INDEX

Explanations

references to something being specifically noted or highlighted

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 this

-0.16

ÑĪÐµ

-0.15

velope

-0.14

iene

-0.14

ucene

-0.13

Ãºc

-0.13

oque

-0.13

nga

-0.13

ahlen

-0.13

ificio

-0.13

POSITIVE LOGITS

URA

0.17

ï¸

0.16

isd

0.16

ura

0.16

isin

0.15

profil

0.15

isy

0.14

idot

0.14

earable

0.14

olin

0.13

Activations Density 0.066%

references to something being specifically noted or highlighted

No Comments

No Known Activations