INDEX

Explanations

promotional and descriptive language related to products and features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

Referanser

-0.39

 Steine

-0.38

 albeit

-0.37

albeit

-0.36

 admittedly

-0.33

Sadly

-0.33

 AssemblyTitle

-0.33

 Ganze

-0.32

 generally

-0.32

"]);

-0.31

POSITIVE LOGITS

WireFormatLite

0.65

 your

0.64

 يتيمه

0.62

PerformLayout

0.54

Accessory

0.54

 pleaſure

0.53

 ваших

0.53

 effortlessly

0.52

 gourmet

0.50

your

0.50

Activations Density 0.166%

promotional and descriptive language related to products and features

No Comments

No Known Activations