INDEX

Explanations

references to various unspecified objects or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 mères

-0.94

 larmes

-0.83

 vœux

-0.78

 Deum

-0.76

 őket

-0.76

 genoux

-0.74

ύτε

-0.72

 AssemblyCompany

-0.71

 respectivos

-0.70

Опыт

-0.69

POSITIVE LOGITS

 things

3.55

Things

2.92

 Things

2.90

things

2.79

 THINGS

2.64

 cosas

2.11

 Dinge

2.06

 coisas

1.99

THINGS

1.96

 dingen

1.92

Activations Density 0.080%

references to various unspecified objects or concepts

No Comments

No Known Activations