INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

 promin

-0.18

 outlier

-0.15

 gran

-0.15

Abb

-0.14

 televis

-0.14

 UIScreen

-0.14

chw

-0.14

 seperate

-0.14

 seper

-0.14

eb

-0.14

POSITIVE LOGITS

 Romanian

0.25

Äĥ

0.23

cea

0.22

 Romania

0.22

 pentru

0.19

ilor

0.18

 ÈĻi

0.17

Roman

0.17

 Ã®n

0.17

nesc

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.