INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

translated

-0.18

 chalk

-0.15

IAS

-0.14

 printable

-0.14

Ð¸ÑĤÐ¾ÑĢ

-0.14

iefs

-0.14

haft

-0.14

gili

-0.13

ubah

-0.13

 Printable

-0.13

POSITIVE LOGITS

 Document

0.20

aket

0.19

bid

0.19

ATA

0.19

 native

0.18

Editing

0.18

CAT

0.18

Native

0.18

.native

0.18

CAT

0.17

Activations Density 0.020%

No Comments

No Known Activations