INDEX

Explanations

phrases related to involvement or responsibility

oai_token-act-pair · gpt-3.5-turbo

instances of the word "in," often related to roles or involvement

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

alli

-0.75

cit

-0.75

types

-0.75

arious

-0.69

intend

-0.68

ibl

-0.67

earchers

-0.67

=-=-

-0.66

bytes

-0.66

icol

-0.65

POSITIVE LOGITS

 shaping

1.69

 influencing

1.48

 facilitating

1.47

 determining

1.44

 creating

1.38

 helping

1.38

 ensuring

1.35

 arranging

1.34

 organising

1.32

 overseeing

1.30

Activations Density 0.150%

phrases related to involvement or responsibility

instances of the word "in," often related to roles or involvement

No Comments

No Known Activations