agu18dec
/

olens-ablations

interpretability

activation-reading

Model card Files Files and versions

olens-ablations

30.9 GB

Ctrl+K

Ctrl+K

1 contributor

History: 3 commits

agu18dec's picture

Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward

a1987eb verified 3 days ago

ar_target
OLens ablations: training text, AR target, RL starting point, reward shape, RL length 5 days ago
bullets
Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward 3 days ago
gradient_target
Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward 3 days ago
reward_shape
OLens ablations: training text, AR target, RL starting point, reward shape, RL length 5 days ago
rl_advantage_normalisation
Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward 3 days ago
rl_co_trained_ar
Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward 3 days ago
rl_length
OLens ablations: training text, AR target, RL starting point, reward shape, RL length 5 days ago
rl_starting_point
Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward 3 days ago
training_text
OLens ablations: training text, AR target, RL starting point, reward shape, RL length 5 days ago
.gitattributes

1.52 kB
initial commit 5 days ago
README.md

11 kB
Add RL variants (co-trained AR, advantage normalisation), bullet readouts, gradient target, bare-model log reward 3 days ago
olens.py

7.97 kB
OLens ablations: training text, AR target, RL starting point, reward shape, RL length 5 days ago