INDEX

Explanations

statements or sentences that express significant findings or conclusions in research.

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

information related to scientific studies and their methodology.

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

the beginning of a new section or topic in a document, often indicated by words like "This study", "This paper", or "This work".

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @bot

"This" followed by academic/technical terms

np_acts-logits-general · gemini-2.0-flash

introducing academic content

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+#+

-0.57

:][

-0.55

 betweenstory

-0.54

 &___

-0.52

 '\\;'

-0.52

 nakalista

-0.52

rinth

-0.50

Abitanti

-0.49

 للاسماء

-0.48

henswürdigkeiten

-0.48

POSITIVE LOGITS

 article

0.74

 paper

0.70

 section

0.66

 page

0.63

 заключается

0.60

article

0.57

tagHelperRunner

0.57

 JpaRepository

0.56

homonymie

0.56

 guide

0.55

Activations Density 0.384%

statements or sentences that express significant findings or conclusions in research.

information related to scientific studies and their methodology.

the beginning of a new section or topic in a document, often indicated by words like "This study", "This paper", or "This work".

"This" followed by academic/technical terms

introducing academic content

No Comments

No Known Activations