INDEX

Explanations

numerical data and mathematical expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Burke

-0.14

unma

-0.14

 numbering

-0.14

 hamburger

-0.13

(?)

-0.13

áº¥y

-0.13

 Kelly

-0.13

 letter

-0.13

odium

-0.13

POSITIVE LOGITS

RAINT

0.16

utral

0.16

uhn

0.15

lsen

0.15

itational

0.14

itud

0.14

 chests

0.14

.gs

0.14

LOCKS

0.14

ecast

0.14

Activations Density 0.006%

numerical data and mathematical expressions

No Comments

No Known Activations