INDEX

Explanations

capital letters and formatted text typically found in headings or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 Outbreak

-0.59

 poffible

-0.59

 chofe

-0.57

 raiſ

-0.57

 ytter

-0.55

 bezeichneter

-0.55

 iconFacebook

-0.55

 kere

-0.55

 不锈钢

-0.55

 sais

-0.55

POSITIVE LOGITS

ThroughAttribute

0.65

省市镇

0.62

tagext

0.59

parsedMessage

0.59

WebElementEntity

0.57

сшта

0.57

0.56

COUVER

0.54

0.53

̣n

0.53

Activations Density 0.291%

capital letters and formatted text typically found in headings or titles

No Comments

No Known Activations