INDEX

Explanations

entries marked by special symbols or specific structured formats within the text

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

Code comments or questions

np_acts-logits-general · gemini-2.0-flash

Stone, MODIFIED, Immun

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

<eos>

-0.60

uests

-0.60

ucson

-0.58

tacts

-0.54

riculture

-0.51

});

-0.49

?>

-0.49

:),

-0.48

-0.47

ercises

-0.47

POSITIVE LOGITS

sidemargin

0.96

+#+#

0.91

Controllo

0.88

 незавершена

0.88

 виправивши

0.87

Autoritní

0.85

 Baillargeon

0.84

DeleteBehavior

0.84

 vPvB

0.82

 ditto

0.81

Activations Density 0.906%

entries marked by special symbols or specific structured formats within the text

Code comments or questions

Stone, MODIFIED, Immun

No Comments

No Known Activations