INDEX

Explanations

phrases related to freedom of choice and personal preference

oai_token-act-pair · gpt-3.5-turbo

phrases related to personal freedom and choice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ļéĨĴ

-0.81

 Immunity

-0.64

advertisement

-0.64

ãĥ¼ãĥĨãĤ£

-0.62

ctuary

-0.62

Imm

-0.59

¿½

-0.59

EStreamFrame

-0.58

ifax

-0.58

dated

-0.58

POSITIVE LOGITS

 want

1.30

 wish

1.29

 desire

1.18

want

1.17

 pleased

1.17

 wished

1.16

 wanna

1.12

 please

1.12

ple

1.10

 desired

1.06

Activations Density 0.154%

phrases related to freedom of choice and personal preference

phrases related to personal freedom and choice

No Comments

No Known Activations