INDEX

Explanations

instances of the word "new" and class definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

gew

-0.58

hassee

-0.54

はじめに

-0.54

боль

-0.52

fun

-0.51

Wiel

-0.49

 JUGA

-0.49

}{||

-0.47

func

-0.47

 cipolla

-0.47

POSITIVE LOGITS

 propOrder

0.87

]--;

0.85

 wireType

0.77

 autorytatywna

0.74



0.70

 مواليد

0.70

KEYCODE

0.69

new

0.69

awtextra

0.68

oa̍t

0.68

Activations Density 0.059%

instances of the word "new" and class definitions

No Comments

No Known Activations