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

tumblr

-0.68

 aggrav

-0.62

 tresp

-0.61

 monitoring

-0.61

 tether

-0.60

 ambul

-0.60

 thrott

-0.60

 bolst

-0.59

 administering

-0.59

 ingested

-0.59

POSITIVE LOGITS

riz

0.89

mented

0.86

ria

0.82

ments

0.81

rants

0.78

opoly

0.76

inelli

0.75

folios

0.74

enment

0.73

eenth

0.72

Activations Density 0.000%

No Known Activations

This feature has no known activations.