INDEX

Explanations

mentions of specific names and entities

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

Ivo/Iv and Tus

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

 ſind

-0.65

transQ

-0.63

 iſt

-0.54

Batman

-0.53

 Crusoe

-0.52

 dévelo

-0.51

Romeo

-0.51

Beethoven

-0.50

 Pushkin

-0.50

NATO

-0.50

POSITIVE LOGITS

 Ivory

2.02

Ivory

1.90

 ivory

1.79

ivory

1.49

Ivoire

1.36

 Ivor

0.89

Iv

0.68

 Abidjan

0.61

ivo

0.61

 Ebony

0.58

Activations Density 0.001%

mentions of specific names and entities

Ivo/Iv and Tus

No Comments

No Known Activations