INDEX

Explanations

numerical data and statistical information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

.openConnection

-0.17

krom

-0.16

erable

-0.16

inja

-0.16

lymp

-0.15

á»įc

-0.15

 Increment

-0.15

.keyboard

-0.15

isl

-0.15

ANDLE

-0.15

POSITIVE LOGITS

 doubles

0.33

 singles

0.25

 Doub

0.25

 wild

0.24

IT

0.22

ITA

0.22

oubles

0.21

 Wild

0.19

 Challenger

0.19

 wildcard

0.19

Activations Density 0.033%

numerical data and statistical information

No Comments

No Known Activations