Upload bigcode/gpt_bigcode-santacoder ctranslate fp16 weights

README.md

@@ -0,0 +1,282 @@
+---
+tags:
+- ctranslate2
+- int8
+- float16
+
+license: openrail
+datasets:
+- bigcode/the-stack
+language:
+- code
+programming_language: 
+- Java
+- JavaScript
+- Python
+pipeline_tag: text-generation
+inference: false
+
+model-index:
+- name: SantaCoder
+  results:
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL HumanEval (Python)
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 0.18
+      verified: false
+    - name: pass@10
+      type: pass@10
+      value: 0.29
+      verified: false
+    - name: pass@100
+      type: pass@100
+      value: 0.49
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL MBPP (Python)
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 0.35
+      verified: false
+    - name: pass@10
+      type: pass@10
+      value: 0.58
+      verified: false
+    - name: pass@100
+      type: pass@100
+      value: 0.77
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL HumanEval (JavaScript)
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 0.16
+      verified: false
+    - name: pass@10
+      type: pass@10
+      value: 0.27
+      verified: false
+    - name: pass@100
+      type: pass@100
+      value: 0.47
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL MBPP (Javascript)
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 0.28
+      verified: false
+    - name: pass@10
+      type: pass@10
+      value: 0.51
+      verified: false
+    - name: pass@100
+      type: pass@100
+      value: 0.70
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL HumanEval (Java)
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 0.15
+      verified: false
+    - name: pass@10
+      type: pass@10
+      value: 0.26
+      verified: false
+    - name: pass@100
+      type: pass@100
+      value: 0.41
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL MBPP (Java)
+    metrics:
+    - name: pass@1
+      type: pass@1
+      value: 0.28
+      verified: false
+    - name: pass@10
+      type: pass@10
+      value: 0.44
+      verified: false
+    - name: pass@100
+      type: pass@100
+      value: 0.59
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: loubnabnl/humaneval_infilling
+      name: HumanEval FIM (Python)
+    metrics:
+    - name: single_line
+      type: exact_match
+      value: 0.44
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL HumanEval FIM (Java)
+    metrics:
+    - name: single_line
+      type: exact_match
+      value: 0.62
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: nuprl/MultiPL-E
+      name: MultiPL HumanEval FIM (JavaScript)
+    metrics:
+    - name: single_line
+      type: exact_match
+      value: 0.60
+      verified: false
+  - task:
+      type: text-generation
+    dataset:
+      type: code_x_glue_ct_code_to_text
+      name: CodeXGLUE code-to-text (Python)
+    metrics:
+    - name: BLEU
+      type: bleu
+      value: 18.13
+      verified: false
+---
+# # Fast-Inference with Ctranslate2
+Speedup inference while reducing memory by 2x-4x using int8 inference in C++ on CPU or GPU.
+
+quantized version of [bigcode/gpt_bigcode-santacoder](https://huggingface.co/bigcode/gpt_bigcode-santacoder)
+```bash
+pip install hf-hub-ctranslate2>=2.0.8 
+```
+Converted on 2023-05-23 using
+```
+ct2-transformers-converter --model bigcode/gpt_bigcode-santacoder --output_dir /home/michael/tmp-ct2fast-gpt_bigcode-santacoder --force --copy_files tokenizer.json README.md tokenizer_config.json special_tokens_map.json .gitattributes --quantization float16
+```
+
+Checkpoint compatible to [ctranslate2>=3.13.0](https://github.com/OpenNMT/CTranslate2) and [hf-hub-ctranslate2>=2.0.6](https://github.com/michaelfeil/hf-hub-ctranslate2)
+- `compute_type=int8_float16` for `device="cuda"` 
+- `compute_type=int8`  for `device="cpu"`
+
+```python
+from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub
+from transformers import AutoTokenizer
+
+model_name = "michaelfeil/ct2fast-gpt_bigcode-santacoder"
+# use either TranslatorCT2fromHfHub or GeneratorCT2fromHfHub here, depending on model.
+model = GeneratorCT2fromHfHub(
+        # load in int8 on CUDA
+        model_name_or_path=model_name, 
+        device="cuda",
+        compute_type="int8_float16",
+        # tokenizer=AutoTokenizer.from_pretrained("bigcode/gpt_bigcode-santacoder")
+)
+outputs = model.generate(
+    text=["How do you call a fast Flan-ingo?", "User: How are you doing? Bot:"],
+    max_length=64
+)
+print(outputs)
+```
+
+# Licence and other remarks:
+This is just a quantized version. Licence conditions are intended to be idential to original huggingface repo.
+
+# Original description
+    
+
+# SantaCoder
+
+![banner](https://huggingface.co/datasets/bigcode/admin/resolve/main/banner.png)
+
+Play with the model on the [SantaCoder Space Demo](https://huggingface.co/spaces/bigcode/santacoder-demo).
+
+#  Table of Contents
+
+1. [Model Summary](#model-summary)
+2. [Use](#use)
+3. [Limitations](#limitations)
+4. [Training](#training)
+5. [License](#license)
+6. [Citation](#citation)
+
+# Model Summary
+
+This is the Megatron-version of [SantaCoder](https://huggingface.co/bigcode/santacoder).
+We refer the reader to the [SantaCoder model page](https://huggingface.co/bigcode/santacoder) for full documentation about this model
+
+
+- **Repository:** [bigcode/Megatron-LM](https://github.com/bigcode-project/Megatron-LM)
+- **Project Website:** [bigcode-project.org](www.bigcode-project.org)
+- **Paper:** [🎅SantaCoder: Don't reach for the stars!🌟](https://t.co/YV3pzUbYOr)
+- **Point of Contact:** [contact@bigcode-project.org](mailto:contact@bigcode-project.org)
+- **Languages:** Python, Java, and JavaScript
+
+There are two versions (branches) of the model:
+* `main`: Uses the `gpt_bigcode` model. [Requires the bigcode fork of transformers](https://github.com/bigcode-project/transformers).
+* `main_custom`: Packaged with its modeling code. Requires `transformers>=4.27`.
+  Alternatively, it can run on older versions by setting the configuration parameter `activation_function = "gelu_pytorch_tanh"`.
+
+# Use
+
+## Intended use
+
+The model was trained on GitHub code. As such it is _not_ an instruction model and commands like "Write a function that computes the square root." do not work well.
+You should phrase commands like they occur in source code such as comments (e.g. `# the following function computes the sqrt`) or write a function signature and docstring and let the model complete the function body.
+
+### Attribution & Other Requirements
+
+The pretraining dataset of the model was filtered for permissive licenses only. Nevertheless, the model can generate source code verbatim from the dataset. The code's license might require attribution and/or other specific requirements that must be respected. We provide a [search index](https://huggingface.co/spaces/bigcode/santacoder-search) that let's you search through the pretraining data to identify where generated code came from and apply the proper attribution to your code.
+
+# Limitations
+
+The model has been trained on source code in Python, Java, and JavaScript. The predominant language in source is English although other languages are also present. As such the model is capable to generate code snippets provided some context but the generated code is not guaranteed to work as intended. It can be inefficient, contain bugs or exploits.
+
+# Training
+
+## Model
+
+- **Architecture:** GPT-2 model with multi-query attention and Fill-in-the-Middle objective
+- **Pretraining steps:** 600K
+- **Pretraining tokens:** 236 billion
+- **Precision:** float16
+
+## Hardware
+
+- **GPUs:** 96 Tesla V100
+- **Training time:** 6.2 days
+- **Total FLOPS:** 2.1 x 10e21
+
+## Software
+
+- **Orchestration:** [Megatron-LM](https://github.com/bigcode-project/Megatron-LM)
+- **Neural networks:** [PyTorch](https://github.com/pytorch/pytorch)
+- **FP16 if applicable:** [apex](https://github.com/NVIDIA/apex)
+
+# License
+The model is licenses under the CodeML Open RAIL-M v0.1 license. You can find the full license [here](https://huggingface.co/spaces/bigcode/license).

config.json

@@ -0,0 +1,5 @@
+{
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "unk_token": "<|endoftext|>"
+}

model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22d86b7c0f6c13408c8e06ced4c22979d55e136c26bdcb586f5c12ad03c2b7e4
+size 2627461067

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "<fim-prefix>",
+    "<fim-middle>",
+    "<fim-suffix>",
+    "<fim-pad>"
+  ]
+}

tokenizer_config.json

@@ -0,0 +1,7 @@
+{
+  "errors": "replace",
+  "tokenizer_class": "GPT2TokenizerFast",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 2048
+}