Training in progress, epoch 7, checkpoint
Browse files- checkpoint-392/added_tokens.json +3 -0
- checkpoint-392/config.json +42 -0
- checkpoint-392/generation_config.json +8 -0
- checkpoint-392/merges.txt +0 -0
- checkpoint-392/optimizer.pt +3 -0
- checkpoint-392/pytorch_model.bin +3 -0
- checkpoint-392/rng_state.pth +3 -0
- checkpoint-392/scheduler.pt +3 -0
- checkpoint-392/special_tokens_map.json +6 -0
- checkpoint-392/tokenizer.json +0 -0
- checkpoint-392/tokenizer_config.json +20 -0
- checkpoint-392/trainer_state.json +633 -0
- checkpoint-392/training_args.bin +3 -0
- checkpoint-392/vocab.json +0 -0
checkpoint-392/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 50256
|
| 3 |
+
}
|
checkpoint-392/config.json
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2-xl",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"do_sample": true,
|
| 10 |
+
"embd_pdrop": 0.1,
|
| 11 |
+
"eos_token_id": 50256,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"max_length": 50,
|
| 15 |
+
"model_type": "gpt2",
|
| 16 |
+
"n_ctx": 1024,
|
| 17 |
+
"n_embd": 1600,
|
| 18 |
+
"n_head": 25,
|
| 19 |
+
"n_inner": null,
|
| 20 |
+
"n_layer": 48,
|
| 21 |
+
"n_positions": 1024,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"reorder_and_upcast_attn": false,
|
| 24 |
+
"resid_pdrop": 0.1,
|
| 25 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 26 |
+
"scale_attn_weights": true,
|
| 27 |
+
"summary_activation": null,
|
| 28 |
+
"summary_first_dropout": 0.1,
|
| 29 |
+
"summary_proj_to_labels": true,
|
| 30 |
+
"summary_type": "cls_index",
|
| 31 |
+
"summary_use_proj": true,
|
| 32 |
+
"task_specific_params": {
|
| 33 |
+
"text-generation": {
|
| 34 |
+
"do_sample": true,
|
| 35 |
+
"max_length": 50
|
| 36 |
+
}
|
| 37 |
+
},
|
| 38 |
+
"torch_dtype": "float32",
|
| 39 |
+
"transformers_version": "4.34.0",
|
| 40 |
+
"use_cache": true,
|
| 41 |
+
"vocab_size": 50257
|
| 42 |
+
}
|
checkpoint-392/generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": 50256,
|
| 6 |
+
"max_length": 50,
|
| 7 |
+
"transformers_version": "4.34.0"
|
| 8 |
+
}
|
checkpoint-392/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-392/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ac7e4baccc3ba736fb6442f1a4c150d46af45fce7d3ebc3c961dc75e1ca6958
|
| 3 |
+
size 12461385454
|
checkpoint-392/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58a1c83093cd8cbb1e6b06218b5a35ef5b71366d9815c3eaa24c25ec778eb94e
|
| 3 |
+
size 6230637102
|
checkpoint-392/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fbcddf0a1b829b613df835c1ff9a04869fb200482d42c0b3d649690572217cb7
|
| 3 |
+
size 14244
|
checkpoint-392/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dd7f41fb683deadba297bb009414464768c1b3d0ff7dfd96f512b85d2aa16d2
|
| 3 |
+
size 1064
|
checkpoint-392/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
checkpoint-392/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-392/tokenizer_config.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"50256": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
}
|
| 12 |
+
},
|
| 13 |
+
"additional_special_tokens": [],
|
| 14 |
+
"bos_token": "<|endoftext|>",
|
| 15 |
+
"clean_up_tokenization_spaces": true,
|
| 16 |
+
"eos_token": "<|endoftext|>",
|
| 17 |
+
"model_max_length": 1024,
|
| 18 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 19 |
+
"unk_token": "<|endoftext|>"
|
| 20 |
+
}
|
checkpoint-392/trainer_state.json
ADDED
|
@@ -0,0 +1,633 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 7.0,
|
| 5 |
+
"eval_steps": 28,
|
| 6 |
+
"global_step": 392,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11,
|
| 13 |
+
"learning_rate": 3e-05,
|
| 14 |
+
"loss": 3.5167,
|
| 15 |
+
"step": 6
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.21,
|
| 19 |
+
"learning_rate": 3e-05,
|
| 20 |
+
"loss": 2.7034,
|
| 21 |
+
"step": 12
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.32,
|
| 25 |
+
"learning_rate": 3e-05,
|
| 26 |
+
"loss": 2.5409,
|
| 27 |
+
"step": 18
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.43,
|
| 31 |
+
"learning_rate": 3e-05,
|
| 32 |
+
"loss": 2.4931,
|
| 33 |
+
"step": 24
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.5,
|
| 37 |
+
"eval_accuracy": 0.5748751060189339,
|
| 38 |
+
"eval_loss": 2.390355348587036,
|
| 39 |
+
"eval_runtime": 25.7729,
|
| 40 |
+
"eval_samples_per_second": 34.261,
|
| 41 |
+
"eval_steps_per_second": 2.173,
|
| 42 |
+
"step": 28
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.5,
|
| 46 |
+
"eval_bleu": 2.6290615444017718,
|
| 47 |
+
"eval_exact_match": 0.0,
|
| 48 |
+
"eval_prefix_exact_match": 0.0,
|
| 49 |
+
"step": 28
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.54,
|
| 53 |
+
"learning_rate": 3e-05,
|
| 54 |
+
"loss": 2.4944,
|
| 55 |
+
"step": 30
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.64,
|
| 59 |
+
"learning_rate": 3e-05,
|
| 60 |
+
"loss": 2.4832,
|
| 61 |
+
"step": 36
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.75,
|
| 65 |
+
"learning_rate": 3e-05,
|
| 66 |
+
"loss": 2.4336,
|
| 67 |
+
"step": 42
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.86,
|
| 71 |
+
"learning_rate": 3e-05,
|
| 72 |
+
"loss": 2.5101,
|
| 73 |
+
"step": 48
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.96,
|
| 77 |
+
"learning_rate": 3e-05,
|
| 78 |
+
"loss": 2.4699,
|
| 79 |
+
"step": 54
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.0,
|
| 83 |
+
"eval_accuracy": 0.6049526162776446,
|
| 84 |
+
"eval_loss": 2.101423740386963,
|
| 85 |
+
"eval_runtime": 23.9781,
|
| 86 |
+
"eval_samples_per_second": 36.825,
|
| 87 |
+
"eval_steps_per_second": 2.335,
|
| 88 |
+
"step": 56
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.0,
|
| 92 |
+
"eval_bleu": 2.731416805620766,
|
| 93 |
+
"eval_exact_match": 0.0,
|
| 94 |
+
"eval_prefix_exact_match": 0.0,
|
| 95 |
+
"step": 56
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.07,
|
| 99 |
+
"learning_rate": 3e-05,
|
| 100 |
+
"loss": 2.2172,
|
| 101 |
+
"step": 60
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.18,
|
| 105 |
+
"learning_rate": 3e-05,
|
| 106 |
+
"loss": 1.9462,
|
| 107 |
+
"step": 66
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.29,
|
| 111 |
+
"learning_rate": 3e-05,
|
| 112 |
+
"loss": 1.8014,
|
| 113 |
+
"step": 72
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.39,
|
| 117 |
+
"learning_rate": 3e-05,
|
| 118 |
+
"loss": 1.8689,
|
| 119 |
+
"step": 78
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.5,
|
| 123 |
+
"learning_rate": 3e-05,
|
| 124 |
+
"loss": 1.9006,
|
| 125 |
+
"step": 84
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.5,
|
| 129 |
+
"eval_accuracy": 0.6392953969397911,
|
| 130 |
+
"eval_loss": 1.8079485893249512,
|
| 131 |
+
"eval_runtime": 24.4889,
|
| 132 |
+
"eval_samples_per_second": 36.057,
|
| 133 |
+
"eval_steps_per_second": 2.287,
|
| 134 |
+
"step": 84
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.5,
|
| 138 |
+
"eval_bleu": 3.957789284907739,
|
| 139 |
+
"eval_exact_match": 0.0,
|
| 140 |
+
"eval_prefix_exact_match": 0.0,
|
| 141 |
+
"step": 84
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 1.61,
|
| 145 |
+
"learning_rate": 3e-05,
|
| 146 |
+
"loss": 1.8975,
|
| 147 |
+
"step": 90
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 1.71,
|
| 151 |
+
"learning_rate": 3e-05,
|
| 152 |
+
"loss": 1.8936,
|
| 153 |
+
"step": 96
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 1.82,
|
| 157 |
+
"learning_rate": 3e-05,
|
| 158 |
+
"loss": 1.8677,
|
| 159 |
+
"step": 102
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 1.93,
|
| 163 |
+
"learning_rate": 3e-05,
|
| 164 |
+
"loss": 1.9317,
|
| 165 |
+
"step": 108
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.0,
|
| 169 |
+
"eval_accuracy": 0.6722017129718149,
|
| 170 |
+
"eval_loss": 1.5510306358337402,
|
| 171 |
+
"eval_runtime": 24.2083,
|
| 172 |
+
"eval_samples_per_second": 36.475,
|
| 173 |
+
"eval_steps_per_second": 2.313,
|
| 174 |
+
"step": 112
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.0,
|
| 178 |
+
"eval_bleu": 4.403977979732999,
|
| 179 |
+
"eval_exact_match": 0.0,
|
| 180 |
+
"eval_prefix_exact_match": 0.0,
|
| 181 |
+
"step": 112
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.04,
|
| 185 |
+
"learning_rate": 3e-05,
|
| 186 |
+
"loss": 1.7801,
|
| 187 |
+
"step": 114
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.14,
|
| 191 |
+
"learning_rate": 3e-05,
|
| 192 |
+
"loss": 1.3602,
|
| 193 |
+
"step": 120
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.25,
|
| 197 |
+
"learning_rate": 3e-05,
|
| 198 |
+
"loss": 1.3667,
|
| 199 |
+
"step": 126
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 2.36,
|
| 203 |
+
"learning_rate": 3e-05,
|
| 204 |
+
"loss": 1.3497,
|
| 205 |
+
"step": 132
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.46,
|
| 209 |
+
"learning_rate": 3e-05,
|
| 210 |
+
"loss": 1.3984,
|
| 211 |
+
"step": 138
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 2.5,
|
| 215 |
+
"eval_accuracy": 0.707495599907831,
|
| 216 |
+
"eval_loss": 1.284958839416504,
|
| 217 |
+
"eval_runtime": 24.875,
|
| 218 |
+
"eval_samples_per_second": 35.498,
|
| 219 |
+
"eval_steps_per_second": 2.251,
|
| 220 |
+
"step": 140
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 2.5,
|
| 224 |
+
"eval_bleu": 7.1028643857654,
|
| 225 |
+
"eval_exact_match": 0.0,
|
| 226 |
+
"eval_prefix_exact_match": 0.0,
|
| 227 |
+
"step": 140
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.57,
|
| 231 |
+
"learning_rate": 3e-05,
|
| 232 |
+
"loss": 1.3143,
|
| 233 |
+
"step": 144
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.68,
|
| 237 |
+
"learning_rate": 3e-05,
|
| 238 |
+
"loss": 1.3887,
|
| 239 |
+
"step": 150
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.79,
|
| 243 |
+
"learning_rate": 3e-05,
|
| 244 |
+
"loss": 1.3563,
|
| 245 |
+
"step": 156
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.89,
|
| 249 |
+
"learning_rate": 3e-05,
|
| 250 |
+
"loss": 1.3749,
|
| 251 |
+
"step": 162
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 3.0,
|
| 255 |
+
"learning_rate": 3e-05,
|
| 256 |
+
"loss": 1.3662,
|
| 257 |
+
"step": 168
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0,
|
| 261 |
+
"eval_accuracy": 0.737386810999495,
|
| 262 |
+
"eval_loss": 1.0899983644485474,
|
| 263 |
+
"eval_runtime": 24.1441,
|
| 264 |
+
"eval_samples_per_second": 36.572,
|
| 265 |
+
"eval_steps_per_second": 2.319,
|
| 266 |
+
"step": 168
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 3.0,
|
| 270 |
+
"eval_bleu": 8.181972004065782,
|
| 271 |
+
"eval_exact_match": 0.0,
|
| 272 |
+
"eval_prefix_exact_match": 0.0,
|
| 273 |
+
"step": 168
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 3.11,
|
| 277 |
+
"learning_rate": 3e-05,
|
| 278 |
+
"loss": 0.9654,
|
| 279 |
+
"step": 174
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.21,
|
| 283 |
+
"learning_rate": 3e-05,
|
| 284 |
+
"loss": 0.9704,
|
| 285 |
+
"step": 180
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 3.32,
|
| 289 |
+
"learning_rate": 3e-05,
|
| 290 |
+
"loss": 0.9269,
|
| 291 |
+
"step": 186
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 3.43,
|
| 295 |
+
"learning_rate": 3e-05,
|
| 296 |
+
"loss": 0.9041,
|
| 297 |
+
"step": 192
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.5,
|
| 301 |
+
"eval_accuracy": 0.7670475994371804,
|
| 302 |
+
"eval_loss": 0.890863835811615,
|
| 303 |
+
"eval_runtime": 24.7388,
|
| 304 |
+
"eval_samples_per_second": 35.693,
|
| 305 |
+
"eval_steps_per_second": 2.264,
|
| 306 |
+
"step": 196
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 3.5,
|
| 310 |
+
"eval_bleu": 12.34433922960259,
|
| 311 |
+
"eval_exact_match": 0.0022650056625141564,
|
| 312 |
+
"eval_prefix_exact_match": 0.0022650056625141564,
|
| 313 |
+
"step": 196
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 3.54,
|
| 317 |
+
"learning_rate": 3e-05,
|
| 318 |
+
"loss": 0.9277,
|
| 319 |
+
"step": 198
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 3.64,
|
| 323 |
+
"learning_rate": 3e-05,
|
| 324 |
+
"loss": 0.8827,
|
| 325 |
+
"step": 204
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 3.75,
|
| 329 |
+
"learning_rate": 3e-05,
|
| 330 |
+
"loss": 0.99,
|
| 331 |
+
"step": 210
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 3.86,
|
| 335 |
+
"learning_rate": 3e-05,
|
| 336 |
+
"loss": 0.9083,
|
| 337 |
+
"step": 216
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.96,
|
| 341 |
+
"learning_rate": 3e-05,
|
| 342 |
+
"loss": 0.9056,
|
| 343 |
+
"step": 222
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 4.0,
|
| 347 |
+
"eval_accuracy": 0.7902908718310757,
|
| 348 |
+
"eval_loss": 0.7502285242080688,
|
| 349 |
+
"eval_runtime": 24.8951,
|
| 350 |
+
"eval_samples_per_second": 35.469,
|
| 351 |
+
"eval_steps_per_second": 2.249,
|
| 352 |
+
"step": 224
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 4.0,
|
| 356 |
+
"eval_bleu": 14.357948484733997,
|
| 357 |
+
"eval_exact_match": 0.0033975084937712344,
|
| 358 |
+
"eval_prefix_exact_match": 0.0033975084937712344,
|
| 359 |
+
"step": 224
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 4.07,
|
| 363 |
+
"learning_rate": 3e-05,
|
| 364 |
+
"loss": 0.7462,
|
| 365 |
+
"step": 228
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.18,
|
| 369 |
+
"learning_rate": 3e-05,
|
| 370 |
+
"loss": 0.583,
|
| 371 |
+
"step": 234
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 4.29,
|
| 375 |
+
"learning_rate": 3e-05,
|
| 376 |
+
"loss": 0.5995,
|
| 377 |
+
"step": 240
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 4.39,
|
| 381 |
+
"learning_rate": 3e-05,
|
| 382 |
+
"loss": 0.6094,
|
| 383 |
+
"step": 246
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.5,
|
| 387 |
+
"learning_rate": 3e-05,
|
| 388 |
+
"loss": 0.6131,
|
| 389 |
+
"step": 252
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.5,
|
| 393 |
+
"eval_accuracy": 0.8066557828732234,
|
| 394 |
+
"eval_loss": 0.6304216980934143,
|
| 395 |
+
"eval_runtime": 25.7927,
|
| 396 |
+
"eval_samples_per_second": 34.234,
|
| 397 |
+
"eval_steps_per_second": 2.171,
|
| 398 |
+
"step": 252
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 4.5,
|
| 402 |
+
"eval_bleu": 21.42036295397234,
|
| 403 |
+
"eval_exact_match": 0.010192525481313703,
|
| 404 |
+
"eval_prefix_exact_match": 0.010192525481313703,
|
| 405 |
+
"step": 252
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.61,
|
| 409 |
+
"learning_rate": 3e-05,
|
| 410 |
+
"loss": 0.6447,
|
| 411 |
+
"step": 258
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.71,
|
| 415 |
+
"learning_rate": 3e-05,
|
| 416 |
+
"loss": 0.6134,
|
| 417 |
+
"step": 264
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.82,
|
| 421 |
+
"learning_rate": 3e-05,
|
| 422 |
+
"loss": 0.611,
|
| 423 |
+
"step": 270
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.93,
|
| 427 |
+
"learning_rate": 3e-05,
|
| 428 |
+
"loss": 0.6101,
|
| 429 |
+
"step": 276
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 5.0,
|
| 433 |
+
"eval_accuracy": 0.82152049535968,
|
| 434 |
+
"eval_loss": 0.5429415106773376,
|
| 435 |
+
"eval_runtime": 24.6601,
|
| 436 |
+
"eval_samples_per_second": 35.807,
|
| 437 |
+
"eval_steps_per_second": 2.271,
|
| 438 |
+
"step": 280
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 5.0,
|
| 442 |
+
"eval_bleu": 26.261017838942095,
|
| 443 |
+
"eval_exact_match": 0.02944507361268403,
|
| 444 |
+
"eval_prefix_exact_match": 0.02944507361268403,
|
| 445 |
+
"step": 280
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 5.04,
|
| 449 |
+
"learning_rate": 3e-05,
|
| 450 |
+
"loss": 0.5416,
|
| 451 |
+
"step": 282
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 5.14,
|
| 455 |
+
"learning_rate": 3e-05,
|
| 456 |
+
"loss": 0.3793,
|
| 457 |
+
"step": 288
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 5.25,
|
| 461 |
+
"learning_rate": 3e-05,
|
| 462 |
+
"loss": 0.3858,
|
| 463 |
+
"step": 294
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 5.36,
|
| 467 |
+
"learning_rate": 3e-05,
|
| 468 |
+
"loss": 0.3977,
|
| 469 |
+
"step": 300
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 5.46,
|
| 473 |
+
"learning_rate": 3e-05,
|
| 474 |
+
"loss": 0.3772,
|
| 475 |
+
"step": 306
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 5.5,
|
| 479 |
+
"eval_accuracy": 0.8287371367779068,
|
| 480 |
+
"eval_loss": 0.48723191022872925,
|
| 481 |
+
"eval_runtime": 25.4066,
|
| 482 |
+
"eval_samples_per_second": 34.755,
|
| 483 |
+
"eval_steps_per_second": 2.204,
|
| 484 |
+
"step": 308
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 5.5,
|
| 488 |
+
"eval_bleu": 38.82587580024761,
|
| 489 |
+
"eval_exact_match": 0.10419026047565119,
|
| 490 |
+
"eval_prefix_exact_match": 0.10419026047565119,
|
| 491 |
+
"step": 308
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.57,
|
| 495 |
+
"learning_rate": 3e-05,
|
| 496 |
+
"loss": 0.3898,
|
| 497 |
+
"step": 312
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 5.68,
|
| 501 |
+
"learning_rate": 3e-05,
|
| 502 |
+
"loss": 0.3916,
|
| 503 |
+
"step": 318
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 5.79,
|
| 507 |
+
"learning_rate": 3e-05,
|
| 508 |
+
"loss": 0.3948,
|
| 509 |
+
"step": 324
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.89,
|
| 513 |
+
"learning_rate": 3e-05,
|
| 514 |
+
"loss": 0.4125,
|
| 515 |
+
"step": 330
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 6.0,
|
| 519 |
+
"learning_rate": 3e-05,
|
| 520 |
+
"loss": 0.4265,
|
| 521 |
+
"step": 336
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 6.0,
|
| 525 |
+
"eval_accuracy": 0.8356841346648821,
|
| 526 |
+
"eval_loss": 0.44368451833724976,
|
| 527 |
+
"eval_runtime": 25.151,
|
| 528 |
+
"eval_samples_per_second": 35.108,
|
| 529 |
+
"eval_steps_per_second": 2.227,
|
| 530 |
+
"step": 336
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 6.0,
|
| 534 |
+
"eval_bleu": 51.490761143615906,
|
| 535 |
+
"eval_exact_match": 0.16987542468856173,
|
| 536 |
+
"eval_prefix_exact_match": 0.16987542468856173,
|
| 537 |
+
"step": 336
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 6.11,
|
| 541 |
+
"learning_rate": 3e-05,
|
| 542 |
+
"loss": 0.2554,
|
| 543 |
+
"step": 342
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 6.21,
|
| 547 |
+
"learning_rate": 3e-05,
|
| 548 |
+
"loss": 0.2562,
|
| 549 |
+
"step": 348
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 6.32,
|
| 553 |
+
"learning_rate": 3e-05,
|
| 554 |
+
"loss": 0.2531,
|
| 555 |
+
"step": 354
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 6.43,
|
| 559 |
+
"learning_rate": 3e-05,
|
| 560 |
+
"loss": 0.2552,
|
| 561 |
+
"step": 360
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 6.5,
|
| 565 |
+
"eval_accuracy": 0.8388953439916067,
|
| 566 |
+
"eval_loss": 0.4225637912750244,
|
| 567 |
+
"eval_runtime": 25.2976,
|
| 568 |
+
"eval_samples_per_second": 34.905,
|
| 569 |
+
"eval_steps_per_second": 2.214,
|
| 570 |
+
"step": 364
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 6.5,
|
| 574 |
+
"eval_bleu": 66.56501737677819,
|
| 575 |
+
"eval_exact_match": 0.3012457531143828,
|
| 576 |
+
"eval_prefix_exact_match": 0.3012457531143828,
|
| 577 |
+
"step": 364
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 6.54,
|
| 581 |
+
"learning_rate": 3e-05,
|
| 582 |
+
"loss": 0.2536,
|
| 583 |
+
"step": 366
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 6.64,
|
| 587 |
+
"learning_rate": 3e-05,
|
| 588 |
+
"loss": 0.2617,
|
| 589 |
+
"step": 372
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 6.75,
|
| 593 |
+
"learning_rate": 3e-05,
|
| 594 |
+
"loss": 0.2763,
|
| 595 |
+
"step": 378
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.86,
|
| 599 |
+
"learning_rate": 3e-05,
|
| 600 |
+
"loss": 0.2614,
|
| 601 |
+
"step": 384
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 6.96,
|
| 605 |
+
"learning_rate": 3e-05,
|
| 606 |
+
"loss": 0.2875,
|
| 607 |
+
"step": 390
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 7.0,
|
| 611 |
+
"eval_accuracy": 0.8417535654228746,
|
| 612 |
+
"eval_loss": 0.4019472599029541,
|
| 613 |
+
"eval_runtime": 24.4914,
|
| 614 |
+
"eval_samples_per_second": 36.053,
|
| 615 |
+
"eval_steps_per_second": 2.287,
|
| 616 |
+
"step": 392
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 7.0,
|
| 620 |
+
"eval_bleu": 73.8359090816191,
|
| 621 |
+
"eval_exact_match": 0.3941109852774632,
|
| 622 |
+
"eval_prefix_exact_match": 0.3941109852774632,
|
| 623 |
+
"step": 392
|
| 624 |
+
}
|
| 625 |
+
],
|
| 626 |
+
"logging_steps": 6,
|
| 627 |
+
"max_steps": 560,
|
| 628 |
+
"num_train_epochs": 10,
|
| 629 |
+
"save_steps": 500,
|
| 630 |
+
"total_flos": 9579770868940800.0,
|
| 631 |
+
"trial_name": null,
|
| 632 |
+
"trial_params": null
|
| 633 |
+
}
|
checkpoint-392/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8b959b77797bfa32386725e4d78839c39c28940487e69e41fb564f8bb31a78b3
|
| 3 |
+
size 4664
|
checkpoint-392/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|