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

ysis

-0.86

ocument

-0.78

INAL

-0.75

apy

-0.74

irez

-0.73

stery

-0.72

resso

-0.72

 captcha

-0.71

inea

-0.71

uyomi

-0.71

POSITIVE LOGITS

 Wraith

0.67

 Plains

0.66

ãĥ»

0.61

stown

0.60

 subtitles

0.59

 Bryant

0.59

 Sands

0.58

 caches

0.57

banks

0.57

heads

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.