INDEX

Explanations

numeric data and statistical references

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

IHS Markit

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

HXLINE

-0.47

 pinulongan

-0.41

abestanden

-0.39

AsUp

-0.39

XmlIgnore

-0.35

 HasFactory

-0.34

 inalámbrica

-0.34

 disambiguazione

-0.32

BaseController

-0.31

Халык

-0.30

POSITIVE LOGITS

clero

0.58

 kasarigan

0.54

icep

0.52

 uits

0.51

enciaga

0.50

 Normdatei

0.50

similiano

0.48

pec

0.48

örm

0.48

IMUM

0.47

Activations Density 0.007%

numeric data and statistical references

IHS Markit

No Comments

No Known Activations