INDEX

Explanations

words related to statements made by individuals

oai_token-act-pair · gpt-3.5-turbo

instances of reported speech or quotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

©¶æ

-0.71

powered

-0.71

ãĥĥãĥĪ

-0.71

flo

-0.69

DragonMagazine

-0.69

atana

-0.68

river

-0.67

circle

-0.65

flying

-0.65

successfully

-0.64

POSITIVE LOGITS

 said

0.77

 Cheong

0.73

 Evan

0.73

Said

0.70

Published

0.70

 Trey

0.70

 McGill

0.69

 Gould

0.68

 Freeman

0.68

 Reverend

0.67

Activations Density 0.387%

words related to statements made by individuals

instances of reported speech or quotations

No Comments

No Known Activations