INDEX

Explanations

phrases and data related to statistics and reports indicating trends

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

 ÚĨÙĩ

-0.15

geb

-0.14

ë¡¤

-0.14

.what

-0.14

 WHETHER

-0.14

allee

-0.13

ÙĨÙĪØ¹

-0.13

à¸³

-0.13

ãģ©ãģĨ

-0.13

_PHP

-0.13

POSITIVE LOGITS

 there

0.31

 only

0.29

 fewer

0.28

 slightly

0.27

 overall

0.25

only

0.25

there

0.24

 Only

0.24

 close

0.24

 altogether

0.23

Activations Density 0.153%

phrases and data related to statistics and reports indicating trends

No Comments

No Known Activations