INDEX

Explanations

emphasis on significant size or scale

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

hnia

-0.60

 canali

-0.59

-0.58

-0.57

nav

-0.57

 eventualmente

-0.56

trab

-0.56

ka

-0.55

my

-0.54

POSITIVE LOGITS

 huge

2.69

huge

2.63

Huge

2.60

 Huge

2.45

 HUGE

2.33

 enormous

2.25

 énorme

2.03

 enorme

1.95

ormous

1.92

 gigantic

1.85

Activations Density 0.070%

emphasis on significant size or scale

No Comments

No Known Activations

emphasis on significant size or scale

No Comments

No Known Activations