Delete checkpoint_5000 (1)/content
Browse files- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/config.json +0 -46
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/generation_config.json +0 -6
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/merges.txt +0 -0
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/optimizer.pt +0 -3
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/pytorch_model.bin +0 -3
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/rng_state.pth +0 -3
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/scheduler.pt +0 -3
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/special_tokens_map.json +0 -6
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/tokenizer.json +0 -0
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/tokenizer_config.json +0 -9
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/trainer_state.json +0 -44
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/training_args.bin +0 -3
- checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/vocab.json +0 -0
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/config.json
DELETED
|
@@ -1,46 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "distilgpt2",
|
| 3 |
-
"_num_labels": 1,
|
| 4 |
-
"activation_function": "gelu_new",
|
| 5 |
-
"architectures": [
|
| 6 |
-
"GPT2LMHeadModel"
|
| 7 |
-
],
|
| 8 |
-
"attn_pdrop": 0.1,
|
| 9 |
-
"bos_token_id": 50256,
|
| 10 |
-
"embd_pdrop": 0.1,
|
| 11 |
-
"eos_token_id": 50256,
|
| 12 |
-
"id2label": {
|
| 13 |
-
"0": "LABEL_0"
|
| 14 |
-
},
|
| 15 |
-
"initializer_range": 0.02,
|
| 16 |
-
"label2id": {
|
| 17 |
-
"LABEL_0": 0
|
| 18 |
-
},
|
| 19 |
-
"layer_norm_epsilon": 1e-05,
|
| 20 |
-
"model_type": "gpt2",
|
| 21 |
-
"n_ctx": 1024,
|
| 22 |
-
"n_embd": 768,
|
| 23 |
-
"n_head": 12,
|
| 24 |
-
"n_inner": null,
|
| 25 |
-
"n_layer": 6,
|
| 26 |
-
"n_positions": 1024,
|
| 27 |
-
"reorder_and_upcast_attn": false,
|
| 28 |
-
"resid_pdrop": 0.1,
|
| 29 |
-
"scale_attn_by_inverse_layer_idx": false,
|
| 30 |
-
"scale_attn_weights": true,
|
| 31 |
-
"summary_activation": null,
|
| 32 |
-
"summary_first_dropout": 0.1,
|
| 33 |
-
"summary_proj_to_labels": true,
|
| 34 |
-
"summary_type": "cls_index",
|
| 35 |
-
"summary_use_proj": true,
|
| 36 |
-
"task_specific_params": {
|
| 37 |
-
"text-generation": {
|
| 38 |
-
"do_sample": true,
|
| 39 |
-
"max_length": 50
|
| 40 |
-
}
|
| 41 |
-
},
|
| 42 |
-
"torch_dtype": "float32",
|
| 43 |
-
"transformers_version": "4.30.2",
|
| 44 |
-
"use_cache": true,
|
| 45 |
-
"vocab_size": 50257
|
| 46 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/generation_config.json
DELETED
|
@@ -1,6 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_from_model_config": true,
|
| 3 |
-
"bos_token_id": 50256,
|
| 4 |
-
"eos_token_id": 50256,
|
| 5 |
-
"transformers_version": "4.30.2"
|
| 6 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/merges.txt
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/optimizer.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:c3e7c7d24f3636847ad37d9dcd9849213590bfe627d32e1880beefe57ae1e771
|
| 3 |
-
size 655345093
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/pytorch_model.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:95b5d445272ce5124b381aa4a9526a0744b9dc9a5da02166678acd8aeb6a6367
|
| 3 |
-
size 327674773
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/rng_state.pth
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:0d52ad5b46c2d84d3d57dd9503e47782a2d05ff2a3bbfb945a7eae7dec648d30
|
| 3 |
-
size 14575
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/scheduler.pt
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:5113844dd81345e733f760270707efba2267b48bd813dd002144c6b864e7f337
|
| 3 |
-
size 627
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/special_tokens_map.json
DELETED
|
@@ -1,6 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"bos_token": "<|endoftext|>",
|
| 3 |
-
"eos_token": "<|endoftext|>",
|
| 4 |
-
"pad_token": "<|endoftext|>",
|
| 5 |
-
"unk_token": "<|endoftext|>"
|
| 6 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/tokenizer.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/tokenizer_config.json
DELETED
|
@@ -1,9 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"add_prefix_space": false,
|
| 3 |
-
"bos_token": "<|endoftext|>",
|
| 4 |
-
"clean_up_tokenization_spaces": true,
|
| 5 |
-
"eos_token": "<|endoftext|>",
|
| 6 |
-
"model_max_length": 1024,
|
| 7 |
-
"tokenizer_class": "GPT2Tokenizer",
|
| 8 |
-
"unk_token": "<|endoftext|>"
|
| 9 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/trainer_state.json
DELETED
|
@@ -1,44 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_metric": null,
|
| 3 |
-
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 2.765486725663717,
|
| 5 |
-
"global_step": 5000,
|
| 6 |
-
"is_hyper_param_search": false,
|
| 7 |
-
"is_local_process_zero": true,
|
| 8 |
-
"is_world_process_zero": true,
|
| 9 |
-
"log_history": [
|
| 10 |
-
{
|
| 11 |
-
"epoch": 1.0,
|
| 12 |
-
"learning_rate": 1.5015040558432561e-05,
|
| 13 |
-
"loss": 4.2302,
|
| 14 |
-
"step": 1807
|
| 15 |
-
},
|
| 16 |
-
{
|
| 17 |
-
"epoch": 1.0,
|
| 18 |
-
"eval_loss": 4.0081586837768555,
|
| 19 |
-
"eval_runtime": 33.5458,
|
| 20 |
-
"eval_samples_per_second": 3435.628,
|
| 21 |
-
"eval_steps_per_second": 13.444,
|
| 22 |
-
"step": 1808
|
| 23 |
-
},
|
| 24 |
-
{
|
| 25 |
-
"epoch": 2.0,
|
| 26 |
-
"learning_rate": 5.020077553310694e-06,
|
| 27 |
-
"loss": 4.0653,
|
| 28 |
-
"step": 3614
|
| 29 |
-
},
|
| 30 |
-
{
|
| 31 |
-
"epoch": 2.0,
|
| 32 |
-
"eval_loss": 3.9669103622436523,
|
| 33 |
-
"eval_runtime": 33.7689,
|
| 34 |
-
"eval_samples_per_second": 3412.936,
|
| 35 |
-
"eval_steps_per_second": 13.355,
|
| 36 |
-
"step": 3616
|
| 37 |
-
}
|
| 38 |
-
],
|
| 39 |
-
"max_steps": 5424,
|
| 40 |
-
"num_train_epochs": 3,
|
| 41 |
-
"total_flos": 3918208057933824.0,
|
| 42 |
-
"trial_name": null,
|
| 43 |
-
"trial_params": null
|
| 44 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/training_args.bin
DELETED
|
@@ -1,3 +0,0 @@
|
|
| 1 |
-
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:e008d35b83974f323bb79f5386e08bec58fd48fd28bd7077782fb62391feda3d
|
| 3 |
-
size 3963
|
|
|
|
|
|
|
|
|
|
|
|
checkpoint_5000 (1)/content/distilgpt2-finetuned-causal/checkpoint-5000/vocab.json
DELETED
|
The diff for this file is too large to render.
See raw diff
|
|
|