INDEX

Explanations

issues related to malfunctioning or broken systems and their resolutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

arten

-0.17

-0.14

awaii

-0.14

agen

-0.14

lasses

-0.14

ewe

-0.13

USR

-0.13

 perl

-0.13

 linen

-0.13

POSITIVE LOGITS

akt

0.16

Ã¶s

0.15

feit

0.15

imd

0.15

hte

0.15

 tarz

0.14

cred

0.14

Ã¨ge

0.14

ÐµÑģÑĮ

0.14

æ²¢

0.14

Activations Density 0.480%

issues related to malfunctioning or broken systems and their resolutions

No Comments

No Known Activations