INDEX

Explanations

mentions of names, specifically related to various cultural or regional contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ERCHANT

-0.20

bero

-0.19

aceous

-0.16

-lfs

-0.16

çİĩ

-0.15

rahim

-0.15

abic

-0.15

swire

-0.15

icycle

-0.15

ized

-0.15

POSITIVE LOGITS

0.23

0.22

lish

0.20

aÃ±

0.19

ering

0.19

erv

0.18

ires

0.18

hausen

0.18

iment

0.18

go

0.17

Activations Density 0.600%

mentions of names, specifically related to various cultural or regional contexts

No Comments

No Known Activations