INDEX

Explanations

blocks of code or comments in programming or legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 XCTest

-0.70

Efq

-0.66

ſelf

-0.65

<>();

-0.64

 UnityEditor

-0.64

 Monfieur

-0.62

 Audiodateien

-0.59

Etr

-0.58

ieszcz

-0.58

 preocupes

-0.58

POSITIVE LOGITS

mobileqq

0.68

TemporalType

0.64

GOTREF

0.62

 <<<<<<<<<<<<<<

0.62

 TType

0.60

 <=",

0.59

 nahilalakip

0.59

✨:

0.58

 autorytatywna

0.57

 defaultstate

0.57

Activations Density 0.049%

blocks of code or comments in programming or legal contexts

No Comments

No Known Activations