INDEX

Explanations

references to memory or the act of recalling information

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

 AssemblyCulture

-0.89

contentLoaded

-0.88

ConstraintMaker

-0.81

ISupport

-0.79

 defaultstate

-0.77

 ostavi

-0.76

 nahilalakip

-0.75

 AssemblyVersion

-0.73

ⓧ

-0.73

istoitu

-0.72

POSITIVE LOGITS

saw

0.53

iddhar

0.47

me

0.47

 deci

0.44

its

0.44

 дре

0.44

 lead

0.44

 thức

0.44

ני

0.43

Regexp

0.43

Activations Density 0.033%

references to memory or the act of recalling information

No Comments

No Known Activations