INDEX

Explanations

references to box office performance and success

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

antu

-0.17

iba

-0.15

encer

-0.15

BOX

-0.14

kw

-0.14

à¸³

-0.14

akin

-0.14

Ð¿ÑĢÐ°

-0.14

ahr

-0.14

bens

-0.14

POSITIVE LOGITS

-office

0.27

 office

0.25

office

0.24

 Office

0.22

_office

0.20

Office

0.20

 BÃ¼

0.18

score

0.18

wick

0.18

scores

0.18

Activations Density 0.005%

references to box office performance and success

No Comments

No Known Activations