INDEX

Explanations

phrases related to performance and updates in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ÐµÐ½Ð·

-0.17

irit

-0.16

ixo

-0.15

_PLUS

-0.14

icip

-0.14

lam

-0.14

mon

-0.14

HH

-0.13

á»¯ng

-0.13

Til

-0.13

POSITIVE LOGITS

XE

0.16

omite

0.15

eus

0.15

aload

0.15

ubber

0.14

INV

0.14

pler

0.14

adero

0.14

adr

0.14

chg

0.14

Activations Density 0.133%

phrases related to performance and updates in various contexts

No Comments

No Known Activations