INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

sworth

-0.80

ãĥ³ãĤ¸

-0.77

ŃĶ

-0.76

ĸļ

-0.71

wolf

-0.69

ãĥ´

-0.67

 freaking

-0.66

 stockp

-0.66

formed

-0.65

 loft

-0.64

POSITIVE LOGITS

ento

0.74

 âĵĺ

0.66

Characters

0.66

 Cantor

0.64

Math

0.63

IDA

0.61

 Proposition

0.61

olor

0.61

 satire

0.61

vati

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.