INDEX

Explanations

terms related to technical and scientific concepts

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

specific nouns followed by common 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

 Económica

-0.47

Παραπομπές

-0.47

 Normdatei

-0.44

transQ

-0.44

hyrchwyd

-0.43

 kasarigan

-0.43

MLLoader

-0.42

GIVEREF

-0.40

 resourceCulture

-0.39

 ligiloj

-0.39

POSITIVE LOGITS

المكان

0.50

Hentet

0.50

licit

0.48

 modeling

0.47

phrag

0.45

obot

0.45

grass

0.44

STA

0.43

 fencing

0.43

 models

0.42

Activations Density 0.099%

terms related to technical and scientific concepts

specific nouns followed by common nouns

No Comments

No Known Activations