INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

Äįka

-0.16

ngrx

-0.15

figcaption

-0.15

rab

-0.15

uto

-0.14

 Playboy

-0.14

SED

-0.14

 ãĢ

-0.14

chia

-0.14

 ÎºÎ±Î¸

-0.14

POSITIVE LOGITS

urf

0.19

ework

0.15

.fill

0.14

ergus

0.14

ifa

0.14

rie

0.14

ToFront

0.14

 Sanford

0.14

 ordinary

0.13

iders

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.