ArimaKana38
commited on
Commit
•
92a0021
1
Parent(s):
69a84ac
Upload 5 files
Browse files- README.md +3 -0
- adapter_config.json +20 -0
- adapter_model.bin +3 -0
- training_log.json +10 -0
- training_parameters.json +26 -0
README.md
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: peft
|
3 |
+
---
|
adapter_config.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"base_model_name_or_path": "models/chavinlo_alpaca-native",
|
3 |
+
"bias": "none",
|
4 |
+
"fan_in_fan_out": false,
|
5 |
+
"inference_mode": true,
|
6 |
+
"init_lora_weights": true,
|
7 |
+
"layers_pattern": null,
|
8 |
+
"layers_to_transform": null,
|
9 |
+
"lora_alpha": 64,
|
10 |
+
"lora_dropout": 0.05,
|
11 |
+
"modules_to_save": null,
|
12 |
+
"peft_type": "LORA",
|
13 |
+
"r": 32,
|
14 |
+
"revision": null,
|
15 |
+
"target_modules": [
|
16 |
+
"q_proj",
|
17 |
+
"v_proj"
|
18 |
+
],
|
19 |
+
"task_type": "CAUSAL_LM"
|
20 |
+
}
|
adapter_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:43de423b165d4d64fe27290ec2099cf45cc5742d3224e5615aed5f83aee1683f
|
3 |
+
size 51642368
|
training_log.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"loss": 1.1535,
|
3 |
+
"learning_rate": 7.5e-05,
|
4 |
+
"epoch": 2.7,
|
5 |
+
"train_runtime": 3302.3968,
|
6 |
+
"train_samples_per_second": 4.12,
|
7 |
+
"train_steps_per_second": 0.004,
|
8 |
+
"total_flos": 1.2484302776303616e+17,
|
9 |
+
"train_loss": 1.434487760066986
|
10 |
+
}
|
training_parameters.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"lora_name": "lora-minnie-test",
|
3 |
+
"always_override": false,
|
4 |
+
"save_steps": 3.0,
|
5 |
+
"micro_batch_size": 4,
|
6 |
+
"batch_size": 128,
|
7 |
+
"epochs": 3.0,
|
8 |
+
"learning_rate": "3e-4",
|
9 |
+
"lr_scheduler_type": "linear",
|
10 |
+
"lora_rank": 32,
|
11 |
+
"lora_alpha": 64,
|
12 |
+
"lora_dropout": 0.05,
|
13 |
+
"cutoff_len": 256,
|
14 |
+
"dataset": "codegen-instruct",
|
15 |
+
"eval_dataset": "None",
|
16 |
+
"format": "alpaca-format",
|
17 |
+
"eval_steps": 100.0,
|
18 |
+
"raw_text_file": "None",
|
19 |
+
"overlap_len": 128,
|
20 |
+
"newline_favor_len": 128,
|
21 |
+
"higher_rank_limit": false,
|
22 |
+
"warmup_steps": 100.0,
|
23 |
+
"optimizer": "adamw_torch",
|
24 |
+
"hard_cut_string": "\\n\\n\\n",
|
25 |
+
"train_only_after": ""
|
26 |
+
}
|