INDEX

Explanations

No Explanations Found

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

asco

-0.15

svp

-0.15

vetica

-0.14

 backwards

-0.14

stvÃŃ

-0.14

 Merk

-0.14

Ð¾Ñī

-0.13

ä¹¾

-0.13

ctica

-0.13

 Tracy

-0.13

POSITIVE LOGITS

'gc

0.16

 Sith

0.15

aida

0.15

 Carrier

0.15

 Lump

0.15

éģİ

0.14

ez

0.14

unded

0.14

à¸¸à¸ĩ

0.14

shr

0.14

Activations Density 0.758%

No Comments

No Known Activations