> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.nvidia.com/nemo/automodel/v0.4/nemo-automodel/nemo_automodel/components/models/hy_mt2/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/automodel/_mcp/server.

# nemo_automodel.components.models.hy_mt2

## Submodules

* **[`nemo_automodel.components.models.hy_mt2.config`](/nemo-automodel/nemo_automodel/components/models/hy_mt2/config)**
* **[`nemo_automodel.components.models.hy_mt2.dispatch`](/nemo-automodel/nemo_automodel/components/models/hy_mt2/dispatch)**
* **[`nemo_automodel.components.models.hy_mt2.layers`](/nemo-automodel/nemo_automodel/components/models/hy_mt2/layers)**
* **[`nemo_automodel.components.models.hy_mt2.model`](/nemo-automodel/nemo_automodel/components/models/hy_mt2/model)**
* **[`nemo_automodel.components.models.hy_mt2.state_dict_adapter`](/nemo-automodel/nemo_automodel/components/models/hy_mt2/state_dict_adapter)**

## Package Contents

### Classes

| Name                                                                                  | Description                       |
| ------------------------------------------------------------------------------------- | --------------------------------- |
| [`HyMT2ForCausalLM`](#nemo_automodel-components-models-hy_mt2-model-HyMT2ForCausalLM) | Hy-MT2-30B-A3B causal-LM wrapper. |

### API

```python
class nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM(
    config: typing.Any,
    moe_config: nemo_automodel.components.moe.config.MoEConfig | None = None,
    backend: nemo_automodel.components.models.common.BackendConfig | None = None,
    kwargs = {}
)
```

**Bases:** [HFCheckpointingMixin](/nemo-automodel/nemo_automodel/components/models/common/hf_checkpointing_mixin#nemo_automodel-components-models-common-hf_checkpointing_mixin-HFCheckpointingMixin), `Module`, [MoEFSDPSyncMixin](/nemo-automodel/nemo_automodel/components/moe/fsdp_mixin#nemo_automodel-components-moe-fsdp_mixin-MoEFSDPSyncMixin)

Hy-MT2-30B-A3B causal-LM wrapper.

Mixes in `MoEFSDPSyncMixin` so EP / FSDP2 expert-gradient sync works
out of the box (set `distributed.ep_size` in the YAML; must divide
`num_experts`=128). The `HFCheckpointingMixin` provides
`from_pretrained` / `save_pretrained` over the HF safetensors layout.

**`_enable_lm_head_fp32`**

---

**`_keep_in_fp32_modules_strict`** `= ['mlp.gate.e_score_correction_bias']`

---

**`backend`** `= backend or BackendConfig()`

---

**`lm_head`**

---

**`model`**

---

**`state_dict_adapter`**

---

**`tie_word_embeddings_support`** `TieSupport = TieSupport.UNTIED_ONLY`

---

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.forward(
    input_ids: torch.Tensor,
    position_ids: torch.Tensor | None = None,
    attention_mask: torch.Tensor | None = None,
    padding_mask: torch.Tensor | None = None,
    logits_to_keep: typing.Union[int, torch.Tensor] = 0,
    output_hidden_states: bool | None = None,
    attn_kwargs: typing.Any = {}
) -> transformers.modeling_outputs.CausalLMOutputWithPast
```

Forward pass returning `CausalLMOutputWithPast`.

**Parameters:**

**`input_ids`** `torch.Tensor`

Token IDs. BSHD: `[B, S]`; THD: `[1, T]` (squeezed internally).

---

**`position_ids`** `torch.Tensor | None` — default: None

Optional position indices.

---

**`attention_mask`** `torch.Tensor | None` — default: None

Optional 2D padding mask.

---

**`padding_mask`** `torch.Tensor | None` — default: None

Optional padding mask used by the THD squeeze helper.

---

**`logits_to_keep`** `Union[int, torch.Tensor]` — default: 0

If 0 (default) compute logits for all positions; if > 0
only compute logits for the last `logits_to_keep` positions
(avoids materialising the full logit matrix during generation /
enables fused-linear cross-entropy in training).

---

**`output_hidden_states`** `bool | None` — default: None

Whether to return the final hidden states (the
input to `lm_head`) on the output. Defaults to the config flag.

---

**`**attn_kwargs`** `Any` — default: \{}

Additional attention kwargs forwarded to the base model
(e.g. qkv\_format, cu\_seqlens, CP kwargs).

---

**Returns:** `CausalLMOutputWithPast`

class:`~transformers.modeling_outputs.CausalLMOutputWithPast` with

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.from_config(
    config: typing.Any,
    moe_config: nemo_automodel.components.moe.config.MoEConfig | None = None,
    backend: nemo_automodel.components.models.common.BackendConfig | None = None,
    kwargs = {}
)
```

classmethod

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.from_pretrained(
    pretrained_model_name_or_path: str,
    model_args = (),
    kwargs = {}
)
```

classmethod

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.get_input_embeddings()
```

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.get_output_embeddings()
```

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.initialize_weights(
    buffer_device: torch.device | None = None,
    dtype: torch.dtype = torch.bfloat16
) -> None
```

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.set_input_embeddings(
    value
)
```

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.set_output_embeddings(
    new_embeddings
)
```

```python
nemo_automodel.components.models.hy_mt2.model.HyMT2ForCausalLM.update_moe_gate_bias() -> None
```

## Docs

- [nemo_automodel.components.models.hy_mt2.config](https://docs.nvidia.com/nemo/automodel/v0.4/nemo-automodel/nemo_automodel/components/models/hy_mt2/config.md)
- [nemo_automodel.components.models.hy_mt2.dispatch](https://docs.nvidia.com/nemo/automodel/v0.4/nemo-automodel/nemo_automodel/components/models/hy_mt2/dispatch.md)
- [nemo_automodel.components.models.hy_mt2.layers](https://docs.nvidia.com/nemo/automodel/v0.4/nemo-automodel/nemo_automodel/components/models/hy_mt2/layers.md)
- [nemo_automodel.components.models.hy_mt2.model](https://docs.nvidia.com/nemo/automodel/v0.4/nemo-automodel/nemo_automodel/components/models/hy_mt2/model.md)
- [nemo_automodel.components.models.hy_mt2.state_dict_adapter](https://docs.nvidia.com/nemo/automodel/v0.4/nemo-automodel/nemo_automodel/components/models/hy_mt2/state_dict_adapter.md)