INDEX

Explanations

phrases that indicate conclusions or outcomes of discussions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

nette

-0.17

loat

-0.16

deaux

-0.16

ocale

-0.15

TestFixture

-0.14

 Sons

-0.14

elda

-0.14

aÄĩ

-0.13

oria

-0.13

ication

-0.13

POSITIVE LOGITS

 boil

0.40

 boiled

0.36

 reduced

0.35

 boils

0.35

 redu

0.32

 reduces

0.31

 boiling

0.31

 reduce

0.30

Bo

0.29

bo

0.29

Activations Density 0.140%

phrases that indicate conclusions or outcomes of discussions

No Known Activations