INDEX

Explanations

references to various types of cells and their characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 Plin

-0.75

hoga

-0.69

 izvē

-0.69

 SPIE

-0.65

parseLong

-0.64

 valmis

-0.64

IRQ

-0.64

 Dinah

-0.62

 Hogarth

-0.62

 Swindon

-0.62

POSITIVE LOGITS

 cell

2.69

 cells

2.64

 Cell

2.57

 CELL

2.56

cell

2.54

 Cells

2.53

Cell

2.48

CELL

2.45

Cells

2.45

cells

2.43

Activations Density 0.138%

references to various types of cells and their characteristics

No Comments

No Known Activations