INDEX

Explanations

terms related to data analysis and scientific research findings

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

quantifiable aspects of outcomes

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

 communiquez

-0.48

GEBURTSDATUM

-0.46

chauss

-0.40

 Taxation

-0.38

 taxing

-0.36

 Trademark

-0.36

Tax

-0.35

 football

-0.35

 roofing

-0.35

 locomo

-0.34

POSITIVE LOGITS

 thereof

0.58

principalColumn

0.57

intios

0.56

 ervan

0.54

ftagPool

0.53

それを

0.52

 them

0.52

SharedDtor

0.52

WithMany

0.51

 EconPapers

0.50

Activations Density 0.235%

terms related to data analysis and scientific research findings

quantifiable aspects of outcomes

No Comments

No Known Activations