INDEX

Explanations

annotations or references in text, particularly those starting with '@'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

iske

-0.18

akra

-0.16

pter

-0.15

icing

-0.14

unami

-0.14

otti

-0.14

.Actions

-0.14

 Manager

-0.13

thon

-0.13

Manager

-0.13

POSITIVE LOGITS

article

0.30

 article

0.29

Article

0.26

misc

0.25

 misc

0.25

.article

0.25

 Article

0.24

_article

0.24

-article

0.22

 Misc

0.20

Activations Density 0.004%

annotations or references in text, particularly those starting with '@'

No Comments

No Known Activations