INDEX

Explanations

terms related to indicating, signaling, or providing evidence

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

"Indicate", "indication", or "indicator"

np_acts-logits-general · gemini-2.0-flash

indication or signal

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 Gomes

-0.38

 Bost

-0.32

 Nutzungs

-0.32

 Verfahrens

-0.31

 journées

-0.31

userType

-0.31

 bezit

-0.31

McM

-0.30

 pasan

-0.30

 fully

-0.29

POSITIVE LOGITS

 Indication

1.05

indication

1.03

indicate

1.02

Indication

1.00

 indication

1.00

 INDIC

1.00

 Indications

0.98

 Indicator

0.98

Indications

0.95

 indications

0.94

Activations Density 0.248%

terms related to indicating, signaling, or providing evidence

"Indicate", "indication", or "indicator"

indication or signal

No Comments

No Known Activations