End of training
Browse files- README.md +89 -0
- adapter_config.json +28 -0
- adapter_model.safetensors +3 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
language:
|
3 |
+
- ar
|
4 |
+
library_name: peft
|
5 |
+
tags:
|
6 |
+
- generated_from_trainer
|
7 |
+
datasets:
|
8 |
+
- dalyaff/darebah-2K
|
9 |
+
base_model: microsoft/phi-2
|
10 |
+
model-index:
|
11 |
+
- name: phi-2
|
12 |
+
results: []
|
13 |
+
---
|
14 |
+
|
15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
16 |
+
should probably proofread and complete it, then remove this comment. -->
|
17 |
+
|
18 |
+
# phi-2
|
19 |
+
|
20 |
+
This model is a fine-tuned version of [microsoftl](https://huggingface.co/microsoftl) on the dalyaff/darebah-2K dataset.
|
21 |
+
It achieves the following results on the evaluation set:
|
22 |
+
- Loss: 0.5861
|
23 |
+
|
24 |
+
## Model description
|
25 |
+
|
26 |
+
More information needed
|
27 |
+
|
28 |
+
## Intended uses & limitations
|
29 |
+
|
30 |
+
More information needed
|
31 |
+
|
32 |
+
## Training and evaluation data
|
33 |
+
|
34 |
+
More information needed
|
35 |
+
|
36 |
+
## Training procedure
|
37 |
+
|
38 |
+
### Training hyperparameters
|
39 |
+
|
40 |
+
The following hyperparameters were used during training:
|
41 |
+
- learning_rate: 2.5e-05
|
42 |
+
- train_batch_size: 2
|
43 |
+
- eval_batch_size: 8
|
44 |
+
- seed: 42
|
45 |
+
- gradient_accumulation_steps: 4
|
46 |
+
- total_train_batch_size: 8
|
47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
48 |
+
- lr_scheduler_type: linear
|
49 |
+
- lr_scheduler_warmup_steps: 5
|
50 |
+
- training_steps: 2500
|
51 |
+
|
52 |
+
### Training results
|
53 |
+
|
54 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
55 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
56 |
+
| 1.2333 | 0.44 | 100 | 1.1475 |
|
57 |
+
| 1.0155 | 0.89 | 200 | 0.9995 |
|
58 |
+
| 0.9331 | 1.33 | 300 | 0.9305 |
|
59 |
+
| 0.8308 | 1.78 | 400 | 0.8755 |
|
60 |
+
| 0.7955 | 2.22 | 500 | 0.8337 |
|
61 |
+
| 0.7788 | 2.67 | 600 | 0.7974 |
|
62 |
+
| 0.7205 | 3.11 | 700 | 0.7685 |
|
63 |
+
| 0.7588 | 3.56 | 800 | 0.7405 |
|
64 |
+
| 0.7154 | 4.0 | 900 | 0.7193 |
|
65 |
+
| 0.6913 | 4.44 | 1000 | 0.7022 |
|
66 |
+
| 0.6049 | 4.89 | 1100 | 0.6864 |
|
67 |
+
| 0.6546 | 5.33 | 1200 | 0.6672 |
|
68 |
+
| 0.6406 | 5.78 | 1300 | 0.6576 |
|
69 |
+
| 0.6237 | 6.22 | 1400 | 0.6433 |
|
70 |
+
| 0.6016 | 6.67 | 1500 | 0.6313 |
|
71 |
+
| 0.5746 | 7.11 | 1600 | 0.6263 |
|
72 |
+
| 0.5786 | 7.56 | 1700 | 0.6156 |
|
73 |
+
| 0.592 | 8.0 | 1800 | 0.6087 |
|
74 |
+
| 0.5908 | 8.44 | 1900 | 0.6030 |
|
75 |
+
| 0.598 | 8.89 | 2000 | 0.5980 |
|
76 |
+
| 0.5998 | 9.33 | 2100 | 0.5937 |
|
77 |
+
| 0.5859 | 9.78 | 2200 | 0.5909 |
|
78 |
+
| 0.5767 | 10.22 | 2300 | 0.5888 |
|
79 |
+
| 0.5786 | 10.67 | 2400 | 0.5872 |
|
80 |
+
| 0.5911 | 11.11 | 2500 | 0.5861 |
|
81 |
+
|
82 |
+
|
83 |
+
### Framework versions
|
84 |
+
|
85 |
+
- PEFT 0.7.2.dev0
|
86 |
+
- Transformers 4.38.0.dev0
|
87 |
+
- Pytorch 2.1.0+cu121
|
88 |
+
- Datasets 2.16.1
|
89 |
+
- Tokenizers 0.15.1
|
adapter_config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"alpha_pattern": {},
|
3 |
+
"auto_mapping": null,
|
4 |
+
"base_model_name_or_path": "microsoft/phi-2",
|
5 |
+
"bias": "none",
|
6 |
+
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
+
"init_lora_weights": true,
|
9 |
+
"layers_pattern": null,
|
10 |
+
"layers_to_transform": null,
|
11 |
+
"loftq_config": {},
|
12 |
+
"lora_alpha": 16,
|
13 |
+
"lora_dropout": 0.05,
|
14 |
+
"megatron_config": null,
|
15 |
+
"megatron_core": "megatron.core",
|
16 |
+
"modules_to_save": null,
|
17 |
+
"peft_type": "LORA",
|
18 |
+
"r": 8,
|
19 |
+
"rank_pattern": {},
|
20 |
+
"revision": null,
|
21 |
+
"target_modules": [
|
22 |
+
"Wqkv",
|
23 |
+
"fc1",
|
24 |
+
"fc2"
|
25 |
+
],
|
26 |
+
"task_type": "CAUSAL_LM",
|
27 |
+
"use_rslora": false
|
28 |
+
}
|
adapter_model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8feb629041bc1f5ffd1fd65f51b83907b3dfd20c7a0d68fddc4a7b2f8721fb79
|
3 |
+
size 26230352
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8199cdde49ef87742220eb1d0da7f7432801f443732f593316fa4e666a1c9663
|
3 |
+
size 4792
|