INDEX

Explanations

terms related to wide-ranging or broad classifications and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

WebVitals

-0.49

новништво

-0.44

 AssemblyCulture

-0.43

UrlResolution

-0.42

Slf

-0.41

 Corpor

-0.40

EndGlobalSection

-0.39

بوابة

-0.38

 صوتيه

-0.38

 <=",

-0.38

POSITIVE LOGITS

NSMutableArray

0.67

 broad

0.66

 broadly

0.64

 getWidth

0.64

 breadth

0.62

 broadest

0.60

Broad

0.60

 vast

0.59

broad

0.58

 spatial

0.57

Activations Density 0.470%

terms related to wide-ranging or broad classifications and concepts

No Comments

No Known Activations