INDEX

Explanations

technical terms and concepts related to data and parameters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 neighborhoods

-1.29

 neighborhood

-1.26

Neighbors

-1.25

 theater

-1.24

neighborhood

-1.24

 neighbors

-1.21

 favor

-1.19

 Neighbors

-1.17

Neighborhood

-1.16

 flavorful

-1.15

POSITIVE LOGITS

Whilst

1.57

 Whilst

1.51

 whilst

1.34

initialise

1.31

 organisers

1.31

 organiser

1.29

 organising

1.25

COLOUR

1.25

 organisation

1.24

 realising

1.24

Activations Density 2.356%

technical terms and concepts related to data and parameters

No Comments

No Known Activations