INDEX

Explanations

references to hyperlinks or web links

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ysa

-0.18

zos

-0.17

endir

-0.16

ient

-0.15

ventus

-0.15

 Instance

-0.14

Zus

-0.14

enha

-0.14

 predecess

-0.14

AO

-0.14

POSITIVE LOGITS

ometr

0.16

ÑĤÑİ

0.15

obus

0.15

OperationException

0.15

scalar

0.15

pons

0.15

ØªÙĪØ±

0.15

lessly

0.14

velope

0.14

ulong

0.14

Activations Density 0.001%

references to hyperlinks or web links

No Comments

No Known Activations