INDEX

Explanations

references to structural elements in programming, specifically related to thread management and counts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_0/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.0.hook_mlp_out

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

 }}$}

-0.64

annica

-0.62

'):

-0.62

름

-0.60

-------

-0.60

"):

-0.60

 bruto

-0.59

 الحره

-0.59

 thước

-0.59

]:

-0.58

POSITIVE LOGITS

 كومونز

0.69

ViewFeatures

0.54

AsUp

0.52

ItemLayout

0.51

 chắn

0.49

 parseFrom

0.49

 بيها

0.48

тельстве

0.48

__(/*!

0.47

 AttributeSet

0.47

Activations Density 0.167%

references to structural elements in programming, specifically related to thread management and counts

No Comments

No Known Activations