Training in progress, step 6500, checkpoint

checkpoint-6500/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-6500/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-6500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a092a98118a1842bb246139c6fe19ad0d8946a7541fa9ff81951031245e70061
+size 435544704

checkpoint-6500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c02343a53a139432bd6967d0b950ad5b84d3153ad3b62e0c1d187e8473ba5ff7
+size 871183627

checkpoint-6500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4978b19bdbcad2009a858eff946d4b6339ba3e6bc9708ff618c62f098adeed07
+size 14709

checkpoint-6500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72ec040f7dbfafb8a0d9c346b901cc53a9760e25574232230ec90bf3c4a5f84d
+size 1383

checkpoint-6500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f7c594f77bd6d3297437260d221d19dad6dc4db8a61429fb27d5e20ef928317
+size 1465

checkpoint-6500/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-6500/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-6500/trainer_state.json

@@ -0,0 +1,281 @@
+{
+  "best_global_step": 5096,
+  "best_metric": 5.88344669342041,
+  "best_model_checkpoint": null,
+  "epoch": 16.581632653061224,
+  "eval_steps": 500,
+  "global_step": 6500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002551020408163265,
+      "grad_norm": 16.079069137573242,
+      "learning_rate": 0.0,
+      "loss": 10.3294,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.764629065990448,
+      "learning_rate": 7.82e-05,
+      "loss": 7.1796,
+      "step": 392
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 6.494222640991211,
+      "eval_runtime": 19.3292,
+      "eval_samples_per_second": 1303.155,
+      "eval_steps_per_second": 5.122,
+      "step": 392
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.7356462478637695,
+      "learning_rate": 9.613387978142077e-05,
+      "loss": 6.3038,
+      "step": 784
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 6.264538764953613,
+      "eval_runtime": 19.2699,
+      "eval_samples_per_second": 1307.171,
+      "eval_steps_per_second": 5.138,
+      "step": 784
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.7015138864517212,
+      "learning_rate": 9.077868852459017e-05,
+      "loss": 6.1364,
+      "step": 1176
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 6.156426906585693,
+      "eval_runtime": 19.3215,
+      "eval_samples_per_second": 1303.677,
+      "eval_steps_per_second": 5.124,
+      "step": 1176
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6908867955207825,
+      "learning_rate": 8.542349726775956e-05,
+      "loss": 6.0346,
+      "step": 1568
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 6.085503101348877,
+      "eval_runtime": 19.2431,
+      "eval_samples_per_second": 1308.99,
+      "eval_steps_per_second": 5.145,
+      "step": 1568
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.6976261138916016,
+      "learning_rate": 8.006830601092896e-05,
+      "loss": 5.9541,
+      "step": 1960
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 6.033597946166992,
+      "eval_runtime": 19.369,
+      "eval_samples_per_second": 1300.483,
+      "eval_steps_per_second": 5.111,
+      "step": 1960
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.7718036770820618,
+      "learning_rate": 7.471311475409837e-05,
+      "loss": 5.8851,
+      "step": 2352
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 5.988277435302734,
+      "eval_runtime": 19.3652,
+      "eval_samples_per_second": 1300.733,
+      "eval_steps_per_second": 5.112,
+      "step": 2352
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.793295681476593,
+      "learning_rate": 6.935792349726776e-05,
+      "loss": 5.8221,
+      "step": 2744
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 5.95566463470459,
+      "eval_runtime": 19.3512,
+      "eval_samples_per_second": 1301.674,
+      "eval_steps_per_second": 5.116,
+      "step": 2744
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.855619490146637,
+      "learning_rate": 6.400273224043716e-05,
+      "loss": 5.762,
+      "step": 3136
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 5.929388999938965,
+      "eval_runtime": 19.3691,
+      "eval_samples_per_second": 1300.472,
+      "eval_steps_per_second": 5.111,
+      "step": 3136
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.9132499098777771,
+      "learning_rate": 5.864754098360656e-05,
+      "loss": 5.7039,
+      "step": 3528
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 5.909368515014648,
+      "eval_runtime": 19.4001,
+      "eval_samples_per_second": 1298.394,
+      "eval_steps_per_second": 5.103,
+      "step": 3528
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.0017693042755127,
+      "learning_rate": 5.329234972677596e-05,
+      "loss": 5.6469,
+      "step": 3920
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 5.896821975708008,
+      "eval_runtime": 19.3535,
+      "eval_samples_per_second": 1301.52,
+      "eval_steps_per_second": 5.115,
+      "step": 3920
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 1.0763376951217651,
+      "learning_rate": 4.793715846994536e-05,
+      "loss": 5.5915,
+      "step": 4312
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 5.8890886306762695,
+      "eval_runtime": 19.3667,
+      "eval_samples_per_second": 1300.637,
+      "eval_steps_per_second": 5.112,
+      "step": 4312
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.1510194540023804,
+      "learning_rate": 4.258196721311475e-05,
+      "loss": 5.5371,
+      "step": 4704
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 5.884430408477783,
+      "eval_runtime": 19.5907,
+      "eval_samples_per_second": 1285.764,
+      "eval_steps_per_second": 5.053,
+      "step": 4704
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.235405445098877,
+      "learning_rate": 3.722677595628416e-05,
+      "loss": 5.4849,
+      "step": 5096
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 5.88344669342041,
+      "eval_runtime": 19.4202,
+      "eval_samples_per_second": 1297.051,
+      "eval_steps_per_second": 5.098,
+      "step": 5096
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.351965308189392,
+      "learning_rate": 3.1871584699453554e-05,
+      "loss": 5.4357,
+      "step": 5488
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 5.889012336730957,
+      "eval_runtime": 19.3667,
+      "eval_samples_per_second": 1300.637,
+      "eval_steps_per_second": 5.112,
+      "step": 5488
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.4173115491867065,
+      "learning_rate": 2.651639344262295e-05,
+      "loss": 5.3896,
+      "step": 5880
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 5.89474630355835,
+      "eval_runtime": 19.3877,
+      "eval_samples_per_second": 1299.227,
+      "eval_steps_per_second": 5.106,
+      "step": 5880
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.5183329582214355,
+      "learning_rate": 2.116120218579235e-05,
+      "loss": 5.3467,
+      "step": 6272
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 5.901671409606934,
+      "eval_runtime": 19.3699,
+      "eval_samples_per_second": 1300.419,
+      "eval_steps_per_second": 5.111,
+      "step": 6272
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7840,
+  "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": 1.08541755260928e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-6500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d2519aa35d901a447f042811330b55309aa805a1585ea631588d8524576d18a2
+size 5905