Standart Qwen3 mimarisine geciliyor: model her arac tarafindan yuklenebilir

Olculdu: Erk-14B'nin 443 tensorunun 443'u de Qwen3-14B ile ayni isimde ve
mimari alanlarin hepsi esit (hidden_size 5120, 40 katman, 40/8 bas, head_dim 128,
intermediate 17408, vocab 151936, rope_theta 1e6, max_pos 40960). Yani
ErkForCausalLM, Qwen3 kodunun yeniden adlandirilmisiydi; islevsel fark yoktu.

Bedeli buyuktu:
- llama.cpp convert_hf_to_gguf.py model_type'a gore dagitim yapar; "erk" tanimsiz
oldugu icin GGUF donusumu bastan basarisiz oluyordu.
- vLLM, SGLang, TGI, Ollama, LM Studio ya reddediyor ya trust_remote_code istiyor.
- Ozel kod yalnizca transformers 5.x API'lerini kullandigi icin 4.x surumlerde
model hic acilmiyordu.
- use_cache false idi: KV onbellegi varsayilan kapali, uretim cok yavas.

Bu islem yalnizca config ve ozel kod dosyalarini degistirir; agirliklara
dokunulmamistir.

config.json

@@ -1,79 +1,30 @@
 {
   "architectures": [
-    "ErkForCausalLM"
+    "Qwen3ForCausalLM"
   ],
   "attention_bias": false,
   "attention_dropout": 0.0,
   "bos_token_id": 151643,
-  "dtype": "float32",
   "eos_token_id": 151645,
   "head_dim": 128,
   "hidden_act": "silu",
   "hidden_size": 5120,
   "initializer_range": 0.02,
   "intermediate_size": 17408,
-  "layer_types": [
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention",
-    "full_attention"
-  ],
   "max_position_embeddings": 40960,
   "max_window_layers": 40,
-  "model_type": "erk",
+  "model_type": "qwen3",
   "num_attention_heads": 40,
   "num_hidden_layers": 40,
   "num_key_value_heads": 8,
-  "pad_token_id": null,
   "rms_norm_eps": 1e-06,
-  "rope_parameters": {
-    "rope_theta": 1000000,
-    "rope_type": "default"
-  },
+  "rope_scaling": null,
+  "rope_theta": 1000000,
   "sliding_window": null,
   "tie_word_embeddings": false,
-  "transformers_version": "5.13.0",
-  "use_cache": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.0",
+  "use_cache": true,
   "use_sliding_window": false,
-  "vocab_size": 151936,
-  "auto_map": {
-    "AutoConfig": "configuration_erk.ErkConfig",
-    "AutoModelForCausalLM": "modeling_erk.ErkForCausalLM"
-  }
-}
+  "vocab_size": 151936
+}

configuration_erk.py

@@ -1,100 +0,0 @@
-# Copyright 2026 eCloud Yazılım Teknolojileri. Based on Apache-2.0 licensed transformer architecture.
-#
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-"""Erk model configuration"""
-
-from huggingface_hub.dataclasses import strict
-
-from transformers.configuration_utils import PreTrainedConfig
-from transformers.modeling_rope_utils import RopeParameters
-from transformers.utils import auto_docstring
-
-
-@auto_docstring(checkpoint="ecloudtech/erk")
-@strict
-class ErkConfig(PreTrainedConfig):
-    r"""
-    ```python
-    >>> from transformers import ErkModel, ErkConfig
-
-    >>> # Initializing a Erk style configuration
-    >>> configuration = ErkConfig()
-
-    >>> # Initializing a model from the Erk-8B style configuration
-    >>> model = ErkModel(configuration)
-
-    >>> # Accessing the model configuration
-    >>> configuration = model.config
-    ```
-    """
-
-    model_type = "erk"
-    keys_to_ignore_at_inference = ["past_key_values"]
-
-    # Default tensor parallel plan for base model `Erk`
-    base_model_tp_plan = {
-        "layers.*.self_attn.q_proj": "colwise",
-        "layers.*.self_attn.k_proj": "colwise",
-        "layers.*.self_attn.v_proj": "colwise",
-        "layers.*.self_attn.q_norm": "replicated_with_grad_allreduce",
-        "layers.*.self_attn.k_norm": "replicated_with_grad_allreduce",
-        "layers.*.self_attn.o_proj": "rowwise",
-        "layers.*.mlp.gate_proj": "colwise",
-        "layers.*.mlp.up_proj": "colwise",
-        "layers.*.mlp.down_proj": "rowwise",
-    }
-    base_model_pp_plan = {
-        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
-        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
-        "norm": (["hidden_states"], ["hidden_states"]),
-    }
-
-    vocab_size: int = 151936
-    hidden_size: int = 4096
-    intermediate_size: int = 22016
-    num_hidden_layers: int = 32
-    num_attention_heads: int = 32
-    num_key_value_heads: int | None = 32
-    head_dim: int = 128
-    hidden_act: str = "silu"
-    max_position_embeddings: int = 32768
-    initializer_range: float = 0.02
-    rms_norm_eps: float = 1e-6
-    use_cache: bool = True
-    tie_word_embeddings: bool = False
-    rope_parameters: RopeParameters | dict | None = None
-    attention_bias: bool = False
-    use_sliding_window: bool = False
-    sliding_window: int | None = 4096
-    max_window_layers: int = 28
-    layer_types: list[str] | None = None
-    attention_dropout: float | int = 0.0
-    pad_token_id: int | None = None
-    bos_token_id: int | None = None
-    eos_token_id: int | list[int] | None = None
-
-    def __post_init__(self, **kwargs):
-        self.sliding_window = self.sliding_window if self.use_sliding_window else None
-        if self.num_key_value_heads is None:
-            self.num_key_value_heads = self.num_attention_heads
-
-        if self.layer_types is None:
-            self.layer_types = [
-                "sliding_attention"
-                if self.sliding_window is not None and i >= self.max_window_layers
-                else "full_attention"
-                for i in range(self.num_hidden_layers)
-            ]
-        super().__post_init__(**kwargs)
-
-
-__all__ = ["ErkConfig"]

modeling_erk.py

@@ -1,538 +0,0 @@
-#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
-#           This file was automatically generated from src/transformers/models/erk/modular_erk.py.
-#               Do NOT edit this file manually as any edits will be overwritten by the generation of
-#             the file from the modular. If any change should be done, please apply the change to the
-#                          modular_erk.py file directly. One of our CI enforces this.
-#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
-# Copyright 2026 eCloud Yazılım Teknolojileri. Based on Apache-2.0 licensed transformer architecture.
-#
-# you may not use this file except in compliance with the License.
-# You may obtain a copy of the License at
-#
-#     http://www.apache.org/licenses/LICENSE-2.0
-#
-# Unless required by applicable law or agreed to in writing, software
-# distributed under the License is distributed on an "AS IS" BASIS,
-# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-# See the License for the specific language governing permissions and
-# limitations under the License.
-
-from collections.abc import Callable
-from typing import Optional
-
-import torch
-from torch import nn
-
-from transformers.activations import ACT2FN
-from transformers.cache_utils import Cache, DynamicCache
-from transformers.generation import GenerationMixin
-from transformers.integrations import use_kernel_forward_from_hub, use_kernel_func_from_hub, use_kernelized_func
-from transformers.masking_utils import create_causal_mask, create_sliding_window_causal_mask
-from transformers.modeling_flash_attention_utils import FlashAttentionKwargs
-from transformers.modeling_layers import (
-    GenericForQuestionAnswering,
-    GenericForSequenceClassification,
-    GenericForTokenClassification,
-    GradientCheckpointingLayer,
-)
-from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast
-from transformers.modeling_rope_utils import ROPE_INIT_FUNCTIONS, dynamic_rope_update
-from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS, PreTrainedModel
-from transformers.processing_utils import Unpack
-from transformers.utils import TransformersKwargs, auto_docstring, can_return_tuple
-from transformers.utils.generic import maybe_autocast, merge_with_config_defaults
-from transformers.utils.output_capturing import capture_outputs
-from .configuration_erk import ErkConfig
-
-
-@use_kernel_forward_from_hub("RMSNorm")
-class ErkRMSNorm(nn.Module):
-    def __init__(self, hidden_size, eps: float = 1e-6) -> None:
-        """
-        ErkRMSNorm is equivalent to T5LayerNorm
-        """
-        super().__init__()
-        self.weight = nn.Parameter(torch.ones(hidden_size))
-        self.variance_epsilon = eps
-
-    def forward(self, hidden_states: torch.Tensor) -> torch.Tensor:
-        input_dtype = hidden_states.dtype
-        hidden_states = hidden_states.to(torch.float32)
-        variance = hidden_states.pow(2).mean(-1, keepdim=True)
-        hidden_states = hidden_states * torch.rsqrt(variance + self.variance_epsilon)
-        return self.weight * hidden_states.to(input_dtype)
-
-    def extra_repr(self):
-        return f"{tuple(self.weight.shape)}, eps={self.variance_epsilon}"
-
-
-class ErkMLP(nn.Module):
-    def __init__(self, config):
-        super().__init__()
-        self.config = config
-        self.hidden_size = config.hidden_size
-        self.intermediate_size = config.intermediate_size
-        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
-        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
-        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
-        self.act_fn = ACT2FN[config.hidden_act]
-
-    def forward(self, x):
-        down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
-        return down_proj
-
-
-class ErkRotaryEmbedding(nn.Module):
-    inv_freq: torch.Tensor  # fix linting for `register_buffer`
-
-    def __init__(self, config: ErkConfig, device=None):
-        super().__init__()
-        self.max_seq_len_cached = config.max_position_embeddings
-        self.original_max_seq_len = config.max_position_embeddings
-
-        self.config = config
-
-        self.rope_type = self.config.rope_parameters["rope_type"]
-        rope_init_fn: Callable = self.compute_default_rope_parameters
-        if self.rope_type != "default":
-            rope_init_fn = ROPE_INIT_FUNCTIONS[self.rope_type]
-        inv_freq, self.attention_scaling = rope_init_fn(self.config, device)
-
-        self.register_buffer("inv_freq", inv_freq, persistent=False)
-        self.register_buffer("original_inv_freq", inv_freq.clone(), persistent=False)
-
-    @staticmethod
-    def compute_default_rope_parameters(
-        config: ErkConfig | None = None,
-        device: Optional["torch.device"] = None,
-        seq_len: int | None = None,
-    ) -> tuple["torch.Tensor", float]:
-        """
-        Computes the inverse frequencies according to the original RoPE implementation
-        Args:
-            config ([`~transformers.PreTrainedConfig`]):
-                The model configuration.
-            device (`torch.device`):
-                The device to use for initialization of the inverse frequencies.
-            seq_len (`int`, *optional*):
-                The current sequence length. Unused for this type of RoPE.
-        Returns:
-            Tuple of (`torch.Tensor`, `float`), containing the inverse frequencies for the RoPE embeddings and the
-            post-processing scaling factor applied to the computed cos/sin (unused in this type of RoPE).
-        """
-        base = config.rope_parameters["rope_theta"]
-        dim = getattr(config, "head_dim", None) or config.hidden_size // config.num_attention_heads
-
-        attention_factor = 1.0  # Unused in this type of RoPE
-
-        # Compute the inverse frequencies
-        inv_freq = 1.0 / (
-            base ** (torch.arange(0, dim, 2, dtype=torch.int64).to(device=device, dtype=torch.float) / dim)
-        )
-        return inv_freq, attention_factor
-
-    @torch.no_grad()
-    @dynamic_rope_update  # power user: used with advanced RoPE types (e.g. dynamic rope)
-    def forward(self, x, position_ids):
-        inv_freq_expanded = self.inv_freq[None, :, None].float().expand(position_ids.shape[0], -1, 1).to(x.device)
-        position_ids_expanded = position_ids[:, None, :].float()
-
-        device_type = x.device.type if isinstance(x.device.type, str) and x.device.type != "mps" else "cpu"
-        with maybe_autocast(device_type=device_type, enabled=False):  # Force float32
-            freqs = (inv_freq_expanded.float() @ position_ids_expanded.float()).transpose(1, 2)
-            emb = torch.cat((freqs, freqs), dim=-1)
-            cos = emb.cos() * self.attention_scaling
-            sin = emb.sin() * self.attention_scaling
-
-        return cos.to(dtype=x.dtype), sin.to(dtype=x.dtype)
-
-
-def rotate_half(x):
-    """Rotates half the hidden dims of the input."""
-    x1 = x[..., : x.shape[-1] // 2]
-    x2 = x[..., x.shape[-1] // 2 :]
-    return torch.cat((-x2, x1), dim=-1)
-
-
-@use_kernel_func_from_hub("rotary_pos_emb")
-def apply_rotary_pos_emb(q, k, cos, sin, unsqueeze_dim=1):
-    """Applies Rotary Position Embedding to the query and key tensors.
-
-    Args:
-        q (`torch.Tensor`): The query tensor.
-        k (`torch.Tensor`): The key tensor.
-        cos (`torch.Tensor`): The cosine part of the rotary embedding.
-        sin (`torch.Tensor`): The sine part of the rotary embedding.
-        unsqueeze_dim (`int`, *optional*, defaults to 1):
-            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
-            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
-            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
-            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
-            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
-            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
-    Returns:
-        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
-    """
-    cos = cos.unsqueeze(unsqueeze_dim)
-    sin = sin.unsqueeze(unsqueeze_dim)
-    q_embed = (q * cos) + (rotate_half(q) * sin)
-    k_embed = (k * cos) + (rotate_half(k) * sin)
-    return q_embed, k_embed
-
-
-def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
-    """
-    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
-    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
-    """
-    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
-    if n_rep == 1:
-        return hidden_states
-    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
-    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
-
-
-def eager_attention_forward(
-    module: nn.Module,
-    query: torch.Tensor,
-    key: torch.Tensor,
-    value: torch.Tensor,
-    attention_mask: torch.Tensor | None,
-    scaling: float,
-    dropout: float = 0.0,
-    **kwargs: Unpack[TransformersKwargs],
-):
-    key_states = repeat_kv(key, module.num_key_value_groups)
-    value_states = repeat_kv(value, module.num_key_value_groups)
-
-    attn_weights = torch.matmul(query, key_states.transpose(2, 3)) * scaling
-    if attention_mask is not None:
-        attn_weights = attn_weights + attention_mask
-
-    attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query.dtype)
-    attn_weights = nn.functional.dropout(attn_weights, p=dropout, training=module.training)
-    attn_output = torch.matmul(attn_weights, value_states)
-    attn_output = attn_output.transpose(1, 2).contiguous()
-
-    return attn_output, attn_weights
-
-
-@use_kernelized_func(apply_rotary_pos_emb)
-class ErkAttention(nn.Module):
-    """Multi-headed attention from 'Attention Is All You Need' paper"""
-
-    def __init__(self, config: ErkConfig, layer_idx: int):
-        super().__init__()
-        self.layer_type = config.layer_types[layer_idx] if hasattr(config, "layer_types") else None
-        self.config = config
-        self.layer_idx = layer_idx
-        self.head_dim = getattr(config, "head_dim", config.hidden_size // config.num_attention_heads)
-        self.num_key_value_groups = config.num_attention_heads // config.num_key_value_heads
-        self.scaling = self.head_dim**-0.5
-        self.attention_dropout = config.attention_dropout
-        self.is_causal = True
-
-        self.q_proj = nn.Linear(
-            config.hidden_size, config.num_attention_heads * self.head_dim, bias=config.attention_bias
-        )
-        self.k_proj = nn.Linear(
-            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
-        )
-        self.v_proj = nn.Linear(
-            config.hidden_size, config.num_key_value_heads * self.head_dim, bias=config.attention_bias
-        )
-        self.o_proj = nn.Linear(
-            config.num_attention_heads * self.head_dim, config.hidden_size, bias=config.attention_bias
-        )
-        self.q_norm = ErkRMSNorm(self.head_dim, eps=config.rms_norm_eps)  # unlike olmo, only on the head dim!
-        self.k_norm = ErkRMSNorm(self.head_dim, eps=config.rms_norm_eps)  # thus post q_norm does not need reshape
-        self.sliding_window = config.sliding_window if self.layer_type == "sliding_attention" else None
-
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        position_embeddings: tuple[torch.Tensor, torch.Tensor],
-        attention_mask: torch.Tensor | None,
-        past_key_values: Cache | None = None,
-        **kwargs: Unpack[FlashAttentionKwargs],
-    ) -> tuple[torch.Tensor, torch.Tensor | None]:
-        input_shape = hidden_states.shape[:-1]
-        hidden_shape = (*input_shape, -1, self.head_dim)
-
-        query_states = self.q_norm(self.q_proj(hidden_states).view(hidden_shape)).transpose(1, 2)
-        key_states = self.k_norm(self.k_proj(hidden_states).view(hidden_shape)).transpose(1, 2)
-        value_states = self.v_proj(hidden_states).view(hidden_shape).transpose(1, 2)
-
-        cos, sin = position_embeddings
-        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin)
-
-        if past_key_values is not None:
-            key_states, value_states = past_key_values.update(key_states, value_states, self.layer_idx)
-
-        attention_interface: Callable = ALL_ATTENTION_FUNCTIONS.get_interface(
-            self.config._attn_implementation, eager_attention_forward
-        )
-
-        attn_output, attn_weights = attention_interface(
-            self,
-            query_states,
-            key_states,
-            value_states,
-            attention_mask,
-            dropout=0.0 if not self.training else self.attention_dropout,
-            scaling=self.scaling,
-            sliding_window=self.sliding_window,  # diff with Llama
-            **kwargs,
-        )
-
-        attn_output = attn_output.reshape(*input_shape, -1).contiguous()
-        attn_output = self.o_proj(attn_output)
-        return attn_output, attn_weights
-
-
-class ErkDecoderLayer(GradientCheckpointingLayer):
-    def __init__(self, config: ErkConfig, layer_idx: int):
-        super().__init__()
-        self.hidden_size = config.hidden_size
-
-        self.self_attn = ErkAttention(config=config, layer_idx=layer_idx)
-
-        self.mlp = ErkMLP(config)
-        self.input_layernorm = ErkRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
-        self.post_attention_layernorm = ErkRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
-
-    def forward(
-        self,
-        hidden_states: torch.Tensor,
-        attention_mask: torch.Tensor | None = None,
-        position_ids: torch.LongTensor | None = None,
-        past_key_values: Cache | None = None,
-        use_cache: bool | None = False,
-        position_embeddings: tuple[torch.Tensor, torch.Tensor] | None = None,
-        **kwargs: Unpack[TransformersKwargs],
-    ) -> torch.Tensor:
-        residual = hidden_states
-        hidden_states = self.input_layernorm(hidden_states)
-        # Self Attention
-        hidden_states, _ = self.self_attn(
-            hidden_states=hidden_states,
-            attention_mask=attention_mask,
-            position_ids=position_ids,
-            past_key_values=past_key_values,
-            use_cache=use_cache,
-            position_embeddings=position_embeddings,
-            **kwargs,
-        )
-        hidden_states = residual + hidden_states
-
-        # Fully Connected
-        residual = hidden_states
-        hidden_states = self.post_attention_layernorm(hidden_states)
-        hidden_states = self.mlp(hidden_states)
-        hidden_states = residual + hidden_states
-        return hidden_states
-
-
-@auto_docstring
-class ErkPreTrainedModel(PreTrainedModel):
-    config: ErkConfig
-    base_model_prefix = "model"
-    supports_gradient_checkpointing = True
-    _no_split_modules = ["ErkDecoderLayer"]
-    _skip_keys_device_placement = ["past_key_values"]
-    _supports_flash_attn = True
-    _supports_sdpa = True
-    _supports_flex_attn = True
-
-    _can_compile_fullgraph = True
-    _supports_attention_backend = True
-    _can_record_outputs = {
-        "hidden_states": ErkDecoderLayer,
-        "attentions": ErkAttention,
-    }
-
-
-@auto_docstring
-class ErkModel(ErkPreTrainedModel):
-    def __init__(self, config: ErkConfig):
-        super().__init__(config)
-        self.padding_idx = config.pad_token_id
-        self.vocab_size = config.vocab_size
-
-        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
-        self.layers = nn.ModuleList(
-            [ErkDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
-        )
-        self.norm = ErkRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
-        self.rotary_emb = ErkRotaryEmbedding(config=config)
-        self.gradient_checkpointing = False
-        self.has_sliding_layers = "sliding_attention" in self.config.layer_types
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    @merge_with_config_defaults
-    @capture_outputs
-    @auto_docstring
-    def forward(
-        self,
-        input_ids: torch.LongTensor | None = None,
-        attention_mask: torch.Tensor | None = None,
-        position_ids: torch.LongTensor | None = None,
-        past_key_values: Cache | None = None,
-        inputs_embeds: torch.FloatTensor | None = None,
-        use_cache: bool | None = None,
-        **kwargs: Unpack[TransformersKwargs],
-    ) -> BaseModelOutputWithPast:
-        if (input_ids is None) ^ (inputs_embeds is not None):
-            raise ValueError("You must specify exactly one of input_ids or inputs_embeds")
-
-        if inputs_embeds is None:
-            inputs_embeds = self.embed_tokens(input_ids)
-
-        if use_cache and past_key_values is None:
-            past_key_values = DynamicCache(config=self.config)
-
-        if position_ids is None:
-            past_seen_tokens = past_key_values.get_seq_length() if past_key_values is not None else 0
-            position_ids = torch.arange(inputs_embeds.shape[1], device=inputs_embeds.device) + past_seen_tokens
-            position_ids = position_ids.unsqueeze(0)
-
-        # It may already have been prepared by e.g. `generate`
-        if not isinstance(causal_mask_mapping := attention_mask, dict):
-            # Prepare mask arguments
-            mask_kwargs = {
-                "config": self.config,
-                "inputs_embeds": inputs_embeds,
-                "attention_mask": attention_mask,
-                "past_key_values": past_key_values,
-                "position_ids": position_ids,
-            }
-            # Create the masks
-            causal_mask_mapping = {
-                "full_attention": create_causal_mask(**mask_kwargs),
-            }
-            # The sliding window alternating layers are not always activated depending on the config
-            if self.has_sliding_layers:
-                causal_mask_mapping["sliding_attention"] = create_sliding_window_causal_mask(**mask_kwargs)
-
-        hidden_states = inputs_embeds
-        position_embeddings = self.rotary_emb(hidden_states, position_ids)
-
-        for i, decoder_layer in enumerate(self.layers[: self.config.num_hidden_layers]):
-            hidden_states = decoder_layer(
-                hidden_states,
-                attention_mask=causal_mask_mapping[self.config.layer_types[i]],
-                position_embeddings=position_embeddings,
-                position_ids=position_ids,
-                past_key_values=past_key_values,
-                use_cache=use_cache,
-                **kwargs,
-            )
-
-        hidden_states = self.norm(hidden_states)
-        return BaseModelOutputWithPast(
-            last_hidden_state=hidden_states,
-            past_key_values=past_key_values if use_cache else None,
-        )
-
-
-@auto_docstring
-class ErkForCausalLM(ErkPreTrainedModel, GenerationMixin):
-    _tied_weights_keys = {"lm_head.weight": "model.embed_tokens.weight"}
-    _tp_plan = {"lm_head": "colwise_gather_output"}
-    _pp_plan = {"lm_head": (["hidden_states"], ["logits"])}
-
-    def __init__(self, config):
-        super().__init__(config)
-        self.model = ErkModel(config)
-        self.vocab_size = config.vocab_size
-        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    @can_return_tuple
-    @auto_docstring
-    def forward(
-        self,
-        input_ids: torch.LongTensor | None = None,
-        attention_mask: torch.Tensor | None = None,
-        position_ids: torch.LongTensor | None = None,
-        past_key_values: Cache | None = None,
-        inputs_embeds: torch.FloatTensor | None = None,
-        labels: torch.LongTensor | None = None,
-        use_cache: bool | None = None,
-        logits_to_keep: int | torch.Tensor = 0,
-        **kwargs: Unpack[TransformersKwargs],
-    ) -> CausalLMOutputWithPast:
-        r"""
-        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
-            Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
-            config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
-            (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
-
-        Example:
-
-        ```python
-        >>> from transformers import AutoTokenizer, ErkForCausalLM
-
-        >>> model = ErkForCausalLM.from_pretrained("ecloudtech/erk")
-        >>> tokenizer = AutoTokenizer.from_pretrained("ecloudtech/erk")
-
-        >>> prompt = "Hey, are you conscious? Can you talk to me?"
-        >>> inputs = tokenizer(prompt, return_tensors="pt")
-
-        >>> # Generate
-        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
-        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
-        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
-        ```"""
-        outputs: BaseModelOutputWithPast = self.model(
-            input_ids=input_ids,
-            attention_mask=attention_mask,
-            position_ids=position_ids,
-            past_key_values=past_key_values,
-            inputs_embeds=inputs_embeds,
-            use_cache=use_cache,
-            **kwargs,
-        )
-
-        hidden_states = outputs.last_hidden_state
-        # Only compute necessary logits, and do not upcast them to float if we are not computing the loss
-        slice_indices = slice(-logits_to_keep, None) if isinstance(logits_to_keep, int) else logits_to_keep
-        logits = self.lm_head(hidden_states[:, slice_indices, :])
-
-        loss = None
-        if labels is not None:
-            loss = self.loss_function(logits=logits, labels=labels, vocab_size=self.config.vocab_size, **kwargs)
-
-        return CausalLMOutputWithPast(
-            loss=loss,
-            logits=logits,
-            past_key_values=outputs.past_key_values,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-
-class ErkForSequenceClassification(GenericForSequenceClassification, ErkPreTrainedModel):
-    pass
-
-
-class ErkForTokenClassification(GenericForTokenClassification, ErkPreTrainedModel):
-    pass
-
-
-class ErkForQuestionAnswering(GenericForQuestionAnswering, ErkPreTrainedModel):
-    base_model_prefix = "transformer"  # For BC, where `transformer` was used instead of `model`
-
-
-__all__ = [
-    "ErkForCausalLM",
-    "ErkForQuestionAnswering",
-    "ErkPreTrainedModel",
-    "ErkModel",
-    "ErkForSequenceClassification",
-    "ErkForTokenClassification",
-]