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

ESCO

-0.71

pron

-0.68

oby

-0.67

IDS

-0.66

OPS

-0.66

onga

-0.64

iddler

-0.64

TRUMP

-0.62

employ

-0.62

 Outs

-0.60

POSITIVE LOGITS

ghai

0.69

 reproduce

0.66

 stabilize

0.66

 gravity

0.65

 concentrated

0.65

 Skydragon

0.64

 partName

0.63

chwitz

0.63

 detectors

0.63

 explode

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.