Text Classification
PEFT
Safetensors
English
decision-model
calibration
lora
multiple-choice
typesafe
qwen3.5
Eval Results (legacy)
Instructions to use jaredpalmer/kev-0.8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use jaredpalmer/kev-0.8b with PEFT:
from peft import PeftModel from transformers import AutoModel base_model = AutoModel.from_pretrained("Qwen/Qwen3.5-0.8B-Base") model = PeftModel.from_pretrained(base_model, "jaredpalmer/kev-0.8b") - Notebooks
- Google Colab
- Kaggle
Kev-0.8B: dates+unknowable delta (locked test 0.834 / 0.684); previous weights at tag v7-base
2256796 verified Download training_config.json from jaredpalmer/kev-0.8b: direct link, hf CLI and curl.
- Browser
- Download file 1.63 kB
-
https://huggingface.co/jaredpalmer/kev-0.8b/resolve/54f4f8777356cd5bbbb6c6919c657f26e6f2f6d8/training_config.json
- Command line
-
hf download hf://jaredpalmer/kev-0.8b@54f4f8777356cd5bbbb6c6919c657f26e6f2f6d8/training_config.json
-
curl -L -o training_config.json https://huggingface.co/jaredpalmer/kev-0.8b/resolve/54f4f8777356cd5bbbb6c6919c657f26e6f2f6d8/training_config.json
1.63 kB
| { | |
| "args": { | |
| "base": "Qwen/Qwen3.5-0.8B-Base", | |
| "n_per_source": 1000, | |
| "epochs": 1, | |
| "lr": 4e-05, | |
| "head_lr": 0.0, | |
| "weight_decay": 0.01, | |
| "lora": 16, | |
| "accum": 1, | |
| "holdout": "", | |
| "perm_kl": 0.0, | |
| "perm_frac": 0.3, | |
| "ord_w": 0.0, | |
| "suite": "/root/evals/v7/decision-v7", | |
| "train_sources": "", | |
| "device": "cuda", | |
| "batch": 8, | |
| "dtype": "bf16", | |
| "weights_dtype": "fp32", | |
| "checkpointing": 0, | |
| "option_isolation": 0, | |
| "special_embeddings": 0, | |
| "head_dim": 256, | |
| "lora_targets": "all", | |
| "base_revision": "dc7cdfe2ee4154fa7e30f5b51ca41bfa40174e68", | |
| "p_none": 0.1, | |
| "p_none_distract": 0.12, | |
| "p_distract": 0.15, | |
| "p_none_pair": 0.25, | |
| "synthetic_repeat": 1, | |
| "public_frac": 1.0, | |
| "anchor": "", | |
| "anchor_w": 0.0, | |
| "anchor_sources": "", | |
| "out": "/runs/night2-08b-du2/00-trial-0/checkpoint", | |
| "data": "evals/night2/dates_unknowable.jsonl", | |
| "replay": 2000, | |
| "init_from": "jaredpalmer/kev-0.8b", | |
| "seed": 1 | |
| }, | |
| "suite_sha256": "a8f50e481b7d90b97da049e0ff6a01cee2f1ed204aed61a8265af0edbb5514d2", | |
| "base_revision": "dc7cdfe2ee4154fa7e30f5b51ca41bfa40174e68", | |
| "init_source": { | |
| "init_from": "jaredpalmer/kev-0.8b", | |
| "resolved": "/__modal/volumes/vo-kEMu8BkBAIrorAQI6V8f2D/hub/models--jaredpalmer--kev-0.8b/snapshots/c917edefdfd72b3e9ba71455584700acc70595f6", | |
| "adapter_sha256": "d9fa619fd3b0490122454c386bfa1b53c23850189d1b5c4346962162e2e39a64", | |
| "head_sha256": "8610dac1c30a64bbcea7715f20a258f4d084d7b862c10f96a1625c732049a32a" | |
| }, | |
| "ordinal_objective": "ranked_probability_score", | |
| "holdout": [] | |
| } | |