INDEX

Explanations

programming constructs related to data structure manipulation

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

adding items to collections

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

findpost

-0.41

 crossorigin

-0.39

verwijspagina

-0.36

 Мексичка

-0.35

 BrowserModule

-0.34

Gdx

-0.33

knex

-0.33

 nargin

-0.33

chakra

-0.32

 Gson

-0.32

POSITIVE LOGITS

 rhestr

0.48

enumi

0.47

 HasFactory

0.47

 مرئيه

0.47

 referenties

0.47

Linki

0.46

primento

0.46

LookAnd

0.45

ServletException

0.43

urlpatterns

0.43

Activations Density 0.057%

programming constructs related to data structure manipulation

adding items to collections

No Comments

No Known Activations