INDEX

Explanations

references to legal or procedural contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

providedIn

-0.68

endregion

-0.59

inflater

-0.57

 rgba

-0.52

氓

-0.47

躇

-0.46

出版年

-0.45

ioutil

-0.45

꼽

-0.44

ში

-0.44

POSITIVE LOGITS

on

1.10

 behalf

0.95

RenderAtEndOf

0.94

 وعلى

0.94

On

0.93

On

0.89

ON

0.85

 vpon

0.85

 onCreateView

0.84

TemporalType

0.83

Activations Density 0.760%

references to legal or procedural contexts

No Comments

No Known Activations