Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language:
|
| 3 |
+
- fr
|
| 4 |
+
- en
|
| 5 |
+
tags:
|
| 6 |
+
- mistral
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# mistral-7B-v0.1
|
| 10 |
+
|
| 11 |
+
Released Sept. 27 by MistralAI with no further information.
|
| 12 |
+
|
| 13 |
+
### Model upload under way
|
| 14 |
+
|
| 15 |
+
Should be up within the hour
|
| 16 |
+
|
| 17 |
+
### Config
|
| 18 |
+
|
| 19 |
+
```json
|
| 20 |
+
{
|
| 21 |
+
"dim": 4096,
|
| 22 |
+
"n_layers": 32,
|
| 23 |
+
"head_dim": 128,
|
| 24 |
+
"hidden_dim": 14336,
|
| 25 |
+
"n_heads": 32,
|
| 26 |
+
"n_kv_heads": 8,
|
| 27 |
+
"norm_eps": 1e-05,
|
| 28 |
+
"sliding_window": 4096,
|
| 29 |
+
"vocab_size": 32000
|
| 30 |
+
}
|
| 31 |
+
```
|
| 32 |
+
|
| 33 |
+
### Training data
|
| 34 |
+
|
| 35 |
+
Potentially, up to 8T tokens, with English, French, Code...
|
| 36 |
+
|
| 37 |
+
https://twitter.com/ManuelFaysse/status/1706949891358859624
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
### Magnet link
|
| 41 |
+
|
| 42 |
+
`magnet:?xt=urn:btih:208b101a0f51514ecf285885a8b0f6fb1a1e4d7d&dn=mistral-7B-v0.1&tr=udp%3A%2F%http://2Ftracker.opentrackr.org%3A1337%2Fannounce&tr=https%3A%2F%http://2Ftracker1.520.jp%3A443%2Fannounce`
|