INDEX

Explanations

programming class definitions and related structures

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

Code, paths, or filenames

np_acts-logits-general · gemini-2.0-flash

<start_of_turn> user

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 milagro

-0.36

 Transporte

-0.35

 coaching

-0.33

 Geografía

-0.33

recognized

-0.32

 practicing

-0.32

 syke

-0.32

 Walkover

-0.32

 libros

-0.31

bemos

-0.31

POSITIVE LOGITS

 localObject

0.81

CppMethod

0.80

########.

0.75

 surla

0.71

 uLocal

0.69

mobileqq

0.68

migrationBuilder

0.66

 CreateTagHelper

0.65

endphp

0.60

WEBPACK

0.60

Activations Density 0.124%

programming class definitions and related structures

Code, paths, or filenames

<start_of_turn> user

No Comments

No Known Activations