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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

workshop

-0.69

 Carnage

-0.67

=]

-0.65

played

-0.65

teenth

-0.64

fort

-0.63

=~

-0.61

hedon

-0.60

 Remem

-0.59

 Oath

-0.58

POSITIVE LOGITS

renheit

0.74

NM

0.72

bler

0.70

OWS

0.69

Nut

0.67

 Muhammad

0.64

heid

0.63

ogical

0.62

OAD

0.61

 nutritional

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.