INDEX

Explanations

phrases indicating ongoing actions or processes within a context

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

 AssemblyCompany

-0.77

 TextInputType

-0.77

stateProvider

-0.77

SequentialGroup

-0.75

AsUp

-0.73

 المعيارى

-0.70

richTextPanel

-0.69

########.

-0.69

 AssemblyVersion

-0.69

Personendaten

-0.68

POSITIVE LOGITS

 beginnetje

0.32

 juke

0.28

Rhestr

0.28

boîte

0.27

 Rathaus

0.27

 kaikk

0.26

 palju

0.26

 Celui

0.25

 Øst

0.25

ltä

0.25

Activations Density 0.054%

phrases indicating ongoing actions or processes within a context

foreign languages

No Comments

No Known Activations