INDEX

Explanations

references and calculations related to numerical data and averages

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

average rates and durations

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

fato

-0.36

 elevators

-0.35

mation

-0.35

isamment

-0.34

出版年

-0.34

kungan

-0.34

 Médico

-0.33

といけない

-0.33

dess

-0.33

ífer

-0.32

POSITIVE LOGITS

andExpect

0.58

AccessorTable

0.54

 BoxFit

0.54

newtheorem

0.53

HasAnnotation

0.50

makedirs

0.48

AnchorStyles

0.48

moveToFirst

0.47

 ExecuteAsync

0.46

 AttributeSet

0.45

Activations Density 0.083%

references and calculations related to numerical data and averages

average rates and durations

No Comments

No Known Activations