Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- it
|
| 4 |
+
pipeline_tag: text-generation
|
| 5 |
+
widget:
|
| 6 |
+
- text: Alessandro è un bel ragazzo che progetta Infissi
|
| 7 |
+
- text: Melissa è una ragazza
|
| 8 |
+
tags:
|
| 9 |
+
- italian
|
| 10 |
+
- italiano
|
| 11 |
+
- llama
|
| 12 |
+
---
|
| 13 |
+
This is a work in progress of an italian llama2 model with train starting from 0 with the redpajama database (currently only 2023-14 ONLY ITALIAN)
|
| 14 |
+
|
| 15 |
+
Train on my server, i have studied and adapted the model starting from the repository https://github.com/karpathy/llama2.c
|
| 16 |
+
|
| 17 |
+
*** TRAIN IS NOT FINISHED ***
|
| 18 |
+
|
| 19 |
+
1024 max sequence length
|
| 20 |
+
|
| 21 |
+
num decayed parameter tensors: 225, with 251,068,416 parameters
|
| 22 |
+
|
| 23 |
+
num non-decayed parameter tensors: 65, with 49,920 parameters
|