INDEX

Explanations

references to academic papers and their structure

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

in this paper

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

parsedMessage

-0.78

 CreateTagHelper

-0.77

 للمعارف

-0.68

Чыгана

-0.68

хьтан

-0.67

WriteLiteral

-0.64

Diweddarwch

-0.63

cestry

-0.62

ParallelGroup

-0.60

 Numerade

-0.60

POSITIVE LOGITS

 paper

0.42

 kveld

0.35

 article

0.34

HomeController

0.33

 Tode

0.32

 publicada

0.31

 przedmio

0.31

 discusses

0.31

msgSender

0.31

 morgon

0.31

Activations Density 0.014%

references to academic papers and their structure

in this paper

No Comments

No Known Activations

references to academic papers and their structure

in this paper

No Comments

No Known Activations