INDEX

Explanations

references to frequency and necessity in contexts involving updates or information compliance

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

conditions and criteria

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

KommentareTeilen

-0.43

 CommonModule

-0.41

 Resistance

-0.40

 avancée

-0.38

zecz

-0.38

 Ruhm

-0.36

 illustrationer

-0.36

 Hubbard

-0.36

 Allein

-0.36

hamma

-0.36

POSITIVE LOGITS

 صوتيه

0.46

SBATCH

0.44

ByVersion

0.44

@"/

0.43

ESOME

0.39

codable

0.38

Վ

0.38

 MainAxisSize

0.38

 Schengen

0.37

 arşivlendi

0.37

Activations Density 0.296%

references to frequency and necessity in contexts involving updates or information compliance

conditions and criteria

No Comments

No Known Activations