INDEX

Explanations

references to characters in various contexts

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

"character" or "characters"

np_acts-logits-general · gemini-2.0-flash

character and its forms

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Bisnis

-0.57

 Blitz

-0.54

zbęd

-0.53

uckoo

-0.52

⏩

-0.52

<bos>

-0.51

 {}));

-0.50

 Abbey

-0.48

eclampsia

-0.47

 <=",

-0.47

POSITIVE LOGITS

Character

0.98

 Character

0.95

 character

0.93

character

0.92

 Characters

0.89

Characters

0.85

 characters

0.84

 CHARACTER

0.84

CHARACTER

0.82

characters

0.81

Activations Density 0.050%

references to characters in various contexts

"character" or "characters"

character and its forms

No Comments

No Known Activations