INDEX

Explanations

concepts related to code structures and variable references

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

parameters, local variables, anonymous classes

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

出版年

-0.76

complexContent

-0.60

TRIBUN

-0.52

astify

-0.52

 surla

-0.50

存于互联网档案馆

-0.49

 JpaRepository

-0.48

tagHelperRunner

-0.48

tvguidetime

-0.47

UserScript

-0.47

POSITIVE LOGITS

encodeWith

0.39

ẹo

0.36

esía

0.35

 виправивши

0.35

 gemeenten

0.32

sendEmail

0.30

legungen

0.29

 gehouden

0.28

 Warszawa

0.27

staticmethod

0.27

Activations Density 0.100%

concepts related to code structures and variable references

parameters, local variables, anonymous classes

No Comments

No Known Activations