INDEX

Explanations

technical terms and concepts related to transformations and data structures

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

transformations, reference, architecture, problem, slots

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

 فريبيس

-0.40

pushFollow

-0.38

itettura

-0.38

getSettings

-0.38

InnerText

-0.37

 practicing

-0.37

 复古

-0.36

 recalled

-0.36

endencies

-0.36

 recalling

-0.36

POSITIVE LOGITS

ContentAsync

0.68

 ExecuteAsync

0.52

AutoScale

0.51

HtmlAttribute

0.48

scriptcase

0.47

enterOuterAlt

0.46

ieR

0.44

RuleContext

0.44

SequentialGroup

0.43

IntoConstraints

0.43

Activations Density 0.110%

technical terms and concepts related to transformations and data structures

transformations, reference, architecture, problem, slots

No Comments

No Known Activations