INDEX

Explanations

phrases indicating imminent action or threats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

Fab

-0.15

 pand

-0.15

dfa

-0.14

Fab

-0.14

erti

-0.14

skill

-0.13

âĨ

-0.13

ÑģÑĤÐµÑĢ

-0.13

 thought

-0.13

ACK

-0.13

POSITIVE LOGITS

 Force

0.17

andom

0.16

orse

0.15

ç´¢

0.15

Å©

0.15

 force

0.14

Handle

0.14

 FORCE

0.14

force

0.14

Force

0.14

Activations Density 0.259%

phrases indicating imminent action or threats

No Comments

No Known Activations