Text Classification
PEFT
Safetensors
English
decision-model
calibration
lora
multiple-choice
typesafe
qwen3.5
Eval Results (legacy)
Instructions to use jaredpalmer/kev-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use jaredpalmer/kev-4b with PEFT:
Task type is invalid.
- Notebooks
- Google Colab
- Kaggle
kev-4b preview, decision-v7 recipe (transfer 0.790 dev / 0.806 locked; held-out pairs 0.73)
2bd3bb9 verified Download training_metrics.json from jaredpalmer/kev-4b: direct link, hf CLI and curl.
- Browser
- Download file 324 Bytes
-
https://huggingface.co/jaredpalmer/kev-4b/resolve/2bd3bb9a9957aff9a4803be7ce91a521cdff0a31/training_metrics.json
- Command line
-
hf download hf://jaredpalmer/kev-4b@2bd3bb9a9957aff9a4803be7ce91a521cdff0a31/training_metrics.json
-
curl -L -o training_metrics.json https://huggingface.co/jaredpalmer/kev-4b/resolve/2bd3bb9a9957aff9a4803be7ce91a521cdff0a31/training_metrics.json
324 Bytes
| { | |
| "wall_seconds": 2378.645946741104, | |
| "records_seen": 30530, | |
| "requested_records": 25152, | |
| "truncated_records": 0, | |
| "rejected_records": 0, | |
| "optimizer_steps": 3144, | |
| "forward_tokens": 5860024, | |
| "peak_device_bytes": 23007304704, | |
| "device": "cuda", | |
| "dtype": "bf16", | |
| "batch": 4, | |
| "peak_rss_bytes": 25029832704 | |
| } | |