INDEX

Explanations

references to specific research papers or academic citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 lenker

-0.57

tanooga

-0.56

orkin

-0.52

thunk

-0.52

Chimp

-0.50

dymyr

-0.48

ungi

-0.48

TIMORE

-0.47

ratic

-0.47

Vau

-0.47

POSITIVE LOGITS

 japon

0.80

 للمعارف

0.78

脚注の使い方

0.78

 Japão

0.78

 Japón

0.74

 Japan

0.74

 Japon

0.74

Japan

0.73

 japan

0.72

 Giappone

0.71

Activations Density 0.508%

references to specific research papers or academic citations

No Comments

No Known Activations