INDEX

Explanations

references to time durations measured in minutes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 minute

-2.05

 Minute

-1.62

Minute

-1.57

minute

-1.55

 MINUTE

-1.39

 minuto

-0.99

MINUTE

-0.91

 minut

-0.88

 minuten

-0.69

Hour

-0.56

POSITIVE LOGITS

SequentialGroup

0.72

RetentionPolicy

0.66

 Paglinawan

0.63

 Vikipedi

0.62

tagext

0.59

AsUp

0.59

 Vogel

0.59

sonaro

0.58

例句

0.57

addCriterion

0.56

Activations Density 0.145%

references to time durations measured in minutes

No Comments

No Known Activations