INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 nackte

-0.18

zas

-0.15

.scalablytyped

-0.14

.bio

-0.14

arem

-0.14

curity

-0.13

obe

-0.13

chwitz

-0.13

zÅĳ

-0.13

formance

-0.13

POSITIVE LOGITS

ÏĦÎ¹ÎºÎ¿

0.15

 Stark

0.15

 rhyme

0.14

éĽª

0.14

Abr

0.14

oz

0.14

 Britt

0.14

 è¦

0.13

acos

0.13

uce

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.