INDEX

Explanations

words indicating past events or states

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

technical terms or foreign words

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

})();

-0.36

 دور

-0.34

]=>

-0.34

Recognizer

-0.31

 ")");

-0.30

)):

-0.29

 sharp

-0.29

 Also

-0.29

 conosciuto

-0.28

://

-0.28

POSITIVE LOGITS

 Autorisations

0.70

 Autorizaciones

0.69

 CreateTagHelper

0.65

 fusca

0.63

 ModelExpression

0.63

InjectAttribute

0.60

StoryboardSegue

0.59

EndContext

0.59

wieś

0.58

 queſta

0.57

Activations Density 0.284%

words indicating past events or states

technical terms or foreign words

No Comments

No Known Activations