Commit
·
c102729
1
Parent(s):
3f7e522
commit from root
Browse filesThis view is limited to 50 files because it contains too many changes.
See raw diff
- README.md +20 -0
- adapter_config.json +21 -0
- adapter_model.bin +3 -0
- all_results.json +7 -0
- checkpoint-1000/README.md +20 -0
- checkpoint-1000/adapter_config.json +21 -0
- checkpoint-1000/adapter_model.bin +3 -0
- checkpoint-1000/finetuning_args.json +16 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state_0.pth +3 -0
- checkpoint-1000/rng_state_1.pth +3 -0
- checkpoint-1000/rng_state_2.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/trainer_state.json +616 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-2000/README.md +20 -0
- checkpoint-2000/adapter_config.json +21 -0
- checkpoint-2000/adapter_model.bin +3 -0
- checkpoint-2000/finetuning_args.json +16 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state_0.pth +3 -0
- checkpoint-2000/rng_state_1.pth +3 -0
- checkpoint-2000/rng_state_2.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/trainer_state.json +1216 -0
- checkpoint-2000/training_args.bin +3 -0
- checkpoint-3000/README.md +20 -0
- checkpoint-3000/adapter_config.json +21 -0
- checkpoint-3000/adapter_model.bin +3 -0
- checkpoint-3000/finetuning_args.json +16 -0
- checkpoint-3000/optimizer.pt +3 -0
- checkpoint-3000/rng_state_0.pth +3 -0
- checkpoint-3000/rng_state_1.pth +3 -0
- checkpoint-3000/rng_state_2.pth +3 -0
- checkpoint-3000/scheduler.pt +3 -0
- checkpoint-3000/trainer_state.json +1816 -0
- checkpoint-3000/training_args.bin +3 -0
- checkpoint-4000/README.md +20 -0
- checkpoint-4000/adapter_config.json +21 -0
- checkpoint-4000/adapter_model.bin +3 -0
- checkpoint-4000/finetuning_args.json +16 -0
- checkpoint-4000/optimizer.pt +3 -0
- checkpoint-4000/rng_state_0.pth +3 -0
- checkpoint-4000/rng_state_1.pth +3 -0
- checkpoint-4000/rng_state_2.pth +3 -0
- checkpoint-4000/scheduler.pt +3 -0
- checkpoint-4000/trainer_state.json +2416 -0
- checkpoint-4000/training_args.bin +3 -0
- checkpoint-5000/README.md +20 -0
- checkpoint-5000/adapter_config.json +21 -0
README.md
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- load_in_8bit: False
|
| 9 |
+
- load_in_4bit: True
|
| 10 |
+
- llm_int8_threshold: 6.0
|
| 11 |
+
- llm_int8_skip_modules: None
|
| 12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
+
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: float16
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.4.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./Llama-2-7b-chat-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 32.0,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4368dd6a27f87f5b0669b99d1bf2c8c71855de07034dbcfe61195d8eefc4f9ce
|
| 3 |
+
size 16821197
|
all_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.0,
|
| 3 |
+
"train_loss": 0.7826863387384284,
|
| 4 |
+
"train_runtime": 100268.7949,
|
| 5 |
+
"train_samples_per_second": 6.462,
|
| 6 |
+
"train_steps_per_second": 0.067
|
| 7 |
+
}
|
checkpoint-1000/README.md
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- load_in_8bit: False
|
| 9 |
+
- load_in_4bit: True
|
| 10 |
+
- llm_int8_threshold: 6.0
|
| 11 |
+
- llm_int8_skip_modules: None
|
| 12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
+
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: float16
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.4.0
|
checkpoint-1000/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./Llama-2-7b-chat-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 32.0,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-1000/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:01d60e225a241cd5560b29edfc885d45b22d33e3621210139dfba429cb06ad0e
|
| 3 |
+
size 16821197
|
checkpoint-1000/finetuning_args.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dpo_beta": 0.1,
|
| 3 |
+
"finetuning_type": "lora",
|
| 4 |
+
"lora_alpha": 32.0,
|
| 5 |
+
"lora_dropout": 0.1,
|
| 6 |
+
"lora_rank": 8,
|
| 7 |
+
"lora_target": [
|
| 8 |
+
"q_proj",
|
| 9 |
+
"v_proj"
|
| 10 |
+
],
|
| 11 |
+
"name_module_trainable": "mlp",
|
| 12 |
+
"num_hidden_layers": 32,
|
| 13 |
+
"num_layer_trainable": 3,
|
| 14 |
+
"ppo_score_norm": false,
|
| 15 |
+
"resume_lora_training": true
|
| 16 |
+
}
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7a1c5d57a78b918b65ef7e05c7e61e1f2ac502bb9ce489b5aa3e1c97e6c81094
|
| 3 |
+
size 33661637
|
checkpoint-1000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c91a43c96592d2f3f8005a8c57078635c83dcd2d505c009e27479785d727aa23
|
| 3 |
+
size 18679
|
checkpoint-1000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3353bea80f838cc0c11067bc759bc1b19feead95c9fb1486b8d9d45ea5c53757
|
| 3 |
+
size 18679
|
checkpoint-1000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:468555d3ee30c066dd6eb5e97490bff31eb9755b7f5fb041f2afce47110ffc54
|
| 3 |
+
size 18679
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2eba20c386c6cddbfa9305a435bd87f8b87262d92ccda73c302a46ba26396e5d
|
| 3 |
+
size 627
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,616 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.29631824579598487,
|
| 5 |
+
"global_step": 1000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 4.9999729068921297e-05,
|
| 13 |
+
"loss": 1.369,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 4.9998916281557476e-05,
|
| 19 |
+
"loss": 1.1704,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 4.999772150012383e-05,
|
| 25 |
+
"loss": 1.131,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
+
"learning_rate": 4.99958792440589e-05,
|
| 31 |
+
"loss": 1.0618,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.01,
|
| 36 |
+
"learning_rate": 4.999401536044942e-05,
|
| 37 |
+
"loss": 1.0539,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
+
"learning_rate": 4.9991197949909295e-05,
|
| 43 |
+
"loss": 1.0157,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"learning_rate": 4.998783886799168e-05,
|
| 49 |
+
"loss": 1.0235,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.02,
|
| 54 |
+
"learning_rate": 4.998393818750294e-05,
|
| 55 |
+
"loss": 0.9841,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
+
"learning_rate": 4.9979495992988316e-05,
|
| 61 |
+
"loss": 0.9787,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.03,
|
| 66 |
+
"learning_rate": 4.997451238073011e-05,
|
| 67 |
+
"loss": 0.9716,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.03,
|
| 72 |
+
"learning_rate": 4.996898745874553e-05,
|
| 73 |
+
"loss": 0.9718,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
+
"learning_rate": 4.996292134678445e-05,
|
| 79 |
+
"loss": 0.9608,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04,
|
| 84 |
+
"learning_rate": 4.995631417632672e-05,
|
| 85 |
+
"loss": 0.9547,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04,
|
| 90 |
+
"learning_rate": 4.994916609057936e-05,
|
| 91 |
+
"loss": 0.959,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.04,
|
| 96 |
+
"learning_rate": 4.9941477244473455e-05,
|
| 97 |
+
"loss": 0.9297,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.05,
|
| 102 |
+
"learning_rate": 4.993324780466081e-05,
|
| 103 |
+
"loss": 0.927,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.05,
|
| 108 |
+
"learning_rate": 4.992447794951029e-05,
|
| 109 |
+
"loss": 0.9352,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.05,
|
| 114 |
+
"learning_rate": 4.9915167869104006e-05,
|
| 115 |
+
"loss": 0.9244,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.06,
|
| 120 |
+
"learning_rate": 4.990531776523317e-05,
|
| 121 |
+
"loss": 0.9185,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06,
|
| 126 |
+
"learning_rate": 4.9894927851393726e-05,
|
| 127 |
+
"loss": 0.9201,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06,
|
| 132 |
+
"learning_rate": 4.988399835278171e-05,
|
| 133 |
+
"loss": 0.9053,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.07,
|
| 138 |
+
"learning_rate": 4.987252950628839e-05,
|
| 139 |
+
"loss": 0.912,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.07,
|
| 144 |
+
"learning_rate": 4.986052156049513e-05,
|
| 145 |
+
"loss": 0.9013,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.07,
|
| 150 |
+
"learning_rate": 4.984797477566799e-05,
|
| 151 |
+
"loss": 0.8942,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.07,
|
| 156 |
+
"learning_rate": 4.9834889423752074e-05,
|
| 157 |
+
"loss": 0.8873,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.08,
|
| 162 |
+
"learning_rate": 4.9821265788365664e-05,
|
| 163 |
+
"loss": 0.9036,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08,
|
| 168 |
+
"learning_rate": 4.980710416479407e-05,
|
| 169 |
+
"loss": 0.901,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08,
|
| 174 |
+
"learning_rate": 4.97924048599832e-05,
|
| 175 |
+
"loss": 0.8941,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.09,
|
| 180 |
+
"learning_rate": 4.977716819253293e-05,
|
| 181 |
+
"loss": 0.8886,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.09,
|
| 186 |
+
"learning_rate": 4.97613944926902e-05,
|
| 187 |
+
"loss": 0.8852,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.09,
|
| 192 |
+
"learning_rate": 4.974508410234187e-05,
|
| 193 |
+
"loss": 0.8877,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.09,
|
| 198 |
+
"learning_rate": 4.972823737500725e-05,
|
| 199 |
+
"loss": 0.8833,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.1,
|
| 204 |
+
"learning_rate": 4.971085467583051e-05,
|
| 205 |
+
"loss": 0.869,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1,
|
| 210 |
+
"learning_rate": 4.969293638157273e-05,
|
| 211 |
+
"loss": 0.877,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1,
|
| 216 |
+
"learning_rate": 4.967448288060372e-05,
|
| 217 |
+
"loss": 0.8793,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.11,
|
| 222 |
+
"learning_rate": 4.9655494572893644e-05,
|
| 223 |
+
"loss": 0.8742,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.11,
|
| 228 |
+
"learning_rate": 4.9635971870004316e-05,
|
| 229 |
+
"loss": 0.8774,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.11,
|
| 234 |
+
"learning_rate": 4.961591519508029e-05,
|
| 235 |
+
"loss": 0.8682,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.12,
|
| 240 |
+
"learning_rate": 4.9595324982839686e-05,
|
| 241 |
+
"loss": 0.8581,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.12,
|
| 246 |
+
"learning_rate": 4.957420167956479e-05,
|
| 247 |
+
"loss": 0.868,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.12,
|
| 252 |
+
"learning_rate": 4.955254574309234e-05,
|
| 253 |
+
"loss": 0.8672,
|
| 254 |
+
"step": 410
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12,
|
| 258 |
+
"learning_rate": 4.9530357642803646e-05,
|
| 259 |
+
"loss": 0.8672,
|
| 260 |
+
"step": 420
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.13,
|
| 264 |
+
"learning_rate": 4.950763785961436e-05,
|
| 265 |
+
"loss": 0.867,
|
| 266 |
+
"step": 430
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.13,
|
| 270 |
+
"learning_rate": 4.948438688596414e-05,
|
| 271 |
+
"loss": 0.8612,
|
| 272 |
+
"step": 440
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.13,
|
| 276 |
+
"learning_rate": 4.946060522580588e-05,
|
| 277 |
+
"loss": 0.8607,
|
| 278 |
+
"step": 450
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.14,
|
| 282 |
+
"learning_rate": 4.943629339459484e-05,
|
| 283 |
+
"loss": 0.8588,
|
| 284 |
+
"step": 460
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.14,
|
| 288 |
+
"learning_rate": 4.9411451919277484e-05,
|
| 289 |
+
"loss": 0.8585,
|
| 290 |
+
"step": 470
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.14,
|
| 294 |
+
"learning_rate": 4.9386081338280035e-05,
|
| 295 |
+
"loss": 0.8574,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"learning_rate": 4.9360182201496785e-05,
|
| 301 |
+
"loss": 0.8562,
|
| 302 |
+
"step": 490
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.15,
|
| 306 |
+
"learning_rate": 4.933375507027824e-05,
|
| 307 |
+
"loss": 0.8634,
|
| 308 |
+
"step": 500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.15,
|
| 312 |
+
"learning_rate": 4.930680051741887e-05,
|
| 313 |
+
"loss": 0.8483,
|
| 314 |
+
"step": 510
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.15,
|
| 318 |
+
"learning_rate": 4.9279319127144775e-05,
|
| 319 |
+
"loss": 0.8559,
|
| 320 |
+
"step": 520
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.16,
|
| 324 |
+
"learning_rate": 4.9251311495100974e-05,
|
| 325 |
+
"loss": 0.8526,
|
| 326 |
+
"step": 530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.16,
|
| 330 |
+
"learning_rate": 4.9222778228338506e-05,
|
| 331 |
+
"loss": 0.85,
|
| 332 |
+
"step": 540
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.16,
|
| 336 |
+
"learning_rate": 4.919371994530125e-05,
|
| 337 |
+
"loss": 0.8547,
|
| 338 |
+
"step": 550
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17,
|
| 342 |
+
"learning_rate": 4.9164137275812597e-05,
|
| 343 |
+
"loss": 0.8533,
|
| 344 |
+
"step": 560
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.17,
|
| 348 |
+
"learning_rate": 4.913403086106167e-05,
|
| 349 |
+
"loss": 0.8431,
|
| 350 |
+
"step": 570
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.17,
|
| 354 |
+
"learning_rate": 4.9103401353589585e-05,
|
| 355 |
+
"loss": 0.8346,
|
| 356 |
+
"step": 580
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.17,
|
| 360 |
+
"learning_rate": 4.9072249417275164e-05,
|
| 361 |
+
"loss": 0.8448,
|
| 362 |
+
"step": 590
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.18,
|
| 366 |
+
"learning_rate": 4.9040575727320606e-05,
|
| 367 |
+
"loss": 0.8477,
|
| 368 |
+
"step": 600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.18,
|
| 372 |
+
"learning_rate": 4.90083809702369e-05,
|
| 373 |
+
"loss": 0.8555,
|
| 374 |
+
"step": 610
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.18,
|
| 378 |
+
"learning_rate": 4.897566584382884e-05,
|
| 379 |
+
"loss": 0.8295,
|
| 380 |
+
"step": 620
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19,
|
| 384 |
+
"learning_rate": 4.8942431057180015e-05,
|
| 385 |
+
"loss": 0.8359,
|
| 386 |
+
"step": 630
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.19,
|
| 390 |
+
"learning_rate": 4.890867733063733e-05,
|
| 391 |
+
"loss": 0.8521,
|
| 392 |
+
"step": 640
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.19,
|
| 396 |
+
"learning_rate": 4.887440539579547e-05,
|
| 397 |
+
"loss": 0.8349,
|
| 398 |
+
"step": 650
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.2,
|
| 402 |
+
"learning_rate": 4.883961599548103e-05,
|
| 403 |
+
"loss": 0.8285,
|
| 404 |
+
"step": 660
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.2,
|
| 408 |
+
"learning_rate": 4.8804309883736377e-05,
|
| 409 |
+
"loss": 0.831,
|
| 410 |
+
"step": 670
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.2,
|
| 414 |
+
"learning_rate": 4.876848782580336e-05,
|
| 415 |
+
"loss": 0.8295,
|
| 416 |
+
"step": 680
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2,
|
| 420 |
+
"learning_rate": 4.873215059810666e-05,
|
| 421 |
+
"loss": 0.8406,
|
| 422 |
+
"step": 690
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21,
|
| 426 |
+
"learning_rate": 4.8695298988237035e-05,
|
| 427 |
+
"loss": 0.8456,
|
| 428 |
+
"step": 700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.21,
|
| 432 |
+
"learning_rate": 4.8657933794934205e-05,
|
| 433 |
+
"loss": 0.8218,
|
| 434 |
+
"step": 710
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.21,
|
| 438 |
+
"learning_rate": 4.8620055828069535e-05,
|
| 439 |
+
"loss": 0.8424,
|
| 440 |
+
"step": 720
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.22,
|
| 444 |
+
"learning_rate": 4.858166590862849e-05,
|
| 445 |
+
"loss": 0.8258,
|
| 446 |
+
"step": 730
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.22,
|
| 450 |
+
"learning_rate": 4.854276486869287e-05,
|
| 451 |
+
"loss": 0.8352,
|
| 452 |
+
"step": 740
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.22,
|
| 456 |
+
"learning_rate": 4.850335355142272e-05,
|
| 457 |
+
"loss": 0.836,
|
| 458 |
+
"step": 750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.23,
|
| 462 |
+
"learning_rate": 4.846343281103809e-05,
|
| 463 |
+
"loss": 0.8388,
|
| 464 |
+
"step": 760
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.23,
|
| 468 |
+
"learning_rate": 4.842300351280054e-05,
|
| 469 |
+
"loss": 0.8287,
|
| 470 |
+
"step": 770
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.23,
|
| 474 |
+
"learning_rate": 4.8382066532994316e-05,
|
| 475 |
+
"loss": 0.8394,
|
| 476 |
+
"step": 780
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.23,
|
| 480 |
+
"learning_rate": 4.834062275890745e-05,
|
| 481 |
+
"loss": 0.8253,
|
| 482 |
+
"step": 790
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.24,
|
| 486 |
+
"learning_rate": 4.829867308881244e-05,
|
| 487 |
+
"loss": 0.82,
|
| 488 |
+
"step": 800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.24,
|
| 492 |
+
"learning_rate": 4.825621843194684e-05,
|
| 493 |
+
"loss": 0.8294,
|
| 494 |
+
"step": 810
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.24,
|
| 498 |
+
"learning_rate": 4.821325970849353e-05,
|
| 499 |
+
"loss": 0.8275,
|
| 500 |
+
"step": 820
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.25,
|
| 504 |
+
"learning_rate": 4.8169797849560764e-05,
|
| 505 |
+
"loss": 0.8129,
|
| 506 |
+
"step": 830
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.25,
|
| 510 |
+
"learning_rate": 4.812583379716202e-05,
|
| 511 |
+
"loss": 0.8215,
|
| 512 |
+
"step": 840
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.25,
|
| 516 |
+
"learning_rate": 4.8081368504195545e-05,
|
| 517 |
+
"loss": 0.84,
|
| 518 |
+
"step": 850
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.25,
|
| 522 |
+
"learning_rate": 4.803640293442373e-05,
|
| 523 |
+
"loss": 0.8246,
|
| 524 |
+
"step": 860
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.26,
|
| 528 |
+
"learning_rate": 4.799093806245218e-05,
|
| 529 |
+
"loss": 0.8179,
|
| 530 |
+
"step": 870
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.26,
|
| 534 |
+
"learning_rate": 4.7944974873708656e-05,
|
| 535 |
+
"loss": 0.8238,
|
| 536 |
+
"step": 880
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.26,
|
| 540 |
+
"learning_rate": 4.7898514364421655e-05,
|
| 541 |
+
"loss": 0.8327,
|
| 542 |
+
"step": 890
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.27,
|
| 546 |
+
"learning_rate": 4.785155754159886e-05,
|
| 547 |
+
"loss": 0.812,
|
| 548 |
+
"step": 900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.27,
|
| 552 |
+
"learning_rate": 4.780410542300527e-05,
|
| 553 |
+
"loss": 0.8293,
|
| 554 |
+
"step": 910
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.27,
|
| 558 |
+
"learning_rate": 4.7756159037141185e-05,
|
| 559 |
+
"loss": 0.8266,
|
| 560 |
+
"step": 920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.28,
|
| 564 |
+
"learning_rate": 4.7707719423219886e-05,
|
| 565 |
+
"loss": 0.8209,
|
| 566 |
+
"step": 930
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.28,
|
| 570 |
+
"learning_rate": 4.765878763114513e-05,
|
| 571 |
+
"loss": 0.8182,
|
| 572 |
+
"step": 940
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.28,
|
| 576 |
+
"learning_rate": 4.760936472148837e-05,
|
| 577 |
+
"loss": 0.8045,
|
| 578 |
+
"step": 950
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.28,
|
| 582 |
+
"learning_rate": 4.755945176546578e-05,
|
| 583 |
+
"loss": 0.8126,
|
| 584 |
+
"step": 960
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.29,
|
| 588 |
+
"learning_rate": 4.7509049844915036e-05,
|
| 589 |
+
"loss": 0.8201,
|
| 590 |
+
"step": 970
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.29,
|
| 594 |
+
"learning_rate": 4.7458160052271883e-05,
|
| 595 |
+
"loss": 0.8133,
|
| 596 |
+
"step": 980
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.29,
|
| 600 |
+
"learning_rate": 4.7406783490546434e-05,
|
| 601 |
+
"loss": 0.8157,
|
| 602 |
+
"step": 990
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.3,
|
| 606 |
+
"learning_rate": 4.735492127329928e-05,
|
| 607 |
+
"loss": 0.817,
|
| 608 |
+
"step": 1000
|
| 609 |
+
}
|
| 610 |
+
],
|
| 611 |
+
"max_steps": 6748,
|
| 612 |
+
"num_train_epochs": 2,
|
| 613 |
+
"total_flos": 1.778504410981204e+18,
|
| 614 |
+
"trial_name": null,
|
| 615 |
+
"trial_params": null
|
| 616 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f710e8a04dbb33457b5b725adb1da191a7b05d7ab01f6ec60c73386df9d8286
|
| 3 |
+
size 3277
|
checkpoint-2000/README.md
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- load_in_8bit: False
|
| 9 |
+
- load_in_4bit: True
|
| 10 |
+
- llm_int8_threshold: 6.0
|
| 11 |
+
- llm_int8_skip_modules: None
|
| 12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
+
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: float16
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.4.0
|
checkpoint-2000/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./Llama-2-7b-chat-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 32.0,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-2000/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0023c72d6e86d5947e659fbe1b0955df3db6f307814c12415748c9326c87b577
|
| 3 |
+
size 16821197
|
checkpoint-2000/finetuning_args.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dpo_beta": 0.1,
|
| 3 |
+
"finetuning_type": "lora",
|
| 4 |
+
"lora_alpha": 32.0,
|
| 5 |
+
"lora_dropout": 0.1,
|
| 6 |
+
"lora_rank": 8,
|
| 7 |
+
"lora_target": [
|
| 8 |
+
"q_proj",
|
| 9 |
+
"v_proj"
|
| 10 |
+
],
|
| 11 |
+
"name_module_trainable": "mlp",
|
| 12 |
+
"num_hidden_layers": 32,
|
| 13 |
+
"num_layer_trainable": 3,
|
| 14 |
+
"ppo_score_norm": false,
|
| 15 |
+
"resume_lora_training": true
|
| 16 |
+
}
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3c55f01b8687b1be7396a2ba0a869bea48c6f4ba7e11402fa621fac14f79eaff
|
| 3 |
+
size 33661637
|
checkpoint-2000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:27bb9d18d06ab8727ee8c0d76ded639732a23bdc12357cce9eb55d3b23ddfd24
|
| 3 |
+
size 18679
|
checkpoint-2000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1e83fe8157b5cc35063bd106e9aaaf436e7c73ce910a1f2155b6e589fb75ba57
|
| 3 |
+
size 18679
|
checkpoint-2000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2a1a9ff0073355b18dffaf8991e71645c238852f1bfa6ec4cc7bccae2cb83a8f
|
| 3 |
+
size 18679
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6991a324121596d5ed6d1fd72fb64a404c9f9d1ce5484efee7e0117d107705d0
|
| 3 |
+
size 627
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,1216 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.5926364915919697,
|
| 5 |
+
"global_step": 2000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 4.9999729068921297e-05,
|
| 13 |
+
"loss": 1.369,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 4.9998916281557476e-05,
|
| 19 |
+
"loss": 1.1704,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 4.999772150012383e-05,
|
| 25 |
+
"loss": 1.131,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
+
"learning_rate": 4.99958792440589e-05,
|
| 31 |
+
"loss": 1.0618,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.01,
|
| 36 |
+
"learning_rate": 4.999401536044942e-05,
|
| 37 |
+
"loss": 1.0539,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
+
"learning_rate": 4.9991197949909295e-05,
|
| 43 |
+
"loss": 1.0157,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"learning_rate": 4.998783886799168e-05,
|
| 49 |
+
"loss": 1.0235,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.02,
|
| 54 |
+
"learning_rate": 4.998393818750294e-05,
|
| 55 |
+
"loss": 0.9841,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
+
"learning_rate": 4.9979495992988316e-05,
|
| 61 |
+
"loss": 0.9787,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.03,
|
| 66 |
+
"learning_rate": 4.997451238073011e-05,
|
| 67 |
+
"loss": 0.9716,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.03,
|
| 72 |
+
"learning_rate": 4.996898745874553e-05,
|
| 73 |
+
"loss": 0.9718,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
+
"learning_rate": 4.996292134678445e-05,
|
| 79 |
+
"loss": 0.9608,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04,
|
| 84 |
+
"learning_rate": 4.995631417632672e-05,
|
| 85 |
+
"loss": 0.9547,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04,
|
| 90 |
+
"learning_rate": 4.994916609057936e-05,
|
| 91 |
+
"loss": 0.959,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.04,
|
| 96 |
+
"learning_rate": 4.9941477244473455e-05,
|
| 97 |
+
"loss": 0.9297,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.05,
|
| 102 |
+
"learning_rate": 4.993324780466081e-05,
|
| 103 |
+
"loss": 0.927,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.05,
|
| 108 |
+
"learning_rate": 4.992447794951029e-05,
|
| 109 |
+
"loss": 0.9352,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.05,
|
| 114 |
+
"learning_rate": 4.9915167869104006e-05,
|
| 115 |
+
"loss": 0.9244,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.06,
|
| 120 |
+
"learning_rate": 4.990531776523317e-05,
|
| 121 |
+
"loss": 0.9185,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06,
|
| 126 |
+
"learning_rate": 4.9894927851393726e-05,
|
| 127 |
+
"loss": 0.9201,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06,
|
| 132 |
+
"learning_rate": 4.988399835278171e-05,
|
| 133 |
+
"loss": 0.9053,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.07,
|
| 138 |
+
"learning_rate": 4.987252950628839e-05,
|
| 139 |
+
"loss": 0.912,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.07,
|
| 144 |
+
"learning_rate": 4.986052156049513e-05,
|
| 145 |
+
"loss": 0.9013,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.07,
|
| 150 |
+
"learning_rate": 4.984797477566799e-05,
|
| 151 |
+
"loss": 0.8942,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.07,
|
| 156 |
+
"learning_rate": 4.9834889423752074e-05,
|
| 157 |
+
"loss": 0.8873,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.08,
|
| 162 |
+
"learning_rate": 4.9821265788365664e-05,
|
| 163 |
+
"loss": 0.9036,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08,
|
| 168 |
+
"learning_rate": 4.980710416479407e-05,
|
| 169 |
+
"loss": 0.901,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08,
|
| 174 |
+
"learning_rate": 4.97924048599832e-05,
|
| 175 |
+
"loss": 0.8941,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.09,
|
| 180 |
+
"learning_rate": 4.977716819253293e-05,
|
| 181 |
+
"loss": 0.8886,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.09,
|
| 186 |
+
"learning_rate": 4.97613944926902e-05,
|
| 187 |
+
"loss": 0.8852,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.09,
|
| 192 |
+
"learning_rate": 4.974508410234187e-05,
|
| 193 |
+
"loss": 0.8877,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.09,
|
| 198 |
+
"learning_rate": 4.972823737500725e-05,
|
| 199 |
+
"loss": 0.8833,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.1,
|
| 204 |
+
"learning_rate": 4.971085467583051e-05,
|
| 205 |
+
"loss": 0.869,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1,
|
| 210 |
+
"learning_rate": 4.969293638157273e-05,
|
| 211 |
+
"loss": 0.877,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1,
|
| 216 |
+
"learning_rate": 4.967448288060372e-05,
|
| 217 |
+
"loss": 0.8793,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.11,
|
| 222 |
+
"learning_rate": 4.9655494572893644e-05,
|
| 223 |
+
"loss": 0.8742,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.11,
|
| 228 |
+
"learning_rate": 4.9635971870004316e-05,
|
| 229 |
+
"loss": 0.8774,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.11,
|
| 234 |
+
"learning_rate": 4.961591519508029e-05,
|
| 235 |
+
"loss": 0.8682,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.12,
|
| 240 |
+
"learning_rate": 4.9595324982839686e-05,
|
| 241 |
+
"loss": 0.8581,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.12,
|
| 246 |
+
"learning_rate": 4.957420167956479e-05,
|
| 247 |
+
"loss": 0.868,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.12,
|
| 252 |
+
"learning_rate": 4.955254574309234e-05,
|
| 253 |
+
"loss": 0.8672,
|
| 254 |
+
"step": 410
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12,
|
| 258 |
+
"learning_rate": 4.9530357642803646e-05,
|
| 259 |
+
"loss": 0.8672,
|
| 260 |
+
"step": 420
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.13,
|
| 264 |
+
"learning_rate": 4.950763785961436e-05,
|
| 265 |
+
"loss": 0.867,
|
| 266 |
+
"step": 430
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.13,
|
| 270 |
+
"learning_rate": 4.948438688596414e-05,
|
| 271 |
+
"loss": 0.8612,
|
| 272 |
+
"step": 440
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.13,
|
| 276 |
+
"learning_rate": 4.946060522580588e-05,
|
| 277 |
+
"loss": 0.8607,
|
| 278 |
+
"step": 450
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.14,
|
| 282 |
+
"learning_rate": 4.943629339459484e-05,
|
| 283 |
+
"loss": 0.8588,
|
| 284 |
+
"step": 460
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.14,
|
| 288 |
+
"learning_rate": 4.9411451919277484e-05,
|
| 289 |
+
"loss": 0.8585,
|
| 290 |
+
"step": 470
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.14,
|
| 294 |
+
"learning_rate": 4.9386081338280035e-05,
|
| 295 |
+
"loss": 0.8574,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"learning_rate": 4.9360182201496785e-05,
|
| 301 |
+
"loss": 0.8562,
|
| 302 |
+
"step": 490
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.15,
|
| 306 |
+
"learning_rate": 4.933375507027824e-05,
|
| 307 |
+
"loss": 0.8634,
|
| 308 |
+
"step": 500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.15,
|
| 312 |
+
"learning_rate": 4.930680051741887e-05,
|
| 313 |
+
"loss": 0.8483,
|
| 314 |
+
"step": 510
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.15,
|
| 318 |
+
"learning_rate": 4.9279319127144775e-05,
|
| 319 |
+
"loss": 0.8559,
|
| 320 |
+
"step": 520
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.16,
|
| 324 |
+
"learning_rate": 4.9251311495100974e-05,
|
| 325 |
+
"loss": 0.8526,
|
| 326 |
+
"step": 530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.16,
|
| 330 |
+
"learning_rate": 4.9222778228338506e-05,
|
| 331 |
+
"loss": 0.85,
|
| 332 |
+
"step": 540
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.16,
|
| 336 |
+
"learning_rate": 4.919371994530125e-05,
|
| 337 |
+
"loss": 0.8547,
|
| 338 |
+
"step": 550
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17,
|
| 342 |
+
"learning_rate": 4.9164137275812597e-05,
|
| 343 |
+
"loss": 0.8533,
|
| 344 |
+
"step": 560
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.17,
|
| 348 |
+
"learning_rate": 4.913403086106167e-05,
|
| 349 |
+
"loss": 0.8431,
|
| 350 |
+
"step": 570
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.17,
|
| 354 |
+
"learning_rate": 4.9103401353589585e-05,
|
| 355 |
+
"loss": 0.8346,
|
| 356 |
+
"step": 580
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.17,
|
| 360 |
+
"learning_rate": 4.9072249417275164e-05,
|
| 361 |
+
"loss": 0.8448,
|
| 362 |
+
"step": 590
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.18,
|
| 366 |
+
"learning_rate": 4.9040575727320606e-05,
|
| 367 |
+
"loss": 0.8477,
|
| 368 |
+
"step": 600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.18,
|
| 372 |
+
"learning_rate": 4.90083809702369e-05,
|
| 373 |
+
"loss": 0.8555,
|
| 374 |
+
"step": 610
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.18,
|
| 378 |
+
"learning_rate": 4.897566584382884e-05,
|
| 379 |
+
"loss": 0.8295,
|
| 380 |
+
"step": 620
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19,
|
| 384 |
+
"learning_rate": 4.8942431057180015e-05,
|
| 385 |
+
"loss": 0.8359,
|
| 386 |
+
"step": 630
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.19,
|
| 390 |
+
"learning_rate": 4.890867733063733e-05,
|
| 391 |
+
"loss": 0.8521,
|
| 392 |
+
"step": 640
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.19,
|
| 396 |
+
"learning_rate": 4.887440539579547e-05,
|
| 397 |
+
"loss": 0.8349,
|
| 398 |
+
"step": 650
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.2,
|
| 402 |
+
"learning_rate": 4.883961599548103e-05,
|
| 403 |
+
"loss": 0.8285,
|
| 404 |
+
"step": 660
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.2,
|
| 408 |
+
"learning_rate": 4.8804309883736377e-05,
|
| 409 |
+
"loss": 0.831,
|
| 410 |
+
"step": 670
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.2,
|
| 414 |
+
"learning_rate": 4.876848782580336e-05,
|
| 415 |
+
"loss": 0.8295,
|
| 416 |
+
"step": 680
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2,
|
| 420 |
+
"learning_rate": 4.873215059810666e-05,
|
| 421 |
+
"loss": 0.8406,
|
| 422 |
+
"step": 690
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21,
|
| 426 |
+
"learning_rate": 4.8695298988237035e-05,
|
| 427 |
+
"loss": 0.8456,
|
| 428 |
+
"step": 700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.21,
|
| 432 |
+
"learning_rate": 4.8657933794934205e-05,
|
| 433 |
+
"loss": 0.8218,
|
| 434 |
+
"step": 710
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.21,
|
| 438 |
+
"learning_rate": 4.8620055828069535e-05,
|
| 439 |
+
"loss": 0.8424,
|
| 440 |
+
"step": 720
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.22,
|
| 444 |
+
"learning_rate": 4.858166590862849e-05,
|
| 445 |
+
"loss": 0.8258,
|
| 446 |
+
"step": 730
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.22,
|
| 450 |
+
"learning_rate": 4.854276486869287e-05,
|
| 451 |
+
"loss": 0.8352,
|
| 452 |
+
"step": 740
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.22,
|
| 456 |
+
"learning_rate": 4.850335355142272e-05,
|
| 457 |
+
"loss": 0.836,
|
| 458 |
+
"step": 750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.23,
|
| 462 |
+
"learning_rate": 4.846343281103809e-05,
|
| 463 |
+
"loss": 0.8388,
|
| 464 |
+
"step": 760
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.23,
|
| 468 |
+
"learning_rate": 4.842300351280054e-05,
|
| 469 |
+
"loss": 0.8287,
|
| 470 |
+
"step": 770
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.23,
|
| 474 |
+
"learning_rate": 4.8382066532994316e-05,
|
| 475 |
+
"loss": 0.8394,
|
| 476 |
+
"step": 780
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.23,
|
| 480 |
+
"learning_rate": 4.834062275890745e-05,
|
| 481 |
+
"loss": 0.8253,
|
| 482 |
+
"step": 790
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.24,
|
| 486 |
+
"learning_rate": 4.829867308881244e-05,
|
| 487 |
+
"loss": 0.82,
|
| 488 |
+
"step": 800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.24,
|
| 492 |
+
"learning_rate": 4.825621843194684e-05,
|
| 493 |
+
"loss": 0.8294,
|
| 494 |
+
"step": 810
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.24,
|
| 498 |
+
"learning_rate": 4.821325970849353e-05,
|
| 499 |
+
"loss": 0.8275,
|
| 500 |
+
"step": 820
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.25,
|
| 504 |
+
"learning_rate": 4.8169797849560764e-05,
|
| 505 |
+
"loss": 0.8129,
|
| 506 |
+
"step": 830
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.25,
|
| 510 |
+
"learning_rate": 4.812583379716202e-05,
|
| 511 |
+
"loss": 0.8215,
|
| 512 |
+
"step": 840
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.25,
|
| 516 |
+
"learning_rate": 4.8081368504195545e-05,
|
| 517 |
+
"loss": 0.84,
|
| 518 |
+
"step": 850
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.25,
|
| 522 |
+
"learning_rate": 4.803640293442373e-05,
|
| 523 |
+
"loss": 0.8246,
|
| 524 |
+
"step": 860
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.26,
|
| 528 |
+
"learning_rate": 4.799093806245218e-05,
|
| 529 |
+
"loss": 0.8179,
|
| 530 |
+
"step": 870
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.26,
|
| 534 |
+
"learning_rate": 4.7944974873708656e-05,
|
| 535 |
+
"loss": 0.8238,
|
| 536 |
+
"step": 880
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.26,
|
| 540 |
+
"learning_rate": 4.7898514364421655e-05,
|
| 541 |
+
"loss": 0.8327,
|
| 542 |
+
"step": 890
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.27,
|
| 546 |
+
"learning_rate": 4.785155754159886e-05,
|
| 547 |
+
"loss": 0.812,
|
| 548 |
+
"step": 900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.27,
|
| 552 |
+
"learning_rate": 4.780410542300527e-05,
|
| 553 |
+
"loss": 0.8293,
|
| 554 |
+
"step": 910
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.27,
|
| 558 |
+
"learning_rate": 4.7756159037141185e-05,
|
| 559 |
+
"loss": 0.8266,
|
| 560 |
+
"step": 920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.28,
|
| 564 |
+
"learning_rate": 4.7707719423219886e-05,
|
| 565 |
+
"loss": 0.8209,
|
| 566 |
+
"step": 930
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.28,
|
| 570 |
+
"learning_rate": 4.765878763114513e-05,
|
| 571 |
+
"loss": 0.8182,
|
| 572 |
+
"step": 940
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.28,
|
| 576 |
+
"learning_rate": 4.760936472148837e-05,
|
| 577 |
+
"loss": 0.8045,
|
| 578 |
+
"step": 950
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.28,
|
| 582 |
+
"learning_rate": 4.755945176546578e-05,
|
| 583 |
+
"loss": 0.8126,
|
| 584 |
+
"step": 960
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.29,
|
| 588 |
+
"learning_rate": 4.7509049844915036e-05,
|
| 589 |
+
"loss": 0.8201,
|
| 590 |
+
"step": 970
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.29,
|
| 594 |
+
"learning_rate": 4.7458160052271883e-05,
|
| 595 |
+
"loss": 0.8133,
|
| 596 |
+
"step": 980
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.29,
|
| 600 |
+
"learning_rate": 4.7406783490546434e-05,
|
| 601 |
+
"loss": 0.8157,
|
| 602 |
+
"step": 990
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.3,
|
| 606 |
+
"learning_rate": 4.735492127329928e-05,
|
| 607 |
+
"loss": 0.817,
|
| 608 |
+
"step": 1000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.3,
|
| 612 |
+
"learning_rate": 4.7302574524617315e-05,
|
| 613 |
+
"loss": 0.8002,
|
| 614 |
+
"step": 1010
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.3,
|
| 618 |
+
"learning_rate": 4.724974437908944e-05,
|
| 619 |
+
"loss": 0.8281,
|
| 620 |
+
"step": 1020
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.31,
|
| 624 |
+
"learning_rate": 4.719643198178193e-05,
|
| 625 |
+
"loss": 0.8129,
|
| 626 |
+
"step": 1030
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.31,
|
| 630 |
+
"learning_rate": 4.71426384882136e-05,
|
| 631 |
+
"loss": 0.8125,
|
| 632 |
+
"step": 1040
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.31,
|
| 636 |
+
"learning_rate": 4.708836506433078e-05,
|
| 637 |
+
"loss": 0.8066,
|
| 638 |
+
"step": 1050
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.31,
|
| 642 |
+
"learning_rate": 4.7033612886482066e-05,
|
| 643 |
+
"loss": 0.8159,
|
| 644 |
+
"step": 1060
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.32,
|
| 648 |
+
"learning_rate": 4.697838314139278e-05,
|
| 649 |
+
"loss": 0.8288,
|
| 650 |
+
"step": 1070
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.32,
|
| 654 |
+
"learning_rate": 4.6922677026139275e-05,
|
| 655 |
+
"loss": 0.8175,
|
| 656 |
+
"step": 1080
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.32,
|
| 660 |
+
"learning_rate": 4.686649574812299e-05,
|
| 661 |
+
"loss": 0.8107,
|
| 662 |
+
"step": 1090
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.33,
|
| 666 |
+
"learning_rate": 4.680984052504426e-05,
|
| 667 |
+
"loss": 0.8127,
|
| 668 |
+
"step": 1100
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.33,
|
| 672 |
+
"learning_rate": 4.6752712584875935e-05,
|
| 673 |
+
"loss": 0.8023,
|
| 674 |
+
"step": 1110
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.33,
|
| 678 |
+
"learning_rate": 4.669511316583678e-05,
|
| 679 |
+
"loss": 0.8066,
|
| 680 |
+
"step": 1120
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.33,
|
| 684 |
+
"learning_rate": 4.663704351636461e-05,
|
| 685 |
+
"loss": 0.8069,
|
| 686 |
+
"step": 1130
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.34,
|
| 690 |
+
"learning_rate": 4.6578504895089245e-05,
|
| 691 |
+
"loss": 0.8129,
|
| 692 |
+
"step": 1140
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.34,
|
| 696 |
+
"learning_rate": 4.651949857080524e-05,
|
| 697 |
+
"loss": 0.8247,
|
| 698 |
+
"step": 1150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.34,
|
| 702 |
+
"learning_rate": 4.6460025822444354e-05,
|
| 703 |
+
"loss": 0.8125,
|
| 704 |
+
"step": 1160
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.35,
|
| 708 |
+
"learning_rate": 4.640008793904785e-05,
|
| 709 |
+
"loss": 0.8141,
|
| 710 |
+
"step": 1170
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.35,
|
| 714 |
+
"learning_rate": 4.6339686219738576e-05,
|
| 715 |
+
"loss": 0.8066,
|
| 716 |
+
"step": 1180
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.35,
|
| 720 |
+
"learning_rate": 4.627882197369276e-05,
|
| 721 |
+
"loss": 0.8052,
|
| 722 |
+
"step": 1190
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.36,
|
| 726 |
+
"learning_rate": 4.621749652011167e-05,
|
| 727 |
+
"loss": 0.8114,
|
| 728 |
+
"step": 1200
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.36,
|
| 732 |
+
"learning_rate": 4.615571118819301e-05,
|
| 733 |
+
"loss": 0.8001,
|
| 734 |
+
"step": 1210
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.36,
|
| 738 |
+
"learning_rate": 4.60934673171021e-05,
|
| 739 |
+
"loss": 0.7938,
|
| 740 |
+
"step": 1220
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.36,
|
| 744 |
+
"learning_rate": 4.6030766255942884e-05,
|
| 745 |
+
"loss": 0.8107,
|
| 746 |
+
"step": 1230
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.37,
|
| 750 |
+
"learning_rate": 4.596760936372865e-05,
|
| 751 |
+
"loss": 0.8096,
|
| 752 |
+
"step": 1240
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.37,
|
| 756 |
+
"learning_rate": 4.5903998009352586e-05,
|
| 757 |
+
"loss": 0.8105,
|
| 758 |
+
"step": 1250
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.37,
|
| 762 |
+
"learning_rate": 4.583993357155813e-05,
|
| 763 |
+
"loss": 0.7995,
|
| 764 |
+
"step": 1260
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.38,
|
| 768 |
+
"learning_rate": 4.577541743890906e-05,
|
| 769 |
+
"loss": 0.8161,
|
| 770 |
+
"step": 1270
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.38,
|
| 774 |
+
"learning_rate": 4.571045100975941e-05,
|
| 775 |
+
"loss": 0.8003,
|
| 776 |
+
"step": 1280
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.38,
|
| 780 |
+
"learning_rate": 4.564503569222315e-05,
|
| 781 |
+
"loss": 0.8114,
|
| 782 |
+
"step": 1290
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.39,
|
| 786 |
+
"learning_rate": 4.5579172904143685e-05,
|
| 787 |
+
"loss": 0.8165,
|
| 788 |
+
"step": 1300
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.39,
|
| 792 |
+
"learning_rate": 4.551286407306312e-05,
|
| 793 |
+
"loss": 0.8022,
|
| 794 |
+
"step": 1310
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.39,
|
| 798 |
+
"learning_rate": 4.54461106361913e-05,
|
| 799 |
+
"loss": 0.8032,
|
| 800 |
+
"step": 1320
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.39,
|
| 804 |
+
"learning_rate": 4.5378914040374684e-05,
|
| 805 |
+
"loss": 0.8067,
|
| 806 |
+
"step": 1330
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.4,
|
| 810 |
+
"learning_rate": 4.531127574206496e-05,
|
| 811 |
+
"loss": 0.789,
|
| 812 |
+
"step": 1340
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.4,
|
| 816 |
+
"learning_rate": 4.524319720728749e-05,
|
| 817 |
+
"loss": 0.8139,
|
| 818 |
+
"step": 1350
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.4,
|
| 822 |
+
"learning_rate": 4.517467991160956e-05,
|
| 823 |
+
"loss": 0.805,
|
| 824 |
+
"step": 1360
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.41,
|
| 828 |
+
"learning_rate": 4.510572534010836e-05,
|
| 829 |
+
"loss": 0.8123,
|
| 830 |
+
"step": 1370
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.41,
|
| 834 |
+
"learning_rate": 4.5036334987338785e-05,
|
| 835 |
+
"loss": 0.7976,
|
| 836 |
+
"step": 1380
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.41,
|
| 840 |
+
"learning_rate": 4.4966510357301094e-05,
|
| 841 |
+
"loss": 0.813,
|
| 842 |
+
"step": 1390
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.41,
|
| 846 |
+
"learning_rate": 4.489625296340828e-05,
|
| 847 |
+
"loss": 0.7975,
|
| 848 |
+
"step": 1400
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.42,
|
| 852 |
+
"learning_rate": 4.4825564328453256e-05,
|
| 853 |
+
"loss": 0.7969,
|
| 854 |
+
"step": 1410
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.42,
|
| 858 |
+
"learning_rate": 4.4754445984575874e-05,
|
| 859 |
+
"loss": 0.8006,
|
| 860 |
+
"step": 1420
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.42,
|
| 864 |
+
"learning_rate": 4.4682899473229714e-05,
|
| 865 |
+
"loss": 0.8041,
|
| 866 |
+
"step": 1430
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.43,
|
| 870 |
+
"learning_rate": 4.461092634514863e-05,
|
| 871 |
+
"loss": 0.7998,
|
| 872 |
+
"step": 1440
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.43,
|
| 876 |
+
"learning_rate": 4.453852816031323e-05,
|
| 877 |
+
"loss": 0.8,
|
| 878 |
+
"step": 1450
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.43,
|
| 882 |
+
"learning_rate": 4.446570648791695e-05,
|
| 883 |
+
"loss": 0.7993,
|
| 884 |
+
"step": 1460
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.44,
|
| 888 |
+
"learning_rate": 4.439246290633216e-05,
|
| 889 |
+
"loss": 0.8113,
|
| 890 |
+
"step": 1470
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.44,
|
| 894 |
+
"learning_rate": 4.431879900307585e-05,
|
| 895 |
+
"loss": 0.8133,
|
| 896 |
+
"step": 1480
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.44,
|
| 900 |
+
"learning_rate": 4.4244716374775274e-05,
|
| 901 |
+
"loss": 0.7963,
|
| 902 |
+
"step": 1490
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.44,
|
| 906 |
+
"learning_rate": 4.4170216627133365e-05,
|
| 907 |
+
"loss": 0.8005,
|
| 908 |
+
"step": 1500
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.45,
|
| 912 |
+
"learning_rate": 4.409530137489386e-05,
|
| 913 |
+
"loss": 0.8016,
|
| 914 |
+
"step": 1510
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.45,
|
| 918 |
+
"learning_rate": 4.4019972241806384e-05,
|
| 919 |
+
"loss": 0.7858,
|
| 920 |
+
"step": 1520
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.45,
|
| 924 |
+
"learning_rate": 4.3944230860591194e-05,
|
| 925 |
+
"loss": 0.8009,
|
| 926 |
+
"step": 1530
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.46,
|
| 930 |
+
"learning_rate": 4.3868078872903814e-05,
|
| 931 |
+
"loss": 0.7929,
|
| 932 |
+
"step": 1540
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.46,
|
| 936 |
+
"learning_rate": 4.3791517929299466e-05,
|
| 937 |
+
"loss": 0.8065,
|
| 938 |
+
"step": 1550
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.46,
|
| 942 |
+
"learning_rate": 4.3714549689197265e-05,
|
| 943 |
+
"loss": 0.792,
|
| 944 |
+
"step": 1560
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.47,
|
| 948 |
+
"learning_rate": 4.363717582084428e-05,
|
| 949 |
+
"loss": 0.7965,
|
| 950 |
+
"step": 1570
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.47,
|
| 954 |
+
"learning_rate": 4.355939800127937e-05,
|
| 955 |
+
"loss": 0.8038,
|
| 956 |
+
"step": 1580
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.47,
|
| 960 |
+
"learning_rate": 4.34812179162968e-05,
|
| 961 |
+
"loss": 0.7816,
|
| 962 |
+
"step": 1590
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.47,
|
| 966 |
+
"learning_rate": 4.340263726040976e-05,
|
| 967 |
+
"loss": 0.8053,
|
| 968 |
+
"step": 1600
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.48,
|
| 972 |
+
"learning_rate": 4.3323657736813614e-05,
|
| 973 |
+
"loss": 0.8009,
|
| 974 |
+
"step": 1610
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.48,
|
| 978 |
+
"learning_rate": 4.324428105734893e-05,
|
| 979 |
+
"loss": 0.8003,
|
| 980 |
+
"step": 1620
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.48,
|
| 984 |
+
"learning_rate": 4.3164508942464495e-05,
|
| 985 |
+
"loss": 0.7973,
|
| 986 |
+
"step": 1630
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.49,
|
| 990 |
+
"learning_rate": 4.3084343121179896e-05,
|
| 991 |
+
"loss": 0.7851,
|
| 992 |
+
"step": 1640
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.49,
|
| 996 |
+
"learning_rate": 4.300378533104814e-05,
|
| 997 |
+
"loss": 0.7982,
|
| 998 |
+
"step": 1650
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.49,
|
| 1002 |
+
"learning_rate": 4.292283731811794e-05,
|
| 1003 |
+
"loss": 0.797,
|
| 1004 |
+
"step": 1660
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.49,
|
| 1008 |
+
"learning_rate": 4.284150083689589e-05,
|
| 1009 |
+
"loss": 0.7974,
|
| 1010 |
+
"step": 1670
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5,
|
| 1014 |
+
"learning_rate": 4.275977765030845e-05,
|
| 1015 |
+
"loss": 0.79,
|
| 1016 |
+
"step": 1680
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.5,
|
| 1020 |
+
"learning_rate": 4.267766952966369e-05,
|
| 1021 |
+
"loss": 0.7997,
|
| 1022 |
+
"step": 1690
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.5,
|
| 1026 |
+
"learning_rate": 4.259517825461296e-05,
|
| 1027 |
+
"loss": 0.7949,
|
| 1028 |
+
"step": 1700
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.51,
|
| 1032 |
+
"learning_rate": 4.2512305613112266e-05,
|
| 1033 |
+
"loss": 0.8015,
|
| 1034 |
+
"step": 1710
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.51,
|
| 1038 |
+
"learning_rate": 4.2429053401383545e-05,
|
| 1039 |
+
"loss": 0.7989,
|
| 1040 |
+
"step": 1720
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.51,
|
| 1044 |
+
"learning_rate": 4.234542342387571e-05,
|
| 1045 |
+
"loss": 0.7897,
|
| 1046 |
+
"step": 1730
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.52,
|
| 1050 |
+
"learning_rate": 4.2261417493225565e-05,
|
| 1051 |
+
"loss": 0.795,
|
| 1052 |
+
"step": 1740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.52,
|
| 1056 |
+
"learning_rate": 4.2177037430218504e-05,
|
| 1057 |
+
"loss": 0.8037,
|
| 1058 |
+
"step": 1750
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.52,
|
| 1062 |
+
"learning_rate": 4.209228506374906e-05,
|
| 1063 |
+
"loss": 0.7814,
|
| 1064 |
+
"step": 1760
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 0.52,
|
| 1068 |
+
"learning_rate": 4.200716223078121e-05,
|
| 1069 |
+
"loss": 0.7957,
|
| 1070 |
+
"step": 1770
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.53,
|
| 1074 |
+
"learning_rate": 4.192167077630865e-05,
|
| 1075 |
+
"loss": 0.7884,
|
| 1076 |
+
"step": 1780
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 0.53,
|
| 1080 |
+
"learning_rate": 4.183581255331475e-05,
|
| 1081 |
+
"loss": 0.8069,
|
| 1082 |
+
"step": 1790
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.53,
|
| 1086 |
+
"learning_rate": 4.174958942273235e-05,
|
| 1087 |
+
"loss": 0.7946,
|
| 1088 |
+
"step": 1800
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.54,
|
| 1092 |
+
"learning_rate": 4.166300325340355e-05,
|
| 1093 |
+
"loss": 0.8016,
|
| 1094 |
+
"step": 1810
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.54,
|
| 1098 |
+
"learning_rate": 4.157605592203908e-05,
|
| 1099 |
+
"loss": 0.7806,
|
| 1100 |
+
"step": 1820
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.54,
|
| 1104 |
+
"learning_rate": 4.148874931317767e-05,
|
| 1105 |
+
"loss": 0.7786,
|
| 1106 |
+
"step": 1830
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 0.55,
|
| 1110 |
+
"learning_rate": 4.140108531914522e-05,
|
| 1111 |
+
"loss": 0.7817,
|
| 1112 |
+
"step": 1840
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.55,
|
| 1116 |
+
"learning_rate": 4.1313065840013775e-05,
|
| 1117 |
+
"loss": 0.7922,
|
| 1118 |
+
"step": 1850
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.55,
|
| 1122 |
+
"learning_rate": 4.1224692783560333e-05,
|
| 1123 |
+
"loss": 0.7761,
|
| 1124 |
+
"step": 1860
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.55,
|
| 1128 |
+
"learning_rate": 4.113596806522548e-05,
|
| 1129 |
+
"loss": 0.7841,
|
| 1130 |
+
"step": 1870
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.56,
|
| 1134 |
+
"learning_rate": 4.104689360807193e-05,
|
| 1135 |
+
"loss": 0.7983,
|
| 1136 |
+
"step": 1880
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.56,
|
| 1140 |
+
"learning_rate": 4.095747134274276e-05,
|
| 1141 |
+
"loss": 0.7753,
|
| 1142 |
+
"step": 1890
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.56,
|
| 1146 |
+
"learning_rate": 4.086770320741965e-05,
|
| 1147 |
+
"loss": 0.7905,
|
| 1148 |
+
"step": 1900
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 0.57,
|
| 1152 |
+
"learning_rate": 4.077759114778081e-05,
|
| 1153 |
+
"loss": 0.7901,
|
| 1154 |
+
"step": 1910
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.57,
|
| 1158 |
+
"learning_rate": 4.068713711695885e-05,
|
| 1159 |
+
"loss": 0.7834,
|
| 1160 |
+
"step": 1920
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.57,
|
| 1164 |
+
"learning_rate": 4.059634307549841e-05,
|
| 1165 |
+
"loss": 0.7752,
|
| 1166 |
+
"step": 1930
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.57,
|
| 1170 |
+
"learning_rate": 4.0505210991313714e-05,
|
| 1171 |
+
"loss": 0.7905,
|
| 1172 |
+
"step": 1940
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.58,
|
| 1176 |
+
"learning_rate": 4.0413742839645855e-05,
|
| 1177 |
+
"loss": 0.7824,
|
| 1178 |
+
"step": 1950
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.58,
|
| 1182 |
+
"learning_rate": 4.032194060302005e-05,
|
| 1183 |
+
"loss": 0.793,
|
| 1184 |
+
"step": 1960
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.58,
|
| 1188 |
+
"learning_rate": 4.02298062712026e-05,
|
| 1189 |
+
"loss": 0.7911,
|
| 1190 |
+
"step": 1970
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.59,
|
| 1194 |
+
"learning_rate": 4.013734184115783e-05,
|
| 1195 |
+
"loss": 0.7927,
|
| 1196 |
+
"step": 1980
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.59,
|
| 1200 |
+
"learning_rate": 4.004454931700475e-05,
|
| 1201 |
+
"loss": 0.7827,
|
| 1202 |
+
"step": 1990
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.59,
|
| 1206 |
+
"learning_rate": 3.995143070997367e-05,
|
| 1207 |
+
"loss": 0.785,
|
| 1208 |
+
"step": 2000
|
| 1209 |
+
}
|
| 1210 |
+
],
|
| 1211 |
+
"max_steps": 6748,
|
| 1212 |
+
"num_train_epochs": 2,
|
| 1213 |
+
"total_flos": 3.5557245451365253e+18,
|
| 1214 |
+
"trial_name": null,
|
| 1215 |
+
"trial_params": null
|
| 1216 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f710e8a04dbb33457b5b725adb1da191a7b05d7ab01f6ec60c73386df9d8286
|
| 3 |
+
size 3277
|
checkpoint-3000/README.md
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- load_in_8bit: False
|
| 9 |
+
- load_in_4bit: True
|
| 10 |
+
- llm_int8_threshold: 6.0
|
| 11 |
+
- llm_int8_skip_modules: None
|
| 12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
+
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: float16
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.4.0
|
checkpoint-3000/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./Llama-2-7b-chat-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 32.0,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-3000/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e109a32722191825f7fd37340782fa97f952ac78cb7b0cf0af5dea4cc2def675
|
| 3 |
+
size 16821197
|
checkpoint-3000/finetuning_args.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dpo_beta": 0.1,
|
| 3 |
+
"finetuning_type": "lora",
|
| 4 |
+
"lora_alpha": 32.0,
|
| 5 |
+
"lora_dropout": 0.1,
|
| 6 |
+
"lora_rank": 8,
|
| 7 |
+
"lora_target": [
|
| 8 |
+
"q_proj",
|
| 9 |
+
"v_proj"
|
| 10 |
+
],
|
| 11 |
+
"name_module_trainable": "mlp",
|
| 12 |
+
"num_hidden_layers": 32,
|
| 13 |
+
"num_layer_trainable": 3,
|
| 14 |
+
"ppo_score_norm": false,
|
| 15 |
+
"resume_lora_training": true
|
| 16 |
+
}
|
checkpoint-3000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:327e7ccbef91d3577c56e4b964ba7122360817d77251e9f188ade4e19d0f9d81
|
| 3 |
+
size 33661637
|
checkpoint-3000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6ffd433067ffc3c9ba6197fe7e985e7dbf5bf4c45d16aebc15c0d6c1592dffbd
|
| 3 |
+
size 18679
|
checkpoint-3000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:314d9ccd56b7cb0d45f74255469e1f371307e223188ecf89dad105bc342ab3f2
|
| 3 |
+
size 18679
|
checkpoint-3000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:909d6a7009763e37015c47c0f211343f0c6c4bd01582e874b492c5102c9fe1b3
|
| 3 |
+
size 18679
|
checkpoint-3000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec6030f4053fdeb7a01f6f3c018d7de200dc3716f2199ce7205263f4fe56bb62
|
| 3 |
+
size 627
|
checkpoint-3000/trainer_state.json
ADDED
|
@@ -0,0 +1,1816 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.8889547373879547,
|
| 5 |
+
"global_step": 3000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 4.9999729068921297e-05,
|
| 13 |
+
"loss": 1.369,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 4.9998916281557476e-05,
|
| 19 |
+
"loss": 1.1704,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 4.999772150012383e-05,
|
| 25 |
+
"loss": 1.131,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
+
"learning_rate": 4.99958792440589e-05,
|
| 31 |
+
"loss": 1.0618,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.01,
|
| 36 |
+
"learning_rate": 4.999401536044942e-05,
|
| 37 |
+
"loss": 1.0539,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
+
"learning_rate": 4.9991197949909295e-05,
|
| 43 |
+
"loss": 1.0157,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"learning_rate": 4.998783886799168e-05,
|
| 49 |
+
"loss": 1.0235,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.02,
|
| 54 |
+
"learning_rate": 4.998393818750294e-05,
|
| 55 |
+
"loss": 0.9841,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
+
"learning_rate": 4.9979495992988316e-05,
|
| 61 |
+
"loss": 0.9787,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.03,
|
| 66 |
+
"learning_rate": 4.997451238073011e-05,
|
| 67 |
+
"loss": 0.9716,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.03,
|
| 72 |
+
"learning_rate": 4.996898745874553e-05,
|
| 73 |
+
"loss": 0.9718,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
+
"learning_rate": 4.996292134678445e-05,
|
| 79 |
+
"loss": 0.9608,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04,
|
| 84 |
+
"learning_rate": 4.995631417632672e-05,
|
| 85 |
+
"loss": 0.9547,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04,
|
| 90 |
+
"learning_rate": 4.994916609057936e-05,
|
| 91 |
+
"loss": 0.959,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.04,
|
| 96 |
+
"learning_rate": 4.9941477244473455e-05,
|
| 97 |
+
"loss": 0.9297,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.05,
|
| 102 |
+
"learning_rate": 4.993324780466081e-05,
|
| 103 |
+
"loss": 0.927,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.05,
|
| 108 |
+
"learning_rate": 4.992447794951029e-05,
|
| 109 |
+
"loss": 0.9352,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.05,
|
| 114 |
+
"learning_rate": 4.9915167869104006e-05,
|
| 115 |
+
"loss": 0.9244,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.06,
|
| 120 |
+
"learning_rate": 4.990531776523317e-05,
|
| 121 |
+
"loss": 0.9185,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06,
|
| 126 |
+
"learning_rate": 4.9894927851393726e-05,
|
| 127 |
+
"loss": 0.9201,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06,
|
| 132 |
+
"learning_rate": 4.988399835278171e-05,
|
| 133 |
+
"loss": 0.9053,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.07,
|
| 138 |
+
"learning_rate": 4.987252950628839e-05,
|
| 139 |
+
"loss": 0.912,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.07,
|
| 144 |
+
"learning_rate": 4.986052156049513e-05,
|
| 145 |
+
"loss": 0.9013,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.07,
|
| 150 |
+
"learning_rate": 4.984797477566799e-05,
|
| 151 |
+
"loss": 0.8942,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.07,
|
| 156 |
+
"learning_rate": 4.9834889423752074e-05,
|
| 157 |
+
"loss": 0.8873,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.08,
|
| 162 |
+
"learning_rate": 4.9821265788365664e-05,
|
| 163 |
+
"loss": 0.9036,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08,
|
| 168 |
+
"learning_rate": 4.980710416479407e-05,
|
| 169 |
+
"loss": 0.901,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08,
|
| 174 |
+
"learning_rate": 4.97924048599832e-05,
|
| 175 |
+
"loss": 0.8941,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.09,
|
| 180 |
+
"learning_rate": 4.977716819253293e-05,
|
| 181 |
+
"loss": 0.8886,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.09,
|
| 186 |
+
"learning_rate": 4.97613944926902e-05,
|
| 187 |
+
"loss": 0.8852,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.09,
|
| 192 |
+
"learning_rate": 4.974508410234187e-05,
|
| 193 |
+
"loss": 0.8877,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.09,
|
| 198 |
+
"learning_rate": 4.972823737500725e-05,
|
| 199 |
+
"loss": 0.8833,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.1,
|
| 204 |
+
"learning_rate": 4.971085467583051e-05,
|
| 205 |
+
"loss": 0.869,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1,
|
| 210 |
+
"learning_rate": 4.969293638157273e-05,
|
| 211 |
+
"loss": 0.877,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1,
|
| 216 |
+
"learning_rate": 4.967448288060372e-05,
|
| 217 |
+
"loss": 0.8793,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.11,
|
| 222 |
+
"learning_rate": 4.9655494572893644e-05,
|
| 223 |
+
"loss": 0.8742,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.11,
|
| 228 |
+
"learning_rate": 4.9635971870004316e-05,
|
| 229 |
+
"loss": 0.8774,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.11,
|
| 234 |
+
"learning_rate": 4.961591519508029e-05,
|
| 235 |
+
"loss": 0.8682,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.12,
|
| 240 |
+
"learning_rate": 4.9595324982839686e-05,
|
| 241 |
+
"loss": 0.8581,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.12,
|
| 246 |
+
"learning_rate": 4.957420167956479e-05,
|
| 247 |
+
"loss": 0.868,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.12,
|
| 252 |
+
"learning_rate": 4.955254574309234e-05,
|
| 253 |
+
"loss": 0.8672,
|
| 254 |
+
"step": 410
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12,
|
| 258 |
+
"learning_rate": 4.9530357642803646e-05,
|
| 259 |
+
"loss": 0.8672,
|
| 260 |
+
"step": 420
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.13,
|
| 264 |
+
"learning_rate": 4.950763785961436e-05,
|
| 265 |
+
"loss": 0.867,
|
| 266 |
+
"step": 430
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.13,
|
| 270 |
+
"learning_rate": 4.948438688596414e-05,
|
| 271 |
+
"loss": 0.8612,
|
| 272 |
+
"step": 440
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.13,
|
| 276 |
+
"learning_rate": 4.946060522580588e-05,
|
| 277 |
+
"loss": 0.8607,
|
| 278 |
+
"step": 450
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.14,
|
| 282 |
+
"learning_rate": 4.943629339459484e-05,
|
| 283 |
+
"loss": 0.8588,
|
| 284 |
+
"step": 460
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.14,
|
| 288 |
+
"learning_rate": 4.9411451919277484e-05,
|
| 289 |
+
"loss": 0.8585,
|
| 290 |
+
"step": 470
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.14,
|
| 294 |
+
"learning_rate": 4.9386081338280035e-05,
|
| 295 |
+
"loss": 0.8574,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"learning_rate": 4.9360182201496785e-05,
|
| 301 |
+
"loss": 0.8562,
|
| 302 |
+
"step": 490
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.15,
|
| 306 |
+
"learning_rate": 4.933375507027824e-05,
|
| 307 |
+
"loss": 0.8634,
|
| 308 |
+
"step": 500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.15,
|
| 312 |
+
"learning_rate": 4.930680051741887e-05,
|
| 313 |
+
"loss": 0.8483,
|
| 314 |
+
"step": 510
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.15,
|
| 318 |
+
"learning_rate": 4.9279319127144775e-05,
|
| 319 |
+
"loss": 0.8559,
|
| 320 |
+
"step": 520
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.16,
|
| 324 |
+
"learning_rate": 4.9251311495100974e-05,
|
| 325 |
+
"loss": 0.8526,
|
| 326 |
+
"step": 530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.16,
|
| 330 |
+
"learning_rate": 4.9222778228338506e-05,
|
| 331 |
+
"loss": 0.85,
|
| 332 |
+
"step": 540
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.16,
|
| 336 |
+
"learning_rate": 4.919371994530125e-05,
|
| 337 |
+
"loss": 0.8547,
|
| 338 |
+
"step": 550
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17,
|
| 342 |
+
"learning_rate": 4.9164137275812597e-05,
|
| 343 |
+
"loss": 0.8533,
|
| 344 |
+
"step": 560
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.17,
|
| 348 |
+
"learning_rate": 4.913403086106167e-05,
|
| 349 |
+
"loss": 0.8431,
|
| 350 |
+
"step": 570
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.17,
|
| 354 |
+
"learning_rate": 4.9103401353589585e-05,
|
| 355 |
+
"loss": 0.8346,
|
| 356 |
+
"step": 580
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.17,
|
| 360 |
+
"learning_rate": 4.9072249417275164e-05,
|
| 361 |
+
"loss": 0.8448,
|
| 362 |
+
"step": 590
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.18,
|
| 366 |
+
"learning_rate": 4.9040575727320606e-05,
|
| 367 |
+
"loss": 0.8477,
|
| 368 |
+
"step": 600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.18,
|
| 372 |
+
"learning_rate": 4.90083809702369e-05,
|
| 373 |
+
"loss": 0.8555,
|
| 374 |
+
"step": 610
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.18,
|
| 378 |
+
"learning_rate": 4.897566584382884e-05,
|
| 379 |
+
"loss": 0.8295,
|
| 380 |
+
"step": 620
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19,
|
| 384 |
+
"learning_rate": 4.8942431057180015e-05,
|
| 385 |
+
"loss": 0.8359,
|
| 386 |
+
"step": 630
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.19,
|
| 390 |
+
"learning_rate": 4.890867733063733e-05,
|
| 391 |
+
"loss": 0.8521,
|
| 392 |
+
"step": 640
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.19,
|
| 396 |
+
"learning_rate": 4.887440539579547e-05,
|
| 397 |
+
"loss": 0.8349,
|
| 398 |
+
"step": 650
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.2,
|
| 402 |
+
"learning_rate": 4.883961599548103e-05,
|
| 403 |
+
"loss": 0.8285,
|
| 404 |
+
"step": 660
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.2,
|
| 408 |
+
"learning_rate": 4.8804309883736377e-05,
|
| 409 |
+
"loss": 0.831,
|
| 410 |
+
"step": 670
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.2,
|
| 414 |
+
"learning_rate": 4.876848782580336e-05,
|
| 415 |
+
"loss": 0.8295,
|
| 416 |
+
"step": 680
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2,
|
| 420 |
+
"learning_rate": 4.873215059810666e-05,
|
| 421 |
+
"loss": 0.8406,
|
| 422 |
+
"step": 690
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21,
|
| 426 |
+
"learning_rate": 4.8695298988237035e-05,
|
| 427 |
+
"loss": 0.8456,
|
| 428 |
+
"step": 700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.21,
|
| 432 |
+
"learning_rate": 4.8657933794934205e-05,
|
| 433 |
+
"loss": 0.8218,
|
| 434 |
+
"step": 710
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.21,
|
| 438 |
+
"learning_rate": 4.8620055828069535e-05,
|
| 439 |
+
"loss": 0.8424,
|
| 440 |
+
"step": 720
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.22,
|
| 444 |
+
"learning_rate": 4.858166590862849e-05,
|
| 445 |
+
"loss": 0.8258,
|
| 446 |
+
"step": 730
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.22,
|
| 450 |
+
"learning_rate": 4.854276486869287e-05,
|
| 451 |
+
"loss": 0.8352,
|
| 452 |
+
"step": 740
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.22,
|
| 456 |
+
"learning_rate": 4.850335355142272e-05,
|
| 457 |
+
"loss": 0.836,
|
| 458 |
+
"step": 750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.23,
|
| 462 |
+
"learning_rate": 4.846343281103809e-05,
|
| 463 |
+
"loss": 0.8388,
|
| 464 |
+
"step": 760
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.23,
|
| 468 |
+
"learning_rate": 4.842300351280054e-05,
|
| 469 |
+
"loss": 0.8287,
|
| 470 |
+
"step": 770
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.23,
|
| 474 |
+
"learning_rate": 4.8382066532994316e-05,
|
| 475 |
+
"loss": 0.8394,
|
| 476 |
+
"step": 780
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.23,
|
| 480 |
+
"learning_rate": 4.834062275890745e-05,
|
| 481 |
+
"loss": 0.8253,
|
| 482 |
+
"step": 790
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.24,
|
| 486 |
+
"learning_rate": 4.829867308881244e-05,
|
| 487 |
+
"loss": 0.82,
|
| 488 |
+
"step": 800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.24,
|
| 492 |
+
"learning_rate": 4.825621843194684e-05,
|
| 493 |
+
"loss": 0.8294,
|
| 494 |
+
"step": 810
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.24,
|
| 498 |
+
"learning_rate": 4.821325970849353e-05,
|
| 499 |
+
"loss": 0.8275,
|
| 500 |
+
"step": 820
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.25,
|
| 504 |
+
"learning_rate": 4.8169797849560764e-05,
|
| 505 |
+
"loss": 0.8129,
|
| 506 |
+
"step": 830
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.25,
|
| 510 |
+
"learning_rate": 4.812583379716202e-05,
|
| 511 |
+
"loss": 0.8215,
|
| 512 |
+
"step": 840
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.25,
|
| 516 |
+
"learning_rate": 4.8081368504195545e-05,
|
| 517 |
+
"loss": 0.84,
|
| 518 |
+
"step": 850
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.25,
|
| 522 |
+
"learning_rate": 4.803640293442373e-05,
|
| 523 |
+
"loss": 0.8246,
|
| 524 |
+
"step": 860
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.26,
|
| 528 |
+
"learning_rate": 4.799093806245218e-05,
|
| 529 |
+
"loss": 0.8179,
|
| 530 |
+
"step": 870
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.26,
|
| 534 |
+
"learning_rate": 4.7944974873708656e-05,
|
| 535 |
+
"loss": 0.8238,
|
| 536 |
+
"step": 880
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.26,
|
| 540 |
+
"learning_rate": 4.7898514364421655e-05,
|
| 541 |
+
"loss": 0.8327,
|
| 542 |
+
"step": 890
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.27,
|
| 546 |
+
"learning_rate": 4.785155754159886e-05,
|
| 547 |
+
"loss": 0.812,
|
| 548 |
+
"step": 900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.27,
|
| 552 |
+
"learning_rate": 4.780410542300527e-05,
|
| 553 |
+
"loss": 0.8293,
|
| 554 |
+
"step": 910
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.27,
|
| 558 |
+
"learning_rate": 4.7756159037141185e-05,
|
| 559 |
+
"loss": 0.8266,
|
| 560 |
+
"step": 920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.28,
|
| 564 |
+
"learning_rate": 4.7707719423219886e-05,
|
| 565 |
+
"loss": 0.8209,
|
| 566 |
+
"step": 930
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.28,
|
| 570 |
+
"learning_rate": 4.765878763114513e-05,
|
| 571 |
+
"loss": 0.8182,
|
| 572 |
+
"step": 940
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.28,
|
| 576 |
+
"learning_rate": 4.760936472148837e-05,
|
| 577 |
+
"loss": 0.8045,
|
| 578 |
+
"step": 950
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.28,
|
| 582 |
+
"learning_rate": 4.755945176546578e-05,
|
| 583 |
+
"loss": 0.8126,
|
| 584 |
+
"step": 960
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.29,
|
| 588 |
+
"learning_rate": 4.7509049844915036e-05,
|
| 589 |
+
"loss": 0.8201,
|
| 590 |
+
"step": 970
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.29,
|
| 594 |
+
"learning_rate": 4.7458160052271883e-05,
|
| 595 |
+
"loss": 0.8133,
|
| 596 |
+
"step": 980
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.29,
|
| 600 |
+
"learning_rate": 4.7406783490546434e-05,
|
| 601 |
+
"loss": 0.8157,
|
| 602 |
+
"step": 990
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.3,
|
| 606 |
+
"learning_rate": 4.735492127329928e-05,
|
| 607 |
+
"loss": 0.817,
|
| 608 |
+
"step": 1000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.3,
|
| 612 |
+
"learning_rate": 4.7302574524617315e-05,
|
| 613 |
+
"loss": 0.8002,
|
| 614 |
+
"step": 1010
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.3,
|
| 618 |
+
"learning_rate": 4.724974437908944e-05,
|
| 619 |
+
"loss": 0.8281,
|
| 620 |
+
"step": 1020
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.31,
|
| 624 |
+
"learning_rate": 4.719643198178193e-05,
|
| 625 |
+
"loss": 0.8129,
|
| 626 |
+
"step": 1030
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.31,
|
| 630 |
+
"learning_rate": 4.71426384882136e-05,
|
| 631 |
+
"loss": 0.8125,
|
| 632 |
+
"step": 1040
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.31,
|
| 636 |
+
"learning_rate": 4.708836506433078e-05,
|
| 637 |
+
"loss": 0.8066,
|
| 638 |
+
"step": 1050
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.31,
|
| 642 |
+
"learning_rate": 4.7033612886482066e-05,
|
| 643 |
+
"loss": 0.8159,
|
| 644 |
+
"step": 1060
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.32,
|
| 648 |
+
"learning_rate": 4.697838314139278e-05,
|
| 649 |
+
"loss": 0.8288,
|
| 650 |
+
"step": 1070
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.32,
|
| 654 |
+
"learning_rate": 4.6922677026139275e-05,
|
| 655 |
+
"loss": 0.8175,
|
| 656 |
+
"step": 1080
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.32,
|
| 660 |
+
"learning_rate": 4.686649574812299e-05,
|
| 661 |
+
"loss": 0.8107,
|
| 662 |
+
"step": 1090
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.33,
|
| 666 |
+
"learning_rate": 4.680984052504426e-05,
|
| 667 |
+
"loss": 0.8127,
|
| 668 |
+
"step": 1100
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.33,
|
| 672 |
+
"learning_rate": 4.6752712584875935e-05,
|
| 673 |
+
"loss": 0.8023,
|
| 674 |
+
"step": 1110
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.33,
|
| 678 |
+
"learning_rate": 4.669511316583678e-05,
|
| 679 |
+
"loss": 0.8066,
|
| 680 |
+
"step": 1120
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.33,
|
| 684 |
+
"learning_rate": 4.663704351636461e-05,
|
| 685 |
+
"loss": 0.8069,
|
| 686 |
+
"step": 1130
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.34,
|
| 690 |
+
"learning_rate": 4.6578504895089245e-05,
|
| 691 |
+
"loss": 0.8129,
|
| 692 |
+
"step": 1140
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.34,
|
| 696 |
+
"learning_rate": 4.651949857080524e-05,
|
| 697 |
+
"loss": 0.8247,
|
| 698 |
+
"step": 1150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.34,
|
| 702 |
+
"learning_rate": 4.6460025822444354e-05,
|
| 703 |
+
"loss": 0.8125,
|
| 704 |
+
"step": 1160
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.35,
|
| 708 |
+
"learning_rate": 4.640008793904785e-05,
|
| 709 |
+
"loss": 0.8141,
|
| 710 |
+
"step": 1170
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.35,
|
| 714 |
+
"learning_rate": 4.6339686219738576e-05,
|
| 715 |
+
"loss": 0.8066,
|
| 716 |
+
"step": 1180
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.35,
|
| 720 |
+
"learning_rate": 4.627882197369276e-05,
|
| 721 |
+
"loss": 0.8052,
|
| 722 |
+
"step": 1190
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.36,
|
| 726 |
+
"learning_rate": 4.621749652011167e-05,
|
| 727 |
+
"loss": 0.8114,
|
| 728 |
+
"step": 1200
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.36,
|
| 732 |
+
"learning_rate": 4.615571118819301e-05,
|
| 733 |
+
"loss": 0.8001,
|
| 734 |
+
"step": 1210
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.36,
|
| 738 |
+
"learning_rate": 4.60934673171021e-05,
|
| 739 |
+
"loss": 0.7938,
|
| 740 |
+
"step": 1220
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.36,
|
| 744 |
+
"learning_rate": 4.6030766255942884e-05,
|
| 745 |
+
"loss": 0.8107,
|
| 746 |
+
"step": 1230
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.37,
|
| 750 |
+
"learning_rate": 4.596760936372865e-05,
|
| 751 |
+
"loss": 0.8096,
|
| 752 |
+
"step": 1240
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.37,
|
| 756 |
+
"learning_rate": 4.5903998009352586e-05,
|
| 757 |
+
"loss": 0.8105,
|
| 758 |
+
"step": 1250
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.37,
|
| 762 |
+
"learning_rate": 4.583993357155813e-05,
|
| 763 |
+
"loss": 0.7995,
|
| 764 |
+
"step": 1260
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.38,
|
| 768 |
+
"learning_rate": 4.577541743890906e-05,
|
| 769 |
+
"loss": 0.8161,
|
| 770 |
+
"step": 1270
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.38,
|
| 774 |
+
"learning_rate": 4.571045100975941e-05,
|
| 775 |
+
"loss": 0.8003,
|
| 776 |
+
"step": 1280
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.38,
|
| 780 |
+
"learning_rate": 4.564503569222315e-05,
|
| 781 |
+
"loss": 0.8114,
|
| 782 |
+
"step": 1290
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.39,
|
| 786 |
+
"learning_rate": 4.5579172904143685e-05,
|
| 787 |
+
"loss": 0.8165,
|
| 788 |
+
"step": 1300
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.39,
|
| 792 |
+
"learning_rate": 4.551286407306312e-05,
|
| 793 |
+
"loss": 0.8022,
|
| 794 |
+
"step": 1310
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.39,
|
| 798 |
+
"learning_rate": 4.54461106361913e-05,
|
| 799 |
+
"loss": 0.8032,
|
| 800 |
+
"step": 1320
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.39,
|
| 804 |
+
"learning_rate": 4.5378914040374684e-05,
|
| 805 |
+
"loss": 0.8067,
|
| 806 |
+
"step": 1330
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.4,
|
| 810 |
+
"learning_rate": 4.531127574206496e-05,
|
| 811 |
+
"loss": 0.789,
|
| 812 |
+
"step": 1340
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.4,
|
| 816 |
+
"learning_rate": 4.524319720728749e-05,
|
| 817 |
+
"loss": 0.8139,
|
| 818 |
+
"step": 1350
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.4,
|
| 822 |
+
"learning_rate": 4.517467991160956e-05,
|
| 823 |
+
"loss": 0.805,
|
| 824 |
+
"step": 1360
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.41,
|
| 828 |
+
"learning_rate": 4.510572534010836e-05,
|
| 829 |
+
"loss": 0.8123,
|
| 830 |
+
"step": 1370
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.41,
|
| 834 |
+
"learning_rate": 4.5036334987338785e-05,
|
| 835 |
+
"loss": 0.7976,
|
| 836 |
+
"step": 1380
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.41,
|
| 840 |
+
"learning_rate": 4.4966510357301094e-05,
|
| 841 |
+
"loss": 0.813,
|
| 842 |
+
"step": 1390
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.41,
|
| 846 |
+
"learning_rate": 4.489625296340828e-05,
|
| 847 |
+
"loss": 0.7975,
|
| 848 |
+
"step": 1400
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.42,
|
| 852 |
+
"learning_rate": 4.4825564328453256e-05,
|
| 853 |
+
"loss": 0.7969,
|
| 854 |
+
"step": 1410
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.42,
|
| 858 |
+
"learning_rate": 4.4754445984575874e-05,
|
| 859 |
+
"loss": 0.8006,
|
| 860 |
+
"step": 1420
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.42,
|
| 864 |
+
"learning_rate": 4.4682899473229714e-05,
|
| 865 |
+
"loss": 0.8041,
|
| 866 |
+
"step": 1430
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.43,
|
| 870 |
+
"learning_rate": 4.461092634514863e-05,
|
| 871 |
+
"loss": 0.7998,
|
| 872 |
+
"step": 1440
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.43,
|
| 876 |
+
"learning_rate": 4.453852816031323e-05,
|
| 877 |
+
"loss": 0.8,
|
| 878 |
+
"step": 1450
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.43,
|
| 882 |
+
"learning_rate": 4.446570648791695e-05,
|
| 883 |
+
"loss": 0.7993,
|
| 884 |
+
"step": 1460
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.44,
|
| 888 |
+
"learning_rate": 4.439246290633216e-05,
|
| 889 |
+
"loss": 0.8113,
|
| 890 |
+
"step": 1470
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.44,
|
| 894 |
+
"learning_rate": 4.431879900307585e-05,
|
| 895 |
+
"loss": 0.8133,
|
| 896 |
+
"step": 1480
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.44,
|
| 900 |
+
"learning_rate": 4.4244716374775274e-05,
|
| 901 |
+
"loss": 0.7963,
|
| 902 |
+
"step": 1490
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.44,
|
| 906 |
+
"learning_rate": 4.4170216627133365e-05,
|
| 907 |
+
"loss": 0.8005,
|
| 908 |
+
"step": 1500
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.45,
|
| 912 |
+
"learning_rate": 4.409530137489386e-05,
|
| 913 |
+
"loss": 0.8016,
|
| 914 |
+
"step": 1510
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.45,
|
| 918 |
+
"learning_rate": 4.4019972241806384e-05,
|
| 919 |
+
"loss": 0.7858,
|
| 920 |
+
"step": 1520
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.45,
|
| 924 |
+
"learning_rate": 4.3944230860591194e-05,
|
| 925 |
+
"loss": 0.8009,
|
| 926 |
+
"step": 1530
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.46,
|
| 930 |
+
"learning_rate": 4.3868078872903814e-05,
|
| 931 |
+
"loss": 0.7929,
|
| 932 |
+
"step": 1540
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.46,
|
| 936 |
+
"learning_rate": 4.3791517929299466e-05,
|
| 937 |
+
"loss": 0.8065,
|
| 938 |
+
"step": 1550
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.46,
|
| 942 |
+
"learning_rate": 4.3714549689197265e-05,
|
| 943 |
+
"loss": 0.792,
|
| 944 |
+
"step": 1560
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.47,
|
| 948 |
+
"learning_rate": 4.363717582084428e-05,
|
| 949 |
+
"loss": 0.7965,
|
| 950 |
+
"step": 1570
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.47,
|
| 954 |
+
"learning_rate": 4.355939800127937e-05,
|
| 955 |
+
"loss": 0.8038,
|
| 956 |
+
"step": 1580
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.47,
|
| 960 |
+
"learning_rate": 4.34812179162968e-05,
|
| 961 |
+
"loss": 0.7816,
|
| 962 |
+
"step": 1590
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.47,
|
| 966 |
+
"learning_rate": 4.340263726040976e-05,
|
| 967 |
+
"loss": 0.8053,
|
| 968 |
+
"step": 1600
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.48,
|
| 972 |
+
"learning_rate": 4.3323657736813614e-05,
|
| 973 |
+
"loss": 0.8009,
|
| 974 |
+
"step": 1610
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.48,
|
| 978 |
+
"learning_rate": 4.324428105734893e-05,
|
| 979 |
+
"loss": 0.8003,
|
| 980 |
+
"step": 1620
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.48,
|
| 984 |
+
"learning_rate": 4.3164508942464495e-05,
|
| 985 |
+
"loss": 0.7973,
|
| 986 |
+
"step": 1630
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.49,
|
| 990 |
+
"learning_rate": 4.3084343121179896e-05,
|
| 991 |
+
"loss": 0.7851,
|
| 992 |
+
"step": 1640
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.49,
|
| 996 |
+
"learning_rate": 4.300378533104814e-05,
|
| 997 |
+
"loss": 0.7982,
|
| 998 |
+
"step": 1650
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.49,
|
| 1002 |
+
"learning_rate": 4.292283731811794e-05,
|
| 1003 |
+
"loss": 0.797,
|
| 1004 |
+
"step": 1660
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.49,
|
| 1008 |
+
"learning_rate": 4.284150083689589e-05,
|
| 1009 |
+
"loss": 0.7974,
|
| 1010 |
+
"step": 1670
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5,
|
| 1014 |
+
"learning_rate": 4.275977765030845e-05,
|
| 1015 |
+
"loss": 0.79,
|
| 1016 |
+
"step": 1680
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.5,
|
| 1020 |
+
"learning_rate": 4.267766952966369e-05,
|
| 1021 |
+
"loss": 0.7997,
|
| 1022 |
+
"step": 1690
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.5,
|
| 1026 |
+
"learning_rate": 4.259517825461296e-05,
|
| 1027 |
+
"loss": 0.7949,
|
| 1028 |
+
"step": 1700
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.51,
|
| 1032 |
+
"learning_rate": 4.2512305613112266e-05,
|
| 1033 |
+
"loss": 0.8015,
|
| 1034 |
+
"step": 1710
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.51,
|
| 1038 |
+
"learning_rate": 4.2429053401383545e-05,
|
| 1039 |
+
"loss": 0.7989,
|
| 1040 |
+
"step": 1720
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.51,
|
| 1044 |
+
"learning_rate": 4.234542342387571e-05,
|
| 1045 |
+
"loss": 0.7897,
|
| 1046 |
+
"step": 1730
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.52,
|
| 1050 |
+
"learning_rate": 4.2261417493225565e-05,
|
| 1051 |
+
"loss": 0.795,
|
| 1052 |
+
"step": 1740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.52,
|
| 1056 |
+
"learning_rate": 4.2177037430218504e-05,
|
| 1057 |
+
"loss": 0.8037,
|
| 1058 |
+
"step": 1750
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.52,
|
| 1062 |
+
"learning_rate": 4.209228506374906e-05,
|
| 1063 |
+
"loss": 0.7814,
|
| 1064 |
+
"step": 1760
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 0.52,
|
| 1068 |
+
"learning_rate": 4.200716223078121e-05,
|
| 1069 |
+
"loss": 0.7957,
|
| 1070 |
+
"step": 1770
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.53,
|
| 1074 |
+
"learning_rate": 4.192167077630865e-05,
|
| 1075 |
+
"loss": 0.7884,
|
| 1076 |
+
"step": 1780
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 0.53,
|
| 1080 |
+
"learning_rate": 4.183581255331475e-05,
|
| 1081 |
+
"loss": 0.8069,
|
| 1082 |
+
"step": 1790
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.53,
|
| 1086 |
+
"learning_rate": 4.174958942273235e-05,
|
| 1087 |
+
"loss": 0.7946,
|
| 1088 |
+
"step": 1800
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.54,
|
| 1092 |
+
"learning_rate": 4.166300325340355e-05,
|
| 1093 |
+
"loss": 0.8016,
|
| 1094 |
+
"step": 1810
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.54,
|
| 1098 |
+
"learning_rate": 4.157605592203908e-05,
|
| 1099 |
+
"loss": 0.7806,
|
| 1100 |
+
"step": 1820
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.54,
|
| 1104 |
+
"learning_rate": 4.148874931317767e-05,
|
| 1105 |
+
"loss": 0.7786,
|
| 1106 |
+
"step": 1830
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 0.55,
|
| 1110 |
+
"learning_rate": 4.140108531914522e-05,
|
| 1111 |
+
"loss": 0.7817,
|
| 1112 |
+
"step": 1840
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.55,
|
| 1116 |
+
"learning_rate": 4.1313065840013775e-05,
|
| 1117 |
+
"loss": 0.7922,
|
| 1118 |
+
"step": 1850
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.55,
|
| 1122 |
+
"learning_rate": 4.1224692783560333e-05,
|
| 1123 |
+
"loss": 0.7761,
|
| 1124 |
+
"step": 1860
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.55,
|
| 1128 |
+
"learning_rate": 4.113596806522548e-05,
|
| 1129 |
+
"loss": 0.7841,
|
| 1130 |
+
"step": 1870
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.56,
|
| 1134 |
+
"learning_rate": 4.104689360807193e-05,
|
| 1135 |
+
"loss": 0.7983,
|
| 1136 |
+
"step": 1880
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.56,
|
| 1140 |
+
"learning_rate": 4.095747134274276e-05,
|
| 1141 |
+
"loss": 0.7753,
|
| 1142 |
+
"step": 1890
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.56,
|
| 1146 |
+
"learning_rate": 4.086770320741965e-05,
|
| 1147 |
+
"loss": 0.7905,
|
| 1148 |
+
"step": 1900
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 0.57,
|
| 1152 |
+
"learning_rate": 4.077759114778081e-05,
|
| 1153 |
+
"loss": 0.7901,
|
| 1154 |
+
"step": 1910
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.57,
|
| 1158 |
+
"learning_rate": 4.068713711695885e-05,
|
| 1159 |
+
"loss": 0.7834,
|
| 1160 |
+
"step": 1920
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.57,
|
| 1164 |
+
"learning_rate": 4.059634307549841e-05,
|
| 1165 |
+
"loss": 0.7752,
|
| 1166 |
+
"step": 1930
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.57,
|
| 1170 |
+
"learning_rate": 4.0505210991313714e-05,
|
| 1171 |
+
"loss": 0.7905,
|
| 1172 |
+
"step": 1940
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.58,
|
| 1176 |
+
"learning_rate": 4.0413742839645855e-05,
|
| 1177 |
+
"loss": 0.7824,
|
| 1178 |
+
"step": 1950
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.58,
|
| 1182 |
+
"learning_rate": 4.032194060302005e-05,
|
| 1183 |
+
"loss": 0.793,
|
| 1184 |
+
"step": 1960
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.58,
|
| 1188 |
+
"learning_rate": 4.02298062712026e-05,
|
| 1189 |
+
"loss": 0.7911,
|
| 1190 |
+
"step": 1970
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.59,
|
| 1194 |
+
"learning_rate": 4.013734184115783e-05,
|
| 1195 |
+
"loss": 0.7927,
|
| 1196 |
+
"step": 1980
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.59,
|
| 1200 |
+
"learning_rate": 4.004454931700475e-05,
|
| 1201 |
+
"loss": 0.7827,
|
| 1202 |
+
"step": 1990
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.59,
|
| 1206 |
+
"learning_rate": 3.995143070997367e-05,
|
| 1207 |
+
"loss": 0.785,
|
| 1208 |
+
"step": 2000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.6,
|
| 1212 |
+
"learning_rate": 3.985798803836256e-05,
|
| 1213 |
+
"loss": 0.7922,
|
| 1214 |
+
"step": 2010
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6,
|
| 1218 |
+
"learning_rate": 3.976422332749331e-05,
|
| 1219 |
+
"loss": 0.7969,
|
| 1220 |
+
"step": 2020
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6,
|
| 1224 |
+
"learning_rate": 3.967013860966787e-05,
|
| 1225 |
+
"loss": 0.7801,
|
| 1226 |
+
"step": 2030
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.6,
|
| 1230 |
+
"learning_rate": 3.957573592412417e-05,
|
| 1231 |
+
"loss": 0.7876,
|
| 1232 |
+
"step": 2040
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 0.61,
|
| 1236 |
+
"learning_rate": 3.948101731699192e-05,
|
| 1237 |
+
"loss": 0.7862,
|
| 1238 |
+
"step": 2050
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.61,
|
| 1242 |
+
"learning_rate": 3.938598484124827e-05,
|
| 1243 |
+
"loss": 0.7855,
|
| 1244 |
+
"step": 2060
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 0.61,
|
| 1248 |
+
"learning_rate": 3.929064055667333e-05,
|
| 1249 |
+
"loss": 0.7917,
|
| 1250 |
+
"step": 2070
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.62,
|
| 1254 |
+
"learning_rate": 3.919498652980546e-05,
|
| 1255 |
+
"loss": 0.7855,
|
| 1256 |
+
"step": 2080
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.62,
|
| 1260 |
+
"learning_rate": 3.909902483389658e-05,
|
| 1261 |
+
"loss": 0.7641,
|
| 1262 |
+
"step": 2090
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.62,
|
| 1266 |
+
"learning_rate": 3.900275754886714e-05,
|
| 1267 |
+
"loss": 0.784,
|
| 1268 |
+
"step": 2100
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.63,
|
| 1272 |
+
"learning_rate": 3.8906186761261095e-05,
|
| 1273 |
+
"loss": 0.7745,
|
| 1274 |
+
"step": 2110
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.63,
|
| 1278 |
+
"learning_rate": 3.880931456420066e-05,
|
| 1279 |
+
"loss": 0.7913,
|
| 1280 |
+
"step": 2120
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.63,
|
| 1284 |
+
"learning_rate": 3.8712143057340935e-05,
|
| 1285 |
+
"loss": 0.7823,
|
| 1286 |
+
"step": 2130
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 0.63,
|
| 1290 |
+
"learning_rate": 3.861467434682442e-05,
|
| 1291 |
+
"loss": 0.7834,
|
| 1292 |
+
"step": 2140
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.64,
|
| 1296 |
+
"learning_rate": 3.851691054523534e-05,
|
| 1297 |
+
"loss": 0.7929,
|
| 1298 |
+
"step": 2150
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.64,
|
| 1302 |
+
"learning_rate": 3.841885377155388e-05,
|
| 1303 |
+
"loss": 0.7751,
|
| 1304 |
+
"step": 2160
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.64,
|
| 1308 |
+
"learning_rate": 3.832050615111023e-05,
|
| 1309 |
+
"loss": 0.7814,
|
| 1310 |
+
"step": 2170
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.65,
|
| 1314 |
+
"learning_rate": 3.8221869815538544e-05,
|
| 1315 |
+
"loss": 0.7833,
|
| 1316 |
+
"step": 2180
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 0.65,
|
| 1320 |
+
"learning_rate": 3.812294690273074e-05,
|
| 1321 |
+
"loss": 0.7829,
|
| 1322 |
+
"step": 2190
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.65,
|
| 1326 |
+
"learning_rate": 3.802373955679011e-05,
|
| 1327 |
+
"loss": 0.773,
|
| 1328 |
+
"step": 2200
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.65,
|
| 1332 |
+
"learning_rate": 3.7924249927984936e-05,
|
| 1333 |
+
"loss": 0.7827,
|
| 1334 |
+
"step": 2210
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.66,
|
| 1338 |
+
"learning_rate": 3.78244801727018e-05,
|
| 1339 |
+
"loss": 0.7863,
|
| 1340 |
+
"step": 2220
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.66,
|
| 1344 |
+
"learning_rate": 3.77244324533989e-05,
|
| 1345 |
+
"loss": 0.7881,
|
| 1346 |
+
"step": 2230
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.66,
|
| 1350 |
+
"learning_rate": 3.762410893855916e-05,
|
| 1351 |
+
"loss": 0.7889,
|
| 1352 |
+
"step": 2240
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.67,
|
| 1356 |
+
"learning_rate": 3.752351180264323e-05,
|
| 1357 |
+
"loss": 0.7754,
|
| 1358 |
+
"step": 2250
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 0.67,
|
| 1362 |
+
"learning_rate": 3.7422643226042344e-05,
|
| 1363 |
+
"loss": 0.7815,
|
| 1364 |
+
"step": 2260
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 0.67,
|
| 1368 |
+
"learning_rate": 3.732150539503109e-05,
|
| 1369 |
+
"loss": 0.7749,
|
| 1370 |
+
"step": 2270
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.68,
|
| 1374 |
+
"learning_rate": 3.722010050172e-05,
|
| 1375 |
+
"loss": 0.7686,
|
| 1376 |
+
"step": 2280
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.68,
|
| 1380 |
+
"learning_rate": 3.711843074400805e-05,
|
| 1381 |
+
"loss": 0.7918,
|
| 1382 |
+
"step": 2290
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.68,
|
| 1386 |
+
"learning_rate": 3.701649832553499e-05,
|
| 1387 |
+
"loss": 0.7779,
|
| 1388 |
+
"step": 2300
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.68,
|
| 1392 |
+
"learning_rate": 3.691430545563365e-05,
|
| 1393 |
+
"loss": 0.7677,
|
| 1394 |
+
"step": 2310
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.69,
|
| 1398 |
+
"learning_rate": 3.6811854349281984e-05,
|
| 1399 |
+
"loss": 0.7758,
|
| 1400 |
+
"step": 2320
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 0.69,
|
| 1404 |
+
"learning_rate": 3.670914722705508e-05,
|
| 1405 |
+
"loss": 0.7635,
|
| 1406 |
+
"step": 2330
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.69,
|
| 1410 |
+
"learning_rate": 3.660618631507706e-05,
|
| 1411 |
+
"loss": 0.7778,
|
| 1412 |
+
"step": 2340
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 0.7,
|
| 1416 |
+
"learning_rate": 3.6502973844972806e-05,
|
| 1417 |
+
"loss": 0.772,
|
| 1418 |
+
"step": 2350
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.7,
|
| 1422 |
+
"learning_rate": 3.6399512053819574e-05,
|
| 1423 |
+
"loss": 0.7747,
|
| 1424 |
+
"step": 2360
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7,
|
| 1428 |
+
"learning_rate": 3.6295803184098546e-05,
|
| 1429 |
+
"loss": 0.7956,
|
| 1430 |
+
"step": 2370
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.71,
|
| 1434 |
+
"learning_rate": 3.619184948364619e-05,
|
| 1435 |
+
"loss": 0.7699,
|
| 1436 |
+
"step": 2380
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.71,
|
| 1440 |
+
"learning_rate": 3.608765320560558e-05,
|
| 1441 |
+
"loss": 0.7876,
|
| 1442 |
+
"step": 2390
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.71,
|
| 1446 |
+
"learning_rate": 3.59832166083775e-05,
|
| 1447 |
+
"loss": 0.7778,
|
| 1448 |
+
"step": 2400
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.71,
|
| 1452 |
+
"learning_rate": 3.587854195557156e-05,
|
| 1453 |
+
"loss": 0.7911,
|
| 1454 |
+
"step": 2410
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 0.72,
|
| 1458 |
+
"learning_rate": 3.577363151595706e-05,
|
| 1459 |
+
"loss": 0.7856,
|
| 1460 |
+
"step": 2420
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.72,
|
| 1464 |
+
"learning_rate": 3.566848756341391e-05,
|
| 1465 |
+
"loss": 0.7818,
|
| 1466 |
+
"step": 2430
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.72,
|
| 1470 |
+
"learning_rate": 3.556311237688327e-05,
|
| 1471 |
+
"loss": 0.7754,
|
| 1472 |
+
"step": 2440
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.73,
|
| 1476 |
+
"learning_rate": 3.545750824031816e-05,
|
| 1477 |
+
"loss": 0.7958,
|
| 1478 |
+
"step": 2450
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.73,
|
| 1482 |
+
"learning_rate": 3.535167744263401e-05,
|
| 1483 |
+
"loss": 0.777,
|
| 1484 |
+
"step": 2460
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.73,
|
| 1488 |
+
"learning_rate": 3.524562227765897e-05,
|
| 1489 |
+
"loss": 0.7774,
|
| 1490 |
+
"step": 2470
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.73,
|
| 1494 |
+
"learning_rate": 3.5139345044084284e-05,
|
| 1495 |
+
"loss": 0.782,
|
| 1496 |
+
"step": 2480
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 0.74,
|
| 1500 |
+
"learning_rate": 3.503284804541438e-05,
|
| 1501 |
+
"loss": 0.7888,
|
| 1502 |
+
"step": 2490
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.74,
|
| 1506 |
+
"learning_rate": 3.492613358991701e-05,
|
| 1507 |
+
"loss": 0.7684,
|
| 1508 |
+
"step": 2500
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.74,
|
| 1512 |
+
"learning_rate": 3.481920399057316e-05,
|
| 1513 |
+
"loss": 0.7783,
|
| 1514 |
+
"step": 2510
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.75,
|
| 1518 |
+
"learning_rate": 3.471206156502698e-05,
|
| 1519 |
+
"loss": 0.7706,
|
| 1520 |
+
"step": 2520
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.75,
|
| 1524 |
+
"learning_rate": 3.460470863553549e-05,
|
| 1525 |
+
"loss": 0.7739,
|
| 1526 |
+
"step": 2530
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.75,
|
| 1530 |
+
"learning_rate": 3.449714752891829e-05,
|
| 1531 |
+
"loss": 0.7786,
|
| 1532 |
+
"step": 2540
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 0.76,
|
| 1536 |
+
"learning_rate": 3.4389380576507136e-05,
|
| 1537 |
+
"loss": 0.7641,
|
| 1538 |
+
"step": 2550
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.76,
|
| 1542 |
+
"learning_rate": 3.428141011409534e-05,
|
| 1543 |
+
"loss": 0.7723,
|
| 1544 |
+
"step": 2560
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.76,
|
| 1548 |
+
"learning_rate": 3.417323848188721e-05,
|
| 1549 |
+
"loss": 0.7827,
|
| 1550 |
+
"step": 2570
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.76,
|
| 1554 |
+
"learning_rate": 3.406486802444731e-05,
|
| 1555 |
+
"loss": 0.7714,
|
| 1556 |
+
"step": 2580
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.77,
|
| 1560 |
+
"learning_rate": 3.395630109064963e-05,
|
| 1561 |
+
"loss": 0.7785,
|
| 1562 |
+
"step": 2590
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.77,
|
| 1566 |
+
"learning_rate": 3.38475400336267e-05,
|
| 1567 |
+
"loss": 0.7715,
|
| 1568 |
+
"step": 2600
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.77,
|
| 1572 |
+
"learning_rate": 3.3738587210718555e-05,
|
| 1573 |
+
"loss": 0.7736,
|
| 1574 |
+
"step": 2610
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 0.78,
|
| 1578 |
+
"learning_rate": 3.362944498342165e-05,
|
| 1579 |
+
"loss": 0.7679,
|
| 1580 |
+
"step": 2620
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.78,
|
| 1584 |
+
"learning_rate": 3.352011571733772e-05,
|
| 1585 |
+
"loss": 0.7754,
|
| 1586 |
+
"step": 2630
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.78,
|
| 1590 |
+
"learning_rate": 3.341060178212242e-05,
|
| 1591 |
+
"loss": 0.786,
|
| 1592 |
+
"step": 2640
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.79,
|
| 1596 |
+
"learning_rate": 3.330090555143405e-05,
|
| 1597 |
+
"loss": 0.7873,
|
| 1598 |
+
"step": 2650
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.79,
|
| 1602 |
+
"learning_rate": 3.319102940288206e-05,
|
| 1603 |
+
"loss": 0.7699,
|
| 1604 |
+
"step": 2660
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 0.79,
|
| 1608 |
+
"learning_rate": 3.308097571797552e-05,
|
| 1609 |
+
"loss": 0.7627,
|
| 1610 |
+
"step": 2670
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.79,
|
| 1614 |
+
"learning_rate": 3.297074688207152e-05,
|
| 1615 |
+
"loss": 0.7803,
|
| 1616 |
+
"step": 2680
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 0.8,
|
| 1620 |
+
"learning_rate": 3.286034528432345e-05,
|
| 1621 |
+
"loss": 0.78,
|
| 1622 |
+
"step": 2690
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 0.8,
|
| 1626 |
+
"learning_rate": 3.274977331762923e-05,
|
| 1627 |
+
"loss": 0.7549,
|
| 1628 |
+
"step": 2700
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.8,
|
| 1632 |
+
"learning_rate": 3.2639033378579445e-05,
|
| 1633 |
+
"loss": 0.7679,
|
| 1634 |
+
"step": 2710
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.81,
|
| 1638 |
+
"learning_rate": 3.252812786740537e-05,
|
| 1639 |
+
"loss": 0.7708,
|
| 1640 |
+
"step": 2720
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.81,
|
| 1644 |
+
"learning_rate": 3.241705918792699e-05,
|
| 1645 |
+
"loss": 0.7662,
|
| 1646 |
+
"step": 2730
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.81,
|
| 1650 |
+
"learning_rate": 3.230582974750088e-05,
|
| 1651 |
+
"loss": 0.7705,
|
| 1652 |
+
"step": 2740
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 0.81,
|
| 1656 |
+
"learning_rate": 3.2194441956968036e-05,
|
| 1657 |
+
"loss": 0.7762,
|
| 1658 |
+
"step": 2750
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 0.82,
|
| 1662 |
+
"learning_rate": 3.2082898230601574e-05,
|
| 1663 |
+
"loss": 0.7689,
|
| 1664 |
+
"step": 2760
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.82,
|
| 1668 |
+
"learning_rate": 3.1971200986054476e-05,
|
| 1669 |
+
"loss": 0.7834,
|
| 1670 |
+
"step": 2770
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.82,
|
| 1674 |
+
"learning_rate": 3.185935264430714e-05,
|
| 1675 |
+
"loss": 0.7784,
|
| 1676 |
+
"step": 2780
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.83,
|
| 1680 |
+
"learning_rate": 3.174735562961491e-05,
|
| 1681 |
+
"loss": 0.7749,
|
| 1682 |
+
"step": 2790
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.83,
|
| 1686 |
+
"learning_rate": 3.1635212369455535e-05,
|
| 1687 |
+
"loss": 0.7776,
|
| 1688 |
+
"step": 2800
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 0.83,
|
| 1692 |
+
"learning_rate": 3.15229252944766e-05,
|
| 1693 |
+
"loss": 0.7708,
|
| 1694 |
+
"step": 2810
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.84,
|
| 1698 |
+
"learning_rate": 3.141049683844275e-05,
|
| 1699 |
+
"loss": 0.7655,
|
| 1700 |
+
"step": 2820
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.84,
|
| 1704 |
+
"learning_rate": 3.129792943818303e-05,
|
| 1705 |
+
"loss": 0.774,
|
| 1706 |
+
"step": 2830
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 0.84,
|
| 1710 |
+
"learning_rate": 3.118522553353799e-05,
|
| 1711 |
+
"loss": 0.7577,
|
| 1712 |
+
"step": 2840
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.84,
|
| 1716 |
+
"learning_rate": 3.107238756730689e-05,
|
| 1717 |
+
"loss": 0.7657,
|
| 1718 |
+
"step": 2850
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.85,
|
| 1722 |
+
"learning_rate": 3.095941798519468e-05,
|
| 1723 |
+
"loss": 0.7783,
|
| 1724 |
+
"step": 2860
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.85,
|
| 1728 |
+
"learning_rate": 3.084631923575901e-05,
|
| 1729 |
+
"loss": 0.7655,
|
| 1730 |
+
"step": 2870
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 0.85,
|
| 1734 |
+
"learning_rate": 3.0733093770357194e-05,
|
| 1735 |
+
"loss": 0.7708,
|
| 1736 |
+
"step": 2880
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 0.86,
|
| 1740 |
+
"learning_rate": 3.0619744043093e-05,
|
| 1741 |
+
"loss": 0.7671,
|
| 1742 |
+
"step": 2890
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 0.86,
|
| 1746 |
+
"learning_rate": 3.0506272510763557e-05,
|
| 1747 |
+
"loss": 0.7774,
|
| 1748 |
+
"step": 2900
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.86,
|
| 1752 |
+
"learning_rate": 3.0392681632806036e-05,
|
| 1753 |
+
"loss": 0.77,
|
| 1754 |
+
"step": 2910
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.87,
|
| 1758 |
+
"learning_rate": 3.0278973871244366e-05,
|
| 1759 |
+
"loss": 0.7754,
|
| 1760 |
+
"step": 2920
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.87,
|
| 1764 |
+
"learning_rate": 3.0165151690635855e-05,
|
| 1765 |
+
"loss": 0.7837,
|
| 1766 |
+
"step": 2930
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.87,
|
| 1770 |
+
"learning_rate": 3.0051217558017812e-05,
|
| 1771 |
+
"loss": 0.7769,
|
| 1772 |
+
"step": 2940
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 0.87,
|
| 1776 |
+
"learning_rate": 2.993717394285403e-05,
|
| 1777 |
+
"loss": 0.7568,
|
| 1778 |
+
"step": 2950
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 0.88,
|
| 1782 |
+
"learning_rate": 2.982302331698128e-05,
|
| 1783 |
+
"loss": 0.7774,
|
| 1784 |
+
"step": 2960
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 0.88,
|
| 1788 |
+
"learning_rate": 2.9708768154555732e-05,
|
| 1789 |
+
"loss": 0.7629,
|
| 1790 |
+
"step": 2970
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 0.88,
|
| 1794 |
+
"learning_rate": 2.959441093199935e-05,
|
| 1795 |
+
"loss": 0.7679,
|
| 1796 |
+
"step": 2980
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.89,
|
| 1800 |
+
"learning_rate": 2.9479954127946187e-05,
|
| 1801 |
+
"loss": 0.772,
|
| 1802 |
+
"step": 2990
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.89,
|
| 1806 |
+
"learning_rate": 2.936540022318866e-05,
|
| 1807 |
+
"loss": 0.7806,
|
| 1808 |
+
"step": 3000
|
| 1809 |
+
}
|
| 1810 |
+
],
|
| 1811 |
+
"max_steps": 6748,
|
| 1812 |
+
"num_train_epochs": 2,
|
| 1813 |
+
"total_flos": 5.337713347120857e+18,
|
| 1814 |
+
"trial_name": null,
|
| 1815 |
+
"trial_params": null
|
| 1816 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f710e8a04dbb33457b5b725adb1da191a7b05d7ab01f6ec60c73386df9d8286
|
| 3 |
+
size 3277
|
checkpoint-4000/README.md
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- load_in_8bit: False
|
| 9 |
+
- load_in_4bit: True
|
| 10 |
+
- llm_int8_threshold: 6.0
|
| 11 |
+
- llm_int8_skip_modules: None
|
| 12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
+
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: float16
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.4.0
|
checkpoint-4000/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./Llama-2-7b-chat-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 32.0,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-4000/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7bf1845cc6e68aa0579e5e88808550297fa08d24fb8a35cb0e38ebc16755731
|
| 3 |
+
size 16821197
|
checkpoint-4000/finetuning_args.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"dpo_beta": 0.1,
|
| 3 |
+
"finetuning_type": "lora",
|
| 4 |
+
"lora_alpha": 32.0,
|
| 5 |
+
"lora_dropout": 0.1,
|
| 6 |
+
"lora_rank": 8,
|
| 7 |
+
"lora_target": [
|
| 8 |
+
"q_proj",
|
| 9 |
+
"v_proj"
|
| 10 |
+
],
|
| 11 |
+
"name_module_trainable": "mlp",
|
| 12 |
+
"num_hidden_layers": 32,
|
| 13 |
+
"num_layer_trainable": 3,
|
| 14 |
+
"ppo_score_norm": false,
|
| 15 |
+
"resume_lora_training": true
|
| 16 |
+
}
|
checkpoint-4000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7f6cb6eb009a6909453a9696bb0e774d7db6df702514b5010c8141e067ed851a
|
| 3 |
+
size 33661637
|
checkpoint-4000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ffd8c2ce4f590d406989075063816ceb18c656ea870887898bccf20b75bdbd77
|
| 3 |
+
size 18679
|
checkpoint-4000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:45689d820d74708b258149c47a789bd97e2300a6ac6fe54628e4f301b8c12364
|
| 3 |
+
size 18679
|
checkpoint-4000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:62a32f853241a440561897c7f8f79f93e194ff2580a7fe9cc1a18cb0d1a4c5a8
|
| 3 |
+
size 18679
|
checkpoint-4000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03e00f5f20dfa01cb13a6cde7397bf7d6d9693e36851f5c15779c6f2e63250c2
|
| 3 |
+
size 627
|
checkpoint-4000/trainer_state.json
ADDED
|
@@ -0,0 +1,2416 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.1852729831839395,
|
| 5 |
+
"global_step": 4000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 4.9999729068921297e-05,
|
| 13 |
+
"loss": 1.369,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 4.9998916281557476e-05,
|
| 19 |
+
"loss": 1.1704,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 4.999772150012383e-05,
|
| 25 |
+
"loss": 1.131,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.01,
|
| 30 |
+
"learning_rate": 4.99958792440589e-05,
|
| 31 |
+
"loss": 1.0618,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.01,
|
| 36 |
+
"learning_rate": 4.999401536044942e-05,
|
| 37 |
+
"loss": 1.0539,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.02,
|
| 42 |
+
"learning_rate": 4.9991197949909295e-05,
|
| 43 |
+
"loss": 1.0157,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"learning_rate": 4.998783886799168e-05,
|
| 49 |
+
"loss": 1.0235,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.02,
|
| 54 |
+
"learning_rate": 4.998393818750294e-05,
|
| 55 |
+
"loss": 0.9841,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.03,
|
| 60 |
+
"learning_rate": 4.9979495992988316e-05,
|
| 61 |
+
"loss": 0.9787,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.03,
|
| 66 |
+
"learning_rate": 4.997451238073011e-05,
|
| 67 |
+
"loss": 0.9716,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.03,
|
| 72 |
+
"learning_rate": 4.996898745874553e-05,
|
| 73 |
+
"loss": 0.9718,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.04,
|
| 78 |
+
"learning_rate": 4.996292134678445e-05,
|
| 79 |
+
"loss": 0.9608,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.04,
|
| 84 |
+
"learning_rate": 4.995631417632672e-05,
|
| 85 |
+
"loss": 0.9547,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04,
|
| 90 |
+
"learning_rate": 4.994916609057936e-05,
|
| 91 |
+
"loss": 0.959,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.04,
|
| 96 |
+
"learning_rate": 4.9941477244473455e-05,
|
| 97 |
+
"loss": 0.9297,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.05,
|
| 102 |
+
"learning_rate": 4.993324780466081e-05,
|
| 103 |
+
"loss": 0.927,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.05,
|
| 108 |
+
"learning_rate": 4.992447794951029e-05,
|
| 109 |
+
"loss": 0.9352,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.05,
|
| 114 |
+
"learning_rate": 4.9915167869104006e-05,
|
| 115 |
+
"loss": 0.9244,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.06,
|
| 120 |
+
"learning_rate": 4.990531776523317e-05,
|
| 121 |
+
"loss": 0.9185,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.06,
|
| 126 |
+
"learning_rate": 4.9894927851393726e-05,
|
| 127 |
+
"loss": 0.9201,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06,
|
| 132 |
+
"learning_rate": 4.988399835278171e-05,
|
| 133 |
+
"loss": 0.9053,
|
| 134 |
+
"step": 210
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.07,
|
| 138 |
+
"learning_rate": 4.987252950628839e-05,
|
| 139 |
+
"loss": 0.912,
|
| 140 |
+
"step": 220
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.07,
|
| 144 |
+
"learning_rate": 4.986052156049513e-05,
|
| 145 |
+
"loss": 0.9013,
|
| 146 |
+
"step": 230
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.07,
|
| 150 |
+
"learning_rate": 4.984797477566799e-05,
|
| 151 |
+
"loss": 0.8942,
|
| 152 |
+
"step": 240
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.07,
|
| 156 |
+
"learning_rate": 4.9834889423752074e-05,
|
| 157 |
+
"loss": 0.8873,
|
| 158 |
+
"step": 250
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.08,
|
| 162 |
+
"learning_rate": 4.9821265788365664e-05,
|
| 163 |
+
"loss": 0.9036,
|
| 164 |
+
"step": 260
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.08,
|
| 168 |
+
"learning_rate": 4.980710416479407e-05,
|
| 169 |
+
"loss": 0.901,
|
| 170 |
+
"step": 270
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08,
|
| 174 |
+
"learning_rate": 4.97924048599832e-05,
|
| 175 |
+
"loss": 0.8941,
|
| 176 |
+
"step": 280
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.09,
|
| 180 |
+
"learning_rate": 4.977716819253293e-05,
|
| 181 |
+
"loss": 0.8886,
|
| 182 |
+
"step": 290
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.09,
|
| 186 |
+
"learning_rate": 4.97613944926902e-05,
|
| 187 |
+
"loss": 0.8852,
|
| 188 |
+
"step": 300
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.09,
|
| 192 |
+
"learning_rate": 4.974508410234187e-05,
|
| 193 |
+
"loss": 0.8877,
|
| 194 |
+
"step": 310
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.09,
|
| 198 |
+
"learning_rate": 4.972823737500725e-05,
|
| 199 |
+
"loss": 0.8833,
|
| 200 |
+
"step": 320
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.1,
|
| 204 |
+
"learning_rate": 4.971085467583051e-05,
|
| 205 |
+
"loss": 0.869,
|
| 206 |
+
"step": 330
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.1,
|
| 210 |
+
"learning_rate": 4.969293638157273e-05,
|
| 211 |
+
"loss": 0.877,
|
| 212 |
+
"step": 340
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.1,
|
| 216 |
+
"learning_rate": 4.967448288060372e-05,
|
| 217 |
+
"loss": 0.8793,
|
| 218 |
+
"step": 350
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.11,
|
| 222 |
+
"learning_rate": 4.9655494572893644e-05,
|
| 223 |
+
"loss": 0.8742,
|
| 224 |
+
"step": 360
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.11,
|
| 228 |
+
"learning_rate": 4.9635971870004316e-05,
|
| 229 |
+
"loss": 0.8774,
|
| 230 |
+
"step": 370
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.11,
|
| 234 |
+
"learning_rate": 4.961591519508029e-05,
|
| 235 |
+
"loss": 0.8682,
|
| 236 |
+
"step": 380
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.12,
|
| 240 |
+
"learning_rate": 4.9595324982839686e-05,
|
| 241 |
+
"loss": 0.8581,
|
| 242 |
+
"step": 390
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.12,
|
| 246 |
+
"learning_rate": 4.957420167956479e-05,
|
| 247 |
+
"loss": 0.868,
|
| 248 |
+
"step": 400
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.12,
|
| 252 |
+
"learning_rate": 4.955254574309234e-05,
|
| 253 |
+
"loss": 0.8672,
|
| 254 |
+
"step": 410
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.12,
|
| 258 |
+
"learning_rate": 4.9530357642803646e-05,
|
| 259 |
+
"loss": 0.8672,
|
| 260 |
+
"step": 420
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.13,
|
| 264 |
+
"learning_rate": 4.950763785961436e-05,
|
| 265 |
+
"loss": 0.867,
|
| 266 |
+
"step": 430
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.13,
|
| 270 |
+
"learning_rate": 4.948438688596414e-05,
|
| 271 |
+
"loss": 0.8612,
|
| 272 |
+
"step": 440
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.13,
|
| 276 |
+
"learning_rate": 4.946060522580588e-05,
|
| 277 |
+
"loss": 0.8607,
|
| 278 |
+
"step": 450
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.14,
|
| 282 |
+
"learning_rate": 4.943629339459484e-05,
|
| 283 |
+
"loss": 0.8588,
|
| 284 |
+
"step": 460
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.14,
|
| 288 |
+
"learning_rate": 4.9411451919277484e-05,
|
| 289 |
+
"loss": 0.8585,
|
| 290 |
+
"step": 470
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.14,
|
| 294 |
+
"learning_rate": 4.9386081338280035e-05,
|
| 295 |
+
"loss": 0.8574,
|
| 296 |
+
"step": 480
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"learning_rate": 4.9360182201496785e-05,
|
| 301 |
+
"loss": 0.8562,
|
| 302 |
+
"step": 490
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.15,
|
| 306 |
+
"learning_rate": 4.933375507027824e-05,
|
| 307 |
+
"loss": 0.8634,
|
| 308 |
+
"step": 500
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.15,
|
| 312 |
+
"learning_rate": 4.930680051741887e-05,
|
| 313 |
+
"loss": 0.8483,
|
| 314 |
+
"step": 510
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.15,
|
| 318 |
+
"learning_rate": 4.9279319127144775e-05,
|
| 319 |
+
"loss": 0.8559,
|
| 320 |
+
"step": 520
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.16,
|
| 324 |
+
"learning_rate": 4.9251311495100974e-05,
|
| 325 |
+
"loss": 0.8526,
|
| 326 |
+
"step": 530
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.16,
|
| 330 |
+
"learning_rate": 4.9222778228338506e-05,
|
| 331 |
+
"loss": 0.85,
|
| 332 |
+
"step": 540
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.16,
|
| 336 |
+
"learning_rate": 4.919371994530125e-05,
|
| 337 |
+
"loss": 0.8547,
|
| 338 |
+
"step": 550
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17,
|
| 342 |
+
"learning_rate": 4.9164137275812597e-05,
|
| 343 |
+
"loss": 0.8533,
|
| 344 |
+
"step": 560
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.17,
|
| 348 |
+
"learning_rate": 4.913403086106167e-05,
|
| 349 |
+
"loss": 0.8431,
|
| 350 |
+
"step": 570
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.17,
|
| 354 |
+
"learning_rate": 4.9103401353589585e-05,
|
| 355 |
+
"loss": 0.8346,
|
| 356 |
+
"step": 580
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.17,
|
| 360 |
+
"learning_rate": 4.9072249417275164e-05,
|
| 361 |
+
"loss": 0.8448,
|
| 362 |
+
"step": 590
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.18,
|
| 366 |
+
"learning_rate": 4.9040575727320606e-05,
|
| 367 |
+
"loss": 0.8477,
|
| 368 |
+
"step": 600
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.18,
|
| 372 |
+
"learning_rate": 4.90083809702369e-05,
|
| 373 |
+
"loss": 0.8555,
|
| 374 |
+
"step": 610
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.18,
|
| 378 |
+
"learning_rate": 4.897566584382884e-05,
|
| 379 |
+
"loss": 0.8295,
|
| 380 |
+
"step": 620
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19,
|
| 384 |
+
"learning_rate": 4.8942431057180015e-05,
|
| 385 |
+
"loss": 0.8359,
|
| 386 |
+
"step": 630
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.19,
|
| 390 |
+
"learning_rate": 4.890867733063733e-05,
|
| 391 |
+
"loss": 0.8521,
|
| 392 |
+
"step": 640
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.19,
|
| 396 |
+
"learning_rate": 4.887440539579547e-05,
|
| 397 |
+
"loss": 0.8349,
|
| 398 |
+
"step": 650
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.2,
|
| 402 |
+
"learning_rate": 4.883961599548103e-05,
|
| 403 |
+
"loss": 0.8285,
|
| 404 |
+
"step": 660
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.2,
|
| 408 |
+
"learning_rate": 4.8804309883736377e-05,
|
| 409 |
+
"loss": 0.831,
|
| 410 |
+
"step": 670
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.2,
|
| 414 |
+
"learning_rate": 4.876848782580336e-05,
|
| 415 |
+
"loss": 0.8295,
|
| 416 |
+
"step": 680
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.2,
|
| 420 |
+
"learning_rate": 4.873215059810666e-05,
|
| 421 |
+
"loss": 0.8406,
|
| 422 |
+
"step": 690
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.21,
|
| 426 |
+
"learning_rate": 4.8695298988237035e-05,
|
| 427 |
+
"loss": 0.8456,
|
| 428 |
+
"step": 700
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.21,
|
| 432 |
+
"learning_rate": 4.8657933794934205e-05,
|
| 433 |
+
"loss": 0.8218,
|
| 434 |
+
"step": 710
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.21,
|
| 438 |
+
"learning_rate": 4.8620055828069535e-05,
|
| 439 |
+
"loss": 0.8424,
|
| 440 |
+
"step": 720
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.22,
|
| 444 |
+
"learning_rate": 4.858166590862849e-05,
|
| 445 |
+
"loss": 0.8258,
|
| 446 |
+
"step": 730
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.22,
|
| 450 |
+
"learning_rate": 4.854276486869287e-05,
|
| 451 |
+
"loss": 0.8352,
|
| 452 |
+
"step": 740
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.22,
|
| 456 |
+
"learning_rate": 4.850335355142272e-05,
|
| 457 |
+
"loss": 0.836,
|
| 458 |
+
"step": 750
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.23,
|
| 462 |
+
"learning_rate": 4.846343281103809e-05,
|
| 463 |
+
"loss": 0.8388,
|
| 464 |
+
"step": 760
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.23,
|
| 468 |
+
"learning_rate": 4.842300351280054e-05,
|
| 469 |
+
"loss": 0.8287,
|
| 470 |
+
"step": 770
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.23,
|
| 474 |
+
"learning_rate": 4.8382066532994316e-05,
|
| 475 |
+
"loss": 0.8394,
|
| 476 |
+
"step": 780
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.23,
|
| 480 |
+
"learning_rate": 4.834062275890745e-05,
|
| 481 |
+
"loss": 0.8253,
|
| 482 |
+
"step": 790
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.24,
|
| 486 |
+
"learning_rate": 4.829867308881244e-05,
|
| 487 |
+
"loss": 0.82,
|
| 488 |
+
"step": 800
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.24,
|
| 492 |
+
"learning_rate": 4.825621843194684e-05,
|
| 493 |
+
"loss": 0.8294,
|
| 494 |
+
"step": 810
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.24,
|
| 498 |
+
"learning_rate": 4.821325970849353e-05,
|
| 499 |
+
"loss": 0.8275,
|
| 500 |
+
"step": 820
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.25,
|
| 504 |
+
"learning_rate": 4.8169797849560764e-05,
|
| 505 |
+
"loss": 0.8129,
|
| 506 |
+
"step": 830
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.25,
|
| 510 |
+
"learning_rate": 4.812583379716202e-05,
|
| 511 |
+
"loss": 0.8215,
|
| 512 |
+
"step": 840
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.25,
|
| 516 |
+
"learning_rate": 4.8081368504195545e-05,
|
| 517 |
+
"loss": 0.84,
|
| 518 |
+
"step": 850
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.25,
|
| 522 |
+
"learning_rate": 4.803640293442373e-05,
|
| 523 |
+
"loss": 0.8246,
|
| 524 |
+
"step": 860
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.26,
|
| 528 |
+
"learning_rate": 4.799093806245218e-05,
|
| 529 |
+
"loss": 0.8179,
|
| 530 |
+
"step": 870
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.26,
|
| 534 |
+
"learning_rate": 4.7944974873708656e-05,
|
| 535 |
+
"loss": 0.8238,
|
| 536 |
+
"step": 880
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.26,
|
| 540 |
+
"learning_rate": 4.7898514364421655e-05,
|
| 541 |
+
"loss": 0.8327,
|
| 542 |
+
"step": 890
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.27,
|
| 546 |
+
"learning_rate": 4.785155754159886e-05,
|
| 547 |
+
"loss": 0.812,
|
| 548 |
+
"step": 900
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.27,
|
| 552 |
+
"learning_rate": 4.780410542300527e-05,
|
| 553 |
+
"loss": 0.8293,
|
| 554 |
+
"step": 910
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.27,
|
| 558 |
+
"learning_rate": 4.7756159037141185e-05,
|
| 559 |
+
"loss": 0.8266,
|
| 560 |
+
"step": 920
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.28,
|
| 564 |
+
"learning_rate": 4.7707719423219886e-05,
|
| 565 |
+
"loss": 0.8209,
|
| 566 |
+
"step": 930
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.28,
|
| 570 |
+
"learning_rate": 4.765878763114513e-05,
|
| 571 |
+
"loss": 0.8182,
|
| 572 |
+
"step": 940
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.28,
|
| 576 |
+
"learning_rate": 4.760936472148837e-05,
|
| 577 |
+
"loss": 0.8045,
|
| 578 |
+
"step": 950
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.28,
|
| 582 |
+
"learning_rate": 4.755945176546578e-05,
|
| 583 |
+
"loss": 0.8126,
|
| 584 |
+
"step": 960
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.29,
|
| 588 |
+
"learning_rate": 4.7509049844915036e-05,
|
| 589 |
+
"loss": 0.8201,
|
| 590 |
+
"step": 970
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.29,
|
| 594 |
+
"learning_rate": 4.7458160052271883e-05,
|
| 595 |
+
"loss": 0.8133,
|
| 596 |
+
"step": 980
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.29,
|
| 600 |
+
"learning_rate": 4.7406783490546434e-05,
|
| 601 |
+
"loss": 0.8157,
|
| 602 |
+
"step": 990
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.3,
|
| 606 |
+
"learning_rate": 4.735492127329928e-05,
|
| 607 |
+
"loss": 0.817,
|
| 608 |
+
"step": 1000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.3,
|
| 612 |
+
"learning_rate": 4.7302574524617315e-05,
|
| 613 |
+
"loss": 0.8002,
|
| 614 |
+
"step": 1010
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.3,
|
| 618 |
+
"learning_rate": 4.724974437908944e-05,
|
| 619 |
+
"loss": 0.8281,
|
| 620 |
+
"step": 1020
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.31,
|
| 624 |
+
"learning_rate": 4.719643198178193e-05,
|
| 625 |
+
"loss": 0.8129,
|
| 626 |
+
"step": 1030
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.31,
|
| 630 |
+
"learning_rate": 4.71426384882136e-05,
|
| 631 |
+
"loss": 0.8125,
|
| 632 |
+
"step": 1040
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.31,
|
| 636 |
+
"learning_rate": 4.708836506433078e-05,
|
| 637 |
+
"loss": 0.8066,
|
| 638 |
+
"step": 1050
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.31,
|
| 642 |
+
"learning_rate": 4.7033612886482066e-05,
|
| 643 |
+
"loss": 0.8159,
|
| 644 |
+
"step": 1060
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.32,
|
| 648 |
+
"learning_rate": 4.697838314139278e-05,
|
| 649 |
+
"loss": 0.8288,
|
| 650 |
+
"step": 1070
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.32,
|
| 654 |
+
"learning_rate": 4.6922677026139275e-05,
|
| 655 |
+
"loss": 0.8175,
|
| 656 |
+
"step": 1080
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.32,
|
| 660 |
+
"learning_rate": 4.686649574812299e-05,
|
| 661 |
+
"loss": 0.8107,
|
| 662 |
+
"step": 1090
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.33,
|
| 666 |
+
"learning_rate": 4.680984052504426e-05,
|
| 667 |
+
"loss": 0.8127,
|
| 668 |
+
"step": 1100
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.33,
|
| 672 |
+
"learning_rate": 4.6752712584875935e-05,
|
| 673 |
+
"loss": 0.8023,
|
| 674 |
+
"step": 1110
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.33,
|
| 678 |
+
"learning_rate": 4.669511316583678e-05,
|
| 679 |
+
"loss": 0.8066,
|
| 680 |
+
"step": 1120
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.33,
|
| 684 |
+
"learning_rate": 4.663704351636461e-05,
|
| 685 |
+
"loss": 0.8069,
|
| 686 |
+
"step": 1130
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.34,
|
| 690 |
+
"learning_rate": 4.6578504895089245e-05,
|
| 691 |
+
"loss": 0.8129,
|
| 692 |
+
"step": 1140
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.34,
|
| 696 |
+
"learning_rate": 4.651949857080524e-05,
|
| 697 |
+
"loss": 0.8247,
|
| 698 |
+
"step": 1150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.34,
|
| 702 |
+
"learning_rate": 4.6460025822444354e-05,
|
| 703 |
+
"loss": 0.8125,
|
| 704 |
+
"step": 1160
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.35,
|
| 708 |
+
"learning_rate": 4.640008793904785e-05,
|
| 709 |
+
"loss": 0.8141,
|
| 710 |
+
"step": 1170
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.35,
|
| 714 |
+
"learning_rate": 4.6339686219738576e-05,
|
| 715 |
+
"loss": 0.8066,
|
| 716 |
+
"step": 1180
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.35,
|
| 720 |
+
"learning_rate": 4.627882197369276e-05,
|
| 721 |
+
"loss": 0.8052,
|
| 722 |
+
"step": 1190
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.36,
|
| 726 |
+
"learning_rate": 4.621749652011167e-05,
|
| 727 |
+
"loss": 0.8114,
|
| 728 |
+
"step": 1200
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.36,
|
| 732 |
+
"learning_rate": 4.615571118819301e-05,
|
| 733 |
+
"loss": 0.8001,
|
| 734 |
+
"step": 1210
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.36,
|
| 738 |
+
"learning_rate": 4.60934673171021e-05,
|
| 739 |
+
"loss": 0.7938,
|
| 740 |
+
"step": 1220
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.36,
|
| 744 |
+
"learning_rate": 4.6030766255942884e-05,
|
| 745 |
+
"loss": 0.8107,
|
| 746 |
+
"step": 1230
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.37,
|
| 750 |
+
"learning_rate": 4.596760936372865e-05,
|
| 751 |
+
"loss": 0.8096,
|
| 752 |
+
"step": 1240
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.37,
|
| 756 |
+
"learning_rate": 4.5903998009352586e-05,
|
| 757 |
+
"loss": 0.8105,
|
| 758 |
+
"step": 1250
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.37,
|
| 762 |
+
"learning_rate": 4.583993357155813e-05,
|
| 763 |
+
"loss": 0.7995,
|
| 764 |
+
"step": 1260
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.38,
|
| 768 |
+
"learning_rate": 4.577541743890906e-05,
|
| 769 |
+
"loss": 0.8161,
|
| 770 |
+
"step": 1270
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.38,
|
| 774 |
+
"learning_rate": 4.571045100975941e-05,
|
| 775 |
+
"loss": 0.8003,
|
| 776 |
+
"step": 1280
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.38,
|
| 780 |
+
"learning_rate": 4.564503569222315e-05,
|
| 781 |
+
"loss": 0.8114,
|
| 782 |
+
"step": 1290
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.39,
|
| 786 |
+
"learning_rate": 4.5579172904143685e-05,
|
| 787 |
+
"loss": 0.8165,
|
| 788 |
+
"step": 1300
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.39,
|
| 792 |
+
"learning_rate": 4.551286407306312e-05,
|
| 793 |
+
"loss": 0.8022,
|
| 794 |
+
"step": 1310
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.39,
|
| 798 |
+
"learning_rate": 4.54461106361913e-05,
|
| 799 |
+
"loss": 0.8032,
|
| 800 |
+
"step": 1320
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.39,
|
| 804 |
+
"learning_rate": 4.5378914040374684e-05,
|
| 805 |
+
"loss": 0.8067,
|
| 806 |
+
"step": 1330
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.4,
|
| 810 |
+
"learning_rate": 4.531127574206496e-05,
|
| 811 |
+
"loss": 0.789,
|
| 812 |
+
"step": 1340
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.4,
|
| 816 |
+
"learning_rate": 4.524319720728749e-05,
|
| 817 |
+
"loss": 0.8139,
|
| 818 |
+
"step": 1350
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.4,
|
| 822 |
+
"learning_rate": 4.517467991160956e-05,
|
| 823 |
+
"loss": 0.805,
|
| 824 |
+
"step": 1360
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.41,
|
| 828 |
+
"learning_rate": 4.510572534010836e-05,
|
| 829 |
+
"loss": 0.8123,
|
| 830 |
+
"step": 1370
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.41,
|
| 834 |
+
"learning_rate": 4.5036334987338785e-05,
|
| 835 |
+
"loss": 0.7976,
|
| 836 |
+
"step": 1380
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.41,
|
| 840 |
+
"learning_rate": 4.4966510357301094e-05,
|
| 841 |
+
"loss": 0.813,
|
| 842 |
+
"step": 1390
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.41,
|
| 846 |
+
"learning_rate": 4.489625296340828e-05,
|
| 847 |
+
"loss": 0.7975,
|
| 848 |
+
"step": 1400
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.42,
|
| 852 |
+
"learning_rate": 4.4825564328453256e-05,
|
| 853 |
+
"loss": 0.7969,
|
| 854 |
+
"step": 1410
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.42,
|
| 858 |
+
"learning_rate": 4.4754445984575874e-05,
|
| 859 |
+
"loss": 0.8006,
|
| 860 |
+
"step": 1420
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.42,
|
| 864 |
+
"learning_rate": 4.4682899473229714e-05,
|
| 865 |
+
"loss": 0.8041,
|
| 866 |
+
"step": 1430
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.43,
|
| 870 |
+
"learning_rate": 4.461092634514863e-05,
|
| 871 |
+
"loss": 0.7998,
|
| 872 |
+
"step": 1440
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.43,
|
| 876 |
+
"learning_rate": 4.453852816031323e-05,
|
| 877 |
+
"loss": 0.8,
|
| 878 |
+
"step": 1450
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.43,
|
| 882 |
+
"learning_rate": 4.446570648791695e-05,
|
| 883 |
+
"loss": 0.7993,
|
| 884 |
+
"step": 1460
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.44,
|
| 888 |
+
"learning_rate": 4.439246290633216e-05,
|
| 889 |
+
"loss": 0.8113,
|
| 890 |
+
"step": 1470
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.44,
|
| 894 |
+
"learning_rate": 4.431879900307585e-05,
|
| 895 |
+
"loss": 0.8133,
|
| 896 |
+
"step": 1480
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.44,
|
| 900 |
+
"learning_rate": 4.4244716374775274e-05,
|
| 901 |
+
"loss": 0.7963,
|
| 902 |
+
"step": 1490
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.44,
|
| 906 |
+
"learning_rate": 4.4170216627133365e-05,
|
| 907 |
+
"loss": 0.8005,
|
| 908 |
+
"step": 1500
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.45,
|
| 912 |
+
"learning_rate": 4.409530137489386e-05,
|
| 913 |
+
"loss": 0.8016,
|
| 914 |
+
"step": 1510
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.45,
|
| 918 |
+
"learning_rate": 4.4019972241806384e-05,
|
| 919 |
+
"loss": 0.7858,
|
| 920 |
+
"step": 1520
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.45,
|
| 924 |
+
"learning_rate": 4.3944230860591194e-05,
|
| 925 |
+
"loss": 0.8009,
|
| 926 |
+
"step": 1530
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.46,
|
| 930 |
+
"learning_rate": 4.3868078872903814e-05,
|
| 931 |
+
"loss": 0.7929,
|
| 932 |
+
"step": 1540
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.46,
|
| 936 |
+
"learning_rate": 4.3791517929299466e-05,
|
| 937 |
+
"loss": 0.8065,
|
| 938 |
+
"step": 1550
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.46,
|
| 942 |
+
"learning_rate": 4.3714549689197265e-05,
|
| 943 |
+
"loss": 0.792,
|
| 944 |
+
"step": 1560
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.47,
|
| 948 |
+
"learning_rate": 4.363717582084428e-05,
|
| 949 |
+
"loss": 0.7965,
|
| 950 |
+
"step": 1570
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.47,
|
| 954 |
+
"learning_rate": 4.355939800127937e-05,
|
| 955 |
+
"loss": 0.8038,
|
| 956 |
+
"step": 1580
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.47,
|
| 960 |
+
"learning_rate": 4.34812179162968e-05,
|
| 961 |
+
"loss": 0.7816,
|
| 962 |
+
"step": 1590
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.47,
|
| 966 |
+
"learning_rate": 4.340263726040976e-05,
|
| 967 |
+
"loss": 0.8053,
|
| 968 |
+
"step": 1600
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.48,
|
| 972 |
+
"learning_rate": 4.3323657736813614e-05,
|
| 973 |
+
"loss": 0.8009,
|
| 974 |
+
"step": 1610
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.48,
|
| 978 |
+
"learning_rate": 4.324428105734893e-05,
|
| 979 |
+
"loss": 0.8003,
|
| 980 |
+
"step": 1620
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.48,
|
| 984 |
+
"learning_rate": 4.3164508942464495e-05,
|
| 985 |
+
"loss": 0.7973,
|
| 986 |
+
"step": 1630
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.49,
|
| 990 |
+
"learning_rate": 4.3084343121179896e-05,
|
| 991 |
+
"loss": 0.7851,
|
| 992 |
+
"step": 1640
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.49,
|
| 996 |
+
"learning_rate": 4.300378533104814e-05,
|
| 997 |
+
"loss": 0.7982,
|
| 998 |
+
"step": 1650
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.49,
|
| 1002 |
+
"learning_rate": 4.292283731811794e-05,
|
| 1003 |
+
"loss": 0.797,
|
| 1004 |
+
"step": 1660
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.49,
|
| 1008 |
+
"learning_rate": 4.284150083689589e-05,
|
| 1009 |
+
"loss": 0.7974,
|
| 1010 |
+
"step": 1670
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.5,
|
| 1014 |
+
"learning_rate": 4.275977765030845e-05,
|
| 1015 |
+
"loss": 0.79,
|
| 1016 |
+
"step": 1680
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.5,
|
| 1020 |
+
"learning_rate": 4.267766952966369e-05,
|
| 1021 |
+
"loss": 0.7997,
|
| 1022 |
+
"step": 1690
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 0.5,
|
| 1026 |
+
"learning_rate": 4.259517825461296e-05,
|
| 1027 |
+
"loss": 0.7949,
|
| 1028 |
+
"step": 1700
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.51,
|
| 1032 |
+
"learning_rate": 4.2512305613112266e-05,
|
| 1033 |
+
"loss": 0.8015,
|
| 1034 |
+
"step": 1710
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 0.51,
|
| 1038 |
+
"learning_rate": 4.2429053401383545e-05,
|
| 1039 |
+
"loss": 0.7989,
|
| 1040 |
+
"step": 1720
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 0.51,
|
| 1044 |
+
"learning_rate": 4.234542342387571e-05,
|
| 1045 |
+
"loss": 0.7897,
|
| 1046 |
+
"step": 1730
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 0.52,
|
| 1050 |
+
"learning_rate": 4.2261417493225565e-05,
|
| 1051 |
+
"loss": 0.795,
|
| 1052 |
+
"step": 1740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.52,
|
| 1056 |
+
"learning_rate": 4.2177037430218504e-05,
|
| 1057 |
+
"loss": 0.8037,
|
| 1058 |
+
"step": 1750
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 0.52,
|
| 1062 |
+
"learning_rate": 4.209228506374906e-05,
|
| 1063 |
+
"loss": 0.7814,
|
| 1064 |
+
"step": 1760
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 0.52,
|
| 1068 |
+
"learning_rate": 4.200716223078121e-05,
|
| 1069 |
+
"loss": 0.7957,
|
| 1070 |
+
"step": 1770
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 0.53,
|
| 1074 |
+
"learning_rate": 4.192167077630865e-05,
|
| 1075 |
+
"loss": 0.7884,
|
| 1076 |
+
"step": 1780
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 0.53,
|
| 1080 |
+
"learning_rate": 4.183581255331475e-05,
|
| 1081 |
+
"loss": 0.8069,
|
| 1082 |
+
"step": 1790
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 0.53,
|
| 1086 |
+
"learning_rate": 4.174958942273235e-05,
|
| 1087 |
+
"loss": 0.7946,
|
| 1088 |
+
"step": 1800
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 0.54,
|
| 1092 |
+
"learning_rate": 4.166300325340355e-05,
|
| 1093 |
+
"loss": 0.8016,
|
| 1094 |
+
"step": 1810
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.54,
|
| 1098 |
+
"learning_rate": 4.157605592203908e-05,
|
| 1099 |
+
"loss": 0.7806,
|
| 1100 |
+
"step": 1820
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 0.54,
|
| 1104 |
+
"learning_rate": 4.148874931317767e-05,
|
| 1105 |
+
"loss": 0.7786,
|
| 1106 |
+
"step": 1830
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 0.55,
|
| 1110 |
+
"learning_rate": 4.140108531914522e-05,
|
| 1111 |
+
"loss": 0.7817,
|
| 1112 |
+
"step": 1840
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 0.55,
|
| 1116 |
+
"learning_rate": 4.1313065840013775e-05,
|
| 1117 |
+
"loss": 0.7922,
|
| 1118 |
+
"step": 1850
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 0.55,
|
| 1122 |
+
"learning_rate": 4.1224692783560333e-05,
|
| 1123 |
+
"loss": 0.7761,
|
| 1124 |
+
"step": 1860
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 0.55,
|
| 1128 |
+
"learning_rate": 4.113596806522548e-05,
|
| 1129 |
+
"loss": 0.7841,
|
| 1130 |
+
"step": 1870
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 0.56,
|
| 1134 |
+
"learning_rate": 4.104689360807193e-05,
|
| 1135 |
+
"loss": 0.7983,
|
| 1136 |
+
"step": 1880
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.56,
|
| 1140 |
+
"learning_rate": 4.095747134274276e-05,
|
| 1141 |
+
"loss": 0.7753,
|
| 1142 |
+
"step": 1890
|
| 1143 |
+
},
|
| 1144 |
+
{
|
| 1145 |
+
"epoch": 0.56,
|
| 1146 |
+
"learning_rate": 4.086770320741965e-05,
|
| 1147 |
+
"loss": 0.7905,
|
| 1148 |
+
"step": 1900
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 0.57,
|
| 1152 |
+
"learning_rate": 4.077759114778081e-05,
|
| 1153 |
+
"loss": 0.7901,
|
| 1154 |
+
"step": 1910
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 0.57,
|
| 1158 |
+
"learning_rate": 4.068713711695885e-05,
|
| 1159 |
+
"loss": 0.7834,
|
| 1160 |
+
"step": 1920
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.57,
|
| 1164 |
+
"learning_rate": 4.059634307549841e-05,
|
| 1165 |
+
"loss": 0.7752,
|
| 1166 |
+
"step": 1930
|
| 1167 |
+
},
|
| 1168 |
+
{
|
| 1169 |
+
"epoch": 0.57,
|
| 1170 |
+
"learning_rate": 4.0505210991313714e-05,
|
| 1171 |
+
"loss": 0.7905,
|
| 1172 |
+
"step": 1940
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 0.58,
|
| 1176 |
+
"learning_rate": 4.0413742839645855e-05,
|
| 1177 |
+
"loss": 0.7824,
|
| 1178 |
+
"step": 1950
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.58,
|
| 1182 |
+
"learning_rate": 4.032194060302005e-05,
|
| 1183 |
+
"loss": 0.793,
|
| 1184 |
+
"step": 1960
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 0.58,
|
| 1188 |
+
"learning_rate": 4.02298062712026e-05,
|
| 1189 |
+
"loss": 0.7911,
|
| 1190 |
+
"step": 1970
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 0.59,
|
| 1194 |
+
"learning_rate": 4.013734184115783e-05,
|
| 1195 |
+
"loss": 0.7927,
|
| 1196 |
+
"step": 1980
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 0.59,
|
| 1200 |
+
"learning_rate": 4.004454931700475e-05,
|
| 1201 |
+
"loss": 0.7827,
|
| 1202 |
+
"step": 1990
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 0.59,
|
| 1206 |
+
"learning_rate": 3.995143070997367e-05,
|
| 1207 |
+
"loss": 0.785,
|
| 1208 |
+
"step": 2000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 0.6,
|
| 1212 |
+
"learning_rate": 3.985798803836256e-05,
|
| 1213 |
+
"loss": 0.7922,
|
| 1214 |
+
"step": 2010
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 0.6,
|
| 1218 |
+
"learning_rate": 3.976422332749331e-05,
|
| 1219 |
+
"loss": 0.7969,
|
| 1220 |
+
"step": 2020
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.6,
|
| 1224 |
+
"learning_rate": 3.967013860966787e-05,
|
| 1225 |
+
"loss": 0.7801,
|
| 1226 |
+
"step": 2030
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.6,
|
| 1230 |
+
"learning_rate": 3.957573592412417e-05,
|
| 1231 |
+
"loss": 0.7876,
|
| 1232 |
+
"step": 2040
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 0.61,
|
| 1236 |
+
"learning_rate": 3.948101731699192e-05,
|
| 1237 |
+
"loss": 0.7862,
|
| 1238 |
+
"step": 2050
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 0.61,
|
| 1242 |
+
"learning_rate": 3.938598484124827e-05,
|
| 1243 |
+
"loss": 0.7855,
|
| 1244 |
+
"step": 2060
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 0.61,
|
| 1248 |
+
"learning_rate": 3.929064055667333e-05,
|
| 1249 |
+
"loss": 0.7917,
|
| 1250 |
+
"step": 2070
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 0.62,
|
| 1254 |
+
"learning_rate": 3.919498652980546e-05,
|
| 1255 |
+
"loss": 0.7855,
|
| 1256 |
+
"step": 2080
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 0.62,
|
| 1260 |
+
"learning_rate": 3.909902483389658e-05,
|
| 1261 |
+
"loss": 0.7641,
|
| 1262 |
+
"step": 2090
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.62,
|
| 1266 |
+
"learning_rate": 3.900275754886714e-05,
|
| 1267 |
+
"loss": 0.784,
|
| 1268 |
+
"step": 2100
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 0.63,
|
| 1272 |
+
"learning_rate": 3.8906186761261095e-05,
|
| 1273 |
+
"loss": 0.7745,
|
| 1274 |
+
"step": 2110
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 0.63,
|
| 1278 |
+
"learning_rate": 3.880931456420066e-05,
|
| 1279 |
+
"loss": 0.7913,
|
| 1280 |
+
"step": 2120
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 0.63,
|
| 1284 |
+
"learning_rate": 3.8712143057340935e-05,
|
| 1285 |
+
"loss": 0.7823,
|
| 1286 |
+
"step": 2130
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 0.63,
|
| 1290 |
+
"learning_rate": 3.861467434682442e-05,
|
| 1291 |
+
"loss": 0.7834,
|
| 1292 |
+
"step": 2140
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.64,
|
| 1296 |
+
"learning_rate": 3.851691054523534e-05,
|
| 1297 |
+
"loss": 0.7929,
|
| 1298 |
+
"step": 2150
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 0.64,
|
| 1302 |
+
"learning_rate": 3.841885377155388e-05,
|
| 1303 |
+
"loss": 0.7751,
|
| 1304 |
+
"step": 2160
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.64,
|
| 1308 |
+
"learning_rate": 3.832050615111023e-05,
|
| 1309 |
+
"loss": 0.7814,
|
| 1310 |
+
"step": 2170
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 0.65,
|
| 1314 |
+
"learning_rate": 3.8221869815538544e-05,
|
| 1315 |
+
"loss": 0.7833,
|
| 1316 |
+
"step": 2180
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 0.65,
|
| 1320 |
+
"learning_rate": 3.812294690273074e-05,
|
| 1321 |
+
"loss": 0.7829,
|
| 1322 |
+
"step": 2190
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 0.65,
|
| 1326 |
+
"learning_rate": 3.802373955679011e-05,
|
| 1327 |
+
"loss": 0.773,
|
| 1328 |
+
"step": 2200
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 0.65,
|
| 1332 |
+
"learning_rate": 3.7924249927984936e-05,
|
| 1333 |
+
"loss": 0.7827,
|
| 1334 |
+
"step": 2210
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 0.66,
|
| 1338 |
+
"learning_rate": 3.78244801727018e-05,
|
| 1339 |
+
"loss": 0.7863,
|
| 1340 |
+
"step": 2220
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 0.66,
|
| 1344 |
+
"learning_rate": 3.77244324533989e-05,
|
| 1345 |
+
"loss": 0.7881,
|
| 1346 |
+
"step": 2230
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.66,
|
| 1350 |
+
"learning_rate": 3.762410893855916e-05,
|
| 1351 |
+
"loss": 0.7889,
|
| 1352 |
+
"step": 2240
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 0.67,
|
| 1356 |
+
"learning_rate": 3.752351180264323e-05,
|
| 1357 |
+
"loss": 0.7754,
|
| 1358 |
+
"step": 2250
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 0.67,
|
| 1362 |
+
"learning_rate": 3.7422643226042344e-05,
|
| 1363 |
+
"loss": 0.7815,
|
| 1364 |
+
"step": 2260
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 0.67,
|
| 1368 |
+
"learning_rate": 3.732150539503109e-05,
|
| 1369 |
+
"loss": 0.7749,
|
| 1370 |
+
"step": 2270
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 0.68,
|
| 1374 |
+
"learning_rate": 3.722010050172e-05,
|
| 1375 |
+
"loss": 0.7686,
|
| 1376 |
+
"step": 2280
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 0.68,
|
| 1380 |
+
"learning_rate": 3.711843074400805e-05,
|
| 1381 |
+
"loss": 0.7918,
|
| 1382 |
+
"step": 2290
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 0.68,
|
| 1386 |
+
"learning_rate": 3.701649832553499e-05,
|
| 1387 |
+
"loss": 0.7779,
|
| 1388 |
+
"step": 2300
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.68,
|
| 1392 |
+
"learning_rate": 3.691430545563365e-05,
|
| 1393 |
+
"loss": 0.7677,
|
| 1394 |
+
"step": 2310
|
| 1395 |
+
},
|
| 1396 |
+
{
|
| 1397 |
+
"epoch": 0.69,
|
| 1398 |
+
"learning_rate": 3.6811854349281984e-05,
|
| 1399 |
+
"loss": 0.7758,
|
| 1400 |
+
"step": 2320
|
| 1401 |
+
},
|
| 1402 |
+
{
|
| 1403 |
+
"epoch": 0.69,
|
| 1404 |
+
"learning_rate": 3.670914722705508e-05,
|
| 1405 |
+
"loss": 0.7635,
|
| 1406 |
+
"step": 2330
|
| 1407 |
+
},
|
| 1408 |
+
{
|
| 1409 |
+
"epoch": 0.69,
|
| 1410 |
+
"learning_rate": 3.660618631507706e-05,
|
| 1411 |
+
"loss": 0.7778,
|
| 1412 |
+
"step": 2340
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 0.7,
|
| 1416 |
+
"learning_rate": 3.6502973844972806e-05,
|
| 1417 |
+
"loss": 0.772,
|
| 1418 |
+
"step": 2350
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 0.7,
|
| 1422 |
+
"learning_rate": 3.6399512053819574e-05,
|
| 1423 |
+
"loss": 0.7747,
|
| 1424 |
+
"step": 2360
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7,
|
| 1428 |
+
"learning_rate": 3.6295803184098546e-05,
|
| 1429 |
+
"loss": 0.7956,
|
| 1430 |
+
"step": 2370
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.71,
|
| 1434 |
+
"learning_rate": 3.619184948364619e-05,
|
| 1435 |
+
"loss": 0.7699,
|
| 1436 |
+
"step": 2380
|
| 1437 |
+
},
|
| 1438 |
+
{
|
| 1439 |
+
"epoch": 0.71,
|
| 1440 |
+
"learning_rate": 3.608765320560558e-05,
|
| 1441 |
+
"loss": 0.7876,
|
| 1442 |
+
"step": 2390
|
| 1443 |
+
},
|
| 1444 |
+
{
|
| 1445 |
+
"epoch": 0.71,
|
| 1446 |
+
"learning_rate": 3.59832166083775e-05,
|
| 1447 |
+
"loss": 0.7778,
|
| 1448 |
+
"step": 2400
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 0.71,
|
| 1452 |
+
"learning_rate": 3.587854195557156e-05,
|
| 1453 |
+
"loss": 0.7911,
|
| 1454 |
+
"step": 2410
|
| 1455 |
+
},
|
| 1456 |
+
{
|
| 1457 |
+
"epoch": 0.72,
|
| 1458 |
+
"learning_rate": 3.577363151595706e-05,
|
| 1459 |
+
"loss": 0.7856,
|
| 1460 |
+
"step": 2420
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 0.72,
|
| 1464 |
+
"learning_rate": 3.566848756341391e-05,
|
| 1465 |
+
"loss": 0.7818,
|
| 1466 |
+
"step": 2430
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 0.72,
|
| 1470 |
+
"learning_rate": 3.556311237688327e-05,
|
| 1471 |
+
"loss": 0.7754,
|
| 1472 |
+
"step": 2440
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.73,
|
| 1476 |
+
"learning_rate": 3.545750824031816e-05,
|
| 1477 |
+
"loss": 0.7958,
|
| 1478 |
+
"step": 2450
|
| 1479 |
+
},
|
| 1480 |
+
{
|
| 1481 |
+
"epoch": 0.73,
|
| 1482 |
+
"learning_rate": 3.535167744263401e-05,
|
| 1483 |
+
"loss": 0.777,
|
| 1484 |
+
"step": 2460
|
| 1485 |
+
},
|
| 1486 |
+
{
|
| 1487 |
+
"epoch": 0.73,
|
| 1488 |
+
"learning_rate": 3.524562227765897e-05,
|
| 1489 |
+
"loss": 0.7774,
|
| 1490 |
+
"step": 2470
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.73,
|
| 1494 |
+
"learning_rate": 3.5139345044084284e-05,
|
| 1495 |
+
"loss": 0.782,
|
| 1496 |
+
"step": 2480
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 0.74,
|
| 1500 |
+
"learning_rate": 3.503284804541438e-05,
|
| 1501 |
+
"loss": 0.7888,
|
| 1502 |
+
"step": 2490
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 0.74,
|
| 1506 |
+
"learning_rate": 3.492613358991701e-05,
|
| 1507 |
+
"loss": 0.7684,
|
| 1508 |
+
"step": 2500
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 0.74,
|
| 1512 |
+
"learning_rate": 3.481920399057316e-05,
|
| 1513 |
+
"loss": 0.7783,
|
| 1514 |
+
"step": 2510
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.75,
|
| 1518 |
+
"learning_rate": 3.471206156502698e-05,
|
| 1519 |
+
"loss": 0.7706,
|
| 1520 |
+
"step": 2520
|
| 1521 |
+
},
|
| 1522 |
+
{
|
| 1523 |
+
"epoch": 0.75,
|
| 1524 |
+
"learning_rate": 3.460470863553549e-05,
|
| 1525 |
+
"loss": 0.7739,
|
| 1526 |
+
"step": 2530
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 0.75,
|
| 1530 |
+
"learning_rate": 3.449714752891829e-05,
|
| 1531 |
+
"loss": 0.7786,
|
| 1532 |
+
"step": 2540
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 0.76,
|
| 1536 |
+
"learning_rate": 3.4389380576507136e-05,
|
| 1537 |
+
"loss": 0.7641,
|
| 1538 |
+
"step": 2550
|
| 1539 |
+
},
|
| 1540 |
+
{
|
| 1541 |
+
"epoch": 0.76,
|
| 1542 |
+
"learning_rate": 3.428141011409534e-05,
|
| 1543 |
+
"loss": 0.7723,
|
| 1544 |
+
"step": 2560
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 0.76,
|
| 1548 |
+
"learning_rate": 3.417323848188721e-05,
|
| 1549 |
+
"loss": 0.7827,
|
| 1550 |
+
"step": 2570
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 0.76,
|
| 1554 |
+
"learning_rate": 3.406486802444731e-05,
|
| 1555 |
+
"loss": 0.7714,
|
| 1556 |
+
"step": 2580
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.77,
|
| 1560 |
+
"learning_rate": 3.395630109064963e-05,
|
| 1561 |
+
"loss": 0.7785,
|
| 1562 |
+
"step": 2590
|
| 1563 |
+
},
|
| 1564 |
+
{
|
| 1565 |
+
"epoch": 0.77,
|
| 1566 |
+
"learning_rate": 3.38475400336267e-05,
|
| 1567 |
+
"loss": 0.7715,
|
| 1568 |
+
"step": 2600
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 0.77,
|
| 1572 |
+
"learning_rate": 3.3738587210718555e-05,
|
| 1573 |
+
"loss": 0.7736,
|
| 1574 |
+
"step": 2610
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 0.78,
|
| 1578 |
+
"learning_rate": 3.362944498342165e-05,
|
| 1579 |
+
"loss": 0.7679,
|
| 1580 |
+
"step": 2620
|
| 1581 |
+
},
|
| 1582 |
+
{
|
| 1583 |
+
"epoch": 0.78,
|
| 1584 |
+
"learning_rate": 3.352011571733772e-05,
|
| 1585 |
+
"loss": 0.7754,
|
| 1586 |
+
"step": 2630
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 0.78,
|
| 1590 |
+
"learning_rate": 3.341060178212242e-05,
|
| 1591 |
+
"loss": 0.786,
|
| 1592 |
+
"step": 2640
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 0.79,
|
| 1596 |
+
"learning_rate": 3.330090555143405e-05,
|
| 1597 |
+
"loss": 0.7873,
|
| 1598 |
+
"step": 2650
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.79,
|
| 1602 |
+
"learning_rate": 3.319102940288206e-05,
|
| 1603 |
+
"loss": 0.7699,
|
| 1604 |
+
"step": 2660
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 0.79,
|
| 1608 |
+
"learning_rate": 3.308097571797552e-05,
|
| 1609 |
+
"loss": 0.7627,
|
| 1610 |
+
"step": 2670
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 0.79,
|
| 1614 |
+
"learning_rate": 3.297074688207152e-05,
|
| 1615 |
+
"loss": 0.7803,
|
| 1616 |
+
"step": 2680
|
| 1617 |
+
},
|
| 1618 |
+
{
|
| 1619 |
+
"epoch": 0.8,
|
| 1620 |
+
"learning_rate": 3.286034528432345e-05,
|
| 1621 |
+
"loss": 0.78,
|
| 1622 |
+
"step": 2690
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 0.8,
|
| 1626 |
+
"learning_rate": 3.274977331762923e-05,
|
| 1627 |
+
"loss": 0.7549,
|
| 1628 |
+
"step": 2700
|
| 1629 |
+
},
|
| 1630 |
+
{
|
| 1631 |
+
"epoch": 0.8,
|
| 1632 |
+
"learning_rate": 3.2639033378579445e-05,
|
| 1633 |
+
"loss": 0.7679,
|
| 1634 |
+
"step": 2710
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 0.81,
|
| 1638 |
+
"learning_rate": 3.252812786740537e-05,
|
| 1639 |
+
"loss": 0.7708,
|
| 1640 |
+
"step": 2720
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.81,
|
| 1644 |
+
"learning_rate": 3.241705918792699e-05,
|
| 1645 |
+
"loss": 0.7662,
|
| 1646 |
+
"step": 2730
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 0.81,
|
| 1650 |
+
"learning_rate": 3.230582974750088e-05,
|
| 1651 |
+
"loss": 0.7705,
|
| 1652 |
+
"step": 2740
|
| 1653 |
+
},
|
| 1654 |
+
{
|
| 1655 |
+
"epoch": 0.81,
|
| 1656 |
+
"learning_rate": 3.2194441956968036e-05,
|
| 1657 |
+
"loss": 0.7762,
|
| 1658 |
+
"step": 2750
|
| 1659 |
+
},
|
| 1660 |
+
{
|
| 1661 |
+
"epoch": 0.82,
|
| 1662 |
+
"learning_rate": 3.2082898230601574e-05,
|
| 1663 |
+
"loss": 0.7689,
|
| 1664 |
+
"step": 2760
|
| 1665 |
+
},
|
| 1666 |
+
{
|
| 1667 |
+
"epoch": 0.82,
|
| 1668 |
+
"learning_rate": 3.1971200986054476e-05,
|
| 1669 |
+
"loss": 0.7834,
|
| 1670 |
+
"step": 2770
|
| 1671 |
+
},
|
| 1672 |
+
{
|
| 1673 |
+
"epoch": 0.82,
|
| 1674 |
+
"learning_rate": 3.185935264430714e-05,
|
| 1675 |
+
"loss": 0.7784,
|
| 1676 |
+
"step": 2780
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 0.83,
|
| 1680 |
+
"learning_rate": 3.174735562961491e-05,
|
| 1681 |
+
"loss": 0.7749,
|
| 1682 |
+
"step": 2790
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.83,
|
| 1686 |
+
"learning_rate": 3.1635212369455535e-05,
|
| 1687 |
+
"loss": 0.7776,
|
| 1688 |
+
"step": 2800
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 0.83,
|
| 1692 |
+
"learning_rate": 3.15229252944766e-05,
|
| 1693 |
+
"loss": 0.7708,
|
| 1694 |
+
"step": 2810
|
| 1695 |
+
},
|
| 1696 |
+
{
|
| 1697 |
+
"epoch": 0.84,
|
| 1698 |
+
"learning_rate": 3.141049683844275e-05,
|
| 1699 |
+
"loss": 0.7655,
|
| 1700 |
+
"step": 2820
|
| 1701 |
+
},
|
| 1702 |
+
{
|
| 1703 |
+
"epoch": 0.84,
|
| 1704 |
+
"learning_rate": 3.129792943818303e-05,
|
| 1705 |
+
"loss": 0.774,
|
| 1706 |
+
"step": 2830
|
| 1707 |
+
},
|
| 1708 |
+
{
|
| 1709 |
+
"epoch": 0.84,
|
| 1710 |
+
"learning_rate": 3.118522553353799e-05,
|
| 1711 |
+
"loss": 0.7577,
|
| 1712 |
+
"step": 2840
|
| 1713 |
+
},
|
| 1714 |
+
{
|
| 1715 |
+
"epoch": 0.84,
|
| 1716 |
+
"learning_rate": 3.107238756730689e-05,
|
| 1717 |
+
"loss": 0.7657,
|
| 1718 |
+
"step": 2850
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 0.85,
|
| 1722 |
+
"learning_rate": 3.095941798519468e-05,
|
| 1723 |
+
"loss": 0.7783,
|
| 1724 |
+
"step": 2860
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.85,
|
| 1728 |
+
"learning_rate": 3.084631923575901e-05,
|
| 1729 |
+
"loss": 0.7655,
|
| 1730 |
+
"step": 2870
|
| 1731 |
+
},
|
| 1732 |
+
{
|
| 1733 |
+
"epoch": 0.85,
|
| 1734 |
+
"learning_rate": 3.0733093770357194e-05,
|
| 1735 |
+
"loss": 0.7708,
|
| 1736 |
+
"step": 2880
|
| 1737 |
+
},
|
| 1738 |
+
{
|
| 1739 |
+
"epoch": 0.86,
|
| 1740 |
+
"learning_rate": 3.0619744043093e-05,
|
| 1741 |
+
"loss": 0.7671,
|
| 1742 |
+
"step": 2890
|
| 1743 |
+
},
|
| 1744 |
+
{
|
| 1745 |
+
"epoch": 0.86,
|
| 1746 |
+
"learning_rate": 3.0506272510763557e-05,
|
| 1747 |
+
"loss": 0.7774,
|
| 1748 |
+
"step": 2900
|
| 1749 |
+
},
|
| 1750 |
+
{
|
| 1751 |
+
"epoch": 0.86,
|
| 1752 |
+
"learning_rate": 3.0392681632806036e-05,
|
| 1753 |
+
"loss": 0.77,
|
| 1754 |
+
"step": 2910
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.87,
|
| 1758 |
+
"learning_rate": 3.0278973871244366e-05,
|
| 1759 |
+
"loss": 0.7754,
|
| 1760 |
+
"step": 2920
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 0.87,
|
| 1764 |
+
"learning_rate": 3.0165151690635855e-05,
|
| 1765 |
+
"loss": 0.7837,
|
| 1766 |
+
"step": 2930
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.87,
|
| 1770 |
+
"learning_rate": 3.0051217558017812e-05,
|
| 1771 |
+
"loss": 0.7769,
|
| 1772 |
+
"step": 2940
|
| 1773 |
+
},
|
| 1774 |
+
{
|
| 1775 |
+
"epoch": 0.87,
|
| 1776 |
+
"learning_rate": 2.993717394285403e-05,
|
| 1777 |
+
"loss": 0.7568,
|
| 1778 |
+
"step": 2950
|
| 1779 |
+
},
|
| 1780 |
+
{
|
| 1781 |
+
"epoch": 0.88,
|
| 1782 |
+
"learning_rate": 2.982302331698128e-05,
|
| 1783 |
+
"loss": 0.7774,
|
| 1784 |
+
"step": 2960
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 0.88,
|
| 1788 |
+
"learning_rate": 2.9708768154555732e-05,
|
| 1789 |
+
"loss": 0.7629,
|
| 1790 |
+
"step": 2970
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 0.88,
|
| 1794 |
+
"learning_rate": 2.959441093199935e-05,
|
| 1795 |
+
"loss": 0.7679,
|
| 1796 |
+
"step": 2980
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 0.89,
|
| 1800 |
+
"learning_rate": 2.9479954127946187e-05,
|
| 1801 |
+
"loss": 0.772,
|
| 1802 |
+
"step": 2990
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 0.89,
|
| 1806 |
+
"learning_rate": 2.936540022318866e-05,
|
| 1807 |
+
"loss": 0.7806,
|
| 1808 |
+
"step": 3000
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.89,
|
| 1812 |
+
"learning_rate": 2.9250751700623814e-05,
|
| 1813 |
+
"loss": 0.7613,
|
| 1814 |
+
"step": 3010
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 0.89,
|
| 1818 |
+
"learning_rate": 2.9136011045199484e-05,
|
| 1819 |
+
"loss": 0.7712,
|
| 1820 |
+
"step": 3020
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 0.9,
|
| 1824 |
+
"learning_rate": 2.9021180743860443e-05,
|
| 1825 |
+
"loss": 0.7768,
|
| 1826 |
+
"step": 3030
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 0.9,
|
| 1830 |
+
"learning_rate": 2.8906263285494468e-05,
|
| 1831 |
+
"loss": 0.7697,
|
| 1832 |
+
"step": 3040
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 0.9,
|
| 1836 |
+
"learning_rate": 2.8791261160878435e-05,
|
| 1837 |
+
"loss": 0.7699,
|
| 1838 |
+
"step": 3050
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 0.91,
|
| 1842 |
+
"learning_rate": 2.8676176862624333e-05,
|
| 1843 |
+
"loss": 0.7687,
|
| 1844 |
+
"step": 3060
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 0.91,
|
| 1848 |
+
"learning_rate": 2.8561012885125194e-05,
|
| 1849 |
+
"loss": 0.7728,
|
| 1850 |
+
"step": 3070
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.91,
|
| 1854 |
+
"learning_rate": 2.8445771724501082e-05,
|
| 1855 |
+
"loss": 0.7716,
|
| 1856 |
+
"step": 3080
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 0.92,
|
| 1860 |
+
"learning_rate": 2.8330455878544937e-05,
|
| 1861 |
+
"loss": 0.768,
|
| 1862 |
+
"step": 3090
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 0.92,
|
| 1866 |
+
"learning_rate": 2.8215067846668493e-05,
|
| 1867 |
+
"loss": 0.7657,
|
| 1868 |
+
"step": 3100
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 0.92,
|
| 1872 |
+
"learning_rate": 2.8099610129848066e-05,
|
| 1873 |
+
"loss": 0.7699,
|
| 1874 |
+
"step": 3110
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 0.92,
|
| 1878 |
+
"learning_rate": 2.7984085230570355e-05,
|
| 1879 |
+
"loss": 0.7643,
|
| 1880 |
+
"step": 3120
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 0.93,
|
| 1884 |
+
"learning_rate": 2.7868495652778208e-05,
|
| 1885 |
+
"loss": 0.7532,
|
| 1886 |
+
"step": 3130
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.93,
|
| 1890 |
+
"learning_rate": 2.775284390181634e-05,
|
| 1891 |
+
"loss": 0.7562,
|
| 1892 |
+
"step": 3140
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.93,
|
| 1896 |
+
"learning_rate": 2.763713248437705e-05,
|
| 1897 |
+
"loss": 0.7578,
|
| 1898 |
+
"step": 3150
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 0.94,
|
| 1902 |
+
"learning_rate": 2.7521363908445864e-05,
|
| 1903 |
+
"loss": 0.7698,
|
| 1904 |
+
"step": 3160
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 0.94,
|
| 1908 |
+
"learning_rate": 2.7405540683247194e-05,
|
| 1909 |
+
"loss": 0.7655,
|
| 1910 |
+
"step": 3170
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 0.94,
|
| 1914 |
+
"learning_rate": 2.728966531918995e-05,
|
| 1915 |
+
"loss": 0.7736,
|
| 1916 |
+
"step": 3180
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 0.95,
|
| 1920 |
+
"learning_rate": 2.7173740327813118e-05,
|
| 1921 |
+
"loss": 0.7739,
|
| 1922 |
+
"step": 3190
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 0.95,
|
| 1926 |
+
"learning_rate": 2.7057768221731345e-05,
|
| 1927 |
+
"loss": 0.76,
|
| 1928 |
+
"step": 3200
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 0.95,
|
| 1932 |
+
"learning_rate": 2.694175151458045e-05,
|
| 1933 |
+
"loss": 0.7769,
|
| 1934 |
+
"step": 3210
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.95,
|
| 1938 |
+
"learning_rate": 2.6825692720962946e-05,
|
| 1939 |
+
"loss": 0.7594,
|
| 1940 |
+
"step": 3220
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 0.96,
|
| 1944 |
+
"learning_rate": 2.670959435639359e-05,
|
| 1945 |
+
"loss": 0.753,
|
| 1946 |
+
"step": 3230
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 0.96,
|
| 1950 |
+
"learning_rate": 2.6593458937244786e-05,
|
| 1951 |
+
"loss": 0.7587,
|
| 1952 |
+
"step": 3240
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 0.96,
|
| 1956 |
+
"learning_rate": 2.6477288980692082e-05,
|
| 1957 |
+
"loss": 0.7565,
|
| 1958 |
+
"step": 3250
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 0.97,
|
| 1962 |
+
"learning_rate": 2.636108700465961e-05,
|
| 1963 |
+
"loss": 0.7599,
|
| 1964 |
+
"step": 3260
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 0.97,
|
| 1968 |
+
"learning_rate": 2.624485552776551e-05,
|
| 1969 |
+
"loss": 0.7628,
|
| 1970 |
+
"step": 3270
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 0.97,
|
| 1974 |
+
"learning_rate": 2.612859706926733e-05,
|
| 1975 |
+
"loss": 0.7702,
|
| 1976 |
+
"step": 3280
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.97,
|
| 1980 |
+
"learning_rate": 2.6012314149007434e-05,
|
| 1981 |
+
"loss": 0.763,
|
| 1982 |
+
"step": 3290
|
| 1983 |
+
},
|
| 1984 |
+
{
|
| 1985 |
+
"epoch": 0.98,
|
| 1986 |
+
"learning_rate": 2.5896009287358393e-05,
|
| 1987 |
+
"loss": 0.7709,
|
| 1988 |
+
"step": 3300
|
| 1989 |
+
},
|
| 1990 |
+
{
|
| 1991 |
+
"epoch": 0.98,
|
| 1992 |
+
"learning_rate": 2.5779685005168326e-05,
|
| 1993 |
+
"loss": 0.7643,
|
| 1994 |
+
"step": 3310
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 0.98,
|
| 1998 |
+
"learning_rate": 2.5663343823706303e-05,
|
| 1999 |
+
"loss": 0.7537,
|
| 2000 |
+
"step": 3320
|
| 2001 |
+
},
|
| 2002 |
+
{
|
| 2003 |
+
"epoch": 0.99,
|
| 2004 |
+
"learning_rate": 2.5546988264607656e-05,
|
| 2005 |
+
"loss": 0.7665,
|
| 2006 |
+
"step": 3330
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 0.99,
|
| 2010 |
+
"learning_rate": 2.5430620849819375e-05,
|
| 2011 |
+
"loss": 0.7771,
|
| 2012 |
+
"step": 3340
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 0.99,
|
| 2016 |
+
"learning_rate": 2.5314244101545382e-05,
|
| 2017 |
+
"loss": 0.7519,
|
| 2018 |
+
"step": 3350
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 1.0,
|
| 2022 |
+
"learning_rate": 2.5197860542191914e-05,
|
| 2023 |
+
"loss": 0.7581,
|
| 2024 |
+
"step": 3360
|
| 2025 |
+
},
|
| 2026 |
+
{
|
| 2027 |
+
"epoch": 1.0,
|
| 2028 |
+
"learning_rate": 2.5081472694312836e-05,
|
| 2029 |
+
"loss": 0.755,
|
| 2030 |
+
"step": 3370
|
| 2031 |
+
},
|
| 2032 |
+
{
|
| 2033 |
+
"epoch": 1.0,
|
| 2034 |
+
"learning_rate": 2.4965083080554964e-05,
|
| 2035 |
+
"loss": 0.7534,
|
| 2036 |
+
"step": 3380
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 1.0,
|
| 2040 |
+
"learning_rate": 2.484869422360338e-05,
|
| 2041 |
+
"loss": 0.752,
|
| 2042 |
+
"step": 3390
|
| 2043 |
+
},
|
| 2044 |
+
{
|
| 2045 |
+
"epoch": 1.01,
|
| 2046 |
+
"learning_rate": 2.4732308646126777e-05,
|
| 2047 |
+
"loss": 0.7502,
|
| 2048 |
+
"step": 3400
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 1.01,
|
| 2052 |
+
"learning_rate": 2.461592887072275e-05,
|
| 2053 |
+
"loss": 0.7783,
|
| 2054 |
+
"step": 3410
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.01,
|
| 2058 |
+
"learning_rate": 2.4499557419863157e-05,
|
| 2059 |
+
"loss": 0.7661,
|
| 2060 |
+
"step": 3420
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 1.02,
|
| 2064 |
+
"learning_rate": 2.4383196815839407e-05,
|
| 2065 |
+
"loss": 0.7487,
|
| 2066 |
+
"step": 3430
|
| 2067 |
+
},
|
| 2068 |
+
{
|
| 2069 |
+
"epoch": 1.02,
|
| 2070 |
+
"learning_rate": 2.4266849580707818e-05,
|
| 2071 |
+
"loss": 0.7581,
|
| 2072 |
+
"step": 3440
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 1.02,
|
| 2076 |
+
"learning_rate": 2.415051823623495e-05,
|
| 2077 |
+
"loss": 0.7625,
|
| 2078 |
+
"step": 3450
|
| 2079 |
+
},
|
| 2080 |
+
{
|
| 2081 |
+
"epoch": 1.03,
|
| 2082 |
+
"learning_rate": 2.4034205303842924e-05,
|
| 2083 |
+
"loss": 0.7671,
|
| 2084 |
+
"step": 3460
|
| 2085 |
+
},
|
| 2086 |
+
{
|
| 2087 |
+
"epoch": 1.03,
|
| 2088 |
+
"learning_rate": 2.391791330455481e-05,
|
| 2089 |
+
"loss": 0.7531,
|
| 2090 |
+
"step": 3470
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 1.03,
|
| 2094 |
+
"learning_rate": 2.3801644758939945e-05,
|
| 2095 |
+
"loss": 0.7576,
|
| 2096 |
+
"step": 3480
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.03,
|
| 2100 |
+
"learning_rate": 2.368540218705932e-05,
|
| 2101 |
+
"loss": 0.7571,
|
| 2102 |
+
"step": 3490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 1.04,
|
| 2106 |
+
"learning_rate": 2.356918810841099e-05,
|
| 2107 |
+
"loss": 0.7586,
|
| 2108 |
+
"step": 3500
|
| 2109 |
+
},
|
| 2110 |
+
{
|
| 2111 |
+
"epoch": 1.04,
|
| 2112 |
+
"learning_rate": 2.3453005041875396e-05,
|
| 2113 |
+
"loss": 0.7614,
|
| 2114 |
+
"step": 3510
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 1.04,
|
| 2118 |
+
"learning_rate": 2.333685550566082e-05,
|
| 2119 |
+
"loss": 0.7538,
|
| 2120 |
+
"step": 3520
|
| 2121 |
+
},
|
| 2122 |
+
{
|
| 2123 |
+
"epoch": 1.05,
|
| 2124 |
+
"learning_rate": 2.3220742017248797e-05,
|
| 2125 |
+
"loss": 0.7614,
|
| 2126 |
+
"step": 3530
|
| 2127 |
+
},
|
| 2128 |
+
{
|
| 2129 |
+
"epoch": 1.05,
|
| 2130 |
+
"learning_rate": 2.310466709333954e-05,
|
| 2131 |
+
"loss": 0.7603,
|
| 2132 |
+
"step": 3540
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 1.05,
|
| 2136 |
+
"learning_rate": 2.2988633249797398e-05,
|
| 2137 |
+
"loss": 0.7656,
|
| 2138 |
+
"step": 3550
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.05,
|
| 2142 |
+
"learning_rate": 2.2872643001596318e-05,
|
| 2143 |
+
"loss": 0.7444,
|
| 2144 |
+
"step": 3560
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 1.06,
|
| 2148 |
+
"learning_rate": 2.275669886276535e-05,
|
| 2149 |
+
"loss": 0.7527,
|
| 2150 |
+
"step": 3570
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 1.06,
|
| 2154 |
+
"learning_rate": 2.264080334633414e-05,
|
| 2155 |
+
"loss": 0.7414,
|
| 2156 |
+
"step": 3580
|
| 2157 |
+
},
|
| 2158 |
+
{
|
| 2159 |
+
"epoch": 1.06,
|
| 2160 |
+
"learning_rate": 2.252495896427847e-05,
|
| 2161 |
+
"loss": 0.7533,
|
| 2162 |
+
"step": 3590
|
| 2163 |
+
},
|
| 2164 |
+
{
|
| 2165 |
+
"epoch": 1.07,
|
| 2166 |
+
"learning_rate": 2.240916822746581e-05,
|
| 2167 |
+
"loss": 0.7577,
|
| 2168 |
+
"step": 3600
|
| 2169 |
+
},
|
| 2170 |
+
{
|
| 2171 |
+
"epoch": 1.07,
|
| 2172 |
+
"learning_rate": 2.22934336456009e-05,
|
| 2173 |
+
"loss": 0.7524,
|
| 2174 |
+
"step": 3610
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 1.07,
|
| 2178 |
+
"learning_rate": 2.217775772717135e-05,
|
| 2179 |
+
"loss": 0.744,
|
| 2180 |
+
"step": 3620
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.08,
|
| 2184 |
+
"learning_rate": 2.2062142979393265e-05,
|
| 2185 |
+
"loss": 0.7679,
|
| 2186 |
+
"step": 3630
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 1.08,
|
| 2190 |
+
"learning_rate": 2.1946591908156918e-05,
|
| 2191 |
+
"loss": 0.7633,
|
| 2192 |
+
"step": 3640
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 1.08,
|
| 2196 |
+
"learning_rate": 2.1831107017972414e-05,
|
| 2197 |
+
"loss": 0.7735,
|
| 2198 |
+
"step": 3650
|
| 2199 |
+
},
|
| 2200 |
+
{
|
| 2201 |
+
"epoch": 1.08,
|
| 2202 |
+
"learning_rate": 2.1715690811915424e-05,
|
| 2203 |
+
"loss": 0.7577,
|
| 2204 |
+
"step": 3660
|
| 2205 |
+
},
|
| 2206 |
+
{
|
| 2207 |
+
"epoch": 1.09,
|
| 2208 |
+
"learning_rate": 2.1600345791572923e-05,
|
| 2209 |
+
"loss": 0.765,
|
| 2210 |
+
"step": 3670
|
| 2211 |
+
},
|
| 2212 |
+
{
|
| 2213 |
+
"epoch": 1.09,
|
| 2214 |
+
"learning_rate": 2.148507445698898e-05,
|
| 2215 |
+
"loss": 0.7444,
|
| 2216 |
+
"step": 3680
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 1.09,
|
| 2220 |
+
"learning_rate": 2.1369879306610548e-05,
|
| 2221 |
+
"loss": 0.7527,
|
| 2222 |
+
"step": 3690
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.1,
|
| 2226 |
+
"learning_rate": 2.125476283723333e-05,
|
| 2227 |
+
"loss": 0.7679,
|
| 2228 |
+
"step": 3700
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 1.1,
|
| 2232 |
+
"learning_rate": 2.1139727543947687e-05,
|
| 2233 |
+
"loss": 0.765,
|
| 2234 |
+
"step": 3710
|
| 2235 |
+
},
|
| 2236 |
+
{
|
| 2237 |
+
"epoch": 1.1,
|
| 2238 |
+
"learning_rate": 2.1024775920084488e-05,
|
| 2239 |
+
"loss": 0.7662,
|
| 2240 |
+
"step": 3720
|
| 2241 |
+
},
|
| 2242 |
+
{
|
| 2243 |
+
"epoch": 1.11,
|
| 2244 |
+
"learning_rate": 2.0909910457161136e-05,
|
| 2245 |
+
"loss": 0.7583,
|
| 2246 |
+
"step": 3730
|
| 2247 |
+
},
|
| 2248 |
+
{
|
| 2249 |
+
"epoch": 1.11,
|
| 2250 |
+
"learning_rate": 2.079513364482753e-05,
|
| 2251 |
+
"loss": 0.7584,
|
| 2252 |
+
"step": 3740
|
| 2253 |
+
},
|
| 2254 |
+
{
|
| 2255 |
+
"epoch": 1.11,
|
| 2256 |
+
"learning_rate": 2.0680447970812123e-05,
|
| 2257 |
+
"loss": 0.7606,
|
| 2258 |
+
"step": 3750
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 1.11,
|
| 2262 |
+
"learning_rate": 2.056585592086798e-05,
|
| 2263 |
+
"loss": 0.7488,
|
| 2264 |
+
"step": 3760
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.12,
|
| 2268 |
+
"learning_rate": 2.0451359978718913e-05,
|
| 2269 |
+
"loss": 0.7525,
|
| 2270 |
+
"step": 3770
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 1.12,
|
| 2274 |
+
"learning_rate": 2.0336962626005654e-05,
|
| 2275 |
+
"loss": 0.7628,
|
| 2276 |
+
"step": 3780
|
| 2277 |
+
},
|
| 2278 |
+
{
|
| 2279 |
+
"epoch": 1.12,
|
| 2280 |
+
"learning_rate": 2.022266634223206e-05,
|
| 2281 |
+
"loss": 0.7413,
|
| 2282 |
+
"step": 3790
|
| 2283 |
+
},
|
| 2284 |
+
{
|
| 2285 |
+
"epoch": 1.13,
|
| 2286 |
+
"learning_rate": 2.0108473604711368e-05,
|
| 2287 |
+
"loss": 0.761,
|
| 2288 |
+
"step": 3800
|
| 2289 |
+
},
|
| 2290 |
+
{
|
| 2291 |
+
"epoch": 1.13,
|
| 2292 |
+
"learning_rate": 1.99943868885125e-05,
|
| 2293 |
+
"loss": 0.7474,
|
| 2294 |
+
"step": 3810
|
| 2295 |
+
},
|
| 2296 |
+
{
|
| 2297 |
+
"epoch": 1.13,
|
| 2298 |
+
"learning_rate": 1.988040866640642e-05,
|
| 2299 |
+
"loss": 0.7581,
|
| 2300 |
+
"step": 3820
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 1.13,
|
| 2304 |
+
"learning_rate": 1.9766541408812537e-05,
|
| 2305 |
+
"loss": 0.7464,
|
| 2306 |
+
"step": 3830
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.14,
|
| 2310 |
+
"learning_rate": 1.965278758374518e-05,
|
| 2311 |
+
"loss": 0.7629,
|
| 2312 |
+
"step": 3840
|
| 2313 |
+
},
|
| 2314 |
+
{
|
| 2315 |
+
"epoch": 1.14,
|
| 2316 |
+
"learning_rate": 1.953914965676006e-05,
|
| 2317 |
+
"loss": 0.7509,
|
| 2318 |
+
"step": 3850
|
| 2319 |
+
},
|
| 2320 |
+
{
|
| 2321 |
+
"epoch": 1.14,
|
| 2322 |
+
"learning_rate": 1.9425630090900877e-05,
|
| 2323 |
+
"loss": 0.7564,
|
| 2324 |
+
"step": 3860
|
| 2325 |
+
},
|
| 2326 |
+
{
|
| 2327 |
+
"epoch": 1.15,
|
| 2328 |
+
"learning_rate": 1.9312231346645892e-05,
|
| 2329 |
+
"loss": 0.7601,
|
| 2330 |
+
"step": 3870
|
| 2331 |
+
},
|
| 2332 |
+
{
|
| 2333 |
+
"epoch": 1.15,
|
| 2334 |
+
"learning_rate": 1.9198955881854645e-05,
|
| 2335 |
+
"loss": 0.7694,
|
| 2336 |
+
"step": 3880
|
| 2337 |
+
},
|
| 2338 |
+
{
|
| 2339 |
+
"epoch": 1.15,
|
| 2340 |
+
"learning_rate": 1.9085806151714645e-05,
|
| 2341 |
+
"loss": 0.7593,
|
| 2342 |
+
"step": 3890
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 1.16,
|
| 2346 |
+
"learning_rate": 1.8972784608688164e-05,
|
| 2347 |
+
"loss": 0.7503,
|
| 2348 |
+
"step": 3900
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.16,
|
| 2352 |
+
"learning_rate": 1.8859893702459076e-05,
|
| 2353 |
+
"loss": 0.748,
|
| 2354 |
+
"step": 3910
|
| 2355 |
+
},
|
| 2356 |
+
{
|
| 2357 |
+
"epoch": 1.16,
|
| 2358 |
+
"learning_rate": 1.8747135879879802e-05,
|
| 2359 |
+
"loss": 0.7675,
|
| 2360 |
+
"step": 3920
|
| 2361 |
+
},
|
| 2362 |
+
{
|
| 2363 |
+
"epoch": 1.16,
|
| 2364 |
+
"learning_rate": 1.8634513584918217e-05,
|
| 2365 |
+
"loss": 0.7553,
|
| 2366 |
+
"step": 3930
|
| 2367 |
+
},
|
| 2368 |
+
{
|
| 2369 |
+
"epoch": 1.17,
|
| 2370 |
+
"learning_rate": 1.85220292586047e-05,
|
| 2371 |
+
"loss": 0.7511,
|
| 2372 |
+
"step": 3940
|
| 2373 |
+
},
|
| 2374 |
+
{
|
| 2375 |
+
"epoch": 1.17,
|
| 2376 |
+
"learning_rate": 1.8409685338979255e-05,
|
| 2377 |
+
"loss": 0.7533,
|
| 2378 |
+
"step": 3950
|
| 2379 |
+
},
|
| 2380 |
+
{
|
| 2381 |
+
"epoch": 1.17,
|
| 2382 |
+
"learning_rate": 1.8297484261038612e-05,
|
| 2383 |
+
"loss": 0.7553,
|
| 2384 |
+
"step": 3960
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 1.18,
|
| 2388 |
+
"learning_rate": 1.8185428456683513e-05,
|
| 2389 |
+
"loss": 0.7563,
|
| 2390 |
+
"step": 3970
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.18,
|
| 2394 |
+
"learning_rate": 1.807352035466594e-05,
|
| 2395 |
+
"loss": 0.7512,
|
| 2396 |
+
"step": 3980
|
| 2397 |
+
},
|
| 2398 |
+
{
|
| 2399 |
+
"epoch": 1.18,
|
| 2400 |
+
"learning_rate": 1.7961762380536522e-05,
|
| 2401 |
+
"loss": 0.7502,
|
| 2402 |
+
"step": 3990
|
| 2403 |
+
},
|
| 2404 |
+
{
|
| 2405 |
+
"epoch": 1.19,
|
| 2406 |
+
"learning_rate": 1.7850156956591943e-05,
|
| 2407 |
+
"loss": 0.7614,
|
| 2408 |
+
"step": 4000
|
| 2409 |
+
}
|
| 2410 |
+
],
|
| 2411 |
+
"max_steps": 6748,
|
| 2412 |
+
"num_train_epochs": 2,
|
| 2413 |
+
"total_flos": 7.117499263600296e+18,
|
| 2414 |
+
"trial_name": null,
|
| 2415 |
+
"trial_params": null
|
| 2416 |
+
}
|
checkpoint-4000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f710e8a04dbb33457b5b725adb1da191a7b05d7ab01f6ec60c73386df9d8286
|
| 3 |
+
size 3277
|
checkpoint-5000/README.md
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
---
|
| 4 |
+
## Training procedure
|
| 5 |
+
|
| 6 |
+
|
| 7 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 8 |
+
- load_in_8bit: False
|
| 9 |
+
- load_in_4bit: True
|
| 10 |
+
- llm_int8_threshold: 6.0
|
| 11 |
+
- llm_int8_skip_modules: None
|
| 12 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 13 |
+
- llm_int8_has_fp16_weight: False
|
| 14 |
+
- bnb_4bit_quant_type: nf4
|
| 15 |
+
- bnb_4bit_use_double_quant: True
|
| 16 |
+
- bnb_4bit_compute_dtype: float16
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.4.0
|
checkpoint-5000/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "./Llama-2-7b-chat-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 32.0,
|
| 11 |
+
"lora_dropout": 0.1,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|