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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 aggro

-0.66

 mercy

-0.65

BP

-0.65

 surfaces

-0.63

 physic

-0.62

 closure

-0.61

 sinners

-0.60

 geometry

-0.60

 whiff

-0.60

 severity

-0.59

POSITIVE LOGITS

retch

0.69

nant

0.68

ouk

0.68

oland

0.68

lyak

0.68

ovich

0.66

cong

0.64

 monop

0.64

oÄŁ

0.64

 Randolph

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.