Commit
•
ca643a8
1
Parent(s):
0854026
added all files
Browse files- README.md +20 -0
- adapter_config.json +21 -0
- adapter_model.bin +3 -0
- checkpoint-1000/README.md +20 -0
- checkpoint-1000/adapter_config.json +21 -0
- checkpoint-1000/adapter_model.bin +3 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/rng_state.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/trainer_state.json +656 -0
- checkpoint-1000/training_args.bin +3 -0
- checkpoint-1200/README.md +20 -0
- checkpoint-1200/adapter_config.json +21 -0
- checkpoint-1200/adapter_model.bin +3 -0
- checkpoint-1200/optimizer.pt +3 -0
- checkpoint-1200/rng_state.pth +3 -0
- checkpoint-1200/scheduler.pt +3 -0
- checkpoint-1200/trainer_state.json +784 -0
- checkpoint-1200/training_args.bin +3 -0
- checkpoint-400/README.md +20 -0
- checkpoint-400/adapter_config.json +21 -0
- checkpoint-400/adapter_model.bin +3 -0
- checkpoint-400/optimizer.pt +3 -0
- checkpoint-400/rng_state.pth +3 -0
- checkpoint-400/scheduler.pt +3 -0
- checkpoint-400/trainer_state.json +272 -0
- checkpoint-400/training_args.bin +3 -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: True
|
| 9 |
+
- load_in_4bit: False
|
| 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: fp4
|
| 15 |
+
- bnb_4bit_use_double_quant: False
|
| 16 |
+
- bnb_4bit_compute_dtype: float32
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.5.0.dev0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "decapoda-research/llama-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 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:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
|
| 3 |
+
size 443
|
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: True
|
| 9 |
+
- load_in_4bit: False
|
| 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: fp4
|
| 15 |
+
- bnb_4bit_use_double_quant: False
|
| 16 |
+
- bnb_4bit_compute_dtype: float32
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.5.0.dev0
|
checkpoint-1000/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "decapoda-research/llama-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 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:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
|
| 3 |
+
size 443
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d0571d8bc50c48174224763a862b12c408a791eae50df287a862c8cd0014b265
|
| 3 |
+
size 33661637
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:174b00cdc553fa2e958c352598f8e4dc68598c5cc8544a0b2a288cdbbb214635
|
| 3 |
+
size 14575
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7e544c4a4fc8e9078624ab7727e43288ef28532a4fdd174337c0af70547061f
|
| 3 |
+
size 627
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,656 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8681809306144714,
|
| 3 |
+
"best_model_checkpoint": "./lora-warewe/checkpoint-400",
|
| 4 |
+
"epoch": 2.483122526577171,
|
| 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": 0.02,
|
| 12 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 13 |
+
"loss": 1.9531,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 19 |
+
"loss": 1.8714,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 8.999999999999999e-05,
|
| 25 |
+
"loss": 1.7052,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 0.00011999999999999999,
|
| 31 |
+
"loss": 1.3201,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.12,
|
| 36 |
+
"learning_rate": 0.00015,
|
| 37 |
+
"loss": 1.1551,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 0.00017999999999999998,
|
| 43 |
+
"loss": 1.1199,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
+
"learning_rate": 0.00020999999999999998,
|
| 49 |
+
"loss": 1.0598,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 0.00023999999999999998,
|
| 55 |
+
"loss": 1.0233,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.22,
|
| 60 |
+
"learning_rate": 0.00027,
|
| 61 |
+
"loss": 1.0086,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
+
"learning_rate": 0.0003,
|
| 67 |
+
"loss": 0.999,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.27,
|
| 72 |
+
"learning_rate": 0.0002972875226039783,
|
| 73 |
+
"loss": 0.9961,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3,
|
| 78 |
+
"learning_rate": 0.00029457504520795656,
|
| 79 |
+
"loss": 0.9831,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.32,
|
| 84 |
+
"learning_rate": 0.0002918625678119349,
|
| 85 |
+
"loss": 0.9637,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.35,
|
| 90 |
+
"learning_rate": 0.00028915009041591315,
|
| 91 |
+
"loss": 0.9546,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.37,
|
| 96 |
+
"learning_rate": 0.0002864376130198915,
|
| 97 |
+
"loss": 0.9575,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.4,
|
| 102 |
+
"learning_rate": 0.00028372513562386974,
|
| 103 |
+
"loss": 0.9337,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.42,
|
| 108 |
+
"learning_rate": 0.00028101265822784807,
|
| 109 |
+
"loss": 0.942,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.45,
|
| 114 |
+
"learning_rate": 0.0002783001808318264,
|
| 115 |
+
"loss": 0.9265,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.47,
|
| 120 |
+
"learning_rate": 0.0002755877034358047,
|
| 121 |
+
"loss": 0.9324,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.5,
|
| 126 |
+
"learning_rate": 0.000272875226039783,
|
| 127 |
+
"loss": 0.9288,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5,
|
| 132 |
+
"eval_loss": 0.9210162162780762,
|
| 133 |
+
"eval_runtime": 559.4854,
|
| 134 |
+
"eval_samples_per_second": 3.575,
|
| 135 |
+
"eval_steps_per_second": 0.447,
|
| 136 |
+
"step": 200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.52,
|
| 140 |
+
"learning_rate": 0.00027016274864376125,
|
| 141 |
+
"loss": 0.9201,
|
| 142 |
+
"step": 210
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.55,
|
| 146 |
+
"learning_rate": 0.00026745027124773957,
|
| 147 |
+
"loss": 0.9242,
|
| 148 |
+
"step": 220
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.57,
|
| 152 |
+
"learning_rate": 0.0002647377938517179,
|
| 153 |
+
"loss": 0.9114,
|
| 154 |
+
"step": 230
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.6,
|
| 158 |
+
"learning_rate": 0.00026202531645569616,
|
| 159 |
+
"loss": 0.9223,
|
| 160 |
+
"step": 240
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.62,
|
| 164 |
+
"learning_rate": 0.0002593128390596745,
|
| 165 |
+
"loss": 0.9018,
|
| 166 |
+
"step": 250
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.65,
|
| 170 |
+
"learning_rate": 0.0002566003616636528,
|
| 171 |
+
"loss": 0.9038,
|
| 172 |
+
"step": 260
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.67,
|
| 176 |
+
"learning_rate": 0.0002538878842676311,
|
| 177 |
+
"loss": 0.9087,
|
| 178 |
+
"step": 270
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.7,
|
| 182 |
+
"learning_rate": 0.0002511754068716094,
|
| 183 |
+
"loss": 0.9141,
|
| 184 |
+
"step": 280
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.72,
|
| 188 |
+
"learning_rate": 0.00024846292947558766,
|
| 189 |
+
"loss": 0.8942,
|
| 190 |
+
"step": 290
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.74,
|
| 194 |
+
"learning_rate": 0.000245750452079566,
|
| 195 |
+
"loss": 0.8865,
|
| 196 |
+
"step": 300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.77,
|
| 200 |
+
"learning_rate": 0.00024303797468354428,
|
| 201 |
+
"loss": 0.8863,
|
| 202 |
+
"step": 310
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.79,
|
| 206 |
+
"learning_rate": 0.00024032549728752258,
|
| 207 |
+
"loss": 0.8981,
|
| 208 |
+
"step": 320
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.82,
|
| 212 |
+
"learning_rate": 0.0002376130198915009,
|
| 213 |
+
"loss": 0.8837,
|
| 214 |
+
"step": 330
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.84,
|
| 218 |
+
"learning_rate": 0.0002349005424954792,
|
| 219 |
+
"loss": 0.8887,
|
| 220 |
+
"step": 340
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.87,
|
| 224 |
+
"learning_rate": 0.0002321880650994575,
|
| 225 |
+
"loss": 0.8781,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.89,
|
| 230 |
+
"learning_rate": 0.00022947558770343578,
|
| 231 |
+
"loss": 0.891,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.92,
|
| 236 |
+
"learning_rate": 0.00022676311030741408,
|
| 237 |
+
"loss": 0.8852,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.94,
|
| 242 |
+
"learning_rate": 0.00022405063291139237,
|
| 243 |
+
"loss": 0.8736,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.97,
|
| 248 |
+
"learning_rate": 0.0002213381555153707,
|
| 249 |
+
"loss": 0.8799,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.99,
|
| 254 |
+
"learning_rate": 0.000218625678119349,
|
| 255 |
+
"loss": 0.8847,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.99,
|
| 260 |
+
"eval_loss": 0.8681809306144714,
|
| 261 |
+
"eval_runtime": 405.865,
|
| 262 |
+
"eval_samples_per_second": 4.928,
|
| 263 |
+
"eval_steps_per_second": 0.616,
|
| 264 |
+
"step": 400
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.02,
|
| 268 |
+
"learning_rate": 0.0002159132007233273,
|
| 269 |
+
"loss": 0.8731,
|
| 270 |
+
"step": 410
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.04,
|
| 274 |
+
"learning_rate": 0.0002132007233273056,
|
| 275 |
+
"loss": 0.8677,
|
| 276 |
+
"step": 420
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.07,
|
| 280 |
+
"learning_rate": 0.00021048824593128388,
|
| 281 |
+
"loss": 0.8784,
|
| 282 |
+
"step": 430
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.09,
|
| 286 |
+
"learning_rate": 0.00020777576853526217,
|
| 287 |
+
"loss": 0.8688,
|
| 288 |
+
"step": 440
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.12,
|
| 292 |
+
"learning_rate": 0.0002050632911392405,
|
| 293 |
+
"loss": 0.8663,
|
| 294 |
+
"step": 450
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.14,
|
| 298 |
+
"learning_rate": 0.0002023508137432188,
|
| 299 |
+
"loss": 0.8748,
|
| 300 |
+
"step": 460
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.17,
|
| 304 |
+
"learning_rate": 0.00019963833634719708,
|
| 305 |
+
"loss": 0.8619,
|
| 306 |
+
"step": 470
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.19,
|
| 310 |
+
"learning_rate": 0.0001969258589511754,
|
| 311 |
+
"loss": 0.8544,
|
| 312 |
+
"step": 480
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.22,
|
| 316 |
+
"learning_rate": 0.0001942133815551537,
|
| 317 |
+
"loss": 0.871,
|
| 318 |
+
"step": 490
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.24,
|
| 322 |
+
"learning_rate": 0.000191500904159132,
|
| 323 |
+
"loss": 0.8748,
|
| 324 |
+
"step": 500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.27,
|
| 328 |
+
"learning_rate": 0.00018933092224231464,
|
| 329 |
+
"loss": 0.8753,
|
| 330 |
+
"step": 510
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.29,
|
| 334 |
+
"learning_rate": 0.00018824593128390595,
|
| 335 |
+
"loss": 1.8069,
|
| 336 |
+
"step": 520
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.32,
|
| 340 |
+
"learning_rate": 0.00018634719710669076,
|
| 341 |
+
"loss": 4.3811,
|
| 342 |
+
"step": 530
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.34,
|
| 346 |
+
"learning_rate": 0.00018390596745027123,
|
| 347 |
+
"loss": 5.8063,
|
| 348 |
+
"step": 540
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.37,
|
| 352 |
+
"learning_rate": 0.00018119349005424955,
|
| 353 |
+
"loss": 5.2762,
|
| 354 |
+
"step": 550
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.39,
|
| 358 |
+
"learning_rate": 0.00017848101265822782,
|
| 359 |
+
"loss": 3.0862,
|
| 360 |
+
"step": 560
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.42,
|
| 364 |
+
"learning_rate": 0.00017576853526220611,
|
| 365 |
+
"loss": 1.7397,
|
| 366 |
+
"step": 570
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.44,
|
| 370 |
+
"learning_rate": 0.00017305605786618444,
|
| 371 |
+
"loss": 1.5362,
|
| 372 |
+
"step": 580
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.47,
|
| 376 |
+
"learning_rate": 0.0001706148282097649,
|
| 377 |
+
"loss": 3.644,
|
| 378 |
+
"step": 590
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.49,
|
| 382 |
+
"learning_rate": 0.0001679023508137432,
|
| 383 |
+
"loss": 6.1529,
|
| 384 |
+
"step": 600
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.49,
|
| 388 |
+
"eval_loss": 6.275171756744385,
|
| 389 |
+
"eval_runtime": 401.4674,
|
| 390 |
+
"eval_samples_per_second": 4.982,
|
| 391 |
+
"eval_steps_per_second": 0.623,
|
| 392 |
+
"step": 600
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.51,
|
| 396 |
+
"learning_rate": 0.00016573236889692584,
|
| 397 |
+
"loss": 4.6504,
|
| 398 |
+
"step": 610
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.54,
|
| 402 |
+
"learning_rate": 0.00016301989150090414,
|
| 403 |
+
"loss": 3.3039,
|
| 404 |
+
"step": 620
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.56,
|
| 408 |
+
"learning_rate": 0.00016030741410488246,
|
| 409 |
+
"loss": 2.7266,
|
| 410 |
+
"step": 630
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.59,
|
| 414 |
+
"learning_rate": 0.00015759493670886075,
|
| 415 |
+
"loss": 4.5256,
|
| 416 |
+
"step": 640
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.61,
|
| 420 |
+
"learning_rate": 0.0001551537070524412,
|
| 421 |
+
"loss": 3.9326,
|
| 422 |
+
"step": 650
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.64,
|
| 426 |
+
"learning_rate": 0.00015244122965641952,
|
| 427 |
+
"loss": 1.6396,
|
| 428 |
+
"step": 660
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.66,
|
| 432 |
+
"learning_rate": 0.00014972875226039781,
|
| 433 |
+
"loss": 1.1304,
|
| 434 |
+
"step": 670
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.69,
|
| 438 |
+
"learning_rate": 0.0001470162748643761,
|
| 439 |
+
"loss": 1.0282,
|
| 440 |
+
"step": 680
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.71,
|
| 444 |
+
"learning_rate": 0.00014430379746835443,
|
| 445 |
+
"loss": 0.9902,
|
| 446 |
+
"step": 690
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.74,
|
| 450 |
+
"learning_rate": 0.00014159132007233273,
|
| 451 |
+
"loss": 0.9785,
|
| 452 |
+
"step": 700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.76,
|
| 456 |
+
"learning_rate": 0.00013887884267631102,
|
| 457 |
+
"loss": 0.9422,
|
| 458 |
+
"step": 710
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.79,
|
| 462 |
+
"learning_rate": 0.00013616636528028932,
|
| 463 |
+
"loss": 0.9466,
|
| 464 |
+
"step": 720
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.81,
|
| 468 |
+
"learning_rate": 0.0001334538878842676,
|
| 469 |
+
"loss": 0.9254,
|
| 470 |
+
"step": 730
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.84,
|
| 474 |
+
"learning_rate": 0.0001307414104882459,
|
| 475 |
+
"loss": 0.9257,
|
| 476 |
+
"step": 740
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.86,
|
| 480 |
+
"learning_rate": 0.00012802893309222423,
|
| 481 |
+
"loss": 0.9179,
|
| 482 |
+
"step": 750
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.89,
|
| 486 |
+
"learning_rate": 0.00012531645569620252,
|
| 487 |
+
"loss": 0.9247,
|
| 488 |
+
"step": 760
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.91,
|
| 492 |
+
"learning_rate": 0.00012260397830018082,
|
| 493 |
+
"loss": 0.9186,
|
| 494 |
+
"step": 770
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.94,
|
| 498 |
+
"learning_rate": 0.00011989150090415911,
|
| 499 |
+
"loss": 0.9253,
|
| 500 |
+
"step": 780
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.96,
|
| 504 |
+
"learning_rate": 0.00011717902350813742,
|
| 505 |
+
"loss": 0.9193,
|
| 506 |
+
"step": 790
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.99,
|
| 510 |
+
"learning_rate": 0.00011446654611211573,
|
| 511 |
+
"loss": 0.9141,
|
| 512 |
+
"step": 800
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.99,
|
| 516 |
+
"eval_loss": 0.9085212349891663,
|
| 517 |
+
"eval_runtime": 400.6517,
|
| 518 |
+
"eval_samples_per_second": 4.992,
|
| 519 |
+
"eval_steps_per_second": 0.624,
|
| 520 |
+
"step": 800
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.01,
|
| 524 |
+
"learning_rate": 0.00011175406871609403,
|
| 525 |
+
"loss": 0.9057,
|
| 526 |
+
"step": 810
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.04,
|
| 530 |
+
"learning_rate": 0.00010904159132007232,
|
| 531 |
+
"loss": 0.9083,
|
| 532 |
+
"step": 820
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.06,
|
| 536 |
+
"learning_rate": 0.00010632911392405062,
|
| 537 |
+
"loss": 0.9091,
|
| 538 |
+
"step": 830
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 2.09,
|
| 542 |
+
"learning_rate": 0.00010361663652802893,
|
| 543 |
+
"loss": 0.9186,
|
| 544 |
+
"step": 840
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 2.11,
|
| 548 |
+
"learning_rate": 0.00010090415913200722,
|
| 549 |
+
"loss": 0.9018,
|
| 550 |
+
"step": 850
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.14,
|
| 554 |
+
"learning_rate": 9.819168173598552e-05,
|
| 555 |
+
"loss": 0.8989,
|
| 556 |
+
"step": 860
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 2.16,
|
| 560 |
+
"learning_rate": 9.547920433996383e-05,
|
| 561 |
+
"loss": 0.9275,
|
| 562 |
+
"step": 870
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.19,
|
| 566 |
+
"learning_rate": 9.276672694394213e-05,
|
| 567 |
+
"loss": 0.9063,
|
| 568 |
+
"step": 880
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.21,
|
| 572 |
+
"learning_rate": 9.005424954792042e-05,
|
| 573 |
+
"loss": 0.8921,
|
| 574 |
+
"step": 890
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 2.23,
|
| 578 |
+
"learning_rate": 8.734177215189872e-05,
|
| 579 |
+
"loss": 0.8981,
|
| 580 |
+
"step": 900
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.26,
|
| 584 |
+
"learning_rate": 8.462929475587703e-05,
|
| 585 |
+
"loss": 0.9099,
|
| 586 |
+
"step": 910
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.28,
|
| 590 |
+
"learning_rate": 8.191681735985534e-05,
|
| 591 |
+
"loss": 0.8956,
|
| 592 |
+
"step": 920
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.31,
|
| 596 |
+
"learning_rate": 7.920433996383362e-05,
|
| 597 |
+
"loss": 0.8983,
|
| 598 |
+
"step": 930
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.33,
|
| 602 |
+
"learning_rate": 7.649186256781193e-05,
|
| 603 |
+
"loss": 0.9192,
|
| 604 |
+
"step": 940
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.36,
|
| 608 |
+
"learning_rate": 7.377938517179023e-05,
|
| 609 |
+
"loss": 0.9199,
|
| 610 |
+
"step": 950
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 2.38,
|
| 614 |
+
"learning_rate": 7.106690777576854e-05,
|
| 615 |
+
"loss": 0.8857,
|
| 616 |
+
"step": 960
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.41,
|
| 620 |
+
"learning_rate": 6.835443037974683e-05,
|
| 621 |
+
"loss": 0.8871,
|
| 622 |
+
"step": 970
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.43,
|
| 626 |
+
"learning_rate": 6.564195298372513e-05,
|
| 627 |
+
"loss": 0.9054,
|
| 628 |
+
"step": 980
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.46,
|
| 632 |
+
"learning_rate": 6.292947558770344e-05,
|
| 633 |
+
"loss": 0.9167,
|
| 634 |
+
"step": 990
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.48,
|
| 638 |
+
"learning_rate": 6.021699819168174e-05,
|
| 639 |
+
"loss": 0.9023,
|
| 640 |
+
"step": 1000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.48,
|
| 644 |
+
"eval_loss": 0.9026673436164856,
|
| 645 |
+
"eval_runtime": 401.9867,
|
| 646 |
+
"eval_samples_per_second": 4.975,
|
| 647 |
+
"eval_steps_per_second": 0.622,
|
| 648 |
+
"step": 1000
|
| 649 |
+
}
|
| 650 |
+
],
|
| 651 |
+
"max_steps": 1206,
|
| 652 |
+
"num_train_epochs": 3,
|
| 653 |
+
"total_flos": 1.256386002199511e+18,
|
| 654 |
+
"trial_name": null,
|
| 655 |
+
"trial_params": null
|
| 656 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0abdd9155585c4e11f0ad9dcfdd6c80dde474221e48df2dc7b337502cbe5dda9
|
| 3 |
+
size 3963
|
checkpoint-1200/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: True
|
| 9 |
+
- load_in_4bit: False
|
| 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: fp4
|
| 15 |
+
- bnb_4bit_use_double_quant: False
|
| 16 |
+
- bnb_4bit_compute_dtype: float32
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.5.0.dev0
|
checkpoint-1200/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "decapoda-research/llama-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-1200/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
|
| 3 |
+
size 443
|
checkpoint-1200/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:373bb964597f8b53a2f95f174acad9572287fcbf35b503b4f6d15017735b85c7
|
| 3 |
+
size 33661637
|
checkpoint-1200/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b20589bb0717142795c61c218abd6727d306a5a34678eacb70af57198a415ad5
|
| 3 |
+
size 14575
|
checkpoint-1200/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a723c813649a929c45e446b1f3fa3a8ff7bd225988798b70d8e7c28eac965548
|
| 3 |
+
size 627
|
checkpoint-1200/trainer_state.json
ADDED
|
@@ -0,0 +1,784 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8681809306144714,
|
| 3 |
+
"best_model_checkpoint": "./lora-warewe/checkpoint-400",
|
| 4 |
+
"epoch": 2.9797470318926047,
|
| 5 |
+
"global_step": 1200,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 13 |
+
"loss": 1.9531,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 19 |
+
"loss": 1.8714,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 8.999999999999999e-05,
|
| 25 |
+
"loss": 1.7052,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 0.00011999999999999999,
|
| 31 |
+
"loss": 1.3201,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.12,
|
| 36 |
+
"learning_rate": 0.00015,
|
| 37 |
+
"loss": 1.1551,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 0.00017999999999999998,
|
| 43 |
+
"loss": 1.1199,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
+
"learning_rate": 0.00020999999999999998,
|
| 49 |
+
"loss": 1.0598,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 0.00023999999999999998,
|
| 55 |
+
"loss": 1.0233,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.22,
|
| 60 |
+
"learning_rate": 0.00027,
|
| 61 |
+
"loss": 1.0086,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
+
"learning_rate": 0.0003,
|
| 67 |
+
"loss": 0.999,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.27,
|
| 72 |
+
"learning_rate": 0.0002972875226039783,
|
| 73 |
+
"loss": 0.9961,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3,
|
| 78 |
+
"learning_rate": 0.00029457504520795656,
|
| 79 |
+
"loss": 0.9831,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.32,
|
| 84 |
+
"learning_rate": 0.0002918625678119349,
|
| 85 |
+
"loss": 0.9637,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.35,
|
| 90 |
+
"learning_rate": 0.00028915009041591315,
|
| 91 |
+
"loss": 0.9546,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.37,
|
| 96 |
+
"learning_rate": 0.0002864376130198915,
|
| 97 |
+
"loss": 0.9575,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.4,
|
| 102 |
+
"learning_rate": 0.00028372513562386974,
|
| 103 |
+
"loss": 0.9337,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.42,
|
| 108 |
+
"learning_rate": 0.00028101265822784807,
|
| 109 |
+
"loss": 0.942,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.45,
|
| 114 |
+
"learning_rate": 0.0002783001808318264,
|
| 115 |
+
"loss": 0.9265,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.47,
|
| 120 |
+
"learning_rate": 0.0002755877034358047,
|
| 121 |
+
"loss": 0.9324,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.5,
|
| 126 |
+
"learning_rate": 0.000272875226039783,
|
| 127 |
+
"loss": 0.9288,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5,
|
| 132 |
+
"eval_loss": 0.9210162162780762,
|
| 133 |
+
"eval_runtime": 559.4854,
|
| 134 |
+
"eval_samples_per_second": 3.575,
|
| 135 |
+
"eval_steps_per_second": 0.447,
|
| 136 |
+
"step": 200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.52,
|
| 140 |
+
"learning_rate": 0.00027016274864376125,
|
| 141 |
+
"loss": 0.9201,
|
| 142 |
+
"step": 210
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.55,
|
| 146 |
+
"learning_rate": 0.00026745027124773957,
|
| 147 |
+
"loss": 0.9242,
|
| 148 |
+
"step": 220
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.57,
|
| 152 |
+
"learning_rate": 0.0002647377938517179,
|
| 153 |
+
"loss": 0.9114,
|
| 154 |
+
"step": 230
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.6,
|
| 158 |
+
"learning_rate": 0.00026202531645569616,
|
| 159 |
+
"loss": 0.9223,
|
| 160 |
+
"step": 240
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.62,
|
| 164 |
+
"learning_rate": 0.0002593128390596745,
|
| 165 |
+
"loss": 0.9018,
|
| 166 |
+
"step": 250
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.65,
|
| 170 |
+
"learning_rate": 0.0002566003616636528,
|
| 171 |
+
"loss": 0.9038,
|
| 172 |
+
"step": 260
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.67,
|
| 176 |
+
"learning_rate": 0.0002538878842676311,
|
| 177 |
+
"loss": 0.9087,
|
| 178 |
+
"step": 270
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.7,
|
| 182 |
+
"learning_rate": 0.0002511754068716094,
|
| 183 |
+
"loss": 0.9141,
|
| 184 |
+
"step": 280
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.72,
|
| 188 |
+
"learning_rate": 0.00024846292947558766,
|
| 189 |
+
"loss": 0.8942,
|
| 190 |
+
"step": 290
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.74,
|
| 194 |
+
"learning_rate": 0.000245750452079566,
|
| 195 |
+
"loss": 0.8865,
|
| 196 |
+
"step": 300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.77,
|
| 200 |
+
"learning_rate": 0.00024303797468354428,
|
| 201 |
+
"loss": 0.8863,
|
| 202 |
+
"step": 310
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.79,
|
| 206 |
+
"learning_rate": 0.00024032549728752258,
|
| 207 |
+
"loss": 0.8981,
|
| 208 |
+
"step": 320
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.82,
|
| 212 |
+
"learning_rate": 0.0002376130198915009,
|
| 213 |
+
"loss": 0.8837,
|
| 214 |
+
"step": 330
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.84,
|
| 218 |
+
"learning_rate": 0.0002349005424954792,
|
| 219 |
+
"loss": 0.8887,
|
| 220 |
+
"step": 340
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.87,
|
| 224 |
+
"learning_rate": 0.0002321880650994575,
|
| 225 |
+
"loss": 0.8781,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.89,
|
| 230 |
+
"learning_rate": 0.00022947558770343578,
|
| 231 |
+
"loss": 0.891,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.92,
|
| 236 |
+
"learning_rate": 0.00022676311030741408,
|
| 237 |
+
"loss": 0.8852,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.94,
|
| 242 |
+
"learning_rate": 0.00022405063291139237,
|
| 243 |
+
"loss": 0.8736,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.97,
|
| 248 |
+
"learning_rate": 0.0002213381555153707,
|
| 249 |
+
"loss": 0.8799,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.99,
|
| 254 |
+
"learning_rate": 0.000218625678119349,
|
| 255 |
+
"loss": 0.8847,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.99,
|
| 260 |
+
"eval_loss": 0.8681809306144714,
|
| 261 |
+
"eval_runtime": 405.865,
|
| 262 |
+
"eval_samples_per_second": 4.928,
|
| 263 |
+
"eval_steps_per_second": 0.616,
|
| 264 |
+
"step": 400
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.02,
|
| 268 |
+
"learning_rate": 0.0002159132007233273,
|
| 269 |
+
"loss": 0.8731,
|
| 270 |
+
"step": 410
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.04,
|
| 274 |
+
"learning_rate": 0.0002132007233273056,
|
| 275 |
+
"loss": 0.8677,
|
| 276 |
+
"step": 420
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.07,
|
| 280 |
+
"learning_rate": 0.00021048824593128388,
|
| 281 |
+
"loss": 0.8784,
|
| 282 |
+
"step": 430
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.09,
|
| 286 |
+
"learning_rate": 0.00020777576853526217,
|
| 287 |
+
"loss": 0.8688,
|
| 288 |
+
"step": 440
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.12,
|
| 292 |
+
"learning_rate": 0.0002050632911392405,
|
| 293 |
+
"loss": 0.8663,
|
| 294 |
+
"step": 450
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.14,
|
| 298 |
+
"learning_rate": 0.0002023508137432188,
|
| 299 |
+
"loss": 0.8748,
|
| 300 |
+
"step": 460
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.17,
|
| 304 |
+
"learning_rate": 0.00019963833634719708,
|
| 305 |
+
"loss": 0.8619,
|
| 306 |
+
"step": 470
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.19,
|
| 310 |
+
"learning_rate": 0.0001969258589511754,
|
| 311 |
+
"loss": 0.8544,
|
| 312 |
+
"step": 480
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.22,
|
| 316 |
+
"learning_rate": 0.0001942133815551537,
|
| 317 |
+
"loss": 0.871,
|
| 318 |
+
"step": 490
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.24,
|
| 322 |
+
"learning_rate": 0.000191500904159132,
|
| 323 |
+
"loss": 0.8748,
|
| 324 |
+
"step": 500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.27,
|
| 328 |
+
"learning_rate": 0.00018933092224231464,
|
| 329 |
+
"loss": 0.8753,
|
| 330 |
+
"step": 510
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.29,
|
| 334 |
+
"learning_rate": 0.00018824593128390595,
|
| 335 |
+
"loss": 1.8069,
|
| 336 |
+
"step": 520
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.32,
|
| 340 |
+
"learning_rate": 0.00018634719710669076,
|
| 341 |
+
"loss": 4.3811,
|
| 342 |
+
"step": 530
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.34,
|
| 346 |
+
"learning_rate": 0.00018390596745027123,
|
| 347 |
+
"loss": 5.8063,
|
| 348 |
+
"step": 540
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.37,
|
| 352 |
+
"learning_rate": 0.00018119349005424955,
|
| 353 |
+
"loss": 5.2762,
|
| 354 |
+
"step": 550
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.39,
|
| 358 |
+
"learning_rate": 0.00017848101265822782,
|
| 359 |
+
"loss": 3.0862,
|
| 360 |
+
"step": 560
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.42,
|
| 364 |
+
"learning_rate": 0.00017576853526220611,
|
| 365 |
+
"loss": 1.7397,
|
| 366 |
+
"step": 570
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.44,
|
| 370 |
+
"learning_rate": 0.00017305605786618444,
|
| 371 |
+
"loss": 1.5362,
|
| 372 |
+
"step": 580
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.47,
|
| 376 |
+
"learning_rate": 0.0001706148282097649,
|
| 377 |
+
"loss": 3.644,
|
| 378 |
+
"step": 590
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.49,
|
| 382 |
+
"learning_rate": 0.0001679023508137432,
|
| 383 |
+
"loss": 6.1529,
|
| 384 |
+
"step": 600
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.49,
|
| 388 |
+
"eval_loss": 6.275171756744385,
|
| 389 |
+
"eval_runtime": 401.4674,
|
| 390 |
+
"eval_samples_per_second": 4.982,
|
| 391 |
+
"eval_steps_per_second": 0.623,
|
| 392 |
+
"step": 600
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.51,
|
| 396 |
+
"learning_rate": 0.00016573236889692584,
|
| 397 |
+
"loss": 4.6504,
|
| 398 |
+
"step": 610
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.54,
|
| 402 |
+
"learning_rate": 0.00016301989150090414,
|
| 403 |
+
"loss": 3.3039,
|
| 404 |
+
"step": 620
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.56,
|
| 408 |
+
"learning_rate": 0.00016030741410488246,
|
| 409 |
+
"loss": 2.7266,
|
| 410 |
+
"step": 630
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.59,
|
| 414 |
+
"learning_rate": 0.00015759493670886075,
|
| 415 |
+
"loss": 4.5256,
|
| 416 |
+
"step": 640
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.61,
|
| 420 |
+
"learning_rate": 0.0001551537070524412,
|
| 421 |
+
"loss": 3.9326,
|
| 422 |
+
"step": 650
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.64,
|
| 426 |
+
"learning_rate": 0.00015244122965641952,
|
| 427 |
+
"loss": 1.6396,
|
| 428 |
+
"step": 660
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.66,
|
| 432 |
+
"learning_rate": 0.00014972875226039781,
|
| 433 |
+
"loss": 1.1304,
|
| 434 |
+
"step": 670
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.69,
|
| 438 |
+
"learning_rate": 0.0001470162748643761,
|
| 439 |
+
"loss": 1.0282,
|
| 440 |
+
"step": 680
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.71,
|
| 444 |
+
"learning_rate": 0.00014430379746835443,
|
| 445 |
+
"loss": 0.9902,
|
| 446 |
+
"step": 690
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.74,
|
| 450 |
+
"learning_rate": 0.00014159132007233273,
|
| 451 |
+
"loss": 0.9785,
|
| 452 |
+
"step": 700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.76,
|
| 456 |
+
"learning_rate": 0.00013887884267631102,
|
| 457 |
+
"loss": 0.9422,
|
| 458 |
+
"step": 710
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.79,
|
| 462 |
+
"learning_rate": 0.00013616636528028932,
|
| 463 |
+
"loss": 0.9466,
|
| 464 |
+
"step": 720
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.81,
|
| 468 |
+
"learning_rate": 0.0001334538878842676,
|
| 469 |
+
"loss": 0.9254,
|
| 470 |
+
"step": 730
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.84,
|
| 474 |
+
"learning_rate": 0.0001307414104882459,
|
| 475 |
+
"loss": 0.9257,
|
| 476 |
+
"step": 740
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.86,
|
| 480 |
+
"learning_rate": 0.00012802893309222423,
|
| 481 |
+
"loss": 0.9179,
|
| 482 |
+
"step": 750
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.89,
|
| 486 |
+
"learning_rate": 0.00012531645569620252,
|
| 487 |
+
"loss": 0.9247,
|
| 488 |
+
"step": 760
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.91,
|
| 492 |
+
"learning_rate": 0.00012260397830018082,
|
| 493 |
+
"loss": 0.9186,
|
| 494 |
+
"step": 770
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.94,
|
| 498 |
+
"learning_rate": 0.00011989150090415911,
|
| 499 |
+
"loss": 0.9253,
|
| 500 |
+
"step": 780
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.96,
|
| 504 |
+
"learning_rate": 0.00011717902350813742,
|
| 505 |
+
"loss": 0.9193,
|
| 506 |
+
"step": 790
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.99,
|
| 510 |
+
"learning_rate": 0.00011446654611211573,
|
| 511 |
+
"loss": 0.9141,
|
| 512 |
+
"step": 800
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.99,
|
| 516 |
+
"eval_loss": 0.9085212349891663,
|
| 517 |
+
"eval_runtime": 400.6517,
|
| 518 |
+
"eval_samples_per_second": 4.992,
|
| 519 |
+
"eval_steps_per_second": 0.624,
|
| 520 |
+
"step": 800
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.01,
|
| 524 |
+
"learning_rate": 0.00011175406871609403,
|
| 525 |
+
"loss": 0.9057,
|
| 526 |
+
"step": 810
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.04,
|
| 530 |
+
"learning_rate": 0.00010904159132007232,
|
| 531 |
+
"loss": 0.9083,
|
| 532 |
+
"step": 820
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.06,
|
| 536 |
+
"learning_rate": 0.00010632911392405062,
|
| 537 |
+
"loss": 0.9091,
|
| 538 |
+
"step": 830
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 2.09,
|
| 542 |
+
"learning_rate": 0.00010361663652802893,
|
| 543 |
+
"loss": 0.9186,
|
| 544 |
+
"step": 840
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 2.11,
|
| 548 |
+
"learning_rate": 0.00010090415913200722,
|
| 549 |
+
"loss": 0.9018,
|
| 550 |
+
"step": 850
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.14,
|
| 554 |
+
"learning_rate": 9.819168173598552e-05,
|
| 555 |
+
"loss": 0.8989,
|
| 556 |
+
"step": 860
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 2.16,
|
| 560 |
+
"learning_rate": 9.547920433996383e-05,
|
| 561 |
+
"loss": 0.9275,
|
| 562 |
+
"step": 870
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.19,
|
| 566 |
+
"learning_rate": 9.276672694394213e-05,
|
| 567 |
+
"loss": 0.9063,
|
| 568 |
+
"step": 880
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.21,
|
| 572 |
+
"learning_rate": 9.005424954792042e-05,
|
| 573 |
+
"loss": 0.8921,
|
| 574 |
+
"step": 890
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 2.23,
|
| 578 |
+
"learning_rate": 8.734177215189872e-05,
|
| 579 |
+
"loss": 0.8981,
|
| 580 |
+
"step": 900
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.26,
|
| 584 |
+
"learning_rate": 8.462929475587703e-05,
|
| 585 |
+
"loss": 0.9099,
|
| 586 |
+
"step": 910
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.28,
|
| 590 |
+
"learning_rate": 8.191681735985534e-05,
|
| 591 |
+
"loss": 0.8956,
|
| 592 |
+
"step": 920
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.31,
|
| 596 |
+
"learning_rate": 7.920433996383362e-05,
|
| 597 |
+
"loss": 0.8983,
|
| 598 |
+
"step": 930
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.33,
|
| 602 |
+
"learning_rate": 7.649186256781193e-05,
|
| 603 |
+
"loss": 0.9192,
|
| 604 |
+
"step": 940
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.36,
|
| 608 |
+
"learning_rate": 7.377938517179023e-05,
|
| 609 |
+
"loss": 0.9199,
|
| 610 |
+
"step": 950
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 2.38,
|
| 614 |
+
"learning_rate": 7.106690777576854e-05,
|
| 615 |
+
"loss": 0.8857,
|
| 616 |
+
"step": 960
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.41,
|
| 620 |
+
"learning_rate": 6.835443037974683e-05,
|
| 621 |
+
"loss": 0.8871,
|
| 622 |
+
"step": 970
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.43,
|
| 626 |
+
"learning_rate": 6.564195298372513e-05,
|
| 627 |
+
"loss": 0.9054,
|
| 628 |
+
"step": 980
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.46,
|
| 632 |
+
"learning_rate": 6.292947558770344e-05,
|
| 633 |
+
"loss": 0.9167,
|
| 634 |
+
"step": 990
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.48,
|
| 638 |
+
"learning_rate": 6.021699819168174e-05,
|
| 639 |
+
"loss": 0.9023,
|
| 640 |
+
"step": 1000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.48,
|
| 644 |
+
"eval_loss": 0.9026673436164856,
|
| 645 |
+
"eval_runtime": 401.9867,
|
| 646 |
+
"eval_samples_per_second": 4.975,
|
| 647 |
+
"eval_steps_per_second": 0.622,
|
| 648 |
+
"step": 1000
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.51,
|
| 652 |
+
"learning_rate": 5.750452079566003e-05,
|
| 653 |
+
"loss": 0.9043,
|
| 654 |
+
"step": 1010
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.53,
|
| 658 |
+
"learning_rate": 5.4792043399638334e-05,
|
| 659 |
+
"loss": 0.9052,
|
| 660 |
+
"step": 1020
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.56,
|
| 664 |
+
"learning_rate": 5.207956600361663e-05,
|
| 665 |
+
"loss": 0.8941,
|
| 666 |
+
"step": 1030
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 2.58,
|
| 670 |
+
"learning_rate": 4.936708860759493e-05,
|
| 671 |
+
"loss": 0.9081,
|
| 672 |
+
"step": 1040
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.61,
|
| 676 |
+
"learning_rate": 4.6654611211573234e-05,
|
| 677 |
+
"loss": 0.9079,
|
| 678 |
+
"step": 1050
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.63,
|
| 682 |
+
"learning_rate": 4.394213381555153e-05,
|
| 683 |
+
"loss": 0.8925,
|
| 684 |
+
"step": 1060
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.66,
|
| 688 |
+
"learning_rate": 4.122965641952984e-05,
|
| 689 |
+
"loss": 0.9041,
|
| 690 |
+
"step": 1070
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.68,
|
| 694 |
+
"learning_rate": 3.851717902350813e-05,
|
| 695 |
+
"loss": 0.902,
|
| 696 |
+
"step": 1080
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.71,
|
| 700 |
+
"learning_rate": 3.5804701627486435e-05,
|
| 701 |
+
"loss": 0.9011,
|
| 702 |
+
"step": 1090
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.73,
|
| 706 |
+
"learning_rate": 3.3092224231464737e-05,
|
| 707 |
+
"loss": 0.8883,
|
| 708 |
+
"step": 1100
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 2.76,
|
| 712 |
+
"learning_rate": 3.0379746835443035e-05,
|
| 713 |
+
"loss": 0.9039,
|
| 714 |
+
"step": 1110
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.78,
|
| 718 |
+
"learning_rate": 2.7667269439421337e-05,
|
| 719 |
+
"loss": 0.9078,
|
| 720 |
+
"step": 1120
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.81,
|
| 724 |
+
"learning_rate": 2.4954792043399636e-05,
|
| 725 |
+
"loss": 0.9065,
|
| 726 |
+
"step": 1130
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 2.83,
|
| 730 |
+
"learning_rate": 2.2242314647377938e-05,
|
| 731 |
+
"loss": 0.9084,
|
| 732 |
+
"step": 1140
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.86,
|
| 736 |
+
"learning_rate": 1.952983725135624e-05,
|
| 737 |
+
"loss": 0.8979,
|
| 738 |
+
"step": 1150
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.88,
|
| 742 |
+
"learning_rate": 1.6817359855334538e-05,
|
| 743 |
+
"loss": 0.9087,
|
| 744 |
+
"step": 1160
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.91,
|
| 748 |
+
"learning_rate": 1.4104882459312838e-05,
|
| 749 |
+
"loss": 0.9059,
|
| 750 |
+
"step": 1170
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 2.93,
|
| 754 |
+
"learning_rate": 1.1392405063291139e-05,
|
| 755 |
+
"loss": 0.8969,
|
| 756 |
+
"step": 1180
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 2.95,
|
| 760 |
+
"learning_rate": 8.679927667269439e-06,
|
| 761 |
+
"loss": 0.9162,
|
| 762 |
+
"step": 1190
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.98,
|
| 766 |
+
"learning_rate": 5.967450271247739e-06,
|
| 767 |
+
"loss": 0.9065,
|
| 768 |
+
"step": 1200
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 2.98,
|
| 772 |
+
"eval_loss": 0.9031401872634888,
|
| 773 |
+
"eval_runtime": 405.8678,
|
| 774 |
+
"eval_samples_per_second": 4.928,
|
| 775 |
+
"eval_steps_per_second": 0.616,
|
| 776 |
+
"step": 1200
|
| 777 |
+
}
|
| 778 |
+
],
|
| 779 |
+
"max_steps": 1206,
|
| 780 |
+
"num_train_epochs": 3,
|
| 781 |
+
"total_flos": 1.5073290925375488e+18,
|
| 782 |
+
"trial_name": null,
|
| 783 |
+
"trial_params": null
|
| 784 |
+
}
|
checkpoint-1200/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0abdd9155585c4e11f0ad9dcfdd6c80dde474221e48df2dc7b337502cbe5dda9
|
| 3 |
+
size 3963
|
checkpoint-400/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: True
|
| 9 |
+
- load_in_4bit: False
|
| 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: fp4
|
| 15 |
+
- bnb_4bit_use_double_quant: False
|
| 16 |
+
- bnb_4bit_compute_dtype: float32
|
| 17 |
+
### Framework versions
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
- PEFT 0.5.0.dev0
|
checkpoint-400/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "decapoda-research/llama-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 8,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-400/adapter_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5e1621f48d9ad8feb1d6d31050275f0aafd080c5c07153301fe2f48411f4406
|
| 3 |
+
size 443
|
checkpoint-400/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b23dece956a49d71a2cb92f6c91d5e5f369d3fc427abeb70ec26fd9d758e7e07
|
| 3 |
+
size 33661637
|
checkpoint-400/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70a9f504e8e9850cadec6064dad9a76f13708f736481e5c21fdf03e9b781524f
|
| 3 |
+
size 14575
|
checkpoint-400/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe26c94c4fc27c447392293953497b05e2b80698f66c299d2795aad9cf68a3ef
|
| 3 |
+
size 627
|
checkpoint-400/trainer_state.json
ADDED
|
@@ -0,0 +1,272 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8681809306144714,
|
| 3 |
+
"best_model_checkpoint": "./lora-warewe/checkpoint-400",
|
| 4 |
+
"epoch": 0.9932490106308683,
|
| 5 |
+
"global_step": 400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 13 |
+
"loss": 1.9531,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 19 |
+
"loss": 1.8714,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 8.999999999999999e-05,
|
| 25 |
+
"loss": 1.7052,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 0.00011999999999999999,
|
| 31 |
+
"loss": 1.3201,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.12,
|
| 36 |
+
"learning_rate": 0.00015,
|
| 37 |
+
"loss": 1.1551,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 0.00017999999999999998,
|
| 43 |
+
"loss": 1.1199,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
+
"learning_rate": 0.00020999999999999998,
|
| 49 |
+
"loss": 1.0598,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 0.00023999999999999998,
|
| 55 |
+
"loss": 1.0233,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.22,
|
| 60 |
+
"learning_rate": 0.00027,
|
| 61 |
+
"loss": 1.0086,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.25,
|
| 66 |
+
"learning_rate": 0.0003,
|
| 67 |
+
"loss": 0.999,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.27,
|
| 72 |
+
"learning_rate": 0.0002972875226039783,
|
| 73 |
+
"loss": 0.9961,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.3,
|
| 78 |
+
"learning_rate": 0.00029457504520795656,
|
| 79 |
+
"loss": 0.9831,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.32,
|
| 84 |
+
"learning_rate": 0.0002918625678119349,
|
| 85 |
+
"loss": 0.9637,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.35,
|
| 90 |
+
"learning_rate": 0.00028915009041591315,
|
| 91 |
+
"loss": 0.9546,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.37,
|
| 96 |
+
"learning_rate": 0.0002864376130198915,
|
| 97 |
+
"loss": 0.9575,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.4,
|
| 102 |
+
"learning_rate": 0.00028372513562386974,
|
| 103 |
+
"loss": 0.9337,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.42,
|
| 108 |
+
"learning_rate": 0.00028101265822784807,
|
| 109 |
+
"loss": 0.942,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.45,
|
| 114 |
+
"learning_rate": 0.0002783001808318264,
|
| 115 |
+
"loss": 0.9265,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.47,
|
| 120 |
+
"learning_rate": 0.0002755877034358047,
|
| 121 |
+
"loss": 0.9324,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.5,
|
| 126 |
+
"learning_rate": 0.000272875226039783,
|
| 127 |
+
"loss": 0.9288,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5,
|
| 132 |
+
"eval_loss": 0.9210162162780762,
|
| 133 |
+
"eval_runtime": 559.4854,
|
| 134 |
+
"eval_samples_per_second": 3.575,
|
| 135 |
+
"eval_steps_per_second": 0.447,
|
| 136 |
+
"step": 200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.52,
|
| 140 |
+
"learning_rate": 0.00027016274864376125,
|
| 141 |
+
"loss": 0.9201,
|
| 142 |
+
"step": 210
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.55,
|
| 146 |
+
"learning_rate": 0.00026745027124773957,
|
| 147 |
+
"loss": 0.9242,
|
| 148 |
+
"step": 220
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.57,
|
| 152 |
+
"learning_rate": 0.0002647377938517179,
|
| 153 |
+
"loss": 0.9114,
|
| 154 |
+
"step": 230
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.6,
|
| 158 |
+
"learning_rate": 0.00026202531645569616,
|
| 159 |
+
"loss": 0.9223,
|
| 160 |
+
"step": 240
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.62,
|
| 164 |
+
"learning_rate": 0.0002593128390596745,
|
| 165 |
+
"loss": 0.9018,
|
| 166 |
+
"step": 250
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.65,
|
| 170 |
+
"learning_rate": 0.0002566003616636528,
|
| 171 |
+
"loss": 0.9038,
|
| 172 |
+
"step": 260
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.67,
|
| 176 |
+
"learning_rate": 0.0002538878842676311,
|
| 177 |
+
"loss": 0.9087,
|
| 178 |
+
"step": 270
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.7,
|
| 182 |
+
"learning_rate": 0.0002511754068716094,
|
| 183 |
+
"loss": 0.9141,
|
| 184 |
+
"step": 280
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.72,
|
| 188 |
+
"learning_rate": 0.00024846292947558766,
|
| 189 |
+
"loss": 0.8942,
|
| 190 |
+
"step": 290
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.74,
|
| 194 |
+
"learning_rate": 0.000245750452079566,
|
| 195 |
+
"loss": 0.8865,
|
| 196 |
+
"step": 300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.77,
|
| 200 |
+
"learning_rate": 0.00024303797468354428,
|
| 201 |
+
"loss": 0.8863,
|
| 202 |
+
"step": 310
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.79,
|
| 206 |
+
"learning_rate": 0.00024032549728752258,
|
| 207 |
+
"loss": 0.8981,
|
| 208 |
+
"step": 320
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.82,
|
| 212 |
+
"learning_rate": 0.0002376130198915009,
|
| 213 |
+
"loss": 0.8837,
|
| 214 |
+
"step": 330
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.84,
|
| 218 |
+
"learning_rate": 0.0002349005424954792,
|
| 219 |
+
"loss": 0.8887,
|
| 220 |
+
"step": 340
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.87,
|
| 224 |
+
"learning_rate": 0.0002321880650994575,
|
| 225 |
+
"loss": 0.8781,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.89,
|
| 230 |
+
"learning_rate": 0.00022947558770343578,
|
| 231 |
+
"loss": 0.891,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.92,
|
| 236 |
+
"learning_rate": 0.00022676311030741408,
|
| 237 |
+
"loss": 0.8852,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.94,
|
| 242 |
+
"learning_rate": 0.00022405063291139237,
|
| 243 |
+
"loss": 0.8736,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 0.97,
|
| 248 |
+
"learning_rate": 0.0002213381555153707,
|
| 249 |
+
"loss": 0.8799,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.99,
|
| 254 |
+
"learning_rate": 0.000218625678119349,
|
| 255 |
+
"loss": 0.8847,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.99,
|
| 260 |
+
"eval_loss": 0.8681809306144714,
|
| 261 |
+
"eval_runtime": 405.865,
|
| 262 |
+
"eval_samples_per_second": 4.928,
|
| 263 |
+
"eval_steps_per_second": 0.616,
|
| 264 |
+
"step": 400
|
| 265 |
+
}
|
| 266 |
+
],
|
| 267 |
+
"max_steps": 1206,
|
| 268 |
+
"num_train_epochs": 3,
|
| 269 |
+
"total_flos": 5.0266814048894976e+17,
|
| 270 |
+
"trial_name": null,
|
| 271 |
+
"trial_params": null
|
| 272 |
+
}
|
checkpoint-400/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0abdd9155585c4e11f0ad9dcfdd6c80dde474221e48df2dc7b337502cbe5dda9
|
| 3 |
+
size 3963
|