INDEX

Explanations

phrases indicating monitoring or observation

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Grounds

-0.34

òa

-0.34

ldorf

-0.33

########.

-0.33

 lenker

-0.31

 geschlossen

-0.31

 Estudio

-0.31

 equivalently

-0.30

Completo

-0.30

 bedingt

-0.30

POSITIVE LOGITS

AndEndTag

0.86

 monitor

0.76

monitoring

0.74

monitor

0.73

 monitoring

0.71

 MainAxisSize

0.71

 monitored

0.71

watch

0.70

 Monitoring

0.69

Monitoring

0.69

Activations Density 0.016%

phrases indicating monitoring or observation

No Comments

No Known Activations