INDEX

Explanations

references to openings or apertures

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

 opening

-0.64

 kereszt

-0.50

 träd

-0.50

 découver

-0.49

 igång

-0.49

 spé

-0.47

 vostri

-0.47

 passione

-0.47

 vägen

-0.47

Garantía

-0.46

POSITIVE LOGITS

ConstraintMaker

0.91

awtextra

0.87

makeConstraints

0.83

 GenerationType

0.82

addPreferredGap

0.77

DispatchToProps

0.74

 CreateTagHelper

0.73

reportWebVitals

0.71

 JAXBElement

0.67

 ExecuteAsync

0.67

Activations Density 0.017%

references to openings or apertures

No Comments

No Known Activations