INDEX

Explanations

phrases indicating emphasis or importance

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "have" used in various contexts

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

drive

-0.71

oshi

-0.70

eem

-0.68

cone

-0.64

tip

-0.63

alter

-0.63

âĺħ

-0.63

imo

-0.63

quote

-0.62

ogo

-0.62

POSITIVE LOGITS

 ourselves

1.17

 seen

1.10

 been

1.06

 witnessed

0.98

 heard

0.95

 gotten

0.91

 begun

0.89

 talked

0.89

 spoken

0.86

our

0.86

Activations Density 0.131%

phrases indicating emphasis or importance

instances of the phrase "have" used in various contexts

No Comments

No Known Activations