Training in progress, epoch 9, checkpoint
Browse files- checkpoint-1179/added_tokens.json +3 -0
- checkpoint-1179/config.json +42 -0
- checkpoint-1179/generation_config.json +8 -0
- checkpoint-1179/merges.txt +0 -0
- checkpoint-1179/optimizer.pt +3 -0
- checkpoint-1179/pytorch_model.bin +3 -0
- checkpoint-1179/rng_state.pth +3 -0
- checkpoint-1179/scheduler.pt +3 -0
- checkpoint-1179/special_tokens_map.json +6 -0
- checkpoint-1179/tokenizer.json +0 -0
- checkpoint-1179/tokenizer_config.json +20 -0
- checkpoint-1179/trainer_state.json +846 -0
- checkpoint-1179/training_args.bin +3 -0
- checkpoint-1179/vocab.json +0 -0
checkpoint-1179/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 50256
|
| 3 |
+
}
|
checkpoint-1179/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-1179/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-1179/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1179/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c00c409e414ba8bfd0de184b01885de203c30ccb66684be9ea107e09824f97a4
|
| 3 |
+
size 12461385454
|
checkpoint-1179/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5d540c198aeed6709c9c422e150e88c3ad5a9dd48c432064a6ae447ea4f13c79
|
| 3 |
+
size 6230637102
|
checkpoint-1179/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91331de4a996849edb488c72af532dc366b219ef9e393ae825630f2696519739
|
| 3 |
+
size 14244
|
checkpoint-1179/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e98453e4059c93abdbfc0047ac1669310799b1753d0ef96bbc94280598ab0d6a
|
| 3 |
+
size 1064
|
checkpoint-1179/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-1179/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1179/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-1179/trainer_state.json
ADDED
|
@@ -0,0 +1,846 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.0,
|
| 5 |
+
"eval_steps": 66,
|
| 6 |
+
"global_step": 1179,
|
| 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": 3.7331,
|
| 15 |
+
"step": 14
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.21,
|
| 19 |
+
"learning_rate": 3e-05,
|
| 20 |
+
"loss": 2.345,
|
| 21 |
+
"step": 28
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.32,
|
| 25 |
+
"learning_rate": 3e-05,
|
| 26 |
+
"loss": 2.1491,
|
| 27 |
+
"step": 42
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.43,
|
| 31 |
+
"learning_rate": 3e-05,
|
| 32 |
+
"loss": 2.0739,
|
| 33 |
+
"step": 56
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.5,
|
| 37 |
+
"eval_accuracy": 0.6206453178068898,
|
| 38 |
+
"eval_loss": 1.8404711484909058,
|
| 39 |
+
"eval_runtime": 11.9883,
|
| 40 |
+
"eval_samples_per_second": 25.024,
|
| 41 |
+
"eval_steps_per_second": 1.585,
|
| 42 |
+
"step": 66
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.5,
|
| 46 |
+
"eval_exact_match": 7.666666666666667,
|
| 47 |
+
"eval_f1": 10.221428571428572,
|
| 48 |
+
"eval_qa_bleu": 1.4042262195131967,
|
| 49 |
+
"eval_qa_exact_match": 0.07333333333333333,
|
| 50 |
+
"eval_recite_bleu": 8.5956480576491,
|
| 51 |
+
"eval_recite_exact_match": 0.0,
|
| 52 |
+
"step": 66
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.53,
|
| 56 |
+
"learning_rate": 3e-05,
|
| 57 |
+
"loss": 1.9722,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.64,
|
| 62 |
+
"learning_rate": 3e-05,
|
| 63 |
+
"loss": 2.014,
|
| 64 |
+
"step": 84
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.75,
|
| 68 |
+
"learning_rate": 3e-05,
|
| 69 |
+
"loss": 1.9812,
|
| 70 |
+
"step": 98
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.85,
|
| 74 |
+
"learning_rate": 3e-05,
|
| 75 |
+
"loss": 1.9213,
|
| 76 |
+
"step": 112
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.96,
|
| 80 |
+
"learning_rate": 3e-05,
|
| 81 |
+
"loss": 1.8806,
|
| 82 |
+
"step": 126
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 1.01,
|
| 86 |
+
"eval_accuracy": 0.6365065502183406,
|
| 87 |
+
"eval_loss": 1.574863314628601,
|
| 88 |
+
"eval_runtime": 9.1867,
|
| 89 |
+
"eval_samples_per_second": 32.656,
|
| 90 |
+
"eval_steps_per_second": 2.068,
|
| 91 |
+
"step": 132
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.01,
|
| 95 |
+
"eval_exact_match": 8.666666666666666,
|
| 96 |
+
"eval_f1": 13.8123783922171,
|
| 97 |
+
"eval_qa_bleu": 3.0725240037081307,
|
| 98 |
+
"eval_qa_exact_match": 0.08666666666666667,
|
| 99 |
+
"eval_recite_bleu": 16.386073781113847,
|
| 100 |
+
"eval_recite_exact_match": 0.0,
|
| 101 |
+
"step": 132
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.07,
|
| 105 |
+
"learning_rate": 3e-05,
|
| 106 |
+
"loss": 1.6177,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.18,
|
| 111 |
+
"learning_rate": 3e-05,
|
| 112 |
+
"loss": 1.4286,
|
| 113 |
+
"step": 154
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.28,
|
| 117 |
+
"learning_rate": 3e-05,
|
| 118 |
+
"loss": 1.4102,
|
| 119 |
+
"step": 168
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.39,
|
| 123 |
+
"learning_rate": 3e-05,
|
| 124 |
+
"loss": 1.4033,
|
| 125 |
+
"step": 182
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.5,
|
| 129 |
+
"learning_rate": 3e-05,
|
| 130 |
+
"loss": 1.3619,
|
| 131 |
+
"step": 196
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.51,
|
| 135 |
+
"eval_accuracy": 0.6533139252789908,
|
| 136 |
+
"eval_loss": 1.342494010925293,
|
| 137 |
+
"eval_runtime": 9.3165,
|
| 138 |
+
"eval_samples_per_second": 32.201,
|
| 139 |
+
"eval_steps_per_second": 2.039,
|
| 140 |
+
"step": 198
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 1.51,
|
| 144 |
+
"eval_exact_match": 7.0,
|
| 145 |
+
"eval_f1": 10.38888888888889,
|
| 146 |
+
"eval_qa_bleu": 5.590724094958645,
|
| 147 |
+
"eval_qa_exact_match": 0.06666666666666667,
|
| 148 |
+
"eval_recite_bleu": 17.070052559724132,
|
| 149 |
+
"eval_recite_exact_match": 0.0,
|
| 150 |
+
"step": 198
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.6,
|
| 154 |
+
"learning_rate": 3e-05,
|
| 155 |
+
"loss": 1.3649,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.71,
|
| 160 |
+
"learning_rate": 3e-05,
|
| 161 |
+
"loss": 1.2886,
|
| 162 |
+
"step": 224
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.82,
|
| 166 |
+
"learning_rate": 3e-05,
|
| 167 |
+
"loss": 1.2787,
|
| 168 |
+
"step": 238
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.92,
|
| 172 |
+
"learning_rate": 3e-05,
|
| 173 |
+
"loss": 1.283,
|
| 174 |
+
"step": 252
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.02,
|
| 178 |
+
"eval_accuracy": 0.6685152838427948,
|
| 179 |
+
"eval_loss": 1.1253471374511719,
|
| 180 |
+
"eval_runtime": 9.3443,
|
| 181 |
+
"eval_samples_per_second": 32.105,
|
| 182 |
+
"eval_steps_per_second": 2.033,
|
| 183 |
+
"step": 264
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.02,
|
| 187 |
+
"eval_exact_match": 9.0,
|
| 188 |
+
"eval_f1": 13.042195767195768,
|
| 189 |
+
"eval_qa_bleu": 5.743589453455261,
|
| 190 |
+
"eval_qa_exact_match": 0.08666666666666667,
|
| 191 |
+
"eval_recite_bleu": 15.4059437725803,
|
| 192 |
+
"eval_recite_exact_match": 0.0,
|
| 193 |
+
"step": 264
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.03,
|
| 197 |
+
"learning_rate": 3e-05,
|
| 198 |
+
"loss": 1.112,
|
| 199 |
+
"step": 266
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 2.14,
|
| 203 |
+
"learning_rate": 3e-05,
|
| 204 |
+
"loss": 0.8377,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.24,
|
| 209 |
+
"learning_rate": 3e-05,
|
| 210 |
+
"loss": 0.8533,
|
| 211 |
+
"step": 294
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 2.35,
|
| 215 |
+
"learning_rate": 3e-05,
|
| 216 |
+
"loss": 0.8407,
|
| 217 |
+
"step": 308
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 2.46,
|
| 221 |
+
"learning_rate": 3e-05,
|
| 222 |
+
"loss": 0.8433,
|
| 223 |
+
"step": 322
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.52,
|
| 227 |
+
"eval_accuracy": 0.6825278990781174,
|
| 228 |
+
"eval_loss": 0.9735248684883118,
|
| 229 |
+
"eval_runtime": 9.3401,
|
| 230 |
+
"eval_samples_per_second": 32.12,
|
| 231 |
+
"eval_steps_per_second": 2.034,
|
| 232 |
+
"step": 330
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 2.52,
|
| 236 |
+
"eval_exact_match": 11.666666666666666,
|
| 237 |
+
"eval_f1": 15.95859788359788,
|
| 238 |
+
"eval_qa_bleu": 6.31601664322449,
|
| 239 |
+
"eval_qa_exact_match": 0.11,
|
| 240 |
+
"eval_recite_bleu": 21.285839887136575,
|
| 241 |
+
"eval_recite_exact_match": 0.0,
|
| 242 |
+
"step": 330
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 2.56,
|
| 246 |
+
"learning_rate": 3e-05,
|
| 247 |
+
"loss": 0.8632,
|
| 248 |
+
"step": 336
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.67,
|
| 252 |
+
"learning_rate": 3e-05,
|
| 253 |
+
"loss": 0.8308,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.78,
|
| 258 |
+
"learning_rate": 3e-05,
|
| 259 |
+
"loss": 0.8475,
|
| 260 |
+
"step": 364
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.89,
|
| 264 |
+
"learning_rate": 3e-05,
|
| 265 |
+
"loss": 0.7829,
|
| 266 |
+
"step": 378
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 2.99,
|
| 270 |
+
"learning_rate": 3e-05,
|
| 271 |
+
"loss": 0.7629,
|
| 272 |
+
"step": 392
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.02,
|
| 276 |
+
"eval_accuracy": 0.6982241630276564,
|
| 277 |
+
"eval_loss": 0.7873561382293701,
|
| 278 |
+
"eval_runtime": 9.0856,
|
| 279 |
+
"eval_samples_per_second": 33.019,
|
| 280 |
+
"eval_steps_per_second": 2.091,
|
| 281 |
+
"step": 396
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 3.02,
|
| 285 |
+
"eval_exact_match": 12.333333333333334,
|
| 286 |
+
"eval_f1": 16.93478835978836,
|
| 287 |
+
"eval_qa_bleu": 7.7589360856745815,
|
| 288 |
+
"eval_qa_exact_match": 0.12,
|
| 289 |
+
"eval_recite_bleu": 22.16978951972916,
|
| 290 |
+
"eval_recite_exact_match": 0.0,
|
| 291 |
+
"step": 396
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 3.1,
|
| 295 |
+
"learning_rate": 3e-05,
|
| 296 |
+
"loss": 0.5537,
|
| 297 |
+
"step": 406
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 3.21,
|
| 301 |
+
"learning_rate": 3e-05,
|
| 302 |
+
"loss": 0.5111,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 3.31,
|
| 307 |
+
"learning_rate": 3e-05,
|
| 308 |
+
"loss": 0.479,
|
| 309 |
+
"step": 434
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.42,
|
| 313 |
+
"learning_rate": 3e-05,
|
| 314 |
+
"loss": 0.4904,
|
| 315 |
+
"step": 448
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 3.53,
|
| 319 |
+
"learning_rate": 3e-05,
|
| 320 |
+
"loss": 0.5058,
|
| 321 |
+
"step": 462
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.53,
|
| 325 |
+
"eval_accuracy": 0.7086026200873362,
|
| 326 |
+
"eval_loss": 0.6920613646507263,
|
| 327 |
+
"eval_runtime": 9.3473,
|
| 328 |
+
"eval_samples_per_second": 32.095,
|
| 329 |
+
"eval_steps_per_second": 2.033,
|
| 330 |
+
"step": 462
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.53,
|
| 334 |
+
"eval_exact_match": 15.333333333333334,
|
| 335 |
+
"eval_f1": 20.8812384782973,
|
| 336 |
+
"eval_qa_bleu": 11.001436952878551,
|
| 337 |
+
"eval_qa_exact_match": 0.14,
|
| 338 |
+
"eval_recite_bleu": 25.319328021621963,
|
| 339 |
+
"eval_recite_exact_match": 0.0,
|
| 340 |
+
"step": 462
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 3.63,
|
| 344 |
+
"learning_rate": 3e-05,
|
| 345 |
+
"loss": 0.5456,
|
| 346 |
+
"step": 476
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 3.74,
|
| 350 |
+
"learning_rate": 3e-05,
|
| 351 |
+
"loss": 0.4998,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.85,
|
| 356 |
+
"learning_rate": 3e-05,
|
| 357 |
+
"loss": 0.5075,
|
| 358 |
+
"step": 504
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.95,
|
| 362 |
+
"learning_rate": 3e-05,
|
| 363 |
+
"loss": 0.4593,
|
| 364 |
+
"step": 518
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 4.03,
|
| 368 |
+
"eval_accuracy": 0.7196943231441048,
|
| 369 |
+
"eval_loss": 0.564062237739563,
|
| 370 |
+
"eval_runtime": 9.2784,
|
| 371 |
+
"eval_samples_per_second": 32.333,
|
| 372 |
+
"eval_steps_per_second": 2.048,
|
| 373 |
+
"step": 528
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 4.03,
|
| 377 |
+
"eval_exact_match": 13.333333333333334,
|
| 378 |
+
"eval_f1": 18.11966301672184,
|
| 379 |
+
"eval_qa_bleu": 4.47266646700781,
|
| 380 |
+
"eval_qa_exact_match": 0.12666666666666668,
|
| 381 |
+
"eval_recite_bleu": 26.136161725029392,
|
| 382 |
+
"eval_recite_exact_match": 0.0033333333333333335,
|
| 383 |
+
"step": 528
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 4.06,
|
| 387 |
+
"learning_rate": 3e-05,
|
| 388 |
+
"loss": 0.3952,
|
| 389 |
+
"step": 532
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 4.17,
|
| 393 |
+
"learning_rate": 3e-05,
|
| 394 |
+
"loss": 0.2996,
|
| 395 |
+
"step": 546
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.27,
|
| 399 |
+
"learning_rate": 3e-05,
|
| 400 |
+
"loss": 0.3354,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 4.38,
|
| 405 |
+
"learning_rate": 3e-05,
|
| 406 |
+
"loss": 0.3179,
|
| 407 |
+
"step": 574
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 4.49,
|
| 411 |
+
"learning_rate": 3e-05,
|
| 412 |
+
"loss": 0.3064,
|
| 413 |
+
"step": 588
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.53,
|
| 417 |
+
"eval_accuracy": 0.7245269286754003,
|
| 418 |
+
"eval_loss": 0.5348330736160278,
|
| 419 |
+
"eval_runtime": 9.5519,
|
| 420 |
+
"eval_samples_per_second": 31.407,
|
| 421 |
+
"eval_steps_per_second": 1.989,
|
| 422 |
+
"step": 594
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 4.53,
|
| 426 |
+
"eval_exact_match": 13.666666666666666,
|
| 427 |
+
"eval_f1": 19.689682539682536,
|
| 428 |
+
"eval_qa_bleu": 4.45260487005976,
|
| 429 |
+
"eval_qa_exact_match": 0.12666666666666668,
|
| 430 |
+
"eval_recite_bleu": 28.855187002245795,
|
| 431 |
+
"eval_recite_exact_match": 0.0,
|
| 432 |
+
"step": 594
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 4.6,
|
| 436 |
+
"learning_rate": 3e-05,
|
| 437 |
+
"loss": 0.3258,
|
| 438 |
+
"step": 602
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 4.7,
|
| 442 |
+
"learning_rate": 3e-05,
|
| 443 |
+
"loss": 0.3384,
|
| 444 |
+
"step": 616
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.81,
|
| 448 |
+
"learning_rate": 3e-05,
|
| 449 |
+
"loss": 0.3258,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 4.92,
|
| 454 |
+
"learning_rate": 3e-05,
|
| 455 |
+
"loss": 0.312,
|
| 456 |
+
"step": 644
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 5.02,
|
| 460 |
+
"learning_rate": 3e-05,
|
| 461 |
+
"loss": 0.2967,
|
| 462 |
+
"step": 658
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 5.04,
|
| 466 |
+
"eval_accuracy": 0.7303784570596797,
|
| 467 |
+
"eval_loss": 0.47703343629837036,
|
| 468 |
+
"eval_runtime": 9.4774,
|
| 469 |
+
"eval_samples_per_second": 31.654,
|
| 470 |
+
"eval_steps_per_second": 2.005,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 5.04,
|
| 475 |
+
"eval_exact_match": 12.333333333333334,
|
| 476 |
+
"eval_f1": 18.61798941798941,
|
| 477 |
+
"eval_qa_bleu": 6.60839076276961,
|
| 478 |
+
"eval_qa_exact_match": 0.12,
|
| 479 |
+
"eval_recite_bleu": 29.5293953590396,
|
| 480 |
+
"eval_recite_exact_match": 0.0,
|
| 481 |
+
"step": 660
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 5.13,
|
| 485 |
+
"learning_rate": 3e-05,
|
| 486 |
+
"loss": 0.2148,
|
| 487 |
+
"step": 672
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 5.24,
|
| 491 |
+
"learning_rate": 3e-05,
|
| 492 |
+
"loss": 0.2275,
|
| 493 |
+
"step": 686
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 5.34,
|
| 497 |
+
"learning_rate": 3e-05,
|
| 498 |
+
"loss": 0.2158,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 5.45,
|
| 503 |
+
"learning_rate": 3e-05,
|
| 504 |
+
"loss": 0.2167,
|
| 505 |
+
"step": 714
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.54,
|
| 509 |
+
"eval_accuracy": 0.732372634643377,
|
| 510 |
+
"eval_loss": 0.458192378282547,
|
| 511 |
+
"eval_runtime": 9.1255,
|
| 512 |
+
"eval_samples_per_second": 32.875,
|
| 513 |
+
"eval_steps_per_second": 2.082,
|
| 514 |
+
"step": 726
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 5.54,
|
| 518 |
+
"eval_exact_match": 14.333333333333334,
|
| 519 |
+
"eval_f1": 19.81243386243386,
|
| 520 |
+
"eval_qa_bleu": 8.961623752889384,
|
| 521 |
+
"eval_qa_exact_match": 0.13666666666666666,
|
| 522 |
+
"eval_recite_bleu": 33.41904095334099,
|
| 523 |
+
"eval_recite_exact_match": 0.0,
|
| 524 |
+
"step": 726
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 5.56,
|
| 528 |
+
"learning_rate": 3e-05,
|
| 529 |
+
"loss": 0.229,
|
| 530 |
+
"step": 728
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 5.66,
|
| 534 |
+
"learning_rate": 3e-05,
|
| 535 |
+
"loss": 0.2275,
|
| 536 |
+
"step": 742
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 5.77,
|
| 540 |
+
"learning_rate": 3e-05,
|
| 541 |
+
"loss": 0.2211,
|
| 542 |
+
"step": 756
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 5.88,
|
| 546 |
+
"learning_rate": 3e-05,
|
| 547 |
+
"loss": 0.2231,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 5.98,
|
| 552 |
+
"learning_rate": 3e-05,
|
| 553 |
+
"loss": 0.2157,
|
| 554 |
+
"step": 784
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 6.05,
|
| 558 |
+
"eval_accuracy": 0.7358369723435225,
|
| 559 |
+
"eval_loss": 0.4307834804058075,
|
| 560 |
+
"eval_runtime": 9.1847,
|
| 561 |
+
"eval_samples_per_second": 32.663,
|
| 562 |
+
"eval_steps_per_second": 2.069,
|
| 563 |
+
"step": 792
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 6.05,
|
| 567 |
+
"eval_exact_match": 16.666666666666668,
|
| 568 |
+
"eval_f1": 21.786772486772488,
|
| 569 |
+
"eval_qa_bleu": 7.511563755726586,
|
| 570 |
+
"eval_qa_exact_match": 0.16,
|
| 571 |
+
"eval_recite_bleu": 37.63384023220464,
|
| 572 |
+
"eval_recite_exact_match": 0.013333333333333334,
|
| 573 |
+
"step": 792
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 6.09,
|
| 577 |
+
"learning_rate": 3e-05,
|
| 578 |
+
"loss": 0.1669,
|
| 579 |
+
"step": 798
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 6.2,
|
| 583 |
+
"learning_rate": 3e-05,
|
| 584 |
+
"loss": 0.1712,
|
| 585 |
+
"step": 812
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 6.31,
|
| 589 |
+
"learning_rate": 3e-05,
|
| 590 |
+
"loss": 0.1601,
|
| 591 |
+
"step": 826
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 6.41,
|
| 595 |
+
"learning_rate": 3e-05,
|
| 596 |
+
"loss": 0.1608,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.52,
|
| 601 |
+
"learning_rate": 3e-05,
|
| 602 |
+
"loss": 0.1597,
|
| 603 |
+
"step": 854
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 6.55,
|
| 607 |
+
"eval_accuracy": 0.7373410965550704,
|
| 608 |
+
"eval_loss": 0.4301389157772064,
|
| 609 |
+
"eval_runtime": 9.7372,
|
| 610 |
+
"eval_samples_per_second": 30.81,
|
| 611 |
+
"eval_steps_per_second": 1.951,
|
| 612 |
+
"step": 858
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 6.55,
|
| 616 |
+
"eval_exact_match": 15.666666666666666,
|
| 617 |
+
"eval_f1": 21.056661856661847,
|
| 618 |
+
"eval_qa_bleu": 12.649140852831426,
|
| 619 |
+
"eval_qa_exact_match": 0.14666666666666667,
|
| 620 |
+
"eval_recite_bleu": 42.22665248887737,
|
| 621 |
+
"eval_recite_exact_match": 0.013333333333333334,
|
| 622 |
+
"step": 858
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 6.63,
|
| 626 |
+
"learning_rate": 3e-05,
|
| 627 |
+
"loss": 0.1623,
|
| 628 |
+
"step": 868
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 6.73,
|
| 632 |
+
"learning_rate": 3e-05,
|
| 633 |
+
"loss": 0.1668,
|
| 634 |
+
"step": 882
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 6.84,
|
| 638 |
+
"learning_rate": 3e-05,
|
| 639 |
+
"loss": 0.1624,
|
| 640 |
+
"step": 896
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 6.95,
|
| 644 |
+
"learning_rate": 3e-05,
|
| 645 |
+
"loss": 0.1648,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 7.05,
|
| 650 |
+
"learning_rate": 3e-05,
|
| 651 |
+
"loss": 0.1481,
|
| 652 |
+
"step": 924
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 7.05,
|
| 656 |
+
"eval_accuracy": 0.7385298398835517,
|
| 657 |
+
"eval_loss": 0.42236796021461487,
|
| 658 |
+
"eval_runtime": 9.3603,
|
| 659 |
+
"eval_samples_per_second": 32.05,
|
| 660 |
+
"eval_steps_per_second": 2.03,
|
| 661 |
+
"step": 924
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 7.05,
|
| 665 |
+
"eval_exact_match": 18.666666666666668,
|
| 666 |
+
"eval_f1": 25.187830687830694,
|
| 667 |
+
"eval_qa_bleu": 8.014835952265651,
|
| 668 |
+
"eval_qa_exact_match": 0.18,
|
| 669 |
+
"eval_recite_bleu": 42.38987173856079,
|
| 670 |
+
"eval_recite_exact_match": 0.016666666666666666,
|
| 671 |
+
"step": 924
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 7.16,
|
| 675 |
+
"learning_rate": 3e-05,
|
| 676 |
+
"loss": 0.1227,
|
| 677 |
+
"step": 938
|
| 678 |
+
},
|
| 679 |
+
{
|
| 680 |
+
"epoch": 7.27,
|
| 681 |
+
"learning_rate": 3e-05,
|
| 682 |
+
"loss": 0.1272,
|
| 683 |
+
"step": 952
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 7.37,
|
| 687 |
+
"learning_rate": 3e-05,
|
| 688 |
+
"loss": 0.1312,
|
| 689 |
+
"step": 966
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 7.48,
|
| 693 |
+
"learning_rate": 3e-05,
|
| 694 |
+
"loss": 0.1293,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.56,
|
| 699 |
+
"eval_accuracy": 0.739422610383309,
|
| 700 |
+
"eval_loss": 0.41248488426208496,
|
| 701 |
+
"eval_runtime": 9.7486,
|
| 702 |
+
"eval_samples_per_second": 30.774,
|
| 703 |
+
"eval_steps_per_second": 1.949,
|
| 704 |
+
"step": 990
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 7.56,
|
| 708 |
+
"eval_exact_match": 15.666666666666666,
|
| 709 |
+
"eval_f1": 22.348196248196246,
|
| 710 |
+
"eval_qa_bleu": 6.260683683577316,
|
| 711 |
+
"eval_qa_exact_match": 0.14666666666666667,
|
| 712 |
+
"eval_recite_bleu": 45.04221504063147,
|
| 713 |
+
"eval_recite_exact_match": 0.023333333333333334,
|
| 714 |
+
"step": 990
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 7.59,
|
| 718 |
+
"learning_rate": 3e-05,
|
| 719 |
+
"loss": 0.1345,
|
| 720 |
+
"step": 994
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 7.69,
|
| 724 |
+
"learning_rate": 3e-05,
|
| 725 |
+
"loss": 0.1325,
|
| 726 |
+
"step": 1008
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 7.8,
|
| 730 |
+
"learning_rate": 3e-05,
|
| 731 |
+
"loss": 0.1273,
|
| 732 |
+
"step": 1022
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 7.91,
|
| 736 |
+
"learning_rate": 3e-05,
|
| 737 |
+
"loss": 0.1362,
|
| 738 |
+
"step": 1036
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 8.02,
|
| 742 |
+
"learning_rate": 3e-05,
|
| 743 |
+
"loss": 0.125,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 8.06,
|
| 748 |
+
"eval_accuracy": 0.7399902959728287,
|
| 749 |
+
"eval_loss": 0.41223272681236267,
|
| 750 |
+
"eval_runtime": 9.2562,
|
| 751 |
+
"eval_samples_per_second": 32.411,
|
| 752 |
+
"eval_steps_per_second": 2.053,
|
| 753 |
+
"step": 1056
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 8.06,
|
| 757 |
+
"eval_exact_match": 18.0,
|
| 758 |
+
"eval_f1": 25.051058201058197,
|
| 759 |
+
"eval_qa_bleu": 10.352483602423003,
|
| 760 |
+
"eval_qa_exact_match": 0.17,
|
| 761 |
+
"eval_recite_bleu": 46.26228887496748,
|
| 762 |
+
"eval_recite_exact_match": 0.04666666666666667,
|
| 763 |
+
"step": 1056
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 8.12,
|
| 767 |
+
"learning_rate": 3e-05,
|
| 768 |
+
"loss": 0.11,
|
| 769 |
+
"step": 1064
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 8.23,
|
| 773 |
+
"learning_rate": 3e-05,
|
| 774 |
+
"loss": 0.1045,
|
| 775 |
+
"step": 1078
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 8.34,
|
| 779 |
+
"learning_rate": 3e-05,
|
| 780 |
+
"loss": 0.1179,
|
| 781 |
+
"step": 1092
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 8.44,
|
| 785 |
+
"learning_rate": 3e-05,
|
| 786 |
+
"loss": 0.1083,
|
| 787 |
+
"step": 1106
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 8.55,
|
| 791 |
+
"learning_rate": 3e-05,
|
| 792 |
+
"loss": 0.1139,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 8.56,
|
| 797 |
+
"eval_accuracy": 0.7406501698204755,
|
| 798 |
+
"eval_loss": 0.40691348910331726,
|
| 799 |
+
"eval_runtime": 9.0014,
|
| 800 |
+
"eval_samples_per_second": 33.328,
|
| 801 |
+
"eval_steps_per_second": 2.111,
|
| 802 |
+
"step": 1122
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 8.56,
|
| 806 |
+
"eval_exact_match": 16.333333333333332,
|
| 807 |
+
"eval_f1": 23.766955266955264,
|
| 808 |
+
"eval_qa_bleu": 11.966740392922118,
|
| 809 |
+
"eval_qa_exact_match": 0.15333333333333332,
|
| 810 |
+
"eval_recite_bleu": 45.74271226096357,
|
| 811 |
+
"eval_recite_exact_match": 0.03333333333333333,
|
| 812 |
+
"step": 1122
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 8.66,
|
| 816 |
+
"learning_rate": 3e-05,
|
| 817 |
+
"loss": 0.1108,
|
| 818 |
+
"step": 1134
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 8.76,
|
| 822 |
+
"learning_rate": 3e-05,
|
| 823 |
+
"loss": 0.1168,
|
| 824 |
+
"step": 1148
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 8.87,
|
| 828 |
+
"learning_rate": 3e-05,
|
| 829 |
+
"loss": 0.1132,
|
| 830 |
+
"step": 1162
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 8.98,
|
| 834 |
+
"learning_rate": 3e-05,
|
| 835 |
+
"loss": 0.1141,
|
| 836 |
+
"step": 1176
|
| 837 |
+
}
|
| 838 |
+
],
|
| 839 |
+
"logging_steps": 14,
|
| 840 |
+
"max_steps": 1310,
|
| 841 |
+
"num_train_epochs": 10,
|
| 842 |
+
"save_steps": 500,
|
| 843 |
+
"total_flos": 5.35677023130624e+16,
|
| 844 |
+
"trial_name": null,
|
| 845 |
+
"trial_params": null
|
| 846 |
+
}
|
checkpoint-1179/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8e2f5bbca2ac6551a76ff4a6e10000fae190ff8cb3817e74f2e51a104a179e48
|
| 3 |
+
size 4728
|
checkpoint-1179/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|