Training in progress, epoch 19, checkpoint
Browse files- checkpoint-41401/added_tokens.json +3 -0
- checkpoint-41401/config.json +42 -0
- checkpoint-41401/generation_config.json +8 -0
- checkpoint-41401/merges.txt +0 -0
- checkpoint-41401/optimizer.pt +3 -0
- checkpoint-41401/pytorch_model.bin +3 -0
- checkpoint-41401/rng_state.pth +3 -0
- checkpoint-41401/scheduler.pt +3 -0
- checkpoint-41401/special_tokens_map.json +6 -0
- checkpoint-41401/tokenizer.json +0 -0
- checkpoint-41401/tokenizer_config.json +20 -0
- checkpoint-41401/trainer_state.json +944 -0
- checkpoint-41401/training_args.bin +3 -0
- checkpoint-41401/vocab.json +0 -0
checkpoint-41401/added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 50256
|
| 3 |
+
}
|
checkpoint-41401/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-41401/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-41401/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-41401/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26b99ec7bd51559cb7a53991997fea2cd2a8a048c4a6b8abf35b6d528b48f3ea
|
| 3 |
+
size 12461385454
|
checkpoint-41401/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a6c29f30fff01cf9d254c6ae544e0b0bbbaff291693437e9154d1cf1ecdb5423
|
| 3 |
+
size 6230637102
|
checkpoint-41401/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:77a06f70a46fa56d544a6e51e1f6a8f2b494776cce7509377e1fc61f891eecdd
|
| 3 |
+
size 14244
|
checkpoint-41401/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f1a066503e9e4aa58eadd310a0b6c40fffff5347f19248bcc5554d6bf3f3263b
|
| 3 |
+
size 1064
|
checkpoint-41401/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-41401/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-41401/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-41401/trainer_state.json
ADDED
|
@@ -0,0 +1,944 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 19.0,
|
| 5 |
+
"eval_steps": 2179,
|
| 6 |
+
"global_step": 41401,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2,
|
| 13 |
+
"learning_rate": 3e-05,
|
| 14 |
+
"loss": 2.2464,
|
| 15 |
+
"step": 436
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.4,
|
| 19 |
+
"learning_rate": 3e-05,
|
| 20 |
+
"loss": 2.0544,
|
| 21 |
+
"step": 872
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.6,
|
| 25 |
+
"learning_rate": 3e-05,
|
| 26 |
+
"loss": 1.9968,
|
| 27 |
+
"step": 1308
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.8,
|
| 31 |
+
"learning_rate": 3e-05,
|
| 32 |
+
"loss": 1.9537,
|
| 33 |
+
"step": 1744
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 1.0,
|
| 37 |
+
"eval_accuracy": 0.6693020521036767,
|
| 38 |
+
"eval_loss": 1.5623117685317993,
|
| 39 |
+
"eval_runtime": 220.2436,
|
| 40 |
+
"eval_samples_per_second": 33.622,
|
| 41 |
+
"eval_steps_per_second": 2.102,
|
| 42 |
+
"step": 2179
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.0,
|
| 46 |
+
"eval_exact_match": 10.24983119513842,
|
| 47 |
+
"eval_f1": 15.257097040945366,
|
| 48 |
+
"eval_qa_bleu": 5.8382901471491575,
|
| 49 |
+
"eval_qa_exact_match": 0.09736664415935178,
|
| 50 |
+
"eval_recite_bleu": 17.0930417063702,
|
| 51 |
+
"eval_recite_exact_match": 0.0,
|
| 52 |
+
"step": 2179
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.0,
|
| 56 |
+
"learning_rate": 3e-05,
|
| 57 |
+
"loss": 1.9029,
|
| 58 |
+
"step": 2180
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 1.2,
|
| 62 |
+
"learning_rate": 3e-05,
|
| 63 |
+
"loss": 1.5215,
|
| 64 |
+
"step": 2616
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 1.4,
|
| 68 |
+
"learning_rate": 3e-05,
|
| 69 |
+
"loss": 1.5077,
|
| 70 |
+
"step": 3052
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 1.6,
|
| 74 |
+
"learning_rate": 3e-05,
|
| 75 |
+
"loss": 1.5016,
|
| 76 |
+
"step": 3488
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 1.8,
|
| 80 |
+
"learning_rate": 3e-05,
|
| 81 |
+
"loss": 1.4734,
|
| 82 |
+
"step": 3924
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 2.0,
|
| 86 |
+
"eval_accuracy": 0.6923618412861704,
|
| 87 |
+
"eval_loss": 1.2099343538284302,
|
| 88 |
+
"eval_runtime": 222.1542,
|
| 89 |
+
"eval_samples_per_second": 33.333,
|
| 90 |
+
"eval_steps_per_second": 2.084,
|
| 91 |
+
"step": 4358
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 2.0,
|
| 95 |
+
"eval_exact_match": 10.61444969615125,
|
| 96 |
+
"eval_f1": 15.935672519396675,
|
| 97 |
+
"eval_qa_bleu": 6.524750946372503,
|
| 98 |
+
"eval_qa_exact_match": 0.10209318028359217,
|
| 99 |
+
"eval_recite_bleu": 18.618913529891735,
|
| 100 |
+
"eval_recite_exact_match": 0.0,
|
| 101 |
+
"step": 4358
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.0,
|
| 105 |
+
"learning_rate": 3e-05,
|
| 106 |
+
"loss": 1.4425,
|
| 107 |
+
"step": 4360
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.2,
|
| 111 |
+
"learning_rate": 3e-05,
|
| 112 |
+
"loss": 1.0827,
|
| 113 |
+
"step": 4796
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.4,
|
| 117 |
+
"learning_rate": 3e-05,
|
| 118 |
+
"loss": 1.0876,
|
| 119 |
+
"step": 5232
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.6,
|
| 123 |
+
"learning_rate": 3e-05,
|
| 124 |
+
"loss": 1.0887,
|
| 125 |
+
"step": 5668
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.8,
|
| 129 |
+
"learning_rate": 3e-05,
|
| 130 |
+
"loss": 1.0665,
|
| 131 |
+
"step": 6104
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.0,
|
| 135 |
+
"eval_accuracy": 0.7147465927772421,
|
| 136 |
+
"eval_loss": 0.9177509546279907,
|
| 137 |
+
"eval_runtime": 226.6441,
|
| 138 |
+
"eval_samples_per_second": 32.672,
|
| 139 |
+
"eval_steps_per_second": 2.043,
|
| 140 |
+
"step": 6537
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 3.0,
|
| 144 |
+
"eval_exact_match": 12.397029034436192,
|
| 145 |
+
"eval_f1": 17.721572805797255,
|
| 146 |
+
"eval_qa_bleu": 8.139751517788653,
|
| 147 |
+
"eval_qa_exact_match": 0.11978392977717758,
|
| 148 |
+
"eval_recite_bleu": 21.769859861022653,
|
| 149 |
+
"eval_recite_exact_match": 0.0,
|
| 150 |
+
"step": 6537
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 3.0,
|
| 154 |
+
"learning_rate": 3e-05,
|
| 155 |
+
"loss": 0.7772,
|
| 156 |
+
"step": 6540
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 3.2,
|
| 160 |
+
"learning_rate": 3e-05,
|
| 161 |
+
"loss": 0.7688,
|
| 162 |
+
"step": 6976
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 3.4,
|
| 166 |
+
"learning_rate": 3e-05,
|
| 167 |
+
"loss": 0.7598,
|
| 168 |
+
"step": 7412
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 3.6,
|
| 172 |
+
"learning_rate": 3e-05,
|
| 173 |
+
"loss": 0.77,
|
| 174 |
+
"step": 7848
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 3.8,
|
| 178 |
+
"learning_rate": 3e-05,
|
| 179 |
+
"loss": 0.7684,
|
| 180 |
+
"step": 8284
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 4.0,
|
| 184 |
+
"eval_accuracy": 0.7331474586575765,
|
| 185 |
+
"eval_loss": 0.6987847089767456,
|
| 186 |
+
"eval_runtime": 221.7249,
|
| 187 |
+
"eval_samples_per_second": 33.397,
|
| 188 |
+
"eval_steps_per_second": 2.088,
|
| 189 |
+
"step": 8716
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 4.0,
|
| 193 |
+
"eval_exact_match": 13.409858203916272,
|
| 194 |
+
"eval_f1": 19.326947132558665,
|
| 195 |
+
"eval_qa_bleu": 8.79837393539171,
|
| 196 |
+
"eval_qa_exact_match": 0.12869682646860228,
|
| 197 |
+
"eval_recite_bleu": 23.848732661358156,
|
| 198 |
+
"eval_recite_exact_match": 0.0,
|
| 199 |
+
"step": 8716
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 4.0,
|
| 203 |
+
"learning_rate": 3e-05,
|
| 204 |
+
"loss": 0.7559,
|
| 205 |
+
"step": 8720
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 4.2,
|
| 209 |
+
"learning_rate": 3e-05,
|
| 210 |
+
"loss": 0.5354,
|
| 211 |
+
"step": 9156
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 4.4,
|
| 215 |
+
"learning_rate": 3e-05,
|
| 216 |
+
"loss": 0.5511,
|
| 217 |
+
"step": 9592
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 4.6,
|
| 221 |
+
"learning_rate": 3e-05,
|
| 222 |
+
"loss": 0.5496,
|
| 223 |
+
"step": 10028
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 4.8,
|
| 227 |
+
"learning_rate": 3e-05,
|
| 228 |
+
"loss": 0.548,
|
| 229 |
+
"step": 10464
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 5.0,
|
| 233 |
+
"eval_accuracy": 0.7465929589970037,
|
| 234 |
+
"eval_loss": 0.5567022562026978,
|
| 235 |
+
"eval_runtime": 226.4909,
|
| 236 |
+
"eval_samples_per_second": 32.694,
|
| 237 |
+
"eval_steps_per_second": 2.044,
|
| 238 |
+
"step": 10895
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 5.0,
|
| 242 |
+
"eval_exact_match": 14.220121539500338,
|
| 243 |
+
"eval_f1": 20.736856637999697,
|
| 244 |
+
"eval_qa_bleu": 9.542509177303282,
|
| 245 |
+
"eval_qa_exact_match": 0.13652937204591492,
|
| 246 |
+
"eval_recite_bleu": 27.315073242003514,
|
| 247 |
+
"eval_recite_exact_match": 0.0009453072248480756,
|
| 248 |
+
"step": 10895
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 5.0,
|
| 252 |
+
"learning_rate": 3e-05,
|
| 253 |
+
"loss": 0.5518,
|
| 254 |
+
"step": 10900
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 5.2,
|
| 258 |
+
"learning_rate": 3e-05,
|
| 259 |
+
"loss": 0.3837,
|
| 260 |
+
"step": 11336
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 5.4,
|
| 264 |
+
"learning_rate": 3e-05,
|
| 265 |
+
"loss": 0.3963,
|
| 266 |
+
"step": 11772
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 5.6,
|
| 270 |
+
"learning_rate": 3e-05,
|
| 271 |
+
"loss": 0.404,
|
| 272 |
+
"step": 12208
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 5.8,
|
| 276 |
+
"learning_rate": 3e-05,
|
| 277 |
+
"loss": 0.4039,
|
| 278 |
+
"step": 12644
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 6.0,
|
| 282 |
+
"eval_accuracy": 0.7551407100982113,
|
| 283 |
+
"eval_loss": 0.4728190004825592,
|
| 284 |
+
"eval_runtime": 223.8973,
|
| 285 |
+
"eval_samples_per_second": 33.073,
|
| 286 |
+
"eval_steps_per_second": 2.068,
|
| 287 |
+
"step": 13074
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 6.0,
|
| 291 |
+
"eval_exact_match": 16.12424037812289,
|
| 292 |
+
"eval_f1": 23.0965386152669,
|
| 293 |
+
"eval_qa_bleu": 10.706681466541065,
|
| 294 |
+
"eval_qa_exact_match": 0.1524645509790682,
|
| 295 |
+
"eval_recite_bleu": 31.825156559426375,
|
| 296 |
+
"eval_recite_exact_match": 0.0036461850101282916,
|
| 297 |
+
"step": 13074
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 6.0,
|
| 301 |
+
"learning_rate": 3e-05,
|
| 302 |
+
"loss": 0.405,
|
| 303 |
+
"step": 13080
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 6.2,
|
| 307 |
+
"learning_rate": 3e-05,
|
| 308 |
+
"loss": 0.285,
|
| 309 |
+
"step": 13516
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 6.4,
|
| 313 |
+
"learning_rate": 3e-05,
|
| 314 |
+
"loss": 0.2984,
|
| 315 |
+
"step": 13952
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 6.6,
|
| 319 |
+
"learning_rate": 3e-05,
|
| 320 |
+
"loss": 0.3051,
|
| 321 |
+
"step": 14388
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 6.8,
|
| 325 |
+
"learning_rate": 3e-05,
|
| 326 |
+
"loss": 0.3044,
|
| 327 |
+
"step": 14824
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 7.0,
|
| 331 |
+
"eval_accuracy": 0.7600033806411041,
|
| 332 |
+
"eval_loss": 0.4375591278076172,
|
| 333 |
+
"eval_runtime": 225.3993,
|
| 334 |
+
"eval_samples_per_second": 32.853,
|
| 335 |
+
"eval_steps_per_second": 2.054,
|
| 336 |
+
"step": 15253
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 7.0,
|
| 340 |
+
"eval_exact_match": 16.948008102633356,
|
| 341 |
+
"eval_f1": 23.931422337180113,
|
| 342 |
+
"eval_qa_bleu": 11.914541362683222,
|
| 343 |
+
"eval_qa_exact_match": 0.16029709655638083,
|
| 344 |
+
"eval_recite_bleu": 35.91072342079755,
|
| 345 |
+
"eval_recite_exact_match": 0.010938555030384874,
|
| 346 |
+
"step": 15253
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 7.0,
|
| 350 |
+
"learning_rate": 3e-05,
|
| 351 |
+
"loss": 0.2046,
|
| 352 |
+
"step": 15260
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 7.2,
|
| 356 |
+
"learning_rate": 3e-05,
|
| 357 |
+
"loss": 0.2265,
|
| 358 |
+
"step": 15696
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 7.4,
|
| 362 |
+
"learning_rate": 3e-05,
|
| 363 |
+
"loss": 0.236,
|
| 364 |
+
"step": 16132
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 7.6,
|
| 368 |
+
"learning_rate": 3e-05,
|
| 369 |
+
"loss": 0.2399,
|
| 370 |
+
"step": 16568
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 7.8,
|
| 374 |
+
"learning_rate": 3e-05,
|
| 375 |
+
"loss": 0.2446,
|
| 376 |
+
"step": 17004
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 8.0,
|
| 380 |
+
"eval_accuracy": 0.7628389388058522,
|
| 381 |
+
"eval_loss": 0.42204150557518005,
|
| 382 |
+
"eval_runtime": 223.5169,
|
| 383 |
+
"eval_samples_per_second": 33.129,
|
| 384 |
+
"eval_steps_per_second": 2.071,
|
| 385 |
+
"step": 17432
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 8.0,
|
| 389 |
+
"eval_exact_match": 18.433490884537473,
|
| 390 |
+
"eval_f1": 25.859990277646197,
|
| 391 |
+
"eval_qa_bleu": 13.293068057422849,
|
| 392 |
+
"eval_qa_exact_match": 0.17609723160027008,
|
| 393 |
+
"eval_recite_bleu": 41.25950959223269,
|
| 394 |
+
"eval_recite_exact_match": 0.015530047265361242,
|
| 395 |
+
"step": 17432
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 8.0,
|
| 399 |
+
"learning_rate": 3e-05,
|
| 400 |
+
"loss": 0.2466,
|
| 401 |
+
"step": 17440
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 8.2,
|
| 405 |
+
"learning_rate": 3e-05,
|
| 406 |
+
"loss": 0.1887,
|
| 407 |
+
"step": 17876
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 8.4,
|
| 411 |
+
"learning_rate": 3e-05,
|
| 412 |
+
"loss": 0.1967,
|
| 413 |
+
"step": 18312
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 8.6,
|
| 417 |
+
"learning_rate": 3e-05,
|
| 418 |
+
"loss": 0.2026,
|
| 419 |
+
"step": 18748
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 8.8,
|
| 423 |
+
"learning_rate": 3e-05,
|
| 424 |
+
"loss": 0.2039,
|
| 425 |
+
"step": 19184
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 9.0,
|
| 429 |
+
"eval_accuracy": 0.7641864696201232,
|
| 430 |
+
"eval_loss": 0.41897761821746826,
|
| 431 |
+
"eval_runtime": 222.9387,
|
| 432 |
+
"eval_samples_per_second": 33.215,
|
| 433 |
+
"eval_steps_per_second": 2.077,
|
| 434 |
+
"step": 19611
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 9.0,
|
| 438 |
+
"eval_exact_match": 18.933153274814316,
|
| 439 |
+
"eval_f1": 26.522905186169627,
|
| 440 |
+
"eval_qa_bleu": 12.546175820192783,
|
| 441 |
+
"eval_qa_exact_match": 0.1797434166103984,
|
| 442 |
+
"eval_recite_bleu": 44.27793915167843,
|
| 443 |
+
"eval_recite_exact_match": 0.02228224172856178,
|
| 444 |
+
"step": 19611
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 9.0,
|
| 448 |
+
"learning_rate": 3e-05,
|
| 449 |
+
"loss": 0.2067,
|
| 450 |
+
"step": 19620
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 9.2,
|
| 454 |
+
"learning_rate": 3e-05,
|
| 455 |
+
"loss": 0.1667,
|
| 456 |
+
"step": 20056
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 9.4,
|
| 460 |
+
"learning_rate": 3e-05,
|
| 461 |
+
"loss": 0.1709,
|
| 462 |
+
"step": 20492
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 9.6,
|
| 466 |
+
"learning_rate": 3e-05,
|
| 467 |
+
"loss": 0.1774,
|
| 468 |
+
"step": 20928
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 9.8,
|
| 472 |
+
"learning_rate": 3e-05,
|
| 473 |
+
"loss": 0.1787,
|
| 474 |
+
"step": 21364
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 10.0,
|
| 478 |
+
"eval_accuracy": 0.7648945867029934,
|
| 479 |
+
"eval_loss": 0.4250437319278717,
|
| 480 |
+
"eval_runtime": 225.6304,
|
| 481 |
+
"eval_samples_per_second": 32.819,
|
| 482 |
+
"eval_steps_per_second": 2.052,
|
| 483 |
+
"step": 21790
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 10.0,
|
| 487 |
+
"eval_exact_match": 20.054017555705606,
|
| 488 |
+
"eval_f1": 27.607870062960497,
|
| 489 |
+
"eval_qa_bleu": 14.597537360428676,
|
| 490 |
+
"eval_qa_exact_match": 0.1913571910871033,
|
| 491 |
+
"eval_recite_bleu": 46.3428836838782,
|
| 492 |
+
"eval_recite_exact_match": 0.029169480081026333,
|
| 493 |
+
"step": 21790
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 10.0,
|
| 497 |
+
"learning_rate": 3e-05,
|
| 498 |
+
"loss": 0.1829,
|
| 499 |
+
"step": 21800
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 10.2,
|
| 503 |
+
"learning_rate": 3e-05,
|
| 504 |
+
"loss": 0.1498,
|
| 505 |
+
"step": 22236
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 10.4,
|
| 509 |
+
"learning_rate": 3e-05,
|
| 510 |
+
"loss": 0.1553,
|
| 511 |
+
"step": 22672
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 10.6,
|
| 515 |
+
"learning_rate": 3e-05,
|
| 516 |
+
"loss": 0.1612,
|
| 517 |
+
"step": 23108
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 10.8,
|
| 521 |
+
"learning_rate": 3e-05,
|
| 522 |
+
"loss": 0.1652,
|
| 523 |
+
"step": 23544
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 11.0,
|
| 527 |
+
"eval_accuracy": 0.7653962302216591,
|
| 528 |
+
"eval_loss": 0.42946067452430725,
|
| 529 |
+
"eval_runtime": 227.2445,
|
| 530 |
+
"eval_samples_per_second": 32.586,
|
| 531 |
+
"eval_steps_per_second": 2.037,
|
| 532 |
+
"step": 23969
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 11.0,
|
| 536 |
+
"eval_exact_match": 19.581363943281566,
|
| 537 |
+
"eval_f1": 27.25909178622117,
|
| 538 |
+
"eval_qa_bleu": 14.227446461856255,
|
| 539 |
+
"eval_qa_exact_match": 0.18595543551654287,
|
| 540 |
+
"eval_recite_bleu": 46.797968452871594,
|
| 541 |
+
"eval_recite_exact_match": 0.03430114787305875,
|
| 542 |
+
"step": 23969
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 11.01,
|
| 546 |
+
"learning_rate": 3e-05,
|
| 547 |
+
"loss": 0.1439,
|
| 548 |
+
"step": 23980
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 11.21,
|
| 552 |
+
"learning_rate": 3e-05,
|
| 553 |
+
"loss": 0.1402,
|
| 554 |
+
"step": 24416
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 11.41,
|
| 558 |
+
"learning_rate": 3e-05,
|
| 559 |
+
"loss": 0.1462,
|
| 560 |
+
"step": 24852
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 11.61,
|
| 564 |
+
"learning_rate": 3e-05,
|
| 565 |
+
"loss": 0.1492,
|
| 566 |
+
"step": 25288
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 11.81,
|
| 570 |
+
"learning_rate": 3e-05,
|
| 571 |
+
"loss": 0.154,
|
| 572 |
+
"step": 25724
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 12.0,
|
| 576 |
+
"eval_accuracy": 0.7654645700633325,
|
| 577 |
+
"eval_loss": 0.43657177686691284,
|
| 578 |
+
"eval_runtime": 220.7713,
|
| 579 |
+
"eval_samples_per_second": 33.541,
|
| 580 |
+
"eval_steps_per_second": 2.097,
|
| 581 |
+
"step": 26148
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 12.0,
|
| 585 |
+
"eval_exact_match": 19.945982444294394,
|
| 586 |
+
"eval_f1": 27.398959656999555,
|
| 587 |
+
"eval_qa_bleu": 13.50737485018698,
|
| 588 |
+
"eval_qa_exact_match": 0.1886563133018231,
|
| 589 |
+
"eval_recite_bleu": 47.72099251148126,
|
| 590 |
+
"eval_recite_exact_match": 0.036326806212018906,
|
| 591 |
+
"step": 26148
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 12.01,
|
| 595 |
+
"learning_rate": 3e-05,
|
| 596 |
+
"loss": 0.1569,
|
| 597 |
+
"step": 26160
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 12.21,
|
| 601 |
+
"learning_rate": 3e-05,
|
| 602 |
+
"loss": 0.1344,
|
| 603 |
+
"step": 26596
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 12.41,
|
| 607 |
+
"learning_rate": 3e-05,
|
| 608 |
+
"loss": 0.14,
|
| 609 |
+
"step": 27032
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 12.61,
|
| 613 |
+
"learning_rate": 3e-05,
|
| 614 |
+
"loss": 0.1427,
|
| 615 |
+
"step": 27468
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 12.81,
|
| 619 |
+
"learning_rate": 3e-05,
|
| 620 |
+
"loss": 0.1441,
|
| 621 |
+
"step": 27904
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 13.0,
|
| 625 |
+
"eval_accuracy": 0.7656634099218181,
|
| 626 |
+
"eval_loss": 0.44285184144973755,
|
| 627 |
+
"eval_runtime": 222.075,
|
| 628 |
+
"eval_samples_per_second": 33.345,
|
| 629 |
+
"eval_steps_per_second": 2.085,
|
| 630 |
+
"step": 28327
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 13.0,
|
| 634 |
+
"eval_exact_match": 20.405131667792034,
|
| 635 |
+
"eval_f1": 28.457441663021587,
|
| 636 |
+
"eval_qa_bleu": 14.596564409389117,
|
| 637 |
+
"eval_qa_exact_match": 0.19446320054017555,
|
| 638 |
+
"eval_recite_bleu": 48.14075990339527,
|
| 639 |
+
"eval_recite_exact_match": 0.03902768399729912,
|
| 640 |
+
"step": 28327
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 13.01,
|
| 644 |
+
"learning_rate": 3e-05,
|
| 645 |
+
"loss": 0.1502,
|
| 646 |
+
"step": 28340
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 13.21,
|
| 650 |
+
"learning_rate": 3e-05,
|
| 651 |
+
"loss": 0.1288,
|
| 652 |
+
"step": 28776
|
| 653 |
+
},
|
| 654 |
+
{
|
| 655 |
+
"epoch": 13.41,
|
| 656 |
+
"learning_rate": 3e-05,
|
| 657 |
+
"loss": 0.1334,
|
| 658 |
+
"step": 29212
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 13.61,
|
| 662 |
+
"learning_rate": 3e-05,
|
| 663 |
+
"loss": 0.1389,
|
| 664 |
+
"step": 29648
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 13.81,
|
| 668 |
+
"learning_rate": 3e-05,
|
| 669 |
+
"loss": 0.143,
|
| 670 |
+
"step": 30084
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 14.0,
|
| 674 |
+
"eval_accuracy": 0.7656877650781592,
|
| 675 |
+
"eval_loss": 0.44177931547164917,
|
| 676 |
+
"eval_runtime": 226.2366,
|
| 677 |
+
"eval_samples_per_second": 32.731,
|
| 678 |
+
"eval_steps_per_second": 2.047,
|
| 679 |
+
"step": 30506
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 14.0,
|
| 683 |
+
"eval_exact_match": 20.324105334233625,
|
| 684 |
+
"eval_f1": 28.258140680396128,
|
| 685 |
+
"eval_qa_bleu": 14.043080335083465,
|
| 686 |
+
"eval_qa_exact_match": 0.19324780553679946,
|
| 687 |
+
"eval_recite_bleu": 49.992383315936785,
|
| 688 |
+
"eval_recite_exact_match": 0.042403781228899394,
|
| 689 |
+
"step": 30506
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 14.01,
|
| 693 |
+
"learning_rate": 3e-05,
|
| 694 |
+
"loss": 0.1434,
|
| 695 |
+
"step": 30520
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 14.21,
|
| 699 |
+
"learning_rate": 3e-05,
|
| 700 |
+
"loss": 0.1265,
|
| 701 |
+
"step": 30956
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 14.41,
|
| 705 |
+
"learning_rate": 3e-05,
|
| 706 |
+
"loss": 0.1291,
|
| 707 |
+
"step": 31392
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 14.61,
|
| 711 |
+
"learning_rate": 3e-05,
|
| 712 |
+
"loss": 0.1346,
|
| 713 |
+
"step": 31828
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 14.81,
|
| 717 |
+
"learning_rate": 3e-05,
|
| 718 |
+
"loss": 0.1366,
|
| 719 |
+
"step": 32264
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 15.0,
|
| 723 |
+
"eval_accuracy": 0.7660609078838914,
|
| 724 |
+
"eval_loss": 0.44691887497901917,
|
| 725 |
+
"eval_runtime": 226.6588,
|
| 726 |
+
"eval_samples_per_second": 32.67,
|
| 727 |
+
"eval_steps_per_second": 2.043,
|
| 728 |
+
"step": 32685
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 15.0,
|
| 732 |
+
"eval_exact_match": 19.878460499662392,
|
| 733 |
+
"eval_f1": 27.391401943316232,
|
| 734 |
+
"eval_qa_bleu": 14.761384752422014,
|
| 735 |
+
"eval_qa_exact_match": 0.18717083051991898,
|
| 736 |
+
"eval_recite_bleu": 49.40428435717536,
|
| 737 |
+
"eval_recite_exact_match": 0.04659014179608373,
|
| 738 |
+
"step": 32685
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 15.01,
|
| 742 |
+
"learning_rate": 3e-05,
|
| 743 |
+
"loss": 0.1234,
|
| 744 |
+
"step": 32700
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 15.21,
|
| 748 |
+
"learning_rate": 3e-05,
|
| 749 |
+
"loss": 0.1235,
|
| 750 |
+
"step": 33136
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 15.41,
|
| 754 |
+
"learning_rate": 3e-05,
|
| 755 |
+
"loss": 0.1285,
|
| 756 |
+
"step": 33572
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 15.61,
|
| 760 |
+
"learning_rate": 3e-05,
|
| 761 |
+
"loss": 0.1296,
|
| 762 |
+
"step": 34008
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 15.81,
|
| 766 |
+
"learning_rate": 3e-05,
|
| 767 |
+
"loss": 0.1335,
|
| 768 |
+
"step": 34444
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 16.0,
|
| 772 |
+
"eval_accuracy": 0.7660959865792183,
|
| 773 |
+
"eval_loss": 0.45172181725502014,
|
| 774 |
+
"eval_runtime": 220.6741,
|
| 775 |
+
"eval_samples_per_second": 33.556,
|
| 776 |
+
"eval_steps_per_second": 2.098,
|
| 777 |
+
"step": 34864
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 16.0,
|
| 781 |
+
"eval_exact_match": 19.648885887913572,
|
| 782 |
+
"eval_f1": 27.41473279396055,
|
| 783 |
+
"eval_qa_bleu": 14.081874080604702,
|
| 784 |
+
"eval_qa_exact_match": 0.1863605671843349,
|
| 785 |
+
"eval_recite_bleu": 49.64601003868678,
|
| 786 |
+
"eval_recite_exact_match": 0.044699527346387574,
|
| 787 |
+
"step": 34864
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 16.01,
|
| 791 |
+
"learning_rate": 3e-05,
|
| 792 |
+
"loss": 0.1335,
|
| 793 |
+
"step": 34880
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 16.21,
|
| 797 |
+
"learning_rate": 3e-05,
|
| 798 |
+
"loss": 0.1195,
|
| 799 |
+
"step": 35316
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 16.41,
|
| 803 |
+
"learning_rate": 3e-05,
|
| 804 |
+
"loss": 0.1245,
|
| 805 |
+
"step": 35752
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 16.61,
|
| 809 |
+
"learning_rate": 3e-05,
|
| 810 |
+
"loss": 0.1278,
|
| 811 |
+
"step": 36188
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 16.81,
|
| 815 |
+
"learning_rate": 3e-05,
|
| 816 |
+
"loss": 0.1299,
|
| 817 |
+
"step": 36624
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 17.0,
|
| 821 |
+
"eval_accuracy": 0.7661592372837458,
|
| 822 |
+
"eval_loss": 0.45338499546051025,
|
| 823 |
+
"eval_runtime": 221.2554,
|
| 824 |
+
"eval_samples_per_second": 33.468,
|
| 825 |
+
"eval_steps_per_second": 2.093,
|
| 826 |
+
"step": 37043
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 17.0,
|
| 830 |
+
"eval_exact_match": 19.60837272113437,
|
| 831 |
+
"eval_f1": 27.45798777202767,
|
| 832 |
+
"eval_qa_bleu": 14.536144503898182,
|
| 833 |
+
"eval_qa_exact_match": 0.1850101282916948,
|
| 834 |
+
"eval_recite_bleu": 49.273704799153755,
|
| 835 |
+
"eval_recite_exact_match": 0.048345712356515864,
|
| 836 |
+
"step": 37043
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 17.01,
|
| 840 |
+
"learning_rate": 3e-05,
|
| 841 |
+
"loss": 0.1325,
|
| 842 |
+
"step": 37060
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 17.21,
|
| 846 |
+
"learning_rate": 3e-05,
|
| 847 |
+
"loss": 0.1188,
|
| 848 |
+
"step": 37496
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 17.41,
|
| 852 |
+
"learning_rate": 3e-05,
|
| 853 |
+
"loss": 0.1226,
|
| 854 |
+
"step": 37932
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 17.61,
|
| 858 |
+
"learning_rate": 3e-05,
|
| 859 |
+
"loss": 0.1263,
|
| 860 |
+
"step": 38368
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 17.81,
|
| 864 |
+
"learning_rate": 3e-05,
|
| 865 |
+
"loss": 0.1271,
|
| 866 |
+
"step": 38804
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 18.0,
|
| 870 |
+
"eval_accuracy": 0.7663929740826603,
|
| 871 |
+
"eval_loss": 0.45788267254829407,
|
| 872 |
+
"eval_runtime": 224.1369,
|
| 873 |
+
"eval_samples_per_second": 33.038,
|
| 874 |
+
"eval_steps_per_second": 2.066,
|
| 875 |
+
"step": 39222
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 18.0,
|
| 879 |
+
"eval_exact_match": 19.959486833220797,
|
| 880 |
+
"eval_f1": 27.5331537130017,
|
| 881 |
+
"eval_qa_bleu": 14.102203160745331,
|
| 882 |
+
"eval_qa_exact_match": 0.1900067521944632,
|
| 883 |
+
"eval_recite_bleu": 50.42598107564746,
|
| 884 |
+
"eval_recite_exact_match": 0.05172180958811614,
|
| 885 |
+
"step": 39222
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 18.01,
|
| 889 |
+
"learning_rate": 3e-05,
|
| 890 |
+
"loss": 0.1298,
|
| 891 |
+
"step": 39240
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 18.21,
|
| 895 |
+
"learning_rate": 3e-05,
|
| 896 |
+
"loss": 0.1152,
|
| 897 |
+
"step": 39676
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 18.41,
|
| 901 |
+
"learning_rate": 3e-05,
|
| 902 |
+
"loss": 0.119,
|
| 903 |
+
"step": 40112
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 18.61,
|
| 907 |
+
"learning_rate": 3e-05,
|
| 908 |
+
"loss": 0.1251,
|
| 909 |
+
"step": 40548
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 18.81,
|
| 913 |
+
"learning_rate": 3e-05,
|
| 914 |
+
"loss": 0.1268,
|
| 915 |
+
"step": 40984
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 19.0,
|
| 919 |
+
"eval_accuracy": 0.7664202373173704,
|
| 920 |
+
"eval_loss": 0.4556055963039398,
|
| 921 |
+
"eval_runtime": 225.0055,
|
| 922 |
+
"eval_samples_per_second": 32.91,
|
| 923 |
+
"eval_steps_per_second": 2.058,
|
| 924 |
+
"step": 41401
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 19.0,
|
| 928 |
+
"eval_exact_match": 21.02633355840648,
|
| 929 |
+
"eval_f1": 28.383806355797034,
|
| 930 |
+
"eval_qa_bleu": 15.303943702882195,
|
| 931 |
+
"eval_qa_exact_match": 0.19905469277515192,
|
| 932 |
+
"eval_recite_bleu": 51.33952244561918,
|
| 933 |
+
"eval_recite_exact_match": 0.05442268737339635,
|
| 934 |
+
"step": 41401
|
| 935 |
+
}
|
| 936 |
+
],
|
| 937 |
+
"logging_steps": 436,
|
| 938 |
+
"max_steps": 43580,
|
| 939 |
+
"num_train_epochs": 20,
|
| 940 |
+
"save_steps": 500,
|
| 941 |
+
"total_flos": 1.9157215780620288e+18,
|
| 942 |
+
"trial_name": null,
|
| 943 |
+
"trial_params": null
|
| 944 |
+
}
|
checkpoint-41401/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cae5a0482131a1253cac2a7ef1083f0d9bb1ac1cf467c769d2a6b8675da83dc9
|
| 3 |
+
size 4728
|
checkpoint-41401/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|