INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ÅĤaw

-0.15

ãģ¦ãģĦ

-0.14

 sonuc

-0.14

obe

-0.14

 Shar

-0.14

Permanent

-0.14

 ìĹ¬ëŁ¬ë¶Ħ

-0.14

.generated

-0.14

Ç

-0.14

preload

-0.14

POSITIVE LOGITS

ContextHolder

0.15

Äįi

0.14

ullo

0.13

ainers

0.13

 Hind

0.13

eve

0.13

 Turing

0.13

odont

0.13

cona

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.