INDEX

Explanations

references to political events or statements involving authority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 25-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_25/width_16k/average_l0_116

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

 Oklahoma

-0.72

Oklahoma

-0.64

 Tijuana

-0.61

CodedInputStream

-0.57

 Saskatchewan

-0.56

 Hanford

-0.55

Sask

-0.55

 Guadalajara

-0.53

 Okla

-0.53

 boomer

-0.52

POSITIVE LOGITS

 Zanzibar

0.96

 Tanzania

0.90

 Maldives

0.89

 Malawi

0.89

 Seychelles

0.81

 Dodo

0.81

 Mauritius

0.80

Sey

0.78

Tanzania

0.78

Mauri

0.78

Activations Density 0.096%

references to political events or statements involving authority

No Comments

No Known Activations