INDEX

Explanations

instances of the word "barely."

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

 barely

-1.46

 hardly

-1.04

 scarcely

-0.87

ostante

-0.86

 едва

-0.82

 largely

-0.65

 scattata

-0.64

 récla

-0.62

 avvic

-0.62

 mostly

-0.61

POSITIVE LOGITS

Демографія

0.73

Tembelea

0.73

printStackTrace

0.72

SourceChecksum

0.66

verwijspagina

0.64

 unknownFields

0.63

tzmann

0.63

 jScrollPane

0.60

DebuggerNonUser

0.59

poons

0.59

Activations Density 0.013%

instances of the word "barely."

No Comments

No Known Activations