INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Chall

-0.15

.MixedReality

-0.15

prene

-0.15

ibal

-0.14

ìĩ

-0.14

nore

-0.14

ql

-0.14

MW

-0.14

kish

-0.14

POSITIVE LOGITS

icus

0.16

antu

0.15

nts

0.15

REFERRED

0.14

serie

0.14

 ÑģÐ²Ñı

0.14

rei

0.14

izon

0.14

.GridView

0.14

ÅĤaw

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.