INDEX

Explanations

references to a specific film or title in a predictable manner

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

IPH

-0.15

æ¼«

-0.15

ween

-0.14

 ?>"/>↵

-0.14

 Abrams

-0.14

etty

-0.14

ipop

-0.14

agal

-0.14

uncture

-0.14

sik

-0.13

POSITIVE LOGITS

-mode

0.15

acom

0.15

orem

0.15

oya

0.15

oria

0.15

-spin

0.14

plural

0.14

allas

0.14

anth

0.14

ombre

0.14

Activations Density 0.091%

references to a specific film or title in a predictable manner

No Comments

No Known Activations