INDEX

Explanations

the word "best" in various contexts related to quality or superiority

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

hip

-0.15

ned

-0.15

enties

-0.15

umas

-0.15

udeau

-0.14

llib

-0.14

itÃ©

-0.14

ittest

-0.13

WEB

-0.13

POSITIVE LOGITS

owing

0.24

seller

0.23

-selling

0.20

-case

0.20

-known

0.19

åĩĨ

0.18

jamin

0.17

/fast

0.17

ow

0.17

wick

0.17

Activations Density 0.058%

the word "best" in various contexts related to quality or superiority

No Comments

No Known Activations