INDEX

Explanations

phrases related to reasons or justifications

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their actions or situations

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

----------------------------------------------------------------

-0.70

ortun

-0.66

 nutshell

-0.66

GGGGGGGG

-0.65

UGE

-0.64

venge

-0.64

 renaissance

-0.63

 unveiling

-0.63

--------------------------------

-0.62

 toget

-0.62

POSITIVE LOGITS

 lacked

1.58

 hadn

1.55

 disagreed

1.32

 objected

1.26

 refused

1.25

 wasn

1.23

 lacks

1.23

 feared

1.19

 doubted

1.17

 didn

1.16

Activations Density 0.370%

phrases related to reasons or justifications

references to individuals and their actions or situations

No Comments

No Known Activations