INDEX

Explanations

expressions of belief and conviction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rouen

-0.59

Ome

-0.58

 Kassel

-0.57

 Gard

-0.56

 Gardiner

-0.54

Riverside

-0.53

terdam

-0.53

 Marius

-0.53

 inexplicable

-0.52

 porto

-0.52

POSITIVE LOGITS

 belief

0.98

 believe

0.97

 Believe

0.93

 BELIEVE

0.93

 Belief

0.92

Believe

0.87

believe

0.85

 believes

0.84

 Belie

0.82

 believed

0.82

Activations Density 0.067%

expressions of belief and conviction

No Comments

No Known Activations