "eval_runtime": 20.3177,
+      "eval_samples_per_second": 1237.347,
+      "eval_steps_per_second": 4.873,
+      "step": 3910
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.3081825971603394,
+      "learning_rate": 4.794520547945205e-05,
+      "loss": 4.016,
+      "step": 4301
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.366354465484619,
+      "eval_runtime": 20.4931,
+      "eval_samples_per_second": 1226.757,
+      "eval_steps_per_second": 4.831,
+      "step": 4301
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.3272957801818848,
+      "learning_rate": 4.258904109589041e-05,
+      "loss": 3.9493,
+      "step": 4692
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.352566242218018,
+      "eval_runtime": 20.7094,
+      "eval_samples_per_second": 1213.941,
+      "eval_steps_per_second": 4.78,
+      "step": 4692
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.4078395366668701,
+      "learning_rate": 3.723287671232877e-05,
+      "loss": 3.8894,
+      "step": 5083
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.34254789352417,
+      "eval_runtime": 20.5627,
+      "eval_samples_per_second": 1222.599,
+      "eval_steps_per_second": 4.815,
+      "step": 5083
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.4349329471588135,
+      "learning_rate": 3.187671232876712e-05,
+      "loss": 3.8358,
+      "step": 5474
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.337207317352295,
+      "eval_runtime": 20.7809,
+      "eval_samples_per_second": 1209.764,
+      "eval_steps_per_second": 4.764,
+      "step": 5474
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.5260998010635376,
+      "learning_rate": 2.6520547945205482e-05,
+      "loss": 3.7876,
+      "step": 5865
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.335202217102051,
+      "eval_runtime": 19.2842,
+      "eval_samples_per_second": 1303.66,
+      "eval_steps_per_second": 5.134,
+      "step": 5865
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.5741251707077026,
+      "learning_rate": 2.1164383561643834e-05,
+      "loss": 3.7442,
+      "step": 6256
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 4.335844993591309,
+      "eval_runtime": 19.3387,
+      "eval_samples_per_second": 1299.984,
+      "eval_steps_per_second": 5.119,
+      "step": 6256
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.6043676137924194,
+      "learning_rate": 1.5808219178082193e-05,
+      "loss": 3.7063,
+      "step": 6647
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 4.336172103881836,
+      "eval_runtime": 19.2842,
+      "eval_samples_per_second": 1303.657,
+      "eval_steps_per_second": 5.134,
+      "step": 6647
+    },
+    {
+      "epoch": 18.0,
+      "grad_norm": 1.656690239906311,
+      "learning_rate": 1.0452054794520549e-05,
+      "loss": 3.6736,
+      "step": 7038
+    },
+    {
+      "epoch": 18.0,
+      "eval_loss": 4.337766170501709,
+      "eval_runtime": 19.3112,
+      "eval_samples_per_second": 1301.835,
+      "eval_steps_per_second": 5.127,
+      "step": 7038
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 1.678371787071228,
+      "learning_rate": 5.095890410958904e-06,
+      "loss": 3.6464,
+      "step": 7429
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 4.340496063232422,
+      "eval_runtime": 19.2219,
+      "eval_samples_per_second": 1307.884,
+      "eval_steps_per_second": 5.15,
+      "step": 7429
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7820,
+  "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": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.3068782272512e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-7820/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6887ebfa76ed33761cecde1e93c6f3a1b50fe97fa9045d3d52e9d218d4faf62
+size 5905