INDEX

Explanations

personal opinions or beliefs expressed with confidence

oai_token-act-pair · gpt-3.5-turbo

statements of personal opinion or subjective experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Equality

-0.69

 Pric

-0.66

 Towns

-0.61

 redistributed

-0.60

raviolet

-0.60

 Layer

-0.59

itect

-0.59

 Communist

-0.57

Catalog

-0.54

 incub

-0.54

POSITIVE LOGITS

zzo

1.07

adow

1.04

adows

1.01

 personally

0.87

asuring

0.84

lees

0.83

 somew

0.79

asured

0.79

andering

0.77

cca

0.76

Activations Density 0.064%

personal opinions or beliefs expressed with confidence

statements of personal opinion or subjective experiences

No Comments

No Known Activations