INDEX

Explanations

proper nouns, particularly names of people and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

mai

-0.14

vor

-0.14

ep

-0.13

raison

-0.13

endregion

-0.13

frog

-0.13

 Gordon

-0.13

ume

-0.13

Mai

-0.13

Â

-0.13

POSITIVE LOGITS

's

0.28

’s

0.25

ìĿĺ

0.20

çļĦ

0.19

çļĦæĥħ

0.19

'options

0.18

çļĦæīĭ

0.17

çļĦå°ı

0.16

’S

0.16

heimer

0.16

Activations Density 0.179%

proper nouns, particularly names of people and organizations

No Comments

No Known Activations