INDEX

Explanations

mentions of "wiki" and related URLs, indicating a focus on Wikipedia references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

abestanden

-0.78

RenderAtEndOf

-0.76

saraba

-0.71

 виправивши

-0.71

webElement

-0.67

]").

-0.65

 externi

-0.65

.*")]

-0.63

windowFixed

-0.63

 AppColors

-0.62

POSITIVE LOGITS

wiki

1.14

Wiki

0.60

 referenties

0.58

wikipedia

0.58

Demografia

0.57

enumi

0.57

ţele

0.53

 Fick

0.53

 WaitForSeconds

0.53

 wiki

0.52

Activations Density 0.003%

mentions of "wiki" and related URLs, indicating a focus on Wikipedia references

No Comments

No Known Activations