INDEX

Explanations

mathematical variables and parameters related to a specific context or formulae in research

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.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

 Sein

-0.37

mô

-0.36

吸収

-0.36

setParent

-0.35

(_

-0.35

(_

-0.35

 akong

-0.34

$_

-0.34

=_

-0.34

atek

-0.34

POSITIVE LOGITS

 HasFactory

0.98

expandindo

0.95

 виправивши

0.93

Portail

0.82

 مرئيه

0.81

ngdoc

0.81

BeginContext

0.79

 snippetHide

0.78

MigrationBuilder

0.77

 ***!

0.77

Activations Density 0.692%

mathematical variables and parameters related to a specific context or formulae in research

No Comments

No Known Activations