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. 17:05

   tinynext-v2-nofim: 12.5M params, 28 min on GPU

   cfahlgren1 HF Staff
   Verified signature

   e5dde63
3. 16:30

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

   cfahlgren1 HF Staff
   Verified signature

   0de7556
4. 16:27

   card: add v2 (word-piece transformer) with eval results, browser numbers and training details

   cfahlgren1 HF Staff
   Verified signature

   c97f9cc
5. 16:26

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

   cfahlgren1 HF Staff
   Verified signature

   33bfb6d
6. 16:10

   no-FIM models are evaluated with prefix-only prompts

   cfahlgren1 HF Staff
   Verified signature

   793c5c9
7. 15:40

   tinynext-v2: 12.5M params, 28 min on GPU

   cfahlgren1 HF Staff
   Verified signature

   1cffd47
8. 14:43

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

   cfahlgren1 HF Staff
   Verified signature

   63b64fd
9. 14:14

   parallel data prep, data on GPU, trackio, safetensors

   cfahlgren1 HF Staff
   Verified signature

   9bba47b
10. 14:06

    fix step count

    cfahlgren1 HF Staff
    Verified signature

    88610cb
11. 13:47

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

    cfahlgren1 HF Staff
    Verified signature

    3f86878
12. 12:29

    shorter badges so they fit on one line

    cfahlgren1 HF Staff
    Verified signature

    b89a598
13. 12:28

    drop the demo screenshot from the card

    cfahlgren1 HF Staff
    Verified signature

    12dcc4d
14. 12:28

    show the stat badges in one row

    cfahlgren1 HF Staff
    Verified signature

    bca9766
15. 12:15

    tinynext card with banner

    cfahlgren1 HF Staff
    Verified signature

    729bd87
16. 12:13

    Duplicate from cfahlgren1/enron-autocomplete

    cfahlgren1 HF Staff
    a028f12