INDEX

Explanations

details related to specific historical figures and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

izza

-0.15

_reads

-0.15

abee

-0.14

ynam

-0.14

afone

-0.14

.cz

-0.13

aroo

-0.13

Adv

-0.13

 bure

-0.13

 atan

-0.13

POSITIVE LOGITS

thanks

0.17

 thanks

0.17

pur

0.16

Ø§ÙĪÛĮ

0.15

ODEV

0.15

æĲŀ

0.15

oose

0.14

fono

0.14

assel

0.14

ven

0.14

Activations Density 0.021%

details related to specific historical figures and events

No Comments

No Known Activations