INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.2.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ĸļ

-0.76

*:

-0.65

izen

-0.65

uded

-0.65

Í

-0.63

ardless

-0.62

 cooler

-0.61

 Apprentice

-0.60

ija

-0.59

 Editors

-0.59

POSITIVE LOGITS

essage

0.84

ammy

0.77

luster

0.72

awaru

0.71

urdue

0.71

kHz

0.66

lay

0.65

oland

0.65

AFL

0.65

aml

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.