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 docs/README.md from bambamdevs/openjev-e4b: direct link, hf CLI and curl.
- Browser
- Download file 1.19 kB
-
https://huggingface.co/bambamdevs/openjev-e4b/resolve/main/docs/README.md
- Command line
-
hf download hf://bambamdevs/openjev-e4b/docs/README.md
-
curl -L -o README.md https://huggingface.co/bambamdevs/openjev-e4b/resolve/main/docs/README.md
1.19 kB
| # Documentation — OpenJEV E4B 1.0 | |
| ## Experiment 1 whitepaper | |
| **OpenJEV E4B: Training a direct-decision readout on Gemma 4 E4B. Experiment 1 method, evaluation, and results.** | |
| PDF: [`OpenJEV_E4B_Experiment_1_Whitepaper.pdf`](./OpenJEV_E4B_Experiment_1_Whitepaper.pdf) | |
| Source: [`source/OpenJEV_E4B_Experiment_1_Whitepaper.docx`](./source/OpenJEV_E4B_Experiment_1_Whitepaper.docx) and [`source/build_whitepaper.js`](./source/build_whitepaper.js). | |
| The report covers the direct-decision interface, development sequence and promotion rule, the choice-panel evaluation, five standard public tasks, limitations, release records, and reproduction steps. Gemma and OpenJEV 1.0 have complete same-row choice-panel results, including options-only controls and paired comparisons. | |
| Current results are in [`EXPERIMENT_1_RESULTS.md`](./EXPERIMENT_1_RESULTS.md). Frozen rows, the construction audit, scoring code, and row-level outputs are in [`../eval/choice_panel/`](../eval/choice_panel/). | |
| ## Experimental image and audio input | |
| [`MULTIMODAL_EXPERIMENTAL.md`](./MULTIMODAL_EXPERIMENTAL.md) describes the optional `multimodal=True` loader path, its smoke test, and its limits. | |