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

 Minotaur

-0.73

 imaginable

-0.67

onis

-0.67

 Pikachu

-0.66

 Golem

-0.65

 poisoning

-0.63

llo

-0.61

 Typh

-0.61

ea

-0.60

 offensively

-0.59

POSITIVE LOGITS

profits

0.72

cedes

0.69

lene

0.69

 actionGroup

0.69

 Colors

0.65

utical

0.65

atures

0.65

arks

0.64

amac

0.63

terness

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.