Views
No views yet
cminst/realistic-bpe5-union-longest-specialist-500m-10b.
Source run directory name: R5FW10B-SpecNA-realistic_bpe5_union_longest-500M-10B-local.cminst/realistic-bpe5-union-longest-10b.model.pt: final model weightsconfig.json: training/model configuration saved by train_v3.pyresults.json: final evaluation metrics and logsstep_*/: intermediate model-only checkpoints (model.pt, config.json, results.json, and string_atlas.json when present)tokenizers/*.json: tokenizer JSON filesrealistic_bpe5_union_longestrealistic_bpe5_union_longest: BPB 0.8940 at step 95421{
2 "anchor": "realistic_bpe5_union_longest",
3 "anchor_weight": 1.0,
4 "completed_steps": 9542,
5 "ctx_len": 2048,
6 "d_model": 1536,
7 "data_format": "packed_tokens",
8 "input_mode": "full",
9 "lr_decay_steps": 9542,
10 "model_size": "500M",
11 "n_heads": 24,
12 "n_layers": 18,
13 "output_mode": "full",
14 "run_name": "R5FW10B-SpecNA-realistic_bpe5_union_longest-500M-10B-local",
15 "tokens_processed": 10000626688,
16 "vocab_sizes": {
17 "realistic_bpe5_union_longest": 50184
18 }
19}