INDEX

Explanations

names of places, people or entities after which something is named

oai_token-act-pair · gpt-3.5-turbo

phrases that reference naming or being named after someone or something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

asive

-0.75

ibles

-0.74

vs

-0.74

Rs

-0.73

VR

-0.72

illance

-0.72

iphate

-0.71

URE

-0.69

alysis

-0.68

idential

-0.68

POSITIVE LOGITS

 initials

0.86

 George

0.84

 Norse

0.84

 Herbert

0.83

its

0.83

 famed

0.83

 Adolf

0.83

Sir

0.82

 Arthur

0.81

 Tolkien

0.80

Activations Density 0.119%

names of places, people or entities after which something is named

phrases that reference naming or being named after someone or something

No Comments

No Known Activations