INDEX

Explanations

phrases related to attitudes or mindsets, particularly ones with a negative connotation

oai_token-act-pair · gpt-3.5-turbo

concepts relating to mentality and mindset

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

ĵ

-0.66

leased

-0.65

 extensions

-0.62

repe

-0.62

aster

-0.61

 Foss

-0.60

umin

-0.60

abel

-0.60

overed

-0.59

ez

-0.58

POSITIVE LOGITS

 mentality

3.59

 mindset

3.32

 attitude

1.88

 ethos

1.67

 psyche

1.61

 worldview

1.43

 demeanor

1.40

 vibe

1.38

 ethic

1.33

 lifestyle

1.32

Activations Density 0.013%

phrases related to attitudes or mindsets, particularly ones with a negative connotation

concepts relating to mentality and mindset

No Comments

No Known Activations