INDEX

Explanations

terms associated with relationships and opposing concepts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_65k/average_l0_73

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

apunov

-0.60

esModule

-0.60

 समीक्षाएं

-0.60

 kaynağından

-0.59

 Chwiliwch

-0.56

 surla

-0.56

ISupport

-0.56

OptionsMenu

-0.54

 firm

-0.54

bers

-0.54

POSITIVE LOGITS

 ſever

0.77

 ſmall

0.71

 Parag

0.71

 Perſ

0.66

 ſeveral

0.65

 parag

0.65

 nimic

0.63

 greateſt

0.63

ösungen

0.60

 againſt

0.59

Activations Density 0.065%

terms associated with relationships and opposing concepts

No Known Activations