INDEX

Explanations

references to historical events and memorials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

ifar

-0.17

 Aviv

-0.16

Dah

-0.16

Sir

-0.15

iland

-0.15

oco

-0.15

lag

-0.15

ìķĶ

-0.14

Sir

-0.14

lehem

-0.14

POSITIVE LOGITS

 Asian

0.30

Asian

0.26

 Oriental

0.26

 Asians

0.25

 orient

0.23

 asian

0.22

Asi

0.22

 Orient

0.22

orient

0.21

 Chin

0.21

Activations Density 0.129%

references to historical events and memorials

No Comments

No Known Activations