INDEX

Explanations

words related to decoration or elaborate presentation

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

neider

-0.15

mans

-0.15

uen

-0.14

@\

-0.14

 PROGMEM

-0.14

 Successful

-0.14

ough

-0.14

iece

-0.13

POSITIVE LOGITS

 liber

0.23

 accordingly

0.21

 entirely

0.19

 ident

0.18

 prof

0.17

 according

0.17

 attract

0.17

 differently

0.17

 simply

0.16

 elabor

0.16

Activations Density 0.073%

words related to decoration or elaborate presentation

No Comments

No Known Activations