INDEX

Explanations

instances of numerals and punctuation marks, particularly focusing on their distribution and usage within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

à¸¥à¸°

-0.15

ç¿Ķ

-0.14

umbo

-0.14

esthes

-0.14

aa

-0.14

_PCI

-0.14

loff

-0.13

uce

-0.13

Ĥ

-0.13

afa

-0.13

POSITIVE LOGITS

who

0.33

 whose

0.28

 whom

0.27

who

0.24

whose

0.23

 which

0.20

 quien

0.19

 Ø§ÙĦØ°ÙĬ

0.19

 ÐºÐ¾ÑĤÐ¾ÑĢÑĭÐ¹

0.19

 kterÃ½

0.18

Activations Density 0.124%

instances of numerals and punctuation marks, particularly focusing on their distribution and usage within the text

No Comments

No Known Activations