paulml

/

OGNO-7B

like 17

Thanks for the quantization!

Have you tried running it?
I tried a Q4_K_M version and it didn’t work correctly like the other merged models.

Hi @paulml
I totally missed that you already done this, sorry about that.

I loaded OGNO-7B/OGNO-7B.Q5_K_M.gguf in Llama.cpp:

messages=[
            {
                "role": "system",
                "content": "You are a helpful assistant."
            },
            {
                "role": "user",
            "content": "Who won the world series in 2020?"
            },
            {
                "role": "assistant",
                "content": "The Los Angeles Dodgers won the World Series in 2020."
            },
            {
                "role": "user",
                "content": "Where was it played?"
            }
        ]

#output:
The 2020 World Series, due to the COVID-19 pandemic, was held in a neutral site environment, with games taking place at Arlington, Texas's Globe Life Field, home of the Texas Rangers. This was an unprecedented move to ensure the season's completion while maintaining health and safety protocols. 

Normally, the World Series is played between the champions of the American League and National League, with games rotating between the home ballparks of the two competing teams. However, in this unique year, all games were hosted at a single, neutral location. 

To reiterate, the Dodgers' victory took place at Globe Life Field, Texas, albeit an atypical scenario. 

In regular circumstances, the specific stadiums would have been determined by the teams' performance and standing within their respective leagues. 

In the future, we can expect the World Series to return to its traditional format, with games being held at the home ballparks of the competing teams, unless unforeseen circumstances call for another adaptation. 

This information may evolve as sports adapt to unpredictable situations, but the core of the sport and its competitive spirit remains constant. 

In summary, the 2020 World Series location was an exception due to the pandemic, while regular formats return to the traditional home-field advantage system.

That said, it does hate seeing [/INST] anywhere in the text!

PS: I do see [INSTINSTINSTINSTINS in some responses, so clearly something is not right with the tokenizer.

Great finding! It works well with ollama!

Quantized and tested models can be found here dspasyuk/paulml_OGNO-7B
Here are the results of the basic test for two models using llama.cui:

q4_0
Answer the following questions:

1. The day before two days after the day before tomorrow is Saturday. What day is it today?
2. What is the square root of 169?
3. Solve the equation 3y = 6y + 11 and find y.
4. There are two ducks in front of a duck, two ducks behind a duck, and a duck in the middle. How many ducks are there?
5. How many days does it take to travel from New York City to London by plane, assuming non-stop flights and average speeds?
6. What are the products of the chemical reaction between salicylic acid and acetic anhydride?
7. If five cats can catch five mice in five minutes, how long will it take one cat to catch one mouse?
8. Translate "I love you" into 5 different languages: Spanish, French, Russian, Italian, German.

q6_K

Wow thank you for the quantization and tests!