INDEX

Explanations

mathematical notation or symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

uling

-0.17

ients

-0.16

roperties

-0.15

ÙģÙĦ

-0.14

Gap

-0.14

fÃ¼hr

-0.14

iper

-0.14

åĪ©

-0.14

iser

-0.14

 Romeo

-0.13

POSITIVE LOGITS

rm

0.21

ghan

0.19

cal

0.18

rm

0.18

ectl

0.15

.updateDynamic

0.15

/script

0.15

RM

0.15

_rm

0.15

 Burr

0.15

Activations Density 0.021%

mathematical notation or symbols

No Comments

No Known Activations