INDEX

Explanations

references to guidance and educational resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

issent

-0.17

rish

-0.16

umer

-0.16

 overl

-0.15

asher

-0.14

fixtures

-0.14

exc

-0.14

Lab

-0.14

orce

-0.13

POSITIVE LOGITS

 anyone

0.29

 audiences

0.27

 readers

0.26

 anybody

0.26

 practicing

0.23

 practitioner

0.23

Anyone

0.23

 Anyone

0.23

 practition

0.23

 interested

0.23

Activations Density 0.210%

references to guidance and educational resources

No Comments

No Known Activations