INDEX

Explanations

phrases indicating limits, thresholds, or maximum capacities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 للمعارف

-0.91

httphttps

-0.61

 שוליים

-0.61

 intptr

-0.58

קישורים

-0.57

AndEndTag

-0.57

 виправивши

-0.54

 चीज़ों

-0.52

})();

-0.51

 فريبيس

-0.51

POSITIVE LOGITS

testens

0.75

 propOrder

0.70

ConstraintMaker

0.58

Conclusiones

0.58

InlineData

0.57

ksesta

0.56

MockBean

0.56

omitempty

0.55

 kuiten

0.55

 maksimum

0.55

Activations Density 0.057%

phrases indicating limits, thresholds, or maximum capacities

No Comments

No Known Activations