INDEX

Explanations

instances of negation or disbelief in 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 @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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±ÑĥÐ´ÑĮ

-0.19

be

-0.16

Be

-0.16

 shall

-0.15

 admitting

-0.15

 STILL

-0.15

.Magenta

-0.14

otine

-0.14

(be

-0.14

Be

-0.14

POSITIVE LOGITS

 originally

0.26

 previously

0.25

 recently

0.23

 mean

0.22

 happen

0.21

 manage

0.21

 receive

0.21

 intend

0.20

 fare

0.20

 expect

0.20

Activations Density 0.059%

instances of negation or disbelief in the text

No Comments

No Known Activations