INDEX

Explanations

statements indicating future plans or evaluations

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

dates and times

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

Negative Logits

 InputDecoration

-0.44

 onOptions

-0.40

RefNanny

-0.38

libft

-0.36

Pyx

-0.35

men

-0.34

men

-0.33

undle

-0.33

hon

-0.33

ent

-0.33

POSITIVE LOGITS

 typelib

0.69

 mendatang

0.63

 feroit

0.61

 devront

0.60

 prochaines

0.60

 éché

0.59

 prochain

0.57

 dovrà

0.56

 autorytatywna

0.56

ьаж

0.55

Activations Density 0.046%

statements indicating future plans or evaluations

dates and times

No Known Activations