Instructions to use XinyuGuan/CICL with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use XinyuGuan/CICL with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-9B") model = PeftModel.from_pretrained(base_model, "XinyuGuan/CICL") - Notebooks
- Google Colab
- Kaggle
| { | |
| "adapter": "training/outputs/opus_v1_qwen35_9b", | |
| "val": "training/data/opus_v1/val.jsonl", | |
| "n_examples": 144, | |
| "n_parsed": 144, | |
| "parse_rate": 1.0, | |
| "n_parse_failures": 0, | |
| "elapsed_sec": 461.1261742115021, | |
| "field_stats": { | |
| "n_parsed": 144, | |
| "action_shift": { | |
| "mae": 0.02395833283662796, | |
| "p50": 0.0, | |
| "p90": 0.10000000149011612, | |
| "max": 0.6000000238418579, | |
| "n": 144 | |
| }, | |
| "necessity": { | |
| "mae": 0.009722222574055195, | |
| "p50": 0.0, | |
| "p90": 0.0, | |
| "max": 0.20000000298023224, | |
| "n": 144 | |
| }, | |
| "expected_outcome_uplift": { | |
| "mae": 0.03263889253139496, | |
| "p50": 0.0, | |
| "p90": 0.10000000149011612, | |
| "max": 0.25, | |
| "n": 144 | |
| }, | |
| "negative_transfer_risk": { | |
| "mae": 0.06388888508081436, | |
| "p50": 0.0, | |
| "p90": 0.3699997067451477, | |
| "max": 0.5, | |
| "n": 144 | |
| }, | |
| "confidence": { | |
| "mae": 0.01875000074505806, | |
| "p50": 0.0, | |
| "p90": 0.05000000074505806, | |
| "max": 0.15000000596046448, | |
| "n": 144 | |
| }, | |
| "no_context_action": { | |
| "exact_match": 0.8958333333333334, | |
| "n": 144 | |
| }, | |
| "with_context_action": { | |
| "exact_match": 0.9097222222222222, | |
| "n": 144 | |
| }, | |
| "reason": { | |
| "nonempty_rate": 1.0, | |
| "pred_char_mean": 150.67361450195312, | |
| "gold_char_mean": 153.86805725097656, | |
| "len_ratio_mean": 0.9893233180046082, | |
| "n": 144 | |
| } | |
| }, | |
| "parse_failures_head": [] | |
| } |