INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

orners

-0.16

ë¹Ļ

-0.15

ãĥ¼ãĥĬ

-0.15

taj

-0.15

owell

-0.14

apolis

-0.14

xad

-0.14

theid

-0.14

iland

-0.14

crest

-0.14

POSITIVE LOGITS

 terminal

0.23

terminal

0.20

 Terminal

0.19

 terminals

0.19

Terminal

0.18

 wire

0.18

 Gros

0.17

 ceiling

0.17

_terminal

0.17

 wires

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.