File size: 3,426 Bytes
8fedb1b 18da4ef 8fedb1b 18da4ef 8fedb1b 18da4ef 8fedb1b 18da4ef 8fedb1b 18da4ef |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 |
---
base_model: unsloth/qwen2.5-3b-instruct-unsloth-bnb-4bit
tags:
- text-generation-inference
- transformers
- unsloth
- qwen2
- trl
- grpo
license: apache-2.0
language:
- en
datasets:
- open-r1/OpenR1-Math-220k
---
This is my experiment with training a reasoning model using TRL's GRPO and Unsloth API.
# Inference:
## Using Unsloth API (For Faster Inference):
```
import torch
from unsloth import FastLanguageModel
from transformers import TextStreamer
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = "ubermenchh/Qwen2.5-3B-open-r1-math",
max_seq_length = 1024,
dtype = torch.bfloat16,
load_in_4bit = True,
)
FastLanguageModel.for_inference(model)
SYSTEM_PROMPT = """
Respond in the following format:
<think>
...
</think>
<answer>
...
</answer>
"""
test_question = """
Let $z \in \mathbf{C}$, satisfying the condition $a z^{n}+b \mathrm{i} z^{n-1}+b \mathrm{i} z-a=0, a, b \in \mathbf{R}, m \in$ $\mathbf{N}$, find $|z|$.
"""
messages = [
{"role": "system", "content": SYSTEM_PROMPT},
{"role": "user", "content": test_question},
]
input_ids = tokenizer.apply_chat_template(
messages,
add_generation_prompt = True,
return_tensors = "pt",
).to("cuda")
text_streamer = TextStreamer(tokenizer, skip_prompt = True)
_ = model.generate(input_ids, streamer = text_streamer, max_new_tokens = 2048, pad_token_id = tokenizer.eos_token_id)
```
## Using Transformers API:
```
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch
model = AutoModelForCausalLM.from_pretrained(
"ubermenchh/Qwen2.5-3B-open-r1-math",
torch_dtype=torch.bfloat16,
device_map="auto",
trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained(
"ubermenchh/Qwen2.5-3B-open-r1-math",
trust_remote_code=True
)
SYSTEM_PROMPT = """
Respond in the following format:
<think>
...
</think>
<answer>
...
</answer>
"""
problem = "Let $z \in \mathbf{C}$, satisfying the condition $a z^{n}+b \mathrm{i} z^{n-1}+b \mathrm{i} z-a=0, a, b \in \mathbf{R}, m \in$ $\mathbf{N}$, find $|z|$."
prompt = [
{"role": "system", "content": SYSTEM_PROMPT},
{"role": "user", "content": problem}
]
input_text = tokenizer.apply_chat_template(prompt, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(input_text, return_tensors="pt").to(model.device)
outputs = model.generate(
**inputs,
max_new_tokens=3000,
temperature=1.3,
num_return_sequences=1,
pad_token_id=tokenizer.pad_token_id,
eos_token_id=tokenizer.eos_token_id
)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print("Question:\n", problem)
print("\n\nResponse:\n", response)
```
## References:
- [https://github.com/HarleyCoops/smolThinker-.5B](https://github.com/HarleyCoops/smolThinker-.5B)
- [https://gist.github.com/willccbb/4676755236bb08cab5f4e54a0475d6fb](https://gist.github.com/willccbb/4676755236bb08cab5f4e54a0475d6fb)
- [https://github.com/huggingface/open-r1](https://github.com/huggingface/open-r1)
# Uploaded model
- **Developed by:** ubermenchh
- **License:** apache-2.0
- **Finetuned from model :** unsloth/qwen2.5-3b-instruct-unsloth-bnb-4bit
This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth) |