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

–

-0.16

sed

-0.15

des

-0.15

ch

-0.15

 Ramp

-0.14

 synthesis

-0.14

 indent

-0.14

 whatsapp

-0.14

mo

-0.14

¬ģ

-0.14

POSITIVE LOGITS

/Dk

0.15

 CONDITION

0.15

OKIE

0.15

intl

0.15

asca

0.15

áº£nh

0.14

ernals

0.14

 Filme

0.14

.grpc

0.14

/left

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.