INDEX

Explanations

references to Nazi-related topics and their historical implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

æ´¥

-0.16

anja

-0.15

ooter

-0.15

 Guerr

-0.14

ÎµÏģÎ³

-0.13

oin

-0.13

Ð¼Ð¾ÑĢ

-0.13

æ²»

-0.13

alet

-0.13

POSITIVE LOGITS

-era

0.16

-leaning

0.16

 Gest

0.14

 Germany

0.14

 tide

0.14

ReuseIdentifier

0.14

UIF

0.14

Bri

0.14

à¥įà¤¯à¤ķ

0.14

 lean

0.13

Activations Density 0.043%

references to Nazi-related topics and their historical implications

No Comments

No Known Activations