INDEX

Explanations

instances of the word "said" indicating speech or quotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Ð´Ð°Ð»Ñĸ

-0.16

ord

-0.15

oning

-0.14

orman

-0.14

plusplus

-0.14

ãĥ¼ãĥĢ

-0.13

jang

-0.13

 further

-0.13

occ

-0.13

ickey

-0.13

POSITIVE LOGITS

 Tuesday

0.20

 Thursday

0.20

 recently

0.19

 Friday

0.19

 Monday

0.19

 Wednesday

0.19

 earlier

0.19

 late

0.18

 during

0.18

recent

0.17

Activations Density 0.049%

instances of the word "said" indicating speech or quotations

No Comments

No Known Activations