INDEX

Explanations

instances of the term "name" and its variations within a text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 34-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_34/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.34.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

utnik

-0.38

reloadData

-0.37

 Deliber

-0.35

 vidare

-0.34

rebel

-0.33

stiff

-0.33

 JSTOR

-0.33

spectrum

-0.32

Sol

-0.30

 dalamnya

-0.30

POSITIVE LOGITS

plate

0.74

plates

0.71

 Name

0.70

 NAME

0.68

rungsseite

0.65

txtName

0.64

NAME

0.61

Surname

0.61

 Names

0.61

 names

0.61

Activations Density 0.126%

instances of the term "name" and its variations within a text

No Comments

No Known Activations