INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

mall

-0.15

ika

-0.15

sed

-0.14

isel

-0.14

ework

-0.14

 fade

-0.14

odal

-0.14

sig

-0.13

 Latter

-0.13

POSITIVE LOGITS

SYS

0.14

íĺ

0.14

.Module

0.14

.gdx

0.14

ÑĸÐ»Ð¸

0.13

Ø®Ùħ

0.13

'gc

0.13

Ã¤ge

0.13

Ú¯Ø±Ø¯

0.13

:async

0.13

Activations Density 0.023%

No Comments

No Known Activations