INDEX

Explanations

mathematical symbols and notation typically associated with formal proofs

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

\]

-1.77

\]

-1.04

Демографія

-0.98

}\]

-0.89

gdx

-0.73

tagHelperRunner

-0.71

 Majefty

-0.68

 Draws

-0.65

enumi

-0.64

riwal

-0.64

POSITIVE LOGITS

 FixedUpdate

0.49

ल्प

0.45

普

0.41

われています

0.40

nam

0.39

blitz

0.39

 invokingState

0.39

LPS

0.39

šča

0.39

stanz

0.39

Activations Density 0.123%

mathematical symbols and notation typically associated with formal proofs

No Comments

No Known Activations