INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

nce

-0.75

 landfall

-0.68

 crafts

-0.68

 Crest

-0.64

 Trace

-0.64

 Vinyl

-0.61

 disbel

-0.60

 Continued

-0.60

UCS

-0.59

oros

-0.59

POSITIVE LOGITS

requ

0.87

rama

0.83

hold

0.82

itated

0.78

angular

0.77

Merit

0.72

stack

0.71

zai

0.71

arist

0.69

itate

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.