INDEX

Explanations

prominent numerical values and references in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

rub

-0.17

à¤¾à¤®à¤¨

-0.15

lei

-0.14

luck

-0.14

anos

-0.14

och

-0.13

eer

-0.13

æ¨¹

-0.13

mess

-0.13

erre

-0.13

POSITIVE LOGITS

 Baker

0.15

Rox

0.14

inand

0.14

PPP

0.13

ettel

0.13

projection

0.13

iegel

0.13

keh

0.13

enders

0.13

acy

0.13

Activations Density 0.080%

prominent numerical values and references in the text

No Comments

No Known Activations