INDEX

Explanations

references to Mexico and its related terms

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 للاسماء

-0.56

 Stringer

-0.54

urance

-0.54

 Fournier

-0.54

pillars

-0.52

 Cárdenas

-0.51

ferrous

-0.50

 Chowdhury

-0.50

 Arnaud

-0.49

白色的

-0.49

POSITIVE LOGITS

 Mexico

1.05

Mexico

1.00

Mex

0.87

 mexico

0.83

 Mexican

0.83

 MEXICO

0.82

Mex

0.80

 Mexique

0.77

MEX

0.77

MEX

0.74

Activations Density 0.005%

references to Mexico and its related terms

No Comments

No Known Activations