INDEX

Explanations

references to academic articles, including their publication details

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

 Mund

-0.15

ander

-0.15

izzie

-0.14

Ø§ÙĦÙĥ

-0.14

 skin

-0.13

tro

-0.13

uya

-0.13

och

-0.13

ï»¿↵↵

-0.13

ses

-0.13

POSITIVE LOGITS

odor

0.16

coc

0.15

ramework

0.14

éĬĢè¡Į

0.14

ç´

0.14

Cpp

0.14

 Rica

0.14

.restore

0.13

aleb

0.13

 hoáº¡ch

0.13

Activations Density 0.034%

references to academic articles, including their publication details

No Comments

No Known Activations