INDEX

Explanations

the word "commute" and related terms

oai_token-act-pair · gpt-3.5-turbo

references to commuting and travel-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

enic

-0.84

sonian

-0.83

ificial

-0.72

iasis

-0.71

reditary

-0.71

cius

-0.71

grave

-0.71

gren

-0.69

vati

-0.66

argon

-0.66

POSITIVE LOGITS

 commute

1.22

 commuting

1.05

 commuters

0.94

 commuter

0.87

 BART

0.75

 arrang

0.72

ï¸

0.68

 shenan

0.68

uters

0.67

 hubs

0.66

Activations Density 0.016%

the word "commute" and related terms

references to commuting and travel-related activities

No Comments

No Known Activations