INDEX

Explanations

mathematical and legal terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

/Z

-0.16

/qu

-0.15

kp

-0.15

.tt

-0.15

ecal

-0.14

tik

-0.14

TRL

-0.14

quip

-0.14

enheim

-0.14

ãĤ¤ãĤ¯

-0.13

POSITIVE LOGITS

Ä±z

0.20

à¤ľ

0.20

iz

0.19

 zaman

0.18

ï¼

0.18

Âłz

0.17

 zone

0.17

ze

0.17

uz

0.17

zaz

0.17

Activations Density 0.074%

mathematical and legal terminology

No Comments

No Known Activations