INDEX

Explanations

references to scientific or medical concepts and entities

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

the first token of a document, particularly focusing on capitalized words that start sentences or proper nouns.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

text with structural/contextual markers like beginning of text (<bos>) and large numerical values.

oai_token-act-pair · claude-3-5-haiku-20241022 Triggered by @neilrathi

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_6/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<eos>

-0.30

</i>

-0.29

</b>

-0.29

ρυ

-0.28

жде

-0.28

bur

-0.28

veland

-0.28

ο

-0.27

onto

-0.27

Weg

-0.27

POSITIVE LOGITS

 myſelf

0.65

 itſelf

0.62

Efq

0.61

 Monfieur

0.58

 Theſe

0.57

 themſelves

0.56

 whoſe

0.56

 himſelf

0.54

 raiſ

0.54

 purpoſe

0.52

Activations Density 46.605%

references to scientific or medical concepts and entities

the first token of a document, particularly focusing on capitalized words that start sentences or proper nouns.

text with structural/contextual markers like beginning of text (<bos>) and large numerical values.

No Comments

No Known Activations