INDEX

Explanations

words related to automation or automated processes

oai_token-act-pair · gpt-3.5-turbo

instances of the word "autom" and its variations, indicating a focus on automation or automated processes

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

Uri

-0.68

Jed

-0.67

 Mara

-0.66

lihood

-0.66

 uncond

-0.62

 harbor

-0.61

Syd

-0.59

Sri

-0.59

 impunity

-0.59

 Pell

-0.59

POSITIVE LOGITS

otive

1.89

obile

1.85

obiles

1.67

obil

1.44

aton

1.36

akers

1.34

aker

1.25

atics

1.21

atically

1.17

ated

1.16

Activations Density 0.025%

words related to automation or automated processes

instances of the word "autom" and its variations, indicating a focus on automation or automated processes

No Comments

No Known Activations