craa commited on 2 days ago

Commit

fdefd91

verified ·

1 Parent(s): 96e6e00

Add files using upload-large-folder tool

Browse files

This view is limited to 50 files because it contains too many changes. See raw diff

Files changed (50) hide show

exp2b/swap_0.3_cost_to_hit_40817/generation_config.json +6 -0
exp2b/swap_0.3_cost_to_hit_40817/merges.txt +0 -0
exp2b/swap_0.3_cost_to_hit_40817/special_tokens_map.json +5 -0
exp2b/swap_0.3_cost_to_hit_40817/tokenizer.json +0 -0
exp2b/swap_0.3_cost_to_hit_40817/tokenizer_config.json +20 -0
exp2b/swap_0.3_cost_to_hit_40817/train_results.json +9 -0
exp2b/swap_0.3_cost_to_hit_40817/trainer_state.json +0 -0
exp2b/swap_0.3_cost_to_hit_40817/vocab.json +0 -0
exp2b/swap_0.3_cost_to_hit_5039/README.md +155 -0
exp2b/swap_0.3_cost_to_hit_5039/all_results.json +16 -0
exp2b/swap_0.3_cost_to_hit_5039/config.json +31 -0
exp2b/swap_0.3_cost_to_hit_5039/eval_results.json +10 -0
exp2b/swap_0.3_cost_to_hit_5039/generation_config.json +6 -0
exp2b/swap_0.3_cost_to_hit_5039/merges.txt +0 -0
exp2b/swap_0.3_cost_to_hit_5039/special_tokens_map.json +5 -0
exp2b/swap_0.3_cost_to_hit_5039/tokenizer.json +0 -0
exp2b/swap_0.3_cost_to_hit_5039/tokenizer_config.json +20 -0
exp2b/swap_0.3_cost_to_hit_5039/train_results.json +9 -0
exp2b/swap_0.3_cost_to_hit_5039/trainer_state.json +0 -0
exp2b/swap_0.3_cost_to_hit_5039/vocab.json +0 -0
exp2b/swap_0.3_cost_to_push_1032/README.md +155 -0
exp2b/swap_0.3_cost_to_push_1032/all_results.json +16 -0
exp2b/swap_0.3_cost_to_push_1032/config.json +31 -0
exp2b/swap_0.3_cost_to_push_1032/eval_results.json +10 -0
exp2b/swap_0.3_cost_to_push_1032/generation_config.json +6 -0
exp2b/swap_0.3_cost_to_push_1032/merges.txt +0 -0
exp2b/swap_0.3_cost_to_push_1032/special_tokens_map.json +5 -0
exp2b/swap_0.3_cost_to_push_1032/tokenizer.json +0 -0
exp2b/swap_0.3_cost_to_push_1032/tokenizer_config.json +20 -0
exp2b/swap_0.3_cost_to_push_1032/train_results.json +9 -0
exp2b/swap_0.3_cost_to_push_1032/trainer_state.json +0 -0
exp2b/swap_0.3_cost_to_push_1032/vocab.json +0 -0
exp2b/swap_0.3_cost_to_push_2128/README.md +179 -0
exp2b/swap_0.3_cost_to_push_2128/all_results.json +16 -0
exp2b/swap_0.3_cost_to_push_2128/config.json +31 -0
exp2b/swap_0.3_cost_to_push_2128/eval_results.json +10 -0
exp2b/swap_0.3_cost_to_push_2128/generation_config.json +6 -0
exp2b/swap_0.3_cost_to_push_2128/merges.txt +0 -0
exp2b/swap_0.3_cost_to_push_2128/special_tokens_map.json +5 -0
exp2b/swap_0.3_cost_to_push_2128/tokenizer.json +0 -0
exp2b/swap_0.3_cost_to_push_2128/tokenizer_config.json +20 -0
exp2b/swap_0.3_cost_to_push_2128/train_results.json +9 -0
exp2b/swap_0.3_cost_to_push_2128/trainer_state.json +0 -0
exp2b/swap_0.3_cost_to_push_2128/vocab.json +0 -0
exp2b/swap_0.3_cost_to_push_3591/README.md +163 -0
exp2b/swap_0.3_cost_to_push_3591/all_results.json +16 -0
exp2b/swap_0.3_cost_to_push_3591/config.json +31 -0
exp2b/swap_0.3_cost_to_push_3591/eval_results.json +10 -0
exp2b/swap_0.3_cost_to_push_3591/generation_config.json +6 -0
exp2b/swap_0.3_cost_to_push_3591/merges.txt +0 -0

exp2b/swap_0.3_cost_to_hit_40817/generation_config.json ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.55.2"
+}

exp2b/swap_0.3_cost_to_hit_40817/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_40817/special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,5 @@

+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_hit_40817/tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_40817/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,20 @@

+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1024,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_hit_40817/train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 26.82215743440233,
+    "total_flos": 1.92300745162752e+18,
+    "train_loss": 3.417878519141156,
+    "train_runtime": 41227.343,
+    "train_samples": 274385,
+    "train_samples_per_second": 332.771,
+    "train_steps_per_second": 4.16
+}

exp2b/swap_0.3_cost_to_hit_40817/trainer_state.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_40817/vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_5039/README.md ADDED Viewed

	@@ -0,0 +1,155 @@

+---
+library_name: transformers
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: exceptions_exp2_swap_0.3_cost_to_hit_5039
+  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. -->
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/craaaa/exceptions_exp2/runs/tgv44bza)
+# exceptions_exp2_swap_0.3_cost_to_hit_5039
+This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.5668
+- Accuracy: 0.3684
+## 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: 0.0006
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 5039
+- gradient_accumulation_steps: 5
+- total_train_batch_size: 80
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 100
+- num_epochs: 50.0
+- mixed_precision_training: Native AMP
+### Training results
+| Training Loss | Epoch   | Step  | Validation Loss | Accuracy |
+|:-------------:|:-------:|:-----:|:---------------:|:--------:|
+| 4.8309        | 0.2915  | 1000  | 4.7499          | 0.2556   |
+| 4.345         | 0.5831  | 2000  | 4.2875          | 0.2987   |
+| 4.1414        | 0.8746  | 3000  | 4.1075          | 0.3145   |
+| 4.0049        | 1.1662  | 4000  | 3.9990          | 0.3238   |
+| 3.9449        | 1.4577  | 5000  | 3.9233          | 0.3306   |
+| 3.8806        | 1.7493  | 6000  | 3.8647          | 0.3361   |
+| 3.7589        | 2.0408  | 7000  | 3.8245          | 0.3401   |
+| 3.7639        | 2.3324  | 8000  | 3.7924          | 0.3430   |
+| 3.7451        | 2.6239  | 9000  | 3.7635          | 0.3455   |
+| 3.7298        | 2.9155  | 10000 | 3.7375          | 0.3483   |
+| 3.6453        | 3.2070  | 11000 | 3.7227          | 0.3503   |
+| 3.6458        | 3.4985  | 12000 | 3.7057          | 0.3519   |
+| 3.6488        | 3.7901  | 13000 | 3.6875          | 0.3536   |
+| 3.5361        | 4.0816  | 14000 | 3.6784          | 0.3549   |
+| 3.5927        | 4.3732  | 15000 | 3.6679          | 0.3560   |
+| 3.5756        | 4.6647  | 16000 | 3.6559          | 0.3574   |
+| 3.586         | 4.9563  | 17000 | 3.6427          | 0.3583   |
+| 3.5027        | 5.2478  | 18000 | 3.6450          | 0.3590   |
+| 3.5385        | 5.5394  | 19000 | 3.6337          | 0.3595   |
+| 3.552         | 5.8309  | 20000 | 3.6221          | 0.3610   |
+| 3.4485        | 6.1224  | 21000 | 3.6269          | 0.3612   |
+| 3.4794        | 6.4140  | 22000 | 3.6180          | 0.3620   |
+| 3.4948        | 6.7055  | 23000 | 3.6064          | 0.3628   |
+| 3.5001        | 6.9971  | 24000 | 3.5977          | 0.3636   |
+| 3.4314        | 7.2886  | 25000 | 3.6059          | 0.3636   |
+| 3.4464        | 7.5802  | 26000 | 3.5965          | 0.3642   |
+| 3.4776        | 7.8717  | 27000 | 3.5903          | 0.3648   |
+| 3.3921        | 8.1633  | 28000 | 3.5977          | 0.3646   |
+| 3.4316        | 8.4548  | 29000 | 3.5908          | 0.3656   |
+| 3.4402        | 8.7464  | 30000 | 3.5824          | 0.3657   |
+| 3.3367        | 9.0379  | 31000 | 3.5883          | 0.3659   |
+| 3.3894        | 9.3294  | 32000 | 3.5854          | 0.3664   |
+| 3.3974        | 9.6210  | 33000 | 3.5767          | 0.3670   |
+| 3.4123        | 9.9125  | 34000 | 3.5709          | 0.3676   |
+| 3.3524        | 10.2041 | 35000 | 3.5815          | 0.3669   |
+| 3.3772        | 10.4956 | 36000 | 3.5750          | 0.3675   |
+| 3.3878        | 10.7872 | 37000 | 3.5677          | 0.3682   |
+| 3.298         | 11.0787 | 38000 | 3.5755          | 0.3679   |
+| 3.3453        | 11.3703 | 39000 | 3.5732          | 0.3682   |
+| 3.3751        | 11.6618 | 40000 | 3.5668          | 0.3684   |
+| 3.3926        | 11.9534 | 41000 | 3.5553          | 0.3693   |
+| 3.3128        | 12.2449 | 42000 | 3.5722          | 0.3685   |
+| 3.3311        | 12.5364 | 43000 | 3.5631          | 0.3690   |
+| 3.3599        | 12.8280 | 44000 | 3.5573          | 0.3695   |
+| 3.2659        | 13.1195 | 45000 | 3.5699          | 0.3693   |
+| 3.303         | 13.4111 | 46000 | 3.5665          | 0.3693   |
+| 3.3408        | 13.7026 | 47000 | 3.5560          | 0.3699   |
+| 3.3585        | 13.9942 | 48000 | 3.5476          | 0.3704   |
+| 3.2907        | 14.2857 | 49000 | 3.5647          | 0.3696   |
+| 3.3178        | 14.5773 | 50000 | 3.5559          | 0.3704   |
+| 3.3274        | 14.8688 | 51000 | 3.5503          | 0.3708   |
+| 3.2627        | 15.1603 | 52000 | 3.5646          | 0.3702   |
+| 3.2929        | 15.4519 | 53000 | 3.5534          | 0.3709   |
+| 3.3045        | 15.7434 | 54000 | 3.5493          | 0.3709   |
+| 3.2083        | 16.0350 | 55000 | 3.5629          | 0.3707   |
+| 3.2545        | 16.3265 | 56000 | 3.5611          | 0.3708   |
+| 3.2819        | 16.6181 | 57000 | 3.5491          | 0.3711   |
+| 3.3014        | 16.9096 | 58000 | 3.5422          | 0.3719   |
+| 3.239         | 17.2012 | 59000 | 3.5569          | 0.3712   |
+| 3.2663        | 17.4927 | 60000 | 3.5540          | 0.3714   |
+| 3.291         | 17.7843 | 61000 | 3.5455          | 0.3717   |
+| 3.2072        | 18.0758 | 62000 | 3.5595          | 0.3713   |
+| 3.244         | 18.3673 | 63000 | 3.5549          | 0.3714   |
+| 3.2755        | 18.6589 | 64000 | 3.5482          | 0.3721   |
+| 3.2776        | 18.9504 | 65000 | 3.5413          | 0.3722   |
+| 3.2227        | 19.2420 | 66000 | 3.5554          | 0.3717   |
+| 3.2532        | 19.5335 | 67000 | 3.5502          | 0.3718   |
+| 3.2725        | 19.8251 | 68000 | 3.5407          | 0.3726   |
+| 3.1924        | 20.1166 | 69000 | 3.5623          | 0.3718   |
+| 3.2142        | 20.4082 | 70000 | 3.5519          | 0.3722   |
+| 3.2487        | 20.6997 | 71000 | 3.5452          | 0.3725   |
+| 3.2701        | 20.9913 | 72000 | 3.5339          | 0.3730   |
+| 3.2145        | 21.2828 | 73000 | 3.5540          | 0.3723   |
+| 3.2264        | 21.5743 | 74000 | 3.5490          | 0.3725   |
+| 3.2383        | 21.8659 | 75000 | 3.5421          | 0.3727   |
+| 3.1783        | 22.1574 | 76000 | 3.5560          | 0.3726   |
+| 3.2161        | 22.4490 | 77000 | 3.5500          | 0.3727   |
+| 3.2258        | 22.7405 | 78000 | 3.5388          | 0.3734   |
+| 3.1377        | 23.0321 | 79000 | 3.5555          | 0.3725   |
+| 3.1801        | 23.3236 | 80000 | 3.5525          | 0.3727   |
+| 3.2211        | 23.6152 | 81000 | 3.5471          | 0.3731   |
+| 3.2173        | 23.9067 | 82000 | 3.5409          | 0.3732   |
+| 3.1689        | 24.1983 | 83000 | 3.5581          | 0.3727   |
+| 3.1936        | 24.4898 | 84000 | 3.5505          | 0.3731   |
+| 3.222         | 24.7813 | 85000 | 3.5415          | 0.3735   |
+| 3.1266        | 25.0729 | 86000 | 3.5552          | 0.3730   |
+| 3.1636        | 25.3644 | 87000 | 3.5557          | 0.3728   |
+| 3.1903        | 25.6560 | 88000 | 3.5496          | 0.3731   |
+| 3.2143        | 25.9475 | 89000 | 3.5409          | 0.3737   |
+| 3.1427        | 26.2391 | 90000 | 3.5571          | 0.3730   |
+| 3.1789        | 26.5306 | 91000 | 3.5457          | 0.3734   |
+| 3.1941        | 26.8222 | 92000 | 3.5442          | 0.3737   |
+### Framework versions
+- Transformers 4.55.2
+- Pytorch 2.8.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.21.4

exp2b/swap_0.3_cost_to_hit_5039/all_results.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+    "epoch": 26.82215743440233,
+    "eval_accuracy": 0.3683566367460562,
+    "eval_loss": 3.566842794418335,
+    "eval_runtime": 53.8293,
+    "eval_samples": 16613,
+    "eval_samples_per_second": 308.624,
+    "eval_steps_per_second": 19.302,
+    "perplexity": 35.40463679310635,
+    "total_flos": 1.92300745162752e+18,
+    "train_loss": 3.4185851040715756,
+    "train_runtime": 40932.283,
+    "train_samples": 274385,
+    "train_samples_per_second": 335.169,
+    "train_steps_per_second": 4.19
+}

exp2b/swap_0.3_cost_to_hit_5039/config.json ADDED Viewed

	@@ -0,0 +1,31 @@

+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "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,
+  "torch_dtype": "float32",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "vocab_size": 50257
+}

exp2b/swap_0.3_cost_to_hit_5039/eval_results.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+    "epoch": 26.82215743440233,
+    "eval_accuracy": 0.3683566367460562,
+    "eval_loss": 3.566842794418335,
+    "eval_runtime": 53.8293,
+    "eval_samples": 16613,
+    "eval_samples_per_second": 308.624,
+    "eval_steps_per_second": 19.302,
+    "perplexity": 35.40463679310635
+}

exp2b/swap_0.3_cost_to_hit_5039/generation_config.json ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.55.2"
+}

exp2b/swap_0.3_cost_to_hit_5039/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_5039/special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,5 @@

+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_hit_5039/tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_5039/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,20 @@

+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1024,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_hit_5039/train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 26.82215743440233,
+    "total_flos": 1.92300745162752e+18,
+    "train_loss": 3.4185851040715756,
+    "train_runtime": 40932.283,
+    "train_samples": 274385,
+    "train_samples_per_second": 335.169,
+    "train_steps_per_second": 4.19
+}

exp2b/swap_0.3_cost_to_hit_5039/trainer_state.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_hit_5039/vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_1032/README.md ADDED Viewed

	@@ -0,0 +1,155 @@

+---
+library_name: transformers
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: exceptions_exp2_swap_0.3_cost_to_push_1032
+  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. -->
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/craaaa/exceptions_exp2/runs/01h6wav8)
+# exceptions_exp2_swap_0.3_cost_to_push_1032
+This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.5841
+- Accuracy: 0.3657
+## 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: 0.0006
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 1032
+- gradient_accumulation_steps: 5
+- total_train_batch_size: 80
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 100
+- num_epochs: 50.0
+- mixed_precision_training: Native AMP
+### Training results
+| Training Loss | Epoch   | Step  | Validation Loss | Accuracy |
+|:-------------:|:-------:|:-----:|:---------------:|:--------:|
+| 4.8321        | 0.2915  | 1000  | 4.7518          | 0.2550   |
+| 4.3375        | 0.5831  | 2000  | 4.2919          | 0.2988   |
+| 4.1644        | 0.8746  | 3000  | 4.1024          | 0.3143   |
+| 4.003         | 1.1662  | 4000  | 3.9976          | 0.3236   |
+| 3.9444        | 1.4577  | 5000  | 3.9255          | 0.3304   |
+| 3.8926        | 1.7493  | 6000  | 3.8643          | 0.3358   |
+| 3.7451        | 2.0408  | 7000  | 3.8239          | 0.3399   |
+| 3.7615        | 2.3324  | 8000  | 3.7920          | 0.3434   |
+| 3.7452        | 2.6239  | 9000  | 3.7633          | 0.3458   |
+| 3.73          | 2.9155  | 10000 | 3.7365          | 0.3483   |
+| 3.6536        | 3.2070  | 11000 | 3.7235          | 0.3500   |
+| 3.6567        | 3.4985  | 12000 | 3.7064          | 0.3522   |
+| 3.649         | 3.7901  | 13000 | 3.6864          | 0.3535   |
+| 3.5482        | 4.0816  | 14000 | 3.6802          | 0.3547   |
+| 3.5743        | 4.3732  | 15000 | 3.6712          | 0.3560   |
+| 3.5969        | 4.6647  | 16000 | 3.6563          | 0.3570   |
+| 3.6047        | 4.9563  | 17000 | 3.6433          | 0.3584   |
+| 3.5127        | 5.2478  | 18000 | 3.6452          | 0.3588   |
+| 3.5235        | 5.5394  | 19000 | 3.6360          | 0.3599   |
+| 3.5338        | 5.8309  | 20000 | 3.6230          | 0.3607   |
+| 3.4444        | 6.1224  | 21000 | 3.6260          | 0.3611   |
+| 3.4886        | 6.4140  | 22000 | 3.6203          | 0.3619   |
+| 3.4905        | 6.7055  | 23000 | 3.6091          | 0.3624   |
+| 3.4926        | 6.9971  | 24000 | 3.6008          | 0.3634   |
+| 3.4256        | 7.2886  | 25000 | 3.6067          | 0.3634   |
+| 3.4546        | 7.5802  | 26000 | 3.6003          | 0.3636   |
+| 3.4669        | 7.8717  | 27000 | 3.5895          | 0.3651   |
+| 3.3773        | 8.1633  | 28000 | 3.5999          | 0.3647   |
+| 3.4317        | 8.4548  | 29000 | 3.5932          | 0.3650   |
+| 3.4366        | 8.7464  | 30000 | 3.5841          | 0.3657   |
+| 3.3397        | 9.0379  | 31000 | 3.5906          | 0.3658   |
+| 3.3884        | 9.3294  | 32000 | 3.5857          | 0.3662   |
+| 3.4081        | 9.6210  | 33000 | 3.5797          | 0.3667   |
+| 3.4228        | 9.9125  | 34000 | 3.5700          | 0.3674   |
+| 3.3465        | 10.2041 | 35000 | 3.5860          | 0.3668   |
+| 3.3828        | 10.4956 | 36000 | 3.5726          | 0.3675   |
+| 3.395         | 10.7872 | 37000 | 3.5661          | 0.3680   |
+| 3.302         | 11.0787 | 38000 | 3.5796          | 0.3674   |
+| 3.3412        | 11.3703 | 39000 | 3.5757          | 0.3678   |
+| 3.3579        | 11.6618 | 40000 | 3.5665          | 0.3687   |
+| 3.3813        | 11.9534 | 41000 | 3.5574          | 0.3692   |
+| 3.3164        | 12.2449 | 42000 | 3.5707          | 0.3685   |
+| 3.3522        | 12.5364 | 43000 | 3.5650          | 0.3690   |
+| 3.3557        | 12.8280 | 44000 | 3.5569          | 0.3696   |
+| 3.2737        | 13.1195 | 45000 | 3.5717          | 0.3690   |
+| 3.309         | 13.4111 | 46000 | 3.5672          | 0.3693   |
+| 3.3406        | 13.7026 | 47000 | 3.5575          | 0.3699   |
+| 3.3615        | 13.9942 | 48000 | 3.5509          | 0.3703   |
+| 3.2756        | 14.2857 | 49000 | 3.5667          | 0.3692   |
+| 3.3088        | 14.5773 | 50000 | 3.5618          | 0.3699   |
+| 3.3286        | 14.8688 | 51000 | 3.5505          | 0.3706   |
+| 3.2409        | 15.1603 | 52000 | 3.5672          | 0.3700   |
+| 3.3068        | 15.4519 | 53000 | 3.5583          | 0.3703   |
+| 3.3104        | 15.7434 | 54000 | 3.5540          | 0.3710   |
+| 3.204         | 16.0350 | 55000 | 3.5639          | 0.3704   |
+| 3.2553        | 16.3265 | 56000 | 3.5618          | 0.3706   |
+| 3.2939        | 16.6181 | 57000 | 3.5557          | 0.3707   |
+| 3.2982        | 16.9096 | 58000 | 3.5438          | 0.3719   |
+| 3.2398        | 17.2012 | 59000 | 3.5617          | 0.3709   |
+| 3.2713        | 17.4927 | 60000 | 3.5542          | 0.3716   |
+| 3.282         | 17.7843 | 61000 | 3.5448          | 0.3719   |
+| 3.2062        | 18.0758 | 62000 | 3.5639          | 0.3710   |
+| 3.232         | 18.3673 | 63000 | 3.5570          | 0.3712   |
+| 3.2714        | 18.6589 | 64000 | 3.5483          | 0.3720   |
+| 3.2893        | 18.9504 | 65000 | 3.5383          | 0.3726   |
+| 3.2164        | 19.2420 | 66000 | 3.5599          | 0.3714   |
+| 3.2522        | 19.5335 | 67000 | 3.5518          | 0.3719   |
+| 3.2767        | 19.8251 | 68000 | 3.5440          | 0.3723   |
+| 3.1893        | 20.1166 | 69000 | 3.5625          | 0.3713   |
+| 3.217         | 20.4082 | 70000 | 3.5573          | 0.3718   |
+| 3.2432        | 20.6997 | 71000 | 3.5459          | 0.3725   |
+| 3.2656        | 20.9913 | 72000 | 3.5371          | 0.3729   |
+| 3.2003        | 21.2828 | 73000 | 3.5589          | 0.3718   |
+| 3.2253        | 21.5743 | 74000 | 3.5498          | 0.3724   |
+| 3.2579        | 21.8659 | 75000 | 3.5422          | 0.3728   |
+| 3.1764        | 22.1574 | 76000 | 3.5602          | 0.3721   |
+| 3.2047        | 22.4490 | 77000 | 3.5530          | 0.3730   |
+| 3.2406        | 22.7405 | 78000 | 3.5476          | 0.3727   |
+| 3.1435        | 23.0321 | 79000 | 3.5563          | 0.3725   |
+| 3.1993        | 23.3236 | 80000 | 3.5572          | 0.3726   |
+| 3.2068        | 23.6152 | 81000 | 3.5499          | 0.3730   |
+| 3.2179        | 23.9067 | 82000 | 3.5395          | 0.3734   |
+| 3.1598        | 24.1983 | 83000 | 3.5586          | 0.3726   |
+| 3.1887        | 24.4898 | 84000 | 3.5542          | 0.3728   |
+| 3.2059        | 24.7813 | 85000 | 3.5452          | 0.3733   |
+| 3.1336        | 25.0729 | 86000 | 3.5628          | 0.3727   |
+| 3.1732        | 25.3644 | 87000 | 3.5540          | 0.3730   |
+| 3.1923        | 25.6560 | 88000 | 3.5496          | 0.3729   |
+| 3.2143        | 25.9475 | 89000 | 3.5382          | 0.3741   |
+| 3.1572        | 26.2391 | 90000 | 3.5592          | 0.3729   |
+| 3.1743        | 26.5306 | 91000 | 3.5485          | 0.3733   |
+| 3.2024        | 26.8222 | 92000 | 3.5458          | 0.3737   |
+### Framework versions
+- Transformers 4.55.2
+- Pytorch 2.8.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.21.4

exp2b/swap_0.3_cost_to_push_1032/all_results.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+    "epoch": 26.82215743440233,
+    "eval_accuracy": 0.3656773800512948,
+    "eval_loss": 3.5841171741485596,
+    "eval_runtime": 53.6893,
+    "eval_samples": 16614,
+    "eval_samples_per_second": 309.447,
+    "eval_steps_per_second": 19.352,
+    "perplexity": 36.02154292815997,
+    "total_flos": 1.92300745162752e+18,
+    "train_loss": 3.418238466179889,
+    "train_runtime": 40992.6486,
+    "train_samples": 274385,
+    "train_samples_per_second": 334.676,
+    "train_steps_per_second": 4.184
+}

exp2b/swap_0.3_cost_to_push_1032/config.json ADDED Viewed

	@@ -0,0 +1,31 @@

+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "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,
+  "torch_dtype": "float32",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "vocab_size": 50257
+}

exp2b/swap_0.3_cost_to_push_1032/eval_results.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+    "epoch": 26.82215743440233,
+    "eval_accuracy": 0.3656773800512948,
+    "eval_loss": 3.5841171741485596,
+    "eval_runtime": 53.6893,
+    "eval_samples": 16614,
+    "eval_samples_per_second": 309.447,
+    "eval_steps_per_second": 19.352,
+    "perplexity": 36.02154292815997
+}

exp2b/swap_0.3_cost_to_push_1032/generation_config.json ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.55.2"
+}

exp2b/swap_0.3_cost_to_push_1032/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_1032/special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,5 @@

+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_push_1032/tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_1032/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,20 @@

+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1024,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_push_1032/train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 26.82215743440233,
+    "total_flos": 1.92300745162752e+18,
+    "train_loss": 3.418238466179889,
+    "train_runtime": 40992.6486,
+    "train_samples": 274385,
+    "train_samples_per_second": 334.676,
+    "train_steps_per_second": 4.184
+}

exp2b/swap_0.3_cost_to_push_1032/trainer_state.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_1032/vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_2128/README.md ADDED Viewed

	@@ -0,0 +1,179 @@

+---
+library_name: transformers
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: exceptions_exp2_swap_0.3_cost_to_push_2128
+  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. -->
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/craaaa/exceptions_exp2/runs/ymxwvv71)
+# exceptions_exp2_swap_0.3_cost_to_push_2128
+This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.5810
+- Accuracy: 0.3661
+## 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: 0.0006
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 2128
+- gradient_accumulation_steps: 5
+- total_train_batch_size: 80
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 100
+- num_epochs: 50.0
+- mixed_precision_training: Native AMP
+### Training results
+| Training Loss | Epoch   | Step   | Validation Loss | Accuracy |
+|:-------------:|:-------:|:------:|:---------------:|:--------:|
+| 4.8475        | 0.2915  | 1000   | 4.7750          | 0.2502   |
+| 4.3456        | 0.5831  | 2000   | 4.2943          | 0.2978   |
+| 4.1456        | 0.8746  | 3000   | 4.1059          | 0.3138   |
+| 3.9944        | 1.1662  | 4000   | 3.9968          | 0.3234   |
+| 3.9491        | 1.4577  | 5000   | 3.9227          | 0.3306   |
+| 3.8924        | 1.7493  | 6000   | 3.8664          | 0.3360   |
+| 3.7617        | 2.0408  | 7000   | 3.8208          | 0.3401   |
+| 3.7602        | 2.3324  | 8000   | 3.7899          | 0.3432   |
+| 3.7334        | 2.6239  | 9000   | 3.7620          | 0.3459   |
+| 3.7346        | 2.9155  | 10000  | 3.7338          | 0.3483   |
+| 3.6444        | 3.2070  | 11000  | 3.7238          | 0.3499   |
+| 3.659         | 3.4985  | 12000  | 3.7059          | 0.3517   |
+| 3.6575        | 3.7901  | 13000  | 3.6863          | 0.3535   |
+| 3.5507        | 4.0816  | 14000  | 3.6783          | 0.3551   |
+| 3.5711        | 4.3732  | 15000  | 3.6675          | 0.3562   |
+| 3.5827        | 4.6647  | 16000  | 3.6550          | 0.3573   |
+| 3.5824        | 4.9563  | 17000  | 3.6408          | 0.3583   |
+| 3.5035        | 5.2478  | 18000  | 3.6454          | 0.3591   |
+| 3.5303        | 5.5394  | 19000  | 3.6323          | 0.3601   |
+| 3.5382        | 5.8309  | 20000  | 3.6206          | 0.3610   |
+| 3.4441        | 6.1224  | 21000  | 3.6226          | 0.3609   |
+| 3.4783        | 6.4140  | 22000  | 3.6176          | 0.3619   |
+| 3.5042        | 6.7055  | 23000  | 3.6065          | 0.3628   |
+| 3.5019        | 6.9971  | 24000  | 3.5965          | 0.3634   |
+| 3.4391        | 7.2886  | 25000  | 3.6066          | 0.3635   |
+| 3.4671        | 7.5802  | 26000  | 3.5964          | 0.3643   |
+| 3.4646        | 7.8717  | 27000  | 3.5879          | 0.3648   |
+| 3.3788        | 8.1633  | 28000  | 3.5949          | 0.3645   |
+| 3.42          | 8.4548  | 29000  | 3.5912          | 0.3657   |
+| 3.4206        | 8.7464  | 30000  | 3.5810          | 0.3661   |
+| 3.3393        | 9.0379  | 31000  | 3.5860          | 0.3663   |
+| 3.3852        | 9.3294  | 32000  | 3.5854          | 0.3662   |
+| 3.4032        | 9.6210  | 33000  | 3.5760          | 0.3670   |
+| 3.4131        | 9.9125  | 34000  | 3.5672          | 0.3674   |
+| 3.3514        | 10.2041 | 35000  | 3.5798          | 0.3672   |
+| 3.3637        | 10.4956 | 36000  | 3.5698          | 0.3677   |
+| 3.3826        | 10.7872 | 37000  | 3.5626          | 0.3685   |
+| 3.3083        | 11.0787 | 38000  | 3.5737          | 0.3679   |
+| 3.3538        | 11.3703 | 39000  | 3.5722          | 0.3681   |
+| 3.363         | 11.6618 | 40000  | 3.5646          | 0.3687   |
+| 3.3675        | 11.9534 | 41000  | 3.5520          | 0.3693   |
+| 3.3242        | 12.2449 | 42000  | 3.5694          | 0.3687   |
+| 3.3425        | 12.5364 | 43000  | 3.5615          | 0.3693   |
+| 3.3599        | 12.8280 | 44000  | 3.5545          | 0.3693   |
+| 3.2748        | 13.1195 | 45000  | 3.5677          | 0.3692   |
+| 3.308         | 13.4111 | 46000  | 3.5632          | 0.3696   |
+| 3.3306        | 13.7026 | 47000  | 3.5567          | 0.3703   |
+| 3.3501        | 13.9942 | 48000  | 3.5477          | 0.3704   |
+| 3.303         | 14.2857 | 49000  | 3.5616          | 0.3700   |
+| 3.3112        | 14.5773 | 50000  | 3.5578          | 0.3704   |
+| 3.3277        | 14.8688 | 51000  | 3.5486          | 0.3709   |
+| 3.2649        | 15.1603 | 52000  | 3.5625          | 0.3701   |
+| 3.2859        | 15.4519 | 53000  | 3.5579          | 0.3706   |
+| 3.3246        | 15.7434 | 54000  | 3.5483          | 0.3712   |
+| 3.2136        | 16.0350 | 55000  | 3.5616          | 0.3705   |
+| 3.2559        | 16.3265 | 56000  | 3.5585          | 0.3707   |
+| 3.2834        | 16.6181 | 57000  | 3.5516          | 0.3713   |
+| 3.2976        | 16.9096 | 58000  | 3.5446          | 0.3715   |
+| 3.2242        | 17.2012 | 59000  | 3.5605          | 0.3707   |
+| 3.2621        | 17.4927 | 60000  | 3.5490          | 0.3716   |
+| 3.2843        | 17.7843 | 61000  | 3.5449          | 0.3718   |
+| 3.1983        | 18.0758 | 62000  | 3.5611          | 0.3714   |
+| 3.2407        | 18.3673 | 63000  | 3.5544          | 0.3718   |
+| 3.2768        | 18.6589 | 64000  | 3.5447          | 0.3720   |
+| 3.2854        | 18.9504 | 65000  | 3.5428          | 0.3721   |
+| 3.2234        | 19.2420 | 66000  | 3.5554          | 0.3717   |
+| 3.2568        | 19.5335 | 67000  | 3.5503          | 0.3724   |
+| 3.2694        | 19.8251 | 68000  | 3.5420          | 0.3728   |
+| 3.1784        | 20.1166 | 69000  | 3.5584          | 0.3718   |
+| 3.2217        | 20.4082 | 70000  | 3.5522          | 0.3721   |
+| 3.2423        | 20.6997 | 71000  | 3.5455          | 0.3725   |
+| 3.2513        | 20.9913 | 72000  | 3.5366          | 0.3730   |
+| 3.2074        | 21.2828 | 73000  | 3.5550          | 0.3723   |
+| 3.2366        | 21.5743 | 74000  | 3.5475          | 0.3726   |
+| 3.2257        | 21.8659 | 75000  | 3.5367          | 0.3732   |
+| 3.1729        | 22.1574 | 76000  | 3.5582          | 0.3723   |
+| 3.2132        | 22.4490 | 77000  | 3.5498          | 0.3726   |
+| 3.2481        | 22.7405 | 78000  | 3.5411          | 0.3731   |
+| 3.1369        | 23.0321 | 79000  | 3.5574          | 0.3725   |
+| 3.1759        | 23.3236 | 80000  | 3.5522          | 0.3729   |
+| 3.2125        | 23.6152 | 81000  | 3.5454          | 0.3733   |
+| 3.2274        | 23.9067 | 82000  | 3.5361          | 0.3737   |
+| 3.1613        | 24.1983 | 83000  | 3.5553          | 0.3729   |
+| 3.1876        | 24.4898 | 84000  | 3.5515          | 0.3730   |
+| 3.2129        | 24.7813 | 85000  | 3.5437          | 0.3735   |
+| 3.1389        | 25.0729 | 86000  | 3.5526          | 0.3732   |
+| 3.1698        | 25.3644 | 87000  | 3.5525          | 0.3731   |
+| 3.1955        | 25.6560 | 88000  | 3.5473          | 0.3735   |
+| 3.1939        | 25.9475 | 89000  | 3.5392          | 0.3737   |
+| 3.1502        | 26.2391 | 90000  | 3.5545          | 0.3731   |
+| 3.1742        | 26.5306 | 91000  | 3.5458          | 0.3737   |
+| 3.1846        | 26.8222 | 92000  | 3.5407          | 0.3741   |
+| 3.1308        | 27.1137 | 93000  | 3.5584          | 0.3731   |
+| 3.1635        | 27.4052 | 94000  | 3.5564          | 0.3734   |
+| 3.192         | 27.6968 | 95000  | 3.5428          | 0.3742   |
+| 3.186         | 27.9883 | 96000  | 3.5344          | 0.3744   |
+| 3.1337        | 28.2799 | 97000  | 3.5569          | 0.3733   |
+| 3.1666        | 28.5714 | 98000  | 3.5477          | 0.3739   |
+| 3.1893        | 28.8630 | 99000  | 3.5428          | 0.3742   |
+| 3.115         | 29.1545 | 100000 | 3.5589          | 0.3734   |
+| 3.1486        | 29.4461 | 101000 | 3.5559          | 0.3735   |
+| 3.1697        | 29.7376 | 102000 | 3.5468          | 0.3742   |
+| 3.0804        | 30.0292 | 103000 | 3.5555          | 0.3737   |
+| 3.12          | 30.3207 | 104000 | 3.5551          | 0.3737   |
+| 3.1463        | 30.6122 | 105000 | 3.5494          | 0.3742   |
+| 3.1608        | 30.9038 | 106000 | 3.5439          | 0.3744   |
+| 3.1076        | 31.1953 | 107000 | 3.5588          | 0.3735   |
+| 3.1425        | 31.4869 | 108000 | 3.5515          | 0.3741   |
+| 3.1472        | 31.7784 | 109000 | 3.5468          | 0.3745   |
+| 3.0671        | 32.0700 | 110000 | 3.5595          | 0.3737   |
+| 3.1161        | 32.3615 | 111000 | 3.5552          | 0.3741   |
+| 3.1468        | 32.6531 | 112000 | 3.5482          | 0.3746   |
+| 3.145         | 32.9446 | 113000 | 3.5437          | 0.3746   |
+| 3.0995        | 33.2362 | 114000 | 3.5615          | 0.3737   |
+| 3.1161        | 33.5277 | 115000 | 3.5526          | 0.3743   |
+| 3.1275        | 33.8192 | 116000 | 3.5474          | 0.3747   |
+### Framework versions
+- Transformers 4.55.2
+- Pytorch 2.8.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.21.4

exp2b/swap_0.3_cost_to_push_2128/all_results.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+    "epoch": 33.819241982507286,
+    "eval_accuracy": 0.36613451932765073,
+    "eval_loss": 3.5809543132781982,
+    "eval_runtime": 80.5332,
+    "eval_samples": 16614,
+    "eval_samples_per_second": 206.3,
+    "eval_steps_per_second": 12.902,
+    "perplexity": 35.907791783891604,
+    "total_flos": 2.42466071740416e+18,
+    "train_loss": 3.3607484947730755,
+    "train_runtime": 84954.2574,
+    "train_samples": 274385,
+    "train_samples_per_second": 161.49,
+    "train_steps_per_second": 2.019
+}

exp2b/swap_0.3_cost_to_push_2128/config.json ADDED Viewed

	@@ -0,0 +1,31 @@

+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "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,
+  "torch_dtype": "float32",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "vocab_size": 50257
+}

exp2b/swap_0.3_cost_to_push_2128/eval_results.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+    "epoch": 33.819241982507286,
+    "eval_accuracy": 0.36613451932765073,
+    "eval_loss": 3.5809543132781982,
+    "eval_runtime": 80.5332,
+    "eval_samples": 16614,
+    "eval_samples_per_second": 206.3,
+    "eval_steps_per_second": 12.902,
+    "perplexity": 35.907791783891604
+}

exp2b/swap_0.3_cost_to_push_2128/generation_config.json ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.55.2"
+}

exp2b/swap_0.3_cost_to_push_2128/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_2128/special_tokens_map.json ADDED Viewed

	@@ -0,0 +1,5 @@

+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_push_2128/tokenizer.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_2128/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,20 @@

+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": {},
+  "model_max_length": 1024,
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

exp2b/swap_0.3_cost_to_push_2128/train_results.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+    "epoch": 33.819241982507286,
+    "total_flos": 2.42466071740416e+18,
+    "train_loss": 3.3607484947730755,
+    "train_runtime": 84954.2574,
+    "train_samples": 274385,
+    "train_samples_per_second": 161.49,
+    "train_steps_per_second": 2.019
+}

exp2b/swap_0.3_cost_to_push_2128/trainer_state.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_2128/vocab.json ADDED Viewed

The diff for this file is too large to render. See raw diff

exp2b/swap_0.3_cost_to_push_3591/README.md ADDED Viewed

	@@ -0,0 +1,163 @@

+---
+library_name: transformers
+tags:
+- generated_from_trainer
+metrics:
+- accuracy
+model-index:
+- name: exceptions_exp2_swap_0.3_cost_to_push_3591
+  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. -->
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/craaaa/exceptions_exp2/runs/rixnpbrc)
+# exceptions_exp2_swap_0.3_cost_to_push_3591
+This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 3.5586
+- Accuracy: 0.3691
+## 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: 0.0006
+- train_batch_size: 16
+- eval_batch_size: 16
+- seed: 3591
+- gradient_accumulation_steps: 5
+- total_train_batch_size: 80
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 100
+- num_epochs: 50.0
+- mixed_precision_training: Native AMP
+### Training results
+| Training Loss | Epoch   | Step   | Accuracy | Validation Loss |
+|:-------------:|:-------:|:------:|:--------:|:---------------:|
+| 4.8593        | 0.2915  | 1000   | 0.2526   | 4.7654          |
+| 4.3289        | 0.5831  | 2000   | 0.2982   | 4.2945          |
+| 4.1479        | 0.8746  | 3000   | 0.3142   | 4.1063          |
+| 3.9972        | 1.1662  | 4000   | 0.3242   | 3.9937          |
+| 3.9327        | 1.4577  | 5000   | 0.3313   | 3.9192          |
+| 3.8835        | 1.7493  | 6000   | 0.3364   | 3.8616          |
+| 3.7384        | 2.0408  | 7000   | 0.3410   | 3.8179          |
+| 3.7687        | 2.3324  | 8000   | 0.3436   | 3.7878          |
+| 3.7421        | 2.6239  | 9000   | 0.3464   | 3.7564          |
+| 3.7386        | 2.9155  | 10000  | 0.3489   | 3.7297          |
+| 3.6418        | 3.2070  | 11000  | 0.3507   | 3.7163          |
+| 3.6416        | 3.4985  | 12000  | 0.3526   | 3.7019          |
+| 3.6519        | 3.7901  | 13000  | 0.3543   | 3.6797          |
+| 3.5328        | 4.0816  | 14000  | 0.3555   | 3.6727          |
+| 3.5737        | 4.3732  | 15000  | 0.3568   | 3.6612          |
+| 3.584         | 4.6647  | 16000  | 0.3577   | 3.6496          |
+| 3.5798        | 4.9563  | 17000  | 0.3591   | 3.6358          |
+| 3.5091        | 5.2478  | 18000  | 0.3597   | 3.6373          |
+| 3.5288        | 5.5394  | 19000  | 0.3603   | 3.6277          |
+| 3.535         | 5.8309  | 20000  | 0.3615   | 3.6166          |
+| 3.4419        | 6.1224  | 21000  | 0.3621   | 3.6196          |
+| 3.4774        | 6.4140  | 22000  | 0.3624   | 3.6104          |
+| 3.487         | 6.7055  | 23000  | 0.3631   | 3.6031          |
+| 3.4925        | 6.9971  | 24000  | 0.3642   | 3.5918          |
+| 3.4321        | 7.2886  | 25000  | 0.3639   | 3.6004          |
+| 3.4443        | 7.5802  | 26000  | 0.3647   | 3.5903          |
+| 3.4627        | 7.8717  | 27000  | 0.3654   | 3.5863          |
+| 3.383         | 8.1633  | 28000  | 0.3651   | 3.5941          |
+| 3.4144        | 8.4548  | 29000  | 0.3660   | 3.5875          |
+| 3.4243        | 8.7464  | 30000  | 0.3667   | 3.5769          |
+| 3.3265        | 9.0379  | 31000  | 0.3664   | 3.5819          |
+| 3.3732        | 9.3294  | 32000  | 0.3666   | 3.5836          |
+| 3.4029        | 9.6210  | 33000  | 0.3673   | 3.5709          |
+| 3.412         | 9.9125  | 34000  | 0.3676   | 3.5678          |
+| 3.3442        | 10.2041 | 35000  | 0.3676   | 3.5765          |
+| 3.367         | 10.4956 | 36000  | 0.3678   | 3.5701          |
+| 3.3839        | 10.7872 | 37000  | 0.3686   | 3.5612          |
+| 3.3034        | 11.0787 | 38000  | 0.3682   | 3.5731          |
+| 3.3383        | 11.3703 | 39000  | 0.3686   | 3.5677          |
+| 3.3669        | 11.6618 | 40000  | 0.3691   | 3.5586          |
+| 3.3862        | 11.9534 | 41000  | 0.3696   | 3.5519          |
+| 3.3147        | 12.2449 | 42000  | 0.3692   | 3.5673          |
+| 3.3332        | 12.5364 | 43000  | 0.3694   | 3.5584          |
+| 3.3562        | 12.8280 | 44000  | 0.3699   | 3.5525          |
+| 3.2737        | 13.1195 | 45000  | 0.3693   | 3.5666          |
+| 3.3177        | 13.4111 | 46000  | 0.3696   | 3.5580          |
+| 3.3253        | 13.7026 | 47000  | 0.3704   | 3.5513          |
+| 3.3387        | 13.9942 | 48000  | 0.3709   | 3.5473          |
+| 3.2835        | 14.2857 | 49000  | 0.3702   | 3.5606          |
+| 3.3224        | 14.5773 | 50000  | 0.3706   | 3.5517          |
+| 3.331         | 14.8688 | 51000  | 0.3711   | 3.5432          |
+| 3.2473        | 15.1603 | 52000  | 0.3705   | 3.5595          |
+| 3.2867        | 15.4519 | 53000  | 0.3709   | 3.5504          |
+| 3.304         | 15.7434 | 54000  | 0.3717   | 3.5452          |
+| 3.2131        | 16.0350 | 55000  | 0.3708   | 3.5580          |
+| 3.2648        | 16.3265 | 56000  | 0.3710   | 3.5541          |
+| 3.2874        | 16.6181 | 57000  | 0.3713   | 3.5483          |
+| 3.3064        | 16.9096 | 58000  | 0.3720   | 3.5407          |
+| 3.2338        | 17.2012 | 59000  | 0.3710   | 3.5592          |
+| 3.2579        | 17.4927 | 60000  | 0.3714   | 3.5543          |
+| 3.2919        | 17.7843 | 61000  | 0.3722   | 3.5450          |
+| 3.2019        | 18.0758 | 62000  | 0.3713   | 3.5572          |
+| 3.235         | 18.3673 | 63000  | 0.3717   | 3.5555          |
+| 3.2601        | 18.6589 | 64000  | 0.3721   | 3.5437          |
+| 3.2662        | 18.9504 | 65000  | 0.3728   | 3.5354          |
+| 3.2107        | 19.2420 | 66000  | 0.3718   | 3.5546          |
+| 3.2552        | 19.5335 | 67000  | 0.3721   | 3.5477          |
+| 3.2683        | 19.8251 | 68000  | 0.3729   | 3.5381          |
+| 3.1937        | 20.1166 | 69000  | 0.3720   | 3.5565          |
+| 3.2322        | 20.4082 | 70000  | 0.3725   | 3.5476          |
+| 3.2443        | 20.6997 | 71000  | 0.3729   | 3.5433          |
+| 3.2505        | 20.9913 | 72000  | 0.3730   | 3.5371          |
+| 3.2086        | 21.2828 | 73000  | 0.3724   | 3.5527          |
+| 3.2317        | 21.5743 | 74000  | 0.3729   | 3.5469          |
+| 3.2386        | 21.8659 | 75000  | 0.3736   | 3.5376          |
+| 3.1664        | 22.1574 | 76000  | 0.3729   | 3.5540          |
+| 3.2043        | 22.4490 | 77000  | 0.3729   | 3.5509          |
+| 3.2342        | 22.7405 | 78000  | 0.3734   | 3.5422          |
+| 3.1303        | 23.0321 | 79000  | 0.3730   | 3.5534          |
+| 3.1847        | 23.3236 | 80000  | 0.3728   | 3.5516          |
+| 3.1867        | 23.6152 | 81000  | 3.5575   | 0.3726          |
+| 3.2139        | 23.9067 | 82000  | 3.5493   | 0.3730          |
+| 3.1722        | 24.1983 | 83000  | 3.5635   | 0.3725          |
+| 3.203         | 24.4898 | 84000  | 3.5515   | 0.3731          |
+| 3.2029        | 24.7813 | 85000  | 3.5395   | 0.3739          |
+| 3.1271        | 25.0729 | 86000  | 3.5533   | 0.3731          |
+| 3.1722        | 25.3644 | 87000  | 3.5511   | 0.3733          |
+| 3.1964        | 25.6560 | 88000  | 3.5440   | 0.3738          |
+| 3.2176        | 25.9475 | 89000  | 3.5378   | 0.3741          |
+| 3.1526        | 26.2391 | 90000  | 3.5583   | 0.3730          |
+| 3.1819        | 26.5306 | 91000  | 3.5454   | 0.3737          |
+| 3.2004        | 26.8222 | 92000  | 3.5387   | 0.3741          |
+| 3.1162        | 27.1137 | 93000  | 3.5589   | 0.3734          |
+| 3.1631        | 27.4052 | 94000  | 3.5503   | 0.3736          |
+| 3.1695        | 27.6968 | 95000  | 3.5414   | 0.3740          |
+| 3.2012        | 27.9883 | 96000  | 3.5377   | 0.3742          |
+| 3.1274        | 28.2799 | 97000  | 3.5568   | 0.3731          |
+| 3.1686        | 28.5714 | 98000  | 3.5454   | 0.3738          |
+| 3.1767        | 28.8630 | 99000  | 3.5356   | 0.3747          |
+| 3.1033        | 29.1545 | 100000 | 3.5560   | 0.3736          |
+### Framework versions
+- Transformers 4.55.2
+- Pytorch 2.8.0+cu128
+- Datasets 4.0.0
+- Tokenizers 0.21.4

exp2b/swap_0.3_cost_to_push_3591/all_results.json ADDED Viewed

	@@ -0,0 +1,16 @@

+{
+    "epoch": 29.154518950437318,
+    "eval_accuracy": 0.36914273369993994,
+    "eval_loss": 3.5585551261901855,
+    "eval_runtime": 54.3298,
+    "eval_samples": 16614,
+    "eval_samples_per_second": 305.799,
+    "eval_steps_per_second": 19.124,
+    "perplexity": 35.11242744921675,
+    "total_flos": 2.09022259396608e+18,
+    "train_loss": 0.633907257232666,
+    "train_runtime": 9007.2474,
+    "train_samples": 274385,
+    "train_samples_per_second": 1523.135,
+    "train_steps_per_second": 19.04
+}

exp2b/swap_0.3_cost_to_push_3591/config.json ADDED Viewed

	@@ -0,0 +1,31 @@

+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "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,
+  "torch_dtype": "float32",
+  "transformers_version": "4.55.2",
+  "use_cache": true,
+  "vocab_size": 50257
+}

exp2b/swap_0.3_cost_to_push_3591/eval_results.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+    "epoch": 29.154518950437318,
+    "eval_accuracy": 0.36914273369993994,
+    "eval_loss": 3.5585551261901855,
+    "eval_runtime": 54.3298,
+    "eval_samples": 16614,
+    "eval_samples_per_second": 305.799,
+    "eval_steps_per_second": 19.124,
+    "perplexity": 35.11242744921675
+}

exp2b/swap_0.3_cost_to_push_3591/generation_config.json ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "4.55.2"
+}

exp2b/swap_0.3_cost_to_push_3591/merges.txt ADDED Viewed

The diff for this file is too large to render. See raw diff