INDEX

Explanations

statistical data and numerical comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

as

-0.63

 SDLK

-0.57

 like

-0.54

 jako

-0.54

 Bolivar

-0.53

saraba

-0.52

 تضيفلها

-0.52

wie

-0.50

 Głów

-0.49

setViewportView

-0.49

POSITIVE LOGITS

RectangleBorder

0.54

pexpr

0.54

bedo

0.54

 Gleaner

0.53

<bos>

0.51

 CURIAM

0.51

厘

0.51

Cyfeiriadau

0.50

!*\

0.50

 transfieras

0.47

Activations Density 0.248%

statistical data and numerical comparisons

No Comments

No Known Activations