INDEX

Explanations

phrases that indicate capability or potential actions related to people or objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

edes

-0.16

Ø§Ø¬

-0.16

ilion

-0.16

rob

-0.16

relude

-0.16

ennie

-0.15

ÑıÐ²

-0.15

ÐµÑĢÐ°Ð»ÑĮ

-0.15

à¤¤à¤°

-0.15

.restore

-0.15

POSITIVE LOGITS

 Handle

0.16

opies

0.15

ä¼ı

0.15

 handle

0.14

ÏĢÎŃ

0.14

.cross

0.14

rapid

0.14

 Morg

0.14

erva

0.14

 Holdings

0.14

Activations Density 0.294%

phrases that indicate capability or potential actions related to people or objects

No Comments

No Known Activations