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

irted

-0.68

uctor

-0.66

ileged

-0.65

agi

-0.65

agogue

-0.65

rious

-0.64

atted

-0.64

 Presents

-0.64

icultural

-0.63

inatory

-0.63

POSITIVE LOGITS

emet

0.66

 Dynam

0.63

slow

0.61

Xer

0.61

sword

0.60

dies

0.59

Nusra

0.58

 Hannibal

0.58

heid

0.58

 Edwin

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.