INDEX

Explanations

referenced concepts related to actions, opportunities, or choices in various contexts

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

foreign languages

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

司

-0.39

AudioManager

-0.39

Fog

-0.39

 stylo

-0.39

 ciphertext

-0.38

 OkHttpClient

-0.38

fog

-0.38

 metálico

-0.37

 Docket

-0.37

 interstate

-0.37

POSITIVE LOGITS

 समीक्षाएं

0.51

Revenir

0.50

rungsseite

0.48

 الحره

0.48

Izvori

0.47

GEBURTS

0.47

ThemeOverlay

0.45

 MainAxisSize

0.45

Tembelea

0.44

ItemBackground

0.44

Activations Density 0.048%

referenced concepts related to actions, opportunities, or choices in various contexts

foreign languages

No Comments

No Known Activations