INDEX

Explanations

technical terms and coding elements related to data validation and structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

achs

-0.18

roys

-0.16

 temper

-0.15

eed

-0.14

ContentLoaded

-0.14

à¸±à¸§à¸£

-0.13

edl

-0.13

odox

-0.13

 PropertyChanged

-0.13

cru

-0.13

POSITIVE LOGITS

Dit

0.18

ophe

0.16

 similarly

0.15

ãģĭãģĹ

0.15

Ð¾ÑģÑĢÐµÐ´

0.15

ombs

0.15

allon

0.15

 likewise

0.14

Ã©f

0.14

lik

0.14

Activations Density 0.252%

technical terms and coding elements related to data validation and structure

No Comments

No Known Activations