INDEX

Explanations

mathematical notations and structures commonly used in equations

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

lander

-0.16

Tos

-0.15

ish

-0.14

tte

-0.14

.BackgroundColor

-0.14

rss

-0.13

elik

-0.13

ist

-0.13

ict

-0.13

POSITIVE LOGITS

lef

0.17

 BEGIN

0.16

 Begin

0.16

begin

0.16

 âĪĢ

0.15

lim

0.15

-begin

0.15

 begin

0.15

Activations Density 0.123%

mathematical notations and structures commonly used in equations

No Comments

No Known Activations