anima CLM 7B PASS-attempt (H_1141)

Continue-trained 7B ByteGPT (vocab256/d4096/36L/32H, 7.25B) from the H_1128 broad-converged 7B on the balanced 5-lang 1.5GB corpus, pushed toward coherence, then run through the FULL frozen gate battery in /7B_PASS_CONDITIONS.md.

Gate tally: G0=P,G1=P,G2=P,G3=P,val=1.1856881439685822 Visibility: PUBLIC (PUBLIC iff G0∧G1∧G2 all PASS — a_hf_autonomous).

  • G0 COHERENCE — kwr>=0.50 vs /usr/share/dict on >=4/5 English concept prompts
  • G1 RECOMBINATION — H_1137 metric (English + >=3/5 langs)
  • G2 NOVELTY — H_1140 metric (>=3 corpus-absent coherent n-grams, control=0)
  • G3 PHILOSOPHY — byte-continuation only (p1-p8), held by construction

ckpt sha256: 4de903714112c26c826a983797e5dfea0c7b3c1f19f15a34dd35822f33e245d9. Lane-G PyTorch-CUDA REFERENCE mouth (a_clm_gen_pipeline), NOT the CORE substrate. a_scale_honest_scope: single 7B rung, surface-level metrics.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for dancinlab/clm-v1-ref-pytorch-cuda-7b-pass-attempt

Finetuned
(1)
this model