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

DH

-0.72

interrupted

-0.70

EngineDebug

-0.69

 Nationwide

-0.66

puter

-0.64

cgi

-0.62

gravity

-0.61

BAT

-0.61

Cum

-0.60

IB

-0.59

POSITIVE LOGITS

ibrary

0.88

utics

0.74

inness

0.74

othal

0.73

inda

0.73

ignt

0.72

zos

0.72

clus

0.71

acha

0.71

ithe

0.71

Activations Density 0.000%

No Known Activations

This feature has no known activations.