INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 Grip

-0.15

unicode

-0.15

UDIO

-0.15

Gad

-0.14

Haz

-0.14

iza

-0.14

cer

-0.14

-0.13

POSITIVE LOGITS

lop

0.16

REEN

0.16

 ÙħØ§Ø¯

0.15

iyon

0.14

Å¡ti

0.14

 rewarded

0.14

adian

0.14

ÙĤÙħ

0.14

ioc

0.14

 Rack

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.