Commit
•
532afd8
1
Parent(s):
14fdda1
End of training
Browse files
README.md
ADDED
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: gemma
|
3 |
+
library_name: peft
|
4 |
+
tags:
|
5 |
+
- trl
|
6 |
+
- sft
|
7 |
+
- generated_from_trainer
|
8 |
+
datasets:
|
9 |
+
- generator
|
10 |
+
base_model: google/gemma-2b
|
11 |
+
model-index:
|
12 |
+
- name: gemma-2b-dolly-qa
|
13 |
+
results: []
|
14 |
+
---
|
15 |
+
|
16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
17 |
+
should probably proofread and complete it, then remove this comment. -->
|
18 |
+
|
19 |
+
# gemma-2b-dolly-qa
|
20 |
+
|
21 |
+
This model is a fine-tuned version of [google/gemma-2b](https://huggingface.co/google/gemma-2b) on the generator dataset.
|
22 |
+
It achieves the following results on the evaluation set:
|
23 |
+
- Loss: 2.0215
|
24 |
+
|
25 |
+
## Model description
|
26 |
+
|
27 |
+
More information needed
|
28 |
+
|
29 |
+
## Intended uses & limitations
|
30 |
+
|
31 |
+
More information needed
|
32 |
+
|
33 |
+
## Training and evaluation data
|
34 |
+
|
35 |
+
More information needed
|
36 |
+
|
37 |
+
## Training procedure
|
38 |
+
|
39 |
+
### Training hyperparameters
|
40 |
+
|
41 |
+
The following hyperparameters were used during training:
|
42 |
+
- learning_rate: 1e-05
|
43 |
+
- train_batch_size: 2
|
44 |
+
- eval_batch_size: 8
|
45 |
+
- seed: 42
|
46 |
+
- gradient_accumulation_steps: 8
|
47 |
+
- total_train_batch_size: 16
|
48 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
49 |
+
- lr_scheduler_type: linear
|
50 |
+
- lr_scheduler_warmup_ratio: 0.05
|
51 |
+
- training_steps: 1480
|
52 |
+
|
53 |
+
### Training results
|
54 |
+
|
55 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
56 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
57 |
+
| 2.9198 | 1.64 | 100 | 2.5675 |
|
58 |
+
| 2.437 | 3.28 | 200 | 2.2818 |
|
59 |
+
| 2.2514 | 4.92 | 300 | 2.1677 |
|
60 |
+
| 2.1587 | 6.56 | 400 | 2.1038 |
|
61 |
+
| 2.116 | 8.2 | 500 | 2.0741 |
|
62 |
+
| 2.0794 | 9.84 | 600 | 2.0576 |
|
63 |
+
| 2.0663 | 11.48 | 700 | 2.0467 |
|
64 |
+
| 2.0494 | 13.11 | 800 | 2.0394 |
|
65 |
+
| 2.0449 | 14.75 | 900 | 2.0336 |
|
66 |
+
| 2.0336 | 16.39 | 1000 | 2.0293 |
|
67 |
+
| 2.0281 | 18.03 | 1100 | 2.0262 |
|
68 |
+
| 2.0172 | 19.67 | 1200 | 2.0240 |
|
69 |
+
| 2.0227 | 21.31 | 1300 | 2.0227 |
|
70 |
+
| 2.0128 | 22.95 | 1400 | 2.0215 |
|
71 |
+
|
72 |
+
|
73 |
+
### Framework versions
|
74 |
+
|
75 |
+
- PEFT 0.10.0
|
76 |
+
- Transformers 4.39.3
|
77 |
+
- Pytorch 2.0.1a0+cxx11.abi
|
78 |
+
- Datasets 2.18.0
|
79 |
+
- Tokenizers 0.15.2
|
adapter_model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 156926880
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:025af3a29f1b422d2ab442ed8fa6f4ce44eabb89b4d516ab0a6e214608ec8eb4
|
3 |
size 156926880
|
runs/Apr11_15-24-17_idc-beta-batch-pvc-node-14/events.out.tfevents.1712874260.idc-beta-batch-pvc-node-14
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
-
size
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:83402221f52ef2f05ed7fb2b42b03fb7dc315b20dfe52e99297cf0b33494af13
|
3 |
+
size 11817
|