INDEX

Explanations

elements related to historical narratives and literary analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 Å¡p

-0.16

 Tart

-0.15

ucker

-0.15

sector

-0.14

èĶµ

-0.14

 Marr

-0.14

çĥ

-0.14

xygen

-0.14

ought

-0.14

 Duffy

-0.14

POSITIVE LOGITS

 Merc

0.26

 Alfred

0.24

 Guth

0.22

 Lindsey

0.22

 Anglo

0.21

Merc

0.19

ang

0.19

 Ã¾

0.18

 Mild

0.18

Ang

0.17

Activations Density 0.012%

elements related to historical narratives and literary analysis

No Comments

No Known Activations