Quantization made by Richard Erkhov.
PowerLM-3b - GGUF
- Model creator: https://huggingface.co/ibm/
- Original model: https://huggingface.co/ibm/PowerLM-3b/
| Name | Quant method | Size |
|---|---|---|
| PowerLM-3b.Q2_K.gguf | Q2_K | 1.25GB |
| PowerLM-3b.IQ3_XS.gguf | IQ3_XS | 1.38GB |
| PowerLM-3b.IQ3_S.gguf | IQ3_S | 1.45GB |
| PowerLM-3b.Q3_K_S.gguf | Q3_K_S | 1.45GB |
| PowerLM-3b.IQ3_M.gguf | IQ3_M | 1.52GB |
| PowerLM-3b.Q3_K.gguf | Q3_K | 1.62GB |
| PowerLM-3b.Q3_K_M.gguf | Q3_K_M | 1.62GB |
| PowerLM-3b.Q3_K_L.gguf | Q3_K_L | 1.76GB |
| PowerLM-3b.IQ4_XS.gguf | IQ4_XS | 1.79GB |
| PowerLM-3b.Q4_0.gguf | Q4_0 | 1.87GB |
| PowerLM-3b.IQ4_NL.gguf | IQ4_NL | 1.89GB |
| PowerLM-3b.Q4_K_S.gguf | Q4_K_S | 1.89GB |
| PowerLM-3b.Q4_K.gguf | Q4_K | 2.0GB |
| PowerLM-3b.Q4_K_M.gguf | Q4_K_M | 2.0GB |
| PowerLM-3b.Q4_1.gguf | Q4_1 | 2.07GB |
| PowerLM-3b.Q5_0.gguf | Q5_0 | 2.27GB |
| PowerLM-3b.Q5_K_S.gguf | Q5_K_S | 2.27GB |
| PowerLM-3b.Q5_K.gguf | Q5_K | 2.33GB |
| PowerLM-3b.Q5_K_M.gguf | Q5_K_M | 2.33GB |
| PowerLM-3b.Q5_1.gguf | Q5_1 | 2.47GB |
| PowerLM-3b.Q6_K.gguf | Q6_K | 2.69GB |
| PowerLM-3b.Q8_0.gguf | Q8_0 | 3.48GB |
Original model description:
pipeline_tag: text-generation inference: false license: apache-2.0 library_name: transformers model-index: - name: ibm/PowerLM-3b results: - task: type: text-generation dataset: type: lm-eval-harness name: ARC metrics: - name: accuracy-norm type: accuracy-norm value: 60.5 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: BoolQ metrics: - name: accuracy type: accuracy value: 72.0 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: Hellaswag metrics: - name: accuracy-norm type: accuracy-norm value: 74.6 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: OpenBookQA metrics: - name: accuracy-norm type: accuracy-norm value: 43.6 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: PIQA metrics: - name: accuracy-norm type: accuracy-norm value: 79.9 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: Winogrande metrics: - name: accuracy-norm type: accuracy-norm value: 70.0 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: MMLU (5 shot) metrics: - name: accuracy type: accuracy value: 49.2 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: GSM8k (5 shot) metrics: - name: accuracy type: accuracy value: 34.9 verified: false - task: type: text-generation dataset: type: lm-eval-harness name: math (4 shot) metrics: - name: accuracy type: accuracy value: 15.2 verified: false - task: type: text-generation dataset: type: bigcode-eval name: humaneval metrics: - name: pass@1 type: pass@1 value: 26.8 verified: false - task: type: text-generation dataset: type: bigcode-eval name: MBPP metrics: - name: pass@1 type: pass@1 value: 33.6 verified: false
Model Summary
PowerLM-3B is a 3B state-of-the-art small language model trained with the Power learning rate scheduler. It is trained on a mix of open-source and proprietary datasets. PowerLM-3B has shown promising results compared to other models in the size categories across various benchmarks, including natural language multi-choices, code generation, and math reasoning. Paper: https://arxiv.org/abs/2408.13359
Usage
Note: Requires installing HF transformers from source.
Generation
This is a simple example of how to use PowerLM-3b model.
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
device = "cuda" # or "cpu"
model_path = "ibm/PowerLM-3b"
tokenizer = AutoTokenizer.from_pretrained(model_path)
# drop device_map if running on CPU
model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device)
model.eval()
# change input text as desired
prompt = "Write a code to find the maximum value in a list of numbers."
# tokenize the text
input_tokens = tokenizer(prompt, return_tensors="pt")
# transfer tokenized inputs to the device
for i in input_tokens:
input_tokens[i] = input_tokens[i].to(device)
# generate output tokens
output = model.generate(**input_tokens, max_new_tokens=100)
# decode output tokens into text
output = tokenizer.batch_decode(output)
# loop over the batch to print, in this example the batch size is 1
for i in output:
print(i)
Additional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more quants, at much higher speed, than I would otherwise be able to.