INDEX

Explanations

special characters or symbols often related to formatting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

asher

-0.16

Prostit

-0.15

ambi

-0.15

mail

-0.14

cano

-0.14

Å¡lo

-0.14

azard

-0.14

washer

-0.14

Mailer

-0.14

 Mail

-0.13

POSITIVE LOGITS

ned

0.21

 vast

0.21

 barn

0.20

pu

0.20

 gÃ¶

0.19

kes

0.19

ned

0.18

 bund

0.18

 bord

0.18

 petty

0.18

Activations Density 0.004%

special characters or symbols often related to formatting

No Comments

No Known Activations