INDEX

Explanations

phrases related to initial occurrences or debuts

oai_token-act-pair · gpt-3.5-turbo

references to someone's first experiences or appearances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

bara

-0.92

stairs

-0.76

holes

-0.75

ules

-0.68

 Vish

-0.67

ppings

-0.65

bleacher

-0.64

thinkable

-0.64

abus

-0.63

stadt

-0.63

POSITIVE LOGITS

 foray

1.41

 outing

1.22

 stint

1.04

 birthday

1.02

 appearance

1.02

 cousin

0.96

 attempt

0.93

 career

0.88

 incarnation

0.88

 taste

0.86

Activations Density 0.102%

phrases related to initial occurrences or debuts

references to someone's first experiences or appearances

No Comments

No Known Activations