Training in progress, epoch 48, checkpoint
Browse files- checkpoint-16370/README.md +9 -0
- checkpoint-16370/adapter_config.json +25 -0
- checkpoint-16370/adapter_model.bin +3 -0
- checkpoint-16370/added_tokens.json +5 -0
- checkpoint-16370/optimizer.pt +3 -0
- checkpoint-16370/rng_state_0.pth +3 -0
- checkpoint-16370/rng_state_1.pth +3 -0
- checkpoint-16370/rng_state_2.pth +3 -0
- checkpoint-16370/rng_state_3.pth +3 -0
- checkpoint-16370/scheduler.pt +3 -0
- checkpoint-16370/special_tokens_map.json +6 -0
- checkpoint-16370/tokenizer.json +0 -0
- checkpoint-16370/tokenizer.model +3 -0
- checkpoint-16370/tokenizer_config.json +40 -0
- checkpoint-16370/trainer_state.json +1666 -0
- checkpoint-16370/training_args.bin +3 -0
checkpoint-16370/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.5.0
|
checkpoint-16370/adapter_config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 16,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"k_proj",
|
| 19 |
+
"v_proj",
|
| 20 |
+
"down_proj",
|
| 21 |
+
"gate_proj",
|
| 22 |
+
"up_proj"
|
| 23 |
+
],
|
| 24 |
+
"task_type": "CAUSAL_LM"
|
| 25 |
+
}
|
checkpoint-16370/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a2393c47fca5b46fdc77f9521f7b837a2a121115b510c3f36262c70fb86d60bf
|
| 3 |
+
size 143269386
|
checkpoint-16370/added_tokens.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"</s>": 2,
|
| 3 |
+
"<s>": 1,
|
| 4 |
+
"<unk>": 0
|
| 5 |
+
}
|
checkpoint-16370/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:282f88cacd6d5ad2e4a5a85ede224c95cf7566a0c1b7814b72d056edd516c3fc
|
| 3 |
+
size 286590610
|
checkpoint-16370/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:29c4c3956338b8c83e029dabb230004664f028a9633e66a8e8a0a5f853229097
|
| 3 |
+
size 15024
|
checkpoint-16370/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5e13f6c4e2ae46614d96625328801c27ef56d59862a8e18a85109da179fa2af1
|
| 3 |
+
size 15024
|
checkpoint-16370/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:51c4ebe9b125285782773ae81c787b2893e90e219f03f4393e7b14f0f6aa4978
|
| 3 |
+
size 15024
|
checkpoint-16370/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:07193be2514596701b46f43273b26650d1384c5712b0e81720f7235b831cf3c6
|
| 3 |
+
size 15024
|
checkpoint-16370/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:536ff24cacf4bb47ee54132c6d5e87cc09e3bc29ca28e8cffb4f6eb86bc74486
|
| 3 |
+
size 1064
|
checkpoint-16370/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"pad_token": "</s>",
|
| 5 |
+
"unk_token": "<unk>"
|
| 6 |
+
}
|
checkpoint-16370/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-16370/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
checkpoint-16370/tokenizer_config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<unk>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "<s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "</s>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
}
|
| 27 |
+
},
|
| 28 |
+
"additional_special_tokens": [],
|
| 29 |
+
"bos_token": "<s>",
|
| 30 |
+
"clean_up_tokenization_spaces": false,
|
| 31 |
+
"eos_token": "</s>",
|
| 32 |
+
"legacy": false,
|
| 33 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 34 |
+
"pad_token": "</s>",
|
| 35 |
+
"padding_side": "left",
|
| 36 |
+
"sp_model_kwargs": {},
|
| 37 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 38 |
+
"unk_token": "<unk>",
|
| 39 |
+
"use_default_system_prompt": true
|
| 40 |
+
}
|
checkpoint-16370/trainer_state.json
ADDED
|
@@ -0,0 +1,1666 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 48.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 16370,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.29,
|
| 13 |
+
"learning_rate": 0.0001,
|
| 14 |
+
"loss": 1.4842,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.59,
|
| 19 |
+
"learning_rate": 0.0001,
|
| 20 |
+
"loss": 1.4016,
|
| 21 |
+
"step": 200
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.88,
|
| 25 |
+
"learning_rate": 0.0001,
|
| 26 |
+
"loss": 1.3892,
|
| 27 |
+
"step": 300
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 1.0,
|
| 31 |
+
"eval_accuracy": 0.4543589743589744,
|
| 32 |
+
"eval_loss": 3.4055869579315186,
|
| 33 |
+
"eval_runtime": 4.1688,
|
| 34 |
+
"eval_samples_per_second": 119.938,
|
| 35 |
+
"eval_steps_per_second": 15.112,
|
| 36 |
+
"step": 341
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 1.0,
|
| 40 |
+
"eval_bleu": 0.09711501224106496,
|
| 41 |
+
"eval_exact_match": 0.0,
|
| 42 |
+
"eval_prefix_exact_match": 0.054,
|
| 43 |
+
"step": 341
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 1.17,
|
| 47 |
+
"learning_rate": 0.0001,
|
| 48 |
+
"loss": 1.3608,
|
| 49 |
+
"step": 400
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.46,
|
| 53 |
+
"learning_rate": 0.0001,
|
| 54 |
+
"loss": 1.3456,
|
| 55 |
+
"step": 500
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 1.76,
|
| 59 |
+
"learning_rate": 0.0001,
|
| 60 |
+
"loss": 1.3499,
|
| 61 |
+
"step": 600
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 2.0,
|
| 65 |
+
"eval_accuracy": 0.4576923076923077,
|
| 66 |
+
"eval_loss": 3.453113317489624,
|
| 67 |
+
"eval_runtime": 4.4031,
|
| 68 |
+
"eval_samples_per_second": 113.556,
|
| 69 |
+
"eval_steps_per_second": 14.308,
|
| 70 |
+
"step": 683
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 2.0,
|
| 74 |
+
"eval_bleu": 0.1225593574040313,
|
| 75 |
+
"eval_exact_match": 0.0,
|
| 76 |
+
"eval_prefix_exact_match": 0.056,
|
| 77 |
+
"step": 683
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 2.05,
|
| 81 |
+
"learning_rate": 0.0001,
|
| 82 |
+
"loss": 1.3238,
|
| 83 |
+
"step": 700
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.34,
|
| 87 |
+
"learning_rate": 0.0001,
|
| 88 |
+
"loss": 1.2486,
|
| 89 |
+
"step": 800
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.64,
|
| 93 |
+
"learning_rate": 0.0001,
|
| 94 |
+
"loss": 1.2452,
|
| 95 |
+
"step": 900
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.93,
|
| 99 |
+
"learning_rate": 0.0001,
|
| 100 |
+
"loss": 1.2427,
|
| 101 |
+
"step": 1000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.0,
|
| 105 |
+
"eval_accuracy": 0.4584102564102564,
|
| 106 |
+
"eval_loss": 3.671137809753418,
|
| 107 |
+
"eval_runtime": 4.4094,
|
| 108 |
+
"eval_samples_per_second": 113.393,
|
| 109 |
+
"eval_steps_per_second": 14.288,
|
| 110 |
+
"step": 1024
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 3.0,
|
| 114 |
+
"eval_bleu": 0.13165889774889974,
|
| 115 |
+
"eval_exact_match": 0.0,
|
| 116 |
+
"eval_prefix_exact_match": 0.056,
|
| 117 |
+
"step": 1024
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 3.22,
|
| 121 |
+
"learning_rate": 0.0001,
|
| 122 |
+
"loss": 1.1507,
|
| 123 |
+
"step": 1100
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 3.51,
|
| 127 |
+
"learning_rate": 0.0001,
|
| 128 |
+
"loss": 1.1218,
|
| 129 |
+
"step": 1200
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 3.81,
|
| 133 |
+
"learning_rate": 0.0001,
|
| 134 |
+
"loss": 1.1231,
|
| 135 |
+
"step": 1300
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 4.0,
|
| 139 |
+
"eval_accuracy": 0.45697435897435895,
|
| 140 |
+
"eval_loss": 3.7999510765075684,
|
| 141 |
+
"eval_runtime": 4.0886,
|
| 142 |
+
"eval_samples_per_second": 122.29,
|
| 143 |
+
"eval_steps_per_second": 15.409,
|
| 144 |
+
"step": 1366
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 4.0,
|
| 148 |
+
"eval_bleu": 0.15413978034957934,
|
| 149 |
+
"eval_exact_match": 0.0,
|
| 150 |
+
"eval_prefix_exact_match": 0.054,
|
| 151 |
+
"step": 1366
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 4.1,
|
| 155 |
+
"learning_rate": 0.0001,
|
| 156 |
+
"loss": 1.0823,
|
| 157 |
+
"step": 1400
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.39,
|
| 161 |
+
"learning_rate": 0.0001,
|
| 162 |
+
"loss": 0.997,
|
| 163 |
+
"step": 1500
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 4.69,
|
| 167 |
+
"learning_rate": 0.0001,
|
| 168 |
+
"loss": 1.0024,
|
| 169 |
+
"step": 1600
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.98,
|
| 173 |
+
"learning_rate": 0.0001,
|
| 174 |
+
"loss": 0.995,
|
| 175 |
+
"step": 1700
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 5.0,
|
| 179 |
+
"eval_accuracy": 0.4552307692307692,
|
| 180 |
+
"eval_loss": 3.953216314315796,
|
| 181 |
+
"eval_runtime": 4.9824,
|
| 182 |
+
"eval_samples_per_second": 100.354,
|
| 183 |
+
"eval_steps_per_second": 12.645,
|
| 184 |
+
"step": 1707
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 5.0,
|
| 188 |
+
"eval_bleu": 0.1611344512465791,
|
| 189 |
+
"eval_exact_match": 0.0,
|
| 190 |
+
"eval_prefix_exact_match": 0.046,
|
| 191 |
+
"step": 1707
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 5.27,
|
| 195 |
+
"learning_rate": 0.0001,
|
| 196 |
+
"loss": 0.8495,
|
| 197 |
+
"step": 1800
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 5.56,
|
| 201 |
+
"learning_rate": 0.0001,
|
| 202 |
+
"loss": 0.854,
|
| 203 |
+
"step": 1900
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 5.86,
|
| 207 |
+
"learning_rate": 0.0001,
|
| 208 |
+
"loss": 0.8693,
|
| 209 |
+
"step": 2000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 6.0,
|
| 213 |
+
"eval_accuracy": 0.45261538461538464,
|
| 214 |
+
"eval_loss": 4.0766072273254395,
|
| 215 |
+
"eval_runtime": 4.8095,
|
| 216 |
+
"eval_samples_per_second": 103.961,
|
| 217 |
+
"eval_steps_per_second": 13.099,
|
| 218 |
+
"step": 2049
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.0,
|
| 222 |
+
"eval_bleu": 0.1489904300924264,
|
| 223 |
+
"eval_exact_match": 0.0,
|
| 224 |
+
"eval_prefix_exact_match": 0.044,
|
| 225 |
+
"step": 2049
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.15,
|
| 229 |
+
"learning_rate": 0.0001,
|
| 230 |
+
"loss": 0.7917,
|
| 231 |
+
"step": 2100
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.44,
|
| 235 |
+
"learning_rate": 0.0001,
|
| 236 |
+
"loss": 0.7061,
|
| 237 |
+
"step": 2200
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 6.73,
|
| 241 |
+
"learning_rate": 0.0001,
|
| 242 |
+
"loss": 0.7302,
|
| 243 |
+
"step": 2300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 7.0,
|
| 247 |
+
"eval_accuracy": 0.4501025641025641,
|
| 248 |
+
"eval_loss": 4.171727180480957,
|
| 249 |
+
"eval_runtime": 4.2169,
|
| 250 |
+
"eval_samples_per_second": 118.57,
|
| 251 |
+
"eval_steps_per_second": 14.94,
|
| 252 |
+
"step": 2390
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 7.0,
|
| 256 |
+
"step": 2390
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 7.03,
|
| 260 |
+
"learning_rate": 0.0001,
|
| 261 |
+
"loss": 0.7227,
|
| 262 |
+
"step": 2400
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 7.32,
|
| 266 |
+
"learning_rate": 0.0001,
|
| 267 |
+
"loss": 0.573,
|
| 268 |
+
"step": 2500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.61,
|
| 272 |
+
"learning_rate": 0.0001,
|
| 273 |
+
"loss": 0.6036,
|
| 274 |
+
"step": 2600
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 7.91,
|
| 278 |
+
"learning_rate": 0.0001,
|
| 279 |
+
"loss": 0.6033,
|
| 280 |
+
"step": 2700
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 8.0,
|
| 284 |
+
"eval_accuracy": 0.448,
|
| 285 |
+
"eval_loss": 4.277780055999756,
|
| 286 |
+
"eval_runtime": 4.0947,
|
| 287 |
+
"eval_samples_per_second": 122.11,
|
| 288 |
+
"eval_steps_per_second": 15.386,
|
| 289 |
+
"step": 2732
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.0,
|
| 293 |
+
"step": 2732
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 8.2,
|
| 297 |
+
"learning_rate": 0.0001,
|
| 298 |
+
"loss": 0.4945,
|
| 299 |
+
"step": 2800
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 8.49,
|
| 303 |
+
"learning_rate": 0.0001,
|
| 304 |
+
"loss": 0.4718,
|
| 305 |
+
"step": 2900
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 8.78,
|
| 309 |
+
"learning_rate": 0.0001,
|
| 310 |
+
"loss": 0.4825,
|
| 311 |
+
"step": 3000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 9.0,
|
| 315 |
+
"eval_accuracy": 0.4462051282051282,
|
| 316 |
+
"eval_loss": 4.34151554107666,
|
| 317 |
+
"eval_runtime": 4.8113,
|
| 318 |
+
"eval_samples_per_second": 103.921,
|
| 319 |
+
"eval_steps_per_second": 13.094,
|
| 320 |
+
"step": 3073
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 9.0,
|
| 324 |
+
"step": 3073
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 9.08,
|
| 328 |
+
"learning_rate": 0.0001,
|
| 329 |
+
"loss": 0.4538,
|
| 330 |
+
"step": 3100
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 9.37,
|
| 334 |
+
"learning_rate": 0.0001,
|
| 335 |
+
"loss": 0.3609,
|
| 336 |
+
"step": 3200
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 9.66,
|
| 340 |
+
"learning_rate": 0.0001,
|
| 341 |
+
"loss": 0.3769,
|
| 342 |
+
"step": 3300
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 9.96,
|
| 346 |
+
"learning_rate": 0.0001,
|
| 347 |
+
"loss": 0.387,
|
| 348 |
+
"step": 3400
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 10.0,
|
| 352 |
+
"eval_accuracy": 0.4463076923076923,
|
| 353 |
+
"eval_loss": 4.413115978240967,
|
| 354 |
+
"eval_runtime": 4.2874,
|
| 355 |
+
"eval_samples_per_second": 116.62,
|
| 356 |
+
"eval_steps_per_second": 14.694,
|
| 357 |
+
"step": 3415
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 10.0,
|
| 361 |
+
"step": 3415
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 10.25,
|
| 365 |
+
"learning_rate": 0.0001,
|
| 366 |
+
"loss": 0.2911,
|
| 367 |
+
"step": 3500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 10.54,
|
| 371 |
+
"learning_rate": 0.0001,
|
| 372 |
+
"loss": 0.2881,
|
| 373 |
+
"step": 3600
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 10.83,
|
| 377 |
+
"learning_rate": 0.0001,
|
| 378 |
+
"loss": 0.2933,
|
| 379 |
+
"step": 3700
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 11.0,
|
| 383 |
+
"eval_accuracy": 0.44343589743589745,
|
| 384 |
+
"eval_loss": 4.490556716918945,
|
| 385 |
+
"eval_runtime": 4.2824,
|
| 386 |
+
"eval_samples_per_second": 116.757,
|
| 387 |
+
"eval_steps_per_second": 14.711,
|
| 388 |
+
"step": 3756
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 11.0,
|
| 392 |
+
"step": 3756
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 11.13,
|
| 396 |
+
"learning_rate": 0.0001,
|
| 397 |
+
"loss": 0.2653,
|
| 398 |
+
"step": 3800
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 11.42,
|
| 402 |
+
"learning_rate": 0.0001,
|
| 403 |
+
"loss": 0.2181,
|
| 404 |
+
"step": 3900
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 11.71,
|
| 408 |
+
"learning_rate": 0.0001,
|
| 409 |
+
"loss": 0.2344,
|
| 410 |
+
"step": 4000
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 12.0,
|
| 414 |
+
"eval_accuracy": 0.44251282051282054,
|
| 415 |
+
"eval_loss": 4.651719570159912,
|
| 416 |
+
"eval_runtime": 4.0941,
|
| 417 |
+
"eval_samples_per_second": 122.126,
|
| 418 |
+
"eval_steps_per_second": 15.388,
|
| 419 |
+
"step": 4098
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 12.0,
|
| 423 |
+
"step": 4098
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 12.01,
|
| 427 |
+
"learning_rate": 0.0001,
|
| 428 |
+
"loss": 0.2414,
|
| 429 |
+
"step": 4100
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 12.3,
|
| 433 |
+
"learning_rate": 0.0001,
|
| 434 |
+
"loss": 0.173,
|
| 435 |
+
"step": 4200
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 12.59,
|
| 439 |
+
"learning_rate": 0.0001,
|
| 440 |
+
"loss": 0.183,
|
| 441 |
+
"step": 4300
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 12.88,
|
| 445 |
+
"learning_rate": 0.0001,
|
| 446 |
+
"loss": 0.1919,
|
| 447 |
+
"step": 4400
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 13.0,
|
| 451 |
+
"eval_accuracy": 0.44076923076923075,
|
| 452 |
+
"eval_loss": 4.751525402069092,
|
| 453 |
+
"eval_runtime": 4.0902,
|
| 454 |
+
"eval_samples_per_second": 122.244,
|
| 455 |
+
"eval_steps_per_second": 15.403,
|
| 456 |
+
"step": 4439
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 13.0,
|
| 460 |
+
"step": 4439
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 13.18,
|
| 464 |
+
"learning_rate": 0.0001,
|
| 465 |
+
"loss": 0.1661,
|
| 466 |
+
"step": 4500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 13.47,
|
| 470 |
+
"learning_rate": 0.0001,
|
| 471 |
+
"loss": 0.1501,
|
| 472 |
+
"step": 4600
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 13.76,
|
| 476 |
+
"learning_rate": 0.0001,
|
| 477 |
+
"loss": 0.1581,
|
| 478 |
+
"step": 4700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 14.0,
|
| 482 |
+
"eval_accuracy": 0.442051282051282,
|
| 483 |
+
"eval_loss": 4.7323455810546875,
|
| 484 |
+
"eval_runtime": 4.8128,
|
| 485 |
+
"eval_samples_per_second": 103.89,
|
| 486 |
+
"eval_steps_per_second": 13.09,
|
| 487 |
+
"step": 4781
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 14.0,
|
| 491 |
+
"eval_bleu": 0.01991302915032707,
|
| 492 |
+
"eval_exact_match": 0.0,
|
| 493 |
+
"eval_prefix_exact_match": 0.0,
|
| 494 |
+
"step": 4781
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 14.06,
|
| 498 |
+
"learning_rate": 0.0001,
|
| 499 |
+
"loss": 0.1563,
|
| 500 |
+
"step": 4800
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 14.35,
|
| 504 |
+
"learning_rate": 0.0001,
|
| 505 |
+
"loss": 0.13,
|
| 506 |
+
"step": 4900
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 14.64,
|
| 510 |
+
"learning_rate": 0.0001,
|
| 511 |
+
"loss": 0.1388,
|
| 512 |
+
"step": 5000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 14.93,
|
| 516 |
+
"learning_rate": 0.0001,
|
| 517 |
+
"loss": 0.1429,
|
| 518 |
+
"step": 5100
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 15.0,
|
| 522 |
+
"eval_accuracy": 0.44066666666666665,
|
| 523 |
+
"eval_loss": 4.810059547424316,
|
| 524 |
+
"eval_runtime": 5.0086,
|
| 525 |
+
"eval_samples_per_second": 99.829,
|
| 526 |
+
"eval_steps_per_second": 12.578,
|
| 527 |
+
"step": 5122
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 15.0,
|
| 531 |
+
"eval_bleu": 0.007355658423274405,
|
| 532 |
+
"eval_exact_match": 0.0,
|
| 533 |
+
"eval_prefix_exact_match": 0.002,
|
| 534 |
+
"step": 5122
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 15.23,
|
| 538 |
+
"learning_rate": 0.0001,
|
| 539 |
+
"loss": 0.1228,
|
| 540 |
+
"step": 5200
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 15.52,
|
| 544 |
+
"learning_rate": 0.0001,
|
| 545 |
+
"loss": 0.1233,
|
| 546 |
+
"step": 5300
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 15.81,
|
| 550 |
+
"learning_rate": 0.0001,
|
| 551 |
+
"loss": 0.1279,
|
| 552 |
+
"step": 5400
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 16.0,
|
| 556 |
+
"eval_accuracy": 0.44061538461538463,
|
| 557 |
+
"eval_loss": 4.84822940826416,
|
| 558 |
+
"eval_runtime": 4.1962,
|
| 559 |
+
"eval_samples_per_second": 119.156,
|
| 560 |
+
"eval_steps_per_second": 15.014,
|
| 561 |
+
"step": 5464
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 16.0,
|
| 565 |
+
"eval_bleu": 0.020518239891916933,
|
| 566 |
+
"eval_exact_match": 0.0,
|
| 567 |
+
"eval_prefix_exact_match": 0.0,
|
| 568 |
+
"step": 5464
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 16.11,
|
| 572 |
+
"learning_rate": 0.0001,
|
| 573 |
+
"loss": 0.1248,
|
| 574 |
+
"step": 5500
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 16.4,
|
| 578 |
+
"learning_rate": 0.0001,
|
| 579 |
+
"loss": 0.1118,
|
| 580 |
+
"step": 5600
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 16.69,
|
| 584 |
+
"learning_rate": 0.0001,
|
| 585 |
+
"loss": 0.1172,
|
| 586 |
+
"step": 5700
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 16.98,
|
| 590 |
+
"learning_rate": 0.0001,
|
| 591 |
+
"loss": 0.1231,
|
| 592 |
+
"step": 5800
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 17.0,
|
| 596 |
+
"eval_accuracy": 0.4410769230769231,
|
| 597 |
+
"eval_loss": 4.973475933074951,
|
| 598 |
+
"eval_runtime": 4.4042,
|
| 599 |
+
"eval_samples_per_second": 113.529,
|
| 600 |
+
"eval_steps_per_second": 14.305,
|
| 601 |
+
"step": 5805
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 17.0,
|
| 605 |
+
"eval_bleu": 0.025070041424787165,
|
| 606 |
+
"eval_exact_match": 0.0,
|
| 607 |
+
"eval_prefix_exact_match": 0.0,
|
| 608 |
+
"step": 5805
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 17.28,
|
| 612 |
+
"learning_rate": 0.0001,
|
| 613 |
+
"loss": 0.1068,
|
| 614 |
+
"step": 5900
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 17.57,
|
| 618 |
+
"learning_rate": 0.0001,
|
| 619 |
+
"loss": 0.111,
|
| 620 |
+
"step": 6000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 17.86,
|
| 624 |
+
"learning_rate": 0.0001,
|
| 625 |
+
"loss": 0.1145,
|
| 626 |
+
"step": 6100
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 18.0,
|
| 630 |
+
"eval_accuracy": 0.44153846153846155,
|
| 631 |
+
"eval_loss": 5.012055397033691,
|
| 632 |
+
"eval_runtime": 4.1955,
|
| 633 |
+
"eval_samples_per_second": 119.174,
|
| 634 |
+
"eval_steps_per_second": 15.016,
|
| 635 |
+
"step": 6147
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 18.0,
|
| 639 |
+
"eval_bleu": 0.0018720973379033757,
|
| 640 |
+
"eval_exact_match": 0.0,
|
| 641 |
+
"eval_prefix_exact_match": 0.0,
|
| 642 |
+
"step": 6147
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 18.16,
|
| 646 |
+
"learning_rate": 0.0001,
|
| 647 |
+
"loss": 0.1093,
|
| 648 |
+
"step": 6200
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 18.45,
|
| 652 |
+
"learning_rate": 0.0001,
|
| 653 |
+
"loss": 0.1051,
|
| 654 |
+
"step": 6300
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 18.74,
|
| 658 |
+
"learning_rate": 0.0001,
|
| 659 |
+
"loss": 0.1087,
|
| 660 |
+
"step": 6400
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 19.0,
|
| 664 |
+
"eval_accuracy": 0.43943589743589745,
|
| 665 |
+
"eval_loss": 4.983563423156738,
|
| 666 |
+
"eval_runtime": 4.0932,
|
| 667 |
+
"eval_samples_per_second": 122.154,
|
| 668 |
+
"eval_steps_per_second": 15.391,
|
| 669 |
+
"step": 6488
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 19.0,
|
| 673 |
+
"eval_bleu": 0.0014273377715875865,
|
| 674 |
+
"eval_exact_match": 0.0,
|
| 675 |
+
"eval_prefix_exact_match": 0.0,
|
| 676 |
+
"step": 6488
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 19.03,
|
| 680 |
+
"learning_rate": 0.0001,
|
| 681 |
+
"loss": 0.1115,
|
| 682 |
+
"step": 6500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 19.33,
|
| 686 |
+
"learning_rate": 0.0001,
|
| 687 |
+
"loss": 0.1001,
|
| 688 |
+
"step": 6600
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 19.62,
|
| 692 |
+
"learning_rate": 0.0001,
|
| 693 |
+
"loss": 0.1049,
|
| 694 |
+
"step": 6700
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 19.91,
|
| 698 |
+
"learning_rate": 0.0001,
|
| 699 |
+
"loss": 0.1084,
|
| 700 |
+
"step": 6800
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 20.0,
|
| 704 |
+
"eval_accuracy": 0.4388205128205128,
|
| 705 |
+
"eval_loss": 5.117064952850342,
|
| 706 |
+
"eval_runtime": 4.0908,
|
| 707 |
+
"eval_samples_per_second": 122.227,
|
| 708 |
+
"eval_steps_per_second": 15.401,
|
| 709 |
+
"step": 6830
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 20.0,
|
| 713 |
+
"eval_bleu": 0.004914730802266865,
|
| 714 |
+
"eval_exact_match": 0.0,
|
| 715 |
+
"eval_prefix_exact_match": 0.0,
|
| 716 |
+
"step": 6830
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 20.2,
|
| 720 |
+
"learning_rate": 0.0001,
|
| 721 |
+
"loss": 0.102,
|
| 722 |
+
"step": 6900
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 20.5,
|
| 726 |
+
"learning_rate": 0.0001,
|
| 727 |
+
"loss": 0.1014,
|
| 728 |
+
"step": 7000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 20.79,
|
| 732 |
+
"learning_rate": 0.0001,
|
| 733 |
+
"loss": 0.1069,
|
| 734 |
+
"step": 7100
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 21.0,
|
| 738 |
+
"eval_accuracy": 0.44051282051282054,
|
| 739 |
+
"eval_loss": 5.011991500854492,
|
| 740 |
+
"eval_runtime": 4.1961,
|
| 741 |
+
"eval_samples_per_second": 119.159,
|
| 742 |
+
"eval_steps_per_second": 15.014,
|
| 743 |
+
"step": 7171
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 21.0,
|
| 747 |
+
"step": 7171
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 21.08,
|
| 751 |
+
"learning_rate": 0.0001,
|
| 752 |
+
"loss": 0.1075,
|
| 753 |
+
"step": 7200
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 21.38,
|
| 757 |
+
"learning_rate": 0.0001,
|
| 758 |
+
"loss": 0.0985,
|
| 759 |
+
"step": 7300
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 21.67,
|
| 763 |
+
"learning_rate": 0.0001,
|
| 764 |
+
"loss": 0.1032,
|
| 765 |
+
"step": 7400
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 21.96,
|
| 769 |
+
"learning_rate": 0.0001,
|
| 770 |
+
"loss": 0.1075,
|
| 771 |
+
"step": 7500
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 22.0,
|
| 775 |
+
"eval_accuracy": 0.44,
|
| 776 |
+
"eval_loss": 5.234306812286377,
|
| 777 |
+
"eval_runtime": 4.4185,
|
| 778 |
+
"eval_samples_per_second": 113.162,
|
| 779 |
+
"eval_steps_per_second": 14.258,
|
| 780 |
+
"step": 7513
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 22.0,
|
| 784 |
+
"step": 7513
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 22.25,
|
| 788 |
+
"learning_rate": 0.0001,
|
| 789 |
+
"loss": 0.0959,
|
| 790 |
+
"step": 7600
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 22.55,
|
| 794 |
+
"learning_rate": 0.0001,
|
| 795 |
+
"loss": 0.0998,
|
| 796 |
+
"step": 7700
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 22.84,
|
| 800 |
+
"learning_rate": 0.0001,
|
| 801 |
+
"loss": 0.1024,
|
| 802 |
+
"step": 7800
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 23.0,
|
| 806 |
+
"eval_accuracy": 0.4408717948717949,
|
| 807 |
+
"eval_loss": 5.150094985961914,
|
| 808 |
+
"eval_runtime": 4.2006,
|
| 809 |
+
"eval_samples_per_second": 119.032,
|
| 810 |
+
"eval_steps_per_second": 14.998,
|
| 811 |
+
"step": 7854
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 23.0,
|
| 815 |
+
"eval_bleu": 0.008507706871345696,
|
| 816 |
+
"eval_exact_match": 0.0,
|
| 817 |
+
"eval_prefix_exact_match": 0.002,
|
| 818 |
+
"step": 7854
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 23.13,
|
| 822 |
+
"learning_rate": 0.0001,
|
| 823 |
+
"loss": 0.0997,
|
| 824 |
+
"step": 7900
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 23.43,
|
| 828 |
+
"learning_rate": 0.0001,
|
| 829 |
+
"loss": 0.0946,
|
| 830 |
+
"step": 8000
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 23.72,
|
| 834 |
+
"learning_rate": 0.0001,
|
| 835 |
+
"loss": 0.0981,
|
| 836 |
+
"step": 8100
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 24.0,
|
| 840 |
+
"eval_accuracy": 0.44025641025641027,
|
| 841 |
+
"eval_loss": 5.080117225646973,
|
| 842 |
+
"eval_runtime": 4.1942,
|
| 843 |
+
"eval_samples_per_second": 119.212,
|
| 844 |
+
"eval_steps_per_second": 15.021,
|
| 845 |
+
"step": 8196
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 24.0,
|
| 849 |
+
"step": 8196
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 24.01,
|
| 853 |
+
"learning_rate": 0.0001,
|
| 854 |
+
"loss": 0.101,
|
| 855 |
+
"step": 8200
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 24.3,
|
| 859 |
+
"learning_rate": 0.0001,
|
| 860 |
+
"loss": 0.0923,
|
| 861 |
+
"step": 8300
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 24.6,
|
| 865 |
+
"learning_rate": 0.0001,
|
| 866 |
+
"loss": 0.095,
|
| 867 |
+
"step": 8400
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 24.89,
|
| 871 |
+
"learning_rate": 0.0001,
|
| 872 |
+
"loss": 0.097,
|
| 873 |
+
"step": 8500
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 25.0,
|
| 877 |
+
"eval_accuracy": 0.44158974358974357,
|
| 878 |
+
"eval_loss": 5.103724956512451,
|
| 879 |
+
"eval_runtime": 4.307,
|
| 880 |
+
"eval_samples_per_second": 116.09,
|
| 881 |
+
"eval_steps_per_second": 14.627,
|
| 882 |
+
"step": 8537
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 25.0,
|
| 886 |
+
"step": 8537
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 25.18,
|
| 890 |
+
"learning_rate": 0.0001,
|
| 891 |
+
"loss": 0.0925,
|
| 892 |
+
"step": 8600
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 25.48,
|
| 896 |
+
"learning_rate": 0.0001,
|
| 897 |
+
"loss": 0.0926,
|
| 898 |
+
"step": 8700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 25.77,
|
| 902 |
+
"learning_rate": 0.0001,
|
| 903 |
+
"loss": 0.0963,
|
| 904 |
+
"step": 8800
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 26.0,
|
| 908 |
+
"eval_accuracy": 0.4397948717948718,
|
| 909 |
+
"eval_loss": 5.206360340118408,
|
| 910 |
+
"eval_runtime": 4.8198,
|
| 911 |
+
"eval_samples_per_second": 103.739,
|
| 912 |
+
"eval_steps_per_second": 13.071,
|
| 913 |
+
"step": 8879
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 26.0,
|
| 917 |
+
"step": 8879
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 26.06,
|
| 921 |
+
"learning_rate": 0.0001,
|
| 922 |
+
"loss": 0.0966,
|
| 923 |
+
"step": 8900
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 26.35,
|
| 927 |
+
"learning_rate": 0.0001,
|
| 928 |
+
"loss": 0.0909,
|
| 929 |
+
"step": 9000
|
| 930 |
+
},
|
| 931 |
+
{
|
| 932 |
+
"epoch": 26.65,
|
| 933 |
+
"learning_rate": 0.0001,
|
| 934 |
+
"loss": 0.0951,
|
| 935 |
+
"step": 9100
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 26.94,
|
| 939 |
+
"learning_rate": 0.0001,
|
| 940 |
+
"loss": 0.0983,
|
| 941 |
+
"step": 9200
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 27.0,
|
| 945 |
+
"eval_accuracy": 0.4413846153846154,
|
| 946 |
+
"eval_loss": 5.066434383392334,
|
| 947 |
+
"eval_runtime": 4.1912,
|
| 948 |
+
"eval_samples_per_second": 119.298,
|
| 949 |
+
"eval_steps_per_second": 15.032,
|
| 950 |
+
"step": 9220
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 27.0,
|
| 954 |
+
"step": 9220
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 27.23,
|
| 958 |
+
"learning_rate": 0.0001,
|
| 959 |
+
"loss": 0.092,
|
| 960 |
+
"step": 9300
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 27.53,
|
| 964 |
+
"learning_rate": 0.0001,
|
| 965 |
+
"loss": 0.0935,
|
| 966 |
+
"step": 9400
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 27.82,
|
| 970 |
+
"learning_rate": 0.0001,
|
| 971 |
+
"loss": 0.0969,
|
| 972 |
+
"step": 9500
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 28.0,
|
| 976 |
+
"eval_accuracy": 0.440974358974359,
|
| 977 |
+
"eval_loss": 5.255864143371582,
|
| 978 |
+
"eval_runtime": 4.1948,
|
| 979 |
+
"eval_samples_per_second": 119.194,
|
| 980 |
+
"eval_steps_per_second": 15.018,
|
| 981 |
+
"step": 9562
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 28.0,
|
| 985 |
+
"step": 9562
|
| 986 |
+
},
|
| 987 |
+
{
|
| 988 |
+
"epoch": 28.11,
|
| 989 |
+
"learning_rate": 0.0001,
|
| 990 |
+
"loss": 0.0947,
|
| 991 |
+
"step": 9600
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 28.4,
|
| 995 |
+
"learning_rate": 0.0001,
|
| 996 |
+
"loss": 0.0915,
|
| 997 |
+
"step": 9700
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 28.7,
|
| 1001 |
+
"learning_rate": 0.0001,
|
| 1002 |
+
"loss": 0.0948,
|
| 1003 |
+
"step": 9800
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 28.99,
|
| 1007 |
+
"learning_rate": 0.0001,
|
| 1008 |
+
"loss": 0.0966,
|
| 1009 |
+
"step": 9900
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 29.0,
|
| 1013 |
+
"eval_accuracy": 0.44035897435897436,
|
| 1014 |
+
"eval_loss": 5.196049690246582,
|
| 1015 |
+
"eval_runtime": 5.1571,
|
| 1016 |
+
"eval_samples_per_second": 96.953,
|
| 1017 |
+
"eval_steps_per_second": 12.216,
|
| 1018 |
+
"step": 9903
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 29.0,
|
| 1022 |
+
"step": 9903
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 29.28,
|
| 1026 |
+
"learning_rate": 0.0001,
|
| 1027 |
+
"loss": 0.0884,
|
| 1028 |
+
"step": 10000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 29.58,
|
| 1032 |
+
"learning_rate": 0.0001,
|
| 1033 |
+
"loss": 0.0913,
|
| 1034 |
+
"step": 10100
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 29.87,
|
| 1038 |
+
"learning_rate": 0.0001,
|
| 1039 |
+
"loss": 0.0954,
|
| 1040 |
+
"step": 10200
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 30.0,
|
| 1044 |
+
"eval_accuracy": 0.43958974358974356,
|
| 1045 |
+
"eval_loss": 5.223780632019043,
|
| 1046 |
+
"eval_runtime": 4.8025,
|
| 1047 |
+
"eval_samples_per_second": 104.112,
|
| 1048 |
+
"eval_steps_per_second": 13.118,
|
| 1049 |
+
"step": 10245
|
| 1050 |
+
},
|
| 1051 |
+
{
|
| 1052 |
+
"epoch": 30.0,
|
| 1053 |
+
"step": 10245
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 30.16,
|
| 1057 |
+
"learning_rate": 0.0001,
|
| 1058 |
+
"loss": 0.0913,
|
| 1059 |
+
"step": 10300
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 30.45,
|
| 1063 |
+
"learning_rate": 0.0001,
|
| 1064 |
+
"loss": 0.0901,
|
| 1065 |
+
"step": 10400
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 30.75,
|
| 1069 |
+
"learning_rate": 0.0001,
|
| 1070 |
+
"loss": 0.0931,
|
| 1071 |
+
"step": 10500
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 31.0,
|
| 1075 |
+
"eval_accuracy": 0.4402051282051282,
|
| 1076 |
+
"eval_loss": 5.219459056854248,
|
| 1077 |
+
"eval_runtime": 4.2447,
|
| 1078 |
+
"eval_samples_per_second": 117.794,
|
| 1079 |
+
"eval_steps_per_second": 14.842,
|
| 1080 |
+
"step": 10586
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 31.0,
|
| 1084 |
+
"step": 10586
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 31.04,
|
| 1088 |
+
"learning_rate": 0.0001,
|
| 1089 |
+
"loss": 0.0939,
|
| 1090 |
+
"step": 10600
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 31.33,
|
| 1094 |
+
"learning_rate": 0.0001,
|
| 1095 |
+
"loss": 0.0879,
|
| 1096 |
+
"step": 10700
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 31.63,
|
| 1100 |
+
"learning_rate": 0.0001,
|
| 1101 |
+
"loss": 0.0902,
|
| 1102 |
+
"step": 10800
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 31.92,
|
| 1106 |
+
"learning_rate": 0.0001,
|
| 1107 |
+
"loss": 0.0923,
|
| 1108 |
+
"step": 10900
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 32.0,
|
| 1112 |
+
"eval_accuracy": 0.44066666666666665,
|
| 1113 |
+
"eval_loss": 5.287062644958496,
|
| 1114 |
+
"eval_runtime": 4.1005,
|
| 1115 |
+
"eval_samples_per_second": 121.936,
|
| 1116 |
+
"eval_steps_per_second": 15.364,
|
| 1117 |
+
"step": 10928
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 32.0,
|
| 1121 |
+
"step": 10928
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 32.21,
|
| 1125 |
+
"learning_rate": 0.0001,
|
| 1126 |
+
"loss": 0.0881,
|
| 1127 |
+
"step": 11000
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 32.5,
|
| 1131 |
+
"learning_rate": 0.0001,
|
| 1132 |
+
"loss": 0.0891,
|
| 1133 |
+
"step": 11100
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 32.8,
|
| 1137 |
+
"learning_rate": 0.0001,
|
| 1138 |
+
"loss": 0.0911,
|
| 1139 |
+
"step": 11200
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 33.0,
|
| 1143 |
+
"eval_accuracy": 0.4391794871794872,
|
| 1144 |
+
"eval_loss": 5.320067405700684,
|
| 1145 |
+
"eval_runtime": 4.1991,
|
| 1146 |
+
"eval_samples_per_second": 119.074,
|
| 1147 |
+
"eval_steps_per_second": 15.003,
|
| 1148 |
+
"step": 11269
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 33.0,
|
| 1152 |
+
"step": 11269
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 33.09,
|
| 1156 |
+
"learning_rate": 0.0001,
|
| 1157 |
+
"loss": 0.0915,
|
| 1158 |
+
"step": 11300
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 33.38,
|
| 1162 |
+
"learning_rate": 0.0001,
|
| 1163 |
+
"loss": 0.0873,
|
| 1164 |
+
"step": 11400
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 33.67,
|
| 1168 |
+
"learning_rate": 0.0001,
|
| 1169 |
+
"loss": 0.0898,
|
| 1170 |
+
"step": 11500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 33.97,
|
| 1174 |
+
"learning_rate": 0.0001,
|
| 1175 |
+
"loss": 0.0934,
|
| 1176 |
+
"step": 11600
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 34.0,
|
| 1180 |
+
"eval_accuracy": 0.4387179487179487,
|
| 1181 |
+
"eval_loss": 5.362847328186035,
|
| 1182 |
+
"eval_runtime": 4.1975,
|
| 1183 |
+
"eval_samples_per_second": 119.119,
|
| 1184 |
+
"eval_steps_per_second": 15.009,
|
| 1185 |
+
"step": 11611
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 34.0,
|
| 1189 |
+
"step": 11611
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 34.26,
|
| 1193 |
+
"learning_rate": 0.0001,
|
| 1194 |
+
"loss": 0.086,
|
| 1195 |
+
"step": 11700
|
| 1196 |
+
},
|
| 1197 |
+
{
|
| 1198 |
+
"epoch": 34.55,
|
| 1199 |
+
"learning_rate": 0.0001,
|
| 1200 |
+
"loss": 0.0883,
|
| 1201 |
+
"step": 11800
|
| 1202 |
+
},
|
| 1203 |
+
{
|
| 1204 |
+
"epoch": 34.85,
|
| 1205 |
+
"learning_rate": 0.0001,
|
| 1206 |
+
"loss": 0.091,
|
| 1207 |
+
"step": 11900
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 35.0,
|
| 1211 |
+
"eval_accuracy": 0.438974358974359,
|
| 1212 |
+
"eval_loss": 5.319708347320557,
|
| 1213 |
+
"eval_runtime": 4.201,
|
| 1214 |
+
"eval_samples_per_second": 119.019,
|
| 1215 |
+
"eval_steps_per_second": 14.996,
|
| 1216 |
+
"step": 11952
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 35.0,
|
| 1220 |
+
"step": 11952
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 35.14,
|
| 1224 |
+
"learning_rate": 0.0001,
|
| 1225 |
+
"loss": 0.0897,
|
| 1226 |
+
"step": 12000
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 35.43,
|
| 1230 |
+
"learning_rate": 0.0001,
|
| 1231 |
+
"loss": 0.0876,
|
| 1232 |
+
"step": 12100
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 35.72,
|
| 1236 |
+
"learning_rate": 0.0001,
|
| 1237 |
+
"loss": 0.0902,
|
| 1238 |
+
"step": 12200
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 36.0,
|
| 1242 |
+
"eval_accuracy": 0.4391282051282051,
|
| 1243 |
+
"eval_loss": 5.186809539794922,
|
| 1244 |
+
"eval_runtime": 4.2,
|
| 1245 |
+
"eval_samples_per_second": 119.046,
|
| 1246 |
+
"eval_steps_per_second": 15.0,
|
| 1247 |
+
"step": 12294
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 36.0,
|
| 1251 |
+
"step": 12294
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 36.02,
|
| 1255 |
+
"learning_rate": 0.0001,
|
| 1256 |
+
"loss": 0.0931,
|
| 1257 |
+
"step": 12300
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 36.31,
|
| 1261 |
+
"learning_rate": 0.0001,
|
| 1262 |
+
"loss": 0.0857,
|
| 1263 |
+
"step": 12400
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 36.6,
|
| 1267 |
+
"learning_rate": 0.0001,
|
| 1268 |
+
"loss": 0.0894,
|
| 1269 |
+
"step": 12500
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 36.9,
|
| 1273 |
+
"learning_rate": 0.0001,
|
| 1274 |
+
"loss": 0.0916,
|
| 1275 |
+
"step": 12600
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 37.0,
|
| 1279 |
+
"eval_accuracy": 0.44235897435897437,
|
| 1280 |
+
"eval_loss": 5.122660160064697,
|
| 1281 |
+
"eval_runtime": 4.4215,
|
| 1282 |
+
"eval_samples_per_second": 113.083,
|
| 1283 |
+
"eval_steps_per_second": 14.248,
|
| 1284 |
+
"step": 12635
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 37.0,
|
| 1288 |
+
"step": 12635
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 37.19,
|
| 1292 |
+
"learning_rate": 0.0001,
|
| 1293 |
+
"loss": 0.0894,
|
| 1294 |
+
"step": 12700
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 37.48,
|
| 1298 |
+
"learning_rate": 0.0001,
|
| 1299 |
+
"loss": 0.0888,
|
| 1300 |
+
"step": 12800
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 37.77,
|
| 1304 |
+
"learning_rate": 0.0001,
|
| 1305 |
+
"loss": 0.0905,
|
| 1306 |
+
"step": 12900
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 38.0,
|
| 1310 |
+
"eval_accuracy": 0.43666666666666665,
|
| 1311 |
+
"eval_loss": 5.221393585205078,
|
| 1312 |
+
"eval_runtime": 4.6657,
|
| 1313 |
+
"eval_samples_per_second": 107.164,
|
| 1314 |
+
"eval_steps_per_second": 13.503,
|
| 1315 |
+
"step": 12977
|
| 1316 |
+
},
|
| 1317 |
+
{
|
| 1318 |
+
"epoch": 38.0,
|
| 1319 |
+
"step": 12977
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 38.07,
|
| 1323 |
+
"learning_rate": 0.0001,
|
| 1324 |
+
"loss": 0.0906,
|
| 1325 |
+
"step": 13000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 38.36,
|
| 1329 |
+
"learning_rate": 0.0001,
|
| 1330 |
+
"loss": 0.0857,
|
| 1331 |
+
"step": 13100
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 38.65,
|
| 1335 |
+
"learning_rate": 0.0001,
|
| 1336 |
+
"loss": 0.0884,
|
| 1337 |
+
"step": 13200
|
| 1338 |
+
},
|
| 1339 |
+
{
|
| 1340 |
+
"epoch": 38.95,
|
| 1341 |
+
"learning_rate": 0.0001,
|
| 1342 |
+
"loss": 0.0907,
|
| 1343 |
+
"step": 13300
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 39.0,
|
| 1347 |
+
"eval_accuracy": 0.4411794871794872,
|
| 1348 |
+
"eval_loss": 5.241151332855225,
|
| 1349 |
+
"eval_runtime": 4.1927,
|
| 1350 |
+
"eval_samples_per_second": 119.255,
|
| 1351 |
+
"eval_steps_per_second": 15.026,
|
| 1352 |
+
"step": 13318
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 39.0,
|
| 1356 |
+
"eval_bleu": 0.05134364405270662,
|
| 1357 |
+
"eval_exact_match": 0.0,
|
| 1358 |
+
"eval_prefix_exact_match": 0.006,
|
| 1359 |
+
"step": 13318
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 39.24,
|
| 1363 |
+
"learning_rate": 0.0001,
|
| 1364 |
+
"loss": 0.0862,
|
| 1365 |
+
"step": 13400
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 39.53,
|
| 1369 |
+
"learning_rate": 0.0001,
|
| 1370 |
+
"loss": 0.0864,
|
| 1371 |
+
"step": 13500
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 39.82,
|
| 1375 |
+
"learning_rate": 0.0001,
|
| 1376 |
+
"loss": 0.0883,
|
| 1377 |
+
"step": 13600
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 40.0,
|
| 1381 |
+
"eval_accuracy": 0.43948717948717947,
|
| 1382 |
+
"eval_loss": 5.301468372344971,
|
| 1383 |
+
"eval_runtime": 4.0986,
|
| 1384 |
+
"eval_samples_per_second": 121.992,
|
| 1385 |
+
"eval_steps_per_second": 15.371,
|
| 1386 |
+
"step": 13660
|
| 1387 |
+
},
|
| 1388 |
+
{
|
| 1389 |
+
"epoch": 40.0,
|
| 1390 |
+
"step": 13660
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 40.12,
|
| 1394 |
+
"learning_rate": 0.0001,
|
| 1395 |
+
"loss": 0.0864,
|
| 1396 |
+
"step": 13700
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 40.41,
|
| 1400 |
+
"learning_rate": 0.0001,
|
| 1401 |
+
"loss": 0.0838,
|
| 1402 |
+
"step": 13800
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 40.7,
|
| 1406 |
+
"learning_rate": 0.0001,
|
| 1407 |
+
"loss": 0.0863,
|
| 1408 |
+
"step": 13900
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 41.0,
|
| 1412 |
+
"learning_rate": 0.0001,
|
| 1413 |
+
"loss": 0.0892,
|
| 1414 |
+
"step": 14000
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 41.0,
|
| 1418 |
+
"eval_accuracy": 0.43923076923076926,
|
| 1419 |
+
"eval_loss": 5.2815632820129395,
|
| 1420 |
+
"eval_runtime": 4.4744,
|
| 1421 |
+
"eval_samples_per_second": 111.747,
|
| 1422 |
+
"eval_steps_per_second": 14.08,
|
| 1423 |
+
"step": 14001
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 41.0,
|
| 1427 |
+
"step": 14001
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 41.29,
|
| 1431 |
+
"learning_rate": 0.0001,
|
| 1432 |
+
"loss": 0.0835,
|
| 1433 |
+
"step": 14100
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 41.58,
|
| 1437 |
+
"learning_rate": 0.0001,
|
| 1438 |
+
"loss": 0.0854,
|
| 1439 |
+
"step": 14200
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 41.87,
|
| 1443 |
+
"learning_rate": 0.0001,
|
| 1444 |
+
"loss": 0.0881,
|
| 1445 |
+
"step": 14300
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 42.0,
|
| 1449 |
+
"eval_accuracy": 0.4350769230769231,
|
| 1450 |
+
"eval_loss": 5.358331203460693,
|
| 1451 |
+
"eval_runtime": 4.3999,
|
| 1452 |
+
"eval_samples_per_second": 113.64,
|
| 1453 |
+
"eval_steps_per_second": 14.319,
|
| 1454 |
+
"step": 14343
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 42.0,
|
| 1458 |
+
"step": 14343
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 42.17,
|
| 1462 |
+
"learning_rate": 0.0001,
|
| 1463 |
+
"loss": 0.0855,
|
| 1464 |
+
"step": 14400
|
| 1465 |
+
},
|
| 1466 |
+
{
|
| 1467 |
+
"epoch": 42.46,
|
| 1468 |
+
"learning_rate": 0.0001,
|
| 1469 |
+
"loss": 0.0855,
|
| 1470 |
+
"step": 14500
|
| 1471 |
+
},
|
| 1472 |
+
{
|
| 1473 |
+
"epoch": 42.75,
|
| 1474 |
+
"learning_rate": 0.0001,
|
| 1475 |
+
"loss": 0.0881,
|
| 1476 |
+
"step": 14600
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 43.0,
|
| 1480 |
+
"eval_accuracy": 0.43651282051282053,
|
| 1481 |
+
"eval_loss": 5.267802715301514,
|
| 1482 |
+
"eval_runtime": 4.9237,
|
| 1483 |
+
"eval_samples_per_second": 101.549,
|
| 1484 |
+
"eval_steps_per_second": 12.795,
|
| 1485 |
+
"step": 14684
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 43.0,
|
| 1489 |
+
"step": 14684
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 43.05,
|
| 1493 |
+
"learning_rate": 0.0001,
|
| 1494 |
+
"loss": 0.0887,
|
| 1495 |
+
"step": 14700
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 43.34,
|
| 1499 |
+
"learning_rate": 0.0001,
|
| 1500 |
+
"loss": 0.0841,
|
| 1501 |
+
"step": 14800
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 43.63,
|
| 1505 |
+
"learning_rate": 0.0001,
|
| 1506 |
+
"loss": 0.0867,
|
| 1507 |
+
"step": 14900
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 43.92,
|
| 1511 |
+
"learning_rate": 0.0001,
|
| 1512 |
+
"loss": 0.0898,
|
| 1513 |
+
"step": 15000
|
| 1514 |
+
},
|
| 1515 |
+
{
|
| 1516 |
+
"epoch": 44.0,
|
| 1517 |
+
"eval_accuracy": 0.4371794871794872,
|
| 1518 |
+
"eval_loss": 5.385359287261963,
|
| 1519 |
+
"eval_runtime": 4.4074,
|
| 1520 |
+
"eval_samples_per_second": 113.445,
|
| 1521 |
+
"eval_steps_per_second": 14.294,
|
| 1522 |
+
"step": 15026
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 44.0,
|
| 1526 |
+
"step": 15026
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 44.28,
|
| 1530 |
+
"learning_rate": 0.0001,
|
| 1531 |
+
"loss": 0.0828,
|
| 1532 |
+
"step": 15100
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 44.57,
|
| 1536 |
+
"learning_rate": 0.0001,
|
| 1537 |
+
"loss": 0.0855,
|
| 1538 |
+
"step": 15200
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 44.87,
|
| 1542 |
+
"learning_rate": 0.0001,
|
| 1543 |
+
"loss": 0.0874,
|
| 1544 |
+
"step": 15300
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 45.0,
|
| 1548 |
+
"eval_accuracy": 0.4391794871794872,
|
| 1549 |
+
"eval_loss": 5.356825828552246,
|
| 1550 |
+
"eval_runtime": 4.2113,
|
| 1551 |
+
"eval_samples_per_second": 118.728,
|
| 1552 |
+
"eval_steps_per_second": 14.96,
|
| 1553 |
+
"step": 15345
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 45.0,
|
| 1557 |
+
"step": 15345
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 45.16,
|
| 1561 |
+
"learning_rate": 0.0001,
|
| 1562 |
+
"loss": 0.086,
|
| 1563 |
+
"step": 15400
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 45.45,
|
| 1567 |
+
"learning_rate": 0.0001,
|
| 1568 |
+
"loss": 0.0848,
|
| 1569 |
+
"step": 15500
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 45.75,
|
| 1573 |
+
"learning_rate": 0.0001,
|
| 1574 |
+
"loss": 0.088,
|
| 1575 |
+
"step": 15600
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 46.0,
|
| 1579 |
+
"eval_accuracy": 0.4357948717948718,
|
| 1580 |
+
"eval_loss": 5.390822410583496,
|
| 1581 |
+
"eval_runtime": 4.4604,
|
| 1582 |
+
"eval_samples_per_second": 112.098,
|
| 1583 |
+
"eval_steps_per_second": 14.124,
|
| 1584 |
+
"step": 15687
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 46.0,
|
| 1588 |
+
"step": 15687
|
| 1589 |
+
},
|
| 1590 |
+
{
|
| 1591 |
+
"epoch": 46.04,
|
| 1592 |
+
"learning_rate": 0.0001,
|
| 1593 |
+
"loss": 0.0884,
|
| 1594 |
+
"step": 15700
|
| 1595 |
+
},
|
| 1596 |
+
{
|
| 1597 |
+
"epoch": 46.33,
|
| 1598 |
+
"learning_rate": 0.0001,
|
| 1599 |
+
"loss": 0.0831,
|
| 1600 |
+
"step": 15800
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 46.62,
|
| 1604 |
+
"learning_rate": 0.0001,
|
| 1605 |
+
"loss": 0.0856,
|
| 1606 |
+
"step": 15900
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 46.92,
|
| 1610 |
+
"learning_rate": 0.0001,
|
| 1611 |
+
"loss": 0.0885,
|
| 1612 |
+
"step": 16000
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 47.0,
|
| 1616 |
+
"eval_accuracy": 0.4366153846153846,
|
| 1617 |
+
"eval_loss": 5.268533229827881,
|
| 1618 |
+
"eval_runtime": 4.8401,
|
| 1619 |
+
"eval_samples_per_second": 103.304,
|
| 1620 |
+
"eval_steps_per_second": 13.016,
|
| 1621 |
+
"step": 16028
|
| 1622 |
+
},
|
| 1623 |
+
{
|
| 1624 |
+
"epoch": 47.0,
|
| 1625 |
+
"step": 16028
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 47.21,
|
| 1629 |
+
"learning_rate": 0.0001,
|
| 1630 |
+
"loss": 0.084,
|
| 1631 |
+
"step": 16100
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 47.5,
|
| 1635 |
+
"learning_rate": 0.0001,
|
| 1636 |
+
"loss": 0.0839,
|
| 1637 |
+
"step": 16200
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 47.8,
|
| 1641 |
+
"learning_rate": 0.0001,
|
| 1642 |
+
"loss": 0.0872,
|
| 1643 |
+
"step": 16300
|
| 1644 |
+
},
|
| 1645 |
+
{
|
| 1646 |
+
"epoch": 48.0,
|
| 1647 |
+
"eval_accuracy": 0.44,
|
| 1648 |
+
"eval_loss": 5.34999418258667,
|
| 1649 |
+
"eval_runtime": 4.4232,
|
| 1650 |
+
"eval_samples_per_second": 113.04,
|
| 1651 |
+
"eval_steps_per_second": 14.243,
|
| 1652 |
+
"step": 16370
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 48.0,
|
| 1656 |
+
"step": 16370
|
| 1657 |
+
}
|
| 1658 |
+
],
|
| 1659 |
+
"logging_steps": 100,
|
| 1660 |
+
"max_steps": 17050,
|
| 1661 |
+
"num_train_epochs": 50,
|
| 1662 |
+
"save_steps": 500,
|
| 1663 |
+
"total_flos": 3.2417395252689633e+18,
|
| 1664 |
+
"trial_name": null,
|
| 1665 |
+
"trial_params": null
|
| 1666 |
+
}
|
checkpoint-16370/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a770a6999335ba0cb69358d4265fd3fc082329f55fd3459b24ceed2e12ecfa54
|
| 3 |
+
size 4728
|