INDEX

Explanations

characters or symbols from non-Latin scripts

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

expandindo

-0.93

Portale

-0.84

Personendaten

-0.81

InitVars

-0.75

AddTagHelper

-0.74

 CreateTagHelper

-0.73

AnchorStyles

-0.72

ValueStyle

-0.72

.*")]

-0.66

MigrationBuilder

-0.66

POSITIVE LOGITS

 Myanmar

0.92

Myanmar

0.87

 Aung

0.74

gyi

0.72

aung

0.67

Zaw

0.67

Py

0.67

My

0.63

Ky

0.63

Win

0.62

Activations Density 0.044%

characters or symbols from non-Latin scripts

No Comments

No Known Activations