INDEX

Explanations

words related to responsibilities or tasks

oai_token-act-pair · gpt-3.5-turbo

words related to responsibility or obligation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Almighty

-0.66

 inexper

-0.62

 name

-0.62

 scraps

-0.61

 smartphones

-0.61

 inspiration

-0.60

 embod

-0.60

 Awakening

-0.60

 shine

-0.59

 Byzantine

-0.59

POSITIVE LOGITS

ort

4.63

orts

2.89

orted

2.40

orting

2.40

ORT

2.27

ortion

1.53

ortex

1.42

orter

1.38

ortal

1.29

ORTS

1.26

Activations Density 0.011%

words related to responsibilities or tasks

words related to responsibility or obligation

No Comments

No Known Activations