INDEX

Explanations

emphasized or specific qualifiers in the text

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 addirittura

-0.68

 persino

-0.63

 jopa

-0.61

 dokonce

-0.60

 Simplemente

-0.60

 simplemente

-0.57

viewDidLoad

-0.57

 lediglich

-0.57

 prostu

-0.56

脚注の使い方

-0.56

POSITIVE LOGITS

 those

1.09

those

0.86

 Those

0.79

Those

0.78

 since

0.78

 concerning

0.77

 noteworthy

0.77

 THOSE

0.75

 considering

0.75

if

0.71

Activations Density 0.111%

emphasized or specific qualifiers in the text

No Comments

No Known Activations