INDEX

Explanations

symbols or characters relating to the representation of numerical data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

iÅ¡

-0.15

lio

-0.14

adesh

-0.14

venir

-0.14

 Basket

-0.14

bes

-0.14

evice

-0.13

ebo

-0.13

lobs

-0.13

lem

-0.13

POSITIVE LOGITS

fold

0.15

rese

0.14

endale

0.14

vor

0.14

allon

0.14

aign

0.13

ucher

0.13

plorer

0.13

 ëĭ¤

0.13

routeParams

0.13

Activations Density 0.011%

symbols or characters relating to the representation of numerical data

No Comments

No Known Activations