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

eb

-0.17

ves

-0.14

loud

-0.14

 behalf

-0.14

uy

-0.13

 suppl

-0.13

.Tables

-0.13

æĸ¯

-0.13

tac

-0.13

elt

-0.13

POSITIVE LOGITS

RootElement

0.15

HU

0.15

erli

0.14

ABEL

0.14

ruba

0.14

 ValueEventListener

0.14

Î³ÎµÎ¹

0.13

 guarante

0.13

 sáºµn

0.13

qrt

0.13

Activations Density 0.049%

No Comments

No Known Activations