INDEX

Explanations

numeric data and formats

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

sterious

-0.76

談社

-0.73

AndEndTag

-0.72

transQ

-0.71

منابع

-0.70

 doInBackground

-0.70

 Mémoires

-0.70

RectangleBorder

-0.67

ToScroll

-0.67

 Monfieur

-0.67

POSITIVE LOGITS

prnewswire

0.51

Revenir

0.49

 Vikipedi

0.42

спеди

0.41

KeepAlive

0.40

selaer

0.40

Inc

0.40

 slalom

0.39

 loudest

0.38

Activations Density 0.557%

numeric data and formats

No Comments

No Known Activations