Update README.md
Browse files
README.md
CHANGED
|
@@ -213,20 +213,6 @@ quantized_by: bartowski
|
|
| 213 |
|
| 214 |
## Exllama v2 Quantizations of Meta-Llama-3-8B-Instruct
|
| 215 |
|
| 216 |
-
If generation refuses to stop, you can edit tokenizer_config.json.
|
| 217 |
-
|
| 218 |
-
Replace line 2055:
|
| 219 |
-
|
| 220 |
-
```
|
| 221 |
-
"eos_token": "<|end_of_text|>",
|
| 222 |
-
```
|
| 223 |
-
|
| 224 |
-
with:
|
| 225 |
-
|
| 226 |
-
```
|
| 227 |
-
"eos_token": "<|eot_id|>",
|
| 228 |
-
```
|
| 229 |
-
|
| 230 |
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.19">turboderp's ExLlamaV2 v0.0.19</a> for quantization.
|
| 231 |
|
| 232 |
<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>
|
|
|
|
| 213 |
|
| 214 |
## Exllama v2 Quantizations of Meta-Llama-3-8B-Instruct
|
| 215 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 216 |
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.19">turboderp's ExLlamaV2 v0.0.19</a> for quantization.
|
| 217 |
|
| 218 |
<b>The "main" branch only contains the measurement.json, download one of the other branches for the model (see below)</b>
|