INDEX

Explanations

statements about what is good for people or entities in a variety of contexts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating what is beneficial or harmful for individuals or groups

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

orbit

-0.85

aults

-0.83

buster

-0.74

peer

-0.72

soever

-0.72

clair

-0.68

heid

-0.67

ieties

-0.67

ault

-0.67

operated

-0.67

POSITIVE LOGITS

 everybody

1.15

us

1.09

ya

1.05

 everyone

1.04

me

1.04

 morale

1.01

 awhile

1.01

 sure

0.99

 them

0.98

him

0.97

Activations Density 0.133%

statements about what is good for people or entities in a variety of contexts

phrases indicating what is beneficial or harmful for individuals or groups

No Comments

No Known Activations