INDEX

Explanations

concepts related to quantum mechanics and neural networks

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

preceding nouns

np_acts-logits-general · gemini-2.0-flash

Specifically, profoundly, ideas, fuel

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

endmodule

-0.33

 royaume

-0.33

 Faites

-0.28

sû

-0.28

 enfans

-0.27

 ſta

-0.27

mû

-0.26

Betyg

-0.26

 trámites

-0.25

 âme

-0.25

POSITIVE LOGITS

httphttps

0.78

ChildScrollView

0.73

@"/

0.73

+#+#

0.72

tagHelperRunner

0.69

<unused14>

0.63

<unused28>

0.63

[@BOS@]

0.63

delwed

0.63

<pad>

0.63

Activations Density 0.535%

concepts related to quantum mechanics and neural networks

preceding nouns

Specifically, profoundly, ideas, fuel

No Comments

No Known Activations