INDEX

Explanations

code related to data manipulation and visualization in Python

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 MediatR

-0.35

Itoa

-0.35

 llan

-0.34

abestanden

-0.33

éndolo

-0.33

 submenu

-0.32

 kapsul

-0.32

xase

-0.31

 renon

-0.31

 princesas

-0.30

POSITIVE LOGITS

 numpy

0.84

TypedDataSet

0.77

mybatisplus

0.71

 jsPsych

0.69

 pylab

0.63

GEBURTSDATUM

0.59

numpy

0.59

setVerticalGroup

0.58

 فريبيس

0.56

 صوتيه

0.54

Activations Density 0.198%

code related to data manipulation and visualization in Python

No Comments

No Known Activations