transformers/pr_33962/en/model_doc/mamba.md

Mamba

Mamba is a selective structured state space model (SSMs) designed to work around Transformers computational inefficiency when dealing with long sequences. It is a completely attention-free architecture, and comprised of a combination of H3 and gated MLP blocks (Mamba block). Mamba's "content-based reasoning" allows it to focus on specific parts of an input depending on the current token. Mamba also uses a new hardware-aware parallel algorithm to compensate for the lack of convolutional operations. As a result, Mamba has fast inference and can scale to very long sequences.

You can find all the original Mamba checkpoints under the State Space Models organization.

This model was contributed by Molbap and AntonV. Click on the Mamba models in the right sidebar for more examples of how to apply Mamba to different language tasks.

The example below demonstrates how to generate text with Pipeline, AutoModel, and from the command line.

import torch
from transformers import pipeline

pipeline = pipeline(
    task="text-generation",
    model="state-spaces/mamba-130m-hf",
    dtype=torch.float16,
    device=0
)
pipeline("Plants create energy through a process known as")

import torch  
from transformers import AutoModelForCausalLM, AutoTokenizer  

tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-130m-hf")
model = AutoModelForCausalLM.from_pretrained("state-spaces/mamba-130m-hf", dtype=torch.float16, device_map="auto",)  
input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to(model.device)  

output = model.generate(**input_ids)  
print(tokenizer.decode(output[0], skip_special_tokens=True)

echo -e "Plants create energy through a process known as" | transformers run --task text-generation --model state-spaces/mamba-130m-hf --device 0

Quantization reduces the memory burden of large models by representing the weights in a lower precision. Refer to the Quantization overview for more available quantization backends.

The example below uses torchao to only quantize the weights to 4-bit integers.

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer, TorchAoConfig
from torchao.quantization import Int4WeightOnlyConfig

quantization_config = Int4WeightOnlyConfig(group_size=128)
quantization_config = TorchAoConfig(quant_type=quant_config)
tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-2.8b-hf")
model = AutoModelForCausalLM.from_pretrained("state-spaces/mamba-2.8b-hf", dtype=torch.bfloat16, quantization_config=quantization_config, device_map="auto",)
input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to(model.device)

output = model.generate(**input_ids)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Notes

• The current implementation uses the original CUDA kernels. The FlashAttention equivalent implementation is hosted in the mamba-ssm and causal_conv1d repositories. Make sure to install them if your hardware supports it!

• Mamba stacks mixer layers which are equivalent to Attention layers. You can find the main logic of Mamba in the MambaMixer class.

• The example below demonstrates how to fine-tune Mamba with PEFT.

  from datasets import load_dataset
  from trl import SFTConfig, SFTTrainer
  from peft import LoraConfig
  
  model_id = "state-spaces/mamba-130m-hf"
  dataset = load_dataset("Abirate/english_quotes", split="train")
  training_args = SFTConfig(dataset_text_field="quote")
  lora_config =  LoraConfig(target_modules=["x_proj", "embeddings", "in_proj", "out_proj"])
  trainer = SFTTrainer(
      model=model_id,
      args=training_args,
      train_dataset=dataset,
      peft_config=lora_config,
  )
  trainer.train()
  

MambaCache[[transformers.MambaCache]]

class transformers.MambaCachetransformers.MambaCachehttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L84[{"name": "config", "val": ": PreTrainedConfig"}, {"name": "max_batch_size", "val": ": int"}, {"name": "dtype", "val": ": dtype = torch.float16"}, {"name": "device", "val": ": typing.Union[torch.device, str, NoneType] = None"}]- config (`PreTrainedConfig) -- The configuration file defining the shape-related attributes required to initialize the static cache.

• max_batch_size (int) -- The maximum batch size with which the model will be used. Note that a new instance must be instantiated if a smaller batch size is used.
• dtype (torch.dtype, optional, defaults to torch.float16) -- The default dtype to use when initializing the layer.
• device (torch.device or str, optional) -- The device on which the cache should be initialized. Should be the same as the layer.0

Cache for mamba model which does not have attention mechanism and key value states.

Example:

>>> import torch
>>> from transformers import AutoTokenizer, MambaForCausalLM, MambaCache

>>> model = MambaForCausalLM.from_pretrained("state-spaces/mamba-130m-hf")
>>> tokenizer = AutoTokenizer.from_pretrained("state-spaces/mamba-130m-hf")

>>> inputs = tokenizer(text="My name is Mamba", return_tensors="pt")

>>> # Prepare a cache class and pass it to model's forward
>>> cache_params = MambaCache(config=model.config, max_batch_size=1, device=model.device, dtype=model.dtype)
>>> cache_position = torch.arange(len(inputs["input_ids"][0]), device=model.device)  # sequence length
>>> outputs = model(**inputs, cache_params=cache_params, cache_position=cache_position, use_cache=True)
>>> outputs.cache_params

update_conv_statetransformers.MambaCache.update_conv_statehttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L158[{"name": "layer_idx", "val": ": int"}, {"name": "new_conv_state", "val": ": Tensor"}, {"name": "cache_position", "val": ": LongTensor"}]

update_ssm_statetransformers.MambaCache.update_ssm_statehttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L175[{"name": "layer_idx", "val": ": int"}, {"name": "new_ssm_state", "val": ": Tensor"}]

resettransformers.MambaCache.resethttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L180[]

MambaConfig[[transformers.MambaConfig]]

class transformers.MambaConfigtransformers.MambaConfighttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/configuration_mamba.py#L26[{"name": "vocab_size", "val": " = 50280"}, {"name": "hidden_size", "val": " = 768"}, {"name": "state_size", "val": " = 16"}, {"name": "num_hidden_layers", "val": " = 32"}, {"name": "layer_norm_epsilon", "val": " = 1e-05"}, {"name": "pad_token_id", "val": " = 0"}, {"name": "bos_token_id", "val": " = 0"}, {"name": "eos_token_id", "val": " = 0"}, {"name": "expand", "val": " = 2"}, {"name": "conv_kernel", "val": " = 4"}, {"name": "use_bias", "val": " = False"}, {"name": "use_conv_bias", "val": " = True"}, {"name": "hidden_act", "val": " = 'silu'"}, {"name": "initializer_range", "val": " = 0.1"}, {"name": "residual_in_fp32", "val": " = True"}, {"name": "time_step_rank", "val": " = 'auto'"}, {"name": "time_step_scale", "val": " = 1.0"}, {"name": "time_step_min", "val": " = 0.001"}, {"name": "time_step_max", "val": " = 0.1"}, {"name": "time_step_init_scheme", "val": " = 'random'"}, {"name": "time_step_floor", "val": " = 0.0001"}, {"name": "rescale_prenorm_residual", "val": " = False"}, {"name": "use_cache", "val": " = True"}, {"name": "use_mambapy", "val": " = False"}, {"name": "**kwargs", "val": ""}]- vocab_size (int, optional, defaults to 50280) -- Vocabulary size of the MAMBA model. Defines the number of different tokens that can be represented by the inputs_ids passed when calling MambaModel.

• hidden_size (int, optional, defaults to 768) -- Dimensionality of the embeddings and hidden states.
• state_size (int, optional, defaults to 16) -- shape of the state space latents.
• num_hidden_layers (int, optional, defaults to 32) -- Number of hidden layers in the model.
• layer_norm_epsilon (float, optional, defaults to 1e-05) -- The epsilon to use in the layer normalization layers.
• pad_token_id (int, optional, defaults to 0) -- Padding token id.
• bos_token_id (int, optional, defaults to 0) -- The id of the beginning of sentence token in the vocabulary.
• eos_token_id (int, optional, defaults to 0) -- The id of the end of sentence token in the vocabulary.
• expand (int, optional, defaults to 2) -- Expanding factor used to determine the intermediate size.
• conv_kernel (int, optional, defaults to 4) -- Size of the convolution kernel.
• use_bias (bool, optional, defaults to False) -- Whether or not to use bias in ["in_proj", "out_proj"] of the mixer block
• use_conv_bias (bool, optional, defaults to True) -- Whether or not to use bias in the convolution layer of the mixer block.
• hidden_act (str, optional, defaults to "silu") -- The non-linear activation function (function or string) in the decoder.
• initializer_range (float, optional, defaults to 0.1) -- The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
• residual_in_fp32 (bool, optional, defaults to True) -- Whether or not residuals should be in float32. If set to False residuals will keep the same dtype as the rest of the model
• time_step_rank (Union[int,str], optional, defaults to "auto") -- Rank of the discretization projection matrix. "auto" means that it will default to math.ceil(self.hidden_size / 16)
• time_step_scale (float, optional, defaults to 1.0) -- Scale used used to scale dt_proj.bias.
• time_step_min (float, optional, defaults to 0.001) -- Minimum time_step used to bound dt_proj.bias.
• time_step_max (float, optional, defaults to 0.1) -- Maximum time_step used to bound dt_proj.bias.
• time_step_init_scheme (float, optional, defaults to "random") -- Init scheme used for dt_proj.weight. Should be one of ["random","uniform"]
• time_step_floor (float, optional, defaults to 0.0001) -- Minimum clamping value of the dt_proj.bias layer initialization.
• rescale_prenorm_residual (bool, optional, defaults to False) -- Whether or not to rescale out_proj weights when initializing.
• use_cache (bool, optional, defaults to True) -- Whether or not the cache should be used.
• use_mambapy (bool, optional, defaults to False) -- Determines the fallback strategy during training if the CUDA-based official implementation of Mamba is not available. If True, the mamba.py implementation is used. If False, the naive and slower implementation is used. Consider switching to the naive version if memory is limited.0

This is the configuration class to store the configuration of a MambaModel. It is used to instantiate a MAMBA model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the MAMBA state-spaces/mamba-2.8b architecture.

Configuration objects inherit from PreTrainedConfig and can be used to control the model outputs. Read the documentation from PreTrainedConfig for more information.

Example:

>>> from transformers import MambaConfig, MambaModel

>>> # Initializing a Mamba configuration
>>> configuration = MambaConfig()

>>> # Initializing a model (with random weights) from the configuration
>>> model = MambaModel(configuration)

>>> # Accessing the model configuration
>>> configuration = model.config

MambaModel[[transformers.MambaModel]]

class transformers.MambaModeltransformers.MambaModelhttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L624[{"name": "config", "val": ""}]- config (MambaModel) -- Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.0

The bare Mamba Model outputting raw hidden-states without any specific head on top.

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forwardtransformers.MambaModel.forwardhttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L649[{"name": "input_ids", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "cache_params", "val": ": typing.Optional[transformers.models.mamba.modeling_mamba.MambaCache] = None"}, {"name": "use_cache", "val": ": typing.Optional[bool] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}, {"name": "cache_position", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.LongTensor] = None"}]- input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• inputs_embeds (torch.LongTensor of shape (batch_size, sequence_length, hidden_size), optional) -- Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model's internal embedding lookup matrix.

• cache_params (MambaCache, optional) -- If passed along, the model uses the previous state in all the blocks (which will give the output for the input_ids provided as if the model add state_input_ids + input_ids as context).

• use_cache (bool, optional) -- If set to True, the cache_params is returned and can be used to quickly generate the next logits.

• output_hidden_states (bool, optional) -- Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.

• return_dict (bool, optional) -- Whether or not to return a ModelOutput instead of a plain tuple.

• cache_position (torch.LongTensor of shape (sequence_length), optional) -- Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.

• attention_mask (torch.LongTensor of shape (batch_size, sequence_length), optional) -- Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

  • 1 for tokens that are not masked,
  • 0 for tokens that are masked.

  What are attention masks?0transformers.models.mamba.modeling_mamba.MambaOutput or tuple(torch.FloatTensor)A transformers.models.mamba.modeling_mamba.MambaOutput or a tuple of

torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MambaConfig) and inputs.

• last_hidden_state (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional, defaults to None) -- Sequence of hidden-states at the output of the last layer of the model.

• cache_params (~models.mamba.modeling_mamba.MambaCache, optional, defaults to None) -- The state of the model at the last time step. Can be used in a forward method with the next input_ids to avoid providing the old input_ids.

  Includes both the State space model state matrices after the selective scan, and the Convolutional states

• hidden_states (tuple[torch.FloatTensor], optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) -- Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

The MambaModel forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

MambaLMHeadModel[[transformers.MambaForCausalLM]]

class transformers.MambaForCausalLMtransformers.MambaForCausalLMhttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L735[{"name": "config", "val": ""}]- config (MambaForCausalLM) -- Model configuration class with all the parameters of the model. Initializing with a config file does not load the weights associated with the model, only the configuration. Check out the from_pretrained() method to load the model weights.0

The MAMBA Model transformer with a language modeling head on top (linear layer with weights tied to the input embeddings).

This model inherits from PreTrainedModel. Check the superclass documentation for the generic methods the library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads etc.)

This model is also a PyTorch torch.nn.Module subclass. Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage and behavior.

forwardtransformers.MambaForCausalLM.forwardhttps://github.com/huggingface/transformers/blob/vr_33962/src/transformers/models/mamba/modeling_mamba.py#L818[{"name": "input_ids", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "attention_mask", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "inputs_embeds", "val": ": typing.Optional[torch.FloatTensor] = None"}, {"name": "cache_params", "val": ": typing.Optional[transformers.models.mamba.modeling_mamba.MambaCache] = None"}, {"name": "labels", "val": ": typing.Optional[torch.LongTensor] = None"}, {"name": "output_hidden_states", "val": ": typing.Optional[bool] = None"}, {"name": "return_dict", "val": ": typing.Optional[bool] = None"}, {"name": "use_cache", "val": ": typing.Optional[bool] = None"}, {"name": "cache_position", "val": ": typing.Optional[torch.Tensor] = None"}, {"name": "**kwargs", "val": ""}]- input_ids (torch.LongTensor of shape (batch_size, sequence_length), optional) -- Indices of input sequence tokens in the vocabulary. Padding will be ignored by default.

Indices can be obtained using AutoTokenizer. See PreTrainedTokenizer.encode() and PreTrainedTokenizer.call() for details.

What are input IDs?

• attention_mask (torch.LongTensor of shape (batch_size, sequence_length), optional) -- Mask to avoid performing attention on padding token indices. Mask values selected in [0, 1]:

  • 1 for tokens that are not masked,
  • 0 for tokens that are masked.

  What are attention masks?

• inputs_embeds (torch.FloatTensor of shape (batch_size, sequence_length, hidden_size), optional) -- Optionally, instead of passing input_ids you can choose to directly pass an embedded representation. This is useful if you want more control over how to convert input_ids indices into associated vectors than the model's internal embedding lookup matrix.

• cache_params (MambaCache, optional) -- If passed along, the model uses the previous state in all the blocks (which will give the output for the input_ids provided as if the model add state_input_ids + input_ids as context).

• labels (torch.LongTensor of shape (batch_size, sequence_length), optional) -- Labels for language modeling. Note that the labels are shifted inside the model, i.e. you can set labels = input_ids Indices are selected in [-100, 0, ..., config.vocab_size] All labels set to -100 are ignored (masked), the loss is only computed for labels in [0, ..., config.vocab_size]

• output_hidden_states (bool, optional) -- Whether or not to return the hidden states of all layers. See hidden_states under returned tensors for more detail.

• return_dict (bool, optional) -- Whether or not to return a ModelOutput instead of a plain tuple.

• use_cache (bool, optional) -- If set to True, the cache_params is returned and can be used to quickly generate the next logits.

• cache_position (torch.Tensor of shape (sequence_length), optional) -- Indices depicting the position of the input sequence tokens in the sequence. Contrarily to position_ids, this tensor is not affected by padding. It is used to update the cache in the correct position and to infer the complete sequence length.0transformers.models.mamba.modeling_mamba.MambaCausalLMOutput or tuple(torch.FloatTensor)A transformers.models.mamba.modeling_mamba.MambaCausalLMOutput or a tuple of torch.FloatTensor (if return_dict=False is passed or when config.return_dict=False) comprising various elements depending on the configuration (MambaConfig) and inputs.

• loss (torch.FloatTensor of shape (1,), optional, returned when labels is provided) -- Language modeling loss (for next-token prediction).

• logits (torch.FloatTensor of shape (batch_size, sequence_length, config.vocab_size)) -- Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).

• cache_params (~models.mamba.modeling_mamba.MambaCache, optional, defaults to None) -- The state of the model at the last time step. Can be used in a forward method with the next input_ids to avoid providing the old input_ids.

  Includes both the State space model state matrices after the selective scan, and the Convolutional states

• hidden_states (tuple[torch.FloatTensor], optional, returned when output_hidden_states=True is passed or when config.output_hidden_states=True) -- Tuple of torch.FloatTensor (one for the output of the embeddings, if the model has an embedding layer, + one for the output of each layer) of shape (batch_size, sequence_length, hidden_size).

  Hidden-states of the model at the output of each layer plus the optional initial embedding outputs.

The MambaForCausalLM forward method, overrides the __call__ special method.

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the pre and post processing steps while the latter silently ignores them.

Example: