INDEX

Explanations

moments or elements that occur behind the scenes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

etz

-0.18

Ã©t

-0.17

eros

-0.16

enga

-0.15

.nano

-0.15

shal

-0.14

owns

-0.14

ib

-0.14

ative

-0.14

zt

-0.14

POSITIVE LOGITS

 closed

0.29

-the

0.28

 schedule

0.25

 Closed

0.24

 bars

0.23

 Enemy

0.23

 Bars

0.22

Closed

0.22

closed

0.21

Activations Density 0.018%

moments or elements that occur behind the scenes

No Comments

No Known Activations