INDEX

Explanations

evaluations of actions or events that are negative or transformative in nature

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

ÙĦÛĮÙĩ

-0.16

Î»Î¯Î±

-0.15

 EITHER

-0.15

Mixin

-0.14

 Ð¿ÑĢÐµÐ´Ð²Ð°ÑĢ

-0.14

neath

-0.14

 Ð¾ÑĩÐµÑĢÐµÐ´

-0.14

 Credits

-0.14

UIApplication

-0.14

agar

-0.13

POSITIVE LOGITS

 downright

0.50

 outright

0.49

 border

0.29

 actual

0.28

 flat

0.26

 altogether

0.26

 actually

0.25

 entirely

0.25

 actively

0.24

 plain

0.23

Activations Density 0.224%

evaluations of actions or events that are negative or transformative in nature

No Comments

No Known Activations