INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

itals

-0.70

boro

-0.67

 Stra

-0.65

wp

-0.65

hood

-0.62

wordpress

-0.62

leans

-0.60

ught

-0.60

uben

-0.59

ilers

-0.58

POSITIVE LOGITS

BACK

0.77

LOAD

0.75

ADRA

0.75

Offline

0.74

Parent

0.72

ivated

0.67

iesel

0.67

aceous

0.67

WIND

0.67

OHN

0.65

Activations Density 0.000%

No Known Activations

This feature has no known activations.