INDEX

Explanations

references to political structures and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 ÑĦÑĥÐ½Ð´Ð°

-0.17

 krev

-0.16

amburger

-0.15

GenerationStrategy

-0.14

opp

-0.14

 Ð¤ÐµÐ´ÐµÑĢÐ°Ð»ÑĮ

-0.14

 Russell

-0.14

Ãµi

-0.14

ITHER

-0.14

 moderator

-0.13

POSITIVE LOGITS

 Party

0.27

 Lenin

0.27

 party

0.23

 Stalin

0.23

 PARTY

0.23

Party

0.21

 Workers

0.20

 Worker

0.20

GPU

0.20

NK

0.19

Activations Density 0.047%

references to political structures and organizations

No Comments

No Known Activations