INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

.cloudflare

-0.15

AML

-0.15

emme

-0.14

ouch

-0.14

ibo

-0.14

ãĤĩ

-0.14

uzzi

-0.13

 Fritz

-0.13

_odd

-0.13

opleft

-0.13

POSITIVE LOGITS

duk

0.17

comings

0.17

 cease

0.17

Ce

0.16

 conditioning

0.15

 names

0.15

ce

0.15

 pleasant

0.15

adv

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.