INDEX

Explanations

references to toys and trails

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

ViewInit

-0.52

sidemargin

-0.52

 AssemblyTitle

-0.43

 Dobson

-0.42

/*

-0.37

 Marsden

-0.35

حال

-0.34

 comigo

-0.34

 CURIAM

-0.33

 Eintritt

-0.33

POSITIVE LOGITS

toy

1.16

Toy

1.14

Toy

1.09

 toys

1.00

 trailer

0.98

toy

0.97

 Toys

0.96

TOY

0.93

 leak

0.91

 Trailer

0.90

Activations Density 0.058%

references to toys and trails

No Comments

No Known Activations