INDEX

Explanations

terms related to diagnostics and statistical models

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_4/width_65k/average_l0_66

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.4.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

 valami

-0.59

 Paglinawan

-0.57

 AssemblyCulture

-0.57

featureID

-0.56

UnusedPrivate

-0.55

الإنجليزية

-0.55

 estekak

-0.55

astify

-0.55

󠁢

-0.54

 fièvre

-0.53

POSITIVE LOGITS

 المدار

0.50

ConstraintMaker

0.48

 SIMBAD

0.47

hlander

0.45

تفصیلات

0.45

 absol

0.45

 visuals

0.44

ổi

0.44

 DBNull

0.44

raining

0.43

Activations Density 0.006%

terms related to diagnostics and statistical models

No Comments

No Known Activations