INDEX

Explanations

variables and parameters in code snippets

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

variable names and code syntax

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

OGND

-0.94

 Мексичка

-0.94

WebVitals

-0.94

WriteTagHelper

-0.90

Spoljašnje

-0.87

TypedDataSet

-0.83

bootstrapcdn

-0.78

EndGlobalSection

-0.78

 surla

-0.77

 المعيارى

-0.75

POSITIVE LOGITS

is

0.45

may

0.39

has

0.38

0.37

 represented

0.36

 parti

0.35

 properly

0.35

was

0.35

 officially

0.34

 held

0.34

Activations Density 0.027%

variables and parameters in code snippets

variable names and code syntax

No Comments

No Known Activations