INDEX

Explanations

mathematical terms and operations related to equations and expressions

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

lanma

-0.17

Median

-0.16

æĭĶ

-0.15

median

-0.15

pong

-0.15

 ÑĢÐ°Ð·Ð¼

-0.14

isel

-0.14

ableView

-0.14

enet

-0.14

.opend

-0.13

POSITIVE LOGITS

 contributions

0.28

 loop

0.27

 corrections

0.25

-loop

0.25

 diagrams

0.24

 loops

0.23

 Loop

0.23

 contribution

0.23

loop

0.23

terms

0.22

Activations Density 0.018%

mathematical terms and operations related to equations and expressions

No Comments

No Known Activations