config.json CHANGED
@@ -20,17 +20,10 @@
20
  "num_hidden_layers": 60,
21
  "num_key_value_heads": 40,
22
  "pad_token_id": 2,
23
- "pretraining_tp": 1,
24
  "rms_norm_eps": 1e-06,
25
- "rope_scaling": null,
26
- "rope_theta": 10000.0,
27
  "tie_word_embeddings": false,
28
- "torch_dtype": "bfloat16",
29
- "transformers_version": "4.33.1",
30
  "use_cache": true,
31
- "vocab_size": 103168,
32
- "rotary": {
33
- "base": 10000,
34
- "type": "dynamic"
35
- }
36
  }
 
20
  "num_hidden_layers": 60,
21
  "num_key_value_heads": 40,
22
  "pad_token_id": 2,
 
23
  "rms_norm_eps": 1e-06,
 
 
24
  "tie_word_embeddings": false,
25
+ "torch_dtype": "float16",
26
+ "transformers_version": "4.33.2",
27
  "use_cache": true,
28
+ "vocab_size": 103168
 
 
 
 
29
  }
configuration_internlm.py CHANGED
@@ -19,8 +19,9 @@
19
  # limitations under the License.
20
  """ InternLM model configuration"""
21
 
22
- from transformers.configuration_utils import PretrainedConfig
23
  from transformers.utils import logging
 
 
24
 
25
  logger = logging.get_logger(__name__)
26
 
@@ -29,9 +30,9 @@ INTERNLM_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
29
 
30
  class InternLMConfig(PretrainedConfig):
31
  r"""
32
- This is the configuration class to store the configuration of a [`InternLMModel`]. It is used to instantiate
33
- an InternLM model according to the specified arguments, defining the model architecture. Instantiating a
34
- configuration with the defaults will yield a similar configuration to that of the InternLM-7B.
35
 
36
  Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
37
  documentation from [`PretrainedConfig`] for more information.
@@ -49,6 +50,19 @@ class InternLMConfig(PretrainedConfig):
49
  Number of hidden layers in the Transformer encoder.
50
  num_attention_heads (`int`, *optional*, defaults to 32):
51
  Number of attention heads for each attention layer in the Transformer encoder.
 
 
 
 
 
 
 
 
 
 
 
 
 
52
  hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
53
  The non-linear activation function (function or string) in the decoder.
54
  max_position_embeddings (`int`, *optional*, defaults to 2048):
@@ -80,13 +94,14 @@ class InternLMConfig(PretrainedConfig):
80
  model_type = "internlm"
81
  _auto_class = "AutoConfig"
82
 
83
- def __init__( # pylint: disable=W0102
84
  self,
85
  vocab_size=103168,
86
  hidden_size=4096,
87
  intermediate_size=11008,
88
  num_hidden_layers=32,
89
  num_attention_heads=32,
 
90
  hidden_act="silu",
91
  max_position_embeddings=2048,
92
  initializer_range=0.02,
@@ -97,7 +112,6 @@ class InternLMConfig(PretrainedConfig):
97
  eos_token_id=2,
98
  tie_word_embeddings=False,
99
  bias=True,
100
- rotary={"base": 10000, "type": "dynamic"}, # pylint: disable=W0102
101
  **kwargs,
102
  ):
103
  self.vocab_size = vocab_size
@@ -106,16 +120,20 @@ class InternLMConfig(PretrainedConfig):
106
  self.intermediate_size = intermediate_size
107
  self.num_hidden_layers = num_hidden_layers
108
  self.num_attention_heads = num_attention_heads
 
 
 
 
 
109
  self.hidden_act = hidden_act
110
  self.initializer_range = initializer_range
111
  self.rms_norm_eps = rms_norm_eps
112
  self.use_cache = use_cache
113
  self.bias = bias
114
- self.rotary = rotary
115
  super().__init__(
116
  pad_token_id=pad_token_id,
117
  bos_token_id=bos_token_id,
118
  eos_token_id=eos_token_id,
119
  tie_word_embeddings=tie_word_embeddings,
120
  **kwargs,
121
- )
 
19
  # limitations under the License.
20
  """ InternLM model configuration"""
21
 
 
22
  from transformers.utils import logging
23
+ from transformers.configuration_utils import PretrainedConfig
24
+
25
 
26
  logger = logging.get_logger(__name__)
27
 
 
30
 
31
  class InternLMConfig(PretrainedConfig):
32
  r"""
33
+ This is the configuration class to store the configuration of a [`InternLMModel`]. It is used to instantiate an InternLM
34
+ model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
35
+ defaults will yield a similar configuration to that of the InternLM-7B.
36
 
37
  Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
38
  documentation from [`PretrainedConfig`] for more information.
 
50
  Number of hidden layers in the Transformer encoder.
51
  num_attention_heads (`int`, *optional*, defaults to 32):
52
  Number of attention heads for each attention layer in the Transformer encoder.
53
+ num_key_value_heads (`int`, *optional*):
54
+ This is the number of key_value heads that should be used to implement Grouped Query Attention. If
55
+ `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
56
+ `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
57
+ converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
58
+ by meanpooling all the original heads within that group. For more details checkout [this
59
+ paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
60
+ `num_attention_heads`.
61
+ pretraining_tp (`int`, *optional*, defaults to `1`):
62
+ Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
63
+ document](https://huggingface.co/docs/transformers/parallelism) to understand more about it. This value is
64
+ necessary to ensure exact reproducibility of the pretraining results. Please refer to [this
65
+ issue](https://github.com/pytorch/pytorch/issues/76232).
66
  hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
67
  The non-linear activation function (function or string) in the decoder.
68
  max_position_embeddings (`int`, *optional*, defaults to 2048):
 
94
  model_type = "internlm"
95
  _auto_class = "AutoConfig"
96
 
97
+ def __init__(
98
  self,
99
  vocab_size=103168,
100
  hidden_size=4096,
101
  intermediate_size=11008,
102
  num_hidden_layers=32,
103
  num_attention_heads=32,
104
+ num_key_value_heads=None,
105
  hidden_act="silu",
106
  max_position_embeddings=2048,
107
  initializer_range=0.02,
 
112
  eos_token_id=2,
113
  tie_word_embeddings=False,
114
  bias=True,
 
115
  **kwargs,
116
  ):
117
  self.vocab_size = vocab_size
 
120
  self.intermediate_size = intermediate_size
121
  self.num_hidden_layers = num_hidden_layers
122
  self.num_attention_heads = num_attention_heads
123
+
124
+ if num_key_value_heads is None:
125
+ num_key_value_heads = num_attention_heads
126
+ self.num_key_value_heads = num_key_value_heads
127
+
128
  self.hidden_act = hidden_act
129
  self.initializer_range = initializer_range
130
  self.rms_norm_eps = rms_norm_eps
131
  self.use_cache = use_cache
132
  self.bias = bias
 
133
  super().__init__(
134
  pad_token_id=pad_token_id,
135
  bos_token_id=bos_token_id,
136
  eos_token_id=eos_token_id,
137
  tie_word_embeddings=tie_word_embeddings,
138
  **kwargs,
139
+ )
generation_config.json CHANGED
@@ -2,5 +2,6 @@
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
- "transformers_version": "4.33.1"
 
6
  }
 
2
  "_from_model_config": true,
3
  "bos_token_id": 1,
4
  "eos_token_id": 2,
5
+ "pad_token_id": 2,
6
+ "transformers_version": "4.33.2"
7
  }
modeling_internlm.py CHANGED
@@ -19,36 +19,26 @@
19
  # limitations under the License.
20
  """ PyTorch InternLM model."""
21
  import math
22
- import queue
23
- import threading
24
  from typing import List, Optional, Tuple, Union
 
25
 
26
  import torch
27
  import torch.utils.checkpoint
28
  from torch import nn
29
  from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
 
30
  from transformers.activations import ACT2FN
31
- from transformers.generation.streamers import BaseStreamer
32
- from transformers.modeling_outputs import (
33
- BaseModelOutputWithPast,
34
- CausalLMOutputWithPast,
35
- SequenceClassifierOutputWithPast,
36
- )
37
  from transformers.modeling_utils import PreTrainedModel
38
- from transformers.utils import (
39
- add_start_docstrings,
40
- add_start_docstrings_to_model_forward,
41
- logging,
42
- replace_return_docstrings,
43
- )
44
-
45
  from .configuration_internlm import InternLMConfig
46
 
 
47
  logger = logging.get_logger(__name__)
48
 
49
  _CONFIG_FOR_DOC = "InternLMConfig"
50
 
51
-
52
  # Copied from transformers.models.bart.modeling_bart._make_causal_mask
53
  def _make_causal_mask(
54
  input_ids_shape: torch.Size, dtype: torch.dtype, device: torch.device, past_key_values_length: int = 0
@@ -81,10 +71,17 @@ def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Optional[int]
81
 
82
  return inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(dtype).min)
83
 
 
 
 
 
 
 
 
 
 
84
 
85
  class InternLMRMSNorm(nn.Module):
86
- """RMSNorm implemention."""
87
-
88
  def __init__(self, hidden_size, eps=1e-6):
89
  """
90
  InternLMRMSNorm is equivalent to T5LayerNorm
@@ -105,14 +102,6 @@ class InternLMRMSNorm(nn.Module):
105
 
106
 
107
  class InternLMRotaryEmbedding(torch.nn.Module):
108
- """Implement InternLM's rotary embedding.
109
-
110
- Args:
111
- dim (int): Characteristic dimension of each self-attentional head.
112
- max_position_embeddings (int, optional): Model's training length. Defaults to 2048.
113
- base (int, optional): The rotation position encodes the rotation Angle base number. Defaults to 10000.
114
- device (Any, optional): Running device. Defaults to None.
115
- """
116
  def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None):
117
  super().__init__()
118
  inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float().to(device) / dim))
@@ -144,66 +133,6 @@ class InternLMRotaryEmbedding(torch.nn.Module):
144
  )
145
 
146
 
147
- class InternLMDynamicNTKScalingRotaryEmbedding(torch.nn.Module):
148
- """Implement InternLM's DyanmicNTK extrapolation method, thereby broadening the model support context to 16K.
149
-
150
- Args:
151
- dim (int): Characteristic dimension of each self-attentional head.
152
- max_position_embeddings (int, optional): Model's training length. Defaults to 2048.
153
- base (int, optional): The rotation position encodes the rotation Angle base number. Defaults to 10000.
154
- device (Any, optional): Running device. Defaults to None.
155
- scaling_factor (float, optional): NTK method extrapolation coefficient. Defaults to 1.0.
156
- """
157
-
158
- def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None, scaling_factor=1.0):
159
- super().__init__()
160
- inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float().to(device) / dim))
161
- self.register_buffer("inv_freq", inv_freq)
162
- self.dim = dim
163
- self.base = base
164
- self.scaling_factor = scaling_factor
165
-
166
- # Build here to make `torch.jit.trace` work.
167
- self.max_position_embeddings = max_position_embeddings
168
- self.max_seq_len_cached = max_position_embeddings
169
- t = torch.arange(self.max_seq_len_cached, device=self.inv_freq.device, dtype=self.inv_freq.dtype)
170
- freqs = torch.einsum("i,j->ij", t, self.inv_freq)
171
- # Different from paper, but it uses a different permutation in order to obtain the same calculation
172
- emb = torch.cat((freqs, freqs), dim=-1)
173
- self.register_buffer("cos_cached", emb.cos()[None, None, :, :], persistent=False)
174
- self.register_buffer("sin_cached", emb.sin()[None, None, :, :], persistent=False)
175
-
176
- def _update_cached(self, x, seq_len=None):
177
- self.max_seq_len_cached = max(seq_len, self.max_position_embeddings)
178
- if seq_len > self.max_position_embeddings:
179
- base = self.base * (
180
- (self.scaling_factor * seq_len / self.max_position_embeddings) - (self.scaling_factor - 1)
181
- ) ** (self.dim / (self.dim - 2))
182
- inv_freq = 1.0 / (base ** (torch.arange(0, self.dim, 2).float().to(x.device) / self.dim))
183
- else:
184
- inv_freq = self.inv_freq
185
- t = torch.arange(self.max_seq_len_cached, device=inv_freq.device, dtype=inv_freq.dtype)
186
- freqs = torch.einsum("i,j->ij", t, inv_freq)
187
- emb = torch.cat((freqs, freqs), dim=-1)
188
- self.register_buffer("cos_cached", emb.cos()[None, None, :, :], persistent=False)
189
- self.register_buffer("sin_cached", emb.sin()[None, None, :, :], persistent=False)
190
-
191
- def forward(self, x, seq_len=None):
192
- # x: [bs, num_attention_heads, seq_len, head_size]
193
- # This `if` block is unlikely to be run after we build sin/cos in `__init__`. Keep the logic here just in case.
194
- if seq_len <= self.max_position_embeddings:
195
- # Reset the tables if the sequence length has changed,
196
- if self.max_seq_len_cached > self.max_position_embeddings:
197
- self._update_cached(x, seq_len)
198
- else:
199
- self._update_cached(x, seq_len)
200
-
201
- return (
202
- self.cos_cached[:, :, :seq_len, ...].to(dtype=x.dtype),
203
- self.sin_cached[:, :, :seq_len, ...].to(dtype=x.dtype),
204
- )
205
-
206
-
207
  def rotate_half(x):
208
  """Rotates half the hidden dims of the input."""
209
  x1 = x[..., : x.shape[-1] // 2]
@@ -215,18 +144,10 @@ def apply_rotary_pos_emb(q, k, cos, sin, position_ids):
215
  # The first two dimensions of cos and sin are always 1, so we can `squeeze` them.
216
  cos = cos.squeeze(1).squeeze(0) # [seq_len, dim]
217
  sin = sin.squeeze(1).squeeze(0) # [seq_len, dim]
218
- cos = cos.unsqueeze(0).unsqueeze(0).expand(len(position_ids), -1, -1, -1)
219
- sin = sin.unsqueeze(0).unsqueeze(0).expand(len(position_ids), -1, -1, -1)
220
- if q.size(2) == 1:
221
- q_embed = (q * cos[:, :, -1, :]) + (rotate_half(q) * sin[:, :, -1, :])
222
- else:
223
- q_embed = (q * cos) + (rotate_half(q) * sin)
224
-
225
- if k.size(2) == 1:
226
- k_embed = (k * cos[:, :, -1, :]) + (rotate_half(k) * sin[:, :, -1, :])
227
- else:
228
- k_embed = (k * cos) + (rotate_half(k) * sin)
229
-
230
  return q_embed, k_embed
231
 
232
 
@@ -256,6 +177,8 @@ class InternLMAttention(nn.Module):
256
  self.hidden_size = config.hidden_size
257
  self.num_heads = config.num_attention_heads
258
  self.head_dim = self.hidden_size // self.num_heads
 
 
259
  self.max_position_embeddings = config.max_position_embeddings
260
 
261
  if (self.head_dim * self.num_heads) != self.hidden_size:
@@ -264,28 +187,10 @@ class InternLMAttention(nn.Module):
264
  f" and `num_heads`: {self.num_heads})."
265
  )
266
  self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
267
- self.k_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
268
- self.v_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
269
  self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=config.bias)
270
- self.rotary_emb = self._init_rope()
271
-
272
- def _init_rope(self):
273
- if self.config.rotary["type"] == "origin":
274
- self.rotary_emb = InternLMRotaryEmbedding(
275
- self.head_dim,
276
- max_position_embeddings=self.max_position_embeddings,
277
- base=self.config.rotary["base"],
278
- )
279
- elif self.config.rotary["type"] == "dynamic":
280
- self.rotary_emb = InternLMDynamicNTKScalingRotaryEmbedding(
281
- self.head_dim,
282
- max_position_embeddings=self.max_position_embeddings,
283
- base=self.config.rotary["base"],
284
- scaling_factor=self.config.rotary.get("scaling_factor", 1.0),
285
- )
286
- else:
287
- raise ValueError("Currently we only support rotary embedding's type being one of ('origin', 'dynamic').")
288
- return self.rotary_emb
289
 
290
  def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
291
  return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
@@ -302,20 +207,25 @@ class InternLMAttention(nn.Module):
302
  bsz, q_len, _ = hidden_states.size()
303
 
304
  query_states = self.q_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
305
- key_states = self.k_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
306
- value_states = self.v_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
 
 
 
 
 
 
 
307
 
308
  if past_key_value is not None:
309
  # reuse k, v, self_attention
310
  key_states = torch.cat([past_key_value[0], key_states], dim=2)
311
  value_states = torch.cat([past_key_value[1], value_states], dim=2)
312
 
313
- # print(use_cache)
314
  past_key_value = (key_states, value_states) if use_cache else None
315
 
316
- kv_seq_len = key_states.shape[-2]
317
- cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
318
- query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
319
 
320
  attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)
321
 
@@ -426,9 +336,11 @@ INTERNLM_START_DOCSTRING = r"""
426
  This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
427
  library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
428
  etc.)
 
429
  This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
430
  Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
431
  and behavior.
 
432
  Parameters:
433
  config ([`InternLMConfig`]):
434
  Model configuration class with all the parameters of the model. Initializing with a config file does not
@@ -469,34 +381,44 @@ INTERNLM_INPUTS_DOCSTRING = r"""
469
  input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
470
  Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
471
  it.
 
472
  Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
473
  [`PreTrainedTokenizer.__call__`] for details.
 
474
  [What are input IDs?](../glossary#input-ids)
475
  attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
476
  Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
 
477
  - 1 for tokens that are **not masked**,
478
  - 0 for tokens that are **masked**.
 
479
  [What are attention masks?](../glossary#attention-mask)
 
480
  Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
481
  [`PreTrainedTokenizer.__call__`] for details.
 
482
  If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
483
  `past_key_values`).
 
484
  If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
485
  and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
486
  information on the default strategy.
 
487
  - 1 indicates the head is **not masked**,
488
  - 0 indicates the head is **masked**.
489
  position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
490
  Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
491
  config.n_positions - 1]`.
 
492
  [What are position IDs?](../glossary#position-ids)
493
- past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or
494
- when `config.use_cache=True`):
495
  Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
496
  `(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of shape
497
  `(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`.
 
498
  Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
499
  blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
 
500
  If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
501
  don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
502
  `decoder_input_ids` of shape `(batch_size, sequence_length)`.
@@ -525,10 +447,10 @@ INTERNLM_INPUTS_DOCSTRING = r"""
525
  class InternLMModel(InternLMPreTrainedModel):
526
  """
527
  Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`InternLMDecoderLayer`]
 
528
  Args:
529
  config: InternLMConfig
530
  """
531
-
532
  _auto_class = "AutoModel"
533
 
534
  def __init__(self, config: InternLMConfig):
@@ -754,14 +676,20 @@ class InternLMForCausalLM(InternLMPreTrainedModel):
754
  Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
755
  config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
756
  (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
 
757
  Returns:
 
758
  Example:
 
759
  ```python
760
  >>> from transformers import AutoTokenizer, InternLMForCausalLM
 
761
  >>> model = InternLMForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
762
  >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
 
763
  >>> prompt = "Hey, are you consciours? Can you talk to me?"
764
  >>> inputs = tokenizer(prompt, return_tensors="pt")
 
765
  >>> # Generate
766
  >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
767
  >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
@@ -851,56 +779,50 @@ class InternLMForCausalLM(InternLMPreTrainedModel):
851
  for layer_past in past_key_values:
852
  reordered_past += (tuple(past_state.index_select(0, beam_idx) for past_state in layer_past),)
853
  return reordered_past
854
-
855
  def build_inputs(self, tokenizer, query: str, history: List[Tuple[str, str]] = []):
856
  prompt = ""
857
  for record in history:
858
  prompt += f"""<|User|>:{record[0]}<eoh>\n<|Bot|>:{record[1]}<eoa>\n"""
859
  prompt += f"""<|User|>:{query}<eoh>\n<|Bot|>:"""
860
  return tokenizer([prompt], return_tensors="pt")
861
-
862
  @torch.no_grad()
863
- def chat(
864
- self,
865
- tokenizer,
866
- query: str,
867
- history: List[Tuple[str, str]] = [],
868
- streamer: Optional[BaseStreamer] = None,
869
- max_new_tokens: int = 1024,
870
- do_sample: bool = True,
871
- temperature: float = 0.8,
872
- top_p: float = 0.8,
873
- **kwargs,
874
- ):
875
  inputs = self.build_inputs(tokenizer, query, history)
876
  inputs = {k: v.to(self.device) for k, v in inputs.items() if torch.is_tensor(v)}
877
- outputs = self.generate(
878
- **inputs,
879
- streamer=streamer,
880
- max_new_tokens=max_new_tokens,
881
- do_sample=do_sample,
882
- temperature=temperature,
883
- top_p=top_p,
884
- **kwargs,
885
- )
886
- outputs = outputs[0].cpu().tolist()[len(inputs["input_ids"][0]) :]
887
  response = tokenizer.decode(outputs, skip_special_tokens=True)
888
  response = response.split("<eoa>")[0]
889
  history = history + [(query, response)]
890
  return response, history
891
-
892
  @torch.no_grad()
893
- def stream_chat(
894
- self,
895
- tokenizer,
896
- query: str,
897
- history: List[Tuple[str, str]] = [],
898
- max_new_tokens: int = 1024,
899
- do_sample: bool = True,
900
- temperature: float = 0.8,
901
- top_p: float = 0.8,
902
- **kwargs,
903
- ):
904
  """
905
  Return a generator in format: (response, history)
906
  Eg.
@@ -946,12 +868,12 @@ class InternLMForCausalLM(InternLMPreTrainedModel):
946
  tokenizer=tokenizer,
947
  query=query,
948
  streamer=ChatStreamer(tokenizer=tokenizer),
949
- history=history,
950
  max_new_tokens=max_new_tokens,
951
  do_sample=do_sample,
952
  temperature=temperature,
953
  top_p=top_p,
954
- **kwargs,
955
  )
956
 
957
  def consumer():
@@ -969,8 +891,10 @@ class InternLMForCausalLM(InternLMPreTrainedModel):
969
  @add_start_docstrings(
970
  """
971
  The InternLM Model transformer with a sequence classification head on top (linear layer).
 
972
  [`InternLMForSequenceClassification`] uses the last token in order to do the classification, as other causal models
973
  (e.g. GPT-2) do.
 
974
  Since it does classification on the last token, it requires to know the position of the last token. If a
975
  `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
976
  no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
 
19
  # limitations under the License.
20
  """ PyTorch InternLM model."""
21
  import math
 
 
22
  from typing import List, Optional, Tuple, Union
23
+ import threading, queue
24
 
25
  import torch
26
  import torch.utils.checkpoint
27
  from torch import nn
28
  from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
29
+
30
  from transformers.activations import ACT2FN
31
+ from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast, SequenceClassifierOutputWithPast
 
 
 
 
 
32
  from transformers.modeling_utils import PreTrainedModel
33
+ from transformers.generation.streamers import BaseStreamer
34
+ from transformers.utils import add_start_docstrings, add_start_docstrings_to_model_forward, logging, replace_return_docstrings
 
 
 
 
 
35
  from .configuration_internlm import InternLMConfig
36
 
37
+
38
  logger = logging.get_logger(__name__)
39
 
40
  _CONFIG_FOR_DOC = "InternLMConfig"
41
 
 
42
  # Copied from transformers.models.bart.modeling_bart._make_causal_mask
43
  def _make_causal_mask(
44
  input_ids_shape: torch.Size, dtype: torch.dtype, device: torch.device, past_key_values_length: int = 0
 
71
 
72
  return inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(dtype).min)
73
 
74
+ def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
75
+ """
76
+ (batch, num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
77
+ """
78
+ batch, num_key_value_heads, slen, head_dim = hidden_states.shape
79
+ if n_rep == 1:
80
+ return hidden_states
81
+ hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
82
+ return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
83
 
84
  class InternLMRMSNorm(nn.Module):
 
 
85
  def __init__(self, hidden_size, eps=1e-6):
86
  """
87
  InternLMRMSNorm is equivalent to T5LayerNorm
 
102
 
103
 
104
  class InternLMRotaryEmbedding(torch.nn.Module):
 
 
 
 
 
 
 
 
105
  def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None):
106
  super().__init__()
107
  inv_freq = 1.0 / (base ** (torch.arange(0, dim, 2).float().to(device) / dim))
 
133
  )
134
 
135
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
136
  def rotate_half(x):
137
  """Rotates half the hidden dims of the input."""
138
  x1 = x[..., : x.shape[-1] // 2]
 
144
  # The first two dimensions of cos and sin are always 1, so we can `squeeze` them.
145
  cos = cos.squeeze(1).squeeze(0) # [seq_len, dim]
146
  sin = sin.squeeze(1).squeeze(0) # [seq_len, dim]
147
+ cos = cos[position_ids].unsqueeze(1) # [bs, 1, seq_len, dim]
148
+ sin = sin[position_ids].unsqueeze(1) # [bs, 1, seq_len, dim]
149
+ q_embed = (q * cos) + (rotate_half(q) * sin)
150
+ k_embed = (k * cos) + (rotate_half(k) * sin)
 
 
 
 
 
 
 
 
151
  return q_embed, k_embed
152
 
153
 
 
177
  self.hidden_size = config.hidden_size
178
  self.num_heads = config.num_attention_heads
179
  self.head_dim = self.hidden_size // self.num_heads
180
+ self.num_key_value_heads = config.num_key_value_heads
181
+ self.num_key_value_groups = self.num_heads // self.num_key_value_heads
182
  self.max_position_embeddings = config.max_position_embeddings
183
 
184
  if (self.head_dim * self.num_heads) != self.hidden_size:
 
187
  f" and `num_heads`: {self.num_heads})."
188
  )
189
  self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.bias)
190
+ self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=False)
191
+ self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=False)
192
  self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=config.bias)
193
+ self.rotary_emb = InternLMRotaryEmbedding(self.head_dim, max_position_embeddings=self.max_position_embeddings)
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
194
 
195
  def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
196
  return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
 
207
  bsz, q_len, _ = hidden_states.size()
208
 
209
  query_states = self.q_proj(hidden_states).view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
210
+ key_states = self.k_proj(hidden_states).view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
211
+ value_states = self.v_proj(hidden_states).view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
212
+
213
+ kv_seq_len = key_states.shape[-2]
214
+ if past_key_value is not None:
215
+ kv_seq_len += past_key_value[0].shape[-2]
216
+ cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
217
+ query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
218
+ # [bsz, nh, t, hd]
219
 
220
  if past_key_value is not None:
221
  # reuse k, v, self_attention
222
  key_states = torch.cat([past_key_value[0], key_states], dim=2)
223
  value_states = torch.cat([past_key_value[1], value_states], dim=2)
224
 
 
225
  past_key_value = (key_states, value_states) if use_cache else None
226
 
227
+ key_states = repeat_kv(key_states, self.num_key_value_groups)
228
+ value_states = repeat_kv(value_states, self.num_key_value_groups)
 
229
 
230
  attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)
231
 
 
336
  This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
337
  library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
338
  etc.)
339
+
340
  This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
341
  Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
342
  and behavior.
343
+
344
  Parameters:
345
  config ([`InternLMConfig`]):
346
  Model configuration class with all the parameters of the model. Initializing with a config file does not
 
381
  input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
382
  Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
383
  it.
384
+
385
  Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
386
  [`PreTrainedTokenizer.__call__`] for details.
387
+
388
  [What are input IDs?](../glossary#input-ids)
389
  attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
390
  Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
391
+
392
  - 1 for tokens that are **not masked**,
393
  - 0 for tokens that are **masked**.
394
+
395
  [What are attention masks?](../glossary#attention-mask)
396
+
397
  Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
398
  [`PreTrainedTokenizer.__call__`] for details.
399
+
400
  If `past_key_values` is used, optionally only the last `decoder_input_ids` have to be input (see
401
  `past_key_values`).
402
+
403
  If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
404
  and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
405
  information on the default strategy.
406
+
407
  - 1 indicates the head is **not masked**,
408
  - 0 indicates the head is **masked**.
409
  position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
410
  Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
411
  config.n_positions - 1]`.
412
+
413
  [What are position IDs?](../glossary#position-ids)
414
+ past_key_values (`tuple(tuple(torch.FloatTensor))`, *optional*, returned when `use_cache=True` is passed or when `config.use_cache=True`):
 
415
  Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of shape
416
  `(batch_size, num_heads, sequence_length, embed_size_per_head)`) and 2 additional tensors of shape
417
  `(batch_size, num_heads, encoder_sequence_length, embed_size_per_head)`.
418
+
419
  Contains pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
420
  blocks) that can be used (see `past_key_values` input) to speed up sequential decoding.
421
+
422
  If `past_key_values` are used, the user can optionally input only the last `decoder_input_ids` (those that
423
  don't have their past key value states given to this model) of shape `(batch_size, 1)` instead of all
424
  `decoder_input_ids` of shape `(batch_size, sequence_length)`.
 
447
  class InternLMModel(InternLMPreTrainedModel):
448
  """
449
  Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`InternLMDecoderLayer`]
450
+
451
  Args:
452
  config: InternLMConfig
453
  """
 
454
  _auto_class = "AutoModel"
455
 
456
  def __init__(self, config: InternLMConfig):
 
676
  Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
677
  config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
678
  (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
679
+
680
  Returns:
681
+
682
  Example:
683
+
684
  ```python
685
  >>> from transformers import AutoTokenizer, InternLMForCausalLM
686
+
687
  >>> model = InternLMForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
688
  >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
689
+
690
  >>> prompt = "Hey, are you consciours? Can you talk to me?"
691
  >>> inputs = tokenizer(prompt, return_tensors="pt")
692
+
693
  >>> # Generate
694
  >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
695
  >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
 
779
  for layer_past in past_key_values:
780
  reordered_past += (tuple(past_state.index_select(0, beam_idx) for past_state in layer_past),)
781
  return reordered_past
782
+
783
  def build_inputs(self, tokenizer, query: str, history: List[Tuple[str, str]] = []):
784
  prompt = ""
785
  for record in history:
786
  prompt += f"""<|User|>:{record[0]}<eoh>\n<|Bot|>:{record[1]}<eoa>\n"""
787
  prompt += f"""<|User|>:{query}<eoh>\n<|Bot|>:"""
788
  return tokenizer([prompt], return_tensors="pt")
789
+
790
  @torch.no_grad()
791
+ def chat(self,
792
+ tokenizer,
793
+ query: str,
794
+ history: List[Tuple[str, str]] = [],
795
+ streamer: Optional[BaseStreamer] = None,
796
+ max_new_tokens: int = 1024,
797
+ do_sample: bool = True,
798
+ temperature: float = 0.8,
799
+ top_p: float = 0.8,
800
+ **kwargs):
 
 
801
  inputs = self.build_inputs(tokenizer, query, history)
802
  inputs = {k: v.to(self.device) for k, v in inputs.items() if torch.is_tensor(v)}
803
+ outputs = self.generate(**inputs,
804
+ streamer=streamer,
805
+ max_new_tokens=max_new_tokens,
806
+ do_sample=do_sample,
807
+ temperature=temperature,
808
+ top_p=top_p,
809
+ **kwargs)
810
+ outputs = outputs[0].cpu().tolist()[len(inputs["input_ids"][0]):]
 
 
811
  response = tokenizer.decode(outputs, skip_special_tokens=True)
812
  response = response.split("<eoa>")[0]
813
  history = history + [(query, response)]
814
  return response, history
815
+
816
  @torch.no_grad()
817
+ def stream_chat(self,
818
+ tokenizer,
819
+ query: str,
820
+ history: List[Tuple[str, str]] = [],
821
+ max_new_tokens: int = 1024,
822
+ do_sample: bool = True,
823
+ temperature: float = 0.8,
824
+ top_p: float = 0.8,
825
+ **kwargs):
 
 
826
  """
827
  Return a generator in format: (response, history)
828
  Eg.
 
868
  tokenizer=tokenizer,
869
  query=query,
870
  streamer=ChatStreamer(tokenizer=tokenizer),
871
+ history=history,
872
  max_new_tokens=max_new_tokens,
873
  do_sample=do_sample,
874
  temperature=temperature,
875
  top_p=top_p,
876
+ **kwargs
877
  )
878
 
879
  def consumer():
 
891
  @add_start_docstrings(
892
  """
893
  The InternLM Model transformer with a sequence classification head on top (linear layer).
894
+
895
  [`InternLMForSequenceClassification`] uses the last token in order to do the classification, as other causal models
896
  (e.g. GPT-2) do.
897
+
898
  Since it does classification on the last token, it requires to know the position of the last token. If a
899
  `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
900
  no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
pytorch_model-00001-of-00005.bin → pytorch_model-00001-of-00006.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:aeba743507872c45e7cf951d7996bce448d8deada841d055d2ac03948af0c2b7
3
- size 9990647029
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c989b1624a481672a7018455d7ff95398ded2a07698ccf2687877db91baf254
3
+ size 7893395149
pytorch_model-00002-of-00005.bin → pytorch_model-00002-of-00006.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a11c8737fce8d6be9a8f6eb0faa44016c94813aed1d50a757ca32abece4ed461
3
- size 9956594199
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11c9b3fc955587d5ea525c787d7677602e0f3d70131259b3c12079e034e68132
3
+ size 7964241876
pytorch_model-00003-of-00005.bin → pytorch_model-00003-of-00006.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5c64167ce104e9a576da50f89a398ac2124734621c45e12ea0addbac99ad87ac
3
- size 9867486361
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:366595a002cc2ce217aec0c4885e7c5f840df751155f9e2510e6472e171c02d2
3
+ size 7896062197
pytorch_model-00004-of-00005.bin → pytorch_model-00004-of-00006.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:40e22421695e3206bc85f0a4839641370bc8277ab689ff0e5d75e708d51f8691
3
- size 9306483281
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:508cfed19500ecf7678f1680a47a1073b73f8ad5597612c094c8b2e7df8d3931
3
+ size 7964241876
pytorch_model-00005-of-00006.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4182de7c0df21447a5b7ed8cbb68162e85be061f23ddab73b8be157049fb9e31
3
+ size 7403239886
pytorch_model-00005-of-00005.bin → pytorch_model-00006-of-00006.bin RENAMED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:263f29c6331d8951fd454d4bbd2991d422bbcfb5b07d4acbb0e75aaf53b1a76c
3
  size 1056441258
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89b0631c7069213a49dd8a3cb9012e52d82ac9328bd96e2bba8383d825720039
3
  size 1056441258
pytorch_model.bin.index.json CHANGED
@@ -3,548 +3,548 @@
3
  "total_size": 40177428480
4
  },
5
  "weight_map": {
6
- "lm_head.weight": "pytorch_model-00005-of-00005.bin",
7
- "model.embed_tokens.weight": "pytorch_model-00001-of-00005.bin",
8
- "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
9
- "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
10
- "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
11
- "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
12
- "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
13
- "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
14
- "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
15
- "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
16
- "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
17
- "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
18
- "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
19
- "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
20
- "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
21
- "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
22
- "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
23
- "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
24
- "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
25
- "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
26
- "model.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
27
- "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
28
- "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
29
- "model.layers.10.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
30
- "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
31
- "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
32
- "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
33
- "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
34
- "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
35
- "model.layers.11.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
36
- "model.layers.11.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
37
- "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
38
- "model.layers.11.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
39
- "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
40
- "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
41
- "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
42
- "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
43
- "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
44
- "model.layers.12.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
45
- "model.layers.12.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
46
- "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
47
- "model.layers.12.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
48
- "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
49
- "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
50
- "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
51
- "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
52
- "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
53
- "model.layers.13.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
54
- "model.layers.13.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
55
- "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
56
- "model.layers.13.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
57
- "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
58
- "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
59
- "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
60
- "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
61
- "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
62
- "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
63
- "model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
64
- "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
65
- "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
66
- "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
67
- "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
68
- "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
69
- "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
70
- "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
71
- "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
72
- "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
73
- "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
74
- "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
75
- "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
76
- "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
77
- "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
78
- "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
79
- "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
80
- "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
81
- "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
82
- "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
83
- "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
84
- "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
85
- "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
86
- "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
87
- "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
88
- "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
89
- "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
90
- "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
91
- "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
92
- "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
93
- "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
94
- "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
95
- "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
96
- "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
97
- "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
98
- "model.layers.18.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
99
- "model.layers.18.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
100
- "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
101
- "model.layers.18.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
102
- "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
103
- "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
104
- "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
105
- "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
106
- "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
107
- "model.layers.19.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
108
- "model.layers.19.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
109
- "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
110
- "model.layers.19.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
111
- "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
112
- "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
113
- "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
114
- "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
115
- "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
116
- "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
117
- "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
118
- "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
119
- "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
120
- "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
121
- "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
122
- "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
123
- "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
124
- "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
125
- "model.layers.20.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
126
- "model.layers.20.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
127
- "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
128
- "model.layers.20.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
129
- "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
130
- "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
131
- "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
132
- "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
133
- "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
134
- "model.layers.21.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
135
- "model.layers.21.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
136
- "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
137
- "model.layers.21.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
138
- "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
139
- "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
140
- "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
141
- "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
142
- "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
143
- "model.layers.22.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
144
- "model.layers.22.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
145
- "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
146
- "model.layers.22.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
147
- "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
148
- "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
149
- "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
150
- "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
151
- "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
152
- "model.layers.23.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
153
- "model.layers.23.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
154
- "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
155
- "model.layers.23.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
156
- "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
157
- "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
158
- "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
159
- "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
160
- "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
161
- "model.layers.24.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
162
- "model.layers.24.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
163
- "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
164
- "model.layers.24.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
165
- "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
166
- "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
167
- "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
168
- "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
169
- "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
170
- "model.layers.25.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
171
- "model.layers.25.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
172
- "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
173
- "model.layers.25.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
174
- "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
175
- "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
176
- "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
177
- "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
178
- "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
179
- "model.layers.26.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
180
- "model.layers.26.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
181
- "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
182
- "model.layers.26.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
183
- "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
184
- "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
185
- "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
186
- "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
187
- "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
188
- "model.layers.27.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
189
- "model.layers.27.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
190
- "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
191
- "model.layers.27.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
192
- "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
193
- "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
194
- "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
195
- "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
196
- "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
197
- "model.layers.28.input_layernorm.weight": "pytorch_model-00002-of-00005.bin",
198
- "model.layers.28.mlp.down_proj.weight": "pytorch_model-00002-of-00005.bin",
199
- "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
200
- "model.layers.28.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
201
- "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00002-of-00005.bin",
202
- "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
203
- "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
204
- "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
205
- "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
206
- "model.layers.29.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
207
- "model.layers.29.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
208
- "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00002-of-00005.bin",
209
- "model.layers.29.mlp.up_proj.weight": "pytorch_model-00002-of-00005.bin",
210
- "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
211
- "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00002-of-00005.bin",
212
- "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00002-of-00005.bin",
213
- "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00002-of-00005.bin",
214
- "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00002-of-00005.bin",
215
- "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
216
- "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
217
- "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
218
- "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
219
- "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
220
- "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
221
- "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
222
- "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
223
- "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
224
- "model.layers.30.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
225
- "model.layers.30.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
226
- "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
227
- "model.layers.30.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
228
- "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
229
- "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
230
- "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
231
- "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
232
- "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
233
- "model.layers.31.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
234
- "model.layers.31.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
235
- "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
236
- "model.layers.31.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
237
- "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
238
- "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
239
- "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
240
- "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
241
- "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
242
- "model.layers.32.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
243
- "model.layers.32.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
244
- "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
245
- "model.layers.32.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
246
- "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
247
- "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
248
- "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
249
- "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
250
- "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
251
- "model.layers.33.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
252
- "model.layers.33.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
253
- "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
254
- "model.layers.33.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
255
- "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
256
- "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
257
- "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
258
- "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
259
- "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
260
- "model.layers.34.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
261
- "model.layers.34.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
262
- "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
263
- "model.layers.34.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
264
- "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
265
- "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
266
- "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
267
- "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
268
- "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
269
- "model.layers.35.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
270
- "model.layers.35.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
271
- "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
272
- "model.layers.35.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
273
- "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
274
- "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
275
- "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
276
- "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
277
- "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
278
- "model.layers.36.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
279
- "model.layers.36.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
280
- "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
281
- "model.layers.36.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
282
- "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
283
- "model.layers.36.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
284
- "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
285
- "model.layers.36.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
286
- "model.layers.36.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
287
- "model.layers.37.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
288
- "model.layers.37.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
289
- "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
290
- "model.layers.37.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
291
- "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
292
- "model.layers.37.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
293
- "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
294
- "model.layers.37.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
295
- "model.layers.37.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
296
- "model.layers.38.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
297
- "model.layers.38.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
298
- "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
299
- "model.layers.38.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
300
- "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
301
- "model.layers.38.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
302
- "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
303
- "model.layers.38.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
304
- "model.layers.38.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
305
- "model.layers.39.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
306
- "model.layers.39.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
307
- "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
308
- "model.layers.39.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
309
- "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
310
- "model.layers.39.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
311
- "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
312
- "model.layers.39.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
313
- "model.layers.39.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
314
- "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
315
- "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
316
- "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
317
- "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
318
- "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
319
- "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
320
- "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
321
- "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
322
- "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
323
- "model.layers.40.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
324
- "model.layers.40.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
325
- "model.layers.40.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
326
- "model.layers.40.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
327
- "model.layers.40.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
328
- "model.layers.40.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
329
- "model.layers.40.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
330
- "model.layers.40.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
331
- "model.layers.40.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
332
- "model.layers.41.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
333
- "model.layers.41.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
334
- "model.layers.41.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
335
- "model.layers.41.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
336
- "model.layers.41.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
337
- "model.layers.41.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
338
- "model.layers.41.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
339
- "model.layers.41.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
340
- "model.layers.41.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
341
- "model.layers.42.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
342
- "model.layers.42.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
343
- "model.layers.42.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
344
- "model.layers.42.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
345
- "model.layers.42.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
346
- "model.layers.42.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
347
- "model.layers.42.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
348
- "model.layers.42.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
349
- "model.layers.42.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
350
- "model.layers.43.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
351
- "model.layers.43.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
352
- "model.layers.43.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
353
- "model.layers.43.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
354
- "model.layers.43.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
355
- "model.layers.43.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
356
- "model.layers.43.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
357
- "model.layers.43.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
358
- "model.layers.43.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
359
- "model.layers.44.input_layernorm.weight": "pytorch_model-00003-of-00005.bin",
360
- "model.layers.44.mlp.down_proj.weight": "pytorch_model-00003-of-00005.bin",
361
- "model.layers.44.mlp.gate_proj.weight": "pytorch_model-00003-of-00005.bin",
362
- "model.layers.44.mlp.up_proj.weight": "pytorch_model-00003-of-00005.bin",
363
- "model.layers.44.post_attention_layernorm.weight": "pytorch_model-00003-of-00005.bin",
364
- "model.layers.44.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
365
- "model.layers.44.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
366
- "model.layers.44.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
367
- "model.layers.44.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
368
- "model.layers.45.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
369
- "model.layers.45.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
370
- "model.layers.45.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
371
- "model.layers.45.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
372
- "model.layers.45.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
373
- "model.layers.45.self_attn.k_proj.weight": "pytorch_model-00003-of-00005.bin",
374
- "model.layers.45.self_attn.o_proj.weight": "pytorch_model-00003-of-00005.bin",
375
- "model.layers.45.self_attn.q_proj.weight": "pytorch_model-00003-of-00005.bin",
376
- "model.layers.45.self_attn.v_proj.weight": "pytorch_model-00003-of-00005.bin",
377
- "model.layers.46.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
378
- "model.layers.46.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
379
- "model.layers.46.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
380
- "model.layers.46.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
381
- "model.layers.46.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
382
- "model.layers.46.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
383
- "model.layers.46.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
384
- "model.layers.46.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
385
- "model.layers.46.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
386
- "model.layers.47.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
387
- "model.layers.47.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
388
- "model.layers.47.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
389
- "model.layers.47.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
390
- "model.layers.47.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
391
- "model.layers.47.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
392
- "model.layers.47.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
393
- "model.layers.47.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
394
- "model.layers.47.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
395
- "model.layers.48.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
396
- "model.layers.48.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
397
- "model.layers.48.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
398
- "model.layers.48.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
399
- "model.layers.48.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
400
- "model.layers.48.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
401
- "model.layers.48.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
402
- "model.layers.48.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
403
- "model.layers.48.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
404
- "model.layers.49.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
405
- "model.layers.49.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
406
- "model.layers.49.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
407
- "model.layers.49.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
408
- "model.layers.49.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
409
- "model.layers.49.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
410
- "model.layers.49.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
411
- "model.layers.49.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
412
- "model.layers.49.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
413
- "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
414
- "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
415
- "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
416
- "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
417
- "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
418
- "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
419
- "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
420
- "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
421
- "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
422
- "model.layers.50.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
423
- "model.layers.50.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
424
- "model.layers.50.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
425
- "model.layers.50.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
426
- "model.layers.50.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
427
- "model.layers.50.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
428
- "model.layers.50.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
429
- "model.layers.50.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
430
- "model.layers.50.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
431
- "model.layers.51.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
432
- "model.layers.51.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
433
- "model.layers.51.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
434
- "model.layers.51.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
435
- "model.layers.51.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
436
- "model.layers.51.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
437
- "model.layers.51.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
438
- "model.layers.51.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
439
- "model.layers.51.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
440
- "model.layers.52.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
441
- "model.layers.52.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
442
- "model.layers.52.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
443
- "model.layers.52.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
444
- "model.layers.52.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
445
- "model.layers.52.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
446
- "model.layers.52.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
447
- "model.layers.52.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
448
- "model.layers.52.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
449
- "model.layers.53.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
450
- "model.layers.53.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
451
- "model.layers.53.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
452
- "model.layers.53.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
453
- "model.layers.53.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
454
- "model.layers.53.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
455
- "model.layers.53.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
456
- "model.layers.53.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
457
- "model.layers.53.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
458
- "model.layers.54.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
459
- "model.layers.54.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
460
- "model.layers.54.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
461
- "model.layers.54.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
462
- "model.layers.54.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
463
- "model.layers.54.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
464
- "model.layers.54.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
465
- "model.layers.54.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
466
- "model.layers.54.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
467
- "model.layers.55.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
468
- "model.layers.55.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
469
- "model.layers.55.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
470
- "model.layers.55.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
471
- "model.layers.55.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
472
- "model.layers.55.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
473
- "model.layers.55.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
474
- "model.layers.55.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
475
- "model.layers.55.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
476
- "model.layers.56.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
477
- "model.layers.56.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
478
- "model.layers.56.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
479
- "model.layers.56.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
480
- "model.layers.56.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
481
- "model.layers.56.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
482
- "model.layers.56.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
483
- "model.layers.56.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
484
- "model.layers.56.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
485
- "model.layers.57.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
486
- "model.layers.57.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
487
- "model.layers.57.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
488
- "model.layers.57.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
489
- "model.layers.57.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
490
- "model.layers.57.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
491
- "model.layers.57.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
492
- "model.layers.57.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
493
- "model.layers.57.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
494
- "model.layers.58.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
495
- "model.layers.58.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
496
- "model.layers.58.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
497
- "model.layers.58.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
498
- "model.layers.58.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
499
- "model.layers.58.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
500
- "model.layers.58.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
501
- "model.layers.58.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
502
- "model.layers.58.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
503
- "model.layers.59.input_layernorm.weight": "pytorch_model-00004-of-00005.bin",
504
- "model.layers.59.mlp.down_proj.weight": "pytorch_model-00004-of-00005.bin",
505
- "model.layers.59.mlp.gate_proj.weight": "pytorch_model-00004-of-00005.bin",
506
- "model.layers.59.mlp.up_proj.weight": "pytorch_model-00004-of-00005.bin",
507
- "model.layers.59.post_attention_layernorm.weight": "pytorch_model-00004-of-00005.bin",
508
- "model.layers.59.self_attn.k_proj.weight": "pytorch_model-00004-of-00005.bin",
509
- "model.layers.59.self_attn.o_proj.weight": "pytorch_model-00004-of-00005.bin",
510
- "model.layers.59.self_attn.q_proj.weight": "pytorch_model-00004-of-00005.bin",
511
- "model.layers.59.self_attn.v_proj.weight": "pytorch_model-00004-of-00005.bin",
512
- "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
513
- "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
514
- "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
515
- "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
516
- "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
517
- "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
518
- "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
519
- "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
520
- "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
521
- "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
522
- "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
523
- "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
524
- "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
525
- "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
526
- "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
527
- "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
528
- "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
529
- "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
530
- "model.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
531
- "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
532
- "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
533
- "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
534
- "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
535
- "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
536
- "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
537
- "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
538
- "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
539
- "model.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00005.bin",
540
- "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00005.bin",
541
- "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00005.bin",
542
- "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00005.bin",
543
- "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00005.bin",
544
- "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00005.bin",
545
- "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00005.bin",
546
- "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00005.bin",
547
- "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00005.bin",
548
- "model.norm.weight": "pytorch_model-00004-of-00005.bin"
549
  }
550
  }
 
3
  "total_size": 40177428480
4
  },
5
  "weight_map": {
6
+ "lm_head.weight": "pytorch_model-00006-of-00006.bin",
7
+ "model.embed_tokens.weight": "pytorch_model-00001-of-00006.bin",
8
+ "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
9
+ "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
10
+ "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
11
+ "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
12
+ "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
13
+ "model.layers.0.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
14
+ "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
15
+ "model.layers.0.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
16
+ "model.layers.0.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
17
+ "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
18
+ "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
19
+ "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
20
+ "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
21
+ "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
22
+ "model.layers.1.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
23
+ "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
24
+ "model.layers.1.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
25
+ "model.layers.1.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
26
+ "model.layers.10.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
27
+ "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
28
+ "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
29
+ "model.layers.10.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
30
+ "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
31
+ "model.layers.10.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
32
+ "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
33
+ "model.layers.10.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
34
+ "model.layers.10.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
35
+ "model.layers.11.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
36
+ "model.layers.11.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
37
+ "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
38
+ "model.layers.11.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
39
+ "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
40
+ "model.layers.11.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
41
+ "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
42
+ "model.layers.11.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
43
+ "model.layers.11.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
44
+ "model.layers.12.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
45
+ "model.layers.12.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
46
+ "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
47
+ "model.layers.12.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
48
+ "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
49
+ "model.layers.12.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
50
+ "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
51
+ "model.layers.12.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
52
+ "model.layers.12.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
53
+ "model.layers.13.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
54
+ "model.layers.13.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
55
+ "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
56
+ "model.layers.13.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
57
+ "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
58
+ "model.layers.13.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
59
+ "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
60
+ "model.layers.13.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
61
+ "model.layers.13.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
62
+ "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
63
+ "model.layers.14.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
64
+ "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
65
+ "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
66
+ "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
67
+ "model.layers.14.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
68
+ "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
69
+ "model.layers.14.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
70
+ "model.layers.14.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
71
+ "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
72
+ "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
73
+ "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
74
+ "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
75
+ "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
76
+ "model.layers.15.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
77
+ "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
78
+ "model.layers.15.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
79
+ "model.layers.15.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
80
+ "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
81
+ "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
82
+ "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
83
+ "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
84
+ "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
85
+ "model.layers.16.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
86
+ "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
87
+ "model.layers.16.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
88
+ "model.layers.16.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
89
+ "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
90
+ "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
91
+ "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
92
+ "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
93
+ "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
94
+ "model.layers.17.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
95
+ "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
96
+ "model.layers.17.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
97
+ "model.layers.17.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
98
+ "model.layers.18.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
99
+ "model.layers.18.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
100
+ "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
101
+ "model.layers.18.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
102
+ "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
103
+ "model.layers.18.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
104
+ "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
105
+ "model.layers.18.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
106
+ "model.layers.18.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
107
+ "model.layers.19.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
108
+ "model.layers.19.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
109
+ "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
110
+ "model.layers.19.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
111
+ "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
112
+ "model.layers.19.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
113
+ "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
114
+ "model.layers.19.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
115
+ "model.layers.19.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
116
+ "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
117
+ "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
118
+ "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
119
+ "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
120
+ "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
121
+ "model.layers.2.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
122
+ "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
123
+ "model.layers.2.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
124
+ "model.layers.2.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
125
+ "model.layers.20.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
126
+ "model.layers.20.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
127
+ "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
128
+ "model.layers.20.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
129
+ "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
130
+ "model.layers.20.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
131
+ "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
132
+ "model.layers.20.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
133
+ "model.layers.20.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
134
+ "model.layers.21.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
135
+ "model.layers.21.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
136
+ "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
137
+ "model.layers.21.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
138
+ "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
139
+ "model.layers.21.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
140
+ "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
141
+ "model.layers.21.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
142
+ "model.layers.21.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
143
+ "model.layers.22.input_layernorm.weight": "pytorch_model-00002-of-00006.bin",
144
+ "model.layers.22.mlp.down_proj.weight": "pytorch_model-00002-of-00006.bin",
145
+ "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00002-of-00006.bin",
146
+ "model.layers.22.mlp.up_proj.weight": "pytorch_model-00002-of-00006.bin",
147
+ "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00002-of-00006.bin",
148
+ "model.layers.22.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
149
+ "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
150
+ "model.layers.22.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
151
+ "model.layers.22.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
152
+ "model.layers.23.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
153
+ "model.layers.23.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
154
+ "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
155
+ "model.layers.23.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
156
+ "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
157
+ "model.layers.23.self_attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
158
+ "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00002-of-00006.bin",
159
+ "model.layers.23.self_attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
160
+ "model.layers.23.self_attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
161
+ "model.layers.24.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
162
+ "model.layers.24.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
163
+ "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
164
+ "model.layers.24.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
165
+ "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
166
+ "model.layers.24.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
167
+ "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
168
+ "model.layers.24.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
169
+ "model.layers.24.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
170
+ "model.layers.25.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
171
+ "model.layers.25.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
172
+ "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
173
+ "model.layers.25.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
174
+ "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
175
+ "model.layers.25.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
176
+ "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
177
+ "model.layers.25.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
178
+ "model.layers.25.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
179
+ "model.layers.26.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
180
+ "model.layers.26.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
181
+ "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
182
+ "model.layers.26.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
183
+ "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
184
+ "model.layers.26.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
185
+ "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
186
+ "model.layers.26.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
187
+ "model.layers.26.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
188
+ "model.layers.27.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
189
+ "model.layers.27.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
190
+ "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
191
+ "model.layers.27.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
192
+ "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
193
+ "model.layers.27.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
194
+ "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
195
+ "model.layers.27.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
196
+ "model.layers.27.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
197
+ "model.layers.28.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
198
+ "model.layers.28.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
199
+ "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
200
+ "model.layers.28.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
201
+ "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
202
+ "model.layers.28.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
203
+ "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
204
+ "model.layers.28.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
205
+ "model.layers.28.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
206
+ "model.layers.29.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
207
+ "model.layers.29.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
208
+ "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
209
+ "model.layers.29.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
210
+ "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
211
+ "model.layers.29.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
212
+ "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
213
+ "model.layers.29.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
214
+ "model.layers.29.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
215
+ "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
216
+ "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
217
+ "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
218
+ "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
219
+ "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
220
+ "model.layers.3.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
221
+ "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
222
+ "model.layers.3.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
223
+ "model.layers.3.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
224
+ "model.layers.30.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
225
+ "model.layers.30.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
226
+ "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
227
+ "model.layers.30.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
228
+ "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
229
+ "model.layers.30.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
230
+ "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
231
+ "model.layers.30.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
232
+ "model.layers.30.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
233
+ "model.layers.31.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
234
+ "model.layers.31.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
235
+ "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
236
+ "model.layers.31.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
237
+ "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
238
+ "model.layers.31.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
239
+ "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
240
+ "model.layers.31.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
241
+ "model.layers.31.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
242
+ "model.layers.32.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
243
+ "model.layers.32.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
244
+ "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
245
+ "model.layers.32.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
246
+ "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
247
+ "model.layers.32.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
248
+ "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
249
+ "model.layers.32.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
250
+ "model.layers.32.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
251
+ "model.layers.33.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
252
+ "model.layers.33.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
253
+ "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
254
+ "model.layers.33.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
255
+ "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
256
+ "model.layers.33.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
257
+ "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
258
+ "model.layers.33.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
259
+ "model.layers.33.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
260
+ "model.layers.34.input_layernorm.weight": "pytorch_model-00003-of-00006.bin",
261
+ "model.layers.34.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
262
+ "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
263
+ "model.layers.34.mlp.up_proj.weight": "pytorch_model-00003-of-00006.bin",
264
+ "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00003-of-00006.bin",
265
+ "model.layers.34.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
266
+ "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
267
+ "model.layers.34.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
268
+ "model.layers.34.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
269
+ "model.layers.35.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
270
+ "model.layers.35.mlp.down_proj.weight": "pytorch_model-00003-of-00006.bin",
271
+ "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00003-of-00006.bin",
272
+ "model.layers.35.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
273
+ "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
274
+ "model.layers.35.self_attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
275
+ "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00003-of-00006.bin",
276
+ "model.layers.35.self_attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
277
+ "model.layers.35.self_attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
278
+ "model.layers.36.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
279
+ "model.layers.36.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
280
+ "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
281
+ "model.layers.36.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
282
+ "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
283
+ "model.layers.36.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
284
+ "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
285
+ "model.layers.36.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
286
+ "model.layers.36.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
287
+ "model.layers.37.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
288
+ "model.layers.37.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
289
+ "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
290
+ "model.layers.37.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
291
+ "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
292
+ "model.layers.37.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
293
+ "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
294
+ "model.layers.37.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
295
+ "model.layers.37.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
296
+ "model.layers.38.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
297
+ "model.layers.38.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
298
+ "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
299
+ "model.layers.38.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
300
+ "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
301
+ "model.layers.38.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
302
+ "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
303
+ "model.layers.38.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
304
+ "model.layers.38.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
305
+ "model.layers.39.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
306
+ "model.layers.39.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
307
+ "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
308
+ "model.layers.39.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
309
+ "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
310
+ "model.layers.39.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
311
+ "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
312
+ "model.layers.39.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
313
+ "model.layers.39.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
314
+ "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
315
+ "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
316
+ "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
317
+ "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
318
+ "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
319
+ "model.layers.4.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
320
+ "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
321
+ "model.layers.4.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
322
+ "model.layers.4.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
323
+ "model.layers.40.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
324
+ "model.layers.40.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
325
+ "model.layers.40.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
326
+ "model.layers.40.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
327
+ "model.layers.40.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
328
+ "model.layers.40.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
329
+ "model.layers.40.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
330
+ "model.layers.40.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
331
+ "model.layers.40.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
332
+ "model.layers.41.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
333
+ "model.layers.41.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
334
+ "model.layers.41.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
335
+ "model.layers.41.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
336
+ "model.layers.41.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
337
+ "model.layers.41.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
338
+ "model.layers.41.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
339
+ "model.layers.41.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
340
+ "model.layers.41.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
341
+ "model.layers.42.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
342
+ "model.layers.42.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
343
+ "model.layers.42.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
344
+ "model.layers.42.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
345
+ "model.layers.42.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
346
+ "model.layers.42.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
347
+ "model.layers.42.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
348
+ "model.layers.42.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
349
+ "model.layers.42.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
350
+ "model.layers.43.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
351
+ "model.layers.43.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
352
+ "model.layers.43.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
353
+ "model.layers.43.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
354
+ "model.layers.43.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
355
+ "model.layers.43.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
356
+ "model.layers.43.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
357
+ "model.layers.43.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
358
+ "model.layers.43.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
359
+ "model.layers.44.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
360
+ "model.layers.44.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
361
+ "model.layers.44.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
362
+ "model.layers.44.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
363
+ "model.layers.44.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
364
+ "model.layers.44.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
365
+ "model.layers.44.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
366
+ "model.layers.44.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
367
+ "model.layers.44.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
368
+ "model.layers.45.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
369
+ "model.layers.45.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
370
+ "model.layers.45.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
371
+ "model.layers.45.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
372
+ "model.layers.45.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
373
+ "model.layers.45.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
374
+ "model.layers.45.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
375
+ "model.layers.45.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
376
+ "model.layers.45.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
377
+ "model.layers.46.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
378
+ "model.layers.46.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
379
+ "model.layers.46.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
380
+ "model.layers.46.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
381
+ "model.layers.46.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
382
+ "model.layers.46.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
383
+ "model.layers.46.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
384
+ "model.layers.46.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
385
+ "model.layers.46.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
386
+ "model.layers.47.input_layernorm.weight": "pytorch_model-00004-of-00006.bin",
387
+ "model.layers.47.mlp.down_proj.weight": "pytorch_model-00004-of-00006.bin",
388
+ "model.layers.47.mlp.gate_proj.weight": "pytorch_model-00004-of-00006.bin",
389
+ "model.layers.47.mlp.up_proj.weight": "pytorch_model-00004-of-00006.bin",
390
+ "model.layers.47.post_attention_layernorm.weight": "pytorch_model-00004-of-00006.bin",
391
+ "model.layers.47.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
392
+ "model.layers.47.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
393
+ "model.layers.47.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
394
+ "model.layers.47.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
395
+ "model.layers.48.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
396
+ "model.layers.48.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
397
+ "model.layers.48.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
398
+ "model.layers.48.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
399
+ "model.layers.48.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
400
+ "model.layers.48.self_attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
401
+ "model.layers.48.self_attn.o_proj.weight": "pytorch_model-00004-of-00006.bin",
402
+ "model.layers.48.self_attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
403
+ "model.layers.48.self_attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
404
+ "model.layers.49.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
405
+ "model.layers.49.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
406
+ "model.layers.49.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
407
+ "model.layers.49.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
408
+ "model.layers.49.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
409
+ "model.layers.49.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
410
+ "model.layers.49.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
411
+ "model.layers.49.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
412
+ "model.layers.49.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
413
+ "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
414
+ "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
415
+ "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
416
+ "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
417
+ "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
418
+ "model.layers.5.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
419
+ "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
420
+ "model.layers.5.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
421
+ "model.layers.5.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
422
+ "model.layers.50.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
423
+ "model.layers.50.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
424
+ "model.layers.50.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
425
+ "model.layers.50.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
426
+ "model.layers.50.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
427
+ "model.layers.50.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
428
+ "model.layers.50.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
429
+ "model.layers.50.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
430
+ "model.layers.50.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
431
+ "model.layers.51.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
432
+ "model.layers.51.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
433
+ "model.layers.51.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
434
+ "model.layers.51.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
435
+ "model.layers.51.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
436
+ "model.layers.51.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
437
+ "model.layers.51.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
438
+ "model.layers.51.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
439
+ "model.layers.51.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
440
+ "model.layers.52.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
441
+ "model.layers.52.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
442
+ "model.layers.52.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
443
+ "model.layers.52.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
444
+ "model.layers.52.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
445
+ "model.layers.52.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
446
+ "model.layers.52.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
447
+ "model.layers.52.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
448
+ "model.layers.52.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
449
+ "model.layers.53.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
450
+ "model.layers.53.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
451
+ "model.layers.53.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
452
+ "model.layers.53.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
453
+ "model.layers.53.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
454
+ "model.layers.53.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
455
+ "model.layers.53.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
456
+ "model.layers.53.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
457
+ "model.layers.53.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
458
+ "model.layers.54.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
459
+ "model.layers.54.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
460
+ "model.layers.54.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
461
+ "model.layers.54.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
462
+ "model.layers.54.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
463
+ "model.layers.54.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
464
+ "model.layers.54.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
465
+ "model.layers.54.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
466
+ "model.layers.54.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
467
+ "model.layers.55.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
468
+ "model.layers.55.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
469
+ "model.layers.55.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
470
+ "model.layers.55.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
471
+ "model.layers.55.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
472
+ "model.layers.55.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
473
+ "model.layers.55.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
474
+ "model.layers.55.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
475
+ "model.layers.55.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
476
+ "model.layers.56.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
477
+ "model.layers.56.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
478
+ "model.layers.56.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
479
+ "model.layers.56.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
480
+ "model.layers.56.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
481
+ "model.layers.56.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
482
+ "model.layers.56.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
483
+ "model.layers.56.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
484
+ "model.layers.56.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
485
+ "model.layers.57.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
486
+ "model.layers.57.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
487
+ "model.layers.57.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
488
+ "model.layers.57.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
489
+ "model.layers.57.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
490
+ "model.layers.57.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
491
+ "model.layers.57.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
492
+ "model.layers.57.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
493
+ "model.layers.57.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
494
+ "model.layers.58.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
495
+ "model.layers.58.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
496
+ "model.layers.58.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
497
+ "model.layers.58.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
498
+ "model.layers.58.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
499
+ "model.layers.58.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
500
+ "model.layers.58.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
501
+ "model.layers.58.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
502
+ "model.layers.58.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
503
+ "model.layers.59.input_layernorm.weight": "pytorch_model-00005-of-00006.bin",
504
+ "model.layers.59.mlp.down_proj.weight": "pytorch_model-00005-of-00006.bin",
505
+ "model.layers.59.mlp.gate_proj.weight": "pytorch_model-00005-of-00006.bin",
506
+ "model.layers.59.mlp.up_proj.weight": "pytorch_model-00005-of-00006.bin",
507
+ "model.layers.59.post_attention_layernorm.weight": "pytorch_model-00005-of-00006.bin",
508
+ "model.layers.59.self_attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
509
+ "model.layers.59.self_attn.o_proj.weight": "pytorch_model-00005-of-00006.bin",
510
+ "model.layers.59.self_attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
511
+ "model.layers.59.self_attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
512
+ "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
513
+ "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
514
+ "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
515
+ "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
516
+ "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
517
+ "model.layers.6.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
518
+ "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
519
+ "model.layers.6.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
520
+ "model.layers.6.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
521
+ "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
522
+ "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
523
+ "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
524
+ "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
525
+ "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
526
+ "model.layers.7.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
527
+ "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
528
+ "model.layers.7.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
529
+ "model.layers.7.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
530
+ "model.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
531
+ "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
532
+ "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
533
+ "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
534
+ "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
535
+ "model.layers.8.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
536
+ "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
537
+ "model.layers.8.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
538
+ "model.layers.8.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
539
+ "model.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00006.bin",
540
+ "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00006.bin",
541
+ "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00006.bin",
542
+ "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00006.bin",
543
+ "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00006.bin",
544
+ "model.layers.9.self_attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
545
+ "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00006.bin",
546
+ "model.layers.9.self_attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
547
+ "model.layers.9.self_attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
548
+ "model.norm.weight": "pytorch_model-00005-of-00006.bin"
549
  }
550
  }