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

rar

-0.15

istrat

-0.13

NgModule

-0.13

phem

-0.13

jar

-0.13

áº¡ng

-0.13

Ð¡Ð¢

-0.13

pNet

-0.12

OrDefault

-0.12

iku

-0.12

POSITIVE LOGITS

als

0.15

 Seks

0.15

 Maurice

0.15

BLEM

0.14

lesc

0.14

bla

0.14

acy

0.14

ä»°

0.13

.ISupportInitialize

0.13

 Gros

0.13

Activations Density 0.035%

No Comments

No Known Activations