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

Chance

-0.79

PI

-0.71

 isEnabled

-0.70

pared

-0.68

cffff

-0.68

Ground

-0.68

Scroll

-0.67

Condition

-0.66

hma

-0.66

olars

-0.66

POSITIVE LOGITS

enko

0.77

roma

0.73

 Vaughan

0.70

atile

0.67

jen

0.66

omic

0.66

otos

0.66

owe

0.63

opers

0.63

 Stevenson

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.