INDEX

Explanations

references to large quantities or sizes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 41-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_41/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.41.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

 large

-0.63

large

-0.56

 small

-0.55

 pequeno

-0.53

 неболь

-0.52

 pequeño

-0.52

 smaller

-0.52

 little

-0.52

 pequena

-0.51

little

-0.50

POSITIVE LOGITS

 scale

0.84

 amounts

0.81

 quantities

0.77

scale

0.70

(>

0.68

 Scale

0.66

 numbers

0.64

 SCALE

0.63

 chunks

0.63

mouth

0.62

Activations Density 0.404%

references to large quantities or sizes

No Comments

No Known Activations