INDEX

Explanations

phrases related to communication or statements made in various types of interviews

oai_token-act-pair · gpt-3.5-turbo

references to interviews and formal discussions

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

 Chau

-0.64

Pg

-0.61

Rated

-0.61

''.

-0.59

Ly

-0.56

 quot

-0.56

"}

-0.56

"></

-0.54

Written

-0.54

malink

-0.54

POSITIVE LOGITS

that

0.79

whether

0.72

 congratulations

0.68

:"

0.68

]:

0.67

 whether

0.67

ouncing

0.66

ounce

0.66

ãĤ¤ãĥĪ

0.66

ogg

0.64

Activations Density 0.447%

phrases related to communication or statements made in various types of interviews

references to interviews and formal discussions

No Comments

No Known Activations