INDEX

Explanations

unique symbols or characters and words related to technology and system issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

FunctionFlags

-0.17

amarin

-0.15

ãĥ©ãĥ³ãĤ¹

-0.14

ëĤĺë¬´

-0.14

 ÑģÐ¸ÑĢ

-0.13

åĩºåı£

-0.13

ipse

-0.13

unkt

-0.13

okol

-0.13

rze

-0.13

POSITIVE LOGITS

 partition

0.29

 Partition

0.26

 factory

0.26

 partitions

0.26

 format

0.26

 formatting

0.25

 clean

0.25

 Factory

0.24

 formatted

0.24

Partition

0.23

Activations Density 0.012%

unique symbols or characters and words related to technology and system issues

No Comments

No Known Activations