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

º

-0.84

ĳå£«

-0.82

ãĥĥãĥī

-0.79

ãģĤ

-0.79

Ģ

-0.79

ortment

-0.74

iesel

-0.71

-0.69

arette

-0.68

romy

-0.68

POSITIVE LOGITS

MpServer

0.88

formed

0.66

hin

0.65

 Emer

0.64

ng

0.61

 takeover

0.61

folk

0.61

rawdownloadcloneembedreportprint

0.60

cyn

0.59

 alphabet

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.