INDEX

Explanations

instances of interceptions in a sports context

oai_token-act-pair · gpt-3.5-turbo

terms related to turnovers in football, specifically interceptions and fumbles

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

hov

-0.77

eways

-0.76

atten

-0.72

oras

-0.72

este

-0.71

ood

-0.71

illac

-0.69

learn

-0.68

travel

-0.67

mates

-0.66

POSITIVE LOGITS

 touchdowns

0.92

 interceptions

0.89

 touchdown

0.88

 catcher

0.87

 fumble

0.78

 interception

0.75

 cornerback

0.74

 kicker

0.72

ception

0.72

 rusher

0.71

Activations Density 0.017%

instances of interceptions in a sports context

terms related to turnovers in football, specifically interceptions and fumbles

No Comments

No Known Activations