INDEX

Explanations

phrases that indicate action or encouragement to engage with material, such as books or movies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.hook_resid_post

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

 المعيارى

-0.76

due

-0.62

 waardoor

-0.62

due

-0.58

ніципа

-0.58

 оригіналу

-0.56

 nonché

-0.56

almost

-0.55

 dovuto

-0.55

Mutagenicity

-0.54

POSITIVE LOGITS

 yourself

1.19

吧

1.00

 please

0.98

 yourselves

0.93

 ASAP

0.87

 nhé

0.87

吧

0.87

NOW

0.86

yourself

0.85

if

0.82

Activations Density 0.326%

phrases that indicate action or encouragement to engage with material, such as books or movies

No Comments

No Known Activations