INDEX

Explanations

instances of the word "coalition."

oai_token-act-pair · gpt-3.5-turbo

references to coalitions or alliances

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

lycer

-0.72

otle

-0.67

hibited

-0.67

nit

-0.65

rors

-0.65

chester

-0.64

ÃĽ

-0.63

things

-0.62

hin

-0.62

Tesla

-0.62

POSITIVE LOGITS

 Provision

1.04

 partners

1.04

 partner

0.92

 Aven

0.90

 alliance

0.88

 coalition

0.87

naire

0.86

 alliances

0.85

 member

0.82

 effort

0.81

Activations Density 0.066%

instances of the word "coalition."

references to coalitions or alliances

No Known Activations