Zero-Shot Classification
Safetensors
PEFT
English
openjev
classification
decision-model
listwise
gemma4
research
Instructions to use bambamdevs/openjev-e4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use bambamdevs/openjev-e4b with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
Download eval/raw_e4b_nf4_control_summary.json from bambamdevs/openjev-e4b: direct link, hf CLI and curl.
- Browser
- Download file 1.32 kB
-
https://huggingface.co/bambamdevs/openjev-e4b/resolve/main/eval/raw_e4b_nf4_control_summary.json
- Command line
-
hf download hf://bambamdevs/openjev-e4b/eval/raw_e4b_nf4_control_summary.json
-
curl -L -o raw_e4b_nf4_control_summary.json https://huggingface.co/bambamdevs/openjev-e4b/resolve/main/eval/raw_e4b_nf4_control_summary.json
1.32 kB
| { | |
| "model": "google/gemma-4-E4B-it", | |
| "precision": "NF4", | |
| "readout": "native LM-head answer-label scoring", | |
| "complete": true, | |
| "task_count": 5, | |
| "source": "final benchmark output recorded 2026-09-26", | |
| "methodological_note": "Quantization-matched base control for the local path. End-to-end system comparison versus OpenJEV because the readout architecture differs.", | |
| "results": { | |
| "arc_easy": { | |
| "accuracy": 0.9558, | |
| "nll": 0.2218, | |
| "ece": 0.0334, | |
| "precision": "NF4", | |
| "readout": "native_lm_head_answer_label_scoring" | |
| }, | |
| "arc_challenge": { | |
| "accuracy": 0.8737, | |
| "nll": 0.6831, | |
| "ece": 0.094, | |
| "precision": "NF4", | |
| "readout": "native_lm_head_answer_label_scoring" | |
| }, | |
| "winogrande": { | |
| "accuracy": 0.6014, | |
| "nll": 1.2789, | |
| "ece": 0.3053, | |
| "precision": "NF4", | |
| "readout": "native_lm_head_answer_label_scoring" | |
| }, | |
| "hellaswag": { | |
| "accuracy": 0.7651, | |
| "nll": 1.1007, | |
| "ece": 0.1655, | |
| "precision": "NF4", | |
| "readout": "native_lm_head_answer_label_scoring" | |
| }, | |
| "mmlu": { | |
| "accuracy": 0.6619, | |
| "nll": 1.6791, | |
| "ece": 0.2418, | |
| "precision": "NF4", | |
| "readout": "native_lm_head_answer_label_scoring" | |
| } | |
| } | |
| } | |