INDEX

Explanations

terms related to modifying, adjusting, and repairing belongings or systems

oai_token-act-pair · gpt-3.5-turbo

conjunctions and phrases indicating addition or modification

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

ISON

-0.78

Soc

-0.72

tale

-0.69

STEM

-0.67

icted

-0.66

Ask

-0.65

 Idle

-0.64

aga

-0.64

UES

-0.63

 Places

-0.61

POSITIVE LOGITS

 reorgan

1.31

 refin

1.29

 rearr

1.27

 refurb

1.26

 redesign

1.25

 remod

1.21

 restruct

1.18

 rede

1.13

 recon

1.10

 reuse

1.10

Activations Density 0.211%

terms related to modifying, adjusting, and repairing belongings or systems

conjunctions and phrases indicating addition or modification

No Comments

No Known Activations