INDEX

Explanations

keywords and identifiers in programming code

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

code constructs and declarations

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

 betweenstory

-0.63

новниш

-0.57

 Réponses

-0.57

 pinulongan

-0.50

تقاوى

-0.50

 also

-0.47

RegressionTest

-0.46

!*\

-0.46

 Meksiku

-0.45

FailureListener

-0.45

POSITIVE LOGITS

 ब्रेकडाउन

0.59

WriteTagHelper

0.46

SequentialGroup

0.45

 للمعارف

0.45

SerializedSize

0.42

WithIOException

0.41

urlpatterns

0.41

 henvisninger

0.41

varande

0.40

Արտաքին

0.40

Activations Density 0.003%

keywords and identifiers in programming code

code constructs and declarations

No Comments

No Known Activations