INDEX

Explanations

instances of structured data formats or mathematical notation

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

Personendaten

-0.64

abestanden

-0.62

 oignon

-0.58

انظر

-0.54

Referências

-0.50

LLLL

-0.50

 pourrais

-0.50

Portale

-0.50

 BorderRadius

-0.50

otesi

-0.49

POSITIVE LOGITS

はじめに

0.69

+#+

0.69

TypedDataSet

0.66

#+#

0.65

ftagPool

0.63

MigrationBuilder

0.53

afficheront

0.52

 թվական

0.52

@[+][

0.52

lias

0.52

Activations Density 0.009%

instances of structured data formats or mathematical notation

No Comments

No Known Activations