INDEX

Explanations

names and terms related to musicians or composers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

es

-0.28

ex

-0.26

ed

-0.25

em

-0.25

ess

-0.22

yonel

-0.22

elle

-0.22

el

-0.22

els

-0.21

esh

-0.21

POSITIVE LOGITS

ateral

0.24

abyrin

0.24

abyrinth

0.22

icious

0.20

ifestyles

0.20

ty

0.20

ts

0.20

mente

0.20

apsed

0.20

iferay

0.20

Activations Density 0.941%

names and terms related to musicians or composers

No Comments

No Known Activations