INDEX

Explanations

references to statistical data and study results

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

 recent

-0.19

 often

-0.15

otre

-0.15

 sometimes

-0.14

 usually

-0.14

 ç¤º

-0.14

zahl

-0.13

 recently

-0.13

 Showcase

-0.13

ÙħÙĪØ¯

-0.13

POSITIVE LOGITS

 Forty

0.29

 Sevent

0.28

 Thirty

0.28

 Fifty

0.27

Nin

0.27

 Twenty

0.26

 Twelve

0.25

Fif

0.25

 Mean

0.25

 Eight

0.24

Activations Density 0.054%

references to statistical data and study results

No Comments

No Known Activations