INDEX

Explanations

words related to action or behavior

oai_token-act-pair · gpt-3.5-turbo

instances of the word "act" and its variations, indicating a focus on actions or behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ickets

-0.69

nic

-0.66

yip

-0.65

ciating

-0.64

fer

-0.62

antha

-0.60

ixels

-0.60

xual

-0.59

burn

-0.59

Teg

-0.59

POSITIVE LOGITS

uate

1.13

uated

1.08

uary

1.04

 accordingly

0.98

 decisively

0.92

 differently

0.92

inic

0.90

 upon

0.90

 impuls

0.88

 appropriately

0.87

Activations Density 0.053%

words related to action or behavior

instances of the word "act" and its variations, indicating a focus on actions or behaviors

No Comments

No Known Activations