INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ì°°

-0.14

oeff

-0.13

onte

-0.13

zeug

-0.13

oxel

-0.13

XPath

-0.13

ICA

-0.13

 Reco

-0.13

å¤§å°ı

-0.12

 ÐºÐ»Ð°Ð´

-0.12

POSITIVE LOGITS

 Dream

0.18

anja

0.17

 dream

0.17

OKIE

0.16

 Essence

0.15

 objectively

0.15

 essence

0.15

CrLf

0.15

dream

0.15

 rodin

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.