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

 astroph

-0.66

itionally

-0.63

 Plot

-0.63

 Inquis

-0.62

 Scots

-0.61

ikarp

-0.60

 crude

-0.60

 LGBTQ

-0.59

Osw

-0.59

 Autob

-0.59

POSITIVE LOGITS

wo

0.75

inator

0.72

wei

0.70

bite

0.69

CAP

0.68

ãĥīãĥ©ãĤ´ãĥ³

0.68

onz

0.68

=-=-

0.68

onson

0.67

ragon

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.