INDEX

Explanations

words related to modifications or enhancements

oai_token-act-pair · gpt-3.5-turbo

references to modifications in games or digital media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Becky

-0.81

 Opportunity

-0.80

 Goose

-0.79

anwhile

-0.77

 Gideon

-0.76

Ago

-0.75

ewater

-0.73

ideon

-0.73

comings

-0.73

stadt

-0.72

POSITIVE LOGITS

mod

3.66

 mods

2.32

Mod

2.26

mod

2.10

 Mods

1.95

Mod

1.95

Mods

1.88

MOD

1.85

MOD

1.60

mods

1.56

Activations Density 0.010%

words related to modifications or enhancements

references to modifications in games or digital media

No Comments

No Known Activations