INDEX

Explanations

references to "top" and "bottom" in relation to various contexts or subjects

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rition

-0.62

 tubercle

-0.59

casila

-0.57

tubers

-0.56

 pleaſure

-0.56

rohr

-0.54

 quæ

-0.54

uparrow

-0.53

 Varanasi

-0.53

 axillary

-0.53

POSITIVE LOGITS

 Мексичка

0.79

adpleegd

0.73

 utafitiHapana

0.69

parseColor

0.67

تقاوى

0.66

 ویکی‌پدیای

0.65

شهاد

0.65

Tembelea

0.65

 MainAxisSize

0.64

onAttach

0.64

Activations Density 0.204%

references to "top" and "bottom" in relation to various contexts or subjects

No Comments

No Known Activations