INDEX

Explanations

instances of strong contrasts or conflicts within narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

à¹Īà¸Ńà¸Ļ

-0.15

ÑĩÐ¸ÑĤ

-0.15

acker

-0.15

 mÃ³

-0.15

ãĥ³ãĤº

-0.14

.pkg

-0.14

 Waste

-0.13

Ð¼Ð¾Ñģ

-0.13

ìĹ´

-0.13

à¸±à¸Ļà¸Ķ

-0.13

POSITIVE LOGITS

 again

0.32

again

0.28

 continued

0.27

 Again

0.26

 continue

0.26

 continues

0.26

Again

0.25

continued

0.23

 aforementioned

0.23

ç»§ç»Ń

0.23

Activations Density 0.119%

instances of strong contrasts or conflicts within narratives

No Comments

No Known Activations