INDEX

Explanations

outcomes or solutions to unknown situations

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the pursuit of information or answers

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

oun

-0.80

ovich

-0.70

ãĤ´ãĥ³

-0.69

uga

-0.67

 entreprene

-0.65

ĸļ

-0.63

ortunately

-0.62

ThumbnailImage

-0.61

ourses

-0.61

luster

-0.61

POSITIVE LOGITS

why

1.18

WHY

1.09

 about

1.08

 what

1.08

 whats

1.07

 whether

1.05

how

1.03

 whence

0.99

who

0.98

 exactly

0.95

Activations Density 0.080%

outcomes or solutions to unknown situations

phrases indicating the pursuit of information or answers

No Comments

No Known Activations