INDEX

Explanations

numerical scores from sports reports

oai_token-act-pair · gpt-3.5-turbo

negative scoring metrics or statistics in sports 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

 Â®

-0.81

ibur

-0.75

ulhu

-0.75

 photoc

-0.71

atel

-0.70

kson

-0.67

iewicz

-0.67

 Conflict

-0.66

izoph

-0.66

 Paradox

-0.65

POSITIVE LOGITS

Miami

1.05

game

1.03

team

0.99

season

0.99

Detroit

0.98

Minnesota

0.97

player

0.96

scoring

0.95

winning

0.95

Atlanta

0.92

Activations Density 0.124%

numerical scores from sports reports

negative scoring metrics or statistics in sports contexts

No Comments

No Known Activations