INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

icker

-0.19

Ã¡p

-0.18

ilde

-0.16

ãĥªãĥ¼ãĤº

-0.15

ABCDEFGHIJKLMNOP

-0.15

ABCDEFGHI

-0.15

.plus

-0.14

ãĢħ

-0.14

egie

-0.14

?url

-0.14

POSITIVE LOGITS

low

0.15

encil

0.14

.Atomic

0.14

»

0.14

LOW

0.14

angel

0.14

-low

0.14

LOW

0.14

 gang

0.14

FT

0.14

Activations Density 0.114%

No Comments

No Known Activations