INDEX

Explanations

programming-related terminology and function calls

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

code and technical terms

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

Geplaatst

-0.34

SuccessListener

-0.34

basicConfig

-0.31

 تعدى

-0.31

তথ্যসূত্র

-0.30

 Wicidata

-0.30

 quoted

-0.30

ucker

-0.29

 undefined

-0.29

ˌ

-0.29

POSITIVE LOGITS

providedIn

0.66

 للمعارف

0.53

QMetaType

0.52

styleType

0.50

andExpect

0.48

__':

0.47

 Italijanski

0.45

ReusableCell

0.45

TagHelper

0.44

criminator

0.44

Activations Density 0.058%

programming-related terminology and function calls

code and technical terms

No Comments

No Known Activations