INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

books

-0.68

 favorites

-0.68

 Sheen

-0.65

 relaxation

-0.64

 nods

-0.64

 affirmative

-0.62

vity

-0.61

pace

-0.61

 meanings

-0.60

 hereafter

-0.60

POSITIVE LOGITS

raq

0.84

aced

0.76

rend

0.68

iji

0.67

*=-

0.66

RP

0.66

ussie

0.63

Gh

0.63

inosaur

0.62

 Assy

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.