INDEX

Explanations

proper nouns, specifically names or references to individuals and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

raw

-0.50

Raw

-0.45

Tetra

-0.43

 Pratt

-0.42

RAW

-0.40

뀜

-0.40

 Tetra

-0.40

 sintético

-0.39

 Cran

-0.39

REQUIRE

-0.39

POSITIVE LOGITS

脚注の使い方

0.57

Tikang

0.47



0.47

########.

0.47

 okuyayım

0.46

WithIOException

0.44

+:+

0.43

 cherchés

0.43

Kjelder

0.41

devamını

0.41

Activations Density 0.382%

proper nouns, specifically names or references to individuals and places

No Comments

No Known Activations