INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

mite

-0.91

orate

-0.79

pend

-0.77

obo

-0.77

 Adin

-0.71

Quote

-0.69

hov

-0.68

GM

-0.65

atis

-0.64

GMT

-0.63

POSITIVE LOGITS

pless

0.76

 Lanka

0.71

 dread

0.67

fel

0.67

 Blacks

0.66

ãĥ¼ãĤ¯

0.65

 occupations

0.64

Il

0.63

eful

0.63

ãĥ©ãĥ³

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.