INDEX

Explanations

programming constructs and syntax elements, particularly conditionals and return statements

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

Non-English words

np_acts-logits-general · gemini-2.0-flash

code syntax

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

+#+#

-0.43

InstanceState

-0.39

IUrlHelper

-0.39

 Deiner

-0.38

 dés

-0.36

 infancia

-0.36

 tillegg

-0.35

 sacré

-0.35

 Gattung

-0.34

 silvestre

-0.34

POSITIVE LOGITS

 مشين

0.77



0.68

 ujednoznacz

0.59

rungsseite

0.57

المكان

0.55

存于互联网档案馆

0.53

 Italijanski

0.52

endregion

0.49

 चीज़ों

0.48

angelog

0.47

Activations Density 0.179%

programming constructs and syntax elements, particularly conditionals and return statements

Non-English words

code syntax

No Comments

No Known Activations