INDEX

Explanations

instances where declarations or statements are made

oai_token-act-pair · gpt-3.5-turbo

phrases and terms related to declarations or official statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

RH

-0.73

alez

-0.70

dayName

-0.68

anuts

-0.68

engers

-0.66

Rh

-0.65

hov

-0.63

umbn

-0.62

Reply

-0.62

Pg

-0.62

POSITIVE LOGITS

 bankruptcy

1.07

phas

0.96

 allegiance

0.94

 unequivocally

0.87

 independence

0.86

 victory

0.83

 aloud

0.80

 caliphate

0.80

ance

0.79

 martial

0.79

Activations Density 0.061%

instances where declarations or statements are made

phrases and terms related to declarations or official statements

No Comments

No Known Activations