INDEX

Explanations

details related to numerical data, schedules, and time commitments

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

assen

-0.14

Ã¤dchen

-0.14

Ø·

-0.13

á»ĥn

-0.12

tone

-0.12

Magnitude

-0.12

Ãªn

-0.12

 ØµÙĨ

-0.12

Ø¶

-0.12

edException

-0.12

POSITIVE LOGITS

 weeks

0.23

 consecutive

0.23

-days

0.22

 hours

0.22

-hour

0.21

 days

0.21

hrs

0.21

hrs

0.21

(one

0.21

 full

0.20

Activations Density 0.106%

details related to numerical data, schedules, and time commitments

No Comments

No Known Activations