Update
Browse files- README.md +21 -3
- adapter_config.json +23 -0
- adapter_model.bin +3 -0
- added_tokens.json +3 -0
- optimizer.pt +3 -0
- rng_state_0.pth +3 -0
- rng_state_1.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +24 -0
- tokenizer.model +3 -0
- tokenizer_config.json +37 -0
- trainer_state.json +1687 -0
- training_args.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,21 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- quant_method: bitsandbytes
|
| 9 |
+
- load_in_8bit: False
|
| 10 |
+
- load_in_4bit: True
|
| 11 |
+
- llm_int8_threshold: 6.0
|
| 12 |
+
- llm_int8_skip_modules: None
|
| 13 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 14 |
+
- llm_int8_has_fp16_weight: False
|
| 15 |
+
- bnb_4bit_quant_type: nf4
|
| 16 |
+
- bnb_4bit_use_double_quant: True
|
| 17 |
+
- bnb_4bit_compute_dtype: float16
|
| 18 |
+
### Framework versions
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
- PEFT 0.5.0.dev0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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": 1024,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 512,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj",
|
| 19 |
+
"o_proj",
|
| 20 |
+
"k_proj"
|
| 21 |
+
],
|
| 22 |
+
"task_type": "CAUSAL_LM"
|
| 23 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c26c5ed712cd4680005313dcd1a59b5b2434676ef858737b8032a45a24325646
|
| 3 |
+
size 2147576970
|
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"[PAD]": 32000
|
| 3 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91d8f66c82c9bcec4d768dfc3cbf89a024f1c2edeeafe0a47bc85c2f0c47054b
|
| 3 |
+
size 4295183910
|
rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bcd37159b6ea83f170d8f8a6452d703884473aa8b9073ab29462b1d75b66e9cd
|
| 3 |
+
size 14512
|
rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d28209c71f32d16acfa3d8352a57a93876b8372f101c4aec0799c67caa8abf31
|
| 3 |
+
size 14512
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:75a7f63d45dd78b4fe8be018a152b447fe3ed04c3bbaa7448ede985142821726
|
| 3 |
+
size 1064
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "[PAD]",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "<s>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": false,
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "</s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"legacy": false,
|
| 22 |
+
"model_max_length": 512,
|
| 23 |
+
"pad_token": null,
|
| 24 |
+
"padding_side": "right",
|
| 25 |
+
"sp_model_kwargs": {},
|
| 26 |
+
"spaces_between_special_tokens": false,
|
| 27 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 28 |
+
"unk_token": {
|
| 29 |
+
"__type": "AddedToken",
|
| 30 |
+
"content": "<unk>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false
|
| 35 |
+
},
|
| 36 |
+
"use_default_system_prompt": true
|
| 37 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1687 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.995790064552343,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 2225,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 14 |
+
"loss": 1.5563,
|
| 15 |
+
"step": 8
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.04,
|
| 19 |
+
"learning_rate": 2.666666666666667e-06,
|
| 20 |
+
"loss": 1.4423,
|
| 21 |
+
"step": 16
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.05,
|
| 25 |
+
"learning_rate": 4.000000000000001e-06,
|
| 26 |
+
"loss": 1.3027,
|
| 27 |
+
"step": 24
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.07,
|
| 31 |
+
"learning_rate": 5.333333333333334e-06,
|
| 32 |
+
"loss": 1.2535,
|
| 33 |
+
"step": 32
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.09,
|
| 37 |
+
"learning_rate": 6.666666666666667e-06,
|
| 38 |
+
"loss": 1.2352,
|
| 39 |
+
"step": 40
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.11,
|
| 43 |
+
"learning_rate": 8.000000000000001e-06,
|
| 44 |
+
"loss": 1.2077,
|
| 45 |
+
"step": 48
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.13,
|
| 49 |
+
"learning_rate": 9.333333333333334e-06,
|
| 50 |
+
"loss": 1.1871,
|
| 51 |
+
"step": 56
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.14,
|
| 55 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 56 |
+
"loss": 1.1708,
|
| 57 |
+
"step": 64
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.16,
|
| 61 |
+
"learning_rate": 1.2e-05,
|
| 62 |
+
"loss": 1.1666,
|
| 63 |
+
"step": 72
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.18,
|
| 67 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 68 |
+
"loss": 1.1394,
|
| 69 |
+
"step": 80
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.2,
|
| 73 |
+
"learning_rate": 1.4666666666666668e-05,
|
| 74 |
+
"loss": 1.1644,
|
| 75 |
+
"step": 88
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.22,
|
| 79 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 80 |
+
"loss": 1.1521,
|
| 81 |
+
"step": 96
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.23,
|
| 85 |
+
"learning_rate": 1.7333333333333336e-05,
|
| 86 |
+
"loss": 1.142,
|
| 87 |
+
"step": 104
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.25,
|
| 91 |
+
"learning_rate": 1.866666666666667e-05,
|
| 92 |
+
"loss": 1.1314,
|
| 93 |
+
"step": 112
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.27,
|
| 97 |
+
"learning_rate": 2e-05,
|
| 98 |
+
"loss": 1.1495,
|
| 99 |
+
"step": 120
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.29,
|
| 103 |
+
"learning_rate": 2.1333333333333335e-05,
|
| 104 |
+
"loss": 1.1468,
|
| 105 |
+
"step": 128
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.31,
|
| 109 |
+
"learning_rate": 2.2666666666666668e-05,
|
| 110 |
+
"loss": 1.1457,
|
| 111 |
+
"step": 136
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.32,
|
| 115 |
+
"learning_rate": 2.4e-05,
|
| 116 |
+
"loss": 1.115,
|
| 117 |
+
"step": 144
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.34,
|
| 121 |
+
"learning_rate": 2.5333333333333337e-05,
|
| 122 |
+
"loss": 1.1279,
|
| 123 |
+
"step": 152
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.36,
|
| 127 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 128 |
+
"loss": 1.1145,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.38,
|
| 133 |
+
"learning_rate": 2.8000000000000003e-05,
|
| 134 |
+
"loss": 1.1209,
|
| 135 |
+
"step": 168
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.4,
|
| 139 |
+
"learning_rate": 2.9333333333333336e-05,
|
| 140 |
+
"loss": 1.1295,
|
| 141 |
+
"step": 176
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.41,
|
| 145 |
+
"learning_rate": 3.066666666666667e-05,
|
| 146 |
+
"loss": 1.1249,
|
| 147 |
+
"step": 184
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.43,
|
| 151 |
+
"learning_rate": 3.2000000000000005e-05,
|
| 152 |
+
"loss": 1.1294,
|
| 153 |
+
"step": 192
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.45,
|
| 157 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 158 |
+
"loss": 1.103,
|
| 159 |
+
"step": 200
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.47,
|
| 163 |
+
"learning_rate": 3.466666666666667e-05,
|
| 164 |
+
"loss": 1.1241,
|
| 165 |
+
"step": 208
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.48,
|
| 169 |
+
"learning_rate": 3.6e-05,
|
| 170 |
+
"loss": 1.1306,
|
| 171 |
+
"step": 216
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.5,
|
| 175 |
+
"learning_rate": 3.733333333333334e-05,
|
| 176 |
+
"loss": 1.1096,
|
| 177 |
+
"step": 224
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.52,
|
| 181 |
+
"learning_rate": 3.866666666666667e-05,
|
| 182 |
+
"loss": 1.1242,
|
| 183 |
+
"step": 232
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.54,
|
| 187 |
+
"learning_rate": 4e-05,
|
| 188 |
+
"loss": 1.1109,
|
| 189 |
+
"step": 240
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.56,
|
| 193 |
+
"learning_rate": 4.133333333333333e-05,
|
| 194 |
+
"loss": 1.1091,
|
| 195 |
+
"step": 248
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.57,
|
| 199 |
+
"learning_rate": 4.266666666666667e-05,
|
| 200 |
+
"loss": 1.1304,
|
| 201 |
+
"step": 256
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.59,
|
| 205 |
+
"learning_rate": 4.4000000000000006e-05,
|
| 206 |
+
"loss": 1.1149,
|
| 207 |
+
"step": 264
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.61,
|
| 211 |
+
"learning_rate": 4.5333333333333335e-05,
|
| 212 |
+
"loss": 1.106,
|
| 213 |
+
"step": 272
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.63,
|
| 217 |
+
"learning_rate": 4.666666666666667e-05,
|
| 218 |
+
"loss": 1.1139,
|
| 219 |
+
"step": 280
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.65,
|
| 223 |
+
"learning_rate": 4.8e-05,
|
| 224 |
+
"loss": 1.1174,
|
| 225 |
+
"step": 288
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.66,
|
| 229 |
+
"learning_rate": 4.933333333333334e-05,
|
| 230 |
+
"loss": 1.0849,
|
| 231 |
+
"step": 296
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.68,
|
| 235 |
+
"learning_rate": 5e-05,
|
| 236 |
+
"loss": 1.1107,
|
| 237 |
+
"step": 304
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.7,
|
| 241 |
+
"learning_rate": 5e-05,
|
| 242 |
+
"loss": 1.0828,
|
| 243 |
+
"step": 312
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.72,
|
| 247 |
+
"learning_rate": 5e-05,
|
| 248 |
+
"loss": 1.1032,
|
| 249 |
+
"step": 320
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.74,
|
| 253 |
+
"learning_rate": 5e-05,
|
| 254 |
+
"loss": 1.15,
|
| 255 |
+
"step": 328
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.75,
|
| 259 |
+
"learning_rate": 5e-05,
|
| 260 |
+
"loss": 1.1088,
|
| 261 |
+
"step": 336
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.77,
|
| 265 |
+
"learning_rate": 5e-05,
|
| 266 |
+
"loss": 1.1008,
|
| 267 |
+
"step": 344
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.79,
|
| 271 |
+
"learning_rate": 5e-05,
|
| 272 |
+
"loss": 1.1174,
|
| 273 |
+
"step": 352
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.81,
|
| 277 |
+
"learning_rate": 5e-05,
|
| 278 |
+
"loss": 1.1043,
|
| 279 |
+
"step": 360
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.83,
|
| 283 |
+
"learning_rate": 5e-05,
|
| 284 |
+
"loss": 1.1196,
|
| 285 |
+
"step": 368
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.84,
|
| 289 |
+
"learning_rate": 5e-05,
|
| 290 |
+
"loss": 1.1339,
|
| 291 |
+
"step": 376
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.86,
|
| 295 |
+
"learning_rate": 5e-05,
|
| 296 |
+
"loss": 1.1186,
|
| 297 |
+
"step": 384
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.88,
|
| 301 |
+
"learning_rate": 5e-05,
|
| 302 |
+
"loss": 1.0837,
|
| 303 |
+
"step": 392
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.9,
|
| 307 |
+
"learning_rate": 5e-05,
|
| 308 |
+
"loss": 1.1105,
|
| 309 |
+
"step": 400
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.92,
|
| 313 |
+
"learning_rate": 5e-05,
|
| 314 |
+
"loss": 1.1249,
|
| 315 |
+
"step": 408
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.93,
|
| 319 |
+
"learning_rate": 5e-05,
|
| 320 |
+
"loss": 1.0846,
|
| 321 |
+
"step": 416
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.95,
|
| 325 |
+
"learning_rate": 5e-05,
|
| 326 |
+
"loss": 1.0925,
|
| 327 |
+
"step": 424
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.97,
|
| 331 |
+
"learning_rate": 5e-05,
|
| 332 |
+
"loss": 1.0917,
|
| 333 |
+
"step": 432
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.99,
|
| 337 |
+
"learning_rate": 5e-05,
|
| 338 |
+
"loss": 1.1177,
|
| 339 |
+
"step": 440
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.01,
|
| 343 |
+
"learning_rate": 5e-05,
|
| 344 |
+
"loss": 1.0964,
|
| 345 |
+
"step": 448
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.02,
|
| 349 |
+
"learning_rate": 5e-05,
|
| 350 |
+
"loss": 1.0494,
|
| 351 |
+
"step": 456
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.04,
|
| 355 |
+
"learning_rate": 5e-05,
|
| 356 |
+
"loss": 1.0444,
|
| 357 |
+
"step": 464
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.06,
|
| 361 |
+
"learning_rate": 5e-05,
|
| 362 |
+
"loss": 1.043,
|
| 363 |
+
"step": 472
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.08,
|
| 367 |
+
"learning_rate": 5e-05,
|
| 368 |
+
"loss": 1.0303,
|
| 369 |
+
"step": 480
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.1,
|
| 373 |
+
"learning_rate": 5e-05,
|
| 374 |
+
"loss": 1.0862,
|
| 375 |
+
"step": 488
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.11,
|
| 379 |
+
"learning_rate": 5e-05,
|
| 380 |
+
"loss": 1.03,
|
| 381 |
+
"step": 496
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.13,
|
| 385 |
+
"learning_rate": 5e-05,
|
| 386 |
+
"loss": 1.0122,
|
| 387 |
+
"step": 504
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.15,
|
| 391 |
+
"learning_rate": 5e-05,
|
| 392 |
+
"loss": 1.0202,
|
| 393 |
+
"step": 512
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 1.17,
|
| 397 |
+
"learning_rate": 5e-05,
|
| 398 |
+
"loss": 1.0331,
|
| 399 |
+
"step": 520
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.19,
|
| 403 |
+
"learning_rate": 5e-05,
|
| 404 |
+
"loss": 1.0644,
|
| 405 |
+
"step": 528
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 1.2,
|
| 409 |
+
"learning_rate": 5e-05,
|
| 410 |
+
"loss": 1.0172,
|
| 411 |
+
"step": 536
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 1.22,
|
| 415 |
+
"learning_rate": 5e-05,
|
| 416 |
+
"loss": 1.0436,
|
| 417 |
+
"step": 544
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.24,
|
| 421 |
+
"learning_rate": 5e-05,
|
| 422 |
+
"loss": 1.0495,
|
| 423 |
+
"step": 552
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.26,
|
| 427 |
+
"learning_rate": 5e-05,
|
| 428 |
+
"loss": 1.0474,
|
| 429 |
+
"step": 560
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.28,
|
| 433 |
+
"learning_rate": 5e-05,
|
| 434 |
+
"loss": 1.0436,
|
| 435 |
+
"step": 568
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 1.29,
|
| 439 |
+
"learning_rate": 5e-05,
|
| 440 |
+
"loss": 1.0226,
|
| 441 |
+
"step": 576
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.31,
|
| 445 |
+
"learning_rate": 5e-05,
|
| 446 |
+
"loss": 1.0431,
|
| 447 |
+
"step": 584
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 1.33,
|
| 451 |
+
"learning_rate": 5e-05,
|
| 452 |
+
"loss": 1.0407,
|
| 453 |
+
"step": 592
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 1.35,
|
| 457 |
+
"learning_rate": 5e-05,
|
| 458 |
+
"loss": 1.033,
|
| 459 |
+
"step": 600
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 1.37,
|
| 463 |
+
"learning_rate": 5e-05,
|
| 464 |
+
"loss": 1.0202,
|
| 465 |
+
"step": 608
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 1.38,
|
| 469 |
+
"learning_rate": 5e-05,
|
| 470 |
+
"loss": 1.0791,
|
| 471 |
+
"step": 616
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.4,
|
| 475 |
+
"learning_rate": 5e-05,
|
| 476 |
+
"loss": 1.0281,
|
| 477 |
+
"step": 624
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.42,
|
| 481 |
+
"learning_rate": 5e-05,
|
| 482 |
+
"loss": 1.0793,
|
| 483 |
+
"step": 632
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 1.44,
|
| 487 |
+
"learning_rate": 5e-05,
|
| 488 |
+
"loss": 1.0557,
|
| 489 |
+
"step": 640
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 1.45,
|
| 493 |
+
"learning_rate": 5e-05,
|
| 494 |
+
"loss": 1.0413,
|
| 495 |
+
"step": 648
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 1.47,
|
| 499 |
+
"learning_rate": 5e-05,
|
| 500 |
+
"loss": 1.0679,
|
| 501 |
+
"step": 656
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 1.49,
|
| 505 |
+
"learning_rate": 5e-05,
|
| 506 |
+
"loss": 1.0334,
|
| 507 |
+
"step": 664
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.51,
|
| 511 |
+
"learning_rate": 5e-05,
|
| 512 |
+
"loss": 1.0163,
|
| 513 |
+
"step": 672
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.53,
|
| 517 |
+
"learning_rate": 5e-05,
|
| 518 |
+
"loss": 1.0454,
|
| 519 |
+
"step": 680
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.54,
|
| 523 |
+
"learning_rate": 5e-05,
|
| 524 |
+
"loss": 1.0489,
|
| 525 |
+
"step": 688
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 1.56,
|
| 529 |
+
"learning_rate": 5e-05,
|
| 530 |
+
"loss": 1.0437,
|
| 531 |
+
"step": 696
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 1.58,
|
| 535 |
+
"learning_rate": 5e-05,
|
| 536 |
+
"loss": 1.0278,
|
| 537 |
+
"step": 704
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 1.6,
|
| 541 |
+
"learning_rate": 5e-05,
|
| 542 |
+
"loss": 1.0165,
|
| 543 |
+
"step": 712
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 1.62,
|
| 547 |
+
"learning_rate": 5e-05,
|
| 548 |
+
"loss": 1.0673,
|
| 549 |
+
"step": 720
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 1.63,
|
| 553 |
+
"learning_rate": 5e-05,
|
| 554 |
+
"loss": 1.0469,
|
| 555 |
+
"step": 728
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.65,
|
| 559 |
+
"learning_rate": 5e-05,
|
| 560 |
+
"loss": 1.0453,
|
| 561 |
+
"step": 736
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 1.67,
|
| 565 |
+
"learning_rate": 5e-05,
|
| 566 |
+
"loss": 1.0523,
|
| 567 |
+
"step": 744
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 1.69,
|
| 571 |
+
"learning_rate": 5e-05,
|
| 572 |
+
"loss": 1.0433,
|
| 573 |
+
"step": 752
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 1.71,
|
| 577 |
+
"learning_rate": 5e-05,
|
| 578 |
+
"loss": 1.0156,
|
| 579 |
+
"step": 760
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 1.72,
|
| 583 |
+
"learning_rate": 5e-05,
|
| 584 |
+
"loss": 1.0424,
|
| 585 |
+
"step": 768
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.74,
|
| 589 |
+
"learning_rate": 5e-05,
|
| 590 |
+
"loss": 1.0216,
|
| 591 |
+
"step": 776
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 1.76,
|
| 595 |
+
"learning_rate": 5e-05,
|
| 596 |
+
"loss": 1.0369,
|
| 597 |
+
"step": 784
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.78,
|
| 601 |
+
"learning_rate": 5e-05,
|
| 602 |
+
"loss": 1.0509,
|
| 603 |
+
"step": 792
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 1.8,
|
| 607 |
+
"learning_rate": 5e-05,
|
| 608 |
+
"loss": 1.0619,
|
| 609 |
+
"step": 800
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 1.81,
|
| 613 |
+
"learning_rate": 5e-05,
|
| 614 |
+
"loss": 1.0164,
|
| 615 |
+
"step": 808
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 1.83,
|
| 619 |
+
"learning_rate": 5e-05,
|
| 620 |
+
"loss": 1.0256,
|
| 621 |
+
"step": 816
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 1.85,
|
| 625 |
+
"learning_rate": 5e-05,
|
| 626 |
+
"loss": 1.0028,
|
| 627 |
+
"step": 824
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 1.87,
|
| 631 |
+
"learning_rate": 5e-05,
|
| 632 |
+
"loss": 1.0244,
|
| 633 |
+
"step": 832
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.89,
|
| 637 |
+
"learning_rate": 5e-05,
|
| 638 |
+
"loss": 1.0494,
|
| 639 |
+
"step": 840
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.9,
|
| 643 |
+
"learning_rate": 5e-05,
|
| 644 |
+
"loss": 1.0387,
|
| 645 |
+
"step": 848
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 1.92,
|
| 649 |
+
"learning_rate": 5e-05,
|
| 650 |
+
"loss": 1.0168,
|
| 651 |
+
"step": 856
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 1.94,
|
| 655 |
+
"learning_rate": 5e-05,
|
| 656 |
+
"loss": 1.0455,
|
| 657 |
+
"step": 864
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 1.96,
|
| 661 |
+
"learning_rate": 5e-05,
|
| 662 |
+
"loss": 1.0622,
|
| 663 |
+
"step": 872
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 1.98,
|
| 667 |
+
"learning_rate": 5e-05,
|
| 668 |
+
"loss": 1.0564,
|
| 669 |
+
"step": 880
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 1.99,
|
| 673 |
+
"learning_rate": 5e-05,
|
| 674 |
+
"loss": 1.0656,
|
| 675 |
+
"step": 888
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 2.01,
|
| 679 |
+
"learning_rate": 5e-05,
|
| 680 |
+
"loss": 0.9658,
|
| 681 |
+
"step": 896
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 2.03,
|
| 685 |
+
"learning_rate": 5e-05,
|
| 686 |
+
"loss": 0.9094,
|
| 687 |
+
"step": 904
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 2.05,
|
| 691 |
+
"learning_rate": 5e-05,
|
| 692 |
+
"loss": 0.9343,
|
| 693 |
+
"step": 912
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 2.07,
|
| 697 |
+
"learning_rate": 5e-05,
|
| 698 |
+
"loss": 0.8941,
|
| 699 |
+
"step": 920
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.08,
|
| 703 |
+
"learning_rate": 5e-05,
|
| 704 |
+
"loss": 0.8903,
|
| 705 |
+
"step": 928
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 2.1,
|
| 709 |
+
"learning_rate": 5e-05,
|
| 710 |
+
"loss": 0.892,
|
| 711 |
+
"step": 936
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 2.12,
|
| 715 |
+
"learning_rate": 5e-05,
|
| 716 |
+
"loss": 0.9425,
|
| 717 |
+
"step": 944
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 2.14,
|
| 721 |
+
"learning_rate": 5e-05,
|
| 722 |
+
"loss": 0.932,
|
| 723 |
+
"step": 952
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.16,
|
| 727 |
+
"learning_rate": 5e-05,
|
| 728 |
+
"loss": 0.9171,
|
| 729 |
+
"step": 960
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 2.17,
|
| 733 |
+
"learning_rate": 5e-05,
|
| 734 |
+
"loss": 0.8976,
|
| 735 |
+
"step": 968
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 2.19,
|
| 739 |
+
"learning_rate": 5e-05,
|
| 740 |
+
"loss": 0.9055,
|
| 741 |
+
"step": 976
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 2.21,
|
| 745 |
+
"learning_rate": 5e-05,
|
| 746 |
+
"loss": 0.9442,
|
| 747 |
+
"step": 984
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 2.23,
|
| 751 |
+
"learning_rate": 5e-05,
|
| 752 |
+
"loss": 0.8948,
|
| 753 |
+
"step": 992
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 2.25,
|
| 757 |
+
"learning_rate": 5e-05,
|
| 758 |
+
"loss": 0.9297,
|
| 759 |
+
"step": 1000
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 2.26,
|
| 763 |
+
"learning_rate": 5e-05,
|
| 764 |
+
"loss": 0.8957,
|
| 765 |
+
"step": 1008
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.28,
|
| 769 |
+
"learning_rate": 5e-05,
|
| 770 |
+
"loss": 0.9264,
|
| 771 |
+
"step": 1016
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 2.3,
|
| 775 |
+
"learning_rate": 5e-05,
|
| 776 |
+
"loss": 0.9391,
|
| 777 |
+
"step": 1024
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 2.32,
|
| 781 |
+
"learning_rate": 5e-05,
|
| 782 |
+
"loss": 0.8936,
|
| 783 |
+
"step": 1032
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 2.34,
|
| 787 |
+
"learning_rate": 5e-05,
|
| 788 |
+
"loss": 0.9537,
|
| 789 |
+
"step": 1040
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.35,
|
| 793 |
+
"learning_rate": 5e-05,
|
| 794 |
+
"loss": 0.9168,
|
| 795 |
+
"step": 1048
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 2.37,
|
| 799 |
+
"learning_rate": 5e-05,
|
| 800 |
+
"loss": 0.9092,
|
| 801 |
+
"step": 1056
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 2.39,
|
| 805 |
+
"learning_rate": 5e-05,
|
| 806 |
+
"loss": 0.9198,
|
| 807 |
+
"step": 1064
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.41,
|
| 811 |
+
"learning_rate": 5e-05,
|
| 812 |
+
"loss": 0.9284,
|
| 813 |
+
"step": 1072
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.42,
|
| 817 |
+
"learning_rate": 5e-05,
|
| 818 |
+
"loss": 0.8926,
|
| 819 |
+
"step": 1080
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 2.44,
|
| 823 |
+
"learning_rate": 5e-05,
|
| 824 |
+
"loss": 0.9338,
|
| 825 |
+
"step": 1088
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 2.46,
|
| 829 |
+
"learning_rate": 5e-05,
|
| 830 |
+
"loss": 0.91,
|
| 831 |
+
"step": 1096
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.48,
|
| 835 |
+
"learning_rate": 5e-05,
|
| 836 |
+
"loss": 0.9444,
|
| 837 |
+
"step": 1104
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 2.5,
|
| 841 |
+
"learning_rate": 5e-05,
|
| 842 |
+
"loss": 0.9326,
|
| 843 |
+
"step": 1112
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 2.51,
|
| 847 |
+
"learning_rate": 5e-05,
|
| 848 |
+
"loss": 0.901,
|
| 849 |
+
"step": 1120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.53,
|
| 853 |
+
"learning_rate": 5e-05,
|
| 854 |
+
"loss": 0.9344,
|
| 855 |
+
"step": 1128
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 2.55,
|
| 859 |
+
"learning_rate": 5e-05,
|
| 860 |
+
"loss": 0.9344,
|
| 861 |
+
"step": 1136
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 2.57,
|
| 865 |
+
"learning_rate": 5e-05,
|
| 866 |
+
"loss": 0.9063,
|
| 867 |
+
"step": 1144
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.59,
|
| 871 |
+
"learning_rate": 5e-05,
|
| 872 |
+
"loss": 0.9123,
|
| 873 |
+
"step": 1152
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 2.6,
|
| 877 |
+
"learning_rate": 5e-05,
|
| 878 |
+
"loss": 0.9287,
|
| 879 |
+
"step": 1160
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 2.62,
|
| 883 |
+
"learning_rate": 5e-05,
|
| 884 |
+
"loss": 0.8998,
|
| 885 |
+
"step": 1168
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 2.64,
|
| 889 |
+
"learning_rate": 5e-05,
|
| 890 |
+
"loss": 0.9101,
|
| 891 |
+
"step": 1176
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.66,
|
| 895 |
+
"learning_rate": 5e-05,
|
| 896 |
+
"loss": 0.9353,
|
| 897 |
+
"step": 1184
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 2.68,
|
| 901 |
+
"learning_rate": 5e-05,
|
| 902 |
+
"loss": 0.942,
|
| 903 |
+
"step": 1192
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 2.69,
|
| 907 |
+
"learning_rate": 5e-05,
|
| 908 |
+
"loss": 0.9194,
|
| 909 |
+
"step": 1200
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.71,
|
| 913 |
+
"learning_rate": 5e-05,
|
| 914 |
+
"loss": 0.9221,
|
| 915 |
+
"step": 1208
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 2.73,
|
| 919 |
+
"learning_rate": 5e-05,
|
| 920 |
+
"loss": 0.9227,
|
| 921 |
+
"step": 1216
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 2.75,
|
| 925 |
+
"learning_rate": 5e-05,
|
| 926 |
+
"loss": 0.9409,
|
| 927 |
+
"step": 1224
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 2.77,
|
| 931 |
+
"learning_rate": 5e-05,
|
| 932 |
+
"loss": 0.9192,
|
| 933 |
+
"step": 1232
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.78,
|
| 937 |
+
"learning_rate": 5e-05,
|
| 938 |
+
"loss": 0.9274,
|
| 939 |
+
"step": 1240
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 2.8,
|
| 943 |
+
"learning_rate": 5e-05,
|
| 944 |
+
"loss": 0.9422,
|
| 945 |
+
"step": 1248
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 2.82,
|
| 949 |
+
"learning_rate": 5e-05,
|
| 950 |
+
"loss": 0.9236,
|
| 951 |
+
"step": 1256
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 2.84,
|
| 955 |
+
"learning_rate": 5e-05,
|
| 956 |
+
"loss": 0.9185,
|
| 957 |
+
"step": 1264
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 2.86,
|
| 961 |
+
"learning_rate": 5e-05,
|
| 962 |
+
"loss": 0.9179,
|
| 963 |
+
"step": 1272
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 2.87,
|
| 967 |
+
"learning_rate": 5e-05,
|
| 968 |
+
"loss": 0.9186,
|
| 969 |
+
"step": 1280
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 2.89,
|
| 973 |
+
"learning_rate": 5e-05,
|
| 974 |
+
"loss": 0.9117,
|
| 975 |
+
"step": 1288
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.91,
|
| 979 |
+
"learning_rate": 5e-05,
|
| 980 |
+
"loss": 0.9058,
|
| 981 |
+
"step": 1296
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 2.93,
|
| 985 |
+
"learning_rate": 5e-05,
|
| 986 |
+
"loss": 0.9486,
|
| 987 |
+
"step": 1304
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 2.95,
|
| 991 |
+
"learning_rate": 5e-05,
|
| 992 |
+
"loss": 0.9157,
|
| 993 |
+
"step": 1312
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 2.96,
|
| 997 |
+
"learning_rate": 5e-05,
|
| 998 |
+
"loss": 0.9119,
|
| 999 |
+
"step": 1320
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 2.98,
|
| 1003 |
+
"learning_rate": 5e-05,
|
| 1004 |
+
"loss": 0.9238,
|
| 1005 |
+
"step": 1328
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 3.0,
|
| 1009 |
+
"learning_rate": 5e-05,
|
| 1010 |
+
"loss": 0.9201,
|
| 1011 |
+
"step": 1336
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 3.02,
|
| 1015 |
+
"learning_rate": 5e-05,
|
| 1016 |
+
"loss": 0.7697,
|
| 1017 |
+
"step": 1344
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 3.04,
|
| 1021 |
+
"learning_rate": 5e-05,
|
| 1022 |
+
"loss": 0.7608,
|
| 1023 |
+
"step": 1352
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 3.05,
|
| 1027 |
+
"learning_rate": 5e-05,
|
| 1028 |
+
"loss": 0.7581,
|
| 1029 |
+
"step": 1360
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 3.07,
|
| 1033 |
+
"learning_rate": 5e-05,
|
| 1034 |
+
"loss": 0.7518,
|
| 1035 |
+
"step": 1368
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 3.09,
|
| 1039 |
+
"learning_rate": 5e-05,
|
| 1040 |
+
"loss": 0.7121,
|
| 1041 |
+
"step": 1376
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 3.11,
|
| 1045 |
+
"learning_rate": 5e-05,
|
| 1046 |
+
"loss": 0.7284,
|
| 1047 |
+
"step": 1384
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 3.13,
|
| 1051 |
+
"learning_rate": 5e-05,
|
| 1052 |
+
"loss": 0.7513,
|
| 1053 |
+
"step": 1392
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 3.14,
|
| 1057 |
+
"learning_rate": 5e-05,
|
| 1058 |
+
"loss": 0.7529,
|
| 1059 |
+
"step": 1400
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 3.16,
|
| 1063 |
+
"learning_rate": 5e-05,
|
| 1064 |
+
"loss": 0.7443,
|
| 1065 |
+
"step": 1408
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 3.18,
|
| 1069 |
+
"learning_rate": 5e-05,
|
| 1070 |
+
"loss": 0.7603,
|
| 1071 |
+
"step": 1416
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 3.2,
|
| 1075 |
+
"learning_rate": 5e-05,
|
| 1076 |
+
"loss": 0.7242,
|
| 1077 |
+
"step": 1424
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 3.22,
|
| 1081 |
+
"learning_rate": 5e-05,
|
| 1082 |
+
"loss": 0.7504,
|
| 1083 |
+
"step": 1432
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 3.23,
|
| 1087 |
+
"learning_rate": 5e-05,
|
| 1088 |
+
"loss": 0.7363,
|
| 1089 |
+
"step": 1440
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 3.25,
|
| 1093 |
+
"learning_rate": 5e-05,
|
| 1094 |
+
"loss": 0.7307,
|
| 1095 |
+
"step": 1448
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 3.27,
|
| 1099 |
+
"learning_rate": 5e-05,
|
| 1100 |
+
"loss": 0.757,
|
| 1101 |
+
"step": 1456
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 3.29,
|
| 1105 |
+
"learning_rate": 5e-05,
|
| 1106 |
+
"loss": 0.7337,
|
| 1107 |
+
"step": 1464
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 3.31,
|
| 1111 |
+
"learning_rate": 5e-05,
|
| 1112 |
+
"loss": 0.7338,
|
| 1113 |
+
"step": 1472
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 3.32,
|
| 1117 |
+
"learning_rate": 5e-05,
|
| 1118 |
+
"loss": 0.7632,
|
| 1119 |
+
"step": 1480
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 3.34,
|
| 1123 |
+
"learning_rate": 5e-05,
|
| 1124 |
+
"loss": 0.7643,
|
| 1125 |
+
"step": 1488
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 3.36,
|
| 1129 |
+
"learning_rate": 5e-05,
|
| 1130 |
+
"loss": 0.7458,
|
| 1131 |
+
"step": 1496
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 3.38,
|
| 1135 |
+
"learning_rate": 5e-05,
|
| 1136 |
+
"loss": 0.7704,
|
| 1137 |
+
"step": 1504
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 3.39,
|
| 1141 |
+
"learning_rate": 5e-05,
|
| 1142 |
+
"loss": 0.755,
|
| 1143 |
+
"step": 1512
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.41,
|
| 1147 |
+
"learning_rate": 5e-05,
|
| 1148 |
+
"loss": 0.7182,
|
| 1149 |
+
"step": 1520
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 3.43,
|
| 1153 |
+
"learning_rate": 5e-05,
|
| 1154 |
+
"loss": 0.7533,
|
| 1155 |
+
"step": 1528
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 3.45,
|
| 1159 |
+
"learning_rate": 5e-05,
|
| 1160 |
+
"loss": 0.7475,
|
| 1161 |
+
"step": 1536
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 3.47,
|
| 1165 |
+
"learning_rate": 5e-05,
|
| 1166 |
+
"loss": 0.7671,
|
| 1167 |
+
"step": 1544
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 3.48,
|
| 1171 |
+
"learning_rate": 5e-05,
|
| 1172 |
+
"loss": 0.7541,
|
| 1173 |
+
"step": 1552
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 3.5,
|
| 1177 |
+
"learning_rate": 5e-05,
|
| 1178 |
+
"loss": 0.7758,
|
| 1179 |
+
"step": 1560
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 3.52,
|
| 1183 |
+
"learning_rate": 5e-05,
|
| 1184 |
+
"loss": 0.7387,
|
| 1185 |
+
"step": 1568
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.54,
|
| 1189 |
+
"learning_rate": 5e-05,
|
| 1190 |
+
"loss": 0.7928,
|
| 1191 |
+
"step": 1576
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 3.56,
|
| 1195 |
+
"learning_rate": 5e-05,
|
| 1196 |
+
"loss": 0.7494,
|
| 1197 |
+
"step": 1584
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 3.57,
|
| 1201 |
+
"learning_rate": 5e-05,
|
| 1202 |
+
"loss": 0.7716,
|
| 1203 |
+
"step": 1592
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 3.59,
|
| 1207 |
+
"learning_rate": 5e-05,
|
| 1208 |
+
"loss": 0.7735,
|
| 1209 |
+
"step": 1600
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 3.61,
|
| 1213 |
+
"learning_rate": 5e-05,
|
| 1214 |
+
"loss": 0.7688,
|
| 1215 |
+
"step": 1608
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 3.63,
|
| 1219 |
+
"learning_rate": 5e-05,
|
| 1220 |
+
"loss": 0.7852,
|
| 1221 |
+
"step": 1616
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 3.65,
|
| 1225 |
+
"learning_rate": 5e-05,
|
| 1226 |
+
"loss": 0.7501,
|
| 1227 |
+
"step": 1624
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 3.66,
|
| 1231 |
+
"learning_rate": 5e-05,
|
| 1232 |
+
"loss": 0.7754,
|
| 1233 |
+
"step": 1632
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 3.68,
|
| 1237 |
+
"learning_rate": 5e-05,
|
| 1238 |
+
"loss": 0.756,
|
| 1239 |
+
"step": 1640
|
| 1240 |
+
},
|
| 1241 |
+
{
|
| 1242 |
+
"epoch": 3.7,
|
| 1243 |
+
"learning_rate": 5e-05,
|
| 1244 |
+
"loss": 0.7778,
|
| 1245 |
+
"step": 1648
|
| 1246 |
+
},
|
| 1247 |
+
{
|
| 1248 |
+
"epoch": 3.72,
|
| 1249 |
+
"learning_rate": 5e-05,
|
| 1250 |
+
"loss": 0.7835,
|
| 1251 |
+
"step": 1656
|
| 1252 |
+
},
|
| 1253 |
+
{
|
| 1254 |
+
"epoch": 3.74,
|
| 1255 |
+
"learning_rate": 5e-05,
|
| 1256 |
+
"loss": 0.7714,
|
| 1257 |
+
"step": 1664
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 3.75,
|
| 1261 |
+
"learning_rate": 5e-05,
|
| 1262 |
+
"loss": 0.7417,
|
| 1263 |
+
"step": 1672
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 3.77,
|
| 1267 |
+
"learning_rate": 5e-05,
|
| 1268 |
+
"loss": 0.7803,
|
| 1269 |
+
"step": 1680
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.79,
|
| 1273 |
+
"learning_rate": 5e-05,
|
| 1274 |
+
"loss": 0.7616,
|
| 1275 |
+
"step": 1688
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 3.81,
|
| 1279 |
+
"learning_rate": 5e-05,
|
| 1280 |
+
"loss": 0.7574,
|
| 1281 |
+
"step": 1696
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 3.83,
|
| 1285 |
+
"learning_rate": 5e-05,
|
| 1286 |
+
"loss": 0.7895,
|
| 1287 |
+
"step": 1704
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 3.84,
|
| 1291 |
+
"learning_rate": 5e-05,
|
| 1292 |
+
"loss": 0.7773,
|
| 1293 |
+
"step": 1712
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 3.86,
|
| 1297 |
+
"learning_rate": 5e-05,
|
| 1298 |
+
"loss": 0.7839,
|
| 1299 |
+
"step": 1720
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 3.88,
|
| 1303 |
+
"learning_rate": 5e-05,
|
| 1304 |
+
"loss": 0.7675,
|
| 1305 |
+
"step": 1728
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 3.9,
|
| 1309 |
+
"learning_rate": 5e-05,
|
| 1310 |
+
"loss": 0.7649,
|
| 1311 |
+
"step": 1736
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.92,
|
| 1315 |
+
"learning_rate": 5e-05,
|
| 1316 |
+
"loss": 0.7656,
|
| 1317 |
+
"step": 1744
|
| 1318 |
+
},
|
| 1319 |
+
{
|
| 1320 |
+
"epoch": 3.93,
|
| 1321 |
+
"learning_rate": 5e-05,
|
| 1322 |
+
"loss": 0.7725,
|
| 1323 |
+
"step": 1752
|
| 1324 |
+
},
|
| 1325 |
+
{
|
| 1326 |
+
"epoch": 3.95,
|
| 1327 |
+
"learning_rate": 5e-05,
|
| 1328 |
+
"loss": 0.7609,
|
| 1329 |
+
"step": 1760
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 3.97,
|
| 1333 |
+
"learning_rate": 5e-05,
|
| 1334 |
+
"loss": 0.7664,
|
| 1335 |
+
"step": 1768
|
| 1336 |
+
},
|
| 1337 |
+
{
|
| 1338 |
+
"epoch": 3.99,
|
| 1339 |
+
"learning_rate": 5e-05,
|
| 1340 |
+
"loss": 0.7945,
|
| 1341 |
+
"step": 1776
|
| 1342 |
+
},
|
| 1343 |
+
{
|
| 1344 |
+
"epoch": 4.01,
|
| 1345 |
+
"learning_rate": 5e-05,
|
| 1346 |
+
"loss": 0.7065,
|
| 1347 |
+
"step": 1784
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 4.02,
|
| 1351 |
+
"learning_rate": 5e-05,
|
| 1352 |
+
"loss": 0.5463,
|
| 1353 |
+
"step": 1792
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 4.04,
|
| 1357 |
+
"learning_rate": 5e-05,
|
| 1358 |
+
"loss": 0.549,
|
| 1359 |
+
"step": 1800
|
| 1360 |
+
},
|
| 1361 |
+
{
|
| 1362 |
+
"epoch": 4.06,
|
| 1363 |
+
"learning_rate": 5e-05,
|
| 1364 |
+
"loss": 0.571,
|
| 1365 |
+
"step": 1808
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 4.08,
|
| 1369 |
+
"learning_rate": 5e-05,
|
| 1370 |
+
"loss": 0.5538,
|
| 1371 |
+
"step": 1816
|
| 1372 |
+
},
|
| 1373 |
+
{
|
| 1374 |
+
"epoch": 4.1,
|
| 1375 |
+
"learning_rate": 5e-05,
|
| 1376 |
+
"loss": 0.5649,
|
| 1377 |
+
"step": 1824
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 4.11,
|
| 1381 |
+
"learning_rate": 5e-05,
|
| 1382 |
+
"loss": 0.5719,
|
| 1383 |
+
"step": 1832
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 4.13,
|
| 1387 |
+
"learning_rate": 5e-05,
|
| 1388 |
+
"loss": 0.5273,
|
| 1389 |
+
"step": 1840
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 4.15,
|
| 1393 |
+
"learning_rate": 5e-05,
|
| 1394 |
+
"loss": 0.5482,
|
| 1395 |
+
"step": 1848
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 4.17,
|
| 1399 |
+
"learning_rate": 5e-05,
|
| 1400 |
+
"loss": 0.5708,
|
| 1401 |
+
"step": 1856
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 4.19,
|
| 1405 |
+
"learning_rate": 5e-05,
|
| 1406 |
+
"loss": 0.5643,
|
| 1407 |
+
"step": 1864
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 4.2,
|
| 1411 |
+
"learning_rate": 5e-05,
|
| 1412 |
+
"loss": 0.5372,
|
| 1413 |
+
"step": 1872
|
| 1414 |
+
},
|
| 1415 |
+
{
|
| 1416 |
+
"epoch": 4.22,
|
| 1417 |
+
"learning_rate": 5e-05,
|
| 1418 |
+
"loss": 0.5623,
|
| 1419 |
+
"step": 1880
|
| 1420 |
+
},
|
| 1421 |
+
{
|
| 1422 |
+
"epoch": 4.24,
|
| 1423 |
+
"learning_rate": 5e-05,
|
| 1424 |
+
"loss": 0.5869,
|
| 1425 |
+
"step": 1888
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 4.26,
|
| 1429 |
+
"learning_rate": 5e-05,
|
| 1430 |
+
"loss": 0.5929,
|
| 1431 |
+
"step": 1896
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 4.28,
|
| 1435 |
+
"learning_rate": 5e-05,
|
| 1436 |
+
"loss": 0.5645,
|
| 1437 |
+
"step": 1904
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 4.29,
|
| 1441 |
+
"learning_rate": 5e-05,
|
| 1442 |
+
"loss": 0.5901,
|
| 1443 |
+
"step": 1912
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 4.31,
|
| 1447 |
+
"learning_rate": 5e-05,
|
| 1448 |
+
"loss": 0.5585,
|
| 1449 |
+
"step": 1920
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 4.33,
|
| 1453 |
+
"learning_rate": 5e-05,
|
| 1454 |
+
"loss": 0.571,
|
| 1455 |
+
"step": 1928
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 4.35,
|
| 1459 |
+
"learning_rate": 5e-05,
|
| 1460 |
+
"loss": 0.5824,
|
| 1461 |
+
"step": 1936
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 4.36,
|
| 1465 |
+
"learning_rate": 5e-05,
|
| 1466 |
+
"loss": 0.5801,
|
| 1467 |
+
"step": 1944
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 4.38,
|
| 1471 |
+
"learning_rate": 5e-05,
|
| 1472 |
+
"loss": 0.5736,
|
| 1473 |
+
"step": 1952
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 4.4,
|
| 1477 |
+
"learning_rate": 5e-05,
|
| 1478 |
+
"loss": 0.5874,
|
| 1479 |
+
"step": 1960
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.42,
|
| 1483 |
+
"learning_rate": 5e-05,
|
| 1484 |
+
"loss": 0.5795,
|
| 1485 |
+
"step": 1968
|
| 1486 |
+
},
|
| 1487 |
+
{
|
| 1488 |
+
"epoch": 4.44,
|
| 1489 |
+
"learning_rate": 5e-05,
|
| 1490 |
+
"loss": 0.5868,
|
| 1491 |
+
"step": 1976
|
| 1492 |
+
},
|
| 1493 |
+
{
|
| 1494 |
+
"epoch": 4.45,
|
| 1495 |
+
"learning_rate": 5e-05,
|
| 1496 |
+
"loss": 0.5649,
|
| 1497 |
+
"step": 1984
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 4.47,
|
| 1501 |
+
"learning_rate": 5e-05,
|
| 1502 |
+
"loss": 0.5769,
|
| 1503 |
+
"step": 1992
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 4.49,
|
| 1507 |
+
"learning_rate": 5e-05,
|
| 1508 |
+
"loss": 0.5944,
|
| 1509 |
+
"step": 2000
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 4.51,
|
| 1513 |
+
"learning_rate": 5e-05,
|
| 1514 |
+
"loss": 0.5876,
|
| 1515 |
+
"step": 2008
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 4.53,
|
| 1519 |
+
"learning_rate": 5e-05,
|
| 1520 |
+
"loss": 0.6052,
|
| 1521 |
+
"step": 2016
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 4.54,
|
| 1525 |
+
"learning_rate": 5e-05,
|
| 1526 |
+
"loss": 0.5994,
|
| 1527 |
+
"step": 2024
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 4.56,
|
| 1531 |
+
"learning_rate": 5e-05,
|
| 1532 |
+
"loss": 0.5684,
|
| 1533 |
+
"step": 2032
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 4.58,
|
| 1537 |
+
"learning_rate": 5e-05,
|
| 1538 |
+
"loss": 0.5927,
|
| 1539 |
+
"step": 2040
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 4.6,
|
| 1543 |
+
"learning_rate": 5e-05,
|
| 1544 |
+
"loss": 0.5857,
|
| 1545 |
+
"step": 2048
|
| 1546 |
+
},
|
| 1547 |
+
{
|
| 1548 |
+
"epoch": 4.62,
|
| 1549 |
+
"learning_rate": 5e-05,
|
| 1550 |
+
"loss": 0.566,
|
| 1551 |
+
"step": 2056
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 4.63,
|
| 1555 |
+
"learning_rate": 5e-05,
|
| 1556 |
+
"loss": 0.5763,
|
| 1557 |
+
"step": 2064
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 4.65,
|
| 1561 |
+
"learning_rate": 5e-05,
|
| 1562 |
+
"loss": 0.5997,
|
| 1563 |
+
"step": 2072
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 4.67,
|
| 1567 |
+
"learning_rate": 5e-05,
|
| 1568 |
+
"loss": 0.5655,
|
| 1569 |
+
"step": 2080
|
| 1570 |
+
},
|
| 1571 |
+
{
|
| 1572 |
+
"epoch": 4.69,
|
| 1573 |
+
"learning_rate": 5e-05,
|
| 1574 |
+
"loss": 0.5859,
|
| 1575 |
+
"step": 2088
|
| 1576 |
+
},
|
| 1577 |
+
{
|
| 1578 |
+
"epoch": 4.71,
|
| 1579 |
+
"learning_rate": 5e-05,
|
| 1580 |
+
"loss": 0.5636,
|
| 1581 |
+
"step": 2096
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 4.72,
|
| 1585 |
+
"learning_rate": 5e-05,
|
| 1586 |
+
"loss": 0.5964,
|
| 1587 |
+
"step": 2104
|
| 1588 |
+
},
|
| 1589 |
+
{
|
| 1590 |
+
"epoch": 4.74,
|
| 1591 |
+
"learning_rate": 5e-05,
|
| 1592 |
+
"loss": 0.5534,
|
| 1593 |
+
"step": 2112
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 4.76,
|
| 1597 |
+
"learning_rate": 5e-05,
|
| 1598 |
+
"loss": 0.5858,
|
| 1599 |
+
"step": 2120
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 4.78,
|
| 1603 |
+
"learning_rate": 5e-05,
|
| 1604 |
+
"loss": 0.5916,
|
| 1605 |
+
"step": 2128
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 4.8,
|
| 1609 |
+
"learning_rate": 5e-05,
|
| 1610 |
+
"loss": 0.5862,
|
| 1611 |
+
"step": 2136
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 4.81,
|
| 1615 |
+
"learning_rate": 5e-05,
|
| 1616 |
+
"loss": 0.6118,
|
| 1617 |
+
"step": 2144
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 4.83,
|
| 1621 |
+
"learning_rate": 5e-05,
|
| 1622 |
+
"loss": 0.6023,
|
| 1623 |
+
"step": 2152
|
| 1624 |
+
},
|
| 1625 |
+
{
|
| 1626 |
+
"epoch": 4.85,
|
| 1627 |
+
"learning_rate": 5e-05,
|
| 1628 |
+
"loss": 0.5927,
|
| 1629 |
+
"step": 2160
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 4.87,
|
| 1633 |
+
"learning_rate": 5e-05,
|
| 1634 |
+
"loss": 0.5911,
|
| 1635 |
+
"step": 2168
|
| 1636 |
+
},
|
| 1637 |
+
{
|
| 1638 |
+
"epoch": 4.89,
|
| 1639 |
+
"learning_rate": 5e-05,
|
| 1640 |
+
"loss": 0.6043,
|
| 1641 |
+
"step": 2176
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 4.9,
|
| 1645 |
+
"learning_rate": 5e-05,
|
| 1646 |
+
"loss": 0.5779,
|
| 1647 |
+
"step": 2184
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 4.92,
|
| 1651 |
+
"learning_rate": 5e-05,
|
| 1652 |
+
"loss": 0.6112,
|
| 1653 |
+
"step": 2192
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 4.94,
|
| 1657 |
+
"learning_rate": 5e-05,
|
| 1658 |
+
"loss": 0.5893,
|
| 1659 |
+
"step": 2200
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 4.96,
|
| 1663 |
+
"learning_rate": 5e-05,
|
| 1664 |
+
"loss": 0.597,
|
| 1665 |
+
"step": 2208
|
| 1666 |
+
},
|
| 1667 |
+
{
|
| 1668 |
+
"epoch": 4.98,
|
| 1669 |
+
"learning_rate": 5e-05,
|
| 1670 |
+
"loss": 0.5992,
|
| 1671 |
+
"step": 2216
|
| 1672 |
+
},
|
| 1673 |
+
{
|
| 1674 |
+
"epoch": 4.99,
|
| 1675 |
+
"learning_rate": 5e-05,
|
| 1676 |
+
"loss": 0.5933,
|
| 1677 |
+
"step": 2224
|
| 1678 |
+
}
|
| 1679 |
+
],
|
| 1680 |
+
"logging_steps": 8,
|
| 1681 |
+
"max_steps": 2225,
|
| 1682 |
+
"num_train_epochs": 5,
|
| 1683 |
+
"save_steps": 500,
|
| 1684 |
+
"total_flos": 1.5950431556342907e+18,
|
| 1685 |
+
"trial_name": null,
|
| 1686 |
+
"trial_params": null
|
| 1687 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bee34909e811e7da9196bddb6478e59b8b4726a3b15aed7e157b96d4b1b65baa
|
| 3 |
+
size 4792
|