Update README.md
Browse files
README.md
CHANGED
@@ -32,6 +32,7 @@ library_name: transformers
|
|
32 |
<h1 style="margin-top:0rem; margin-bottom: 0rem;">🐋 DeepSeek-R1-0528-Qwen3-8B Usage Guidelines</h1>
|
33 |
</div>
|
34 |
|
|
|
35 |
- Set the temperature between **0.5–0.7 (0.6 recommended)** to reduce repetition and incoherence.
|
36 |
- Set Top_P value of **0.95 (recommended)**
|
37 |
- R1-0528 uses the same chat template as the original R1 model:
|
|
|
32 |
<h1 style="margin-top:0rem; margin-bottom: 0rem;">🐋 DeepSeek-R1-0528-Qwen3-8B Usage Guidelines</h1>
|
33 |
</div>
|
34 |
|
35 |
+
- For Ollama do `ollama run hf.co/unsloth/DeepSeek-R1-0528-Qwen3-8B-GGUF:Q4_K_XL` - it'll auto get the correct chat template and all settings
|
36 |
- Set the temperature between **0.5–0.7 (0.6 recommended)** to reduce repetition and incoherence.
|
37 |
- Set Top_P value of **0.95 (recommended)**
|
38 |
- R1-0528 uses the same chat template as the original R1 model:
|