Update README.md
Browse files
README.md
CHANGED
@@ -11,26 +11,24 @@ model-index:
|
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
should probably proofread and complete it, then remove this comment. -->
|
13 |
|
14 |
-
# 2.
|
15 |
|
16 |
-
|
|
|
|
|
17 |
It achieves the following results on the evaluation set:
|
18 |
- Loss: 2.4748
|
19 |
- Accuracy: 0.4671
|
20 |
|
21 |
## Model description
|
22 |
|
23 |
-
|
24 |
-
|
25 |
-
## Intended uses & limitations
|
26 |
|
27 |
-
More information needed
|
28 |
|
29 |
-
## Training
|
30 |
|
31 |
-
|
32 |
|
33 |
-
## Training procedure
|
34 |
|
35 |
### Training hyperparameters
|
36 |
|
@@ -49,7 +47,7 @@ The following hyperparameters were used during training:
|
|
49 |
|
50 |
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|
51 |
|:-------------:|:-----:|:-----:|:--------:|:---------------:|
|
52 |
-
| 2.3941 |
|
53 |
|
54 |
|
55 |
### Framework versions
|
|
|
11 |
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
12 |
should probably proofread and complete it, then remove this comment. -->
|
13 |
|
14 |
+
# Neo-VN-2.7B
|
15 |
|
16 |
+
Work-in-progress. Current version is very bad.
|
17 |
+
|
18 |
+
This model is a fine-tuned version of [huggingface.co/EleutherAI/opt-neo-2.7b](https://huggingface.co//notebooks/convogpt/gpt-neo-2.7B/) on an unknown dataset.
|
19 |
It achieves the following results on the evaluation set:
|
20 |
- Loss: 2.4748
|
21 |
- Accuracy: 0.4671
|
22 |
|
23 |
## Model description
|
24 |
|
25 |
+
Neo-VN-2.7B is a transformer model based on EleutherAI's replication of GPT-3.
|
|
|
|
|
26 |
|
|
|
27 |
|
28 |
+
## Training procedure
|
29 |
|
30 |
+
Fine-tuned on 233MB of conversational data.
|
31 |
|
|
|
32 |
|
33 |
### Training hyperparameters
|
34 |
|
|
|
47 |
|
48 |
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|
49 |
|:-------------:|:-----:|:-----:|:--------:|:---------------:|
|
50 |
+
| 2.3941 | 1 | 20460 | 0.4671 | 2.4748 |
|
51 |
|
52 |
|
53 |
### Framework versions
|