ChrisMcCormick
/

decoderstack-d24

Model card Files Files and versions

decoderstack-d24

53.1 GB

Ctrl+K

Ctrl+K

1 contributor

History: 19 commits

ChrisMcCormick's picture

ladder-fmt05 r233: the fix-ladder's working RL recipe

5913dab verified 10 days ago

base_checkpoints
meta_005568.json: add user_config so chat_sft inherits the run's LRs 10 days ago
chatrl_checkpoints
ladder-fmt05 r233: the fix-ladder's working RL recipe 10 days ago
chatsft_checkpoints
Tool-free SFT of d24_decoderstack step 5568 (462 steps, val bpb 0.2732, ChatCORE 0.2527); model + meta + w1 optim shard 10 days ago
checkpoints
Upload folder using huggingface_hub 11 days ago
code
Add model card, meta.json, logs, tokenizer, training source, and nanochat converter 10 days ago
logs
Add model card, meta.json, logs, tokenizer, training source, and nanochat converter 10 days ago
runs
ep1-ppr32: full-epoch stock-algorithm RL — collapse by r140; rollouts + metrics 10 days ago
tokenizer
tokenizer: keep the pre-fix token_bytes so published bpb numbers stay reproducible 10 days ago
.gitattributes

1.52 kB
initial commit 11 days ago
README.md

17.1 kB
Model card: correct the recalibration — leaderboard is best-fit only, so the like-for-like gap is +0.019; both checkpoints now measured under both packings 10 days ago
convert_ckpt_to_nanochat.py

26.3 kB
Converter now emits nanochat optimizer shards for an SFT warm-start 10 days ago
test_convert_ckpt_to_nanochat.py

13.9 kB
Converter now emits nanochat optimizer shards for an SFT warm-start 10 days ago