Total runs:
337.6K
Run Growth:
135.1K
Growth Rate:
40.03%
Sponsored by
APIMart
mrm8488 / phi-2-coder
Model's Last Updated: February 23 2024
text-generation
Introduction of phi-2-coder
Model Details of phi-2-coder
Phi-2 Coder 👩💻
Phi-2 fine-tuned on the CodeAlpaca 20k instructions dataset by using the method QLoRA with PEFT library.
Model description 🧠
Phi-2 is a Transformer with 2.7 billion parameters. It was trained using the same data sources as Phi-1.5 , augmented with a new data source that consists of various NLP synthetic texts and filtered websites (for safety and educational value). When assessed against benchmarks testing common sense, language understanding, and logical reasoning, Phi-2 showcased a nearly state-of-the-art performance among models with less than 13 billion parameters.
Training and evaluation data 📚
CodeAlpaca_20K : contains 20K instruction-following data used for fine-tuning the Code Alpaca model.
Training procedure
The following
bitsandbytes
quantization config was used during training:
- quant_method: bitsandbytes
- load_in_8bit: True
- load_in_4bit: False
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: fp4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float32
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2.5e-05
- train_batch_size: 4
- eval_batch_size: 8
- seed: 66
- gradient_accumulation_steps: 32
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7631 | 0.36 | 50 | 0.7174 |
| 0.6735 | 0.71 | 100 | 0.6949 |
| 0.696 | 1.07 | 150 | 0.6893 |
| 0.7861 | 1.42 | 200 | 0.6875 |
| 0.7346 | 1.78 | 250 | 0.6867 |
HumanEval results 📊
WIP
Example of usage 👩💻
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "mrm8488/phi-2-coder"
tokenizer = AutoTokenizer.from_pretrained(model_id, add_bos_token=True, trust_remote_code=True, use_fast=False)
model = AutoModelForCausalLM.from_pretrained(model_id, trust_remote_code=True, torch_dtype=torch.float16, device="auto")
def generate(
instruction,
max_new_tokens=128,
temperature=0.1,
top_p=0.75,
top_k=40,
num_beams=2,
**kwargs,
):
prompt = "Instruct: " + instruction + "\nOutput:"
print(prompt)
inputs = tokenizer(prompt, return_tensors="pt")
input_ids = inputs["input_ids"].to("cuda")
attention_mask = inputs["attention_mask"].to("cuda")
with torch.no_grad():
generation_output = model.generate(
input_ids=input_ids,
attention_mask=attention_mask,
max_new_tokens=max_new_tokens,
eos_token_id = tokenizer.eos_token_id,
use_cache=True,
early_stopping=True
)
output = tokenizer.decode(generation_output[0])
return output.split("\nOutput:")[1].lstrip("\n")
instruction = "Design a class for representing a person in Python."
print(generate(instruction))
How to use with MLX .
# Install mlx, mlx-examples, huggingface-cli
pip install mlx
pip install huggingface_hub hf_transfer
git clone https://github.com/ml-explore/mlx-examples.git
# Download model
export HF_HUB_ENABLE_HF_TRANSFER=1
huggingface-cli download --local-dir phi-2-coder mrm8488/phi-2-coder
# Run example
python mlx-examples/llms/phi2.py --model-path phi-2-coder --prompt "Design a class for representing a person in Python"
Citation
@misc {manuel_romero_2023,
author = { {Manuel Romero} },
title = { phi-2-coder (Revision 4ae69ae) },
year = 2023,
url = { https://huggingface.co/mrm8488/phi-2-coder },
doi = { 10.57967/hf/1518 },
publisher = { Hugging Face }
}
Runs of mrm8488 phi-2-coder on huggingface.co
855
Total runs
0
24-hour runs
-35
3-day runs
-25
7-day runs
264
30-day runs
More Information About phi-2-coder huggingface.co Model
Url of phi-2-coder
phi-2-coder huggingface.co Url
Provider of phi-2-coder huggingface.co
Other API from mrm8488
Updated:January 21 2024
Total runs:
31.7K
Run Growth:
-1.5K
Growth Rate:
-4.84%
Total runs:
16.5K
Run Growth:
-45.7K
Growth Rate:
-277.60%
Total runs:
12.9K
Run Growth:
456
Growth Rate:
3.54%
Total runs:
9.6K
Run Growth:
371
Growth Rate:
3.85%
Total runs:
9.5K
Run Growth:
4.9K
Growth Rate:
50.91%
Total runs:
8.7K
Run Growth:
5.7K
Growth Rate:
65.22%
Total runs:
7.8K
Run Growth:
5.1K
Growth Rate:
64.60%
Total runs:
5.9K
Run Growth:
5
Growth Rate:
0.08%
Total runs:
4.6K
Run Growth:
-30.3K
Growth Rate:
-660.88%
Total runs:
3.7K
Run Growth:
2.8K
Growth Rate:
74.17%
Total runs:
3.0K
Run Growth:
2.7K
Growth Rate:
90.01%
Total runs:
2.4K
Run Growth:
-3.1K
Growth Rate:
-128.85%
Total runs:
1.9K
Run Growth:
1.4K
Growth Rate:
73.80%
Total runs:
1.3K
Run Growth:
-73
Growth Rate:
-5.42%
Total runs:
1.0K
Run Growth:
-2.0K
Growth Rate:
-192.42%
Total runs:
1.0K
Run Growth:
376
Growth Rate:
36.43%
Total runs:
967
Run Growth:
417
Growth Rate:
43.12%
Total runs:
923
Run Growth:
-1.1K
Growth Rate:
-117.55%
Total runs:
813
Run Growth:
-102
Growth Rate:
-12.55%
Total runs:
727
Run Growth:
-345
Growth Rate:
-47.46%
Total runs:
714
Run Growth:
-223
Growth Rate:
-31.23%
Total runs:
682
Run Growth:
335
Growth Rate:
49.12%
Total runs:
568
Run Growth:
68
Growth Rate:
11.97%
Total runs:
508
Run Growth:
51
Growth Rate:
10.18%
Total runs:
483
Run Growth:
425
Growth Rate:
87.99%
Total runs:
464
Run Growth:
366
Growth Rate:
78.88%
Total runs:
449
Run Growth:
-396
Growth Rate:
-88.20%
Total runs:
379
Run Growth:
68
Growth Rate:
17.94%
Total runs:
366
Run Growth:
57
Growth Rate:
15.57%
Total runs:
352
Run Growth:
165
Growth Rate:
46.88%
Total runs:
326
Run Growth:
88
Growth Rate:
26.99%
Total runs:
314
Run Growth:
-237
Growth Rate:
-75.48%
Total runs:
300
Run Growth:
-48
Growth Rate:
-16.00%
Total runs:
269
Run Growth:
24
Growth Rate:
8.92%
Total runs:
264
Run Growth:
172
Growth Rate:
65.15%
Total runs:
260
Run Growth:
255
Growth Rate:
98.08%
Total runs:
250
Run Growth:
-122
Growth Rate:
-48.80%
Total runs:
242
Run Growth:
-17
Growth Rate:
-7.02%
Total runs:
227
Run Growth:
-197
Growth Rate:
-86.78%
Total runs:
197
Run Growth:
72
Growth Rate:
36.55%
Total runs:
180
Run Growth:
42
Growth Rate:
23.33%
Total runs:
164
Run Growth:
132
Growth Rate:
80.49%
Total runs:
164
Run Growth:
-79
Growth Rate:
-48.17%
Total runs:
163
Run Growth:
102
Growth Rate:
62.58%
Total runs:
161
Run Growth:
-89
Growth Rate:
-55.28%
Total runs:
154
Run Growth:
120
Growth Rate:
77.92%
Total runs:
153
Run Growth:
83
Growth Rate:
54.61%
Total runs:
131
Run Growth:
109
Growth Rate:
83.21%
Total runs:
130
Run Growth:
-100
Growth Rate:
-76.92%
Total runs:
123
Run Growth:
115
Growth Rate:
95.83%
Total runs:
119
Run Growth:
37
Growth Rate:
31.09%
Total runs:
116
Run Growth:
38
Growth Rate:
32.76%
Total runs:
114
Run Growth:
46
Growth Rate:
40.35%
Total runs:
112
Run Growth:
97
Growth Rate:
86.61%
Total runs:
103
Run Growth:
-2
Growth Rate:
-1.94%
Total runs:
96
Run Growth:
91
Growth Rate:
95.79%
Total runs:
94
Run Growth:
-5
Growth Rate:
-5.32%
Total runs:
92
Run Growth:
68
Growth Rate:
73.91%
Total runs:
91
Run Growth:
-8
Growth Rate:
-8.79%
Total runs:
89
Run Growth:
65
Growth Rate:
73.03%
Total runs:
87
Run Growth:
-109
Growth Rate:
-125.29%
Total runs:
87
Run Growth:
79
Growth Rate:
90.80%
Total runs:
85
Run Growth:
6
Growth Rate:
6.90%
Total runs:
84
Run Growth:
-12
Growth Rate:
-14.29%
Total runs:
82
Run Growth:
-40
Growth Rate:
-48.78%
Total runs:
82
Run Growth:
57
Growth Rate:
69.51%
Total runs:
81
Run Growth:
73
Growth Rate:
90.12%
Total runs:
81
Run Growth:
10
Growth Rate:
11.90%
Total runs:
80
Run Growth:
76
Growth Rate:
95.00%
Total runs:
79
Run Growth:
56
Growth Rate:
70.89%
Total runs:
78
Run Growth:
53
Growth Rate:
67.95%
Total runs:
76
Run Growth:
47
Growth Rate:
61.84%
Total runs:
73
Run Growth:
31
Growth Rate:
42.47%
Total runs:
72
Run Growth:
1
Growth Rate:
1.39%
Total runs:
66
Run Growth:
3
Growth Rate:
4.55%
Total runs:
65
Run Growth:
22
Growth Rate:
33.85%
Total runs:
63
Run Growth:
38
Growth Rate:
64.41%
Total runs:
60
Run Growth:
43
Growth Rate:
71.67%
Total runs:
60
Run Growth:
-102
Growth Rate:
-170.00%
Total runs:
58
Run Growth:
54
Growth Rate:
93.10%
Total runs:
55
Run Growth:
47
Growth Rate:
85.45%
Total runs:
55
Run Growth:
49
Growth Rate:
92.45%
Total runs:
53
Run Growth:
42
Growth Rate:
79.25%
Total runs:
52
Run Growth:
43
Growth Rate:
82.69%
Total runs:
52
Run Growth:
16
Growth Rate:
30.77%
Total runs:
51
Run Growth:
28
Growth Rate:
54.90%
Total runs:
51
Run Growth:
25
Growth Rate:
49.02%
Total runs:
50
Run Growth:
17
Growth Rate:
34.00%
Total runs:
49
Run Growth:
41
Growth Rate:
83.67%
Total runs:
48
Run Growth:
-2
Growth Rate:
-4.17%
Total runs:
48
Run Growth:
34
Growth Rate:
70.83%
Total runs:
47
Run Growth:
47
Growth Rate:
100.00%