INDEX

Explanations

phrases related to a reason for doing something or making decisions

oai_token-act-pair · gpt-3.5-turbo

phrases that express purpose or justification

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Lyme

-0.66

 intimidating

-0.63

 hordes

-0.63

 landscapes

-0.63

NetMessage

-0.63

 Feld

-0.62

bor

-0.62

 campuses

-0.62

 autoimmune

-0.61

 densely

-0.60

POSITIVE LOGITS

atsu

0.94

SourceFile

0.88

urities

0.87

peria

0.86

ames

0.85

auga

0.81

rifice

0.80

cia

0.79

uin

0.78

ilial

0.77

Activations Density 0.007%

phrases related to a reason for doing something or making decisions

phrases that express purpose or justification

No Comments

No Known Activations