INDEX

Explanations

phrases related to discussions or dialogues

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate conjunctions or connected thoughts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 whis

-0.73

 horizont

-0.66

onite

-0.65

 forced

-0.65

 hemisphere

-0.62

 menace

-0.62

 landfill

-0.62

 pretend

-0.61

 stranded

-0.61

 bedrooms

-0.59

POSITIVE LOGITS

âĢ

1.59

Â»

1.29

âĸł

1.27

âĸº

1.22

Â«

1.22

ãĢ

1.20

"(

1.18

^^

1.16

Activations Density 0.502%

phrases related to discussions or dialogues

phrases that indicate conjunctions or connected thoughts

No Comments

No Known Activations