Views
No views yet
nanochat implementation. It is not
yet converted to the Hugging Face Transformers from_pretrained format.tr_d20_bpe_32768_chinchilla2017100bpe_32768/ari/users/nunal/nanochat-turk-d20-bpe32k/ari/users/nunal/nanochat-turk-d20-bpe32k/base_checkpoints/tr_d20_bpe_32768_chinchilla20/ari/users/nunal/nanochat-turk-d20-bpe32k/tokenizers/bpe_32768492421unknown1{
2 "n_embd": 1280,
3 "n_head": 10,
4 "n_kv_head": 10,
5 "n_layer": 20,
6 "sequence_len": 2048,
7 "vocab_size": 32768,
8 "window_pattern": "L"
9}2032768204841048576Lcheckpoint/model_017100.ptcheckpoint/meta_017100.jsoncheckpoint/optim_017100_rank*.pt if optimizer shards were uploadedtokenizer/tokenizer.pkltokenizer/tokenizer_config.jsontokenizer/token_bytes.ptreport/ and logs/ when availablecetvel_out/ when CETVEL has completedprovenance/upload_manifest.jsonunknownunknownFalse2026-06-08T11:46:49.296475+00:00