INDEX

Explanations

punctuation and indicators of thought or reflection

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

liegende

-0.43

 continúas

-0.42

 étendue

-0.41

ariConfig

-0.39

tų

-0.38

 Vans

-0.38

Configured

-0.38

 aggregates

-0.37

 âme

-0.37

minify

-0.37

POSITIVE LOGITS

LookAnd

0.65

 oprot

0.57

 Мексичка

0.55

httphttps

0.48

 ModelExpression

0.48

MarshalTo

0.48

:✨

0.48

zzleHttp

0.47

 resourceCulture

0.47

UnsafeEnabled

0.47

Activations Density 0.004%

punctuation and indicators of thought or reflection

No Comments

No Known Activations