INDEX

Explanations

references to visiting or accessing websites and related links

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

loom

-0.14

Ð¸Ð»Ð¾

-0.13

 Shore

-0.13

ervas

-0.13

rou

-0.13

hai

-0.13

leo

-0.13

postalcode

-0.13

setProperty

-0.13

wid

-0.13

POSITIVE LOGITS

0.15

/ion

0.15

 BaÄŁ

0.14

alara

0.14

ÑıÐ½

0.14

_published

0.14

uhn

0.13

 ResourceType

0.13

 Foley

0.13

erture

0.13

Activations Density 0.032%

references to visiting or accessing websites and related links

No Comments

No Known Activations