INDEX

Explanations

mentions of "Foreign" entities or activities

oai_token-act-pair · gpt-3.5-turbo

references to foreign affairs and international relations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

displayText

-0.85

CRIP

-0.83

DIT

-0.73

ÅŁ

-0.72

heet

-0.71

illation

-0.67

odcast

-0.67

FUL

-0.66

 DRAG

-0.66

ript

-0.64

POSITIVE LOGITS

 Affairs

0.92

policy

0.92

 Foreign

0.92

 nationals

0.90

Foreign

0.85

 Fighters

0.85

 affairs

0.81

 Languages

0.76

 Immunity

0.74

foreign

0.72

Activations Density 0.004%

mentions of "Foreign" entities or activities

references to foreign affairs and international relations

No Comments

No Known Activations