INDEX

Explanations

questions and requests for help or instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

WriteTagHelper

-0.60

 Camb

-0.48

curricular

-0.47

fromCharCode

-0.46

cro

-0.44

nor

-0.44

©

-0.44

大和

-0.43

pholes

-0.43

 bode

-0.43

POSITIVE LOGITS

 houſe

0.78

 perſon

0.76

 myſelf

0.76

Efq

0.68

 ſta

0.68

 ſtate

0.68

 diſt

0.67

GraphicsUnit

0.67

 Monfieur

0.66

enterOuterAlt

0.66

Activations Density 0.341%

questions and requests for help or instructions

No Comments

No Known Activations