INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 Adin

-0.70

 Leban

-0.68

 HELP

-0.67

thren

-0.66

comfort

-0.65

 remorse

-0.64

 Takeru

-0.64

phrine

-0.64

 Virtue

-0.64

 Watts

-0.63

POSITIVE LOGITS

 until

0.89

ayn

0.74

wo

0.72

ptr

0.72

emort

0.67

ucci

0.67

izoph

0.65

 till

0.65

 before

0.64

conserv

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.