INDEX

Explanations

terms related to optimization and maximizing outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

sworth

-0.19

ëĬ¥

-0.16

lasses

-0.15

Smy

-0.15

azt

-0.14

Î»Î¿Ïį

-0.14

ivar

-0.14

xbc

-0.14

ares

-0.14

POSITIVE LOGITS

/max

0.25

=max

0.22

/min

0.20

-eff

0.18

 chances

0.17

(Max

0.17

 effort

0.17

 maximize

0.17

imize

0.17

 maxim

0.16

Activations Density 0.046%

terms related to optimization and maximizing outcomes

No Comments

No Known Activations