INDEX

Explanations

phrases related to emergence or coming out from a particular state or situation

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate origins or sources

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

oir

-0.67

amount

-0.66

intend

-0.66

selves

-0.66

 minded

-0.63

nant

-0.61

iveness

-0.61

ivities

-0.60

 entit

-0.60

lement

-0.60

POSITIVE LOGITS

 nowhere

1.30

 hiber

1.21

 obscurity

1.18

 hiding

0.99

 hiatus

0.94

 exile

0.94

 whence

0.92

 captivity

0.87

 retirement

0.85

 abroad

0.83

Activations Density 0.105%

phrases related to emergence or coming out from a particular state or situation

phrases that indicate origins or sources

No Comments

No Known Activations