INDEX

Explanations

instances of numerical data or counts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

GEBURTSDATUM

-0.90

 uſed

-0.85

 ―――――

-0.81

 propOrder

-0.80

 raiſ

-0.78

 poffible

-0.77

ंदीखरीदारी

-0.74

RenderAtEndOf

-0.73

 miſ

-0.73

 itſelf

-0.73

POSITIVE LOGITS

etc

0.95

etc

0.84

 whatever

0.74

or

0.72

whatever

0.70

 whatnot

0.69

Etc

0.69

 maybe

0.67

you

0.61

 ataupun

0.59

Activations Density 0.317%

instances of numerical data or counts

No Comments

No Known Activations