Hugging Face's logo

danielcherubini
/
Qwen3.5-DeltaCoder-9B

Text Generation
Safetensors
English
qwen3.5
code
tool-calling
lora
sft
dpo
unsloth
reasoning
chain-of-thought
conversational
Model card Files
xet
 Community
danielcherubini commited on
Commit
09ef04a
·
verified ·
1 Parent(s): fba73ed

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +8 -4
README.md CHANGED
@@ -149,10 +149,14 @@ tokenizer = AutoTokenizer.from_pretrained("danielcherubini/Qwen3.5-DeltaCoder-9B
149
 
150
  ## Recommended Sampling Settings
151
 
152
- | Profile | temperature | top_k | top_p | min_p | presence_penalty |
153
- |---------|-------------|-------|-------|-------|-----------------|
154
- | **Coding** | 0.6 | 20 | 0.95 | 0.0 | 0.0 |
155
- | **Chat** | 1.0 | 20 | 0.95 | 0.0 | 1.5 |
 
 
 
 
156
 
157
  > [!WARNING]
158
  > **Do not use temperature below 0.5** — low temperatures cause deterministic looping in multi-turn agentic use.
 
149
 
150
  ## Recommended Sampling Settings
151
 
152
+ | Parameter | Value |
153
+ |-----------|-------|
154
+ | temperature | 0.6 |
155
+ | top_k | 20 |
156
+ | top_p | 0.95 |
157
+ | min_p | 0.0 |
158
+ | presence_penalty | 0.0 |
159
+ | repeat_penalty | 1.0 |
160
 
161
  > [!WARNING]
162
  > **Do not use temperature below 0.5** — low temperatures cause deterministic looping in multi-turn agentic use.