INDEX

Explanations

terms related to data and statistics

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

all content published

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

甫

-0.35

 might

-0.30

 much

-0.30

疏

-0.29

 another

-0.29

 Bere

-0.29

Dragging

-0.29

ven

-0.29

pap

-0.28

 math

-0.28

POSITIVE LOGITS

 ویکی‌پدی

0.98

Personensuche

0.89

 CanadaChoose

0.79

 Audiodateien

0.73

 pinulongan

0.69

Hentet

0.68

RTCF

0.67

Datuak

0.66

 виправивши

0.66

WriteBarrier

0.64

Activations Density 0.119%

terms related to data and statistics

all content published

No Comments

No Known Activations