INDEX

Explanations

numeric identifiers and values

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

Negative Logits

ignant

-0.30

<h2>

-0.30

</h1>

-0.30

 Spence

-0.30

rom

-0.30

catat

-0.30

blo

-0.30

 بلند

-0.29

ch

-0.29

POSITIVE LOGITS

ValueStyle

0.85

 nahilalakip

0.81

 propOrder

0.72

 pinulongan

0.69

ſelf

0.69

 myſelf

0.68

 للاسماء

0.67

ſelves

0.66

دانشنامهٔ

0.64

点此举报

0.61

Activations Density 2.173%

numeric identifiers and values

No Known Activations