INDEX

Explanations

references to any kind of group or collective entity

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

 spesso

-0.63

 sometimes

-0.62

 algumas

-0.62

 некоторые

-0.61

 many

-0.60

many

-0.60

 algunas

-0.60

 algunos

-0.59

 often

-0.59

 muitas

-0.58

POSITIVE LOGITS

 whatsoever

1.52

 imaginable

1.24

 conceivable

0.99

 apapun

0.92

 anywhere

0.84

anything

0.83

 anybody

0.83

 anyone

0.83

 Anywhere

0.79

anywhere

0.79

Activations Density 0.312%

references to any kind of group or collective entity

No Comments

No Known Activations