INDEX

Explanations

proper nouns or notable names related to various subjects

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

start of capitalized words

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

 Taktlose

-0.84

出版年

-0.81

 autorytatywna

-0.80

хьтан

-0.79

IntoConstraints

-0.79

oa̍t

-0.78

 kaarangay

-0.77

 OMITBAD

-0.77

Хьажоргаш

-0.76

 للمعارف

-0.75

POSITIVE LOGITS

is

0.47

0.41

has

0.36

The

0.35

was

0.35

 tuot

0.35

0.33

0.30

 chercheurs

0.30

had

0.29

Activations Density 0.052%

proper nouns or notable names related to various subjects

start of capitalized words

No Comments

No Known Activations