INDEX

Explanations

conditional phrases related to events and actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

are

-0.18

 ought

-0.17

 doit

-0.16

ieron

-0.16

ÏĥÏĦÎ±

-0.16

is

-0.15

was

-0.15

ãĥ¼ãĥĨãĤ£

-0.15

ilio

-0.15

adol

-0.14

POSITIVE LOGITS

fos

0.41

fos

0.32

 fuera

0.29

 haya

0.22

isse

0.22

uisse

0.22

 foss

0.22

hub

0.22

asse

0.22

 olsun

0.20

Activations Density 0.020%

conditional phrases related to events and actions

No Comments

No Known Activations