macadeliccc/gemma-orchid-7b-dpo

Gemma Orchid 7b

This model is the second checkpoint of a future project. Its capable of function calling as well as having a strong base in communicational skills.

This model has been finetuned on roughly 80k samples so far.

Training

• Time to complete: ~20 hours
• Datasets: Thermostatic/flowers, Intel/orca_dpo_pairs, jondurbin/truthy-dpo-v0.1, glaiveai/glaive_function_calling_v2
• Evaluation loss: 0.69
• Method: LoRa
• Prompt Format: ChatML

Thermostatic/flowers is a blend of open source model generations formatted in ShareGPT. It also includes all of capybara.

This model has been exposed to a wide variety of data. macadeliccc/gemma-function-calling-7b is suitable to finetune further with the dataset of your choosing.

Running the model on a CPU

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")
model = AutoModelForCausalLM.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

Running the model on a single / multi GPU

# pip install accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")
model = AutoModelForCausalLM.from_pretrained("macadeliccc/gemma-orchid-7b-dpo", device_map="auto")

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

Running the model on a GPU using different precisions

• Using torch.float16

# pip install accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")
model = AutoModelForCausalLM.from_pretrained("macadeliccc/gemma-orchid-7b-dpo", device_map="auto", torch_dtype=torch.float16)

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

• Using torch.bfloat16

# pip install accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")
model = AutoModelForCausalLM.from_pretrained("macadeliccc/gemma-orchid-7b-dpo", device_map="auto", torch_dtype=torch.bfloat16)

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

Quantized Versions through bitsandbytes

• Using 8-bit precision (int8)

# pip install bitsandbytes accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig

quantization_config = BitsAndBytesConfig(load_in_8bit=True)

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")
model = AutoModelForCausalLM.from_pretrained("macadeliccc/gemma-orchid-7b-dpo", quantization_config=quantization_config)

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

• Using 4-bit precision

# pip install bitsandbytes accelerate
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig

quantization_config = BitsAndBytesConfig(load_in_4bit=True)

tokenizer = AutoTokenizer.from_pretrained("macadeliccc/gemma-orchid-7b-dpo")
model = AutoModelForCausalLM.from_pretrained("macadeliccc/gemma-orchid-7b-dpo", quantization_config=quantization_config)

input_text = "Write me a poem about Machine Learning."
input_ids = tokenizer(input_text, return_tensors="pt").to("cuda")

outputs = model.generate(**input_ids)
print(tokenizer.decode(outputs[0]))

Other optimizations

• Flash Attention 2

First make sure to install flash-attn in your environment pip install flash-attn

model = AutoModelForCausalLM.from_pretrained(
    model_id, 
    torch_dtype=torch.float16, 
+   attn_implementation="flash_attention_2"
).to(0)

Inputs and outputs

• Input: Text string, such as a question, a prompt, or a document to be summarized.
• Output: Generated English-language text in response to the input, such as an answer to a question, or a summary of a document.

Evaluations

In progress

ExLlamaV2

Available here

Open LLM Leaderboard Evaluation Results

Detailed results can be found here

Metric	Value
Avg.	64.37
AI2 Reasoning Challenge (25-Shot)	62.88
HellaSwag (10-Shot)	80.95
MMLU (5-Shot)	61.41
TruthfulQA (0-shot)	53.27
Winogrande (5-shot)	77.51
GSM8k (5-shot)	50.19