INDEX

Explanations

references to publication years and citation details in academic contexts

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

owo

-0.18

oux

-0.17

krit

-0.16

inoa

-0.15

outh

-0.15

yaml

-0.15

hift

-0.15

ingroup

-0.15

idot

-0.14

catch

-0.14

POSITIVE LOGITS

udiant

0.18

uld

0.17

alc

0.17

ilm

0.17

agh

0.17

strate

0.16

phant

0.16

ascript

0.16

perse

0.15

aÃ±

0.15

Activations Density 0.033%

references to publication years and citation details in academic contexts

No Comments

No Known Activations