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

Republic

-0.74

!/

-0.72

 Frag

-0.68

Frag

-0.66

Micro

-0.65

Ly

-0.64

 Canaveral

-0.63

ndra

-0.62

zyme

-0.62

Gy

-0.62

POSITIVE LOGITS

esthes

0.87

 stalk

0.75

adata

0.68

ptin

0.68

sheets

0.67

stood

0.66

arij

0.65

 pillow

0.63

 quartz

0.60

kies

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.