INDEX

Explanations

variations of the word "Cho" in different contexts

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

cho and choke

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

معلومات

-0.44

 américaine

-0.42

année

-0.41

นาน

-0.41

 Tennant

-0.40

 années

-0.40

 invierno

-0.40

 prêtre

-0.39

 Stedman

-0.39

 antiguas

-0.39

POSITIVE LOGITS

Cho

2.17

Cho

1.99

cho

1.79

CHO

1.50

cho

1.41

CHO

1.23

 choke

1.22

 Choo

1.20

 Chopper

1.14

 choo

1.11

Activations Density 0.003%

variations of the word "Cho" in different contexts

cho and choke

No Comments

No Known Activations