Instructions to use Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40 with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-4B") model = PeftModel.from_pretrained(base_model, "Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40") - Notebooks
- Google Colab
- Kaggle
Download prompt_contract.json from Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40: direct link, hf CLI and curl.
- Browser
- Download file 314 Bytes
-
https://huggingface.co/Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40/resolve/main/prompt_contract.json
- Command line
-
hf download hf://Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40/prompt_contract.json
-
curl -L -o prompt_contract.json https://huggingface.co/Jazhyc/Gleipnir-4B-ToolTrajectories-InjectionAugmented40/resolve/main/prompt_contract.json
314 Bytes
| { | |
| "prompt_set_id": "tool-trajectory-binary-v1", | |
| "template_sha256": "f104aa18fe9f04094efdca14f9cebe65f7b18d976f454c134fa4282bd5162eaa", | |
| "trajectory_open": "<agent_trajectory>", | |
| "trajectory_close": "</agent_trajectory>", | |
| "decision_prefix": "Prediction:", | |
| "negative_label": "0", | |
| "positive_label": "1" | |
| } | |