Upload folder using huggingface_hub

README.md

@@ -0,0 +1,68 @@
+---
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: GPT2_natural_function_53
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# GPT2_natural_function_53
+
+This model was trained from scratch on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.7165
+- Accuracy: 0.3685
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 128
+- eval_batch_size: 128
+- seed: 53
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 10000
+- num_epochs: 10
+
+### Training results
+
+| Training Loss | Epoch | Step  | Validation Loss | Accuracy |
+|:-------------:|:-----:|:-----:|:---------------:|:--------:|
+| 4.9681        | 1.0   | 6422  | 4.7441          | 0.2809   |
+| 4.3361        | 2.0   | 12844 | 4.1924          | 0.3208   |
+| 4.0681        | 3.0   | 19266 | 4.0019          | 0.3379   |
+| 3.9431        | 4.0   | 25688 | 3.9031          | 0.3468   |
+| 3.8435        | 5.0   | 32110 | 3.8400          | 0.3539   |
+| 3.7716        | 6.0   | 38532 | 3.7953          | 0.3586   |
+| 3.7115        | 7.0   | 44954 | 3.7655          | 0.3618   |
+| 3.6569        | 8.0   | 51376 | 3.7415          | 0.3647   |
+| 3.6087        | 9.0   | 57798 | 3.7231          | 0.3674   |
+| 3.5685        | 10.0  | 64220 | 3.7165          | 0.3685   |
+
+
+### Framework versions
+
+- Transformers 4.30.2
+- Pytorch 2.9.1+cu128
+- Datasets 4.1.1
+- Tokenizers 0.13.3

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.36847452642273704,
+    "eval_loss": 3.716495990753174,
+    "eval_runtime": 2.2543,
+    "eval_samples": 1178,
+    "eval_samples_per_second": 522.557,
+    "eval_steps_per_second": 4.436,
+    "perplexity": 41.120056318248146,
+    "train_loss": 4.052648230083113,
+    "train_runtime": 32739.8326,
+    "train_samples": 821994,
+    "train_samples_per_second": 251.068,
+    "train_steps_per_second": 1.962
+}

config.json

@@ -0,0 +1,40 @@
+{
+  "_name_or_path": "models/GPT2_natural_function_53",
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 2,
+  "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": 128,
+  "pad_token_id": 1,
+  "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,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.30.2",
+  "use_cache": true,
+  "vocab_size": 32768
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.36847452642273704,
+    "eval_loss": 3.716495990753174,
+    "eval_runtime": 2.2543,
+    "eval_samples": 1178,
+    "eval_samples_per_second": 522.557,
+    "eval_steps_per_second": 4.436,
+    "perplexity": 41.120056318248146
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 2,
+  "pad_token_id": 1,
+  "transformers_version": "4.30.2"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8b39a127ccb1c98a00727cff8001422fd9b5fef949b1086873eb6384b75c7d7f
+size 441329395

special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

tokenizer_config.json

@@ -0,0 +1,11 @@
+{
+  "add_prefix_space": false,
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "token": null,
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 4.052648230083113,
+    "train_runtime": 32739.8326,
+    "train_samples": 821994,
+    "train_samples_per_second": 251.068,
+    "train_steps_per_second": 1.962
+}

trainer_state.json

@@ -0,0 +1,499 @@
+{
+  "best_metric": 3.716495990753174,
+  "best_model_checkpoint": "models/GPT2_natural_function_53/checkpoint-64220",
+  "epoch": 10.0,
+  "global_step": 64220,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 1e-05,
+      "loss": 7.6082,
+      "step": 1000
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 2e-05,
+      "loss": 6.1937,
+      "step": 2000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 3e-05,
+      "loss": 5.7372,
+      "step": 3000
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4e-05,
+      "loss": 5.4128,
+      "step": 4000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 5e-05,
+      "loss": 5.1655,
+      "step": 5000
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 6e-05,
+      "loss": 4.9681,
+      "step": 6000
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.2809045091774394,
+      "eval_loss": 4.744062900543213,
+      "eval_runtime": 2.2559,
+      "eval_samples_per_second": 522.188,
+      "eval_steps_per_second": 4.433,
+      "step": 6422
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 7e-05,
+      "loss": 4.7985,
+      "step": 7000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 8e-05,
+      "loss": 4.668,
+      "step": 8000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 9e-05,
+      "loss": 4.5619,
+      "step": 9000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0001,
+      "loss": 4.4759,
+      "step": 10000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 9.815750645518259e-05,
+      "loss": 4.4017,
+      "step": 11000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 9.631316857248248e-05,
+      "loss": 4.3361,
+      "step": 12000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.32079595738138844,
+      "eval_loss": 4.192352771759033,
+      "eval_runtime": 2.2616,
+      "eval_samples_per_second": 520.877,
+      "eval_steps_per_second": 4.422,
+      "step": 12844
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 9.447067502766507e-05,
+      "loss": 4.274,
+      "step": 13000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 9.262633714496496e-05,
+      "loss": 4.1982,
+      "step": 14000
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 9.078384360014756e-05,
+      "loss": 4.1693,
+      "step": 15000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 8.893950571744743e-05,
+      "loss": 4.1416,
+      "step": 16000
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 8.709701217263003e-05,
+      "loss": 4.1148,
+      "step": 17000
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 8.525267428992992e-05,
+      "loss": 4.0924,
+      "step": 18000
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 8.341018074511251e-05,
+      "loss": 4.0681,
+      "step": 19000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.33793430744756225,
+      "eval_loss": 4.001856803894043,
+      "eval_runtime": 2.2566,
+      "eval_samples_per_second": 522.015,
+      "eval_steps_per_second": 4.431,
+      "step": 19266
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 8.15658428624124e-05,
+      "loss": 4.0058,
+      "step": 20000
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 7.972334931759499e-05,
+      "loss": 3.9829,
+      "step": 21000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 7.788085577277757e-05,
+      "loss": 3.974,
+      "step": 22000
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 7.603651789007746e-05,
+      "loss": 3.9622,
+      "step": 23000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 7.419402434526006e-05,
+      "loss": 3.9505,
+      "step": 24000
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 7.234968646255994e-05,
+      "loss": 3.9431,
+      "step": 25000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.34683769367538736,
+      "eval_loss": 3.903092384338379,
+      "eval_runtime": 2.2582,
+      "eval_samples_per_second": 521.658,
+      "eval_steps_per_second": 4.428,
+      "step": 25688
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 7.050719291774254e-05,
+      "loss": 3.9077,
+      "step": 26000
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 6.866285503504243e-05,
+      "loss": 3.8605,
+      "step": 27000
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 6.681851715234231e-05,
+      "loss": 3.8619,
+      "step": 28000
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 6.497602360752491e-05,
+      "loss": 3.8582,
+      "step": 29000
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 6.31316857248248e-05,
+      "loss": 3.8527,
+      "step": 30000
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 6.128734784212468e-05,
+      "loss": 3.8506,
+      "step": 31000
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 5.9444854297307275e-05,
+      "loss": 3.8435,
+      "step": 32000
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.35392297100383674,
+      "eval_loss": 3.8399806022644043,
+      "eval_runtime": 2.2614,
+      "eval_samples_per_second": 520.907,
+      "eval_steps_per_second": 4.422,
+      "step": 32110
+    },
+    {
+      "epoch": 5.14,
+      "learning_rate": 5.7600516414607167e-05,
+      "loss": 3.7772,
+      "step": 33000
+    },
+    {
+      "epoch": 5.29,
+      "learning_rate": 5.5756178531907044e-05,
+      "loss": 3.7753,
+      "step": 34000
+    },
+    {
+      "epoch": 5.45,
+      "learning_rate": 5.3911840649206936e-05,
+      "loss": 3.7785,
+      "step": 35000
+    },
+    {
+      "epoch": 5.61,
+      "learning_rate": 5.2069347104389535e-05,
+      "loss": 3.7769,
+      "step": 36000
+    },
+    {
+      "epoch": 5.76,
+      "learning_rate": 5.022500922168941e-05,
+      "loss": 3.7748,
+      "step": 37000
+    },
+    {
+      "epoch": 5.92,
+      "learning_rate": 4.8382515676872005e-05,
+      "loss": 3.7716,
+      "step": 38000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.35863534884964504,
+      "eval_loss": 3.7953338623046875,
+      "eval_runtime": 2.2523,
+      "eval_samples_per_second": 523.018,
+      "eval_steps_per_second": 4.44,
+      "step": 38532
+    },
+    {
+      "epoch": 6.07,
+      "learning_rate": 4.6538177794171896e-05,
+      "loss": 3.7357,
+      "step": 39000
+    },
+    {
+      "epoch": 6.23,
+      "learning_rate": 4.469568424935449e-05,
+      "loss": 3.7059,
+      "step": 40000
+    },
+    {
+      "epoch": 6.38,
+      "learning_rate": 4.285134636665437e-05,
+      "loss": 3.7108,
+      "step": 41000
+    },
+    {
+      "epoch": 6.54,
+      "learning_rate": 4.1007008483954264e-05,
+      "loss": 3.7127,
+      "step": 42000
+    },
+    {
+      "epoch": 6.7,
+      "learning_rate": 3.9164514939136856e-05,
+      "loss": 3.7119,
+      "step": 43000
+    },
+    {
+      "epoch": 6.85,
+      "learning_rate": 3.732017705643674e-05,
+      "loss": 3.7115,
+      "step": 44000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.3617635656323944,
+      "eval_loss": 3.765533924102783,
+      "eval_runtime": 2.2532,
+      "eval_samples_per_second": 522.805,
+      "eval_steps_per_second": 4.438,
+      "step": 44954
+    },
+    {
+      "epoch": 7.01,
+      "learning_rate": 3.547768351161933e-05,
+      "loss": 3.706,
+      "step": 45000
+    },
+    {
+      "epoch": 7.16,
+      "learning_rate": 3.3633345628919224e-05,
+      "loss": 3.6454,
+      "step": 46000
+    },
+    {
+      "epoch": 7.32,
+      "learning_rate": 3.178900774621911e-05,
+      "loss": 3.6521,
+      "step": 47000
+    },
+    {
+      "epoch": 7.47,
+      "learning_rate": 2.9946514201401697e-05,
+      "loss": 3.6557,
+      "step": 48000
+    },
+    {
+      "epoch": 7.63,
+      "learning_rate": 2.810217631870159e-05,
+      "loss": 3.6567,
+      "step": 49000
+    },
+    {
+      "epoch": 7.79,
+      "learning_rate": 2.6257838436001473e-05,
+      "loss": 3.6575,
+      "step": 50000
+    },
+    {
+      "epoch": 7.94,
+      "learning_rate": 2.4413500553301365e-05,
+      "loss": 3.6569,
+      "step": 51000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.36471130836998517,
+      "eval_loss": 3.741542100906372,
+      "eval_runtime": 2.2527,
+      "eval_samples_per_second": 522.938,
+      "eval_steps_per_second": 4.439,
+      "step": 51376
+    },
+    {
+      "epoch": 8.1,
+      "learning_rate": 2.2571007008483954e-05,
+      "loss": 3.6211,
+      "step": 52000
+    },
+    {
+      "epoch": 8.25,
+      "learning_rate": 2.0726669125783845e-05,
+      "loss": 3.6044,
+      "step": 53000
+    },
+    {
+      "epoch": 8.41,
+      "learning_rate": 1.8884175580966434e-05,
+      "loss": 3.6072,
+      "step": 54000
+    },
+    {
+      "epoch": 8.56,
+      "learning_rate": 1.703983769826632e-05,
+      "loss": 3.6075,
+      "step": 55000
+    },
+    {
+      "epoch": 8.72,
+      "learning_rate": 1.5195499815566213e-05,
+      "loss": 3.6107,
+      "step": 56000
+    },
+    {
+      "epoch": 8.88,
+      "learning_rate": 1.3351161932866103e-05,
+      "loss": 3.6087,
+      "step": 57000
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.36737162948010105,
+      "eval_loss": 3.723080635070801,
+      "eval_runtime": 2.2548,
+      "eval_samples_per_second": 522.439,
+      "eval_steps_per_second": 4.435,
+      "step": 57798
+    },
+    {
+      "epoch": 9.03,
+      "learning_rate": 1.1508668388048691e-05,
+      "loss": 3.5986,
+      "step": 58000
+    },
+    {
+      "epoch": 9.19,
+      "learning_rate": 9.66617484323128e-06,
+      "loss": 3.5673,
+      "step": 59000
+    },
+    {
+      "epoch": 9.34,
+      "learning_rate": 7.82183696053117e-06,
+      "loss": 3.5667,
+      "step": 60000
+    },
+    {
+      "epoch": 9.5,
+      "learning_rate": 5.9774990778310595e-06,
+      "loss": 3.567,
+      "step": 61000
+    },
+    {
+      "epoch": 9.65,
+      "learning_rate": 4.133161195130948e-06,
+      "loss": 3.5674,
+      "step": 62000
+    },
+    {
+      "epoch": 9.81,
+      "learning_rate": 2.2906676503135374e-06,
+      "loss": 3.569,
+      "step": 63000
+    },
+    {
+      "epoch": 9.97,
+      "learning_rate": 4.463297676134268e-07,
+      "loss": 3.5685,
+      "step": 64000
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.36847452642273704,
+      "eval_loss": 3.716495990753174,
+      "eval_runtime": 2.2559,
+      "eval_samples_per_second": 522.193,
+      "eval_steps_per_second": 4.433,
+      "step": 64220
+    },
+    {
+      "epoch": 10.0,
+      "step": 64220,
+      "total_flos": 5.3695120637952e+17,
+      "train_loss": 4.052648230083113,
+      "train_runtime": 32739.8326,
+      "train_samples_per_second": 251.068,
+      "train_steps_per_second": 1.962
+    }
+  ],
+  "max_steps": 64220,
+  "num_train_epochs": 10,
+  "total_flos": 5.3695120637952e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ba90b3914e9313c9f5f69726681804f660a35df74387b6eec9611fe829bdc73
+size 4881