INDEX

Explanations

numeric sequences or patterns 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 @ 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

thâu

-0.65

kiwa

-0.60

diyesi

-0.59

 שוליים

-0.56

 tech

-0.56

fP

-0.56

RenderAtEndOf

-0.55

ValueStyle

-0.55

WriteAttribute

-0.55

FAD

-0.55

POSITIVE LOGITS

cc

0.85

cc

0.79

xx

0.76

CC

0.74

ZZ

0.74

dd

0.72

BB

0.71

bb

0.71

ZZ

0.70

xxx

0.70

Activations Density 0.653%

numeric sequences or patterns in the text

No Comments

No Known Activations