INDEX

Explanations

evidence of exhibited behaviors or symptoms

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

exhibited characteristics or behavior

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

Combien

-0.30

 trône

-0.30

 évêque

-0.30

饭

-0.30

 Wahrheit

-0.29

 Geburt

-0.28

 huesos

-0.28

 générations

-0.28

 Landwirtschaft

-0.27

 τού

-0.26

POSITIVE LOGITS

 Showing

0.84

 exhibited

0.84

 SHOWING

0.83

 showing

0.80

 autorytatywna

0.79

 showed

0.79

Showing

0.78

@[+][

0.77

 exhibiting

0.77

showed

0.75

Activations Density 0.062%

evidence of exhibited behaviors or symptoms

exhibited characteristics or behavior

No Comments

No Known Activations