INDEX

Explanations

phrases indicating the beginning stages or initial conditions of a situation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

eldorf

-0.15

Äįen

-0.14

odal

-0.14

zew

-0.14

Î»Ïī

-0.14

ournÃ©e

-0.14

ogene

-0.14

usercontent

-0.13

lene

-0.13

OLLOW

-0.13

POSITIVE LOGITS

 start

0.35

 starts

0.28

 Start

0.28

start

0.26

Start

0.26

 START

0.25

 started

0.24

-start

0.24

START

0.23

.start

0.22

Activations Density 0.018%

phrases indicating the beginning stages or initial conditions of a situation

No Comments

No Known Activations