INDEX

Explanations

statistical and predictive information related to study outcomes and health analyses

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

suggests indicates conclusion predictive

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

хьтан

-0.79

ſammen

-0.68

 Administrativna

-0.65

MigrationBuilder

-0.63

 AssemblyCulture

-0.63

 Numerade

-0.59

 Италијани

-0.59

rollment

-0.59

 nakalista

-0.58

󠁣

-0.57

POSITIVE LOGITS

 lecz

0.32

 collectie

0.32

 Erscheinung

0.30

 Factbook

0.30

 muhimu

0.29

)}

0.27

 zatem

0.27

 Werken

0.26

及び

0.26

 лікар

0.25

Activations Density 0.165%

statistical and predictive information related to study outcomes and health analyses

suggests indicates conclusion predictive

No Comments

No Known Activations