INDEX

Explanations

technical references to programming and coding practices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

-0.18

-0.14

.train

-0.14

ipples

-0.14

 Ð·Ð°ÐºÐ»Ð°Ð´

-0.14

thon

-0.14

emption

-0.13

 Retrie

-0.13

ãĥ³ãĥ

-0.13

to

-0.13

POSITIVE LOGITS

weg

0.18

ovit

0.17

-scrollbar

0.16

itler

0.16

gli

0.15

eyJ

0.14

FindBy

0.14

gles

0.14

 Prem

0.14

oods

0.13

Activations Density 0.015%

technical references to programming and coding practices

No Comments

No Known Activations