TheNexus42
/

bloom-1b7-head-surgery

Text Generation

attention-heads

surgical-reinitialization

Model card Files Files and versions

bloom-1b7-head-surgery / evaluation

650 kB

Ctrl+K

Ctrl+K

1 contributor

History: 10 commits

TheNexus42's picture

Add C4 validation PPL evaluation (closes distribution shift gap)

530be61 verified 6 months ago

architecture_map.json

151 kB
Upload evaluation/architecture_map.json with huggingface_hub 6 months ago
c4_baseline_trajectory.json

14.4 kB
Upload evaluation/c4_baseline_trajectory.json with huggingface_hub 6 months ago
c4_validation_ppl.json

25.5 kB
Add C4 validation PPL evaluation (closes distribution shift gap) 6 months ago
completion_evaluation.json

107 kB
Upload evaluation/completion_evaluation.json with huggingface_hub 6 months ago
evaluation_3way.json

175 kB
Upload evaluation/evaluation_3way.json with huggingface_hub 6 months ago
frozen_drift_analysis.json

18.4 kB
Upload evaluation/frozen_drift_analysis.json with huggingface_hub 6 months ago
h5_trajectory.json

8.44 kB
Upload evaluation/h5_trajectory.json with huggingface_hub 6 months ago
pass1_evaluation.json

117 kB
Upload evaluation/pass1_evaluation.json with huggingface_hub 6 months ago
pass1_trajectory.json

14.6 kB
Upload evaluation/pass1_trajectory.json with huggingface_hub 6 months ago
pass2_trajectory.json

18.5 kB
Upload evaluation/pass2_trajectory.json with huggingface_hub 6 months ago