INDEX

Explanations

references to captions and outputs related to visual content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Daniels

-0.42

/*

-0.39

 mysqli

-0.38

-0.37

<h2>

-0.37

wieś

-0.36

vec

-0.36

 radi

-0.36

ждый

-0.35

titian

-0.35

POSITIVE LOGITS

ApiOperation

3.50

 AssemblyTitle

0.92

 CreateTagHelper

0.78

 للاسماء

0.76

 EconPapers

0.73

 صوتيه

0.72

featureID

0.72

ArgumentParser

0.70

 autorytatywna

0.68

Ecotoxicity

0.68

Activations Density 0.000%

references to captions and outputs related to visual content

No Comments

No Known Activations