Upload folder using huggingface_hub
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- RM-alfworld/checkpoint-460/adapter_model/README.md +9 -0
- RM-alfworld/checkpoint-460/adapter_model/lora_default/adapter_config.json +26 -0
- RM-alfworld/checkpoint-460/adapter_model/lora_default/adapter_model.bin +3 -0
- RM-alfworld/checkpoint-460/config.json +10 -0
- RM-alfworld/checkpoint-460/reward_head +0 -0
- RM-alfworld/checkpoint-460/rng_state_0.pth +3 -0
- RM-alfworld/checkpoint-460/rng_state_1.pth +3 -0
- RM-alfworld/checkpoint-460/rng_state_2.pth +3 -0
- RM-alfworld/checkpoint-460/rng_state_3.pth +3 -0
- RM-alfworld/checkpoint-460/scheduler.pt +3 -0
- RM-alfworld/checkpoint-460/special_tokens_map.json +24 -0
- RM-alfworld/checkpoint-460/tokenizer.model +3 -0
- RM-alfworld/checkpoint-460/tokenizer_config.json +72 -0
- RM-alfworld/checkpoint-460/trainer_state.json +1902 -0
- RM-alfworld/checkpoint-460/training_args.bin +3 -0
- RM-game24/checkpoint-1500/adapter_model/README.md +9 -0
- RM-game24/checkpoint-1500/adapter_model/lora_default/adapter_config.json +26 -0
- RM-game24/checkpoint-1500/adapter_model/lora_default/adapter_model.bin +3 -0
- RM-game24/checkpoint-1500/config.json +10 -0
- RM-game24/checkpoint-1500/reward_head +0 -0
- RM-game24/checkpoint-1500/rng_state_0.pth +3 -0
- RM-game24/checkpoint-1500/rng_state_1.pth +3 -0
- RM-game24/checkpoint-1500/rng_state_2.pth +3 -0
- RM-game24/checkpoint-1500/rng_state_3.pth +3 -0
- RM-game24/checkpoint-1500/scheduler.pt +3 -0
- RM-game24/checkpoint-1500/special_tokens_map.json +24 -0
- RM-game24/checkpoint-1500/tokenizer.model +3 -0
- RM-game24/checkpoint-1500/tokenizer_config.json +72 -0
- RM-game24/checkpoint-1500/trainer_state.json +0 -0
- RM-game24/checkpoint-1500/training_args.bin +3 -0
- RM-sciworld/checkpoint-120/adapter_model/README.md +9 -0
- RM-sciworld/checkpoint-120/adapter_model/lora_default/adapter_config.json +26 -0
- RM-sciworld/checkpoint-120/adapter_model/lora_default/adapter_model.bin +3 -0
- RM-sciworld/checkpoint-120/config.json +10 -0
- RM-sciworld/checkpoint-120/reward_head +0 -0
- RM-sciworld/checkpoint-120/rng_state_0.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_1.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_2.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_3.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_4.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_5.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_6.pth +3 -0
- RM-sciworld/checkpoint-120/rng_state_7.pth +3 -0
- RM-sciworld/checkpoint-120/scheduler.pt +3 -0
- RM-sciworld/checkpoint-120/special_tokens_map.json +24 -0
- RM-sciworld/checkpoint-120/tokenizer.model +3 -0
- RM-sciworld/checkpoint-120/tokenizer_config.json +72 -0
- RM-sciworld/checkpoint-120/trainer_state.json +442 -0
- RM-sciworld/checkpoint-120/training_args.bin +3 -0
- RM-webshop/checkpoint-1100/adapter_model/README.md +9 -0
RM-alfworld/checkpoint-460/adapter_model/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.4.0
|
RM-alfworld/checkpoint-460/adapter_model/lora_default/adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "/nobackup/users/zfchen/cdl/LLaVA-RLHF/VILA1.5-3b/llm",
|
| 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": 0.0,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 64,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"k_proj",
|
| 19 |
+
"v_proj",
|
| 20 |
+
"o_proj",
|
| 21 |
+
"gate_proj",
|
| 22 |
+
"up_proj",
|
| 23 |
+
"down_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
RM-alfworld/checkpoint-460/adapter_model/lora_default/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e258baa77264425ae0abbafbb7af65ca20e6ec14bf7a03e8ba1a96e5ae8222d
|
| 3 |
+
size 224393877
|
RM-alfworld/checkpoint-460/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RewardModel"
|
| 4 |
+
],
|
| 5 |
+
"backbone_model_name_or_path": "/nobackup/users/zfchen/cdl/LLaVA-RLHF/VILA1.5-3b/llm",
|
| 6 |
+
"mm_use_im_start_end": false,
|
| 7 |
+
"model_type": "reward_model",
|
| 8 |
+
"torch_dtype": "bfloat16",
|
| 9 |
+
"transformers_version": "4.30.2"
|
| 10 |
+
}
|
RM-alfworld/checkpoint-460/reward_head
ADDED
|
Binary file (11.3 kB). View file
|
|
|
RM-alfworld/checkpoint-460/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9a1197811f1a939dfbf0cb1cdd76120973cb8b73234a4aaaff4bbb85ec4f0b3
|
| 3 |
+
size 17655
|
RM-alfworld/checkpoint-460/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e91584c5da5fb3890c82bd485d367b56d34cff1ca99b435c1c4ffdfcbbca9fcd
|
| 3 |
+
size 17655
|
RM-alfworld/checkpoint-460/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8bc95e2a9a42eb285d0d838616b12da1bf146889ffc123af5212e906eaf53d88
|
| 3 |
+
size 17655
|
RM-alfworld/checkpoint-460/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5887ea61b512c6f77d8e2e072d4ac517fe341d1e626315a494a9e59019ac4ea5
|
| 3 |
+
size 17655
|
RM-alfworld/checkpoint-460/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:405fdfc89ca52eaccc190e6c9e1eda63c47ba993dec818a66cb932e02a7beb45
|
| 3 |
+
size 627
|
RM-alfworld/checkpoint-460/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": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
RM-alfworld/checkpoint-460/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7aedb3582ecda9fa99ee9242c17a9658f6744db083ee6ebdc8fb14857f84d220
|
| 3 |
+
size 499723
|
RM-alfworld/checkpoint-460/tokenizer_config.json
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": {
|
| 32 |
+
"__type": "AddedToken",
|
| 33 |
+
"content": "<s>",
|
| 34 |
+
"lstrip": false,
|
| 35 |
+
"normalized": true,
|
| 36 |
+
"rstrip": false,
|
| 37 |
+
"single_word": false
|
| 38 |
+
},
|
| 39 |
+
"clean_up_tokenization_spaces": false,
|
| 40 |
+
"eos_token": {
|
| 41 |
+
"__type": "AddedToken",
|
| 42 |
+
"content": "</s>",
|
| 43 |
+
"lstrip": false,
|
| 44 |
+
"normalized": true,
|
| 45 |
+
"rstrip": false,
|
| 46 |
+
"single_word": false
|
| 47 |
+
},
|
| 48 |
+
"legacy": false,
|
| 49 |
+
"model_max_length": 4096,
|
| 50 |
+
"pad_token": {
|
| 51 |
+
"__type": "AddedToken",
|
| 52 |
+
"content": "<unk>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": true,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false
|
| 57 |
+
},
|
| 58 |
+
"padding_side": "left",
|
| 59 |
+
"sp_model_kwargs": {},
|
| 60 |
+
"spaces_between_special_tokens": false,
|
| 61 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 62 |
+
"truncation_side": "right",
|
| 63 |
+
"unk_token": {
|
| 64 |
+
"__type": "AddedToken",
|
| 65 |
+
"content": "<unk>",
|
| 66 |
+
"lstrip": false,
|
| 67 |
+
"normalized": true,
|
| 68 |
+
"rstrip": false,
|
| 69 |
+
"single_word": false
|
| 70 |
+
},
|
| 71 |
+
"use_default_system_prompt": false
|
| 72 |
+
}
|
RM-alfworld/checkpoint-460/trainer_state.json
ADDED
|
@@ -0,0 +1,1902 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.484536082474227,
|
| 5 |
+
"global_step": 460,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 2.0833333333333333e-07,
|
| 13 |
+
"loss": 0.7393,
|
| 14 |
+
"step": 2
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"learning_rate": 4.1666666666666667e-07,
|
| 19 |
+
"loss": 0.7516,
|
| 20 |
+
"step": 4
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.12,
|
| 24 |
+
"learning_rate": 6.25e-07,
|
| 25 |
+
"loss": 0.7506,
|
| 26 |
+
"step": 6
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.16,
|
| 30 |
+
"learning_rate": 8.333333333333333e-07,
|
| 31 |
+
"loss": 0.7494,
|
| 32 |
+
"step": 8
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.21,
|
| 36 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 37 |
+
"loss": 0.7602,
|
| 38 |
+
"step": 10
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.21,
|
| 42 |
+
"eval_accuracy": 0.6200000047683716,
|
| 43 |
+
"eval_average_score": 0.6218893527984619,
|
| 44 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 45 |
+
"eval_loss": 0.7480455040931702,
|
| 46 |
+
"eval_runtime": 3.4088,
|
| 47 |
+
"eval_samples_per_second": 14.668,
|
| 48 |
+
"eval_steps_per_second": 3.814,
|
| 49 |
+
"step": 10
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 0.25,
|
| 53 |
+
"learning_rate": 1.25e-06,
|
| 54 |
+
"loss": 0.7501,
|
| 55 |
+
"step": 12
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 0.29,
|
| 59 |
+
"learning_rate": 1.4583333333333335e-06,
|
| 60 |
+
"loss": 0.743,
|
| 61 |
+
"step": 14
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 0.33,
|
| 65 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 66 |
+
"loss": 0.7412,
|
| 67 |
+
"step": 16
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.37,
|
| 71 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 72 |
+
"loss": 0.7553,
|
| 73 |
+
"step": 18
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.41,
|
| 77 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 78 |
+
"loss": 0.7355,
|
| 79 |
+
"step": 20
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.41,
|
| 83 |
+
"eval_accuracy": 0.5400000214576721,
|
| 84 |
+
"eval_average_score": 0.5707897543907166,
|
| 85 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 86 |
+
"eval_loss": 0.7415069341659546,
|
| 87 |
+
"eval_runtime": 3.405,
|
| 88 |
+
"eval_samples_per_second": 14.684,
|
| 89 |
+
"eval_steps_per_second": 3.818,
|
| 90 |
+
"step": 20
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.45,
|
| 94 |
+
"learning_rate": 2.2916666666666666e-06,
|
| 95 |
+
"loss": 0.7402,
|
| 96 |
+
"step": 22
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.49,
|
| 100 |
+
"learning_rate": 2.5e-06,
|
| 101 |
+
"loss": 0.7427,
|
| 102 |
+
"step": 24
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.54,
|
| 106 |
+
"learning_rate": 2.7083333333333334e-06,
|
| 107 |
+
"loss": 0.7286,
|
| 108 |
+
"step": 26
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.58,
|
| 112 |
+
"learning_rate": 2.916666666666667e-06,
|
| 113 |
+
"loss": 0.727,
|
| 114 |
+
"step": 28
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.62,
|
| 118 |
+
"learning_rate": 3.125e-06,
|
| 119 |
+
"loss": 0.7305,
|
| 120 |
+
"step": 30
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.62,
|
| 124 |
+
"eval_accuracy": 0.6000000238418579,
|
| 125 |
+
"eval_average_score": 0.4747978150844574,
|
| 126 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 127 |
+
"eval_loss": 0.7272794842720032,
|
| 128 |
+
"eval_runtime": 3.41,
|
| 129 |
+
"eval_samples_per_second": 14.663,
|
| 130 |
+
"eval_steps_per_second": 3.812,
|
| 131 |
+
"step": 30
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.66,
|
| 135 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 136 |
+
"loss": 0.7186,
|
| 137 |
+
"step": 32
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.7,
|
| 141 |
+
"learning_rate": 3.5416666666666673e-06,
|
| 142 |
+
"loss": 0.7398,
|
| 143 |
+
"step": 34
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.74,
|
| 147 |
+
"learning_rate": 3.7500000000000005e-06,
|
| 148 |
+
"loss": 0.7166,
|
| 149 |
+
"step": 36
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.78,
|
| 153 |
+
"learning_rate": 3.958333333333333e-06,
|
| 154 |
+
"loss": 0.7148,
|
| 155 |
+
"step": 38
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.82,
|
| 159 |
+
"learning_rate": 4.166666666666667e-06,
|
| 160 |
+
"loss": 0.7116,
|
| 161 |
+
"step": 40
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.82,
|
| 165 |
+
"eval_accuracy": 0.6000000238418579,
|
| 166 |
+
"eval_average_score": 0.3107444643974304,
|
| 167 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 168 |
+
"eval_loss": 0.7109410166740417,
|
| 169 |
+
"eval_runtime": 3.4072,
|
| 170 |
+
"eval_samples_per_second": 14.675,
|
| 171 |
+
"eval_steps_per_second": 3.815,
|
| 172 |
+
"step": 40
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.87,
|
| 176 |
+
"learning_rate": 4.3750000000000005e-06,
|
| 177 |
+
"loss": 0.7221,
|
| 178 |
+
"step": 42
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.91,
|
| 182 |
+
"learning_rate": 4.583333333333333e-06,
|
| 183 |
+
"loss": 0.7042,
|
| 184 |
+
"step": 44
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.95,
|
| 188 |
+
"learning_rate": 4.791666666666668e-06,
|
| 189 |
+
"loss": 0.7011,
|
| 190 |
+
"step": 46
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.99,
|
| 194 |
+
"learning_rate": 5e-06,
|
| 195 |
+
"loss": 0.6987,
|
| 196 |
+
"step": 48
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.03,
|
| 200 |
+
"learning_rate": 5.208333333333334e-06,
|
| 201 |
+
"loss": 0.6926,
|
| 202 |
+
"step": 50
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.03,
|
| 206 |
+
"eval_accuracy": 0.6800000071525574,
|
| 207 |
+
"eval_average_score": 0.07106868177652359,
|
| 208 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 209 |
+
"eval_loss": 0.6947080492973328,
|
| 210 |
+
"eval_runtime": 3.4106,
|
| 211 |
+
"eval_samples_per_second": 14.66,
|
| 212 |
+
"eval_steps_per_second": 3.812,
|
| 213 |
+
"step": 50
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.07,
|
| 217 |
+
"learning_rate": 5.416666666666667e-06,
|
| 218 |
+
"loss": 0.695,
|
| 219 |
+
"step": 52
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.11,
|
| 223 |
+
"learning_rate": 5.625e-06,
|
| 224 |
+
"loss": 0.6917,
|
| 225 |
+
"step": 54
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 1.15,
|
| 229 |
+
"learning_rate": 5.833333333333334e-06,
|
| 230 |
+
"loss": 0.7115,
|
| 231 |
+
"step": 56
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.2,
|
| 235 |
+
"learning_rate": 6.041666666666667e-06,
|
| 236 |
+
"loss": 0.6858,
|
| 237 |
+
"step": 58
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.24,
|
| 241 |
+
"learning_rate": 6.25e-06,
|
| 242 |
+
"loss": 0.6867,
|
| 243 |
+
"step": 60
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.24,
|
| 247 |
+
"eval_accuracy": 0.6000000238418579,
|
| 248 |
+
"eval_average_score": -0.15788692235946655,
|
| 249 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 250 |
+
"eval_loss": 0.6935771703720093,
|
| 251 |
+
"eval_runtime": 3.4133,
|
| 252 |
+
"eval_samples_per_second": 14.648,
|
| 253 |
+
"eval_steps_per_second": 3.809,
|
| 254 |
+
"step": 60
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.28,
|
| 258 |
+
"learning_rate": 6.458333333333334e-06,
|
| 259 |
+
"loss": 0.6868,
|
| 260 |
+
"step": 62
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.32,
|
| 264 |
+
"learning_rate": 6.666666666666667e-06,
|
| 265 |
+
"loss": 0.6837,
|
| 266 |
+
"step": 64
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.36,
|
| 270 |
+
"learning_rate": 6.875e-06,
|
| 271 |
+
"loss": 0.6883,
|
| 272 |
+
"step": 66
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 1.4,
|
| 276 |
+
"learning_rate": 7.083333333333335e-06,
|
| 277 |
+
"loss": 0.689,
|
| 278 |
+
"step": 68
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.44,
|
| 282 |
+
"learning_rate": 7.291666666666667e-06,
|
| 283 |
+
"loss": 0.6761,
|
| 284 |
+
"step": 70
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.44,
|
| 288 |
+
"eval_accuracy": 0.7799999713897705,
|
| 289 |
+
"eval_average_score": -0.07916055619716644,
|
| 290 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 291 |
+
"eval_loss": 0.6847817897796631,
|
| 292 |
+
"eval_runtime": 3.4111,
|
| 293 |
+
"eval_samples_per_second": 14.658,
|
| 294 |
+
"eval_steps_per_second": 3.811,
|
| 295 |
+
"step": 70
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.48,
|
| 299 |
+
"learning_rate": 7.500000000000001e-06,
|
| 300 |
+
"loss": 0.6863,
|
| 301 |
+
"step": 72
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 1.53,
|
| 305 |
+
"learning_rate": 7.708333333333334e-06,
|
| 306 |
+
"loss": 0.6841,
|
| 307 |
+
"step": 74
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 1.57,
|
| 311 |
+
"learning_rate": 7.916666666666667e-06,
|
| 312 |
+
"loss": 0.6719,
|
| 313 |
+
"step": 76
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.61,
|
| 317 |
+
"learning_rate": 8.125000000000001e-06,
|
| 318 |
+
"loss": 0.6802,
|
| 319 |
+
"step": 78
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.65,
|
| 323 |
+
"learning_rate": 8.333333333333334e-06,
|
| 324 |
+
"loss": 0.672,
|
| 325 |
+
"step": 80
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.65,
|
| 329 |
+
"eval_accuracy": 0.8199999928474426,
|
| 330 |
+
"eval_average_score": -0.042581118643283844,
|
| 331 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 332 |
+
"eval_loss": 0.6735163331031799,
|
| 333 |
+
"eval_runtime": 3.4082,
|
| 334 |
+
"eval_samples_per_second": 14.67,
|
| 335 |
+
"eval_steps_per_second": 3.814,
|
| 336 |
+
"step": 80
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.69,
|
| 340 |
+
"learning_rate": 8.541666666666666e-06,
|
| 341 |
+
"loss": 0.6675,
|
| 342 |
+
"step": 82
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.73,
|
| 346 |
+
"learning_rate": 8.750000000000001e-06,
|
| 347 |
+
"loss": 0.6649,
|
| 348 |
+
"step": 84
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.77,
|
| 352 |
+
"learning_rate": 8.958333333333334e-06,
|
| 353 |
+
"loss": 0.6804,
|
| 354 |
+
"step": 86
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.81,
|
| 358 |
+
"learning_rate": 9.166666666666666e-06,
|
| 359 |
+
"loss": 0.6692,
|
| 360 |
+
"step": 88
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.86,
|
| 364 |
+
"learning_rate": 9.375000000000001e-06,
|
| 365 |
+
"loss": 0.654,
|
| 366 |
+
"step": 90
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.86,
|
| 370 |
+
"eval_accuracy": 0.8600000143051147,
|
| 371 |
+
"eval_average_score": -0.11977884918451309,
|
| 372 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 373 |
+
"eval_loss": 0.652625560760498,
|
| 374 |
+
"eval_runtime": 3.4105,
|
| 375 |
+
"eval_samples_per_second": 14.661,
|
| 376 |
+
"eval_steps_per_second": 3.812,
|
| 377 |
+
"step": 90
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 1.9,
|
| 381 |
+
"learning_rate": 9.583333333333335e-06,
|
| 382 |
+
"loss": 0.6495,
|
| 383 |
+
"step": 92
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.94,
|
| 387 |
+
"learning_rate": 9.791666666666666e-06,
|
| 388 |
+
"loss": 0.6472,
|
| 389 |
+
"step": 94
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.98,
|
| 393 |
+
"learning_rate": 1e-05,
|
| 394 |
+
"loss": 0.6499,
|
| 395 |
+
"step": 96
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 2.02,
|
| 399 |
+
"learning_rate": 1e-05,
|
| 400 |
+
"loss": 0.6247,
|
| 401 |
+
"step": 98
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 2.06,
|
| 405 |
+
"learning_rate": 1e-05,
|
| 406 |
+
"loss": 0.6422,
|
| 407 |
+
"step": 100
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 2.06,
|
| 411 |
+
"eval_accuracy": 0.8799999952316284,
|
| 412 |
+
"eval_average_score": -0.13170765340328217,
|
| 413 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 414 |
+
"eval_loss": 0.608478307723999,
|
| 415 |
+
"eval_runtime": 3.4104,
|
| 416 |
+
"eval_samples_per_second": 14.661,
|
| 417 |
+
"eval_steps_per_second": 3.812,
|
| 418 |
+
"step": 100
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 2.1,
|
| 422 |
+
"learning_rate": 1e-05,
|
| 423 |
+
"loss": 0.6252,
|
| 424 |
+
"step": 102
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.14,
|
| 428 |
+
"learning_rate": 1e-05,
|
| 429 |
+
"loss": 0.6065,
|
| 430 |
+
"step": 104
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 2.19,
|
| 434 |
+
"learning_rate": 1e-05,
|
| 435 |
+
"loss": 0.5829,
|
| 436 |
+
"step": 106
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 2.23,
|
| 440 |
+
"learning_rate": 1e-05,
|
| 441 |
+
"loss": 0.5936,
|
| 442 |
+
"step": 108
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 2.27,
|
| 446 |
+
"learning_rate": 1e-05,
|
| 447 |
+
"loss": 0.5528,
|
| 448 |
+
"step": 110
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 2.27,
|
| 452 |
+
"eval_accuracy": 0.8600000143051147,
|
| 453 |
+
"eval_average_score": -0.0412043035030365,
|
| 454 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 455 |
+
"eval_loss": 0.5438380241394043,
|
| 456 |
+
"eval_runtime": 3.4096,
|
| 457 |
+
"eval_samples_per_second": 14.665,
|
| 458 |
+
"eval_steps_per_second": 3.813,
|
| 459 |
+
"step": 110
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.31,
|
| 463 |
+
"learning_rate": 1e-05,
|
| 464 |
+
"loss": 0.5529,
|
| 465 |
+
"step": 112
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.35,
|
| 469 |
+
"learning_rate": 1e-05,
|
| 470 |
+
"loss": 0.5658,
|
| 471 |
+
"step": 114
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.39,
|
| 475 |
+
"learning_rate": 1e-05,
|
| 476 |
+
"loss": 0.5421,
|
| 477 |
+
"step": 116
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.43,
|
| 481 |
+
"learning_rate": 1e-05,
|
| 482 |
+
"loss": 0.5503,
|
| 483 |
+
"step": 118
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.47,
|
| 487 |
+
"learning_rate": 1e-05,
|
| 488 |
+
"loss": 0.4854,
|
| 489 |
+
"step": 120
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.47,
|
| 493 |
+
"eval_accuracy": 0.8999999761581421,
|
| 494 |
+
"eval_average_score": -0.40412548184394836,
|
| 495 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 496 |
+
"eval_loss": 0.46706727147102356,
|
| 497 |
+
"eval_runtime": 3.4091,
|
| 498 |
+
"eval_samples_per_second": 14.667,
|
| 499 |
+
"eval_steps_per_second": 3.813,
|
| 500 |
+
"step": 120
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 2.52,
|
| 504 |
+
"learning_rate": 1e-05,
|
| 505 |
+
"loss": 0.5268,
|
| 506 |
+
"step": 122
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.56,
|
| 510 |
+
"learning_rate": 1e-05,
|
| 511 |
+
"loss": 0.5251,
|
| 512 |
+
"step": 124
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.6,
|
| 516 |
+
"learning_rate": 1e-05,
|
| 517 |
+
"loss": 0.4857,
|
| 518 |
+
"step": 126
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.64,
|
| 522 |
+
"learning_rate": 1e-05,
|
| 523 |
+
"loss": 0.5307,
|
| 524 |
+
"step": 128
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.68,
|
| 528 |
+
"learning_rate": 1e-05,
|
| 529 |
+
"loss": 0.4672,
|
| 530 |
+
"step": 130
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 2.68,
|
| 534 |
+
"eval_accuracy": 0.8999999761581421,
|
| 535 |
+
"eval_average_score": -0.14681239426136017,
|
| 536 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 537 |
+
"eval_loss": 0.391941636800766,
|
| 538 |
+
"eval_runtime": 3.4094,
|
| 539 |
+
"eval_samples_per_second": 14.665,
|
| 540 |
+
"eval_steps_per_second": 3.813,
|
| 541 |
+
"step": 130
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.72,
|
| 545 |
+
"learning_rate": 1e-05,
|
| 546 |
+
"loss": 0.4077,
|
| 547 |
+
"step": 132
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 2.76,
|
| 551 |
+
"learning_rate": 1e-05,
|
| 552 |
+
"loss": 0.3849,
|
| 553 |
+
"step": 134
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.8,
|
| 557 |
+
"learning_rate": 1e-05,
|
| 558 |
+
"loss": 0.4358,
|
| 559 |
+
"step": 136
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.85,
|
| 563 |
+
"learning_rate": 1e-05,
|
| 564 |
+
"loss": 0.3896,
|
| 565 |
+
"step": 138
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 2.89,
|
| 569 |
+
"learning_rate": 1e-05,
|
| 570 |
+
"loss": 0.4903,
|
| 571 |
+
"step": 140
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.89,
|
| 575 |
+
"eval_accuracy": 0.8999999761581421,
|
| 576 |
+
"eval_average_score": -0.11637383699417114,
|
| 577 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 578 |
+
"eval_loss": 0.3501974940299988,
|
| 579 |
+
"eval_runtime": 3.4098,
|
| 580 |
+
"eval_samples_per_second": 14.664,
|
| 581 |
+
"eval_steps_per_second": 3.813,
|
| 582 |
+
"step": 140
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.93,
|
| 586 |
+
"learning_rate": 1e-05,
|
| 587 |
+
"loss": 0.3395,
|
| 588 |
+
"step": 142
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 2.97,
|
| 592 |
+
"learning_rate": 1e-05,
|
| 593 |
+
"loss": 0.3976,
|
| 594 |
+
"step": 144
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 3.01,
|
| 598 |
+
"learning_rate": 1e-05,
|
| 599 |
+
"loss": 0.3826,
|
| 600 |
+
"step": 146
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 3.05,
|
| 604 |
+
"learning_rate": 1e-05,
|
| 605 |
+
"loss": 0.4534,
|
| 606 |
+
"step": 148
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 3.09,
|
| 610 |
+
"learning_rate": 1e-05,
|
| 611 |
+
"loss": 0.3473,
|
| 612 |
+
"step": 150
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 3.09,
|
| 616 |
+
"eval_accuracy": 0.9200000166893005,
|
| 617 |
+
"eval_average_score": -0.3039152920246124,
|
| 618 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 619 |
+
"eval_loss": 0.3362675607204437,
|
| 620 |
+
"eval_runtime": 3.4088,
|
| 621 |
+
"eval_samples_per_second": 14.668,
|
| 622 |
+
"eval_steps_per_second": 3.814,
|
| 623 |
+
"step": 150
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 3.13,
|
| 627 |
+
"learning_rate": 1e-05,
|
| 628 |
+
"loss": 0.368,
|
| 629 |
+
"step": 152
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 3.18,
|
| 633 |
+
"learning_rate": 1e-05,
|
| 634 |
+
"loss": 0.4096,
|
| 635 |
+
"step": 154
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 3.22,
|
| 639 |
+
"learning_rate": 1e-05,
|
| 640 |
+
"loss": 0.3829,
|
| 641 |
+
"step": 156
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 3.26,
|
| 645 |
+
"learning_rate": 1e-05,
|
| 646 |
+
"loss": 0.389,
|
| 647 |
+
"step": 158
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 3.3,
|
| 651 |
+
"learning_rate": 1e-05,
|
| 652 |
+
"loss": 0.4191,
|
| 653 |
+
"step": 160
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 3.3,
|
| 657 |
+
"eval_accuracy": 0.8999999761581421,
|
| 658 |
+
"eval_average_score": 0.41642922163009644,
|
| 659 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 660 |
+
"eval_loss": 0.31328877806663513,
|
| 661 |
+
"eval_runtime": 3.415,
|
| 662 |
+
"eval_samples_per_second": 14.641,
|
| 663 |
+
"eval_steps_per_second": 3.807,
|
| 664 |
+
"step": 160
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 3.34,
|
| 668 |
+
"learning_rate": 1e-05,
|
| 669 |
+
"loss": 0.2698,
|
| 670 |
+
"step": 162
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 3.38,
|
| 674 |
+
"learning_rate": 1e-05,
|
| 675 |
+
"loss": 0.4149,
|
| 676 |
+
"step": 164
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.42,
|
| 680 |
+
"learning_rate": 1e-05,
|
| 681 |
+
"loss": 0.3844,
|
| 682 |
+
"step": 166
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 3.46,
|
| 686 |
+
"learning_rate": 1e-05,
|
| 687 |
+
"loss": 0.3549,
|
| 688 |
+
"step": 168
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 3.51,
|
| 692 |
+
"learning_rate": 1e-05,
|
| 693 |
+
"loss": 0.3571,
|
| 694 |
+
"step": 170
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 3.51,
|
| 698 |
+
"eval_accuracy": 0.8999999761581421,
|
| 699 |
+
"eval_average_score": -0.14143863320350647,
|
| 700 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 701 |
+
"eval_loss": 0.3141113221645355,
|
| 702 |
+
"eval_runtime": 3.4148,
|
| 703 |
+
"eval_samples_per_second": 14.642,
|
| 704 |
+
"eval_steps_per_second": 3.807,
|
| 705 |
+
"step": 170
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.55,
|
| 709 |
+
"learning_rate": 1e-05,
|
| 710 |
+
"loss": 0.2812,
|
| 711 |
+
"step": 172
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.59,
|
| 715 |
+
"learning_rate": 1e-05,
|
| 716 |
+
"loss": 0.4222,
|
| 717 |
+
"step": 174
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.63,
|
| 721 |
+
"learning_rate": 1e-05,
|
| 722 |
+
"loss": 0.3538,
|
| 723 |
+
"step": 176
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 3.67,
|
| 727 |
+
"learning_rate": 1e-05,
|
| 728 |
+
"loss": 0.2896,
|
| 729 |
+
"step": 178
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 3.71,
|
| 733 |
+
"learning_rate": 1e-05,
|
| 734 |
+
"loss": 0.3285,
|
| 735 |
+
"step": 180
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 3.71,
|
| 739 |
+
"eval_accuracy": 0.8999999761581421,
|
| 740 |
+
"eval_average_score": -0.04474750533699989,
|
| 741 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 742 |
+
"eval_loss": 0.30486491322517395,
|
| 743 |
+
"eval_runtime": 3.4082,
|
| 744 |
+
"eval_samples_per_second": 14.67,
|
| 745 |
+
"eval_steps_per_second": 3.814,
|
| 746 |
+
"step": 180
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 3.75,
|
| 750 |
+
"learning_rate": 1e-05,
|
| 751 |
+
"loss": 0.3278,
|
| 752 |
+
"step": 182
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 3.79,
|
| 756 |
+
"learning_rate": 1e-05,
|
| 757 |
+
"loss": 0.3715,
|
| 758 |
+
"step": 184
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 3.84,
|
| 762 |
+
"learning_rate": 1e-05,
|
| 763 |
+
"loss": 0.3234,
|
| 764 |
+
"step": 186
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 3.88,
|
| 768 |
+
"learning_rate": 1e-05,
|
| 769 |
+
"loss": 0.2756,
|
| 770 |
+
"step": 188
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 3.92,
|
| 774 |
+
"learning_rate": 1e-05,
|
| 775 |
+
"loss": 0.3535,
|
| 776 |
+
"step": 190
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 3.92,
|
| 780 |
+
"eval_accuracy": 0.8999999761581421,
|
| 781 |
+
"eval_average_score": -0.3523316979408264,
|
| 782 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 783 |
+
"eval_loss": 0.3148648738861084,
|
| 784 |
+
"eval_runtime": 3.4081,
|
| 785 |
+
"eval_samples_per_second": 14.671,
|
| 786 |
+
"eval_steps_per_second": 3.814,
|
| 787 |
+
"step": 190
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 3.96,
|
| 791 |
+
"learning_rate": 1e-05,
|
| 792 |
+
"loss": 0.3098,
|
| 793 |
+
"step": 192
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 4.0,
|
| 797 |
+
"learning_rate": 1e-05,
|
| 798 |
+
"loss": 0.2694,
|
| 799 |
+
"step": 194
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 4.04,
|
| 803 |
+
"learning_rate": 1e-05,
|
| 804 |
+
"loss": 0.2731,
|
| 805 |
+
"step": 196
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 4.08,
|
| 809 |
+
"learning_rate": 1e-05,
|
| 810 |
+
"loss": 0.309,
|
| 811 |
+
"step": 198
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 4.12,
|
| 815 |
+
"learning_rate": 1e-05,
|
| 816 |
+
"loss": 0.3961,
|
| 817 |
+
"step": 200
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 4.12,
|
| 821 |
+
"eval_accuracy": 0.8999999761581421,
|
| 822 |
+
"eval_average_score": 0.40198975801467896,
|
| 823 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 824 |
+
"eval_loss": 0.2921418249607086,
|
| 825 |
+
"eval_runtime": 3.4104,
|
| 826 |
+
"eval_samples_per_second": 14.661,
|
| 827 |
+
"eval_steps_per_second": 3.812,
|
| 828 |
+
"step": 200
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 4.16,
|
| 832 |
+
"learning_rate": 1e-05,
|
| 833 |
+
"loss": 0.3488,
|
| 834 |
+
"step": 202
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 4.21,
|
| 838 |
+
"learning_rate": 1e-05,
|
| 839 |
+
"loss": 0.399,
|
| 840 |
+
"step": 204
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 4.25,
|
| 844 |
+
"learning_rate": 1e-05,
|
| 845 |
+
"loss": 0.348,
|
| 846 |
+
"step": 206
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 4.29,
|
| 850 |
+
"learning_rate": 1e-05,
|
| 851 |
+
"loss": 0.2844,
|
| 852 |
+
"step": 208
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 4.33,
|
| 856 |
+
"learning_rate": 1e-05,
|
| 857 |
+
"loss": 0.3491,
|
| 858 |
+
"step": 210
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"epoch": 4.33,
|
| 862 |
+
"eval_accuracy": 0.8999999761581421,
|
| 863 |
+
"eval_average_score": 0.008343610912561417,
|
| 864 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 865 |
+
"eval_loss": 0.2992297410964966,
|
| 866 |
+
"eval_runtime": 3.4089,
|
| 867 |
+
"eval_samples_per_second": 14.667,
|
| 868 |
+
"eval_steps_per_second": 3.814,
|
| 869 |
+
"step": 210
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 4.37,
|
| 873 |
+
"learning_rate": 1e-05,
|
| 874 |
+
"loss": 0.2381,
|
| 875 |
+
"step": 212
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 4.41,
|
| 879 |
+
"learning_rate": 1e-05,
|
| 880 |
+
"loss": 0.3823,
|
| 881 |
+
"step": 214
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 4.45,
|
| 885 |
+
"learning_rate": 1e-05,
|
| 886 |
+
"loss": 0.2863,
|
| 887 |
+
"step": 216
|
| 888 |
+
},
|
| 889 |
+
{
|
| 890 |
+
"epoch": 4.49,
|
| 891 |
+
"learning_rate": 1e-05,
|
| 892 |
+
"loss": 0.2731,
|
| 893 |
+
"step": 218
|
| 894 |
+
},
|
| 895 |
+
{
|
| 896 |
+
"epoch": 4.54,
|
| 897 |
+
"learning_rate": 1e-05,
|
| 898 |
+
"loss": 0.2552,
|
| 899 |
+
"step": 220
|
| 900 |
+
},
|
| 901 |
+
{
|
| 902 |
+
"epoch": 4.54,
|
| 903 |
+
"eval_accuracy": 0.8999999761581421,
|
| 904 |
+
"eval_average_score": 0.21405518054962158,
|
| 905 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 906 |
+
"eval_loss": 0.28597450256347656,
|
| 907 |
+
"eval_runtime": 3.4094,
|
| 908 |
+
"eval_samples_per_second": 14.665,
|
| 909 |
+
"eval_steps_per_second": 3.813,
|
| 910 |
+
"step": 220
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 4.58,
|
| 914 |
+
"learning_rate": 1e-05,
|
| 915 |
+
"loss": 0.3438,
|
| 916 |
+
"step": 222
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 4.62,
|
| 920 |
+
"learning_rate": 1e-05,
|
| 921 |
+
"loss": 0.3672,
|
| 922 |
+
"step": 224
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 4.66,
|
| 926 |
+
"learning_rate": 1e-05,
|
| 927 |
+
"loss": 0.2783,
|
| 928 |
+
"step": 226
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 4.7,
|
| 932 |
+
"learning_rate": 1e-05,
|
| 933 |
+
"loss": 0.2705,
|
| 934 |
+
"step": 228
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 4.74,
|
| 938 |
+
"learning_rate": 1e-05,
|
| 939 |
+
"loss": 0.1839,
|
| 940 |
+
"step": 230
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 4.74,
|
| 944 |
+
"eval_accuracy": 0.9200000166893005,
|
| 945 |
+
"eval_average_score": -0.29983624815940857,
|
| 946 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 947 |
+
"eval_loss": 0.2997009754180908,
|
| 948 |
+
"eval_runtime": 3.409,
|
| 949 |
+
"eval_samples_per_second": 14.667,
|
| 950 |
+
"eval_steps_per_second": 3.813,
|
| 951 |
+
"step": 230
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 4.78,
|
| 955 |
+
"learning_rate": 1e-05,
|
| 956 |
+
"loss": 0.2776,
|
| 957 |
+
"step": 232
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 4.82,
|
| 961 |
+
"learning_rate": 1e-05,
|
| 962 |
+
"loss": 0.284,
|
| 963 |
+
"step": 234
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 4.87,
|
| 967 |
+
"learning_rate": 1e-05,
|
| 968 |
+
"loss": 0.3313,
|
| 969 |
+
"step": 236
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 4.91,
|
| 973 |
+
"learning_rate": 1e-05,
|
| 974 |
+
"loss": 0.2521,
|
| 975 |
+
"step": 238
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 4.95,
|
| 979 |
+
"learning_rate": 1e-05,
|
| 980 |
+
"loss": 0.2896,
|
| 981 |
+
"step": 240
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 4.95,
|
| 985 |
+
"eval_accuracy": 0.8999999761581421,
|
| 986 |
+
"eval_average_score": 0.08488038927316666,
|
| 987 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 988 |
+
"eval_loss": 0.2867457866668701,
|
| 989 |
+
"eval_runtime": 3.4098,
|
| 990 |
+
"eval_samples_per_second": 14.664,
|
| 991 |
+
"eval_steps_per_second": 3.813,
|
| 992 |
+
"step": 240
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 4.99,
|
| 996 |
+
"learning_rate": 1e-05,
|
| 997 |
+
"loss": 0.3284,
|
| 998 |
+
"step": 242
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 5.03,
|
| 1002 |
+
"learning_rate": 1e-05,
|
| 1003 |
+
"loss": 0.3234,
|
| 1004 |
+
"step": 244
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 5.07,
|
| 1008 |
+
"learning_rate": 1e-05,
|
| 1009 |
+
"loss": 0.2818,
|
| 1010 |
+
"step": 246
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 5.11,
|
| 1014 |
+
"learning_rate": 1e-05,
|
| 1015 |
+
"loss": 0.3356,
|
| 1016 |
+
"step": 248
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 5.15,
|
| 1020 |
+
"learning_rate": 1e-05,
|
| 1021 |
+
"loss": 0.3125,
|
| 1022 |
+
"step": 250
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 5.15,
|
| 1026 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1027 |
+
"eval_average_score": -0.0647800862789154,
|
| 1028 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1029 |
+
"eval_loss": 0.28697526454925537,
|
| 1030 |
+
"eval_runtime": 3.4101,
|
| 1031 |
+
"eval_samples_per_second": 14.663,
|
| 1032 |
+
"eval_steps_per_second": 3.812,
|
| 1033 |
+
"step": 250
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 5.2,
|
| 1037 |
+
"learning_rate": 1e-05,
|
| 1038 |
+
"loss": 0.2619,
|
| 1039 |
+
"step": 252
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 5.24,
|
| 1043 |
+
"learning_rate": 1e-05,
|
| 1044 |
+
"loss": 0.2484,
|
| 1045 |
+
"step": 254
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 5.28,
|
| 1049 |
+
"learning_rate": 1e-05,
|
| 1050 |
+
"loss": 0.3222,
|
| 1051 |
+
"step": 256
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 5.32,
|
| 1055 |
+
"learning_rate": 1e-05,
|
| 1056 |
+
"loss": 0.1985,
|
| 1057 |
+
"step": 258
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 5.36,
|
| 1061 |
+
"learning_rate": 1e-05,
|
| 1062 |
+
"loss": 0.3923,
|
| 1063 |
+
"step": 260
|
| 1064 |
+
},
|
| 1065 |
+
{
|
| 1066 |
+
"epoch": 5.36,
|
| 1067 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1068 |
+
"eval_average_score": 0.1731201559305191,
|
| 1069 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1070 |
+
"eval_loss": 0.2790464162826538,
|
| 1071 |
+
"eval_runtime": 3.4092,
|
| 1072 |
+
"eval_samples_per_second": 14.666,
|
| 1073 |
+
"eval_steps_per_second": 3.813,
|
| 1074 |
+
"step": 260
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 5.4,
|
| 1078 |
+
"learning_rate": 1e-05,
|
| 1079 |
+
"loss": 0.3787,
|
| 1080 |
+
"step": 262
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 5.44,
|
| 1084 |
+
"learning_rate": 1e-05,
|
| 1085 |
+
"loss": 0.2585,
|
| 1086 |
+
"step": 264
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 5.48,
|
| 1090 |
+
"learning_rate": 1e-05,
|
| 1091 |
+
"loss": 0.2569,
|
| 1092 |
+
"step": 266
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 5.53,
|
| 1096 |
+
"learning_rate": 1e-05,
|
| 1097 |
+
"loss": 0.2574,
|
| 1098 |
+
"step": 268
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 5.57,
|
| 1102 |
+
"learning_rate": 1e-05,
|
| 1103 |
+
"loss": 0.3044,
|
| 1104 |
+
"step": 270
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 5.57,
|
| 1108 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1109 |
+
"eval_average_score": 0.16195712983608246,
|
| 1110 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1111 |
+
"eval_loss": 0.28128278255462646,
|
| 1112 |
+
"eval_runtime": 3.4082,
|
| 1113 |
+
"eval_samples_per_second": 14.67,
|
| 1114 |
+
"eval_steps_per_second": 3.814,
|
| 1115 |
+
"step": 270
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 5.61,
|
| 1119 |
+
"learning_rate": 1e-05,
|
| 1120 |
+
"loss": 0.2866,
|
| 1121 |
+
"step": 272
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 5.65,
|
| 1125 |
+
"learning_rate": 1e-05,
|
| 1126 |
+
"loss": 0.2532,
|
| 1127 |
+
"step": 274
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 5.69,
|
| 1131 |
+
"learning_rate": 1e-05,
|
| 1132 |
+
"loss": 0.2748,
|
| 1133 |
+
"step": 276
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 5.73,
|
| 1137 |
+
"learning_rate": 1e-05,
|
| 1138 |
+
"loss": 0.3158,
|
| 1139 |
+
"step": 278
|
| 1140 |
+
},
|
| 1141 |
+
{
|
| 1142 |
+
"epoch": 5.77,
|
| 1143 |
+
"learning_rate": 1e-05,
|
| 1144 |
+
"loss": 0.3294,
|
| 1145 |
+
"step": 280
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 5.77,
|
| 1149 |
+
"eval_accuracy": 0.8999999761581421,
|
| 1150 |
+
"eval_average_score": 0.017967911437153816,
|
| 1151 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1152 |
+
"eval_loss": 0.280609667301178,
|
| 1153 |
+
"eval_runtime": 3.4082,
|
| 1154 |
+
"eval_samples_per_second": 14.67,
|
| 1155 |
+
"eval_steps_per_second": 3.814,
|
| 1156 |
+
"step": 280
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 5.81,
|
| 1160 |
+
"learning_rate": 1e-05,
|
| 1161 |
+
"loss": 0.2155,
|
| 1162 |
+
"step": 282
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 5.86,
|
| 1166 |
+
"learning_rate": 1e-05,
|
| 1167 |
+
"loss": 0.2301,
|
| 1168 |
+
"step": 284
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 5.9,
|
| 1172 |
+
"learning_rate": 1e-05,
|
| 1173 |
+
"loss": 0.3384,
|
| 1174 |
+
"step": 286
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 5.94,
|
| 1178 |
+
"learning_rate": 1e-05,
|
| 1179 |
+
"loss": 0.2534,
|
| 1180 |
+
"step": 288
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 5.98,
|
| 1184 |
+
"learning_rate": 1e-05,
|
| 1185 |
+
"loss": 0.2902,
|
| 1186 |
+
"step": 290
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 5.98,
|
| 1190 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1191 |
+
"eval_average_score": 0.3276750147342682,
|
| 1192 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1193 |
+
"eval_loss": 0.2765599191188812,
|
| 1194 |
+
"eval_runtime": 3.4103,
|
| 1195 |
+
"eval_samples_per_second": 14.661,
|
| 1196 |
+
"eval_steps_per_second": 3.812,
|
| 1197 |
+
"step": 290
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 6.02,
|
| 1201 |
+
"learning_rate": 1e-05,
|
| 1202 |
+
"loss": 0.3011,
|
| 1203 |
+
"step": 292
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 6.06,
|
| 1207 |
+
"learning_rate": 1e-05,
|
| 1208 |
+
"loss": 0.3248,
|
| 1209 |
+
"step": 294
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 6.1,
|
| 1213 |
+
"learning_rate": 1e-05,
|
| 1214 |
+
"loss": 0.37,
|
| 1215 |
+
"step": 296
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 6.14,
|
| 1219 |
+
"learning_rate": 1e-05,
|
| 1220 |
+
"loss": 0.278,
|
| 1221 |
+
"step": 298
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 6.19,
|
| 1225 |
+
"learning_rate": 1e-05,
|
| 1226 |
+
"loss": 0.275,
|
| 1227 |
+
"step": 300
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 6.19,
|
| 1231 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1232 |
+
"eval_average_score": -0.1893548220396042,
|
| 1233 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1234 |
+
"eval_loss": 0.2832108438014984,
|
| 1235 |
+
"eval_runtime": 3.4099,
|
| 1236 |
+
"eval_samples_per_second": 14.663,
|
| 1237 |
+
"eval_steps_per_second": 3.812,
|
| 1238 |
+
"step": 300
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 6.23,
|
| 1242 |
+
"learning_rate": 1e-05,
|
| 1243 |
+
"loss": 0.262,
|
| 1244 |
+
"step": 302
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 6.27,
|
| 1248 |
+
"learning_rate": 1e-05,
|
| 1249 |
+
"loss": 0.2765,
|
| 1250 |
+
"step": 304
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 6.31,
|
| 1254 |
+
"learning_rate": 1e-05,
|
| 1255 |
+
"loss": 0.2069,
|
| 1256 |
+
"step": 306
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 6.35,
|
| 1260 |
+
"learning_rate": 1e-05,
|
| 1261 |
+
"loss": 0.2567,
|
| 1262 |
+
"step": 308
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 6.39,
|
| 1266 |
+
"learning_rate": 1e-05,
|
| 1267 |
+
"loss": 0.2939,
|
| 1268 |
+
"step": 310
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 6.39,
|
| 1272 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1273 |
+
"eval_average_score": -0.16238823533058167,
|
| 1274 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1275 |
+
"eval_loss": 0.2794664204120636,
|
| 1276 |
+
"eval_runtime": 3.407,
|
| 1277 |
+
"eval_samples_per_second": 14.676,
|
| 1278 |
+
"eval_steps_per_second": 3.816,
|
| 1279 |
+
"step": 310
|
| 1280 |
+
},
|
| 1281 |
+
{
|
| 1282 |
+
"epoch": 6.43,
|
| 1283 |
+
"learning_rate": 1e-05,
|
| 1284 |
+
"loss": 0.3302,
|
| 1285 |
+
"step": 312
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 6.47,
|
| 1289 |
+
"learning_rate": 1e-05,
|
| 1290 |
+
"loss": 0.2244,
|
| 1291 |
+
"step": 314
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 6.52,
|
| 1295 |
+
"learning_rate": 1e-05,
|
| 1296 |
+
"loss": 0.2158,
|
| 1297 |
+
"step": 316
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 6.56,
|
| 1301 |
+
"learning_rate": 1e-05,
|
| 1302 |
+
"loss": 0.2536,
|
| 1303 |
+
"step": 318
|
| 1304 |
+
},
|
| 1305 |
+
{
|
| 1306 |
+
"epoch": 6.6,
|
| 1307 |
+
"learning_rate": 1e-05,
|
| 1308 |
+
"loss": 0.1905,
|
| 1309 |
+
"step": 320
|
| 1310 |
+
},
|
| 1311 |
+
{
|
| 1312 |
+
"epoch": 6.6,
|
| 1313 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1314 |
+
"eval_average_score": 0.26926523447036743,
|
| 1315 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1316 |
+
"eval_loss": 0.26797807216644287,
|
| 1317 |
+
"eval_runtime": 3.4064,
|
| 1318 |
+
"eval_samples_per_second": 14.678,
|
| 1319 |
+
"eval_steps_per_second": 3.816,
|
| 1320 |
+
"step": 320
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 6.64,
|
| 1324 |
+
"learning_rate": 1e-05,
|
| 1325 |
+
"loss": 0.2607,
|
| 1326 |
+
"step": 322
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 6.68,
|
| 1330 |
+
"learning_rate": 1e-05,
|
| 1331 |
+
"loss": 0.2542,
|
| 1332 |
+
"step": 324
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 6.72,
|
| 1336 |
+
"learning_rate": 1e-05,
|
| 1337 |
+
"loss": 0.265,
|
| 1338 |
+
"step": 326
|
| 1339 |
+
},
|
| 1340 |
+
{
|
| 1341 |
+
"epoch": 6.76,
|
| 1342 |
+
"learning_rate": 1e-05,
|
| 1343 |
+
"loss": 0.2763,
|
| 1344 |
+
"step": 328
|
| 1345 |
+
},
|
| 1346 |
+
{
|
| 1347 |
+
"epoch": 6.8,
|
| 1348 |
+
"learning_rate": 1e-05,
|
| 1349 |
+
"loss": 0.3119,
|
| 1350 |
+
"step": 330
|
| 1351 |
+
},
|
| 1352 |
+
{
|
| 1353 |
+
"epoch": 6.8,
|
| 1354 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1355 |
+
"eval_average_score": -0.12742717564105988,
|
| 1356 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1357 |
+
"eval_loss": 0.2834363877773285,
|
| 1358 |
+
"eval_runtime": 3.4081,
|
| 1359 |
+
"eval_samples_per_second": 14.671,
|
| 1360 |
+
"eval_steps_per_second": 3.814,
|
| 1361 |
+
"step": 330
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 6.85,
|
| 1365 |
+
"learning_rate": 1e-05,
|
| 1366 |
+
"loss": 0.1726,
|
| 1367 |
+
"step": 332
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 6.89,
|
| 1371 |
+
"learning_rate": 1e-05,
|
| 1372 |
+
"loss": 0.2636,
|
| 1373 |
+
"step": 334
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 6.93,
|
| 1377 |
+
"learning_rate": 1e-05,
|
| 1378 |
+
"loss": 0.3579,
|
| 1379 |
+
"step": 336
|
| 1380 |
+
},
|
| 1381 |
+
{
|
| 1382 |
+
"epoch": 6.97,
|
| 1383 |
+
"learning_rate": 1e-05,
|
| 1384 |
+
"loss": 0.2635,
|
| 1385 |
+
"step": 338
|
| 1386 |
+
},
|
| 1387 |
+
{
|
| 1388 |
+
"epoch": 7.01,
|
| 1389 |
+
"learning_rate": 1e-05,
|
| 1390 |
+
"loss": 0.2602,
|
| 1391 |
+
"step": 340
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 7.01,
|
| 1395 |
+
"eval_accuracy": 0.9399999976158142,
|
| 1396 |
+
"eval_average_score": -0.0478409007191658,
|
| 1397 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1398 |
+
"eval_loss": 0.27865225076675415,
|
| 1399 |
+
"eval_runtime": 3.4081,
|
| 1400 |
+
"eval_samples_per_second": 14.671,
|
| 1401 |
+
"eval_steps_per_second": 3.814,
|
| 1402 |
+
"step": 340
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 7.05,
|
| 1406 |
+
"learning_rate": 1e-05,
|
| 1407 |
+
"loss": 0.2111,
|
| 1408 |
+
"step": 342
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 7.09,
|
| 1412 |
+
"learning_rate": 1e-05,
|
| 1413 |
+
"loss": 0.2853,
|
| 1414 |
+
"step": 344
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 7.13,
|
| 1418 |
+
"learning_rate": 1e-05,
|
| 1419 |
+
"loss": 0.2853,
|
| 1420 |
+
"step": 346
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 7.18,
|
| 1424 |
+
"learning_rate": 1e-05,
|
| 1425 |
+
"loss": 0.3026,
|
| 1426 |
+
"step": 348
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 7.22,
|
| 1430 |
+
"learning_rate": 1e-05,
|
| 1431 |
+
"loss": 0.2446,
|
| 1432 |
+
"step": 350
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 7.22,
|
| 1436 |
+
"eval_accuracy": 0.9399999976158142,
|
| 1437 |
+
"eval_average_score": -0.1700994372367859,
|
| 1438 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1439 |
+
"eval_loss": 0.2747596800327301,
|
| 1440 |
+
"eval_runtime": 3.4061,
|
| 1441 |
+
"eval_samples_per_second": 14.679,
|
| 1442 |
+
"eval_steps_per_second": 3.817,
|
| 1443 |
+
"step": 350
|
| 1444 |
+
},
|
| 1445 |
+
{
|
| 1446 |
+
"epoch": 7.26,
|
| 1447 |
+
"learning_rate": 1e-05,
|
| 1448 |
+
"loss": 0.2469,
|
| 1449 |
+
"step": 352
|
| 1450 |
+
},
|
| 1451 |
+
{
|
| 1452 |
+
"epoch": 7.3,
|
| 1453 |
+
"learning_rate": 1e-05,
|
| 1454 |
+
"loss": 0.2372,
|
| 1455 |
+
"step": 354
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 7.34,
|
| 1459 |
+
"learning_rate": 1e-05,
|
| 1460 |
+
"loss": 0.2655,
|
| 1461 |
+
"step": 356
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 7.38,
|
| 1465 |
+
"learning_rate": 1e-05,
|
| 1466 |
+
"loss": 0.2661,
|
| 1467 |
+
"step": 358
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 7.42,
|
| 1471 |
+
"learning_rate": 1e-05,
|
| 1472 |
+
"loss": 0.2731,
|
| 1473 |
+
"step": 360
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 7.42,
|
| 1477 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1478 |
+
"eval_average_score": 0.01858169585466385,
|
| 1479 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1480 |
+
"eval_loss": 0.2665027379989624,
|
| 1481 |
+
"eval_runtime": 3.4061,
|
| 1482 |
+
"eval_samples_per_second": 14.68,
|
| 1483 |
+
"eval_steps_per_second": 3.817,
|
| 1484 |
+
"step": 360
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 7.46,
|
| 1488 |
+
"learning_rate": 1e-05,
|
| 1489 |
+
"loss": 0.1845,
|
| 1490 |
+
"step": 362
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 7.51,
|
| 1494 |
+
"learning_rate": 1e-05,
|
| 1495 |
+
"loss": 0.2881,
|
| 1496 |
+
"step": 364
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 7.55,
|
| 1500 |
+
"learning_rate": 1e-05,
|
| 1501 |
+
"loss": 0.2038,
|
| 1502 |
+
"step": 366
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 7.59,
|
| 1506 |
+
"learning_rate": 1e-05,
|
| 1507 |
+
"loss": 0.2111,
|
| 1508 |
+
"step": 368
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 7.63,
|
| 1512 |
+
"learning_rate": 1e-05,
|
| 1513 |
+
"loss": 0.2992,
|
| 1514 |
+
"step": 370
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 7.63,
|
| 1518 |
+
"eval_accuracy": 0.9399999976158142,
|
| 1519 |
+
"eval_average_score": 0.4874734878540039,
|
| 1520 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1521 |
+
"eval_loss": 0.2591661810874939,
|
| 1522 |
+
"eval_runtime": 3.4058,
|
| 1523 |
+
"eval_samples_per_second": 14.681,
|
| 1524 |
+
"eval_steps_per_second": 3.817,
|
| 1525 |
+
"step": 370
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 7.67,
|
| 1529 |
+
"learning_rate": 1e-05,
|
| 1530 |
+
"loss": 0.1714,
|
| 1531 |
+
"step": 372
|
| 1532 |
+
},
|
| 1533 |
+
{
|
| 1534 |
+
"epoch": 7.71,
|
| 1535 |
+
"learning_rate": 1e-05,
|
| 1536 |
+
"loss": 0.3,
|
| 1537 |
+
"step": 374
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 7.75,
|
| 1541 |
+
"learning_rate": 1e-05,
|
| 1542 |
+
"loss": 0.2276,
|
| 1543 |
+
"step": 376
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 7.79,
|
| 1547 |
+
"learning_rate": 1e-05,
|
| 1548 |
+
"loss": 0.3777,
|
| 1549 |
+
"step": 378
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 7.84,
|
| 1553 |
+
"learning_rate": 1e-05,
|
| 1554 |
+
"loss": 0.2173,
|
| 1555 |
+
"step": 380
|
| 1556 |
+
},
|
| 1557 |
+
{
|
| 1558 |
+
"epoch": 7.84,
|
| 1559 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1560 |
+
"eval_average_score": -0.07412362843751907,
|
| 1561 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1562 |
+
"eval_loss": 0.274651437997818,
|
| 1563 |
+
"eval_runtime": 3.4072,
|
| 1564 |
+
"eval_samples_per_second": 14.675,
|
| 1565 |
+
"eval_steps_per_second": 3.815,
|
| 1566 |
+
"step": 380
|
| 1567 |
+
},
|
| 1568 |
+
{
|
| 1569 |
+
"epoch": 7.88,
|
| 1570 |
+
"learning_rate": 1e-05,
|
| 1571 |
+
"loss": 0.2491,
|
| 1572 |
+
"step": 382
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 7.92,
|
| 1576 |
+
"learning_rate": 1e-05,
|
| 1577 |
+
"loss": 0.2425,
|
| 1578 |
+
"step": 384
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 7.96,
|
| 1582 |
+
"learning_rate": 1e-05,
|
| 1583 |
+
"loss": 0.2925,
|
| 1584 |
+
"step": 386
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 8.0,
|
| 1588 |
+
"learning_rate": 1e-05,
|
| 1589 |
+
"loss": 0.2723,
|
| 1590 |
+
"step": 388
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 8.04,
|
| 1594 |
+
"learning_rate": 1e-05,
|
| 1595 |
+
"loss": 0.3516,
|
| 1596 |
+
"step": 390
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 8.04,
|
| 1600 |
+
"eval_accuracy": 0.9399999976158142,
|
| 1601 |
+
"eval_average_score": 0.005836525000631809,
|
| 1602 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1603 |
+
"eval_loss": 0.26507002115249634,
|
| 1604 |
+
"eval_runtime": 3.4088,
|
| 1605 |
+
"eval_samples_per_second": 14.668,
|
| 1606 |
+
"eval_steps_per_second": 3.814,
|
| 1607 |
+
"step": 390
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 8.08,
|
| 1611 |
+
"learning_rate": 1e-05,
|
| 1612 |
+
"loss": 0.2519,
|
| 1613 |
+
"step": 392
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 8.12,
|
| 1617 |
+
"learning_rate": 1e-05,
|
| 1618 |
+
"loss": 0.2991,
|
| 1619 |
+
"step": 394
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 8.16,
|
| 1623 |
+
"learning_rate": 1e-05,
|
| 1624 |
+
"loss": 0.2234,
|
| 1625 |
+
"step": 396
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 8.21,
|
| 1629 |
+
"learning_rate": 1e-05,
|
| 1630 |
+
"loss": 0.2711,
|
| 1631 |
+
"step": 398
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 8.25,
|
| 1635 |
+
"learning_rate": 1e-05,
|
| 1636 |
+
"loss": 0.3236,
|
| 1637 |
+
"step": 400
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 8.25,
|
| 1641 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1642 |
+
"eval_average_score": 0.04313167557120323,
|
| 1643 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1644 |
+
"eval_loss": 0.25931546092033386,
|
| 1645 |
+
"eval_runtime": 3.4059,
|
| 1646 |
+
"eval_samples_per_second": 14.681,
|
| 1647 |
+
"eval_steps_per_second": 3.817,
|
| 1648 |
+
"step": 400
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 8.29,
|
| 1652 |
+
"learning_rate": 1e-05,
|
| 1653 |
+
"loss": 0.1869,
|
| 1654 |
+
"step": 402
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 8.33,
|
| 1658 |
+
"learning_rate": 1e-05,
|
| 1659 |
+
"loss": 0.2314,
|
| 1660 |
+
"step": 404
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 8.37,
|
| 1664 |
+
"learning_rate": 1e-05,
|
| 1665 |
+
"loss": 0.2405,
|
| 1666 |
+
"step": 406
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 8.41,
|
| 1670 |
+
"learning_rate": 1e-05,
|
| 1671 |
+
"loss": 0.1779,
|
| 1672 |
+
"step": 408
|
| 1673 |
+
},
|
| 1674 |
+
{
|
| 1675 |
+
"epoch": 8.45,
|
| 1676 |
+
"learning_rate": 1e-05,
|
| 1677 |
+
"loss": 0.2458,
|
| 1678 |
+
"step": 410
|
| 1679 |
+
},
|
| 1680 |
+
{
|
| 1681 |
+
"epoch": 8.45,
|
| 1682 |
+
"eval_accuracy": 0.8999999761581421,
|
| 1683 |
+
"eval_average_score": -0.21540480852127075,
|
| 1684 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1685 |
+
"eval_loss": 0.2680150270462036,
|
| 1686 |
+
"eval_runtime": 3.4085,
|
| 1687 |
+
"eval_samples_per_second": 14.669,
|
| 1688 |
+
"eval_steps_per_second": 3.814,
|
| 1689 |
+
"step": 410
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 8.49,
|
| 1693 |
+
"learning_rate": 1e-05,
|
| 1694 |
+
"loss": 0.1723,
|
| 1695 |
+
"step": 412
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 8.54,
|
| 1699 |
+
"learning_rate": 1e-05,
|
| 1700 |
+
"loss": 0.3219,
|
| 1701 |
+
"step": 414
|
| 1702 |
+
},
|
| 1703 |
+
{
|
| 1704 |
+
"epoch": 8.58,
|
| 1705 |
+
"learning_rate": 1e-05,
|
| 1706 |
+
"loss": 0.243,
|
| 1707 |
+
"step": 416
|
| 1708 |
+
},
|
| 1709 |
+
{
|
| 1710 |
+
"epoch": 8.62,
|
| 1711 |
+
"learning_rate": 1e-05,
|
| 1712 |
+
"loss": 0.2095,
|
| 1713 |
+
"step": 418
|
| 1714 |
+
},
|
| 1715 |
+
{
|
| 1716 |
+
"epoch": 8.66,
|
| 1717 |
+
"learning_rate": 1e-05,
|
| 1718 |
+
"loss": 0.1976,
|
| 1719 |
+
"step": 420
|
| 1720 |
+
},
|
| 1721 |
+
{
|
| 1722 |
+
"epoch": 8.66,
|
| 1723 |
+
"eval_accuracy": 0.8999999761581421,
|
| 1724 |
+
"eval_average_score": 0.17054100334644318,
|
| 1725 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1726 |
+
"eval_loss": 0.2620932459831238,
|
| 1727 |
+
"eval_runtime": 3.4133,
|
| 1728 |
+
"eval_samples_per_second": 14.648,
|
| 1729 |
+
"eval_steps_per_second": 3.809,
|
| 1730 |
+
"step": 420
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 8.7,
|
| 1734 |
+
"learning_rate": 1e-05,
|
| 1735 |
+
"loss": 0.1717,
|
| 1736 |
+
"step": 422
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 8.74,
|
| 1740 |
+
"learning_rate": 1e-05,
|
| 1741 |
+
"loss": 0.2759,
|
| 1742 |
+
"step": 424
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 8.78,
|
| 1746 |
+
"learning_rate": 1e-05,
|
| 1747 |
+
"loss": 0.3572,
|
| 1748 |
+
"step": 426
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 8.82,
|
| 1752 |
+
"learning_rate": 1e-05,
|
| 1753 |
+
"loss": 0.1688,
|
| 1754 |
+
"step": 428
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 8.87,
|
| 1758 |
+
"learning_rate": 1e-05,
|
| 1759 |
+
"loss": 0.2358,
|
| 1760 |
+
"step": 430
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 8.87,
|
| 1764 |
+
"eval_accuracy": 0.8999999761581421,
|
| 1765 |
+
"eval_average_score": 0.2345839887857437,
|
| 1766 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1767 |
+
"eval_loss": 0.2618144154548645,
|
| 1768 |
+
"eval_runtime": 3.406,
|
| 1769 |
+
"eval_samples_per_second": 14.68,
|
| 1770 |
+
"eval_steps_per_second": 3.817,
|
| 1771 |
+
"step": 430
|
| 1772 |
+
},
|
| 1773 |
+
{
|
| 1774 |
+
"epoch": 8.91,
|
| 1775 |
+
"learning_rate": 1e-05,
|
| 1776 |
+
"loss": 0.1975,
|
| 1777 |
+
"step": 432
|
| 1778 |
+
},
|
| 1779 |
+
{
|
| 1780 |
+
"epoch": 8.95,
|
| 1781 |
+
"learning_rate": 1e-05,
|
| 1782 |
+
"loss": 0.3316,
|
| 1783 |
+
"step": 434
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 8.99,
|
| 1787 |
+
"learning_rate": 1e-05,
|
| 1788 |
+
"loss": 0.1964,
|
| 1789 |
+
"step": 436
|
| 1790 |
+
},
|
| 1791 |
+
{
|
| 1792 |
+
"epoch": 9.03,
|
| 1793 |
+
"learning_rate": 1e-05,
|
| 1794 |
+
"loss": 0.2148,
|
| 1795 |
+
"step": 438
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 9.07,
|
| 1799 |
+
"learning_rate": 1e-05,
|
| 1800 |
+
"loss": 0.1855,
|
| 1801 |
+
"step": 440
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 9.07,
|
| 1805 |
+
"eval_accuracy": 0.8999999761581421,
|
| 1806 |
+
"eval_average_score": 0.04021953418850899,
|
| 1807 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1808 |
+
"eval_loss": 0.2684229612350464,
|
| 1809 |
+
"eval_runtime": 3.4104,
|
| 1810 |
+
"eval_samples_per_second": 14.661,
|
| 1811 |
+
"eval_steps_per_second": 3.812,
|
| 1812 |
+
"step": 440
|
| 1813 |
+
},
|
| 1814 |
+
{
|
| 1815 |
+
"epoch": 9.11,
|
| 1816 |
+
"learning_rate": 1e-05,
|
| 1817 |
+
"loss": 0.2186,
|
| 1818 |
+
"step": 442
|
| 1819 |
+
},
|
| 1820 |
+
{
|
| 1821 |
+
"epoch": 9.15,
|
| 1822 |
+
"learning_rate": 1e-05,
|
| 1823 |
+
"loss": 0.1604,
|
| 1824 |
+
"step": 444
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 9.2,
|
| 1828 |
+
"learning_rate": 1e-05,
|
| 1829 |
+
"loss": 0.1649,
|
| 1830 |
+
"step": 446
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 9.24,
|
| 1834 |
+
"learning_rate": 1e-05,
|
| 1835 |
+
"loss": 0.2496,
|
| 1836 |
+
"step": 448
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 9.28,
|
| 1840 |
+
"learning_rate": 1e-05,
|
| 1841 |
+
"loss": 0.3004,
|
| 1842 |
+
"step": 450
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 9.28,
|
| 1846 |
+
"eval_accuracy": 0.8999999761581421,
|
| 1847 |
+
"eval_average_score": 0.08485334366559982,
|
| 1848 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1849 |
+
"eval_loss": 0.26888754963874817,
|
| 1850 |
+
"eval_runtime": 3.4093,
|
| 1851 |
+
"eval_samples_per_second": 14.666,
|
| 1852 |
+
"eval_steps_per_second": 3.813,
|
| 1853 |
+
"step": 450
|
| 1854 |
+
},
|
| 1855 |
+
{
|
| 1856 |
+
"epoch": 9.32,
|
| 1857 |
+
"learning_rate": 1e-05,
|
| 1858 |
+
"loss": 0.2056,
|
| 1859 |
+
"step": 452
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 9.36,
|
| 1863 |
+
"learning_rate": 1e-05,
|
| 1864 |
+
"loss": 0.2284,
|
| 1865 |
+
"step": 454
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 9.4,
|
| 1869 |
+
"learning_rate": 1e-05,
|
| 1870 |
+
"loss": 0.2529,
|
| 1871 |
+
"step": 456
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 9.44,
|
| 1875 |
+
"learning_rate": 1e-05,
|
| 1876 |
+
"loss": 0.2847,
|
| 1877 |
+
"step": 458
|
| 1878 |
+
},
|
| 1879 |
+
{
|
| 1880 |
+
"epoch": 9.48,
|
| 1881 |
+
"learning_rate": 1e-05,
|
| 1882 |
+
"loss": 0.2804,
|
| 1883 |
+
"step": 460
|
| 1884 |
+
},
|
| 1885 |
+
{
|
| 1886 |
+
"epoch": 9.48,
|
| 1887 |
+
"eval_accuracy": 0.9200000166893005,
|
| 1888 |
+
"eval_average_score": -0.19331374764442444,
|
| 1889 |
+
"eval_label_positive_rate": 0.47999998927116394,
|
| 1890 |
+
"eval_loss": 0.2731383442878723,
|
| 1891 |
+
"eval_runtime": 3.4089,
|
| 1892 |
+
"eval_samples_per_second": 14.668,
|
| 1893 |
+
"eval_steps_per_second": 3.814,
|
| 1894 |
+
"step": 460
|
| 1895 |
+
}
|
| 1896 |
+
],
|
| 1897 |
+
"max_steps": 480,
|
| 1898 |
+
"num_train_epochs": 10,
|
| 1899 |
+
"total_flos": 8.790703793881743e+17,
|
| 1900 |
+
"trial_name": null,
|
| 1901 |
+
"trial_params": null
|
| 1902 |
+
}
|
RM-alfworld/checkpoint-460/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:564182d5169ca566b583554da00a2543eaff9147c109688d8ca6f76822de599e
|
| 3 |
+
size 4795
|
RM-game24/checkpoint-1500/adapter_model/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.4.0
|
RM-game24/checkpoint-1500/adapter_model/lora_default/adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "/nobackup/users/zfchen/cdl/LLaVA-RLHF/VILA1.5-3b/llm",
|
| 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": 0.0,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 64,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"k_proj",
|
| 19 |
+
"v_proj",
|
| 20 |
+
"o_proj",
|
| 21 |
+
"gate_proj",
|
| 22 |
+
"up_proj",
|
| 23 |
+
"down_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
RM-game24/checkpoint-1500/adapter_model/lora_default/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1a0f8cb20ebef4180afa57c9a2050c04f2463843ee165a8628f588cb328c07a
|
| 3 |
+
size 224393877
|
RM-game24/checkpoint-1500/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RewardModel"
|
| 4 |
+
],
|
| 5 |
+
"backbone_model_name_or_path": "/nobackup/users/zfchen/cdl/LLaVA-RLHF/VILA1.5-3b/llm",
|
| 6 |
+
"mm_use_im_start_end": false,
|
| 7 |
+
"model_type": "reward_model",
|
| 8 |
+
"torch_dtype": "bfloat16",
|
| 9 |
+
"transformers_version": "4.30.2"
|
| 10 |
+
}
|
RM-game24/checkpoint-1500/reward_head
ADDED
|
Binary file (11.3 kB). View file
|
|
|
RM-game24/checkpoint-1500/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c12c3d61bd1047f32032390ba429811e5970389348951a181848cf08492da65b
|
| 3 |
+
size 17655
|
RM-game24/checkpoint-1500/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:153d31109edaf107200441dc4fee4aafb47a47f77c4f0d77e620dcc18681e57f
|
| 3 |
+
size 17655
|
RM-game24/checkpoint-1500/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:234eab2363fb4cce2ecb5bc297368e9a2d10a63ba2337679e7e4dc240b78a45f
|
| 3 |
+
size 17655
|
RM-game24/checkpoint-1500/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9508ca8daf95d7e82371151ebcfdcbb5383b78450086fe24a911c5e3e3a6cfe2
|
| 3 |
+
size 17655
|
RM-game24/checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7faa563a95139d32825b4d758cea69277ce77204a6269f6119d485f6d58426a4
|
| 3 |
+
size 627
|
RM-game24/checkpoint-1500/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": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
RM-game24/checkpoint-1500/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7aedb3582ecda9fa99ee9242c17a9658f6744db083ee6ebdc8fb14857f84d220
|
| 3 |
+
size 499723
|
RM-game24/checkpoint-1500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": {
|
| 32 |
+
"__type": "AddedToken",
|
| 33 |
+
"content": "<s>",
|
| 34 |
+
"lstrip": false,
|
| 35 |
+
"normalized": true,
|
| 36 |
+
"rstrip": false,
|
| 37 |
+
"single_word": false
|
| 38 |
+
},
|
| 39 |
+
"clean_up_tokenization_spaces": false,
|
| 40 |
+
"eos_token": {
|
| 41 |
+
"__type": "AddedToken",
|
| 42 |
+
"content": "</s>",
|
| 43 |
+
"lstrip": false,
|
| 44 |
+
"normalized": true,
|
| 45 |
+
"rstrip": false,
|
| 46 |
+
"single_word": false
|
| 47 |
+
},
|
| 48 |
+
"legacy": false,
|
| 49 |
+
"model_max_length": 4000,
|
| 50 |
+
"pad_token": {
|
| 51 |
+
"__type": "AddedToken",
|
| 52 |
+
"content": "<unk>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": true,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false
|
| 57 |
+
},
|
| 58 |
+
"padding_side": "left",
|
| 59 |
+
"sp_model_kwargs": {},
|
| 60 |
+
"spaces_between_special_tokens": false,
|
| 61 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 62 |
+
"truncation_side": "right",
|
| 63 |
+
"unk_token": {
|
| 64 |
+
"__type": "AddedToken",
|
| 65 |
+
"content": "<unk>",
|
| 66 |
+
"lstrip": false,
|
| 67 |
+
"normalized": true,
|
| 68 |
+
"rstrip": false,
|
| 69 |
+
"single_word": false
|
| 70 |
+
},
|
| 71 |
+
"use_default_system_prompt": false
|
| 72 |
+
}
|
RM-game24/checkpoint-1500/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
RM-game24/checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92b14e35676490aec5d4400f6f3ba19bbeacd9e1ebaa851d9b191e8fe7e395b0
|
| 3 |
+
size 4795
|
RM-sciworld/checkpoint-120/adapter_model/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.4.0
|
RM-sciworld/checkpoint-120/adapter_model/lora_default/adapter_config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "/nobackup/users/zfchen/cdl/LLaVA-RLHF/VILA1.5-3b/llm",
|
| 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": 0.0,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 64,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"k_proj",
|
| 19 |
+
"v_proj",
|
| 20 |
+
"o_proj",
|
| 21 |
+
"gate_proj",
|
| 22 |
+
"up_proj",
|
| 23 |
+
"down_proj"
|
| 24 |
+
],
|
| 25 |
+
"task_type": "CAUSAL_LM"
|
| 26 |
+
}
|
RM-sciworld/checkpoint-120/adapter_model/lora_default/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:091c4020c80e9d9fc315c44e8bf529be50c3430db06b6f8c68428f5a7c3ada74
|
| 3 |
+
size 224393877
|
RM-sciworld/checkpoint-120/config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RewardModel"
|
| 4 |
+
],
|
| 5 |
+
"backbone_model_name_or_path": "/nobackup/users/zfchen/cdl/LLaVA-RLHF/VILA1.5-3b/llm",
|
| 6 |
+
"mm_use_im_start_end": false,
|
| 7 |
+
"model_type": "reward_model",
|
| 8 |
+
"torch_dtype": "bfloat16",
|
| 9 |
+
"transformers_version": "4.30.2"
|
| 10 |
+
}
|
RM-sciworld/checkpoint-120/reward_head
ADDED
|
Binary file (11.3 kB). View file
|
|
|
RM-sciworld/checkpoint-120/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9108380eb8e76c1064593403bcea3798532ffc999f60aa83d77f1986c3111852
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:23ccbfcfaa29f8718318d2ca568c15317eeb460c340c72d82fecc183d56d17ba
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02dd6b8354207613ffddba9ad82b40cd300d39a17a5b5bc7461d9a0e9dbbfb1e
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:28db48b3220dbbb391220d436cc3aa0a9dc31b7f44552c0ab3cb87b13905d62a
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4b76df961261b4c1808f6de8fd12ddfb9adaa71836e4192a0e5e0588b4d5011f
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fcaabcf235f388013089a0731f3185918b9bcf3efc3be8cafd3048d57e5229e2
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dc00b4d49ee413bc73f995a9ce3c0ecd7fd6c1628c392094668f46d0635029d8
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b8b14a8a29425691a14fc71d787608a6554eded1f16c908c18ada686cc97ec42
|
| 3 |
+
size 21687
|
RM-sciworld/checkpoint-120/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f735eb293f032dfeb9e458f35501d617ae5cf159a29a9b6ac465d72d0667b44e
|
| 3 |
+
size 627
|
RM-sciworld/checkpoint-120/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": "<unk>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
RM-sciworld/checkpoint-120/tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7aedb3582ecda9fa99ee9242c17a9658f6744db083ee6ebdc8fb14857f84d220
|
| 3 |
+
size 499723
|
RM-sciworld/checkpoint-120/tokenizer_config.json
ADDED
|
@@ -0,0 +1,72 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": true,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": {
|
| 32 |
+
"__type": "AddedToken",
|
| 33 |
+
"content": "<s>",
|
| 34 |
+
"lstrip": false,
|
| 35 |
+
"normalized": true,
|
| 36 |
+
"rstrip": false,
|
| 37 |
+
"single_word": false
|
| 38 |
+
},
|
| 39 |
+
"clean_up_tokenization_spaces": false,
|
| 40 |
+
"eos_token": {
|
| 41 |
+
"__type": "AddedToken",
|
| 42 |
+
"content": "</s>",
|
| 43 |
+
"lstrip": false,
|
| 44 |
+
"normalized": true,
|
| 45 |
+
"rstrip": false,
|
| 46 |
+
"single_word": false
|
| 47 |
+
},
|
| 48 |
+
"legacy": false,
|
| 49 |
+
"model_max_length": 4096,
|
| 50 |
+
"pad_token": {
|
| 51 |
+
"__type": "AddedToken",
|
| 52 |
+
"content": "<unk>",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": true,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false
|
| 57 |
+
},
|
| 58 |
+
"padding_side": "left",
|
| 59 |
+
"sp_model_kwargs": {},
|
| 60 |
+
"spaces_between_special_tokens": false,
|
| 61 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 62 |
+
"truncation_side": "right",
|
| 63 |
+
"unk_token": {
|
| 64 |
+
"__type": "AddedToken",
|
| 65 |
+
"content": "<unk>",
|
| 66 |
+
"lstrip": false,
|
| 67 |
+
"normalized": true,
|
| 68 |
+
"rstrip": false,
|
| 69 |
+
"single_word": false
|
| 70 |
+
},
|
| 71 |
+
"use_default_system_prompt": false
|
| 72 |
+
}
|
RM-sciworld/checkpoint-120/trainer_state.json
ADDED
|
@@ -0,0 +1,442 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.304932735426009,
|
| 5 |
+
"global_step": 120,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.07,
|
| 12 |
+
"learning_rate": 3.7037037037037036e-07,
|
| 13 |
+
"loss": 0.6848,
|
| 14 |
+
"step": 2
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"learning_rate": 7.407407407407407e-07,
|
| 19 |
+
"loss": 0.6805,
|
| 20 |
+
"step": 4
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.22,
|
| 24 |
+
"learning_rate": 1.111111111111111e-06,
|
| 25 |
+
"loss": 0.6842,
|
| 26 |
+
"step": 6
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.29,
|
| 30 |
+
"learning_rate": 1.4814814814814815e-06,
|
| 31 |
+
"loss": 0.6893,
|
| 32 |
+
"step": 8
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.36,
|
| 36 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 37 |
+
"loss": 0.6787,
|
| 38 |
+
"step": 10
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.43,
|
| 42 |
+
"learning_rate": 2.222222222222222e-06,
|
| 43 |
+
"loss": 0.6789,
|
| 44 |
+
"step": 12
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5,
|
| 48 |
+
"learning_rate": 2.5925925925925925e-06,
|
| 49 |
+
"loss": 0.685,
|
| 50 |
+
"step": 14
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.57,
|
| 54 |
+
"learning_rate": 2.962962962962963e-06,
|
| 55 |
+
"loss": 0.6823,
|
| 56 |
+
"step": 16
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.65,
|
| 60 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 61 |
+
"loss": 0.6836,
|
| 62 |
+
"step": 18
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.72,
|
| 66 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 67 |
+
"loss": 0.6756,
|
| 68 |
+
"step": 20
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.72,
|
| 72 |
+
"eval_accuracy": 0.5419999957084656,
|
| 73 |
+
"eval_average_score": 0.5289427638053894,
|
| 74 |
+
"eval_label_positive_rate": 0.47600001096725464,
|
| 75 |
+
"eval_loss": 0.6793817281723022,
|
| 76 |
+
"eval_runtime": 8.697,
|
| 77 |
+
"eval_samples_per_second": 57.491,
|
| 78 |
+
"eval_steps_per_second": 7.244,
|
| 79 |
+
"step": 20
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.79,
|
| 83 |
+
"learning_rate": 4.074074074074074e-06,
|
| 84 |
+
"loss": 0.6845,
|
| 85 |
+
"step": 22
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.86,
|
| 89 |
+
"learning_rate": 4.444444444444444e-06,
|
| 90 |
+
"loss": 0.6837,
|
| 91 |
+
"step": 24
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.93,
|
| 95 |
+
"learning_rate": 4.814814814814815e-06,
|
| 96 |
+
"loss": 0.6743,
|
| 97 |
+
"step": 26
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 1.0,
|
| 101 |
+
"learning_rate": 5.185185185185185e-06,
|
| 102 |
+
"loss": 0.6718,
|
| 103 |
+
"step": 28
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 1.08,
|
| 107 |
+
"learning_rate": 5.555555555555557e-06,
|
| 108 |
+
"loss": 0.6777,
|
| 109 |
+
"step": 30
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.15,
|
| 113 |
+
"learning_rate": 5.925925925925926e-06,
|
| 114 |
+
"loss": 0.6702,
|
| 115 |
+
"step": 32
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.22,
|
| 119 |
+
"learning_rate": 6.296296296296297e-06,
|
| 120 |
+
"loss": 0.6743,
|
| 121 |
+
"step": 34
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 1.29,
|
| 125 |
+
"learning_rate": 6.666666666666667e-06,
|
| 126 |
+
"loss": 0.6722,
|
| 127 |
+
"step": 36
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 1.36,
|
| 131 |
+
"learning_rate": 7.0370370370370375e-06,
|
| 132 |
+
"loss": 0.6697,
|
| 133 |
+
"step": 38
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 1.43,
|
| 137 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 138 |
+
"loss": 0.6654,
|
| 139 |
+
"step": 40
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 1.43,
|
| 143 |
+
"eval_accuracy": 0.7080000042915344,
|
| 144 |
+
"eval_average_score": 0.4319504201412201,
|
| 145 |
+
"eval_label_positive_rate": 0.47600001096725464,
|
| 146 |
+
"eval_loss": 0.6608079075813293,
|
| 147 |
+
"eval_runtime": 8.6968,
|
| 148 |
+
"eval_samples_per_second": 57.492,
|
| 149 |
+
"eval_steps_per_second": 7.244,
|
| 150 |
+
"step": 40
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 1.51,
|
| 154 |
+
"learning_rate": 7.77777777777778e-06,
|
| 155 |
+
"loss": 0.6572,
|
| 156 |
+
"step": 42
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.58,
|
| 160 |
+
"learning_rate": 8.148148148148148e-06,
|
| 161 |
+
"loss": 0.6621,
|
| 162 |
+
"step": 44
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 1.65,
|
| 166 |
+
"learning_rate": 8.518518518518519e-06,
|
| 167 |
+
"loss": 0.6575,
|
| 168 |
+
"step": 46
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 1.72,
|
| 172 |
+
"learning_rate": 8.888888888888888e-06,
|
| 173 |
+
"loss": 0.6455,
|
| 174 |
+
"step": 48
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 1.79,
|
| 178 |
+
"learning_rate": 9.25925925925926e-06,
|
| 179 |
+
"loss": 0.6367,
|
| 180 |
+
"step": 50
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 1.87,
|
| 184 |
+
"learning_rate": 9.62962962962963e-06,
|
| 185 |
+
"loss": 0.64,
|
| 186 |
+
"step": 52
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 1.94,
|
| 190 |
+
"learning_rate": 1e-05,
|
| 191 |
+
"loss": 0.6316,
|
| 192 |
+
"step": 54
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 2.01,
|
| 196 |
+
"learning_rate": 1e-05,
|
| 197 |
+
"loss": 0.6121,
|
| 198 |
+
"step": 56
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 2.08,
|
| 202 |
+
"learning_rate": 1e-05,
|
| 203 |
+
"loss": 0.6104,
|
| 204 |
+
"step": 58
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 2.15,
|
| 208 |
+
"learning_rate": 1e-05,
|
| 209 |
+
"loss": 0.5916,
|
| 210 |
+
"step": 60
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 2.15,
|
| 214 |
+
"eval_accuracy": 0.9419999718666077,
|
| 215 |
+
"eval_average_score": 0.13356663286685944,
|
| 216 |
+
"eval_label_positive_rate": 0.47600001096725464,
|
| 217 |
+
"eval_loss": 0.5838245153427124,
|
| 218 |
+
"eval_runtime": 8.6962,
|
| 219 |
+
"eval_samples_per_second": 57.496,
|
| 220 |
+
"eval_steps_per_second": 7.245,
|
| 221 |
+
"step": 60
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 2.22,
|
| 225 |
+
"learning_rate": 1e-05,
|
| 226 |
+
"loss": 0.5801,
|
| 227 |
+
"step": 62
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 2.3,
|
| 231 |
+
"learning_rate": 1e-05,
|
| 232 |
+
"loss": 0.5677,
|
| 233 |
+
"step": 64
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.37,
|
| 237 |
+
"learning_rate": 1e-05,
|
| 238 |
+
"loss": 0.5472,
|
| 239 |
+
"step": 66
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.44,
|
| 243 |
+
"learning_rate": 1e-05,
|
| 244 |
+
"loss": 0.5267,
|
| 245 |
+
"step": 68
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 2.51,
|
| 249 |
+
"learning_rate": 1e-05,
|
| 250 |
+
"loss": 0.5224,
|
| 251 |
+
"step": 70
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.58,
|
| 255 |
+
"learning_rate": 1e-05,
|
| 256 |
+
"loss": 0.49,
|
| 257 |
+
"step": 72
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 2.65,
|
| 261 |
+
"learning_rate": 1e-05,
|
| 262 |
+
"loss": 0.4607,
|
| 263 |
+
"step": 74
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.73,
|
| 267 |
+
"learning_rate": 1e-05,
|
| 268 |
+
"loss": 0.4295,
|
| 269 |
+
"step": 76
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.8,
|
| 273 |
+
"learning_rate": 1e-05,
|
| 274 |
+
"loss": 0.393,
|
| 275 |
+
"step": 78
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 2.87,
|
| 279 |
+
"learning_rate": 1e-05,
|
| 280 |
+
"loss": 0.3463,
|
| 281 |
+
"step": 80
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.87,
|
| 285 |
+
"eval_accuracy": 0.9800000190734863,
|
| 286 |
+
"eval_average_score": -0.13676369190216064,
|
| 287 |
+
"eval_label_positive_rate": 0.47600001096725464,
|
| 288 |
+
"eval_loss": 0.30906087160110474,
|
| 289 |
+
"eval_runtime": 8.6996,
|
| 290 |
+
"eval_samples_per_second": 57.474,
|
| 291 |
+
"eval_steps_per_second": 7.242,
|
| 292 |
+
"step": 80
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 2.94,
|
| 296 |
+
"learning_rate": 1e-05,
|
| 297 |
+
"loss": 0.3167,
|
| 298 |
+
"step": 82
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 3.01,
|
| 302 |
+
"learning_rate": 1e-05,
|
| 303 |
+
"loss": 0.2559,
|
| 304 |
+
"step": 84
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 3.09,
|
| 308 |
+
"learning_rate": 1e-05,
|
| 309 |
+
"loss": 0.2119,
|
| 310 |
+
"step": 86
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 3.16,
|
| 314 |
+
"learning_rate": 1e-05,
|
| 315 |
+
"loss": 0.1874,
|
| 316 |
+
"step": 88
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.23,
|
| 320 |
+
"learning_rate": 1e-05,
|
| 321 |
+
"loss": 0.1407,
|
| 322 |
+
"step": 90
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 3.3,
|
| 326 |
+
"learning_rate": 1e-05,
|
| 327 |
+
"loss": 0.1037,
|
| 328 |
+
"step": 92
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.37,
|
| 332 |
+
"learning_rate": 1e-05,
|
| 333 |
+
"loss": 0.0717,
|
| 334 |
+
"step": 94
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 3.44,
|
| 338 |
+
"learning_rate": 1e-05,
|
| 339 |
+
"loss": 0.0692,
|
| 340 |
+
"step": 96
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 3.52,
|
| 344 |
+
"learning_rate": 1e-05,
|
| 345 |
+
"loss": 0.0708,
|
| 346 |
+
"step": 98
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 3.59,
|
| 350 |
+
"learning_rate": 1e-05,
|
| 351 |
+
"loss": 0.0426,
|
| 352 |
+
"step": 100
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 3.59,
|
| 356 |
+
"eval_accuracy": 0.9879999756813049,
|
| 357 |
+
"eval_average_score": -0.7325789332389832,
|
| 358 |
+
"eval_label_positive_rate": 0.47600001096725464,
|
| 359 |
+
"eval_loss": 0.03151561692357063,
|
| 360 |
+
"eval_runtime": 8.708,
|
| 361 |
+
"eval_samples_per_second": 57.418,
|
| 362 |
+
"eval_steps_per_second": 7.235,
|
| 363 |
+
"step": 100
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 3.66,
|
| 367 |
+
"learning_rate": 1e-05,
|
| 368 |
+
"loss": 0.0315,
|
| 369 |
+
"step": 102
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 3.73,
|
| 373 |
+
"learning_rate": 1e-05,
|
| 374 |
+
"loss": 0.0227,
|
| 375 |
+
"step": 104
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 3.8,
|
| 379 |
+
"learning_rate": 1e-05,
|
| 380 |
+
"loss": 0.0272,
|
| 381 |
+
"step": 106
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 3.87,
|
| 385 |
+
"learning_rate": 1e-05,
|
| 386 |
+
"loss": 0.0203,
|
| 387 |
+
"step": 108
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 3.95,
|
| 391 |
+
"learning_rate": 1e-05,
|
| 392 |
+
"loss": 0.0151,
|
| 393 |
+
"step": 110
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 4.02,
|
| 397 |
+
"learning_rate": 1e-05,
|
| 398 |
+
"loss": 0.0199,
|
| 399 |
+
"step": 112
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.09,
|
| 403 |
+
"learning_rate": 1e-05,
|
| 404 |
+
"loss": 0.0171,
|
| 405 |
+
"step": 114
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 4.16,
|
| 409 |
+
"learning_rate": 1e-05,
|
| 410 |
+
"loss": 0.0094,
|
| 411 |
+
"step": 116
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 4.23,
|
| 415 |
+
"learning_rate": 1e-05,
|
| 416 |
+
"loss": 0.0166,
|
| 417 |
+
"step": 118
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.3,
|
| 421 |
+
"learning_rate": 1e-05,
|
| 422 |
+
"loss": 0.0079,
|
| 423 |
+
"step": 120
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.3,
|
| 427 |
+
"eval_accuracy": 1.0,
|
| 428 |
+
"eval_average_score": -1.2080471515655518,
|
| 429 |
+
"eval_label_positive_rate": 0.47600001096725464,
|
| 430 |
+
"eval_loss": 0.009560000151395798,
|
| 431 |
+
"eval_runtime": 8.7049,
|
| 432 |
+
"eval_samples_per_second": 57.439,
|
| 433 |
+
"eval_steps_per_second": 7.237,
|
| 434 |
+
"step": 120
|
| 435 |
+
}
|
| 436 |
+
],
|
| 437 |
+
"max_steps": 270,
|
| 438 |
+
"num_train_epochs": 10,
|
| 439 |
+
"total_flos": 3.914291024788193e+17,
|
| 440 |
+
"trial_name": null,
|
| 441 |
+
"trial_params": null
|
| 442 |
+
}
|
RM-sciworld/checkpoint-120/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57b146ab149226a1406ae4dbc206a2fc5340b734501c6a044f3f0daf846ec023
|
| 3 |
+
size 4795
|
RM-webshop/checkpoint-1100/adapter_model/README.md
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
### Framework versions
|
| 7 |
+
|
| 8 |
+
|
| 9 |
+
- PEFT 0.4.0
|