INDEX

Explanations

references to emerging trends and entities across various fields

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

emerging paired with nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Picchu

-0.45

 aDecoder

-0.45

blico

-0.44

 onOptions

-0.44

CPtr

-0.43

SPATH

-0.42

jacency

-0.41

 SDLK

-0.41

 letzt

-0.40

 torebka

-0.40

POSITIVE LOGITS

 emerging

0.73

Emerging

0.65

 Emerging

0.63

 emergent

0.59

 prospects

0.59

new

0.59

upcoming

0.59

 incipient

0.59

 budding

0.58

 nascent

0.58

Activations Density 0.008%

references to emerging trends and entities across various fields

emerging paired with nouns

No Comments

No Known Activations