Text Generation
PEFT
Safetensors
Chinese
qwen3
lora
dialogue-act-planning
psychological-counseling
conversational
Instructions to use UEC-InabaLab/CounselTraj with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use UEC-InabaLab/CounselTraj with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B") model = PeftModel.from_pretrained(base_model, "UEC-InabaLab/CounselTraj") - Notebooks
- Google Colab
- Kaggle
Download model_details.json from UEC-InabaLab/CounselTraj: direct link, hf CLI and curl.
- Browser
- Download file 964 Bytes
-
https://huggingface.co/UEC-InabaLab/CounselTraj/resolve/main/model_details.json
- Command line
-
hf download hf://UEC-InabaLab/CounselTraj/model_details.json
-
curl -L -o model_details.json https://huggingface.co/UEC-InabaLab/CounselTraj/resolve/main/model_details.json
964 Bytes
| { | |
| "base_model": "Qwen/Qwen3-8B", | |
| "adapter_type": "LoRA", | |
| "selected_checkpoint": "checkpoint-145", | |
| "epoch": 5.0, | |
| "global_step": 145, | |
| "validation_loss": 1.1678476333618164, | |
| "adapter_parameters": 21823488, | |
| "rank": 8, | |
| "alpha": 16, | |
| "dropout": 0.05, | |
| "learning_rate": 0.0002, | |
| "warmup_ratio": 0.03, | |
| "training_epochs": 5, | |
| "planner_version": "v4", | |
| "prior_dialogues": 1010, | |
| "script_changes": [ | |
| "Resolve adapter, prior and example paths relative to this release.", | |
| "Set dialogue_chars default to 900, matching the saved production configuration.", | |
| "Include client Resistance in the candidate inventory and prompt definitions." | |
| ], | |
| "client_candidate_tags": [ | |
| "Greeting", | |
| "SimpleAnswer", | |
| "DetailedDescription", | |
| "SelfDisclosure", | |
| "Backchannel", | |
| "Agreement", | |
| "PositiveEmotion", | |
| "Hesitation", | |
| "Resistance", | |
| "Question", | |
| "Request", | |
| "Thanking", | |
| "UnderstandingCheck", | |
| "Other" | |
| ] | |
| } | |