Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model:
|
| 4 |
+
- SicariusSicariiStuff/Llama-3.1-Nemotron-8B-UltraLong-1M-Instruct_Abliterated
|
| 5 |
+
tags:
|
| 6 |
+
- finetune
|
| 7 |
+
- llama
|
| 8 |
+
- raven
|
| 9 |
+
- poe
|
| 10 |
+
- gothic
|
| 11 |
+
- horror
|
| 12 |
+
- creative writing
|
| 13 |
+
- RP
|
| 14 |
+
datasets:
|
| 15 |
+
- DarkArtsForge/Poe_v1
|
| 16 |
+
language:
|
| 17 |
+
- en
|
| 18 |
+
library_name: transformers
|
| 19 |
+
widget:
|
| 20 |
+
- text: "Raven 8B v1"
|
| 21 |
+
output:
|
| 22 |
+
url: https://cdn-uploads.huggingface.co/production/uploads/68e840caa318194c44ec2a04/gPKk1RgcW0QN0NAVpf4lh.jpeg
|
| 23 |
+
---
|
| 24 |
+
|
| 25 |
+
> [!CAUTION]
|
| 26 |
+
> <span style="color:red; font-weight:bold">⚠️ Warning:</span> This model can produce narratives and RP that contain violent and graphic erotic content. Adjust your system prompt accordingly, and use **Llama 3** chat template.
|
| 27 |
+
>
|
| 28 |
+
|
| 29 |
+
# Raven 8B v1
|
| 30 |
+
A [fully uncensored](https://huggingface.co/SicariusSicariiStuff/Llama-3.1-Nemotron-8B-UltraLong-1M-Instruct_Abliterated) finetune of Llama-3.1-Nemotron-8B trained on a small dataset of [Edgar Allan Poe](https://huggingface.co/datasets/DarkArtsForge/Poe_v1) corpus. Cooked for 5 epochs using PMPF.
|
| 31 |
+
|
| 32 |
+
```
|
| 33 |
+
{'loss': 0.1136, 'grad_norm': 1.0182174444198608, 'learning_rate': 1.685173482438018e-08, 'entropy': 0.18156841583549976, 'num_tokens': 99475.0, 'mean_token_accuracy': 0.9738506525754929, 'epoch': 5.0}
|
| 34 |
+
{'train_runtime': 590.173, 'train_samples_per_second': 0.847, 'train_steps_per_second': 0.212, 'train_loss': 1.036527609705925, 'epoch': 5.0}
|
| 35 |
+
```
|
| 36 |
+
|
| 37 |
+
|