INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 IMDb

-0.15

itemap

-0.14

 Harmony

-0.14

Ã³ng

-0.14

è«

-0.14

inne

-0.14

subpackage

-0.13

à¹ģà¸«à¸¥

-0.13

 Harmon

-0.13

LastError

-0.13

POSITIVE LOGITS

 letter

0.21

aut

0.18

 Letter

0.18

beta

0.17

 paper

0.17

 global

0.17

letter

0.17

 Paper

0.17

Letter

0.17

global

0.16

Activations Density 0.000%

No Known Activations

This feature has no known activations.