Yelban
commited on
Commit
•
9f9b025
1
Parent(s):
6e22181
add LoRA
Browse files- 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-600/README.md +20 -0
- checkpoint-600/adapter_config.json +21 -0
- checkpoint-600/adapter_model.bin +3 -0
- checkpoint-600/optimizer.pt +3 -0
- checkpoint-600/rng_state.pth +3 -0
- checkpoint-600/scheduler.pt +3 -0
- checkpoint-600/trainer_state.json +400 -0
- checkpoint-600/training_args.bin +3 -0
- checkpoint-800/README.md +20 -0
- checkpoint-800/adapter_config.json +21 -0
- checkpoint-800/adapter_model.bin +3 -0
- checkpoint-800/optimizer.pt +3 -0
- checkpoint-800/rng_state.pth +3 -0
- checkpoint-800/scheduler.pt +3 -0
- checkpoint-800/trainer_state.json +528 -0
- checkpoint-800/training_args.bin +3 -0
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": "meta-llama/Llama-2-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 16,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"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 sha256:17ccdb3f813d3112cc5a319e23f283e62a69e15cbdcd67f105af1f6a269c356a
|
| 3 |
+
size 67216581
|
checkpoint-1000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31092f61007b84621ac1eabf82487538d0534f4fc882db88a8020f37d034ebaa
|
| 3 |
+
size 14575
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c44d6cc36fc995015bd384d8be7c41e3a3accab39d9b5ba013de1980754d947
|
| 3 |
+
size 627
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,656 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7937892079353333,
|
| 3 |
+
"best_model_checkpoint": "./lora-alpaca/checkpoint-1000",
|
| 4 |
+
"epoch": 2.556727388942154,
|
| 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": 0.03,
|
| 12 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 13 |
+
"loss": 1.6613,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 19 |
+
"loss": 1.6025,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.08,
|
| 24 |
+
"learning_rate": 8.999999999999999e-05,
|
| 25 |
+
"loss": 1.4512,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 0.00011999999999999999,
|
| 31 |
+
"loss": 1.1653,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.13,
|
| 36 |
+
"learning_rate": 0.00015,
|
| 37 |
+
"loss": 0.9588,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 0.00017999999999999998,
|
| 43 |
+
"loss": 0.9058,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.18,
|
| 48 |
+
"learning_rate": 0.00020999999999999998,
|
| 49 |
+
"loss": 0.8655,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 0.00023999999999999998,
|
| 55 |
+
"loss": 0.8493,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.23,
|
| 60 |
+
"learning_rate": 0.00027,
|
| 61 |
+
"loss": 0.8462,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.26,
|
| 66 |
+
"learning_rate": 0.0003,
|
| 67 |
+
"loss": 0.8353,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.28,
|
| 72 |
+
"learning_rate": 0.0002972041006523765,
|
| 73 |
+
"loss": 0.833,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.31,
|
| 78 |
+
"learning_rate": 0.000294408201304753,
|
| 79 |
+
"loss": 0.8322,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.33,
|
| 84 |
+
"learning_rate": 0.0002916123019571295,
|
| 85 |
+
"loss": 0.8239,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.36,
|
| 90 |
+
"learning_rate": 0.00028881640260950607,
|
| 91 |
+
"loss": 0.8174,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.38,
|
| 96 |
+
"learning_rate": 0.00028602050326188253,
|
| 97 |
+
"loss": 0.8178,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.41,
|
| 102 |
+
"learning_rate": 0.00028322460391425904,
|
| 103 |
+
"loss": 0.8321,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.43,
|
| 108 |
+
"learning_rate": 0.0002804287045666356,
|
| 109 |
+
"loss": 0.8213,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.46,
|
| 114 |
+
"learning_rate": 0.0002776328052190121,
|
| 115 |
+
"loss": 0.8088,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.49,
|
| 120 |
+
"learning_rate": 0.00027483690587138857,
|
| 121 |
+
"loss": 0.8021,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.51,
|
| 126 |
+
"learning_rate": 0.00027204100652376514,
|
| 127 |
+
"loss": 0.8297,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.51,
|
| 132 |
+
"eval_loss": 0.8141016960144043,
|
| 133 |
+
"eval_runtime": 87.4274,
|
| 134 |
+
"eval_samples_per_second": 22.876,
|
| 135 |
+
"eval_steps_per_second": 2.86,
|
| 136 |
+
"step": 200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.54,
|
| 140 |
+
"learning_rate": 0.00026924510717614165,
|
| 141 |
+
"loss": 0.8148,
|
| 142 |
+
"step": 210
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.56,
|
| 146 |
+
"learning_rate": 0.00026644920782851816,
|
| 147 |
+
"loss": 0.8204,
|
| 148 |
+
"step": 220
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.59,
|
| 152 |
+
"learning_rate": 0.00026365330848089467,
|
| 153 |
+
"loss": 0.8181,
|
| 154 |
+
"step": 230
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.61,
|
| 158 |
+
"learning_rate": 0.0002608574091332712,
|
| 159 |
+
"loss": 0.8146,
|
| 160 |
+
"step": 240
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.64,
|
| 164 |
+
"learning_rate": 0.0002580615097856477,
|
| 165 |
+
"loss": 0.8108,
|
| 166 |
+
"step": 250
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.66,
|
| 170 |
+
"learning_rate": 0.0002552656104380242,
|
| 171 |
+
"loss": 0.8069,
|
| 172 |
+
"step": 260
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.69,
|
| 176 |
+
"learning_rate": 0.0002524697110904007,
|
| 177 |
+
"loss": 0.8219,
|
| 178 |
+
"step": 270
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.72,
|
| 182 |
+
"learning_rate": 0.0002496738117427772,
|
| 183 |
+
"loss": 0.8076,
|
| 184 |
+
"step": 280
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.74,
|
| 188 |
+
"learning_rate": 0.00024687791239515373,
|
| 189 |
+
"loss": 0.8076,
|
| 190 |
+
"step": 290
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.77,
|
| 194 |
+
"learning_rate": 0.00024408201304753027,
|
| 195 |
+
"loss": 0.8173,
|
| 196 |
+
"step": 300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.79,
|
| 200 |
+
"learning_rate": 0.00024128611369990678,
|
| 201 |
+
"loss": 0.8032,
|
| 202 |
+
"step": 310
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.82,
|
| 206 |
+
"learning_rate": 0.0002384902143522833,
|
| 207 |
+
"loss": 0.8185,
|
| 208 |
+
"step": 320
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.84,
|
| 212 |
+
"learning_rate": 0.0002356943150046598,
|
| 213 |
+
"loss": 0.8046,
|
| 214 |
+
"step": 330
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.87,
|
| 218 |
+
"learning_rate": 0.00023289841565703632,
|
| 219 |
+
"loss": 0.812,
|
| 220 |
+
"step": 340
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.89,
|
| 224 |
+
"learning_rate": 0.00023010251630941285,
|
| 225 |
+
"loss": 0.807,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.92,
|
| 230 |
+
"learning_rate": 0.00022730661696178936,
|
| 231 |
+
"loss": 0.8003,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.95,
|
| 236 |
+
"learning_rate": 0.00022451071761416585,
|
| 237 |
+
"loss": 0.802,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.97,
|
| 242 |
+
"learning_rate": 0.00022171481826654239,
|
| 243 |
+
"loss": 0.8073,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.0,
|
| 248 |
+
"learning_rate": 0.0002189189189189189,
|
| 249 |
+
"loss": 0.8055,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.02,
|
| 254 |
+
"learning_rate": 0.00021612301957129543,
|
| 255 |
+
"loss": 0.8132,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.02,
|
| 260 |
+
"eval_loss": 0.8033702373504639,
|
| 261 |
+
"eval_runtime": 87.6065,
|
| 262 |
+
"eval_samples_per_second": 22.829,
|
| 263 |
+
"eval_steps_per_second": 2.854,
|
| 264 |
+
"step": 400
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.05,
|
| 268 |
+
"learning_rate": 0.00021332712022367195,
|
| 269 |
+
"loss": 0.7992,
|
| 270 |
+
"step": 410
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.07,
|
| 274 |
+
"learning_rate": 0.00021053122087604843,
|
| 275 |
+
"loss": 0.7945,
|
| 276 |
+
"step": 420
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.1,
|
| 280 |
+
"learning_rate": 0.00020773532152842497,
|
| 281 |
+
"loss": 0.7917,
|
| 282 |
+
"step": 430
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.12,
|
| 286 |
+
"learning_rate": 0.00020493942218080148,
|
| 287 |
+
"loss": 0.8002,
|
| 288 |
+
"step": 440
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.15,
|
| 292 |
+
"learning_rate": 0.00020214352283317796,
|
| 293 |
+
"loss": 0.7934,
|
| 294 |
+
"step": 450
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.18,
|
| 298 |
+
"learning_rate": 0.0001993476234855545,
|
| 299 |
+
"loss": 0.7982,
|
| 300 |
+
"step": 460
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.2,
|
| 304 |
+
"learning_rate": 0.000196551724137931,
|
| 305 |
+
"loss": 0.7995,
|
| 306 |
+
"step": 470
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.23,
|
| 310 |
+
"learning_rate": 0.00019375582479030755,
|
| 311 |
+
"loss": 0.7935,
|
| 312 |
+
"step": 480
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.25,
|
| 316 |
+
"learning_rate": 0.00019095992544268406,
|
| 317 |
+
"loss": 0.8003,
|
| 318 |
+
"step": 490
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.28,
|
| 322 |
+
"learning_rate": 0.00018816402609506054,
|
| 323 |
+
"loss": 0.7967,
|
| 324 |
+
"step": 500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.3,
|
| 328 |
+
"learning_rate": 0.00018536812674743708,
|
| 329 |
+
"loss": 0.7962,
|
| 330 |
+
"step": 510
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.33,
|
| 334 |
+
"learning_rate": 0.0001825722273998136,
|
| 335 |
+
"loss": 0.7956,
|
| 336 |
+
"step": 520
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.36,
|
| 340 |
+
"learning_rate": 0.00017977632805219013,
|
| 341 |
+
"loss": 0.7958,
|
| 342 |
+
"step": 530
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.38,
|
| 346 |
+
"learning_rate": 0.00017698042870456661,
|
| 347 |
+
"loss": 0.7944,
|
| 348 |
+
"step": 540
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.41,
|
| 352 |
+
"learning_rate": 0.00017418452935694312,
|
| 353 |
+
"loss": 0.7945,
|
| 354 |
+
"step": 550
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.43,
|
| 358 |
+
"learning_rate": 0.00017138863000931966,
|
| 359 |
+
"loss": 0.802,
|
| 360 |
+
"step": 560
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.46,
|
| 364 |
+
"learning_rate": 0.00016859273066169617,
|
| 365 |
+
"loss": 0.7923,
|
| 366 |
+
"step": 570
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.48,
|
| 370 |
+
"learning_rate": 0.00016579683131407266,
|
| 371 |
+
"loss": 0.8024,
|
| 372 |
+
"step": 580
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.51,
|
| 376 |
+
"learning_rate": 0.0001630009319664492,
|
| 377 |
+
"loss": 0.7902,
|
| 378 |
+
"step": 590
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.53,
|
| 382 |
+
"learning_rate": 0.0001602050326188257,
|
| 383 |
+
"loss": 0.8066,
|
| 384 |
+
"step": 600
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.53,
|
| 388 |
+
"eval_loss": 0.7989646196365356,
|
| 389 |
+
"eval_runtime": 87.5148,
|
| 390 |
+
"eval_samples_per_second": 22.853,
|
| 391 |
+
"eval_steps_per_second": 2.857,
|
| 392 |
+
"step": 600
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.56,
|
| 396 |
+
"learning_rate": 0.00015740913327120224,
|
| 397 |
+
"loss": 0.7958,
|
| 398 |
+
"step": 610
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.59,
|
| 402 |
+
"learning_rate": 0.00015461323392357873,
|
| 403 |
+
"loss": 0.7837,
|
| 404 |
+
"step": 620
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.61,
|
| 408 |
+
"learning_rate": 0.00015181733457595524,
|
| 409 |
+
"loss": 0.8008,
|
| 410 |
+
"step": 630
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.64,
|
| 414 |
+
"learning_rate": 0.00014902143522833175,
|
| 415 |
+
"loss": 0.7964,
|
| 416 |
+
"step": 640
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.66,
|
| 420 |
+
"learning_rate": 0.0001462255358807083,
|
| 421 |
+
"loss": 0.7951,
|
| 422 |
+
"step": 650
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.69,
|
| 426 |
+
"learning_rate": 0.0001434296365330848,
|
| 427 |
+
"loss": 0.8012,
|
| 428 |
+
"step": 660
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.71,
|
| 432 |
+
"learning_rate": 0.0001406337371854613,
|
| 433 |
+
"loss": 0.7928,
|
| 434 |
+
"step": 670
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.74,
|
| 438 |
+
"learning_rate": 0.00013783783783783782,
|
| 439 |
+
"loss": 0.7903,
|
| 440 |
+
"step": 680
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.76,
|
| 444 |
+
"learning_rate": 0.00013504193849021433,
|
| 445 |
+
"loss": 0.7949,
|
| 446 |
+
"step": 690
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.79,
|
| 450 |
+
"learning_rate": 0.00013224603914259084,
|
| 451 |
+
"loss": 0.7866,
|
| 452 |
+
"step": 700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.82,
|
| 456 |
+
"learning_rate": 0.00012945013979496738,
|
| 457 |
+
"loss": 0.7889,
|
| 458 |
+
"step": 710
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.84,
|
| 462 |
+
"learning_rate": 0.0001266542404473439,
|
| 463 |
+
"loss": 0.7928,
|
| 464 |
+
"step": 720
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.87,
|
| 468 |
+
"learning_rate": 0.0001238583410997204,
|
| 469 |
+
"loss": 0.7932,
|
| 470 |
+
"step": 730
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.89,
|
| 474 |
+
"learning_rate": 0.00012106244175209691,
|
| 475 |
+
"loss": 0.7963,
|
| 476 |
+
"step": 740
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.92,
|
| 480 |
+
"learning_rate": 0.00011826654240447344,
|
| 481 |
+
"loss": 0.7979,
|
| 482 |
+
"step": 750
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.94,
|
| 486 |
+
"learning_rate": 0.00011547064305684995,
|
| 487 |
+
"loss": 0.7943,
|
| 488 |
+
"step": 760
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.97,
|
| 492 |
+
"learning_rate": 0.00011267474370922644,
|
| 493 |
+
"loss": 0.7983,
|
| 494 |
+
"step": 770
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.99,
|
| 498 |
+
"learning_rate": 0.00010987884436160297,
|
| 499 |
+
"loss": 0.801,
|
| 500 |
+
"step": 780
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 2.02,
|
| 504 |
+
"learning_rate": 0.0001070829450139795,
|
| 505 |
+
"loss": 0.7836,
|
| 506 |
+
"step": 790
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.05,
|
| 510 |
+
"learning_rate": 0.000104287045666356,
|
| 511 |
+
"loss": 0.785,
|
| 512 |
+
"step": 800
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.05,
|
| 516 |
+
"eval_loss": 0.7959330677986145,
|
| 517 |
+
"eval_runtime": 88.4745,
|
| 518 |
+
"eval_samples_per_second": 22.605,
|
| 519 |
+
"eval_steps_per_second": 2.826,
|
| 520 |
+
"step": 800
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 2.07,
|
| 524 |
+
"learning_rate": 0.00010149114631873253,
|
| 525 |
+
"loss": 0.8007,
|
| 526 |
+
"step": 810
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 2.1,
|
| 530 |
+
"learning_rate": 9.869524697110903e-05,
|
| 531 |
+
"loss": 0.7778,
|
| 532 |
+
"step": 820
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.12,
|
| 536 |
+
"learning_rate": 9.589934762348555e-05,
|
| 537 |
+
"loss": 0.7916,
|
| 538 |
+
"step": 830
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 2.15,
|
| 542 |
+
"learning_rate": 9.310344827586206e-05,
|
| 543 |
+
"loss": 0.7921,
|
| 544 |
+
"step": 840
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 2.17,
|
| 548 |
+
"learning_rate": 9.030754892823859e-05,
|
| 549 |
+
"loss": 0.7825,
|
| 550 |
+
"step": 850
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.2,
|
| 554 |
+
"learning_rate": 8.751164958061508e-05,
|
| 555 |
+
"loss": 0.7797,
|
| 556 |
+
"step": 860
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 2.22,
|
| 560 |
+
"learning_rate": 8.471575023299161e-05,
|
| 561 |
+
"loss": 0.7904,
|
| 562 |
+
"step": 870
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.25,
|
| 566 |
+
"learning_rate": 8.191985088536812e-05,
|
| 567 |
+
"loss": 0.7848,
|
| 568 |
+
"step": 880
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.28,
|
| 572 |
+
"learning_rate": 7.912395153774464e-05,
|
| 573 |
+
"loss": 0.777,
|
| 574 |
+
"step": 890
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 2.3,
|
| 578 |
+
"learning_rate": 7.632805219012114e-05,
|
| 579 |
+
"loss": 0.7881,
|
| 580 |
+
"step": 900
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 2.33,
|
| 584 |
+
"learning_rate": 7.353215284249766e-05,
|
| 585 |
+
"loss": 0.7841,
|
| 586 |
+
"step": 910
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 2.35,
|
| 590 |
+
"learning_rate": 7.073625349487418e-05,
|
| 591 |
+
"loss": 0.7732,
|
| 592 |
+
"step": 920
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.38,
|
| 596 |
+
"learning_rate": 6.794035414725069e-05,
|
| 597 |
+
"loss": 0.7862,
|
| 598 |
+
"step": 930
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 2.4,
|
| 602 |
+
"learning_rate": 6.514445479962721e-05,
|
| 603 |
+
"loss": 0.784,
|
| 604 |
+
"step": 940
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.43,
|
| 608 |
+
"learning_rate": 6.234855545200372e-05,
|
| 609 |
+
"loss": 0.7873,
|
| 610 |
+
"step": 950
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 2.45,
|
| 614 |
+
"learning_rate": 5.955265610438024e-05,
|
| 615 |
+
"loss": 0.7966,
|
| 616 |
+
"step": 960
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 2.48,
|
| 620 |
+
"learning_rate": 5.6756756756756757e-05,
|
| 621 |
+
"loss": 0.7861,
|
| 622 |
+
"step": 970
|
| 623 |
+
},
|
| 624 |
+
{
|
| 625 |
+
"epoch": 2.51,
|
| 626 |
+
"learning_rate": 5.396085740913327e-05,
|
| 627 |
+
"loss": 0.789,
|
| 628 |
+
"step": 980
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.53,
|
| 632 |
+
"learning_rate": 5.1164958061509785e-05,
|
| 633 |
+
"loss": 0.786,
|
| 634 |
+
"step": 990
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.56,
|
| 638 |
+
"learning_rate": 4.8369058713886296e-05,
|
| 639 |
+
"loss": 0.777,
|
| 640 |
+
"step": 1000
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 2.56,
|
| 644 |
+
"eval_loss": 0.7937892079353333,
|
| 645 |
+
"eval_runtime": 87.4958,
|
| 646 |
+
"eval_samples_per_second": 22.858,
|
| 647 |
+
"eval_steps_per_second": 2.857,
|
| 648 |
+
"step": 1000
|
| 649 |
+
}
|
| 650 |
+
],
|
| 651 |
+
"max_steps": 1173,
|
| 652 |
+
"num_train_epochs": 3,
|
| 653 |
+
"total_flos": 1.3003597459862323e+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 sha256:7dc5fdae5752fa38ccdc7bdda6792bd163b3cd2401b9536f9886aa0aba7720a1
|
| 3 |
+
size 3899
|
checkpoint-600/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-600/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 16,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-600/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-600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:54524daca5cccec2fc45f8e7055cd80aea8838bbf5b55e94174529738e3fd173
|
| 3 |
+
size 67216581
|
checkpoint-600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:09c14dcb8bcbc2587537c2db7fb30839af26c8587367ddbb6475706e9be89059
|
| 3 |
+
size 14575
|
checkpoint-600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba4e43dd1fc9d82ff8fa51149544c9eea7f5fe801185eeda44c6be5431b8bc37
|
| 3 |
+
size 627
|
checkpoint-600/trainer_state.json
ADDED
|
@@ -0,0 +1,400 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7989646196365356,
|
| 3 |
+
"best_model_checkpoint": "./lora-alpaca/checkpoint-600",
|
| 4 |
+
"epoch": 1.5340364333652925,
|
| 5 |
+
"global_step": 600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 13 |
+
"loss": 1.6613,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 19 |
+
"loss": 1.6025,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.08,
|
| 24 |
+
"learning_rate": 8.999999999999999e-05,
|
| 25 |
+
"loss": 1.4512,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 0.00011999999999999999,
|
| 31 |
+
"loss": 1.1653,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.13,
|
| 36 |
+
"learning_rate": 0.00015,
|
| 37 |
+
"loss": 0.9588,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 0.00017999999999999998,
|
| 43 |
+
"loss": 0.9058,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.18,
|
| 48 |
+
"learning_rate": 0.00020999999999999998,
|
| 49 |
+
"loss": 0.8655,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 0.00023999999999999998,
|
| 55 |
+
"loss": 0.8493,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.23,
|
| 60 |
+
"learning_rate": 0.00027,
|
| 61 |
+
"loss": 0.8462,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.26,
|
| 66 |
+
"learning_rate": 0.0003,
|
| 67 |
+
"loss": 0.8353,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.28,
|
| 72 |
+
"learning_rate": 0.0002972041006523765,
|
| 73 |
+
"loss": 0.833,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.31,
|
| 78 |
+
"learning_rate": 0.000294408201304753,
|
| 79 |
+
"loss": 0.8322,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.33,
|
| 84 |
+
"learning_rate": 0.0002916123019571295,
|
| 85 |
+
"loss": 0.8239,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.36,
|
| 90 |
+
"learning_rate": 0.00028881640260950607,
|
| 91 |
+
"loss": 0.8174,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.38,
|
| 96 |
+
"learning_rate": 0.00028602050326188253,
|
| 97 |
+
"loss": 0.8178,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.41,
|
| 102 |
+
"learning_rate": 0.00028322460391425904,
|
| 103 |
+
"loss": 0.8321,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.43,
|
| 108 |
+
"learning_rate": 0.0002804287045666356,
|
| 109 |
+
"loss": 0.8213,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.46,
|
| 114 |
+
"learning_rate": 0.0002776328052190121,
|
| 115 |
+
"loss": 0.8088,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.49,
|
| 120 |
+
"learning_rate": 0.00027483690587138857,
|
| 121 |
+
"loss": 0.8021,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.51,
|
| 126 |
+
"learning_rate": 0.00027204100652376514,
|
| 127 |
+
"loss": 0.8297,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.51,
|
| 132 |
+
"eval_loss": 0.8141016960144043,
|
| 133 |
+
"eval_runtime": 87.4274,
|
| 134 |
+
"eval_samples_per_second": 22.876,
|
| 135 |
+
"eval_steps_per_second": 2.86,
|
| 136 |
+
"step": 200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.54,
|
| 140 |
+
"learning_rate": 0.00026924510717614165,
|
| 141 |
+
"loss": 0.8148,
|
| 142 |
+
"step": 210
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.56,
|
| 146 |
+
"learning_rate": 0.00026644920782851816,
|
| 147 |
+
"loss": 0.8204,
|
| 148 |
+
"step": 220
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.59,
|
| 152 |
+
"learning_rate": 0.00026365330848089467,
|
| 153 |
+
"loss": 0.8181,
|
| 154 |
+
"step": 230
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.61,
|
| 158 |
+
"learning_rate": 0.0002608574091332712,
|
| 159 |
+
"loss": 0.8146,
|
| 160 |
+
"step": 240
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.64,
|
| 164 |
+
"learning_rate": 0.0002580615097856477,
|
| 165 |
+
"loss": 0.8108,
|
| 166 |
+
"step": 250
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.66,
|
| 170 |
+
"learning_rate": 0.0002552656104380242,
|
| 171 |
+
"loss": 0.8069,
|
| 172 |
+
"step": 260
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.69,
|
| 176 |
+
"learning_rate": 0.0002524697110904007,
|
| 177 |
+
"loss": 0.8219,
|
| 178 |
+
"step": 270
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.72,
|
| 182 |
+
"learning_rate": 0.0002496738117427772,
|
| 183 |
+
"loss": 0.8076,
|
| 184 |
+
"step": 280
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.74,
|
| 188 |
+
"learning_rate": 0.00024687791239515373,
|
| 189 |
+
"loss": 0.8076,
|
| 190 |
+
"step": 290
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.77,
|
| 194 |
+
"learning_rate": 0.00024408201304753027,
|
| 195 |
+
"loss": 0.8173,
|
| 196 |
+
"step": 300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.79,
|
| 200 |
+
"learning_rate": 0.00024128611369990678,
|
| 201 |
+
"loss": 0.8032,
|
| 202 |
+
"step": 310
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.82,
|
| 206 |
+
"learning_rate": 0.0002384902143522833,
|
| 207 |
+
"loss": 0.8185,
|
| 208 |
+
"step": 320
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.84,
|
| 212 |
+
"learning_rate": 0.0002356943150046598,
|
| 213 |
+
"loss": 0.8046,
|
| 214 |
+
"step": 330
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.87,
|
| 218 |
+
"learning_rate": 0.00023289841565703632,
|
| 219 |
+
"loss": 0.812,
|
| 220 |
+
"step": 340
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.89,
|
| 224 |
+
"learning_rate": 0.00023010251630941285,
|
| 225 |
+
"loss": 0.807,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.92,
|
| 230 |
+
"learning_rate": 0.00022730661696178936,
|
| 231 |
+
"loss": 0.8003,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.95,
|
| 236 |
+
"learning_rate": 0.00022451071761416585,
|
| 237 |
+
"loss": 0.802,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.97,
|
| 242 |
+
"learning_rate": 0.00022171481826654239,
|
| 243 |
+
"loss": 0.8073,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.0,
|
| 248 |
+
"learning_rate": 0.0002189189189189189,
|
| 249 |
+
"loss": 0.8055,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.02,
|
| 254 |
+
"learning_rate": 0.00021612301957129543,
|
| 255 |
+
"loss": 0.8132,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.02,
|
| 260 |
+
"eval_loss": 0.8033702373504639,
|
| 261 |
+
"eval_runtime": 87.6065,
|
| 262 |
+
"eval_samples_per_second": 22.829,
|
| 263 |
+
"eval_steps_per_second": 2.854,
|
| 264 |
+
"step": 400
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.05,
|
| 268 |
+
"learning_rate": 0.00021332712022367195,
|
| 269 |
+
"loss": 0.7992,
|
| 270 |
+
"step": 410
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.07,
|
| 274 |
+
"learning_rate": 0.00021053122087604843,
|
| 275 |
+
"loss": 0.7945,
|
| 276 |
+
"step": 420
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.1,
|
| 280 |
+
"learning_rate": 0.00020773532152842497,
|
| 281 |
+
"loss": 0.7917,
|
| 282 |
+
"step": 430
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.12,
|
| 286 |
+
"learning_rate": 0.00020493942218080148,
|
| 287 |
+
"loss": 0.8002,
|
| 288 |
+
"step": 440
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.15,
|
| 292 |
+
"learning_rate": 0.00020214352283317796,
|
| 293 |
+
"loss": 0.7934,
|
| 294 |
+
"step": 450
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.18,
|
| 298 |
+
"learning_rate": 0.0001993476234855545,
|
| 299 |
+
"loss": 0.7982,
|
| 300 |
+
"step": 460
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.2,
|
| 304 |
+
"learning_rate": 0.000196551724137931,
|
| 305 |
+
"loss": 0.7995,
|
| 306 |
+
"step": 470
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.23,
|
| 310 |
+
"learning_rate": 0.00019375582479030755,
|
| 311 |
+
"loss": 0.7935,
|
| 312 |
+
"step": 480
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.25,
|
| 316 |
+
"learning_rate": 0.00019095992544268406,
|
| 317 |
+
"loss": 0.8003,
|
| 318 |
+
"step": 490
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.28,
|
| 322 |
+
"learning_rate": 0.00018816402609506054,
|
| 323 |
+
"loss": 0.7967,
|
| 324 |
+
"step": 500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.3,
|
| 328 |
+
"learning_rate": 0.00018536812674743708,
|
| 329 |
+
"loss": 0.7962,
|
| 330 |
+
"step": 510
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.33,
|
| 334 |
+
"learning_rate": 0.0001825722273998136,
|
| 335 |
+
"loss": 0.7956,
|
| 336 |
+
"step": 520
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.36,
|
| 340 |
+
"learning_rate": 0.00017977632805219013,
|
| 341 |
+
"loss": 0.7958,
|
| 342 |
+
"step": 530
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.38,
|
| 346 |
+
"learning_rate": 0.00017698042870456661,
|
| 347 |
+
"loss": 0.7944,
|
| 348 |
+
"step": 540
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.41,
|
| 352 |
+
"learning_rate": 0.00017418452935694312,
|
| 353 |
+
"loss": 0.7945,
|
| 354 |
+
"step": 550
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.43,
|
| 358 |
+
"learning_rate": 0.00017138863000931966,
|
| 359 |
+
"loss": 0.802,
|
| 360 |
+
"step": 560
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.46,
|
| 364 |
+
"learning_rate": 0.00016859273066169617,
|
| 365 |
+
"loss": 0.7923,
|
| 366 |
+
"step": 570
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.48,
|
| 370 |
+
"learning_rate": 0.00016579683131407266,
|
| 371 |
+
"loss": 0.8024,
|
| 372 |
+
"step": 580
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.51,
|
| 376 |
+
"learning_rate": 0.0001630009319664492,
|
| 377 |
+
"loss": 0.7902,
|
| 378 |
+
"step": 590
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.53,
|
| 382 |
+
"learning_rate": 0.0001602050326188257,
|
| 383 |
+
"loss": 0.8066,
|
| 384 |
+
"step": 600
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.53,
|
| 388 |
+
"eval_loss": 0.7989646196365356,
|
| 389 |
+
"eval_runtime": 87.5148,
|
| 390 |
+
"eval_samples_per_second": 22.853,
|
| 391 |
+
"eval_steps_per_second": 2.857,
|
| 392 |
+
"step": 600
|
| 393 |
+
}
|
| 394 |
+
],
|
| 395 |
+
"max_steps": 1173,
|
| 396 |
+
"num_train_epochs": 3,
|
| 397 |
+
"total_flos": 7.80240616893186e+17,
|
| 398 |
+
"trial_name": null,
|
| 399 |
+
"trial_params": null
|
| 400 |
+
}
|
checkpoint-600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dc5fdae5752fa38ccdc7bdda6792bd163b3cd2401b9536f9886aa0aba7720a1
|
| 3 |
+
size 3899
|
checkpoint-800/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-800/adapter_config.json
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"auto_mapping": null,
|
| 3 |
+
"base_model_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"lora_alpha": 16,
|
| 11 |
+
"lora_dropout": 0.05,
|
| 12 |
+
"modules_to_save": null,
|
| 13 |
+
"peft_type": "LORA",
|
| 14 |
+
"r": 16,
|
| 15 |
+
"revision": null,
|
| 16 |
+
"target_modules": [
|
| 17 |
+
"q_proj",
|
| 18 |
+
"v_proj"
|
| 19 |
+
],
|
| 20 |
+
"task_type": "CAUSAL_LM"
|
| 21 |
+
}
|
checkpoint-800/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-800/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:50880fdb0d632e86027b933470f8ed3b30672eb8b87fad7355ebe939ea3216a5
|
| 3 |
+
size 67216581
|
checkpoint-800/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1d16b27ee45106b68fc8c05dd7f883cff7b9fdd0ec416a2165e5dc7d013cb876
|
| 3 |
+
size 14575
|
checkpoint-800/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f200c0fb32253ad1d9bbd10d25da30d194c32e235170770e314055c37d8fd903
|
| 3 |
+
size 627
|
checkpoint-800/trainer_state.json
ADDED
|
@@ -0,0 +1,528 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7959330677986145,
|
| 3 |
+
"best_model_checkpoint": "./lora-alpaca/checkpoint-800",
|
| 4 |
+
"epoch": 2.045381911153723,
|
| 5 |
+
"global_step": 800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 2.9999999999999997e-05,
|
| 13 |
+
"loss": 1.6613,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 19 |
+
"loss": 1.6025,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.08,
|
| 24 |
+
"learning_rate": 8.999999999999999e-05,
|
| 25 |
+
"loss": 1.4512,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.1,
|
| 30 |
+
"learning_rate": 0.00011999999999999999,
|
| 31 |
+
"loss": 1.1653,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.13,
|
| 36 |
+
"learning_rate": 0.00015,
|
| 37 |
+
"loss": 0.9588,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.15,
|
| 42 |
+
"learning_rate": 0.00017999999999999998,
|
| 43 |
+
"loss": 0.9058,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.18,
|
| 48 |
+
"learning_rate": 0.00020999999999999998,
|
| 49 |
+
"loss": 0.8655,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.2,
|
| 54 |
+
"learning_rate": 0.00023999999999999998,
|
| 55 |
+
"loss": 0.8493,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.23,
|
| 60 |
+
"learning_rate": 0.00027,
|
| 61 |
+
"loss": 0.8462,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.26,
|
| 66 |
+
"learning_rate": 0.0003,
|
| 67 |
+
"loss": 0.8353,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.28,
|
| 72 |
+
"learning_rate": 0.0002972041006523765,
|
| 73 |
+
"loss": 0.833,
|
| 74 |
+
"step": 110
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.31,
|
| 78 |
+
"learning_rate": 0.000294408201304753,
|
| 79 |
+
"loss": 0.8322,
|
| 80 |
+
"step": 120
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.33,
|
| 84 |
+
"learning_rate": 0.0002916123019571295,
|
| 85 |
+
"loss": 0.8239,
|
| 86 |
+
"step": 130
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.36,
|
| 90 |
+
"learning_rate": 0.00028881640260950607,
|
| 91 |
+
"loss": 0.8174,
|
| 92 |
+
"step": 140
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.38,
|
| 96 |
+
"learning_rate": 0.00028602050326188253,
|
| 97 |
+
"loss": 0.8178,
|
| 98 |
+
"step": 150
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.41,
|
| 102 |
+
"learning_rate": 0.00028322460391425904,
|
| 103 |
+
"loss": 0.8321,
|
| 104 |
+
"step": 160
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.43,
|
| 108 |
+
"learning_rate": 0.0002804287045666356,
|
| 109 |
+
"loss": 0.8213,
|
| 110 |
+
"step": 170
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.46,
|
| 114 |
+
"learning_rate": 0.0002776328052190121,
|
| 115 |
+
"loss": 0.8088,
|
| 116 |
+
"step": 180
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.49,
|
| 120 |
+
"learning_rate": 0.00027483690587138857,
|
| 121 |
+
"loss": 0.8021,
|
| 122 |
+
"step": 190
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.51,
|
| 126 |
+
"learning_rate": 0.00027204100652376514,
|
| 127 |
+
"loss": 0.8297,
|
| 128 |
+
"step": 200
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.51,
|
| 132 |
+
"eval_loss": 0.8141016960144043,
|
| 133 |
+
"eval_runtime": 87.4274,
|
| 134 |
+
"eval_samples_per_second": 22.876,
|
| 135 |
+
"eval_steps_per_second": 2.86,
|
| 136 |
+
"step": 200
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.54,
|
| 140 |
+
"learning_rate": 0.00026924510717614165,
|
| 141 |
+
"loss": 0.8148,
|
| 142 |
+
"step": 210
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.56,
|
| 146 |
+
"learning_rate": 0.00026644920782851816,
|
| 147 |
+
"loss": 0.8204,
|
| 148 |
+
"step": 220
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.59,
|
| 152 |
+
"learning_rate": 0.00026365330848089467,
|
| 153 |
+
"loss": 0.8181,
|
| 154 |
+
"step": 230
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.61,
|
| 158 |
+
"learning_rate": 0.0002608574091332712,
|
| 159 |
+
"loss": 0.8146,
|
| 160 |
+
"step": 240
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.64,
|
| 164 |
+
"learning_rate": 0.0002580615097856477,
|
| 165 |
+
"loss": 0.8108,
|
| 166 |
+
"step": 250
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.66,
|
| 170 |
+
"learning_rate": 0.0002552656104380242,
|
| 171 |
+
"loss": 0.8069,
|
| 172 |
+
"step": 260
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.69,
|
| 176 |
+
"learning_rate": 0.0002524697110904007,
|
| 177 |
+
"loss": 0.8219,
|
| 178 |
+
"step": 270
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.72,
|
| 182 |
+
"learning_rate": 0.0002496738117427772,
|
| 183 |
+
"loss": 0.8076,
|
| 184 |
+
"step": 280
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.74,
|
| 188 |
+
"learning_rate": 0.00024687791239515373,
|
| 189 |
+
"loss": 0.8076,
|
| 190 |
+
"step": 290
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.77,
|
| 194 |
+
"learning_rate": 0.00024408201304753027,
|
| 195 |
+
"loss": 0.8173,
|
| 196 |
+
"step": 300
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.79,
|
| 200 |
+
"learning_rate": 0.00024128611369990678,
|
| 201 |
+
"loss": 0.8032,
|
| 202 |
+
"step": 310
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.82,
|
| 206 |
+
"learning_rate": 0.0002384902143522833,
|
| 207 |
+
"loss": 0.8185,
|
| 208 |
+
"step": 320
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.84,
|
| 212 |
+
"learning_rate": 0.0002356943150046598,
|
| 213 |
+
"loss": 0.8046,
|
| 214 |
+
"step": 330
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.87,
|
| 218 |
+
"learning_rate": 0.00023289841565703632,
|
| 219 |
+
"loss": 0.812,
|
| 220 |
+
"step": 340
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.89,
|
| 224 |
+
"learning_rate": 0.00023010251630941285,
|
| 225 |
+
"loss": 0.807,
|
| 226 |
+
"step": 350
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.92,
|
| 230 |
+
"learning_rate": 0.00022730661696178936,
|
| 231 |
+
"loss": 0.8003,
|
| 232 |
+
"step": 360
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.95,
|
| 236 |
+
"learning_rate": 0.00022451071761416585,
|
| 237 |
+
"loss": 0.802,
|
| 238 |
+
"step": 370
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 0.97,
|
| 242 |
+
"learning_rate": 0.00022171481826654239,
|
| 243 |
+
"loss": 0.8073,
|
| 244 |
+
"step": 380
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.0,
|
| 248 |
+
"learning_rate": 0.0002189189189189189,
|
| 249 |
+
"loss": 0.8055,
|
| 250 |
+
"step": 390
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.02,
|
| 254 |
+
"learning_rate": 0.00021612301957129543,
|
| 255 |
+
"loss": 0.8132,
|
| 256 |
+
"step": 400
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.02,
|
| 260 |
+
"eval_loss": 0.8033702373504639,
|
| 261 |
+
"eval_runtime": 87.6065,
|
| 262 |
+
"eval_samples_per_second": 22.829,
|
| 263 |
+
"eval_steps_per_second": 2.854,
|
| 264 |
+
"step": 400
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 1.05,
|
| 268 |
+
"learning_rate": 0.00021332712022367195,
|
| 269 |
+
"loss": 0.7992,
|
| 270 |
+
"step": 410
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 1.07,
|
| 274 |
+
"learning_rate": 0.00021053122087604843,
|
| 275 |
+
"loss": 0.7945,
|
| 276 |
+
"step": 420
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.1,
|
| 280 |
+
"learning_rate": 0.00020773532152842497,
|
| 281 |
+
"loss": 0.7917,
|
| 282 |
+
"step": 430
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 1.12,
|
| 286 |
+
"learning_rate": 0.00020493942218080148,
|
| 287 |
+
"loss": 0.8002,
|
| 288 |
+
"step": 440
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.15,
|
| 292 |
+
"learning_rate": 0.00020214352283317796,
|
| 293 |
+
"loss": 0.7934,
|
| 294 |
+
"step": 450
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 1.18,
|
| 298 |
+
"learning_rate": 0.0001993476234855545,
|
| 299 |
+
"loss": 0.7982,
|
| 300 |
+
"step": 460
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.2,
|
| 304 |
+
"learning_rate": 0.000196551724137931,
|
| 305 |
+
"loss": 0.7995,
|
| 306 |
+
"step": 470
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 1.23,
|
| 310 |
+
"learning_rate": 0.00019375582479030755,
|
| 311 |
+
"loss": 0.7935,
|
| 312 |
+
"step": 480
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 1.25,
|
| 316 |
+
"learning_rate": 0.00019095992544268406,
|
| 317 |
+
"loss": 0.8003,
|
| 318 |
+
"step": 490
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.28,
|
| 322 |
+
"learning_rate": 0.00018816402609506054,
|
| 323 |
+
"loss": 0.7967,
|
| 324 |
+
"step": 500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.3,
|
| 328 |
+
"learning_rate": 0.00018536812674743708,
|
| 329 |
+
"loss": 0.7962,
|
| 330 |
+
"step": 510
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 1.33,
|
| 334 |
+
"learning_rate": 0.0001825722273998136,
|
| 335 |
+
"loss": 0.7956,
|
| 336 |
+
"step": 520
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.36,
|
| 340 |
+
"learning_rate": 0.00017977632805219013,
|
| 341 |
+
"loss": 0.7958,
|
| 342 |
+
"step": 530
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 1.38,
|
| 346 |
+
"learning_rate": 0.00017698042870456661,
|
| 347 |
+
"loss": 0.7944,
|
| 348 |
+
"step": 540
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.41,
|
| 352 |
+
"learning_rate": 0.00017418452935694312,
|
| 353 |
+
"loss": 0.7945,
|
| 354 |
+
"step": 550
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.43,
|
| 358 |
+
"learning_rate": 0.00017138863000931966,
|
| 359 |
+
"loss": 0.802,
|
| 360 |
+
"step": 560
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.46,
|
| 364 |
+
"learning_rate": 0.00016859273066169617,
|
| 365 |
+
"loss": 0.7923,
|
| 366 |
+
"step": 570
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.48,
|
| 370 |
+
"learning_rate": 0.00016579683131407266,
|
| 371 |
+
"loss": 0.8024,
|
| 372 |
+
"step": 580
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.51,
|
| 376 |
+
"learning_rate": 0.0001630009319664492,
|
| 377 |
+
"loss": 0.7902,
|
| 378 |
+
"step": 590
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.53,
|
| 382 |
+
"learning_rate": 0.0001602050326188257,
|
| 383 |
+
"loss": 0.8066,
|
| 384 |
+
"step": 600
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.53,
|
| 388 |
+
"eval_loss": 0.7989646196365356,
|
| 389 |
+
"eval_runtime": 87.5148,
|
| 390 |
+
"eval_samples_per_second": 22.853,
|
| 391 |
+
"eval_steps_per_second": 2.857,
|
| 392 |
+
"step": 600
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 1.56,
|
| 396 |
+
"learning_rate": 0.00015740913327120224,
|
| 397 |
+
"loss": 0.7958,
|
| 398 |
+
"step": 610
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 1.59,
|
| 402 |
+
"learning_rate": 0.00015461323392357873,
|
| 403 |
+
"loss": 0.7837,
|
| 404 |
+
"step": 620
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 1.61,
|
| 408 |
+
"learning_rate": 0.00015181733457595524,
|
| 409 |
+
"loss": 0.8008,
|
| 410 |
+
"step": 630
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.64,
|
| 414 |
+
"learning_rate": 0.00014902143522833175,
|
| 415 |
+
"loss": 0.7964,
|
| 416 |
+
"step": 640
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.66,
|
| 420 |
+
"learning_rate": 0.0001462255358807083,
|
| 421 |
+
"loss": 0.7951,
|
| 422 |
+
"step": 650
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.69,
|
| 426 |
+
"learning_rate": 0.0001434296365330848,
|
| 427 |
+
"loss": 0.8012,
|
| 428 |
+
"step": 660
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.71,
|
| 432 |
+
"learning_rate": 0.0001406337371854613,
|
| 433 |
+
"loss": 0.7928,
|
| 434 |
+
"step": 670
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.74,
|
| 438 |
+
"learning_rate": 0.00013783783783783782,
|
| 439 |
+
"loss": 0.7903,
|
| 440 |
+
"step": 680
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.76,
|
| 444 |
+
"learning_rate": 0.00013504193849021433,
|
| 445 |
+
"loss": 0.7949,
|
| 446 |
+
"step": 690
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.79,
|
| 450 |
+
"learning_rate": 0.00013224603914259084,
|
| 451 |
+
"loss": 0.7866,
|
| 452 |
+
"step": 700
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.82,
|
| 456 |
+
"learning_rate": 0.00012945013979496738,
|
| 457 |
+
"loss": 0.7889,
|
| 458 |
+
"step": 710
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.84,
|
| 462 |
+
"learning_rate": 0.0001266542404473439,
|
| 463 |
+
"loss": 0.7928,
|
| 464 |
+
"step": 720
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.87,
|
| 468 |
+
"learning_rate": 0.0001238583410997204,
|
| 469 |
+
"loss": 0.7932,
|
| 470 |
+
"step": 730
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.89,
|
| 474 |
+
"learning_rate": 0.00012106244175209691,
|
| 475 |
+
"loss": 0.7963,
|
| 476 |
+
"step": 740
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.92,
|
| 480 |
+
"learning_rate": 0.00011826654240447344,
|
| 481 |
+
"loss": 0.7979,
|
| 482 |
+
"step": 750
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 1.94,
|
| 486 |
+
"learning_rate": 0.00011547064305684995,
|
| 487 |
+
"loss": 0.7943,
|
| 488 |
+
"step": 760
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 1.97,
|
| 492 |
+
"learning_rate": 0.00011267474370922644,
|
| 493 |
+
"loss": 0.7983,
|
| 494 |
+
"step": 770
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 1.99,
|
| 498 |
+
"learning_rate": 0.00010987884436160297,
|
| 499 |
+
"loss": 0.801,
|
| 500 |
+
"step": 780
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 2.02,
|
| 504 |
+
"learning_rate": 0.0001070829450139795,
|
| 505 |
+
"loss": 0.7836,
|
| 506 |
+
"step": 790
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 2.05,
|
| 510 |
+
"learning_rate": 0.000104287045666356,
|
| 511 |
+
"loss": 0.785,
|
| 512 |
+
"step": 800
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 2.05,
|
| 516 |
+
"eval_loss": 0.7959330677986145,
|
| 517 |
+
"eval_runtime": 88.4745,
|
| 518 |
+
"eval_samples_per_second": 22.605,
|
| 519 |
+
"eval_steps_per_second": 2.826,
|
| 520 |
+
"step": 800
|
| 521 |
+
}
|
| 522 |
+
],
|
| 523 |
+
"max_steps": 1173,
|
| 524 |
+
"num_train_epochs": 3,
|
| 525 |
+
"total_flos": 1.040223650649342e+18,
|
| 526 |
+
"trial_name": null,
|
| 527 |
+
"trial_params": null
|
| 528 |
+
}
|
checkpoint-800/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7dc5fdae5752fa38ccdc7bdda6792bd163b3cd2401b9536f9886aa0aba7720a1
|
| 3 |
+
size 3899
|