Upload mosaicml/mpt-7b ctranslate fp16 weights

README.md

@@ -21,38 +21,40 @@ Speedup inference while reducing memory by 2x-4x using int8 inference in C++ on
 
 quantized version of [mosaicml/mpt-7b](https://huggingface.co/mosaicml/mpt-7b)
 ```bash
-pip install hf-hub-ctranslate2>=2.0.8 ctranslate2>=3.14.0
-```
-Converted on 2023-05-31 using
-```
-ct2-transformers-converter --model mosaicml/mpt-7b --output_dir /home/michael/tmp-ct2fast-mpt-7b --force --copy_files tokenizer.json README.md tokenizer_config.json generation_config.json special_tokens_map.json .gitattributes --quantization float16 --trust_remote_code
+pip install hf-hub-ctranslate2>=2.12.0 ctranslate2>=3.16.0
 ```
 
-Checkpoint compatible to [ctranslate2>=3.14.0](https://github.com/OpenNMT/CTranslate2) and [hf-hub-ctranslate2>=2.0.8](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
-
+# from transformers import AutoTokenizer
 model_name = "michaelfeil/ct2fast-mpt-7b"
-# use either TranslatorCT2fromHfHub or GeneratorCT2fromHfHub here, depending on model.
+
+
+from hf_hub_ctranslate2 import GeneratorCT2fromHfHub
 model = GeneratorCT2fromHfHub(
         # load in int8 on CUDA
-        model_name_or_path=model_name, 
+        model_name_or_path=model_name,
         device="cuda",
         compute_type="int8_float16",
-        # tokenizer=AutoTokenizer.from_pretrained("mosaicml/mpt-7b")
+        # tokenizer=AutoTokenizer.from_pretrained("{ORG}/{NAME}")
 )
 outputs = model.generate(
-    text=["How do you call a fast Flan-ingo?", "User: How are you doing? Bot:"],
-    max_length=64, 
+    text=["def fibonnaci(", "User: How are you doing? Bot:"],
+    max_length=64,
     include_prompt_in_result=False
 )
 print(outputs)
 ```
 
+Checkpoint compatible to [ctranslate2>=3.16.0](https://github.com/OpenNMT/CTranslate2)
+and [hf-hub-ctranslate2>=2.12.0](https://github.com/michaelfeil/hf-hub-ctranslate2)
+- `compute_type=int8_float16` for `device="cuda"`
+- `compute_type=int8`  for `device="cpu"`
+
+Converted on 2023-06-27 using
+```
+ct2-transformers-converter --model mosaicml/mpt-7b --output_dir ~/tmp-ct2fast-mpt-7b --force --copy_files tokenizer.json README.md tokenizer_config.json generation_config.json special_tokens_map.json requirements.txt .gitattributes --quantization int8_float16 --trust_remote_code
+```
+
 # Licence and other remarks:
 This is just a quantized version. Licence conditions are intended to be idential to original huggingface repo.
 
@@ -64,12 +66,12 @@ This is just a quantized version. Licence conditions are intended to be idential
 MPT-7B is a decoder-style transformer pretrained from scratch on 1T tokens of English text and code.
 This model was trained by [MosaicML](https://www.mosaicml.com).
 
-MPT-7B is part of the family of MosaicPretrainedTransformer (MPT) models, which use a modified transformer architecture optimized for efficient training and inference. 
+MPT-7B is part of the family of MosaicPretrainedTransformer (MPT) models, which use a modified transformer architecture optimized for efficient training and inference.
 
-These architectural changes include performance-optimized layer implementations and the elimination of context length limits by replacing 
-positional embeddings with Attention with Linear Biases ([ALiBi](https://arxiv.org/abs/2108.12409)). 
-Thanks to these modifications, MPT models can be trained with high throughput efficiency and stable convergence. 
-MPT models can also be served efficiently with both standard HuggingFace pipelines and NVIDIA's [FasterTransformer](https://github.com/NVIDIA/FasterTransformer). 
+These architectural changes include performance-optimized layer implementations and the elimination of context length limits by replacing
+positional embeddings with Attention with Linear Biases ([ALiBi](https://arxiv.org/abs/2108.12409)).
+Thanks to these modifications, MPT models can be trained with high throughput efficiency and stable convergence.
+MPT models can also be served efficiently with both standard HuggingFace pipelines and NVIDIA's [FasterTransformer](https://github.com/NVIDIA/FasterTransformer).
 
 This model uses the MosaicML LLM codebase, which can be found in the [llm-foundry repository](https://github.com/mosaicml/llm-foundry). It was trained by MosaicML’s NLP team on the [MosaicML platform](https://www.mosaicml.com/training) for LLM pretraining, finetuning, and inference.
 
@@ -94,7 +96,7 @@ We demonstrate generations as long as 80k tokens on a single A100-80GB GPU in ou
   * License: Apache 2.0
 
 * [MPT-7B-Instruct](https://huggingface.co/mosaicml/mpt-7b-instruct): a model for short-form instruction following.
-Built by finetuning MPT-7B on a [dataset](https://huggingface.co/datasets/mosaicml/dolly_hhrlhf) we also release, derived from the [Databricks Dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k) and the [Anthropic Helpful and Harmless (HH-RLHF)](https://huggingface.co/datasets/Anthropic/hh-rlhf) datasets. 
+Built by finetuning MPT-7B on a [dataset](https://huggingface.co/datasets/mosaicml/dolly_hhrlhf) we also release, derived from the [Databricks Dolly-15k](https://huggingface.co/datasets/databricks/databricks-dolly-15k) and the [Anthropic Helpful and Harmless (HH-RLHF)](https://huggingface.co/datasets/Anthropic/hh-rlhf) datasets.
   * License: _CC-By-SA-3.0_
   * [Demo on Hugging Face Spaces](https://huggingface.co/spaces/mosaicml/mpt-7b-instruct)
 
@@ -130,37 +132,41 @@ model = transformers.AutoModelForCausalLM.from_pretrained(
   trust_remote_code=True
 )
 ```
-Note: This model requires that `trust_remote_code=True` be passed to the `from_pretrained` method. 
+Note: This model requires that `trust_remote_code=True` be passed to the `from_pretrained` method.
 This is because we use a custom `MPT` model architecture that is not yet part of the Hugging Face `transformers` package.
 `MPT` includes options for many training efficiency features such as [FlashAttention](https://arxiv.org/pdf/2205.14135.pdf), [ALiBi](https://arxiv.org/abs/2108.12409), [QK LayerNorm](https://arxiv.org/abs/2010.04245), and more.
 
-To use the optimized [triton implementation](https://github.com/openai/triton) of FlashAttention, you can load the model with `attn_impl='triton'` and move the model to `bfloat16`:
+To use the optimized [triton implementation](https://github.com/openai/triton) of FlashAttention, you can load the model on GPU (`cuda:0`) with `attn_impl='triton'` and with `bfloat16` precision:
 ```python
-config = transformers.AutoConfig.from_pretrained(
-  'mosaicml/mpt-7b',
-  trust_remote_code=True
-)
+import torch
+import transformers
+
+name = 'mosaicml/mpt-7b'
+
+config = transformers.AutoConfig.from_pretrained(name, trust_remote_code=True)
 config.attn_config['attn_impl'] = 'triton'
+config.init_device = 'cuda:0' # For fast initialization directly on GPU!
 
 model = transformers.AutoModelForCausalLM.from_pretrained(
-  'mosaicml/mpt-7b',
+  name,
   config=config,
-  torch_dtype=torch.bfloat16,
+  torch_dtype=torch.bfloat16, # Load model weights in bfloat16
   trust_remote_code=True
 )
-model.to(device='cuda:0')
 ```
 
 Although the model was trained with a sequence length of 2048, ALiBi enables users to increase the maximum sequence length during finetuning and/or inference. For example:
 
 ```python
-config = transformers.AutoConfig.from_pretrained(
-  'mosaicml/mpt-7b',
-  trust_remote_code=True
-)
-config.update({"max_seq_len": 4096})
+import transformers
+
+name = 'mosaicml/mpt-7b'
+
+config = transformers.AutoConfig.from_pretrained(name, trust_remote_code=True)
+config.max_seq_len = 4096 # (input + output) tokens can now be up to 4096
+
 model = transformers.AutoModelForCausalLM.from_pretrained(
-  'mosaicml/mpt-7b',
+  name,
   config=config,
   trust_remote_code=True
 )
@@ -170,7 +176,23 @@ This model was trained with the [EleutherAI/gpt-neox-20b](https://huggingface.co
 
 ```python
 from transformers import AutoTokenizer
-tokenizer = AutoTokenizer.from_pretrained("EleutherAI/gpt-neox-20b")
+tokenizer = AutoTokenizer.from_pretrained('EleutherAI/gpt-neox-20b')
+```
+
+The model can then be used, for example, within a text-generation pipeline.  
+Note: when running Torch modules in lower precision, it is best practice to use the [torch.autocast context manager](https://pytorch.org/docs/stable/amp.html).
+
+```python
+from transformers import pipeline
+
+pipe = pipeline('text-generation', model=model, tokenizer=tokenizer, device='cuda:0')
+
+with torch.autocast('cuda', dtype=torch.bfloat16):
+    print(
+        pipe('Here is a recipe for vegan banana bread:\n',
+            max_new_tokens=100,
+            do_sample=True,
+            use_cache=True))
 ```
 
 ## Model Description
@@ -198,7 +220,7 @@ The model has been modified from a standard transformer in the following ways:
 
 ### Streaming Datasets
 
-Data was formatted using the MosaicML [StreamingDataset](https://github.com/mosaicml/streaming) library to host our data in object storage and efficiently stream it to our compute cluster during training. 
+Data was formatted using the MosaicML [StreamingDataset](https://github.com/mosaicml/streaming) library to host our data in object storage and efficiently stream it to our compute cluster during training.
 StreamingDataset obviates the need to download the whole dataset before starting training, and allows instant resumption of training from any point in the dataset.
 
 
@@ -223,24 +245,24 @@ The model was trained for 1T tokens (with batch size 1760 and sequence length 20
 Samples for each batch were selected from one of the datasets with the probability specified above.
 The examples were shuffled within each dataset, and each example was constructed from as many sequences from that dataset as were necessary to fill the 2048 sequence length.
 
-The data was tokenized using the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer. This BPE tokenizer has a number of desirable characteristics, 
-most of which are relevant for tokenizing code: 
-(1) It was trained on a diverse mix of data that includes code (The Pile) 
-(2) It applies consistent space delimitation, unlike the GPT2 tokenizer which tokenizes inconsistently depending on the presence of prefix spaces 
-(3) It contains tokens for repeated space characters, which allows superior compression of text with large amounts of repeated space characters. 
+The data was tokenized using the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer. This BPE tokenizer has a number of desirable characteristics,
+most of which are relevant for tokenizing code:
+(1) It was trained on a diverse mix of data that includes code (The Pile)
+(2) It applies consistent space delimitation, unlike the GPT2 tokenizer which tokenizes inconsistently depending on the presence of prefix spaces
+(3) It contains tokens for repeated space characters, which allows superior compression of text with large amounts of repeated space characters.
 
 The model vocabulary size of 50432 was set to be a multiple of 128 (as in [MEGATRON-LM](https://arxiv.org/abs/1909.08053)), model flop utilization (MFU) increased by up to four percentage points.
 
 ### Training Configuration
 
-This model was trained on 440 A100-40GBs for about 9.5 days using the [MosaicML Platform](https://www.mosaicml.com/platform). 
-The model was trained with sharded data parallelism using [FSDP](https://pytorch.org/docs/stable/fsdp.html) and used the [LION](https://arxiv.org/abs/2302.06675) optimizer. 
+This model was trained on 440 A100-40GBs for about 9.5 days using the [MosaicML Platform](https://www.mosaicml.com/platform).
+The model was trained with sharded data parallelism using [FSDP](https://pytorch.org/docs/stable/fsdp.html) and used the [LION](https://arxiv.org/abs/2302.06675) optimizer.
 
 ## Limitations and Biases
 
 _The following language is modified from [EleutherAI's GPT-NeoX-20B](https://huggingface.co/EleutherAI/gpt-neox-20b)_
 
-MPT-7B (Base) is **not** intended for deployment without finetuning. 
+MPT-7B (Base) is **not** intended for deployment without finetuning.
 It should not be used for human-facing interactions without further guardrails and user consent.
 
 MPT-7B can produce factually incorrect output, and should not be relied on to produce factually accurate information.
@@ -263,11 +285,11 @@ Please cite this model using the following format:
 ```
 @online{MosaicML2023Introducing,
     author    = {MosaicML NLP Team},
-    title     = {Introducing MPT-7B: A New Standard for Open-Source, 
+    title     = {Introducing MPT-7B: A New Standard for Open-Source,
     ly Usable LLMs},
     year      = {2023},
     url       = {www.mosaicml.com/blog/mpt-7b},
     note      = {Accessed: 2023-03-28}, % change this date
     urldate   = {2023-03-28} % change this date
 }
-```
+```

config.json

@@ -1,5 +1,56 @@
 {
-  "bos_token": "<|endoftext|>",
-  "eos_token": "<|endoftext|>",
-  "unk_token": "<|endoftext|>"
-}
+    "architectures": [
+        "MPTForCausalLM"
+    ],
+    "attn_config": {
+        "alibi": true,
+        "alibi_bias_max": 8,
+        "attn_impl": "torch",
+        "attn_pdrop": 0,
+        "attn_type": "multihead_attention",
+        "attn_uses_sequence_id": false,
+        "clip_qkv": null,
+        "prefix_lm": false,
+        "qk_ln": false,
+        "softmax_scale": null
+    },
+    "auto_map": {
+        "AutoConfig": "configuration_mpt.MPTConfig",
+        "AutoModelForCausalLM": "modeling_mpt.MPTForCausalLM"
+    },
+    "d_model": 4096,
+    "emb_pdrop": 0,
+    "embedding_fraction": 1.0,
+    "expansion_ratio": 4,
+    "init_config": {
+        "emb_init_std": null,
+        "emb_init_uniform_lim": null,
+        "fan_mode": "fan_in",
+        "init_div_is_residual": true,
+        "init_gain": 0,
+        "init_nonlinearity": "relu",
+        "init_std": 0.02,
+        "name": "kaiming_normal_",
+        "verbose": 0
+    },
+    "init_device": "cpu",
+    "learned_pos_emb": true,
+    "logit_scale": null,
+    "max_seq_len": 2048,
+    "model_type": "mpt",
+    "n_heads": 32,
+    "n_layers": 32,
+    "no_bias": true,
+    "norm_type": "low_precision_layernorm",
+    "resid_pdrop": 0,
+    "tokenizer_name": "EleutherAI/gpt-neox-20b",
+    "torch_dtype": "bfloat16",
+    "transformers_version": "4.28.1",
+    "use_cache": false,
+    "verbose": 0,
+    "vocab_size": 50432,
+    "bos_token": "<|endoftext|>",
+    "eos_token": "<|endoftext|>",
+    "layer_norm_epsilon": null,
+    "unk_token": "<|endoftext|>"
+}

model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6e0b56675dcfb2208f90b599b216e647e808b1835f4e1e176877d5e77546566e
-size 13298599938
+oid sha256:1d82106e0ac05df8469ebba696197da8a3b1eaec83c858b7af823c61073a03fa
+size 6654505904

requirements.txt

@@ -0,0 +1,2 @@
+einops==0.5.0
+triton-pre-mlir@git+https://github.com/vchiley/triton.git@triton_pre_mlir_sm90#subdirectory=python