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

 impeachment

-0.87

ãĥĥãĤ¯

-0.77

ismo

-0.75

 Frie

-0.70

 Indonesian

-0.69

ĳå£«

-0.65

ICLE

-0.65

cade

-0.65

 arson

-0.65

 Theft

-0.64

POSITIVE LOGITS

eson

0.81

REF

0.71

culosis

0.67

ental

0.64

successfully

0.62

lass

0.61

lasses

0.61

VIS

0.61

 rehearsal

0.60

ËĪ

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.