INDEX

Explanations

terms related to location, specifically "Locator."

oai_token-act-pair · gpt-3.5-turbo

references to location-based technology or systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

child

-0.86

uyomi

-0.74

iscopal

-0.72

ACA

-0.70

manship

-0.67

mast

-0.67

RIS

-0.67

older

-0.67

ħĭ

-0.65

dates

-0.65

POSITIVE LOGITS

atell

0.98

ombo

0.94

Loc

0.93

abulary

0.86

ality

0.82

ally

0.81

entric

0.80

Loc

0.80

ocity

0.78

 corrid

0.78

Activations Density 0.009%

terms related to location, specifically "Locator."

references to location-based technology or systems

No Comments

No Known Activations