INDEX

Explanations

references to various types of 'cab' or 'cabin' related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

yk

-0.18

avian

-0.17

åłĤ

-0.17

ackson

-0.16

isans

-0.16

 smiles

-0.15

ÑģÑı

-0.15

 Hint

-0.15

_DECLS

-0.14

POSITIVE LOGITS

aret

0.36

oose

0.32

ernet

0.29

rio

0.27

ildo

0.25

oodle

0.24

anas

0.23

ecera

0.23

drivers

0.21

by

0.20

Activations Density 0.007%

references to various types of 'cab' or 'cabin' related terms

No Comments

No Known Activations