Hugging Face's logo

allenai
/
tmax-8b

Safetensors
qwen3
Model card Files
xet
 Community
1
hamishivi commited on
Commit
5e4c096
·
verified ·
1 Parent(s): a349916

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -73,7 +73,7 @@ For more details on evaluation, please see [our codebase](https://github.com/ham
73
  ### Hyperparameters
74
 
75
  This model was trained using DPPO with the following hyperparameters:
76
- - **base model**: [hamishivi/Qwen3.5-4B](https://huggingface.co/hamishivi/Qwen3.5-4B)
77
  - **Dataset**: [tmax 15K](https://huggingface.co/datasets/allenai/tmax-15k-open-instruct)
78
  - **Max prompt tokens**: 2048
79
  - **Max per-turn tokens**: 16384
 
73
  ### Hyperparameters
74
 
75
  This model was trained using DPPO with the following hyperparameters:
76
+ - **base model**: [allenai/tmax-sft-8b](https://huggingface.co/allenai/tmax-sft-8b)
77
  - **Dataset**: [tmax 15K](https://huggingface.co/datasets/allenai/tmax-15k-open-instruct)
78
  - **Max prompt tokens**: 2048
79
  - **Max per-turn tokens**: 16384