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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 hosted

-0.67

 shared

-0.66

 collaborators

-0.63

 kits

-0.63

WAS

-0.62

mails

-0.62

Wid

-0.61

HAM

-0.61

 sorts

-0.61

 married

-0.61

POSITIVE LOGITS

Pwr

0.87

 reflection

0.73

wcs

0.72

operator

0.72

chief

0.71

pillar

0.69

PDATE

0.69

riad

0.69

Leader

0.68

Redd

0.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.