INDEX

Explanations

references to chapters or section headings in a document

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

MessageTagHelper

-0.69

komst

-0.65

醐

-0.65

brecht

-0.60

StartPosition

-0.60

/*

-0.58

 Mühle

-0.58

atori

-0.58

américa

-0.57

interopRequire

-0.57

POSITIVE LOGITS

featureID

0.68

Климат

0.63

;#

0.63

 Chom

0.57

CH

0.57

CHP

0.57

Ch

0.56

endwhile

0.55

 Référence

0.54

flashdata

0.54

Activations Density 0.376%

references to chapters or section headings in a document

No Comments

No Known Activations