INDEX

Explanations

locations or settings where major events or incidents took place

oai_token-act-pair · gpt-3.5-turbo

references to locations and places

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

Not in Any Lists

No Comments

Negative Logits

-0.84

HUD

-0.79

TPP

-0.79

é¾į

-0.76

IRED

-0.76

Crunch

-0.74

NA

-0.74

Mail

-0.73

Netflix

-0.72

inker

-0.72

POSITIVE LOGITS

 area

1.53

 scene

1.49

 vicinity

1.47

 premises

1.30

 facility

1.25

 compound

1.23

 situation

1.20

 site

1.17

 perimeter

1.14

 fray

1.12

Activations Density 0.224%

locations or settings where major events or incidents took place

references to locations and places

No Comments

No Known Activations