INDEX

Explanations

references to political parties and election dynamics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 Lobby

-0.18

lobby

-0.18

 lobby

-0.16

 diplomacy

-0.14

 FOREIGN

-0.14

 lobbying

-0.13

 diplomatic

-0.13

 diplomat

-0.13

(æĹ¥

-0.13

POSITIVE LOGITS

 flipped

0.28

 flipping

0.28

 seats

0.27

 flips

0.26

 control

0.26

 pickups

0.25

 races

0.25

 flip

0.25

 competitive

0.23

Seats

0.23

Activations Density 0.047%

references to political parties and election dynamics

No Comments

No Known Activations