OPT 2.7b Prompt Tuned

config.json

@@ -0,0 +1,35 @@
+{
+  "_name_or_path": "facebook/opt-2.7b",
+  "_remove_final_layer_norm": false,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "architectures": [
+    "OPT_PromptTuned_For_SentimentAnalysis"
+  ],
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoModel": "prompt_tuned_opt.OPT_PromptTuned_For_SentimentAnalysis"
+  },
+  "bos_token_id": 2,
+  "do_layer_norm_before": true,
+  "dropout": 0.1,
+  "enable_bias": true,
+  "eos_token_id": 1,
+  "ffn_dim": 10240,
+  "hidden_size": 2560,
+  "init_std": 0.02,
+  "layer_norm_elementwise_affine": true,
+  "layerdrop": 0.0,
+  "max_position_embeddings": 2048,
+  "model_type": "opt",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "pad_token_id": 1,
+  "prefix": "</s>",
+  "prompt_dict_path": "./prompt_embedding",
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.1",
+  "use_cache": true,
+  "vocab_size": 50272,
+  "word_embed_proj_dim": 2560
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 2,
+  "eos_token_id": 1,
+  "pad_token_id": 1,
+  "transformers_version": "4.33.1"
+}

prompt_tuned_opt.py

@@ -0,0 +1,239 @@
+from transformers import  AutoConfig, AutoModelForCausalLM, AutoTokenizer, PreTrainedModel
+import torch
+import torch.nn as nn
+from typing import List, Optional, Tuple, Union
+
+from transformers.utils import (
+    add_code_sample_docstrings,
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    logging,
+    replace_return_docstrings,
+)
+
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPast,
+    CausalLMOutputWithPast,
+    QuestionAnsweringModelOutput,
+    SequenceClassifierOutputWithPast,
+)
+
+from transformers import OPTConfig
+from transformers.models.opt.modeling_opt import OPTModel
+from transformers.models.opt.modeling_opt import OPTPreTrainedModel
+
+class OPT_PromptTuned_For_SentimentAnalysis(OPTPreTrainedModel):
+    _tied_weights_keys = ["lm_head.weight"]
+    _CONFIG_FOR_DOC = "OPTConfig"
+    config_class = OPTConfig
+
+
+    def __init__(self, config):
+
+        self.config = config
+        super().__init__(config)
+        self.model = OPTModel(config)
+        self.lm_head = nn.Linear(config.word_embed_proj_dim, config.vocab_size, bias=False)
+        self.embedding = nn.Embedding(8, config.word_embed_proj_dim)
+        self.post_init()
+
+
+    def get_input_embeddings(self):
+        return self.model.decoder.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.decoder.embed_tokens = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model.decoder = decoder
+
+    def get_decoder(self):
+        return self.model.decoder
+
+    def load_prompts(self):
+        self.embedding.load_state_dict(torch.load(self.config.prompt_dict_path))
+        return self
+
+    @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        head_mask: Optional[torch.Tensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None
+    ) -> Union[Tuple, CausalLMOutputWithPast]:
+        r"""
+        Args:
+            input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
+                Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you
+                provide it.
+
+                Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+                [`PreTrainedTokenizer.__call__`] for details.
+
+                [What are input IDs?](../glossary#input-ids)
+            attention_mask (`torch.Tensor` 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?](../glossary#attention-mask)
+            head_mask (`torch.Tensor` of shape `(num_hidden_layers, num_attention_heads)`, *optional*):
+                Mask to nullify selected heads of the attention modules. Mask values selected in `[0, 1]`:
+
+                - 1 indicates the head is **not masked**,
+                - 0 indicates the head is **masked**.
+
+            past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
+                Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of
+                shape `(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of
+                shape `(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`. The two additional
+                tensors are only required when the model is used as a decoder in a Sequence to Sequence model.
+
+                Contains pre-computed hidden-states (key and values in the self-attention blocks and in the
+                cross-attention blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
+
+                If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those
+                that don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of
+                all `decoder_input_ids` of shape `(batch_size, sequence_length)`.
+            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.
+            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]`.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            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 [`~utils.ModelOutput`] instead of a plain tuple.
+
+        Returns:
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer, OPTForCausalLM
+
+        >>> model = OPTForCausalLM.from_pretrained("facebook/opt-350m")
+        >>> tokenizer = AutoTokenizer.from_pretrained("facebook/opt-350m")
+
+        >>> 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. I'm just a little bit of a weirdo."
+        ```"""
+
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+
+
+        outputs = self.model.decoder(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            head_mask=head_mask,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=True,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict
+        )
+
+        logits = self.lm_head(outputs[0]).contiguous()
+
+        loss = None
+        if labels is not None:
+            # move labels to correct device to enable model parallelism
+            labels = labels.to(logits.device)
+            # Shift so that tokens < n predict n
+            shift_logits = logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss()
+            loss = loss_fct(shift_logits.view(-1, self.config.vocab_size), shift_labels.view(-1))
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return (loss,) + output if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+    def prepare_inputs_for_generation(
+        self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs
+    ):
+            
+        if past_key_values:
+            input_ids = input_ids[:, -1:]
+        
+        if inputs_embeds is not None and past_key_values is None:
+            input = torch.tensor([0,1,2,3,4,5,6,7]).to(inputs_embeds.device)
+            inputs_embeds = torch.cat([self.embedding(input).unsqueeze(0), inputs_embeds], dim=1)
+            attention_mask = torch.cat(
+                        [torch.ones((attention_mask.shape[0], 8), 
+                                    device=attention_mask.device), 
+                         attention_mask], dim=1)
+            model_inputs = {"inputs_embeds": inputs_embeds}
+        else:
+            model_inputs = {"input_ids": input_ids}
+            attention_mask = torch.cat(
+            [torch.ones((attention_mask.shape[0], 8), 
+                        device=attention_mask.device), 
+             attention_mask], dim=1)
+
+        model_inputs.update(
+            {
+                "past_key_values": past_key_values,
+                "use_cache": kwargs.get("use_cache"),
+                "attention_mask": attention_mask,
+            }
+        )
+        return model_inputs
+
+    def generate(self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs):
+        max_new_tokens = 3
+        input_embeddings = self.get_input_embeddings()(input_ids).to(input_ids.device)
+        # Call the original generate method
+        return super().generate(input_ids=input_ids, inputs_embeds=input_embeddings,max_new_tokens=max_new_tokens,attention_mask=attention_mask, **kwargs)
+
+    @staticmethod
+    def _reorder_cache(past_key_values, beam_idx):
+        reordered_past = ()
+        for layer_past in past_key_values:
+            reordered_past += (tuple(past_state.index_select(0, beam_idx) for past_state in layer_past),)
+        return reordered_past

pytorch_model-00001-of-00002.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a212da4c1bfe10de10138ba74183d2bd080301b911fa774c238a73e57e9026c
+size 9977150467

pytorch_model-00002-of-00002.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b819842995682ab6473d63013d778c3d7cb55f34bfff5a7fe3036d6cfc97ccf7
+size 629505249

pytorch_model.bin.index.json

@@ -0,0 +1,525 @@
+{
+  "metadata": {
+    "total_size": 10606469120
+  },
+  "weight_map": {
+    "embedding.weight": "pytorch_model-00002-of-00002.bin",
+    "lm_head.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.embed_positions.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.embed_tokens.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.0.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.1.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.10.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.11.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.12.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.13.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.14.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.15.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.16.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.17.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.18.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.19.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.2.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.20.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.21.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.22.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.23.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.24.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.25.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.26.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.27.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.28.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.29.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.3.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.30.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.final_layer_norm.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.final_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn_layer_norm.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.30.self_attn_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.fc1.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.fc1.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.fc2.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.fc2.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.final_layer_norm.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.final_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.k_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.k_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.out_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.out_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.q_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.q_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.v_proj.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn.v_proj.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn_layer_norm.bias": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.31.self_attn_layer_norm.weight": "pytorch_model-00002-of-00002.bin",
+    "model.decoder.layers.4.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.4.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.5.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.6.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.7.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.8.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.fc1.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.fc1.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.fc2.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.fc2.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.final_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.final_layer_norm.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.k_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.out_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.out_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.q_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.v_proj.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn_layer_norm.bias": "pytorch_model-00001-of-00002.bin",
+    "model.decoder.layers.9.self_attn_layer_norm.weight": "pytorch_model-00001-of-00002.bin"
+  }
+}