INDEX

Explanations

concepts related to file sharing, community structures, and interest extraction algorithms

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

 satur

-0.14

pic

-0.14

 exercises

-0.14

atur

-0.14

 Weinstein

-0.14

zar

-0.14

stras

-0.14

yses

-0.14

van

-0.13

ulas

-0.13

POSITIVE LOGITS

-based

0.24

-Based

0.22

GAN

0.21

Net

0.21

nets

0.20

based

0.19

 Nets

0.18

 nets

0.17

 models

0.17

onet

0.17

Activations Density 0.180%

concepts related to file sharing, community structures, and interest extraction algorithms

No Comments

No Known Activations