INDEX

Explanations

numerical and percentage values

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

seau

-0.17

mpr

-0.15

mug

-0.15

Î¸Î»Î·

-0.14

 FRAME

-0.14

ceso

-0.13

EventManager

-0.13

 tile

-0.13

.UInt

-0.13

aus

-0.13

POSITIVE LOGITS

aturas

0.16

ero

0.15

ASA

0.14

Im

0.14

 grÃ¡tis

0.14

 deposit

0.14

eh

0.14

Pon

0.13

ruptcy

0.13

ach

0.13

Activations Density 0.011%

numerical and percentage values

No Comments

No Known Activations