INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

Gio

-0.15

zet

-0.15

ades

-0.14

 Vend

-0.14

ones

-0.14

izens

-0.14

uai

-0.14

pcs

-0.14

ntag

-0.13

nten

-0.13

POSITIVE LOGITS

uitka

0.19

cona

0.15

_trampoline

0.14

OTA

0.14

ãĥ¬ãĥ¼

0.13

âĤ¬âĦ¢

0.13

ãħĭãħĭãħĭãħĭ

0.13

ÑĢÐ¾Ð³ÑĢÐ°

0.13

ìĨį

0.13

osh

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.