INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 latitude

-0.73

lod

-0.72

ricular

-0.72

astical

-0.63

ional

-0.62

 disob

-0.62

":[{"

-0.61

hift

-0.61

 resc

-0.60

erential

-0.60

POSITIVE LOGITS

ubis

0.78

lang

0.74

neum

0.73

avery

0.72

 mins

0.68

umbn

0.66

amation

0.65

 PAGE

0.65

bia

0.65

AGE

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.