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 sha256:6d5a4a316db8c3d7446f7f37f85482dc27ce25bde07759dc18a1a106637b3419
+size 435544704

checkpoint-8500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c000eda4e65b965ed436e3d1a73c946930a64f4a485cc9cdbb6c77e4b60aa58
+size 871183627

checkpoint-8500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d5e91b4001791fd7e39b756210e379d76dbfd9e2e0c8e3583afd040802e2d7d
+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": 4455,
+  "best_metric": 2.70760178565979,
+  "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": 29.966644287109375,
+      "learning_rate": 0.0,
+      "loss": 10.3432,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.0022364854812622,
+      "learning_rate": 9.88e-05,
+      "loss": 4.4821,
+      "step": 495
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 3.2837753295898438,
+      "eval_runtime": 24.5317,
+      "eval_samples_per_second": 1297.138,
+      "eval_steps_per_second": 5.095,
+      "step": 495
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8058988451957703,
+      "learning_rate": 9.478678038379531e-05,
+      "loss": 3.0363,
+      "step": 990
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 2.975402355194092,
+      "eval_runtime": 24.6895,
+      "eval_samples_per_second": 1288.849,
+      "eval_steps_per_second": 5.063,
+      "step": 990
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.6841588616371155,
+      "learning_rate": 8.950959488272922e-05,
+      "loss": 2.8089,
+      "step": 1485
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 2.8664374351501465,
+      "eval_runtime": 24.5488,
+      "eval_samples_per_second": 1296.234,
+      "eval_steps_per_second": 5.092,
+      "step": 1485
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.6922294497489929,
+      "learning_rate": 8.42324093816631e-05,
+      "loss": 2.6892,
+      "step": 1980
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 2.8051071166992188,
+      "eval_runtime": 24.9233,
+      "eval_samples_per_second": 1276.756,
+      "eval_steps_per_second": 5.015,
+      "step": 1980
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7276241183280945,
+      "learning_rate": 7.895522388059702e-05,
+      "loss": 2.6067,
+      "step": 2475
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 2.7664873600006104,
+      "eval_runtime": 24.5368,
+      "eval_samples_per_second": 1296.867,
+      "eval_steps_per_second": 5.094,
+      "step": 2475
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.7834751605987549,
+      "learning_rate": 7.367803837953093e-05,
+      "loss": 2.5413,
+      "step": 2970
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 2.7397351264953613,
+      "eval_runtime": 24.467,
+      "eval_samples_per_second": 1300.566,
+      "eval_steps_per_second": 5.109,
+      "step": 2970
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.7579982280731201,
+      "learning_rate": 6.840085287846483e-05,
+      "loss": 2.4849,
+      "step": 3465
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 2.724043130874634,
+      "eval_runtime": 24.442,
+      "eval_samples_per_second": 1301.899,
+      "eval_steps_per_second": 5.114,
+      "step": 3465
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.8136159777641296,
+      "learning_rate": 6.312366737739872e-05,
+      "loss": 2.4344,
+      "step": 3960
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 2.714533567428589,
+      "eval_runtime": 24.6731,
+      "eval_samples_per_second": 1289.702,
+      "eval_steps_per_second": 5.066,
+      "step": 3960
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.8327829837799072,
+      "learning_rate": 5.784648187633263e-05,
+      "loss": 2.3871,
+      "step": 4455
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 2.70760178565979,
+      "eval_runtime": 24.7482,
+      "eval_samples_per_second": 1285.791,
+      "eval_steps_per_second": 5.051,
+      "step": 4455
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.8686661124229431,
+      "learning_rate": 5.256929637526653e-05,
+      "loss": 2.3423,
+      "step": 4950
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 2.708085536956787,
+      "eval_runtime": 24.409,
+      "eval_samples_per_second": 1303.659,
+      "eval_steps_per_second": 5.121,
+      "step": 4950
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.9367507100105286,
+      "learning_rate": 4.7292110874200426e-05,
+      "loss": 2.2993,
+      "step": 5445
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 2.7078373432159424,
+      "eval_runtime": 24.4382,
+      "eval_samples_per_second": 1302.101,
+      "eval_steps_per_second": 5.115,
+      "step": 5445
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 1.0120211839675903,
+      "learning_rate": 4.201492537313433e-05,
+      "loss": 2.2581,
+      "step": 5940
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 2.713834047317505,
+      "eval_runtime": 24.7104,
+      "eval_samples_per_second": 1287.758,
+      "eval_steps_per_second": 5.059,
+      "step": 5940
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 1.062059760093689,
+      "learning_rate": 3.673773987206823e-05,
+      "loss": 2.2187,
+      "step": 6435
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 2.722933053970337,
+      "eval_runtime": 24.479,
+      "eval_samples_per_second": 1299.931,
+      "eval_steps_per_second": 5.106,
+      "step": 6435
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 1.1194597482681274,
+      "learning_rate": 3.1460554371002134e-05,
+      "loss": 2.1811,
+      "step": 6930
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 2.7325448989868164,
+      "eval_runtime": 24.4339,
+      "eval_samples_per_second": 1302.33,
+      "eval_steps_per_second": 5.116,
+      "step": 6930
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 1.1889002323150635,
+      "learning_rate": 2.6183368869936037e-05,
+      "loss": 2.1447,
+      "step": 7425
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 2.741835594177246,
+      "eval_runtime": 24.5268,
+      "eval_samples_per_second": 1297.395,
+      "eval_steps_per_second": 5.096,
+      "step": 7425
+    },
+    {
+      "epoch": 16.0,
+      "grad_norm": 1.2749454975128174,
+      "learning_rate": 2.0906183368869936e-05,
+      "loss": 2.1127,
+      "step": 7920
+    },
+    {
+      "epoch": 16.0,
+      "eval_loss": 2.7535629272460938,
+      "eval_runtime": 24.6402,
+      "eval_samples_per_second": 1291.424,
+      "eval_steps_per_second": 5.073,
+      "step": 7920
+    },
+    {
+      "epoch": 17.0,
+      "grad_norm": 1.3005127906799316,
+      "learning_rate": 1.562899786780384e-05,
+      "loss": 2.0843,
+      "step": 8415
+    },
+    {
+      "epoch": 17.0,
+      "eval_loss": 2.761566638946533,
+      "eval_runtime": 24.4821,
+      "eval_samples_per_second": 1299.769,
+      "eval_steps_per_second": 5.106,
+      "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": 1.4208178839552e+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 sha256:139ca21c6338347f7f8337e2425636928e5a2955437ad042fe0e226309b09e4d
+size 5969