INDEX

Explanations

references to historical figures and events, particularly related to specific monarchs and their reigns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istory

-0.16

abic

-0.15

.fac

-0.14

usunda

-0.14

ä¾į

-0.13

eer

-0.13

Î±Î½Î´

-0.13

DA

-0.13

ute

-0.12

ears

-0.12

POSITIVE LOGITS

 himself

0.19

ÑĢÐ¾Ð²Ð¸Ñĩ

0.17

 Magnus

0.15

-dropdown

0.15

Ð´ÐµÐ¹ÑģÑĤÐ²

0.14

son

0.14

ROW

0.14

son

0.14

who

0.14

Overlap

0.14

Activations Density 0.095%

references to historical figures and events, particularly related to specific monarchs and their reigns

No Comments

No Known Activations