INDEX

Explanations

references or identifiers for names in data structures or code

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

 незавершена

-0.78

_{-\

-0.75

adalupe

-0.75

 Smal

-0.73

 Paglinawan

-0.69

 فريبيس

-0.69

 ddelweddau

-0.66

GPP

-0.64

rawler

-0.64

Eloquent

-0.64

POSITIVE LOGITS

Name

1.92

 Name

1.26

 NAME

0.96

NAME

0.91

 Namely

0.79

 rempliss

0.77

 inferiore

0.75

Nombre

0.73

 serviceName

0.72

 QName

0.72

Activations Density 0.093%

references or identifiers for names in data structures or code

No Comments

No Known Activations