INDEX

Explanations

the identification and evaluation of resources or values in different 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

lect

-0.17

ises

-0.15

-tra

-0.15

aris

-0.15

ro

-0.15

<*>

-0.14

fty

-0.14

 ÑģÑĥÐ¿

-0.14

_AI

-0.14

ÑĢÐ¸Ð¹

-0.14

POSITIVE LOGITS

 danger

0.15

lal

0.15

HEY

0.15

ValuePair

0.14

à¸ļà¸£à¸£

0.14

edException

0.14

_fixture

0.14

yleft

0.14

éħ

0.14

 Marcus

0.13

Activations Density 0.187%

the identification and evaluation of resources or values in different contexts

No Comments

No Known Activations