INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

icer

-0.73

artifacts

-0.71

rites

-0.70

 indo

-0.69

ributes

-0.69

MpServer

-0.69

agents

-0.66

resents

-0.65

oshenko

-0.63

henko

-0.63

POSITIVE LOGITS

 pred

1.24

 scramble

0.65

 æľ

0.64

0.63

 harbor

0.63

mouth

0.59

,...

0.59

okia

0.59

code

0.57

cca

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.