README.md

@@ -65,16 +65,6 @@ For development dependencies:
 pip install -e ".[dev]"
 ```
 
-#### Download Pre-trained Model
-
-Download pre-trained model weights (1B bilingual checkpoint) to `params/`:
-
-```bash
-python scripts/demo/download.py
-```
-
-Or download manually from [HuggingFace](https://huggingface.co/ViperEk/KHAOSZ) into `params/`.
-
 #### Train a Model
 
 ```bash

assets/docs/README-zh-CN.md

@@ -71,16 +71,6 @@ pip install -e .
 pip install -e ".[dev]"
 ```
 
-#### 下载预训练模型
-
-下载预训练模型权重（1B 双语检查点）到 `params/` 目录：
-
-```bash
-python scripts/demo/download.py
-```
-
-或从 [HuggingFace](https://huggingface.co/ViperEk/KHAOSZ) 手动下载放入 `params/`。
-
 #### 训练模型
 
 ```bash

assets/docs/dataflow.md

@@ -88,7 +88,7 @@ flowchart LR
 - **`DecoderBlock`**: GQA attention + residual + MLP + RMSNorm
 - **`GQA`**: Grouped Query Attention (also `MLA` for multi-latent attention)
 - **`MLP`**: `SiLU(gate(x)) * up(x)` → down projection
-- **`RotaryEmbedding`**: RoPE complex cache (freqs_cis)
+- **`RotaryEmbedding`**: RoPE cos/sin cache
 - **`RMSNorm`**: Layer normalization
 
 ### 4. Training Module
@@ -104,23 +104,22 @@ The training loop is nested: **epoch** → **batch** (with step phase interspers
 ```
 on_train_begin
   on_epoch_begin
-    for each accumulation window of batches:        ← step phase
+    for each batch:
-      on_step_begin
+      if iteration % accumulation_steps == 0:        ← step phase
-        for each batch in window:                    ← batch phase
+        on_step_begin → optimizer.step() → zero_grad → on_step_end
+                                                      ← batch phase
       on_batch_begin → strategy(batch) → loss → backward → on_batch_end
       iteration += 1
-      on_step_end
-      optimizer.step() → zero_grad
 
     on_epoch_end
 on_train_end
 ```
 
 Key points:
-- `on_step_*` fires every `accumulation_steps` batches, wrapping optimizer step AFTER the hook
+- `on_step_*` wraps optimizer step (fires every `accumulation_steps` batches)
-- `on_batch_*` fires every batch, wrapping loss computation
+- `on_batch_*` wraps loss computation (fires every batch)
-- `GradientClippingCallback` fires on `on_step_end`
+- `SchedulerCallback` fires on `on_batch_end` — LR scheduler steps every batch
-- LR scheduler steps inline (no `SchedulerCallback` class)
+- `GradientClippingCallback` fires on `on_step_begin`
 
 #### 4.3 Strategy (`strategy.py`)
 - **`SEQStrategy`**: Next-token prediction, cross-entropy with label smoothing
@@ -137,7 +136,8 @@ Key points:
 - **`CheckpointCallback`**: Saves safetensors at `ckpt_interval` iterations
 - **`ProgressBarCallback`**: tqdm progress display
 - **`MetricLoggerCallback`**: Writes JSONL metrics to `{ckpt_dir}/logs/`
-- **`GradientClippingCallback`**: `clip_grad_norm_` on `on_step_end`
+- **`GradientClippingCallback`**: `clip_grad_norm_` on `on_step_begin`
+- **`SchedulerCallback`**: `scheduler.step()` on `on_batch_end`
 
 ### 5. Inference Module
 

assets/docs/design.md

@@ -91,8 +91,8 @@ classDiagram
         }
 
         class BaseStorage {
-            +MultiSegmentFetcher _fetcher
+            +Dict segments
-            +keys (property)
+            +List keys
             +load(load_path, tokenizer)
             +fetch(begin, end, keys)
             +__len__()
@@ -145,7 +145,7 @@ classDiagram
             +ModelConfig config
             +Registry _registry
             +register(model_type) decorator
-            +get_component_class(model_type) Type
+            +get_model_class(model_type) Type
             +from_pretrained(path, disable_random_init) nn.Module
             +save_pretrained(save_directory)
             +to(*args, **kwargs) Self
@@ -214,7 +214,7 @@ classDiagram
             +int dim
             +int max_len
             +float base
-            +forward(x, position_ids=None) Tensor
+            +forward(x, position_ids=None) Tuple[Tensor, Tensor]
         }
 
         class Embedding {
@@ -225,10 +225,13 @@ classDiagram
 
     namespace tokenize {
         class AutoTokenizer {
+            +List[int] stop_ids
+            +int bos_id
+            +int eos_id
+            +int pad_id
             +vocab_size int
             +encode(tokens, out_ids, add_special_tokens) List[int]
             +decode(tokens, skip_special_tokens) str
-            +__getattr__(name) Any (bos_id, eos_id, pad_id, stop_ids)
             +apply_chat_template(messages, tokenize) Union[str, List[int]]
             +set_chat_template(template)
             +load(path)
@@ -322,8 +325,6 @@ classDiagram
             +float clip_eps
             +float kl_coef
             +int group_size
-            +str reduction
-            +int sync_interval
             +compute_loss(batch) Tensor
         }
 
@@ -368,6 +369,11 @@ classDiagram
             +on_step_begin(context)
         }
 
+        class SchedulerCallback {
+            +on_train_begin(context)
+            +on_batch_end(context)
+        }
+
         class CheckpointCallback {
             +str save_dir
             +int interval
@@ -403,6 +409,8 @@ classDiagram
             +nn.Module model
             +AutoTokenizer tokenizer
             +InferenceScheduler scheduler
+            +int max_batch_size
+            +Optional int max_seq_len
             +generate(prompt, stream, max_tokens, temperature, top_p, top_k) Union[Generator, str, List[str]]
             +generate_with_request(request) Union[Generator, str, List[str]]
             +generate_async(prompt, max_tokens, temperature, top_p, top_k) AsyncGenerator
@@ -413,12 +421,13 @@ classDiagram
         class InferenceScheduler {
             +nn.Module model
             +AutoTokenizer tokenizer
-            +KVCache _page_cache
+            +KVCache page_cache
             +int max_batch_size
             +int max_seq_len
             +int max_prompt_len
             +int page_size
-            +TaskManager _task_mgr
+            +List waiting_queue
+            +List active_tasks
             +add_task(prompt, max_tokens, temperature, top_p, top_k, stream_callback) str
             +remove_task(task_id)
             +start()
@@ -559,7 +568,7 @@ classDiagram
         }
 
         class GenerateResult {
-            +List[Tuple[int, str]] tokens
+            +List[str] tokens
             +List[str] results
             +List[bool] _done
             +append(token, idx)
@@ -634,6 +643,7 @@ classDiagram
     BaseScheduler <|-- SGDRScheduler
     CallbackFactory ..> TrainCallback : creates
     TrainCallback <|-- GradientClippingCallback
+    TrainCallback <|-- SchedulerCallback
     TrainCallback <|-- CheckpointCallback
     TrainCallback <|-- ProgressBarCallback
     TrainCallback <|-- MetricLoggerCallback

astrai/__init__.py

@@ -1,4 +1,4 @@
-__version__ = "1.3.5"
+__version__ = "1.3.4"
 __author__ = "ViperEkura"
 
 from astrai.config import (

astrai/config/model_config.py

@@ -1,92 +1,12 @@
 import json
-import sys
+from dataclasses import asdict, dataclass
-from dataclasses import dataclass, fields
+from typing import Optional, Self
-from typing import Any, Dict, Optional, Self, get_type_hints
 
 
 @dataclass
-class BaseModelConfig:
+class ModelConfig:
-    """Field-aware JSON load/save for dataclass configs.
+    # basic config
-
-    Subclass with additional fields. The base ``model_type`` field
-    enables ``AutoModel`` to pick the correct subclass.
-    """
-
     model_type: Optional[str] = None
-
-    def load(self, config_path: str) -> Self:
-        raw: Dict[str, Any] = {}
-        with open(config_path, "r") as f:
-            raw.update(json.load(f))
-
-        hints = get_type_hints(type(self))
-        valid = {fld.name for fld in fields(self)}
-        for key, value in raw.items():
-            if key not in valid:
-                sys.stderr.write(f"WARNING: unknown config key '{key}'\n")
-                continue
-
-            target_type = self._unwrap_optional(hints.get(key))
-            if target_type is None:
-                continue
-
-            try:
-                value = self._coerce(value, target_type)
-            except (TypeError, ValueError):
-                sys.stderr.write(
-                    f"WARNING: cannot coerce '{key}' = {value!r} to {target_type}\n"
-                )
-                continue
-
-            setattr(self, key, value)
-
-        return self
-
-    def save(self, config_path: str):
-        config_dict: Dict[str, Any] = {}
-        for fld in fields(self):
-            v = getattr(self, fld.name)
-            if v is not None:
-                config_dict[fld.name] = v
-        with open(config_path, "w") as f:
-            json.dump(config_dict, f, indent=4)
-
-    @staticmethod
-    def _unwrap_optional(tp: type) -> Optional[type]:
-        if tp is None:
-            return None
-        origin = getattr(tp, "__origin__", None)
-        if origin is not None:
-            args = getattr(tp, "__args__", ())
-            non_none = [a for a in args if a is not type(None)]
-            return non_none[0] if non_none else None
-        return tp
-
-    @staticmethod
-    def _coerce(value: Any, target_type: type) -> Any:
-        if target_type is bool and isinstance(value, bool):
-            return value
-        if (
-            target_type is int
-            and isinstance(value, (int, float))
-            and not isinstance(value, bool)
-        ):
-            return int(value)
-        if (
-            target_type is float
-            and isinstance(value, (int, float))
-            and not isinstance(value, bool)
-        ):
-            return float(value)
-        if target_type is str and isinstance(value, str):
-            return value
-        if isinstance(value, target_type):
-            return value
-        raise TypeError
-
-
-@dataclass
-class ModelConfig(BaseModelConfig):
     vocab_size: Optional[int] = None
     dim: Optional[int] = None
 
@@ -99,16 +19,24 @@ class ModelConfig(BaseModelConfig):
     max_len: Optional[int] = None
     rope_theta: Optional[float] = None
 
-    # attention
+    # GQA
-    attn_type: str = "gqa"
     n_heads: Optional[int] = None
     n_kv_heads: Optional[int] = None
     use_qk_norm: Optional[bool] = None
     use_gated_attention: Optional[bool] = None
 
-    # MoE
+    def load(self, config_path: str) -> Self:
-    ffn_type: str = "mlp"
+        config = {}
-    n_routed_experts: Optional[int] = None
+        with open(config_path, "r") as f:
-    n_shared_experts: Optional[int] = None
+            config.update(json.load(f))
-    n_activated_experts: Optional[int] = None
+
-    moe_topk_method: Optional[str] = None
+        for key, value in config.items():
+            if hasattr(self, key):
+                setattr(self, key, value)
+
+        return self
+
+    def save(self, config_path: str):
+        config_dict = {k: v for k, v in asdict(self).items() if v is not None}
+        with open(config_path, "w") as f:
+            json.dump(config_dict, f, indent=4)

astrai/model/__init__.py

@@ -1,9 +1,11 @@
 from astrai.model.automodel import AutoModel
-from astrai.model.components.attention import GQA
+from astrai.model.module import (
-from astrai.model.components.decoder_block import DecoderBlock
+    GQA,
-from astrai.model.components.linear import Linear
+    MLP,
-from astrai.model.components.mlp import MLP
+    DecoderBlock,
-from astrai.model.components.norm import RMSNorm
+    Linear,
+    RMSNorm,
+)
 from astrai.model.transformer import Transformer
 
 __all__ = [

astrai/model/components/__init__.py

@@ -1,25 +0,0 @@
-from astrai.model.components.attention import GQA, MLA, repeat_kv
-from astrai.model.components.decoder_block import DecoderBlock
-from astrai.model.components.embedding import Embedding
-from astrai.model.components.linear import Linear
-from astrai.model.components.mlp import MLP
-from astrai.model.components.norm import RMSNorm
-from astrai.model.components.rope import (
-    RotaryEmbedding,
-    apply_rotary_emb,
-    get_rotary_emb,
-)
-
-__all__ = [
-    "Linear",
-    "RMSNorm",
-    "MLP",
-    "Embedding",
-    "GQA",
-    "MLA",
-    "DecoderBlock",
-    "RotaryEmbedding",
-    "apply_rotary_emb",
-    "get_rotary_emb",
-    "repeat_kv",
-]

astrai/model/components/decoder_block.py

@@ -1,58 +0,0 @@
-from typing import Optional
-
-import torch.nn as nn
-from torch import Tensor
-
-from astrai.inference.core.cache import KvcacheView
-from astrai.model.components.attention import AttnFactory
-from astrai.model.components.mlp import FFNFactory
-from astrai.model.components.norm import RMSNorm
-
-
-class DecoderBlock(nn.Module):
-    def __init__(
-        self,
-        dim: int,
-        n_heads: int,
-        dim_ffn: int,
-        n_kv_heads: int,
-        norm_eps: int,
-        use_qk_norm: bool,
-        use_gated_attention: bool,
-        layer_id: int,
-        attn_type: str = "gqa",
-        ffn_type: str = "mlp",
-        **moe_kwargs,
-    ):
-        super().__init__()
-        self.attention = AttnFactory.create(
-            attn_type,
-            dim=dim,
-            n_heads=n_heads,
-            n_kv_heads=n_kv_heads,
-            use_qk_norm=use_qk_norm,
-            norm_eps=norm_eps,
-            use_gated_attention=use_gated_attention,
-            layer_id=layer_id,
-        )
-        self.input_norm = RMSNorm(dim, norm_eps)
-        self.post_attention_norm = RMSNorm(dim, norm_eps)
-        self.mlp = FFNFactory.create(ffn_type, dim, dim_ffn, **moe_kwargs)
-
-    def forward(
-        self,
-        x: Tensor,
-        rotary_emb: Tensor,
-        attention_mask: Optional[Tensor] = None,
-        paged_cache: Optional[KvcacheView] = None,
-    ) -> Tensor:
-        attn_output = self.attention(
-            self.input_norm(x),
-            rotary_emb,
-            attention_mask,
-            paged_cache,
-        )
-        x = attn_output + x
-        x = self.mlp(self.post_attention_norm(x)) + x
-
-        return x

astrai/model/components/embedding.py

@@ -1,13 +0,0 @@
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from torch import Tensor
-
-
-class Embedding(nn.Module):
-    def __init__(self, vocab_size: int, embedding_dim: int):
-        super().__init__()
-        self.weight = nn.Parameter(torch.empty((vocab_size, embedding_dim)))
-
-    def forward(self, x: Tensor) -> Tensor:
-        return F.embedding(x, self.weight)

astrai/model/components/linear.py

@@ -1,14 +0,0 @@
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from torch import Tensor
-
-
-class Linear(nn.Module):
-    def __init__(self, in_dim: int, out_dim: int, bias: bool = False):
-        super().__init__()
-        self.weight = nn.Parameter(torch.empty((out_dim, in_dim)))
-        self.bias = nn.Parameter(torch.zeros(out_dim)) if bias else None
-
-    def forward(self, x: Tensor) -> Tensor:
-        return F.linear(x, self.weight, self.bias)

astrai/model/components/mlp.py

@@ -1,94 +0,0 @@
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from torch import Tensor
-
-from astrai.factory import BaseFactory
-from astrai.model.components.linear import Linear
-
-
-class FFNFactory(BaseFactory[nn.Module]):
-    @classmethod
-    def create(cls, ffn_type: str, dim: int, dim_ffn: int, **kwargs) -> nn.Module:
-        return super().create(ffn_type, dim, dim_ffn, **kwargs)
-
-
-@FFNFactory.register("mlp")
-class MLP(nn.Module):
-    def __init__(self, dim: int, dim_feed_forward: int, **kwargs):
-        super().__init__()
-        self.up = Linear(dim, dim_feed_forward)
-        self.gate = Linear(dim, dim_feed_forward)
-        self.down = Linear(dim_feed_forward, dim)
-
-    def forward(self, x: Tensor) -> Tensor:
-        gated = self.up(x) * F.silu(self.gate(x))
-        out = self.down(gated)
-        return out
-
-
-@FFNFactory.register("moe")
-class DeepSeekMoE(nn.Module):
-    def __init__(
-        self,
-        dim: int,
-        dim_feed_forward: int,
-        n_routed_experts: int,
-        n_shared_experts: int = 1,
-        n_activated_experts: int = 2,
-        topk_method: str = "greedy",
-        **kwargs,
-    ):
-        super().__init__()
-        self.dim = dim
-        self.n_routed_experts = n_routed_experts
-        self.n_shared_experts = n_shared_experts
-        self.n_activated_experts = n_activated_experts
-        self.topk_method = topk_method
-
-        self.router = Linear(dim, n_routed_experts, bias=False)
-
-        self.shared_experts = nn.ModuleList(
-            [MLP(dim, dim_feed_forward) for _ in range(n_shared_experts)]
-        )
-        self.routed_experts = nn.ModuleList(
-            [MLP(dim, dim_feed_forward) for _ in range(n_routed_experts)]
-        )
-
-    def forward(self, x: Tensor) -> Tensor:
-        bsz, seq_len, dim = x.shape
-        x_flat = x.view(-1, dim)
-
-        shared_out = self._shared_forward(x_flat)
-        routed_out = self._routed_forward(x_flat)
-
-        out = (shared_out + routed_out).view(bsz, seq_len, dim)
-        return out
-
-    def _shared_forward(self, x: Tensor) -> Tensor:
-        if self.n_shared_experts == 0:
-            return torch.zeros_like(x)
-        return sum(e(x) for e in self.shared_experts) / self.n_shared_experts
-
-    def _routed_forward(self, x: Tensor) -> Tensor:
-        N, D = x.shape
-        K = self.n_activated_experts
-
-        router_logits = self.router(x)
-        router_probs = torch.softmax(router_logits.float(), dim=-1).to(x.dtype)
-
-        topk_weights, topk_indices = torch.topk(router_probs, K, dim=-1)
-        topk_weights = topk_weights / topk_weights.sum(dim=-1, keepdim=True)
-
-        output = torch.zeros(N, D, device=x.device, dtype=x.dtype)
-        for expert_idx in range(self.n_routed_experts):
-            expert_mask = topk_indices == expert_idx
-            token_idx, k_idx = expert_mask.nonzero(as_tuple=True)
-            if token_idx.numel() == 0:
-                continue
-            expert_input = x[token_idx]
-            expert_output = self.routed_experts[expert_idx](expert_input)
-            weights = topk_weights[token_idx, k_idx].unsqueeze(-1)
-            output.index_add_(0, token_idx, expert_output * weights)
-
-        return output

astrai/model/components/norm.py

@@ -1,15 +0,0 @@
-import torch
-import torch.nn as nn
-import torch.nn.functional as F
-from torch import Tensor
-
-
-class RMSNorm(nn.Module):
-    def __init__(self, dim, norm_eps):
-        super().__init__()
-        self.weight = nn.Parameter(torch.ones(dim))
-        self.normalized_shape = (dim,)
-        self.norm_eps = norm_eps
-
-    def forward(self, x: Tensor) -> Tensor:
-        return F.rms_norm(x, self.normalized_shape, self.weight, self.norm_eps)

astrai/model/components/rope.py

@@ -1,53 +0,0 @@
-from typing import Optional
-
-import torch
-import torch.nn as nn
-from torch import Tensor
-
-
-def get_rotary_emb(
-    dim: int,
-    max_len: int,
-    base: float = 10000,
-    device: Optional[torch.device] = None,
-) -> Tensor:
-    theta = base ** (-torch.arange(0, dim, 2, dtype=torch.float64, device=device) / dim)
-    t = torch.arange(0, max_len, dtype=torch.float64, device=device)
-    freqs = torch.outer(t, theta).float()
-    cos = torch.cos(freqs)
-    sin = torch.sin(freqs)
-    return torch.complex(cos, sin)
-
-
-def apply_rotary_emb(x: torch.Tensor, freqs_cis: Tensor) -> Tensor:
-    dtype = x.dtype
-    x_ = x.float().reshape(*x.shape[:-1], -1, 2)
-    x_complex = torch.view_as_complex(x_)
-    freqs_cis = freqs_cis.unsqueeze(2)
-    x_rotated = x_complex * freqs_cis
-    x_out = torch.view_as_real(x_rotated).flatten(-2)
-    return x_out.to(dtype)
-
-
-class RotaryEmbedding(nn.Module):
-    def __init__(self, dim: int, max_len: int, base: int = 10000):
-        super().__init__()
-        self.dim = dim
-        self.max_len = max_len
-        self.base = base
-        self._set_rotary_buffer(self.max_len)
-
-    def _set_rotary_buffer(self, max_len: int):
-        rotary_emb = get_rotary_emb(self.dim, max_len, self.base)
-        freqs_cis = torch.view_as_real(rotary_emb)
-        self.register_buffer("freqs_cis", freqs_cis, persistent=False)
-
-    def forward(self, x: Tensor, position_ids: Optional[Tensor] = None) -> Tensor:
-        if position_ids is None:
-            position_ids = (
-                torch.arange(x.size(1), device=x.device)
-                .unsqueeze(0)
-                .expand(x.size(0), -1)
-            )
-        position_freq_cis = self.freqs_cis[position_ids].float()
-        return torch.view_as_complex(position_freq_cis)

astrai/model/module.py

@@ -5,14 +5,11 @@ import torch.nn as nn
 import torch.nn.functional as F
 from torch import Tensor
 
-from astrai.factory import BaseFactory
 from astrai.inference.core.cache import KvcacheView
-from astrai.model.components.linear import Linear
-from astrai.model.components.norm import RMSNorm
-from astrai.model.components.rope import apply_rotary_emb
 
 
 def repeat_kv(x: Tensor, n_rep: int) -> Tensor:
+    """Repeat KV heads n_rep times for GQA."""
     bs, slen, n_heads, head_dim = x.shape
     if n_rep == 1:
         return x
@@ -23,13 +20,88 @@ def repeat_kv(x: Tensor, n_rep: int) -> Tensor:
     )
 
 
-class AttnFactory(BaseFactory[nn.Module]):
+def get_rotary_emb(
-    @classmethod
+    dim: int,
-    def create(cls, attn_type: str, **kwargs) -> nn.Module:
+    max_len: int,
-        return super().create(attn_type, **kwargs)
+    base: float = 10000,
+    device: Optional[torch.device] = None,
+) -> Tensor:
+    theta = base ** (-torch.arange(0, dim, 2, dtype=torch.float64, device=device) / dim)
+    t = torch.arange(0, max_len, dtype=torch.float64, device=device)
+    freqs = torch.outer(t, theta).float()
+    cos = torch.cos(freqs)
+    sin = torch.sin(freqs)
+    return torch.complex(cos, sin)
+
+
+def apply_rotary_emb(x: torch.Tensor, freqs_cis: Tensor) -> Tensor:
+    dtype = x.dtype
+    x_ = x.float().reshape(*x.shape[:-1], -1, 2)
+    x_complex = torch.view_as_complex(x_)
+    freqs_cis = freqs_cis.unsqueeze(2)
+    x_rotated = x_complex * freqs_cis
+    x_out = torch.view_as_real(x_rotated).flatten(-2)
+    return x_out.to(dtype)
+
+
+class RotaryEmbedding(nn.Module):
+    def __init__(self, dim: int, max_len: int, base: int = 10000):
+        super().__init__()
+        self.dim = dim
+        self.max_len = max_len
+        self.base = base
+        self._set_rotary_buffer(self.max_len)
+
+    def _set_rotary_buffer(self, max_len: int):
+        rotary_emb = get_rotary_emb(self.dim, max_len, self.base)
+        freqs_cis = torch.view_as_real(rotary_emb)
+        self.register_buffer("freqs_cis", freqs_cis, persistent=False)
+
+    def forward(self, x: Tensor, position_ids: Optional[Tensor] = None) -> Tensor:
+        if position_ids is None:
+            position_ids = (
+                torch.arange(x.size(1), device=x.device)
+                .unsqueeze(0)
+                .expand(x.size(0), -1)
+            )
+        position_freq_cis = self.freqs_cis[position_ids].float()
+        return torch.view_as_complex(position_freq_cis)
+
+
+class Linear(nn.Module):
+    def __init__(self, in_dim: int, out_dim: int, bias: bool = False):
+        super().__init__()
+        self.weight = nn.Parameter(torch.empty((out_dim, in_dim)))
+        self.bias = nn.Parameter(torch.zeros(out_dim)) if bias else None
+
+    def forward(self, x: Tensor) -> Tensor:
+        return F.linear(x, self.weight, self.bias)
+
+
+class RMSNorm(nn.Module):
+    def __init__(self, dim, norm_eps):
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(dim))
+        self.normalized_shape = (dim,)
+        self.norm_eps = norm_eps
+
+    def forward(self, x: Tensor) -> Tensor:
+        return F.rms_norm(x, self.normalized_shape, self.weight, self.norm_eps)
+
+
+class MLP(nn.Module):
+    def __init__(self, dim: int, dim_feed_forward: int):
+        super().__init__()
+        self.up = Linear(dim, dim_feed_forward)
+        self.gate = Linear(dim, dim_feed_forward)
+        self.down = Linear(dim_feed_forward, dim)
+
+    def forward(self, x: Tensor) -> Tensor:
+        gated = self.up(x) * F.silu(self.gate(x))
+        out = self.down(gated)
+        return out
 
 
-@AttnFactory.register("gqa")
 class GQA(nn.Module):
     def __init__(
         self,
@@ -40,7 +112,6 @@ class GQA(nn.Module):
         norm_eps: float,
         use_gated_attention: bool,
         layer_id: int,
-        **kwargs,
     ):
         super().__init__()
         assert dim % n_heads == 0
@@ -81,6 +152,7 @@ class GQA(nn.Module):
     ) -> Tensor:
         is_causal = attn_mask is None
 
+        # (bsz, seq_len, dim) -> (bsz, seq_len, n_heads, head_dim)
         q = self._split_heads(self.q_proj(x), self.n_heads)
         k = self._split_heads(self.k_proj(x), self.n_kv_heads)
         v = self._split_heads(self.v_proj(x), self.n_kv_heads)
@@ -95,6 +167,7 @@ class GQA(nn.Module):
 
         k, v = repeat_kv(k, self.n_rep), repeat_kv(v, self.n_rep)
 
+        # (bsz, seq_len, n_heads, head_dim) -> (bsz, n_heads, seq_len, head_dim)
         q, k, v = q.permute(0, 2, 1, 3), k.permute(0, 2, 1, 3), v.permute(0, 2, 1, 3)
         sdqa_out = (
             F.scaled_dot_product_attention(q, k, v, attn_mask, is_causal=is_causal)
@@ -110,7 +183,6 @@ class GQA(nn.Module):
         return out
 
 
-@AttnFactory.register("mla")
 class MLA(nn.Module):
     def __init__(
         self,
@@ -123,7 +195,6 @@ class MLA(nn.Module):
         norm_eps: float,
         use_gated_attention: bool,
         layer_id: int,
-        **kwargs,
     ):
         super().__init__()
         self.dim = dim
@@ -141,6 +212,7 @@ class MLA(nn.Module):
         self.kv_a_proj = Linear(dim, kv_lora_rank, bias=False)
         self.kv_norm = RMSNorm(kv_lora_rank, norm_eps)
 
+        # fused KV: (k_nope, k_rope, v)
         self.kv_b_proj = Linear(
             kv_lora_rank,
             n_kv_heads * (self.head_dim + qk_rope_head_dim + self.head_dim),
@@ -202,3 +274,57 @@ class MLA(nn.Module):
 
         out = self.o_proj(attn_out)
         return out
+
+
+class DecoderBlock(nn.Module):
+    def __init__(
+        self,
+        dim: int,
+        n_heads: int,
+        dim_ffn: int,
+        n_kv_heads: int,
+        norm_eps: int,
+        use_qk_norm: bool,
+        use_gated_attention: bool,
+        layer_id: int,
+    ):
+        super().__init__()
+        self.attention = GQA(
+            dim,
+            n_heads,
+            n_kv_heads,
+            use_qk_norm,
+            norm_eps,
+            use_gated_attention,
+            layer_id,
+        )
+        self.input_norm = RMSNorm(dim, norm_eps)
+        self.mlp = MLP(dim, dim_ffn)
+        self.post_attention_norm = RMSNorm(dim, norm_eps)
+
+    def forward(
+        self,
+        x: Tensor,
+        rotary_emb: Tensor,
+        attention_mask: Optional[Tensor] = None,
+        paged_cache: Optional[KvcacheView] = None,
+    ) -> Tensor:
+        attn_output = self.attention(
+            self.input_norm(x),
+            rotary_emb,
+            attention_mask,
+            paged_cache,
+        )
+        x = attn_output + x
+        x = self.mlp(self.post_attention_norm(x)) + x
+
+        return x
+
+
+class Embedding(nn.Module):
+    def __init__(self, vocab_size: int, embedding_dim: int):
+        super().__init__()
+        self.weight = nn.Parameter(torch.empty((vocab_size, embedding_dim)))
+
+    def forward(self, x: Tensor) -> Tensor:
+        return F.embedding(x, self.weight)

astrai/model/transformer.py

@@ -7,11 +7,13 @@ from torch import Tensor
 from astrai.config.model_config import ModelConfig
 from astrai.inference.core.cache import KvcacheView
 from astrai.model.automodel import AutoModel
-from astrai.model.components.decoder_block import DecoderBlock
+from astrai.model.module import (
-from astrai.model.components.embedding import Embedding
+    DecoderBlock,
-from astrai.model.components.linear import Linear
+    Embedding,
-from astrai.model.components.norm import RMSNorm
+    Linear,
-from astrai.model.components.rope import RotaryEmbedding
+    RMSNorm,
+    RotaryEmbedding,
+)
 
 
 def process_attention_mask(
@@ -69,12 +71,6 @@ class Transformer(AutoModel):
                     config.use_qk_norm,
                     config.use_gated_attention,
                     layer_id,
-                    attn_type=config.attn_type,
-                    ffn_type=config.ffn_type,
-                    n_routed_experts=config.n_routed_experts,
-                    n_shared_experts=config.n_shared_experts,
-                    n_activated_experts=config.n_activated_experts,
-                    topk_method=config.moe_topk_method,
                 )
                 for layer_id in range(config.n_layers)
             ]