INDEX

Explanations

phrases indicating membership or belonging

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

IUrlHelper

-0.52

AsUp

-0.47

참고

-0.44

یکن

-0.42

ajn

-0.40

 gyhoeddwyd

-0.40

Hein

-0.40

 تضيفلها

-0.39

ież

-0.39

contrast

-0.39

POSITIVE LOGITS

 belong

1.13

belong

1.12

 belongs

1.09

 belonged

1.01

belongs

1.01

belongsTo

0.97

 Belong

0.96

 belonging

0.94

 pertenece

0.91

 perten

0.90

Activations Density 0.365%

phrases indicating membership or belonging

No Comments

No Known Activations