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

pig

-0.19

ache

-0.17

ãĥ¼ãĤº

-0.17

uner

-0.16

inia

-0.16

Ð¾Ð¿Ð¸Ñģ

-0.15

éĩ

-0.15

exo

-0.15

taj

-0.15

 Ð½Ð°Ð»

-0.15

POSITIVE LOGITS

ieren

0.16

midt

0.16

inf

0.15

Inf

0.15

 style

0.14

iale

0.13

Rud

0.13

 effect

0.13

»

0.13

.inf

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.