INDEX

Explanations

references to the name "Nan" or related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

-0.17

iy

-0.15

 Western

-0.15

kl

-0.15

woods

-0.14

kÃ¡

-0.14

klÃ¤

-0.14

á»§ng

-0.14

yes

-0.14

POSITIVE LOGITS

otechnology

0.21

oseconds

0.20

igans

0.18

DED

0.17

ette

0.16

jing

0.16

ncy

0.16

aim

0.15

_expand

0.15

alyze

0.15

Activations Density 0.005%

references to the name "Nan" or related terms

No Comments

No Known Activations