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

instein

-0.15

 hers

-0.14

Ð·Ñĭ

-0.14

 Hers

-0.14

WWW

-0.14

dy

-0.13

stein

-0.13

.astype

-0.13

ATEST

-0.13

ì¶ľ

-0.13

POSITIVE LOGITS

DITION

0.17

ÙĪÙĨØ§

0.15

odge

0.15

ephy

0.14

oley

0.14

indsight

0.14

.tsv

0.14

okus

0.14

0.13

tÄ±r

0.13

Activations Density 0.783%

No Comments

No Known Activations