INDEX

Explanations

phrases related to analysis and optimization in data processing and algorithms

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

Followed by "more," "as," "obtaining," "again."

np_acts-logits-general · gemini-2.0-flash

even more

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

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

 decent

-0.33

Jeografia

-0.33

ười

-0.32

indirizzo

-0.31

parametrize

-0.30

Decent

-0.29

нке

-0.29

AsUp

-0.28

 Modular

-0.28

 Contrary

-0.28

POSITIVE LOGITS

 further

2.64

further

2.44

Further

2.13

 FURTHER

2.09

 Further

2.09

さらに

2.06

更に

1.89

 even

1.77

urther

1.72

更是

1.68

Activations Density 0.913%

phrases related to analysis and optimization in data processing and algorithms

Followed by "more," "as," "obtaining," "again."

even more

No Comments

No Known Activations