INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 practicable

-0.71

endment

-0.69

cffffcc

-0.68

JO

-0.67

chie

-0.66

PLA

-0.66

lass

-0.63

 Hughes

-0.62

Fair

-0.62

.>>

-0.61

POSITIVE LOGITS

 Pascal

0.81

icht

0.75

ipe

0.71

ioch

0.70

wine

0.66

getic

0.66

 souls

0.65

 Amar

0.65

ternity

0.65

inger

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.