INDEX

Explanations

sports-related terms related to competitions or matches

oai_token-act-pair · gpt-3.5-turbo

terms related to competitive matchups and events

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

mental

-0.78

employment

-0.72

growth

-0.68

serving

-0.65

vest

-0.65

ako

-0.64

za

-0.63

redo

-0.62

 Shelter

-0.62

ascript

-0.62

POSITIVE LOGITS

 pitted

0.94

 against

0.90

 matchup

0.89

 between

0.87

 matchups

0.85

roy

0.84

 Royale

0.82

 rivals

0.81

 duel

0.79

iques

0.77

Activations Density 0.141%

sports-related terms related to competitions or matches

terms related to competitive matchups and events

No Comments

No Known Activations