INDEX

Explanations

improvements or enhancements related to design and efficiency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 officially

-0.55

tagHelperRunner

-0.53

 procéder

-0.51

]!=

-0.50

BeginInit

-0.50

{?>

-0.50

})]

-0.50

 differing

-0.49

 tips

-0.49

tende

-0.48

POSITIVE LOGITS

 snippetHide

0.72

 stability

0.68

 visibility

0.66

 fluidity

0.66

 adaptability

0.65

 confidence

0.64

LookAnd

0.64

 credibility

0.63

eability

0.62

 morale

0.61

Activations Density 0.415%

improvements or enhancements related to design and efficiency

No Comments

No Known Activations