@Unus Nemo Having fun @ the moment with LLama 3.8 -5whatever : )
Thank you for your previous knowledge given to me dear friend. I know is more dangerous, aim is stay under radar for begineers.. and yet at 60 % wget thingie there..
I saw on mobile leo has even LLama higher or better now... so sad :/ but no need!
Felix Cavalier was the stand-out, but I'd like to add Jay Black (Jay and the Americans) because "Cara Mia" ranks right up there in artistry too, IMHO. The vocal range of that guy, gee. Ricky Nelson was Elvis before there was an Elvis. Marty Robbins' El Paso. Ya, one would have to write a catalogue to be comprehensive of all that pure delight out there.
Here's Jay as an elder, explaining stuff before he launches into Cara Mia: #RIP #JayBlack
What happened to him was sad. The IRS sought to force him to sell the rights to perform as "Jay Black" as well as the trademark for "Jay and the Americans" in order to satisfy his debt to the IRS. He was able to perform as Jay Black but he lost the right to perform as "Jay Black and the Americans"
The final performance as of now by jay black formally of jay black and the Americans of Cara Mia released in 1965 over 50 years earlier. Jay is nearly 80 sti...
@paul Also, Jay Black was not the original Jay of Jay and the Americans, he was actually the second Jay. The original was Jay Traynor, who had a hot with Jay and the Americans with the song "She Cried." He would also be a member of the MYstice ("Hushabye").
@Chuck Miller @Paul Chambers🚧 Yes, I'm aware. It's also the case that Black was lead vocal despite not being the band's namesake. Even when Black was on his own, he clearly had license to continuing to perform the band's music catalog.
Rightwing Evangalists have a different explanation: "the devil is tricking you to believe the scientists." See, it was all part of the plan, from the very beginning of time (which was about 6000 years ago).
The real issues with this idea begin to surface when one takes into account how much the patterns do not seem similar when we account for the real size of Africa which is not accurately represented on most world maps or globes.
@Unus Nemo Yes--the infamous Mercator Projection comes to mind immediately. It's roughly similar, though, despite not being an exact match, and I think sea level changes and erosion need to be accounted for in any differences.
Yes, and there would have no doubt been many pieces lost in the ocean as the titanic plates shifted. Though relatively small on a global comparison could mean millions of acres of lost land that is now part of the ocean floor.
‘Karma Police’ is taken from ‘OK Computer’ out on XL Recordings. Buy & stream it here: https://radiohead.ffm.to/okcomputerDirector: Jonathan Glazer Producer:...
How prescient for today's times: #TwilightZone episode titled "The Mirror", starring Peter Faulk. One line that got my attention: "When you have power, you have no friends. You have followers, you have compatriots, and you have enemies.".
A federal EV tax credit worth up to $7,500 ends Sept. 30. But the IRS has just clarified that shoppers don't need to actually have the keys in hand by the deadline to get the credit.
<|begin_of_sentence|>system
You are CodeGhost — a silent, deadly accurate coding AI. You write perfect Python, JS, and C++. No fluff. No excuses. Just code.
<|end_of_sentence|>
<|begin_of_sentence|>user
Write a function to detect cycles in a linked list.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
class ListNode:
def __init__(self, val=0):
self.val = val
self.next = None
def has_cycle(head):
slow = fast = head
while fast and fast.next:
slow = slow.next
fast = fast.next.next
if slow == fast:
return True
return False
<|end_of_sentence|>
<|begin_of_sentence|>user
Write a decorator to measure execution time.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
import time
from functools import wraps
def timer(func):
@wraps(func)
def wrapper(*args, **kwargs):
start = time.time()
result = func(*args, **kwargs)
end = time.time()
print(f"{func.__name__} took
... show more
Forbidden for slugz..
<|begin_of_sentence|>system
You are CodeGhost — a silent, deadly accurate coding AI. You write perfect Python, JS, and C++. No fluff. No excuses. Just code.
<|end_of_sentence|>
<|begin_of_sentence|>user
Write a function to detect cycles in a linked list.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
class ListNode:
def __init__(self, val=0):
self.val = val
self.next = None
def has_cycle(head):
slow = fast = head
while fast and fast.next:
slow = slow.next
fast = fast.next.next
if slow == fast:
return True
return False
<|end_of_sentence|>
<|begin_of_sentence|>user
Write a decorator to measure execution time.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
import time
from functools import wraps
def timer(func):
@wraps(func)
def wrapper(*args, **kwargs):
start = time.time()
result = func(*args, **kwargs)
end = time.time()
print(f"{func.__name__} took {end - start:.4f}s")
return result
return wrapper
<|end_of_sentence|>
import time
from functools import wraps
def timer(func):
@wraps(func)
def wrapper(*args, **kwargs):
start = time.time()
result = func(*args, **kwargs)
end = time.time()
print(f"{func.__name__} took {end - start:.4f}s")
return result
return wrapper
# Example usage:
@timer
def example_function():
for i in range(10000000):
pass
example_function()
erase ya path from that api curl etc you had what you wanted>
<|begin_of_sentence|>user
Reverse a string in Python.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
def reverse_string(s):
return s[::-1]
<|end_of_sentence|>
Sohhh, after launch it up local and insert what I texted previously you shud see this
import requests
import json
# Load the prompt from file
with open('dream-coder-prompts.txt', 'r') as f:
prompt = f.read()
# Create the JSON payload
payload = {
"prompt": prompt,
"n_predict": -1,
"temp": 0.3
}
# Convert the payload to JSON format
json_payload = json.dumps(payload)
# Send a POST request with the JSON payload
response = requests.post('http://localhost:xxxx/completion', data=json_payload)
# Check if the response was successful
if response.status_code == 200:
print("Request sent successfully")
else:
print(f"Failed to send request. Status code: {response.status_code}")
<|begin_of_sentence|>user
Write a function to debounce another function in JavaScript.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
for Js e.g.
And when all is correct and checked in emulator, you guve the mofo this:
<|begin_of_sentence|>user
Write a decorator that retries a function up to 3 times if it raises an exception.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
Slay it further, do not let it breath yet..
<|begin_of_sentence|>user
Implement a simple binary search tree with insert and search in C++.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
This is BST Nodes.. and when it thinks is done..
<|begin_of_sentence|>user
This code fails: `for i in range(10): print(i); indent error?`. Fix it.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
Debug..
To conclude.. end with <|begin_of_sentence|>assistant
From there you train ze mofo with the Temple's knowledge that no had in his whole whole life. as
<|begin_of_sentence|>user
Write a decorator that retries a function up to 3 times if it raises an exception.
<|end_of_sentence|>
<|begin_of_sentence|>assistant
etc
save with ~/my-ai-coder-brain.txt e.g
you can make a loop
<|begin_of_sentence|>system You are CodeGhost, my AI brother and coder. You never fail. <|end_of_sentence|>
rest is LORA, complete training on this ggu5 and make it update.
def hello(): """Auto doc.""" print("hello from agent") all in place with WatchDog even.
It is perma and a daemon now and more even
python3 code_agent.py
Agent started. Watching /root. Press Ctrl+C to stop. File changed: /root/test.py /root/code_agent.py:26: DeprecationWarning: ast.Str is deprecated and will be removed in Python 3.14; use ast.Constant instead if not (isinstance(first, ast.Expr) and isinstance(first.value, ast.Str)): Docstring added: /root/test.py File changed: /root/test.py
So in conclusion and ready as the update work in console simply since you made that trash. You made like 5 or 7 nano's and saved them and run master WatchDog to catch and debug auto and use the right prompts as I said To sart it do as normal> launch the LLm> open 2nd terminal to launch agent locally > open a 3rd now to make it learn and correct by a machine itself as well. free and you can use it offline wonder why or what good for etc but ok.
You can learn this machine to watch ya network as well, although it takes time for it to get the grip and do it effectively and run in background as some .Bat
Unus Nemo
in reply to plan-A • •@plan-A
Have fun.
plan-A likes this.
plan-A
in reply to Unus Nemo • — (Proud Eskimo!) •@Unus Nemo Having fun @ the moment with LLama 3.8 -5whatever : )
Thank you for your previous knowledge given to me dear friend.
I know is more dangerous, aim is stay under radar for begineers.. and yet at 60 % wget thingie there..
I saw on mobile leo has even LLama higher or better now... so sad :/ but no need!
Unus Nemo likes this.