INDEX

Explanations

phrases related to expressing agreement or positive sentiment towards something

oai_token-act-pair · gpt-3.5-turbo

variations of the word "like" in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

tein

-0.75

 exting

-0.71

Catal

-0.70

 destro

-0.69

 enthusi

-0.68

inas

-0.68

Ire

-0.67

ilts

-0.66

 Parsons

-0.65

éŃĶ

-0.65

POSITIVE LOGITS

lihood

2.03

liest

1.17

minded

1.09

 minded

1.08

lier

1.03

liness

1.02

ability

1.01

able

0.89

joy

0.78

ably

0.73

Activations Density 0.048%

phrases related to expressing agreement or positive sentiment towards something

variations of the word "like" in different contexts

No Comments

No Known Activations