INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Ful

-0.71

eson

-0.71

 MacArthur

-0.66

 Inspector

-0.65

psc

-0.65

 Dear

-0.65

 Funk

-0.64

 Edit

-0.61

 Etsy

-0.61

 Statement

-0.60

POSITIVE LOGITS

 thous

0.80

ãĤ¨ãĥ«

0.78

 adolesc

0.77

 juven

0.76

 proport

0.76

aution

0.75

pora

0.74

ngth

0.73

ãĥīãĥ©

0.73

auga

0.72

Activations Density 0.000%

No Known Activations

This feature has no known activations.