dimitarpg13
/

semsimula-fock-parflm-depthcond-vtheta-openwebtext

Model card Files Files and versions

semsimula-fock-parflm-depthcond-vtheta-openwebtext

2.57 GB

Ctrl+K

Ctrl+K

1 contributor

History: 22 commits

dimitarpg13's picture

Mark as superseded by the anisotropic Gaussian V_theta + Fock-reg OpenWebText line

2ce4a2b verified 6 days ago

checkpoints
Add checkpoint fock_dcvt_owt_xi5long_topk16_dt32da16_mh4_dcvt5x8_ob_untied_wsd_e5c_plgate_rep0.05_step99000_best.pt about 1 month ago
code
Add model card, self-contained model code, notebooks, val/logfreq data, training logs, config about 1 month ago
data
Add model card, self-contained model code, notebooks, val/logfreq data, training logs, config about 1 month ago
geodesic
Add geodesic preservation experiment: PPL vs damped geodesic residual overlay, per-layer heatmap, results data about 1 month ago
notebooks
Add model card, self-contained model code, notebooks, val/logfreq data, training logs, config about 1 month ago
results
Add model card, self-contained model code, notebooks, val/logfreq data, training logs, config about 1 month ago
.gitattributes

198 Bytes
Add geodesic preservation experiment: PPL vs damped geodesic residual overlay, per-layer heatmap, results data about 1 month ago
README.md

38.1 kB
Mark as superseded by the anisotropic Gaussian V_theta + Fock-reg OpenWebText line 6 days ago
config.json

3.77 kB
Add model card, self-contained model code, notebooks, val/logfreq data, training logs, config about 1 month ago