INDEX

Explanations

references to geographical locations and names associated with places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

rizona

-0.68

oa̍t

-0.65

 Kabul

-0.58

+:+

-0.57

 vVar

-0.57

 Leid

-0.54

 تضيفلها

-0.53

 utafitiHapana

-0.51

XmlAccessType

-0.50

 "..\..\..\

-0.50

POSITIVE LOGITS

 Southampton

0.78

 Islanders

0.71

Southampton

0.70

 Monmouth

0.62

 Suffolk

0.58

 Freehold

0.58

 Lakewood

0.58

 Staten

0.55

 Islander

0.54

 Holm

0.53

Activations Density 0.250%

references to geographical locations and names associated with places

No Comments

No Known Activations