Nov 13 Today

1. 17:10

   export the compressed browser format from train.py instead of int8 activations

   cfahlgren1 HF Staff
   Verified signature

   4dece7f
2. 16:30

   fetch the shared eval from v2/ even when uploading to an experiment folder

   cfahlgren1 HF Staff
   Verified signature

   0de7556
3. 16:26

   v2: 13 MB browser build (int8 download, float compute), export/compress scripts, onnx evals

   cfahlgren1 HF Staff
   Verified signature

   33bfb6d
4. 16:10

   no-FIM models are evaluated with prefix-only prompts

   cfahlgren1 HF Staff
   Verified signature

   793c5c9
5. 14:43

   prepare data on a CPU job, train from the prepared files

   cfahlgren1 HF Staff
   Verified signature

   63b64fd
6. 14:14

   parallel data prep, data on GPU, trackio, safetensors

   cfahlgren1 HF Staff
   Verified signature

   9bba47b
7. 14:06

   fix step count

   cfahlgren1 HF Staff
   Verified signature

   88610cb
8. 13:47

   v2: training script, eval harness, eval set and baseline metrics

   cfahlgren1 HF Staff
   Verified signature

   3f86878