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

channelAvailability

-0.93

lace

-0.75

flush

-0.71

 Lent

-0.71

dq

-0.70

 ACTIONS

-0.70

 Frie

-0.70

seq

-0.68

lations

-0.66

NetMessage

-0.65

POSITIVE LOGITS

Kov

0.65

Rac

0.63

Bow

0.62

 muzzle

0.61

agh

0.60

 Pond

0.59

ocon

0.59

ivil

0.58

rav

0.58

playing

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.