INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Nem

-0.14

odel

-0.14

lops

-0.14

/features

-0.14

 trimest

-0.14

ops

-0.13

 Labour

-0.13

Ð±Ð¸ÑĢÐ°

-0.13

ifr

-0.13

 glob

-0.13

POSITIVE LOGITS

Ø®ÙĪ

0.15

ymb

0.15

ihat

0.14

USR

0.14

yii

0.14

yy

0.14

 å¹³æĸ¹

0.14

Jad

0.14

 Blasio

0.14

adc

0.13

Activations Density 0.204%

No Comments

No Known Activations