INDEX

Explanations

references to data analysis and correlation in scientific research

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

descriptions of figures and tables

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

AddTagHelper

-0.63

AntiForgeryToken

-0.59

MessageTagHelper

-0.59

ReusableCell

-0.58

帖最后由

-0.54

DoubleQuotes

-0.52

principalColumn

-0.52

новништво

-0.51

errHandler

-0.50

oneofs

-0.50

POSITIVE LOGITS

 GlobalKey

0.35

 повто

0.32

 repeat

0.32

 Réponses

0.32

}]

0.31

 Repeat

0.29

 reiterated

0.29

 dzięki

0.29

 mempel

0.28

veti

0.28

Activations Density 0.175%

references to data analysis and correlation in scientific research

descriptions of figures and tables

No Comments

No Known Activations