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

#$

-0.70

Rae

-0.69

ornia

-0.65

Ja

-0.63

Â¨

-0.62

ÃŃa

-0.60

rought

-0.60

 halves

-0.60

 Rafael

-0.60

aid

-0.60

POSITIVE LOGITS

metadata

0.77

spect

0.66

erness

0.63

phabet

0.63

 STOR

0.62

wagon

0.62

shown

0.60

abouts

0.59

ARCH

0.59

 PATH

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.