INDEX

Explanations

occurrences of system or programming-related terms, particularly in a technical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 insuffisamment

-0.73

 tartalomajánló

-0.59

fjspx

-0.58

DockStyle

-0.56

EXTERN

-0.54

 stanovnika

-0.54

AnchorStyles

-0.53

 chi̍t

-0.53

liferay

-0.52

 SHARE

-0.51

POSITIVE LOGITS

CodeAttribute

1.51

TypedDataSet

0.81

GeneratedCode

0.67

CloseOperation

0.66

providedIn

0.62

ADELPHIA

0.61

CodeDom

0.58

出版年

0.57

Geplaatst

0.57

setupUi

0.55

Activations Density 0.036%

occurrences of system or programming-related terms, particularly in a technical context

No Comments

No Known Activations