INDEX

Explanations

phrases indicating quantities or measurements related to topics

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

antan

-0.18

&a

-0.17

onu

-0.16

ait

-0.15

ied

-0.15

uyáº¿t

-0.15

Î³ÏĮ

-0.14

emmel

-0.14

elf

-0.14

frey

-0.13

POSITIVE LOGITS

 nowhere

0.35

 bounds

0.26

 sight

0.24

 reach

0.23

 necessity

0.21

town

0.21

 Bounds

0.20

 pocket

0.20

 sheer

0.19

 town

0.19

Activations Density 0.055%

phrases indicating quantities or measurements related to topics

No Comments

No Known Activations