INDEX

Explanations

specific identifiers or alphanumeric strings, possibly related to coding or data attributes

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

PerformLayout

-0.82

DeleteBehavior

-0.73

parsedMessage

-0.69

mybatisplus

-0.68

reportWebVitals

-0.68

oredCriteria

-0.68

 Baillargeon

-0.67

]--;

-0.59

 ویکی‌پدی

-0.58

POSITIVE LOGITS

<bos>

0.58

gouv

0.45

peicher

0.41

タジー

0.40

ECC

0.40

ASMA

0.40

âce

0.39

ppelin

0.38

Ald

0.38

nitř

0.38

Activations Density 0.541%

specific identifiers or alphanumeric strings, possibly related to coding or data attributes

No Known Activations