INDEX

Explanations

phrases related to collaboration and working together towards a common goal

oai_token-act-pair · gpt-3.5-turbo

references to collaboration with various organizations and authorities

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

Quotes

-0.64

 spite

-0.63

ording

-0.62

ãĥĩãĤ£

-0.61

]}

-0.61

Wr

-0.61

Spoiler

-0.59

must

-0.58

Ö¼

-0.58

)=

-0.57

POSITIVE LOGITS

to

1.01

 counterparts

0.88

 partners

0.85

 toward

0.83

 regarding

0.78

on

0.75

 towards

0.75

 stakeholders

0.72

 representatives

0.71

assetsadobe

0.70

Activations Density 0.242%

phrases related to collaboration and working together towards a common goal

references to collaboration with various organizations and authorities

No Comments

No Known Activations