INDEX

Explanations

references to political figures and their actions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 protesting

-0.15

AGMA

-0.15

lt

-0.14

reib

-0.14

Ð°Ð½ÐºÐ°

-0.14

anca

-0.14

 Reporting

-0.14

 protests

-0.13

 protested

-0.13

Badge

-0.13

POSITIVE LOGITS

 personally

0.29

 appoint

0.23

 ordering

0.22

 himself

0.21

 persÃ¶n

0.20

 Personally

0.20

 personal

0.20

 Ð»Ð¸Ñĩ

0.20

 orders

0.19

 cabinet

0.19

Activations Density 0.347%

references to political figures and their actions

No Known Activations