INDEX

Explanations

terms related to quantitative research and analysis methods

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

Followed by a question marker or abbreviation

np_acts-logits-general · gemini-2.0-flash

control, restriction, access, memory

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 cherchés

-0.49

!*\

-0.46

 souverain

-0.46

 Consumo

-0.45

 mihi

-0.45

javadoc

-0.45

рав

-0.44

 Russie

-0.44

 viac

-0.43

nija

-0.43

POSITIVE LOGITS

 يتيمه

0.74

 lenker

0.65

matchCondition

0.65

potranspiration

0.62

modelBuilder

0.62

adaptiveStyles

0.58

 utafitiHapana

0.57

structors

0.57

washingtonpost

0.57

calaure

0.56

Activations Density 0.798%

terms related to quantitative research and analysis methods

Followed by a question marker or abbreviation

control, restriction, access, memory

No Comments

No Known Activations