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

 Petersen

-0.65

 Emerson

-0.65

Runner

-0.64

rition

-0.62

 stumbling

-0.62

ihil

-0.62

 Slater

-0.62

 Canberra

-0.61

 stomp

-0.61

apego

-0.59

POSITIVE LOGITS

Adds

0.78

tains

0.73

CVE

0.73

alde

0.72

 agre

0.71

codes

0.67

 pleas

0.67

urate

0.66

kin

0.66

rez

0.66

Activations Density 0.000%

No Known Activations

This feature has no known activations.