End of training
Browse files- README.md +77 -0
- adapter_config.json +20 -0
- adapter_model.bin +3 -0
- training_args.bin +3 -0
README.md
ADDED
@@ -0,0 +1,77 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: other
|
3 |
+
base_model: microsoft/phi-1_5
|
4 |
+
tags:
|
5 |
+
- generated_from_trainer
|
6 |
+
datasets:
|
7 |
+
- scitldr
|
8 |
+
model-index:
|
9 |
+
- name: arphiv
|
10 |
+
results: []
|
11 |
+
---
|
12 |
+
|
13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
14 |
+
should probably proofread and complete it, then remove this comment. -->
|
15 |
+
|
16 |
+
# arphiv
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [microsoft/phi-1_5](https://huggingface.co/microsoft/phi-1_5) on the scitldr dataset.
|
19 |
+
It achieves the following results on the evaluation set:
|
20 |
+
- Loss: 1.9806
|
21 |
+
|
22 |
+
## Model description
|
23 |
+
|
24 |
+
More information needed
|
25 |
+
|
26 |
+
## Intended uses & limitations
|
27 |
+
|
28 |
+
More information needed
|
29 |
+
|
30 |
+
## Training and evaluation data
|
31 |
+
|
32 |
+
More information needed
|
33 |
+
|
34 |
+
## Training procedure
|
35 |
+
|
36 |
+
### Training hyperparameters
|
37 |
+
|
38 |
+
The following hyperparameters were used during training:
|
39 |
+
- learning_rate: 0.001
|
40 |
+
- train_batch_size: 1
|
41 |
+
- eval_batch_size: 1
|
42 |
+
- seed: 42
|
43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
44 |
+
- lr_scheduler_type: linear
|
45 |
+
- num_epochs: 2
|
46 |
+
|
47 |
+
### Training results
|
48 |
+
|
49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
51 |
+
| 1.9802 | 0.1 | 200 | 2.0909 |
|
52 |
+
| 1.9764 | 0.2 | 400 | 2.0620 |
|
53 |
+
| 1.9247 | 0.3 | 600 | 2.0534 |
|
54 |
+
| 2.0485 | 0.4 | 800 | 2.0619 |
|
55 |
+
| 1.9351 | 0.5 | 1000 | 2.0143 |
|
56 |
+
| 1.9793 | 0.6 | 1200 | 2.0255 |
|
57 |
+
| 1.9704 | 0.7 | 1400 | 2.0190 |
|
58 |
+
| 1.9144 | 0.8 | 1600 | 1.9922 |
|
59 |
+
| 2.0139 | 0.9 | 1800 | 2.0024 |
|
60 |
+
| 1.9789 | 1.0 | 2000 | 1.9845 |
|
61 |
+
| 1.622 | 1.1 | 2200 | 2.0068 |
|
62 |
+
| 1.6057 | 1.2 | 2400 | 2.0123 |
|
63 |
+
| 1.6747 | 1.31 | 2600 | 1.9998 |
|
64 |
+
| 1.5369 | 1.41 | 2800 | 2.0055 |
|
65 |
+
| 1.5566 | 1.51 | 3000 | 2.0138 |
|
66 |
+
| 1.3799 | 1.61 | 3200 | 1.9924 |
|
67 |
+
| 1.5956 | 1.71 | 3400 | 1.9849 |
|
68 |
+
| 1.3932 | 1.81 | 3600 | 1.9926 |
|
69 |
+
| 1.4757 | 1.91 | 3800 | 1.9806 |
|
70 |
+
|
71 |
+
|
72 |
+
### Framework versions
|
73 |
+
|
74 |
+
- Transformers 4.34.1
|
75 |
+
- Pytorch 2.1.0+cu118
|
76 |
+
- Datasets 2.14.6
|
77 |
+
- Tokenizers 0.14.1
|
adapter_config.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"auto_mapping": null,
|
3 |
+
"base_model_name_or_path": "microsoft/phi-1_5",
|
4 |
+
"bias": "none",
|
5 |
+
"fan_in_fan_out": false,
|
6 |
+
"inference_mode": true,
|
7 |
+
"init_lora_weights": true,
|
8 |
+
"layers_pattern": null,
|
9 |
+
"layers_to_transform": null,
|
10 |
+
"lora_alpha": 32,
|
11 |
+
"lora_dropout": 0.05,
|
12 |
+
"modules_to_save": null,
|
13 |
+
"peft_type": "LORA",
|
14 |
+
"r": 32,
|
15 |
+
"revision": null,
|
16 |
+
"target_modules": [
|
17 |
+
"Wqkv"
|
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:b71caa32a5df32dd7e5cb39ce2ea6523ac79fda24648fdd88aa49bcb3bc11601
|
3 |
+
size 25182782
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:10bf4d9458caa71d28b0e45a298d5a62408362c9c6a6d3b7e1f33c87fa7b4d12
|
3 |
+
size 4472
|