INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

ToF

-0.16

ergus

-0.15

Lorem

-0.15

chwitz

-0.15

Crop

-0.15

 credited

-0.14

UTERS

-0.14

ju

-0.14

 Grim

-0.14

 Lantern

-0.14

POSITIVE LOGITS

IPS

0.16

reeze

0.15

usement

0.15

azzi

0.15

Ð°Ð½Ð¸ÑĨ

0.15

abay

0.15

arm

0.14

ash

0.14

rella

0.14

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.