INDEX

Explanations

attendance figures or mentions

oai_token-act-pair · gpt-3.5-turbo

references to attendance in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enthal

-0.75

enegger

-0.70

geon

-0.70

pex

-0.69

grave

-0.68

thodox

-0.67

imov

-0.66

imb

-0.65

 centr

-0.61

itars

-0.61

POSITIVE LOGITS

 attendance

1.00

 Attend

0.98

 attendees

0.83

enance

0.81

oided

0.78

 attends

0.73

 attend

0.71

 tickets

0.68

 meetings

0.66

goers

0.66

Activations Density 0.014%

attendance figures or mentions

references to attendance in various contexts

No Comments

No Known Activations