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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

byn

-0.87

ocard

-0.79

ategory

-0.74

ogene

-0.70

eanor

-0.66

folios

-0.65

gart

-0.65

rosis

-0.65

agging

-0.64

anchez

-0.64

POSITIVE LOGITS

bleacher

0.65

 NIGHT

0.63

 Dance

0.63

 Raiders

0.62

Cast

0.62

 Sorce

0.61

nah

0.60

bread

0.59

Hell

0.59

inis

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.