INDEX

Explanations

technical terms and conditions related to data processing and encoding in programming

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ancode

-0.15

eless

-0.15

ubbo

-0.14

eson

-0.14

ONO

-0.14

uetype

-0.14

ÑĤÑİ

-0.14

ê³¤

-0.13

.blur

-0.13

unas

-0.13

POSITIVE LOGITS

undy

0.15

 automatically

0.14

amil

0.14

adu

0.14

 Pond

0.14

Ð»Ð¸Ñı

0.14

dux

0.13

ãĥ³ãĥĳ

0.13

AMPLE

0.13

Activations Density 0.191%

technical terms and conditions related to data processing and encoding in programming

No Comments

No Known Activations