MollyHexapotato commited on
Commit
05d5176
·
verified ·
1 Parent(s): e6907e8

Upload folder using huggingface_hub

Browse files
chat_template.jinja ADDED
@@ -0,0 +1 @@
 
 
1
+ {% if not add_generation_prompt is defined %}{% set add_generation_prompt = false %}{% endif %}{% set ns = namespace(is_first=false, is_tool=false, is_output_first=true, system_prompt='', is_first_sp=true) %}{%- for message in messages %}{%- if message['role'] == 'system' %}{%- if ns.is_first_sp %}{% set ns.system_prompt = ns.system_prompt + message['content'] %}{% set ns.is_first_sp = false %}{%- else %}{% set ns.system_prompt = ns.system_prompt + '\n\n' + message['content'] %}{%- endif %}{%- endif %}{%- endfor %}{{ bos_token }}{{ ns.system_prompt }}{%- for message in messages %}{%- if message['role'] == 'user' %}{%- set ns.is_tool = false -%}{{'<|User|>' + message['content']}}{%- endif %}{%- if message['role'] == 'assistant' and 'tool_calls' in message %}{%- set ns.is_tool = false -%}{%- for tool in message['tool_calls'] %}{%- if not ns.is_first %}{%- if message['content'] is none %}{{'<|Assistant|><|tool▁calls▁begin|><|tool▁call▁begin|>' + tool['type'] + '<|tool▁sep|>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<|tool▁call▁end|>'}}{%- else %}{{'<|Assistant|>' + message['content'] + '<|tool▁calls▁begin|><|tool▁call▁begin|>' + tool['type'] + '<|tool▁sep|>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<|tool▁call▁end|>'}}{%- endif %}{%- set ns.is_first = true -%}{%- else %}{{'\n' + '<|tool▁call▁begin|>' + tool['type'] + '<|tool▁sep|>' + tool['function']['name'] + '\n' + '```json' + '\n' + tool['function']['arguments'] + '\n' + '```' + '<|tool▁call▁end|>'}}{%- endif %}{%- endfor %}{{'<|tool▁calls▁end|><|end▁of▁sentence|>'}}{%- endif %}{%- if message['role'] == 'assistant' and 'tool_calls' not in message %}{%- if ns.is_tool %}{{'<|tool▁outputs▁end|>' + message['content'] + '<|end▁of▁sentence|>'}}{%- set ns.is_tool = false -%}{%- else %}{% set content = message['content'] %}{% if '</think>' in content %}{% set content = content.split('</think>')[-1] %}{% endif %}{{'<|Assistant|>' + content + '<|end▁of▁sentence|>'}}{%- endif %}{%- endif %}{%- if message['role'] == 'tool' %}{%- set ns.is_tool = true -%}{%- if ns.is_output_first %}{{'<|tool▁outputs▁begin|><|tool▁output▁begin|>' + message['content'] + '<|tool▁output▁end|>'}}{%- set ns.is_output_first = false %}{%- else %}{{'<|tool▁output▁begin|>' + message['content'] + '<|tool▁output▁end|>'}}{%- endif %}{%- endif %}{%- endfor -%}{% if ns.is_tool %}{{'<|tool▁outputs▁end|>'}}{% endif %}{% if add_generation_prompt and not ns.is_tool %}{{'<|Assistant|><think>\n'}}{% endif %}
config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DeepseekV3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "auto_map": {
8
+ "AutoConfig": "configuration_deepseek.DeepseekV3Config",
9
+ "AutoModel": "modeling_deepseek.DeepseekV3Model",
10
+ "AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
11
+ },
12
+ "bos_token_id": 0,
13
+ "eos_token_id": 1,
14
+ "ep_size": 1,
15
+ "first_k_dense_replace": 3,
16
+ "head_dim": 64,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 7168,
19
+ "initializer_range": 0.02,
20
+ "intermediate_size": 18432,
21
+ "kv_lora_rank": 512,
22
+ "max_position_embeddings": 163840,
23
+ "model_type": "deepseek_v3",
24
+ "moe_intermediate_size": 2048,
25
+ "moe_layer_freq": 1,
26
+ "n_group": 8,
27
+ "n_routed_experts": 256,
28
+ "n_shared_experts": 1,
29
+ "norm_topk_prob": true,
30
+ "num_attention_heads": 128,
31
+ "num_experts_per_tok": 8,
32
+ "num_hidden_layers": 4,
33
+ "num_key_value_heads": 128,
34
+ "num_nextn_predict_layers": 1,
35
+ "pretraining_tp": 1,
36
+ "q_lora_rank": 1536,
37
+ "qk_head_dim": 192,
38
+ "qk_nope_head_dim": 128,
39
+ "qk_rope_head_dim": 64,
40
+ "quantization_config": {
41
+ "activation_scheme": "dynamic",
42
+ "fmt": "e4m3",
43
+ "quant_method": "fp8",
44
+ "weight_block_size": [
45
+ 128,
46
+ 128
47
+ ]
48
+ },
49
+ "rms_norm_eps": 1e-06,
50
+ "rope_interleave": true,
51
+ "rope_scaling": {
52
+ "beta_fast": 32.0,
53
+ "beta_slow": 1.0,
54
+ "factor": 40.0,
55
+ "mscale": 1.0,
56
+ "mscale_all_dim": 1.0,
57
+ "original_max_position_embeddings": 4096,
58
+ "rope_type": "yarn",
59
+ "type": "yarn"
60
+ },
61
+ "rope_theta": 10000,
62
+ "routed_scaling_factor": 2.5,
63
+ "scoring_func": "sigmoid",
64
+ "tie_word_embeddings": false,
65
+ "topk_group": 4,
66
+ "topk_method": "noaux_tc",
67
+ "torch_dtype": "bfloat16",
68
+ "transformers_version": "4.53.0",
69
+ "use_cache": true,
70
+ "v_head_dim": 128,
71
+ "vocab_size": 129280
72
+ }
configuration_deepseek.py ADDED
@@ -0,0 +1,199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ from transformers.configuration_utils import PretrainedConfig
2
+ from transformers.utils import logging
3
+
4
+ logger = logging.get_logger(__name__)
5
+
6
+ DEEPSEEK_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
7
+ class DeepseekV3Config(PretrainedConfig):
8
+ r"""
9
+ This is the configuration class to store the configuration of a [`DeepseekV3Model`]. It is used to instantiate an DeepSeek
10
+ model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
11
+ defaults will yield a similar configuration to that of the DeepSeek-V3.
12
+
13
+ Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
14
+ documentation from [`PretrainedConfig`] for more information.
15
+
16
+
17
+ Args:
18
+ vocab_size (`int`, *optional*, defaults to 129280):
19
+ Vocabulary size of the Deep model. Defines the number of different tokens that can be represented by the
20
+ `inputs_ids` passed when calling [`DeepseekV3Model`]
21
+ hidden_size (`int`, *optional*, defaults to 4096):
22
+ Dimension of the hidden representations.
23
+ intermediate_size (`int`, *optional*, defaults to 11008):
24
+ Dimension of the MLP representations.
25
+ moe_intermediate_size (`int`, *optional*, defaults to 1407):
26
+ Dimension of the MoE representations.
27
+ num_hidden_layers (`int`, *optional*, defaults to 32):
28
+ Number of hidden layers in the Transformer decoder.
29
+ num_nextn_predict_layers (`int`, *optional*, defaults to 1):
30
+ Number of nextn predict layers in the DeepSeekV3 Model.
31
+ num_attention_heads (`int`, *optional*, defaults to 32):
32
+ Number of attention heads for each attention layer in the Transformer decoder.
33
+ n_shared_experts (`int`, *optional*, defaults to None):
34
+ Number of shared experts, None means dense model.
35
+ n_routed_experts (`int`, *optional*, defaults to None):
36
+ Number of routed experts, None means dense model.
37
+ routed_scaling_factor (`float`, *optional*, defaults to 1.0):
38
+ Scaling factor or routed experts.
39
+ topk_method (`str`, *optional*, defaults to `gready`):
40
+ Topk method used in routed gate.
41
+ n_group (`int`, *optional*, defaults to None):
42
+ Number of groups for routed experts.
43
+ topk_group (`int`, *optional*, defaults to None):
44
+ Number of selected groups for each token(for each token, ensuring the selected experts is only within `topk_group` groups).
45
+ num_experts_per_tok (`int`, *optional*, defaults to None):
46
+ Number of selected experts, None means dense model.
47
+ moe_layer_freq (`int`, *optional*, defaults to 1):
48
+ The frequency of the MoE layer: one expert layer for every `moe_layer_freq - 1` dense layers.
49
+ first_k_dense_replace (`int`, *optional*, defaults to 0):
50
+ Number of dense layers in shallow layers(embed->dense->dense->...->dense->moe->moe...->lm_head).
51
+ \--k dense layers--/
52
+ norm_topk_prob (`bool`, *optional*, defaults to False):
53
+ Whether to normalize the weights of the routed experts.
54
+ scoring_func (`str`, *optional*, defaults to 'softmax'):
55
+ Method of computing expert weights.
56
+ aux_loss_alpha (`float`, *optional*, defaults to 0.001):
57
+ Auxiliary loss weight coefficient.
58
+ seq_aux = (`bool`, *optional*, defaults to True):
59
+ Whether to compute the auxiliary loss for each individual sample.
60
+ num_key_value_heads (`int`, *optional*):
61
+ This is the number of key_value heads that should be used to implement Grouped Query Attention. If
62
+ `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
63
+ `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
64
+ converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
65
+ by meanpooling all the original heads within that group. For more details checkout [this
66
+ paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
67
+ `num_attention_heads`.
68
+ hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
69
+ The non-linear activation function (function or string) in the decoder.
70
+ max_position_embeddings (`int`, *optional*, defaults to 2048):
71
+ The maximum sequence length that this model might ever be used with.
72
+ initializer_range (`float`, *optional*, defaults to 0.02):
73
+ The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
74
+ rms_norm_eps (`float`, *optional*, defaults to 1e-06):
75
+ The epsilon used by the rms normalization layers.
76
+ use_cache (`bool`, *optional*, defaults to `True`):
77
+ Whether or not the model should return the last key/values attentions (not used by all models). Only
78
+ relevant if `config.is_decoder=True`.
79
+ pad_token_id (`int`, *optional*):
80
+ Padding token id.
81
+ bos_token_id (`int`, *optional*, defaults to 1):
82
+ Beginning of stream token id.
83
+ eos_token_id (`int`, *optional*, defaults to 2):
84
+ End of stream token id.
85
+ tie_word_embeddings (`bool`, *optional*, defaults to `False`):
86
+ Whether to tie weight embeddings
87
+ rope_theta (`float`, *optional*, defaults to 10000.0):
88
+ The base period of the RoPE embeddings.
89
+ rope_scaling (`Dict`, *optional*):
90
+ Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
91
+ strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
92
+ `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
93
+ `max_position_embeddings` to the expected new maximum.
94
+ attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
95
+ Whether to use a bias in the query, key, value and output projection layers during self-attention.
96
+ attention_dropout (`float`, *optional*, defaults to 0.0):
97
+ The dropout ratio for the attention probabilities.
98
+
99
+ ```python
100
+ >>> from transformers import DeepseekV3Model, DeepseekV3Config
101
+
102
+ >>> # Initializing a Deepseek-V3 style configuration
103
+ >>> configuration = DeepseekV3Config()
104
+
105
+ >>> # Accessing the model configuration
106
+ >>> configuration = model.config
107
+ ```"""
108
+
109
+ model_type = "deepseek_v3"
110
+ keys_to_ignore_at_inference = ["past_key_values"]
111
+
112
+ def __init__(
113
+ self,
114
+ vocab_size=129280,
115
+ hidden_size=7168,
116
+ intermediate_size=18432,
117
+ moe_intermediate_size = 2048,
118
+ num_hidden_layers=61,
119
+ num_nextn_predict_layers=1,
120
+ num_attention_heads=128,
121
+ num_key_value_heads=128,
122
+ n_shared_experts = 1,
123
+ n_routed_experts = 256,
124
+ ep_size = 1,
125
+ routed_scaling_factor = 2.5,
126
+ kv_lora_rank = 512,
127
+ q_lora_rank = 1536,
128
+ qk_rope_head_dim = 64,
129
+ v_head_dim = 128,
130
+ qk_nope_head_dim = 128,
131
+ topk_method = 'noaux_tc',
132
+ n_group = 8,
133
+ topk_group = 4,
134
+ num_experts_per_tok = 8,
135
+ moe_layer_freq = 1,
136
+ first_k_dense_replace = 3,
137
+ norm_topk_prob = True,
138
+ scoring_func = 'sigmoid',
139
+ hidden_act="silu",
140
+ max_position_embeddings=4096,
141
+ initializer_range=0.02,
142
+ rms_norm_eps=1e-6,
143
+ use_cache=True,
144
+ pad_token_id=None,
145
+ bos_token_id=0,
146
+ eos_token_id=1,
147
+ tie_word_embeddings=False,
148
+ rope_theta=10000.0,
149
+ rope_scaling=None,
150
+ attention_bias=False,
151
+ attention_dropout=0.0,
152
+ **kwargs,
153
+ ):
154
+ self.vocab_size = vocab_size
155
+ self.max_position_embeddings = max_position_embeddings
156
+ self.hidden_size = hidden_size
157
+ self.intermediate_size = intermediate_size
158
+ self.moe_intermediate_size = moe_intermediate_size
159
+ self.num_hidden_layers = num_hidden_layers
160
+ self.num_nextn_predict_layers = num_nextn_predict_layers
161
+ self.num_attention_heads = num_attention_heads
162
+ self.n_shared_experts = n_shared_experts
163
+ self.n_routed_experts = n_routed_experts
164
+ self.ep_size = ep_size
165
+ self.routed_scaling_factor = routed_scaling_factor
166
+ self.kv_lora_rank = kv_lora_rank
167
+ self.q_lora_rank = q_lora_rank
168
+ self.qk_rope_head_dim = qk_rope_head_dim
169
+ self.v_head_dim = v_head_dim
170
+ self.qk_nope_head_dim = qk_nope_head_dim
171
+ self.topk_method = topk_method
172
+ self.n_group = n_group
173
+ self.topk_group = topk_group
174
+ self.num_experts_per_tok = num_experts_per_tok
175
+ self.moe_layer_freq = moe_layer_freq
176
+ self.first_k_dense_replace = first_k_dense_replace
177
+ self.norm_topk_prob = norm_topk_prob
178
+ self.scoring_func = scoring_func
179
+ # for backward compatibility
180
+ if num_key_value_heads is None:
181
+ num_key_value_heads = num_attention_heads
182
+
183
+ self.num_key_value_heads = num_key_value_heads
184
+ self.hidden_act = hidden_act
185
+ self.initializer_range = initializer_range
186
+ self.rms_norm_eps = rms_norm_eps
187
+ self.use_cache = use_cache
188
+ self.rope_theta = rope_theta
189
+ self.rope_scaling = rope_scaling
190
+ self.attention_bias = attention_bias
191
+ self.attention_dropout = attention_dropout
192
+
193
+ super().__init__(
194
+ pad_token_id=pad_token_id,
195
+ bos_token_id=bos_token_id,
196
+ eos_token_id=eos_token_id,
197
+ tie_word_embeddings=tie_word_embeddings,
198
+ **kwargs,
199
+ )
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "transformers_version": "4.53.0"
6
+ }
model-00001-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07a88fb4b94274fa4d6b770738bed4c6ade5b4f7530ab6fa60b0a85440cc4dfe
3
+ size 4825751392
model-00002-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dffab665851d5085174f8c53b15744ed1a4d81d997480452786728b1d52d1013
3
+ size 4983801968
model-00003-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c60cc29e358a7f30857e7b8cf67da93871da154265d62f7a22f7002a2f84c667
3
+ size 4991242864
model-00004-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2091e0ee20e861c16bb6003c34024dbdc2d27f62d8484596ff254858f441de30
3
+ size 4991243024
model-00005-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81f8d7530561b350fd12e7cbcb16990bbd21e59d0eb5f45333d85a499c9927bc
3
+ size 4991243032
model-00006-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ded685f8e99080e9365af3a9ba3f1cfd748a6763f91f64005f300a47d402d08
3
+ size 3585664952
model-00007-of-00007.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bbcf901a69be4dd1db9f280dfd876bf24642abdc02250c4da396f87a9266538a
3
+ size 1853358208
model.safetensors.index.json ADDED
@@ -0,0 +1,829 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_parameters": 15111101440,
4
+ "total_size": 30222203392
5
+ },
6
+ "weight_map": {
7
+ "lm_head.weight": "model-00007-of-00007.safetensors",
8
+ "model.embed_tokens.weight": "model-00001-of-00007.safetensors",
9
+ "model.layers.0.input_layernorm.weight": "model-00001-of-00007.safetensors",
10
+ "model.layers.0.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
11
+ "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
12
+ "model.layers.0.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
13
+ "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
14
+ "model.layers.0.self_attn.kv_a_layernorm.weight": "model-00001-of-00007.safetensors",
15
+ "model.layers.0.self_attn.kv_a_proj_with_mqa.weight": "model-00001-of-00007.safetensors",
16
+ "model.layers.0.self_attn.kv_b_proj.weight": "model-00001-of-00007.safetensors",
17
+ "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
18
+ "model.layers.0.self_attn.q_a_layernorm.weight": "model-00001-of-00007.safetensors",
19
+ "model.layers.0.self_attn.q_a_proj.weight": "model-00001-of-00007.safetensors",
20
+ "model.layers.0.self_attn.q_b_proj.weight": "model-00001-of-00007.safetensors",
21
+ "model.layers.1.input_layernorm.weight": "model-00001-of-00007.safetensors",
22
+ "model.layers.1.mlp.down_proj.weight": "model-00001-of-00007.safetensors",
23
+ "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
24
+ "model.layers.1.mlp.up_proj.weight": "model-00001-of-00007.safetensors",
25
+ "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00007.safetensors",
26
+ "model.layers.1.self_attn.kv_a_layernorm.weight": "model-00001-of-00007.safetensors",
27
+ "model.layers.1.self_attn.kv_a_proj_with_mqa.weight": "model-00001-of-00007.safetensors",
28
+ "model.layers.1.self_attn.kv_b_proj.weight": "model-00001-of-00007.safetensors",
29
+ "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
30
+ "model.layers.1.self_attn.q_a_layernorm.weight": "model-00001-of-00007.safetensors",
31
+ "model.layers.1.self_attn.q_a_proj.weight": "model-00001-of-00007.safetensors",
32
+ "model.layers.1.self_attn.q_b_proj.weight": "model-00001-of-00007.safetensors",
33
+ "model.layers.2.input_layernorm.weight": "model-00002-of-00007.safetensors",
34
+ "model.layers.2.mlp.down_proj.weight": "model-00002-of-00007.safetensors",
35
+ "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00007.safetensors",
36
+ "model.layers.2.mlp.up_proj.weight": "model-00002-of-00007.safetensors",
37
+ "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00007.safetensors",
38
+ "model.layers.2.self_attn.kv_a_layernorm.weight": "model-00001-of-00007.safetensors",
39
+ "model.layers.2.self_attn.kv_a_proj_with_mqa.weight": "model-00001-of-00007.safetensors",
40
+ "model.layers.2.self_attn.kv_b_proj.weight": "model-00001-of-00007.safetensors",
41
+ "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00007.safetensors",
42
+ "model.layers.2.self_attn.q_a_layernorm.weight": "model-00001-of-00007.safetensors",
43
+ "model.layers.2.self_attn.q_a_proj.weight": "model-00001-of-00007.safetensors",
44
+ "model.layers.2.self_attn.q_b_proj.weight": "model-00001-of-00007.safetensors",
45
+ "model.layers.3.input_layernorm.weight": "model-00006-of-00007.safetensors",
46
+ "model.layers.3.mlp.experts.0.down_proj.weight": "model-00002-of-00007.safetensors",
47
+ "model.layers.3.mlp.experts.0.gate_proj.weight": "model-00002-of-00007.safetensors",
48
+ "model.layers.3.mlp.experts.0.up_proj.weight": "model-00002-of-00007.safetensors",
49
+ "model.layers.3.mlp.experts.1.down_proj.weight": "model-00002-of-00007.safetensors",
50
+ "model.layers.3.mlp.experts.1.gate_proj.weight": "model-00002-of-00007.safetensors",
51
+ "model.layers.3.mlp.experts.1.up_proj.weight": "model-00002-of-00007.safetensors",
52
+ "model.layers.3.mlp.experts.10.down_proj.weight": "model-00002-of-00007.safetensors",
53
+ "model.layers.3.mlp.experts.10.gate_proj.weight": "model-00002-of-00007.safetensors",
54
+ "model.layers.3.mlp.experts.10.up_proj.weight": "model-00002-of-00007.safetensors",
55
+ "model.layers.3.mlp.experts.100.down_proj.weight": "model-00003-of-00007.safetensors",
56
+ "model.layers.3.mlp.experts.100.gate_proj.weight": "model-00003-of-00007.safetensors",
57
+ "model.layers.3.mlp.experts.100.up_proj.weight": "model-00003-of-00007.safetensors",
58
+ "model.layers.3.mlp.experts.101.down_proj.weight": "model-00003-of-00007.safetensors",
59
+ "model.layers.3.mlp.experts.101.gate_proj.weight": "model-00003-of-00007.safetensors",
60
+ "model.layers.3.mlp.experts.101.up_proj.weight": "model-00003-of-00007.safetensors",
61
+ "model.layers.3.mlp.experts.102.down_proj.weight": "model-00003-of-00007.safetensors",
62
+ "model.layers.3.mlp.experts.102.gate_proj.weight": "model-00003-of-00007.safetensors",
63
+ "model.layers.3.mlp.experts.102.up_proj.weight": "model-00003-of-00007.safetensors",
64
+ "model.layers.3.mlp.experts.103.down_proj.weight": "model-00004-of-00007.safetensors",
65
+ "model.layers.3.mlp.experts.103.gate_proj.weight": "model-00004-of-00007.safetensors",
66
+ "model.layers.3.mlp.experts.103.up_proj.weight": "model-00004-of-00007.safetensors",
67
+ "model.layers.3.mlp.experts.104.down_proj.weight": "model-00004-of-00007.safetensors",
68
+ "model.layers.3.mlp.experts.104.gate_proj.weight": "model-00004-of-00007.safetensors",
69
+ "model.layers.3.mlp.experts.104.up_proj.weight": "model-00004-of-00007.safetensors",
70
+ "model.layers.3.mlp.experts.105.down_proj.weight": "model-00004-of-00007.safetensors",
71
+ "model.layers.3.mlp.experts.105.gate_proj.weight": "model-00004-of-00007.safetensors",
72
+ "model.layers.3.mlp.experts.105.up_proj.weight": "model-00004-of-00007.safetensors",
73
+ "model.layers.3.mlp.experts.106.down_proj.weight": "model-00004-of-00007.safetensors",
74
+ "model.layers.3.mlp.experts.106.gate_proj.weight": "model-00004-of-00007.safetensors",
75
+ "model.layers.3.mlp.experts.106.up_proj.weight": "model-00004-of-00007.safetensors",
76
+ "model.layers.3.mlp.experts.107.down_proj.weight": "model-00004-of-00007.safetensors",
77
+ "model.layers.3.mlp.experts.107.gate_proj.weight": "model-00004-of-00007.safetensors",
78
+ "model.layers.3.mlp.experts.107.up_proj.weight": "model-00004-of-00007.safetensors",
79
+ "model.layers.3.mlp.experts.108.down_proj.weight": "model-00004-of-00007.safetensors",
80
+ "model.layers.3.mlp.experts.108.gate_proj.weight": "model-00004-of-00007.safetensors",
81
+ "model.layers.3.mlp.experts.108.up_proj.weight": "model-00004-of-00007.safetensors",
82
+ "model.layers.3.mlp.experts.109.down_proj.weight": "model-00004-of-00007.safetensors",
83
+ "model.layers.3.mlp.experts.109.gate_proj.weight": "model-00004-of-00007.safetensors",
84
+ "model.layers.3.mlp.experts.109.up_proj.weight": "model-00004-of-00007.safetensors",
85
+ "model.layers.3.mlp.experts.11.down_proj.weight": "model-00002-of-00007.safetensors",
86
+ "model.layers.3.mlp.experts.11.gate_proj.weight": "model-00002-of-00007.safetensors",
87
+ "model.layers.3.mlp.experts.11.up_proj.weight": "model-00002-of-00007.safetensors",
88
+ "model.layers.3.mlp.experts.110.down_proj.weight": "model-00004-of-00007.safetensors",
89
+ "model.layers.3.mlp.experts.110.gate_proj.weight": "model-00004-of-00007.safetensors",
90
+ "model.layers.3.mlp.experts.110.up_proj.weight": "model-00004-of-00007.safetensors",
91
+ "model.layers.3.mlp.experts.111.down_proj.weight": "model-00004-of-00007.safetensors",
92
+ "model.layers.3.mlp.experts.111.gate_proj.weight": "model-00004-of-00007.safetensors",
93
+ "model.layers.3.mlp.experts.111.up_proj.weight": "model-00004-of-00007.safetensors",
94
+ "model.layers.3.mlp.experts.112.down_proj.weight": "model-00004-of-00007.safetensors",
95
+ "model.layers.3.mlp.experts.112.gate_proj.weight": "model-00004-of-00007.safetensors",
96
+ "model.layers.3.mlp.experts.112.up_proj.weight": "model-00004-of-00007.safetensors",
97
+ "model.layers.3.mlp.experts.113.down_proj.weight": "model-00004-of-00007.safetensors",
98
+ "model.layers.3.mlp.experts.113.gate_proj.weight": "model-00004-of-00007.safetensors",
99
+ "model.layers.3.mlp.experts.113.up_proj.weight": "model-00004-of-00007.safetensors",
100
+ "model.layers.3.mlp.experts.114.down_proj.weight": "model-00004-of-00007.safetensors",
101
+ "model.layers.3.mlp.experts.114.gate_proj.weight": "model-00004-of-00007.safetensors",
102
+ "model.layers.3.mlp.experts.114.up_proj.weight": "model-00004-of-00007.safetensors",
103
+ "model.layers.3.mlp.experts.115.down_proj.weight": "model-00004-of-00007.safetensors",
104
+ "model.layers.3.mlp.experts.115.gate_proj.weight": "model-00004-of-00007.safetensors",
105
+ "model.layers.3.mlp.experts.115.up_proj.weight": "model-00004-of-00007.safetensors",
106
+ "model.layers.3.mlp.experts.116.down_proj.weight": "model-00004-of-00007.safetensors",
107
+ "model.layers.3.mlp.experts.116.gate_proj.weight": "model-00004-of-00007.safetensors",
108
+ "model.layers.3.mlp.experts.116.up_proj.weight": "model-00004-of-00007.safetensors",
109
+ "model.layers.3.mlp.experts.117.down_proj.weight": "model-00004-of-00007.safetensors",
110
+ "model.layers.3.mlp.experts.117.gate_proj.weight": "model-00004-of-00007.safetensors",
111
+ "model.layers.3.mlp.experts.117.up_proj.weight": "model-00004-of-00007.safetensors",
112
+ "model.layers.3.mlp.experts.118.down_proj.weight": "model-00004-of-00007.safetensors",
113
+ "model.layers.3.mlp.experts.118.gate_proj.weight": "model-00004-of-00007.safetensors",
114
+ "model.layers.3.mlp.experts.118.up_proj.weight": "model-00004-of-00007.safetensors",
115
+ "model.layers.3.mlp.experts.119.down_proj.weight": "model-00004-of-00007.safetensors",
116
+ "model.layers.3.mlp.experts.119.gate_proj.weight": "model-00004-of-00007.safetensors",
117
+ "model.layers.3.mlp.experts.119.up_proj.weight": "model-00004-of-00007.safetensors",
118
+ "model.layers.3.mlp.experts.12.down_proj.weight": "model-00002-of-00007.safetensors",
119
+ "model.layers.3.mlp.experts.12.gate_proj.weight": "model-00002-of-00007.safetensors",
120
+ "model.layers.3.mlp.experts.12.up_proj.weight": "model-00002-of-00007.safetensors",
121
+ "model.layers.3.mlp.experts.120.down_proj.weight": "model-00004-of-00007.safetensors",
122
+ "model.layers.3.mlp.experts.120.gate_proj.weight": "model-00004-of-00007.safetensors",
123
+ "model.layers.3.mlp.experts.120.up_proj.weight": "model-00004-of-00007.safetensors",
124
+ "model.layers.3.mlp.experts.121.down_proj.weight": "model-00004-of-00007.safetensors",
125
+ "model.layers.3.mlp.experts.121.gate_proj.weight": "model-00004-of-00007.safetensors",
126
+ "model.layers.3.mlp.experts.121.up_proj.weight": "model-00004-of-00007.safetensors",
127
+ "model.layers.3.mlp.experts.122.down_proj.weight": "model-00004-of-00007.safetensors",
128
+ "model.layers.3.mlp.experts.122.gate_proj.weight": "model-00004-of-00007.safetensors",
129
+ "model.layers.3.mlp.experts.122.up_proj.weight": "model-00004-of-00007.safetensors",
130
+ "model.layers.3.mlp.experts.123.down_proj.weight": "model-00004-of-00007.safetensors",
131
+ "model.layers.3.mlp.experts.123.gate_proj.weight": "model-00004-of-00007.safetensors",
132
+ "model.layers.3.mlp.experts.123.up_proj.weight": "model-00004-of-00007.safetensors",
133
+ "model.layers.3.mlp.experts.124.down_proj.weight": "model-00004-of-00007.safetensors",
134
+ "model.layers.3.mlp.experts.124.gate_proj.weight": "model-00004-of-00007.safetensors",
135
+ "model.layers.3.mlp.experts.124.up_proj.weight": "model-00004-of-00007.safetensors",
136
+ "model.layers.3.mlp.experts.125.down_proj.weight": "model-00004-of-00007.safetensors",
137
+ "model.layers.3.mlp.experts.125.gate_proj.weight": "model-00004-of-00007.safetensors",
138
+ "model.layers.3.mlp.experts.125.up_proj.weight": "model-00004-of-00007.safetensors",
139
+ "model.layers.3.mlp.experts.126.down_proj.weight": "model-00004-of-00007.safetensors",
140
+ "model.layers.3.mlp.experts.126.gate_proj.weight": "model-00004-of-00007.safetensors",
141
+ "model.layers.3.mlp.experts.126.up_proj.weight": "model-00004-of-00007.safetensors",
142
+ "model.layers.3.mlp.experts.127.down_proj.weight": "model-00004-of-00007.safetensors",
143
+ "model.layers.3.mlp.experts.127.gate_proj.weight": "model-00004-of-00007.safetensors",
144
+ "model.layers.3.mlp.experts.127.up_proj.weight": "model-00004-of-00007.safetensors",
145
+ "model.layers.3.mlp.experts.128.down_proj.weight": "model-00004-of-00007.safetensors",
146
+ "model.layers.3.mlp.experts.128.gate_proj.weight": "model-00004-of-00007.safetensors",
147
+ "model.layers.3.mlp.experts.128.up_proj.weight": "model-00004-of-00007.safetensors",
148
+ "model.layers.3.mlp.experts.129.down_proj.weight": "model-00004-of-00007.safetensors",
149
+ "model.layers.3.mlp.experts.129.gate_proj.weight": "model-00004-of-00007.safetensors",
150
+ "model.layers.3.mlp.experts.129.up_proj.weight": "model-00004-of-00007.safetensors",
151
+ "model.layers.3.mlp.experts.13.down_proj.weight": "model-00002-of-00007.safetensors",
152
+ "model.layers.3.mlp.experts.13.gate_proj.weight": "model-00002-of-00007.safetensors",
153
+ "model.layers.3.mlp.experts.13.up_proj.weight": "model-00002-of-00007.safetensors",
154
+ "model.layers.3.mlp.experts.130.down_proj.weight": "model-00004-of-00007.safetensors",
155
+ "model.layers.3.mlp.experts.130.gate_proj.weight": "model-00004-of-00007.safetensors",
156
+ "model.layers.3.mlp.experts.130.up_proj.weight": "model-00004-of-00007.safetensors",
157
+ "model.layers.3.mlp.experts.131.down_proj.weight": "model-00004-of-00007.safetensors",
158
+ "model.layers.3.mlp.experts.131.gate_proj.weight": "model-00004-of-00007.safetensors",
159
+ "model.layers.3.mlp.experts.131.up_proj.weight": "model-00004-of-00007.safetensors",
160
+ "model.layers.3.mlp.experts.132.down_proj.weight": "model-00004-of-00007.safetensors",
161
+ "model.layers.3.mlp.experts.132.gate_proj.weight": "model-00004-of-00007.safetensors",
162
+ "model.layers.3.mlp.experts.132.up_proj.weight": "model-00004-of-00007.safetensors",
163
+ "model.layers.3.mlp.experts.133.down_proj.weight": "model-00004-of-00007.safetensors",
164
+ "model.layers.3.mlp.experts.133.gate_proj.weight": "model-00004-of-00007.safetensors",
165
+ "model.layers.3.mlp.experts.133.up_proj.weight": "model-00004-of-00007.safetensors",
166
+ "model.layers.3.mlp.experts.134.down_proj.weight": "model-00004-of-00007.safetensors",
167
+ "model.layers.3.mlp.experts.134.gate_proj.weight": "model-00004-of-00007.safetensors",
168
+ "model.layers.3.mlp.experts.134.up_proj.weight": "model-00004-of-00007.safetensors",
169
+ "model.layers.3.mlp.experts.135.down_proj.weight": "model-00004-of-00007.safetensors",
170
+ "model.layers.3.mlp.experts.135.gate_proj.weight": "model-00004-of-00007.safetensors",
171
+ "model.layers.3.mlp.experts.135.up_proj.weight": "model-00004-of-00007.safetensors",
172
+ "model.layers.3.mlp.experts.136.down_proj.weight": "model-00004-of-00007.safetensors",
173
+ "model.layers.3.mlp.experts.136.gate_proj.weight": "model-00004-of-00007.safetensors",
174
+ "model.layers.3.mlp.experts.136.up_proj.weight": "model-00004-of-00007.safetensors",
175
+ "model.layers.3.mlp.experts.137.down_proj.weight": "model-00004-of-00007.safetensors",
176
+ "model.layers.3.mlp.experts.137.gate_proj.weight": "model-00004-of-00007.safetensors",
177
+ "model.layers.3.mlp.experts.137.up_proj.weight": "model-00004-of-00007.safetensors",
178
+ "model.layers.3.mlp.experts.138.down_proj.weight": "model-00004-of-00007.safetensors",
179
+ "model.layers.3.mlp.experts.138.gate_proj.weight": "model-00004-of-00007.safetensors",
180
+ "model.layers.3.mlp.experts.138.up_proj.weight": "model-00004-of-00007.safetensors",
181
+ "model.layers.3.mlp.experts.139.down_proj.weight": "model-00004-of-00007.safetensors",
182
+ "model.layers.3.mlp.experts.139.gate_proj.weight": "model-00004-of-00007.safetensors",
183
+ "model.layers.3.mlp.experts.139.up_proj.weight": "model-00004-of-00007.safetensors",
184
+ "model.layers.3.mlp.experts.14.down_proj.weight": "model-00002-of-00007.safetensors",
185
+ "model.layers.3.mlp.experts.14.gate_proj.weight": "model-00002-of-00007.safetensors",
186
+ "model.layers.3.mlp.experts.14.up_proj.weight": "model-00002-of-00007.safetensors",
187
+ "model.layers.3.mlp.experts.140.down_proj.weight": "model-00004-of-00007.safetensors",
188
+ "model.layers.3.mlp.experts.140.gate_proj.weight": "model-00004-of-00007.safetensors",
189
+ "model.layers.3.mlp.experts.140.up_proj.weight": "model-00004-of-00007.safetensors",
190
+ "model.layers.3.mlp.experts.141.down_proj.weight": "model-00004-of-00007.safetensors",
191
+ "model.layers.3.mlp.experts.141.gate_proj.weight": "model-00004-of-00007.safetensors",
192
+ "model.layers.3.mlp.experts.141.up_proj.weight": "model-00004-of-00007.safetensors",
193
+ "model.layers.3.mlp.experts.142.down_proj.weight": "model-00004-of-00007.safetensors",
194
+ "model.layers.3.mlp.experts.142.gate_proj.weight": "model-00004-of-00007.safetensors",
195
+ "model.layers.3.mlp.experts.142.up_proj.weight": "model-00004-of-00007.safetensors",
196
+ "model.layers.3.mlp.experts.143.down_proj.weight": "model-00004-of-00007.safetensors",
197
+ "model.layers.3.mlp.experts.143.gate_proj.weight": "model-00004-of-00007.safetensors",
198
+ "model.layers.3.mlp.experts.143.up_proj.weight": "model-00004-of-00007.safetensors",
199
+ "model.layers.3.mlp.experts.144.down_proj.weight": "model-00004-of-00007.safetensors",
200
+ "model.layers.3.mlp.experts.144.gate_proj.weight": "model-00004-of-00007.safetensors",
201
+ "model.layers.3.mlp.experts.144.up_proj.weight": "model-00004-of-00007.safetensors",
202
+ "model.layers.3.mlp.experts.145.down_proj.weight": "model-00004-of-00007.safetensors",
203
+ "model.layers.3.mlp.experts.145.gate_proj.weight": "model-00004-of-00007.safetensors",
204
+ "model.layers.3.mlp.experts.145.up_proj.weight": "model-00004-of-00007.safetensors",
205
+ "model.layers.3.mlp.experts.146.down_proj.weight": "model-00004-of-00007.safetensors",
206
+ "model.layers.3.mlp.experts.146.gate_proj.weight": "model-00004-of-00007.safetensors",
207
+ "model.layers.3.mlp.experts.146.up_proj.weight": "model-00004-of-00007.safetensors",
208
+ "model.layers.3.mlp.experts.147.down_proj.weight": "model-00004-of-00007.safetensors",
209
+ "model.layers.3.mlp.experts.147.gate_proj.weight": "model-00004-of-00007.safetensors",
210
+ "model.layers.3.mlp.experts.147.up_proj.weight": "model-00004-of-00007.safetensors",
211
+ "model.layers.3.mlp.experts.148.down_proj.weight": "model-00004-of-00007.safetensors",
212
+ "model.layers.3.mlp.experts.148.gate_proj.weight": "model-00004-of-00007.safetensors",
213
+ "model.layers.3.mlp.experts.148.up_proj.weight": "model-00004-of-00007.safetensors",
214
+ "model.layers.3.mlp.experts.149.down_proj.weight": "model-00004-of-00007.safetensors",
215
+ "model.layers.3.mlp.experts.149.gate_proj.weight": "model-00004-of-00007.safetensors",
216
+ "model.layers.3.mlp.experts.149.up_proj.weight": "model-00004-of-00007.safetensors",
217
+ "model.layers.3.mlp.experts.15.down_proj.weight": "model-00002-of-00007.safetensors",
218
+ "model.layers.3.mlp.experts.15.gate_proj.weight": "model-00002-of-00007.safetensors",
219
+ "model.layers.3.mlp.experts.15.up_proj.weight": "model-00002-of-00007.safetensors",
220
+ "model.layers.3.mlp.experts.150.down_proj.weight": "model-00004-of-00007.safetensors",
221
+ "model.layers.3.mlp.experts.150.gate_proj.weight": "model-00004-of-00007.safetensors",
222
+ "model.layers.3.mlp.experts.150.up_proj.weight": "model-00004-of-00007.safetensors",
223
+ "model.layers.3.mlp.experts.151.down_proj.weight": "model-00004-of-00007.safetensors",
224
+ "model.layers.3.mlp.experts.151.gate_proj.weight": "model-00004-of-00007.safetensors",
225
+ "model.layers.3.mlp.experts.151.up_proj.weight": "model-00004-of-00007.safetensors",
226
+ "model.layers.3.mlp.experts.152.down_proj.weight": "model-00004-of-00007.safetensors",
227
+ "model.layers.3.mlp.experts.152.gate_proj.weight": "model-00004-of-00007.safetensors",
228
+ "model.layers.3.mlp.experts.152.up_proj.weight": "model-00004-of-00007.safetensors",
229
+ "model.layers.3.mlp.experts.153.down_proj.weight": "model-00004-of-00007.safetensors",
230
+ "model.layers.3.mlp.experts.153.gate_proj.weight": "model-00004-of-00007.safetensors",
231
+ "model.layers.3.mlp.experts.153.up_proj.weight": "model-00004-of-00007.safetensors",
232
+ "model.layers.3.mlp.experts.154.down_proj.weight": "model-00004-of-00007.safetensors",
233
+ "model.layers.3.mlp.experts.154.gate_proj.weight": "model-00004-of-00007.safetensors",
234
+ "model.layers.3.mlp.experts.154.up_proj.weight": "model-00004-of-00007.safetensors",
235
+ "model.layers.3.mlp.experts.155.down_proj.weight": "model-00004-of-00007.safetensors",
236
+ "model.layers.3.mlp.experts.155.gate_proj.weight": "model-00004-of-00007.safetensors",
237
+ "model.layers.3.mlp.experts.155.up_proj.weight": "model-00004-of-00007.safetensors",
238
+ "model.layers.3.mlp.experts.156.down_proj.weight": "model-00004-of-00007.safetensors",
239
+ "model.layers.3.mlp.experts.156.gate_proj.weight": "model-00004-of-00007.safetensors",
240
+ "model.layers.3.mlp.experts.156.up_proj.weight": "model-00004-of-00007.safetensors",
241
+ "model.layers.3.mlp.experts.157.down_proj.weight": "model-00004-of-00007.safetensors",
242
+ "model.layers.3.mlp.experts.157.gate_proj.weight": "model-00004-of-00007.safetensors",
243
+ "model.layers.3.mlp.experts.157.up_proj.weight": "model-00004-of-00007.safetensors",
244
+ "model.layers.3.mlp.experts.158.down_proj.weight": "model-00004-of-00007.safetensors",
245
+ "model.layers.3.mlp.experts.158.gate_proj.weight": "model-00004-of-00007.safetensors",
246
+ "model.layers.3.mlp.experts.158.up_proj.weight": "model-00004-of-00007.safetensors",
247
+ "model.layers.3.mlp.experts.159.down_proj.weight": "model-00005-of-00007.safetensors",
248
+ "model.layers.3.mlp.experts.159.gate_proj.weight": "model-00004-of-00007.safetensors",
249
+ "model.layers.3.mlp.experts.159.up_proj.weight": "model-00004-of-00007.safetensors",
250
+ "model.layers.3.mlp.experts.16.down_proj.weight": "model-00002-of-00007.safetensors",
251
+ "model.layers.3.mlp.experts.16.gate_proj.weight": "model-00002-of-00007.safetensors",
252
+ "model.layers.3.mlp.experts.16.up_proj.weight": "model-00002-of-00007.safetensors",
253
+ "model.layers.3.mlp.experts.160.down_proj.weight": "model-00005-of-00007.safetensors",
254
+ "model.layers.3.mlp.experts.160.gate_proj.weight": "model-00005-of-00007.safetensors",
255
+ "model.layers.3.mlp.experts.160.up_proj.weight": "model-00005-of-00007.safetensors",
256
+ "model.layers.3.mlp.experts.161.down_proj.weight": "model-00005-of-00007.safetensors",
257
+ "model.layers.3.mlp.experts.161.gate_proj.weight": "model-00005-of-00007.safetensors",
258
+ "model.layers.3.mlp.experts.161.up_proj.weight": "model-00005-of-00007.safetensors",
259
+ "model.layers.3.mlp.experts.162.down_proj.weight": "model-00005-of-00007.safetensors",
260
+ "model.layers.3.mlp.experts.162.gate_proj.weight": "model-00005-of-00007.safetensors",
261
+ "model.layers.3.mlp.experts.162.up_proj.weight": "model-00005-of-00007.safetensors",
262
+ "model.layers.3.mlp.experts.163.down_proj.weight": "model-00005-of-00007.safetensors",
263
+ "model.layers.3.mlp.experts.163.gate_proj.weight": "model-00005-of-00007.safetensors",
264
+ "model.layers.3.mlp.experts.163.up_proj.weight": "model-00005-of-00007.safetensors",
265
+ "model.layers.3.mlp.experts.164.down_proj.weight": "model-00005-of-00007.safetensors",
266
+ "model.layers.3.mlp.experts.164.gate_proj.weight": "model-00005-of-00007.safetensors",
267
+ "model.layers.3.mlp.experts.164.up_proj.weight": "model-00005-of-00007.safetensors",
268
+ "model.layers.3.mlp.experts.165.down_proj.weight": "model-00005-of-00007.safetensors",
269
+ "model.layers.3.mlp.experts.165.gate_proj.weight": "model-00005-of-00007.safetensors",
270
+ "model.layers.3.mlp.experts.165.up_proj.weight": "model-00005-of-00007.safetensors",
271
+ "model.layers.3.mlp.experts.166.down_proj.weight": "model-00005-of-00007.safetensors",
272
+ "model.layers.3.mlp.experts.166.gate_proj.weight": "model-00005-of-00007.safetensors",
273
+ "model.layers.3.mlp.experts.166.up_proj.weight": "model-00005-of-00007.safetensors",
274
+ "model.layers.3.mlp.experts.167.down_proj.weight": "model-00005-of-00007.safetensors",
275
+ "model.layers.3.mlp.experts.167.gate_proj.weight": "model-00005-of-00007.safetensors",
276
+ "model.layers.3.mlp.experts.167.up_proj.weight": "model-00005-of-00007.safetensors",
277
+ "model.layers.3.mlp.experts.168.down_proj.weight": "model-00005-of-00007.safetensors",
278
+ "model.layers.3.mlp.experts.168.gate_proj.weight": "model-00005-of-00007.safetensors",
279
+ "model.layers.3.mlp.experts.168.up_proj.weight": "model-00005-of-00007.safetensors",
280
+ "model.layers.3.mlp.experts.169.down_proj.weight": "model-00005-of-00007.safetensors",
281
+ "model.layers.3.mlp.experts.169.gate_proj.weight": "model-00005-of-00007.safetensors",
282
+ "model.layers.3.mlp.experts.169.up_proj.weight": "model-00005-of-00007.safetensors",
283
+ "model.layers.3.mlp.experts.17.down_proj.weight": "model-00002-of-00007.safetensors",
284
+ "model.layers.3.mlp.experts.17.gate_proj.weight": "model-00002-of-00007.safetensors",
285
+ "model.layers.3.mlp.experts.17.up_proj.weight": "model-00002-of-00007.safetensors",
286
+ "model.layers.3.mlp.experts.170.down_proj.weight": "model-00005-of-00007.safetensors",
287
+ "model.layers.3.mlp.experts.170.gate_proj.weight": "model-00005-of-00007.safetensors",
288
+ "model.layers.3.mlp.experts.170.up_proj.weight": "model-00005-of-00007.safetensors",
289
+ "model.layers.3.mlp.experts.171.down_proj.weight": "model-00005-of-00007.safetensors",
290
+ "model.layers.3.mlp.experts.171.gate_proj.weight": "model-00005-of-00007.safetensors",
291
+ "model.layers.3.mlp.experts.171.up_proj.weight": "model-00005-of-00007.safetensors",
292
+ "model.layers.3.mlp.experts.172.down_proj.weight": "model-00005-of-00007.safetensors",
293
+ "model.layers.3.mlp.experts.172.gate_proj.weight": "model-00005-of-00007.safetensors",
294
+ "model.layers.3.mlp.experts.172.up_proj.weight": "model-00005-of-00007.safetensors",
295
+ "model.layers.3.mlp.experts.173.down_proj.weight": "model-00005-of-00007.safetensors",
296
+ "model.layers.3.mlp.experts.173.gate_proj.weight": "model-00005-of-00007.safetensors",
297
+ "model.layers.3.mlp.experts.173.up_proj.weight": "model-00005-of-00007.safetensors",
298
+ "model.layers.3.mlp.experts.174.down_proj.weight": "model-00005-of-00007.safetensors",
299
+ "model.layers.3.mlp.experts.174.gate_proj.weight": "model-00005-of-00007.safetensors",
300
+ "model.layers.3.mlp.experts.174.up_proj.weight": "model-00005-of-00007.safetensors",
301
+ "model.layers.3.mlp.experts.175.down_proj.weight": "model-00005-of-00007.safetensors",
302
+ "model.layers.3.mlp.experts.175.gate_proj.weight": "model-00005-of-00007.safetensors",
303
+ "model.layers.3.mlp.experts.175.up_proj.weight": "model-00005-of-00007.safetensors",
304
+ "model.layers.3.mlp.experts.176.down_proj.weight": "model-00005-of-00007.safetensors",
305
+ "model.layers.3.mlp.experts.176.gate_proj.weight": "model-00005-of-00007.safetensors",
306
+ "model.layers.3.mlp.experts.176.up_proj.weight": "model-00005-of-00007.safetensors",
307
+ "model.layers.3.mlp.experts.177.down_proj.weight": "model-00005-of-00007.safetensors",
308
+ "model.layers.3.mlp.experts.177.gate_proj.weight": "model-00005-of-00007.safetensors",
309
+ "model.layers.3.mlp.experts.177.up_proj.weight": "model-00005-of-00007.safetensors",
310
+ "model.layers.3.mlp.experts.178.down_proj.weight": "model-00005-of-00007.safetensors",
311
+ "model.layers.3.mlp.experts.178.gate_proj.weight": "model-00005-of-00007.safetensors",
312
+ "model.layers.3.mlp.experts.178.up_proj.weight": "model-00005-of-00007.safetensors",
313
+ "model.layers.3.mlp.experts.179.down_proj.weight": "model-00005-of-00007.safetensors",
314
+ "model.layers.3.mlp.experts.179.gate_proj.weight": "model-00005-of-00007.safetensors",
315
+ "model.layers.3.mlp.experts.179.up_proj.weight": "model-00005-of-00007.safetensors",
316
+ "model.layers.3.mlp.experts.18.down_proj.weight": "model-00002-of-00007.safetensors",
317
+ "model.layers.3.mlp.experts.18.gate_proj.weight": "model-00002-of-00007.safetensors",
318
+ "model.layers.3.mlp.experts.18.up_proj.weight": "model-00002-of-00007.safetensors",
319
+ "model.layers.3.mlp.experts.180.down_proj.weight": "model-00005-of-00007.safetensors",
320
+ "model.layers.3.mlp.experts.180.gate_proj.weight": "model-00005-of-00007.safetensors",
321
+ "model.layers.3.mlp.experts.180.up_proj.weight": "model-00005-of-00007.safetensors",
322
+ "model.layers.3.mlp.experts.181.down_proj.weight": "model-00005-of-00007.safetensors",
323
+ "model.layers.3.mlp.experts.181.gate_proj.weight": "model-00005-of-00007.safetensors",
324
+ "model.layers.3.mlp.experts.181.up_proj.weight": "model-00005-of-00007.safetensors",
325
+ "model.layers.3.mlp.experts.182.down_proj.weight": "model-00005-of-00007.safetensors",
326
+ "model.layers.3.mlp.experts.182.gate_proj.weight": "model-00005-of-00007.safetensors",
327
+ "model.layers.3.mlp.experts.182.up_proj.weight": "model-00005-of-00007.safetensors",
328
+ "model.layers.3.mlp.experts.183.down_proj.weight": "model-00005-of-00007.safetensors",
329
+ "model.layers.3.mlp.experts.183.gate_proj.weight": "model-00005-of-00007.safetensors",
330
+ "model.layers.3.mlp.experts.183.up_proj.weight": "model-00005-of-00007.safetensors",
331
+ "model.layers.3.mlp.experts.184.down_proj.weight": "model-00005-of-00007.safetensors",
332
+ "model.layers.3.mlp.experts.184.gate_proj.weight": "model-00005-of-00007.safetensors",
333
+ "model.layers.3.mlp.experts.184.up_proj.weight": "model-00005-of-00007.safetensors",
334
+ "model.layers.3.mlp.experts.185.down_proj.weight": "model-00005-of-00007.safetensors",
335
+ "model.layers.3.mlp.experts.185.gate_proj.weight": "model-00005-of-00007.safetensors",
336
+ "model.layers.3.mlp.experts.185.up_proj.weight": "model-00005-of-00007.safetensors",
337
+ "model.layers.3.mlp.experts.186.down_proj.weight": "model-00005-of-00007.safetensors",
338
+ "model.layers.3.mlp.experts.186.gate_proj.weight": "model-00005-of-00007.safetensors",
339
+ "model.layers.3.mlp.experts.186.up_proj.weight": "model-00005-of-00007.safetensors",
340
+ "model.layers.3.mlp.experts.187.down_proj.weight": "model-00005-of-00007.safetensors",
341
+ "model.layers.3.mlp.experts.187.gate_proj.weight": "model-00005-of-00007.safetensors",
342
+ "model.layers.3.mlp.experts.187.up_proj.weight": "model-00005-of-00007.safetensors",
343
+ "model.layers.3.mlp.experts.188.down_proj.weight": "model-00005-of-00007.safetensors",
344
+ "model.layers.3.mlp.experts.188.gate_proj.weight": "model-00005-of-00007.safetensors",
345
+ "model.layers.3.mlp.experts.188.up_proj.weight": "model-00005-of-00007.safetensors",
346
+ "model.layers.3.mlp.experts.189.down_proj.weight": "model-00005-of-00007.safetensors",
347
+ "model.layers.3.mlp.experts.189.gate_proj.weight": "model-00005-of-00007.safetensors",
348
+ "model.layers.3.mlp.experts.189.up_proj.weight": "model-00005-of-00007.safetensors",
349
+ "model.layers.3.mlp.experts.19.down_proj.weight": "model-00002-of-00007.safetensors",
350
+ "model.layers.3.mlp.experts.19.gate_proj.weight": "model-00002-of-00007.safetensors",
351
+ "model.layers.3.mlp.experts.19.up_proj.weight": "model-00002-of-00007.safetensors",
352
+ "model.layers.3.mlp.experts.190.down_proj.weight": "model-00005-of-00007.safetensors",
353
+ "model.layers.3.mlp.experts.190.gate_proj.weight": "model-00005-of-00007.safetensors",
354
+ "model.layers.3.mlp.experts.190.up_proj.weight": "model-00005-of-00007.safetensors",
355
+ "model.layers.3.mlp.experts.191.down_proj.weight": "model-00005-of-00007.safetensors",
356
+ "model.layers.3.mlp.experts.191.gate_proj.weight": "model-00005-of-00007.safetensors",
357
+ "model.layers.3.mlp.experts.191.up_proj.weight": "model-00005-of-00007.safetensors",
358
+ "model.layers.3.mlp.experts.192.down_proj.weight": "model-00005-of-00007.safetensors",
359
+ "model.layers.3.mlp.experts.192.gate_proj.weight": "model-00005-of-00007.safetensors",
360
+ "model.layers.3.mlp.experts.192.up_proj.weight": "model-00005-of-00007.safetensors",
361
+ "model.layers.3.mlp.experts.193.down_proj.weight": "model-00005-of-00007.safetensors",
362
+ "model.layers.3.mlp.experts.193.gate_proj.weight": "model-00005-of-00007.safetensors",
363
+ "model.layers.3.mlp.experts.193.up_proj.weight": "model-00005-of-00007.safetensors",
364
+ "model.layers.3.mlp.experts.194.down_proj.weight": "model-00005-of-00007.safetensors",
365
+ "model.layers.3.mlp.experts.194.gate_proj.weight": "model-00005-of-00007.safetensors",
366
+ "model.layers.3.mlp.experts.194.up_proj.weight": "model-00005-of-00007.safetensors",
367
+ "model.layers.3.mlp.experts.195.down_proj.weight": "model-00005-of-00007.safetensors",
368
+ "model.layers.3.mlp.experts.195.gate_proj.weight": "model-00005-of-00007.safetensors",
369
+ "model.layers.3.mlp.experts.195.up_proj.weight": "model-00005-of-00007.safetensors",
370
+ "model.layers.3.mlp.experts.196.down_proj.weight": "model-00005-of-00007.safetensors",
371
+ "model.layers.3.mlp.experts.196.gate_proj.weight": "model-00005-of-00007.safetensors",
372
+ "model.layers.3.mlp.experts.196.up_proj.weight": "model-00005-of-00007.safetensors",
373
+ "model.layers.3.mlp.experts.197.down_proj.weight": "model-00005-of-00007.safetensors",
374
+ "model.layers.3.mlp.experts.197.gate_proj.weight": "model-00005-of-00007.safetensors",
375
+ "model.layers.3.mlp.experts.197.up_proj.weight": "model-00005-of-00007.safetensors",
376
+ "model.layers.3.mlp.experts.198.down_proj.weight": "model-00005-of-00007.safetensors",
377
+ "model.layers.3.mlp.experts.198.gate_proj.weight": "model-00005-of-00007.safetensors",
378
+ "model.layers.3.mlp.experts.198.up_proj.weight": "model-00005-of-00007.safetensors",
379
+ "model.layers.3.mlp.experts.199.down_proj.weight": "model-00005-of-00007.safetensors",
380
+ "model.layers.3.mlp.experts.199.gate_proj.weight": "model-00005-of-00007.safetensors",
381
+ "model.layers.3.mlp.experts.199.up_proj.weight": "model-00005-of-00007.safetensors",
382
+ "model.layers.3.mlp.experts.2.down_proj.weight": "model-00002-of-00007.safetensors",
383
+ "model.layers.3.mlp.experts.2.gate_proj.weight": "model-00002-of-00007.safetensors",
384
+ "model.layers.3.mlp.experts.2.up_proj.weight": "model-00002-of-00007.safetensors",
385
+ "model.layers.3.mlp.experts.20.down_proj.weight": "model-00002-of-00007.safetensors",
386
+ "model.layers.3.mlp.experts.20.gate_proj.weight": "model-00002-of-00007.safetensors",
387
+ "model.layers.3.mlp.experts.20.up_proj.weight": "model-00002-of-00007.safetensors",
388
+ "model.layers.3.mlp.experts.200.down_proj.weight": "model-00005-of-00007.safetensors",
389
+ "model.layers.3.mlp.experts.200.gate_proj.weight": "model-00005-of-00007.safetensors",
390
+ "model.layers.3.mlp.experts.200.up_proj.weight": "model-00005-of-00007.safetensors",
391
+ "model.layers.3.mlp.experts.201.down_proj.weight": "model-00005-of-00007.safetensors",
392
+ "model.layers.3.mlp.experts.201.gate_proj.weight": "model-00005-of-00007.safetensors",
393
+ "model.layers.3.mlp.experts.201.up_proj.weight": "model-00005-of-00007.safetensors",
394
+ "model.layers.3.mlp.experts.202.down_proj.weight": "model-00005-of-00007.safetensors",
395
+ "model.layers.3.mlp.experts.202.gate_proj.weight": "model-00005-of-00007.safetensors",
396
+ "model.layers.3.mlp.experts.202.up_proj.weight": "model-00005-of-00007.safetensors",
397
+ "model.layers.3.mlp.experts.203.down_proj.weight": "model-00005-of-00007.safetensors",
398
+ "model.layers.3.mlp.experts.203.gate_proj.weight": "model-00005-of-00007.safetensors",
399
+ "model.layers.3.mlp.experts.203.up_proj.weight": "model-00005-of-00007.safetensors",
400
+ "model.layers.3.mlp.experts.204.down_proj.weight": "model-00005-of-00007.safetensors",
401
+ "model.layers.3.mlp.experts.204.gate_proj.weight": "model-00005-of-00007.safetensors",
402
+ "model.layers.3.mlp.experts.204.up_proj.weight": "model-00005-of-00007.safetensors",
403
+ "model.layers.3.mlp.experts.205.down_proj.weight": "model-00005-of-00007.safetensors",
404
+ "model.layers.3.mlp.experts.205.gate_proj.weight": "model-00005-of-00007.safetensors",
405
+ "model.layers.3.mlp.experts.205.up_proj.weight": "model-00005-of-00007.safetensors",
406
+ "model.layers.3.mlp.experts.206.down_proj.weight": "model-00005-of-00007.safetensors",
407
+ "model.layers.3.mlp.experts.206.gate_proj.weight": "model-00005-of-00007.safetensors",
408
+ "model.layers.3.mlp.experts.206.up_proj.weight": "model-00005-of-00007.safetensors",
409
+ "model.layers.3.mlp.experts.207.down_proj.weight": "model-00005-of-00007.safetensors",
410
+ "model.layers.3.mlp.experts.207.gate_proj.weight": "model-00005-of-00007.safetensors",
411
+ "model.layers.3.mlp.experts.207.up_proj.weight": "model-00005-of-00007.safetensors",
412
+ "model.layers.3.mlp.experts.208.down_proj.weight": "model-00005-of-00007.safetensors",
413
+ "model.layers.3.mlp.experts.208.gate_proj.weight": "model-00005-of-00007.safetensors",
414
+ "model.layers.3.mlp.experts.208.up_proj.weight": "model-00005-of-00007.safetensors",
415
+ "model.layers.3.mlp.experts.209.down_proj.weight": "model-00005-of-00007.safetensors",
416
+ "model.layers.3.mlp.experts.209.gate_proj.weight": "model-00005-of-00007.safetensors",
417
+ "model.layers.3.mlp.experts.209.up_proj.weight": "model-00005-of-00007.safetensors",
418
+ "model.layers.3.mlp.experts.21.down_proj.weight": "model-00002-of-00007.safetensors",
419
+ "model.layers.3.mlp.experts.21.gate_proj.weight": "model-00002-of-00007.safetensors",
420
+ "model.layers.3.mlp.experts.21.up_proj.weight": "model-00002-of-00007.safetensors",
421
+ "model.layers.3.mlp.experts.210.down_proj.weight": "model-00005-of-00007.safetensors",
422
+ "model.layers.3.mlp.experts.210.gate_proj.weight": "model-00005-of-00007.safetensors",
423
+ "model.layers.3.mlp.experts.210.up_proj.weight": "model-00005-of-00007.safetensors",
424
+ "model.layers.3.mlp.experts.211.down_proj.weight": "model-00005-of-00007.safetensors",
425
+ "model.layers.3.mlp.experts.211.gate_proj.weight": "model-00005-of-00007.safetensors",
426
+ "model.layers.3.mlp.experts.211.up_proj.weight": "model-00005-of-00007.safetensors",
427
+ "model.layers.3.mlp.experts.212.down_proj.weight": "model-00005-of-00007.safetensors",
428
+ "model.layers.3.mlp.experts.212.gate_proj.weight": "model-00005-of-00007.safetensors",
429
+ "model.layers.3.mlp.experts.212.up_proj.weight": "model-00005-of-00007.safetensors",
430
+ "model.layers.3.mlp.experts.213.down_proj.weight": "model-00005-of-00007.safetensors",
431
+ "model.layers.3.mlp.experts.213.gate_proj.weight": "model-00005-of-00007.safetensors",
432
+ "model.layers.3.mlp.experts.213.up_proj.weight": "model-00005-of-00007.safetensors",
433
+ "model.layers.3.mlp.experts.214.down_proj.weight": "model-00005-of-00007.safetensors",
434
+ "model.layers.3.mlp.experts.214.gate_proj.weight": "model-00005-of-00007.safetensors",
435
+ "model.layers.3.mlp.experts.214.up_proj.weight": "model-00005-of-00007.safetensors",
436
+ "model.layers.3.mlp.experts.215.down_proj.weight": "model-00005-of-00007.safetensors",
437
+ "model.layers.3.mlp.experts.215.gate_proj.weight": "model-00005-of-00007.safetensors",
438
+ "model.layers.3.mlp.experts.215.up_proj.weight": "model-00005-of-00007.safetensors",
439
+ "model.layers.3.mlp.experts.216.down_proj.weight": "model-00006-of-00007.safetensors",
440
+ "model.layers.3.mlp.experts.216.gate_proj.weight": "model-00005-of-00007.safetensors",
441
+ "model.layers.3.mlp.experts.216.up_proj.weight": "model-00006-of-00007.safetensors",
442
+ "model.layers.3.mlp.experts.217.down_proj.weight": "model-00006-of-00007.safetensors",
443
+ "model.layers.3.mlp.experts.217.gate_proj.weight": "model-00006-of-00007.safetensors",
444
+ "model.layers.3.mlp.experts.217.up_proj.weight": "model-00006-of-00007.safetensors",
445
+ "model.layers.3.mlp.experts.218.down_proj.weight": "model-00006-of-00007.safetensors",
446
+ "model.layers.3.mlp.experts.218.gate_proj.weight": "model-00006-of-00007.safetensors",
447
+ "model.layers.3.mlp.experts.218.up_proj.weight": "model-00006-of-00007.safetensors",
448
+ "model.layers.3.mlp.experts.219.down_proj.weight": "model-00006-of-00007.safetensors",
449
+ "model.layers.3.mlp.experts.219.gate_proj.weight": "model-00006-of-00007.safetensors",
450
+ "model.layers.3.mlp.experts.219.up_proj.weight": "model-00006-of-00007.safetensors",
451
+ "model.layers.3.mlp.experts.22.down_proj.weight": "model-00002-of-00007.safetensors",
452
+ "model.layers.3.mlp.experts.22.gate_proj.weight": "model-00002-of-00007.safetensors",
453
+ "model.layers.3.mlp.experts.22.up_proj.weight": "model-00002-of-00007.safetensors",
454
+ "model.layers.3.mlp.experts.220.down_proj.weight": "model-00006-of-00007.safetensors",
455
+ "model.layers.3.mlp.experts.220.gate_proj.weight": "model-00006-of-00007.safetensors",
456
+ "model.layers.3.mlp.experts.220.up_proj.weight": "model-00006-of-00007.safetensors",
457
+ "model.layers.3.mlp.experts.221.down_proj.weight": "model-00006-of-00007.safetensors",
458
+ "model.layers.3.mlp.experts.221.gate_proj.weight": "model-00006-of-00007.safetensors",
459
+ "model.layers.3.mlp.experts.221.up_proj.weight": "model-00006-of-00007.safetensors",
460
+ "model.layers.3.mlp.experts.222.down_proj.weight": "model-00006-of-00007.safetensors",
461
+ "model.layers.3.mlp.experts.222.gate_proj.weight": "model-00006-of-00007.safetensors",
462
+ "model.layers.3.mlp.experts.222.up_proj.weight": "model-00006-of-00007.safetensors",
463
+ "model.layers.3.mlp.experts.223.down_proj.weight": "model-00006-of-00007.safetensors",
464
+ "model.layers.3.mlp.experts.223.gate_proj.weight": "model-00006-of-00007.safetensors",
465
+ "model.layers.3.mlp.experts.223.up_proj.weight": "model-00006-of-00007.safetensors",
466
+ "model.layers.3.mlp.experts.224.down_proj.weight": "model-00006-of-00007.safetensors",
467
+ "model.layers.3.mlp.experts.224.gate_proj.weight": "model-00006-of-00007.safetensors",
468
+ "model.layers.3.mlp.experts.224.up_proj.weight": "model-00006-of-00007.safetensors",
469
+ "model.layers.3.mlp.experts.225.down_proj.weight": "model-00006-of-00007.safetensors",
470
+ "model.layers.3.mlp.experts.225.gate_proj.weight": "model-00006-of-00007.safetensors",
471
+ "model.layers.3.mlp.experts.225.up_proj.weight": "model-00006-of-00007.safetensors",
472
+ "model.layers.3.mlp.experts.226.down_proj.weight": "model-00006-of-00007.safetensors",
473
+ "model.layers.3.mlp.experts.226.gate_proj.weight": "model-00006-of-00007.safetensors",
474
+ "model.layers.3.mlp.experts.226.up_proj.weight": "model-00006-of-00007.safetensors",
475
+ "model.layers.3.mlp.experts.227.down_proj.weight": "model-00006-of-00007.safetensors",
476
+ "model.layers.3.mlp.experts.227.gate_proj.weight": "model-00006-of-00007.safetensors",
477
+ "model.layers.3.mlp.experts.227.up_proj.weight": "model-00006-of-00007.safetensors",
478
+ "model.layers.3.mlp.experts.228.down_proj.weight": "model-00006-of-00007.safetensors",
479
+ "model.layers.3.mlp.experts.228.gate_proj.weight": "model-00006-of-00007.safetensors",
480
+ "model.layers.3.mlp.experts.228.up_proj.weight": "model-00006-of-00007.safetensors",
481
+ "model.layers.3.mlp.experts.229.down_proj.weight": "model-00006-of-00007.safetensors",
482
+ "model.layers.3.mlp.experts.229.gate_proj.weight": "model-00006-of-00007.safetensors",
483
+ "model.layers.3.mlp.experts.229.up_proj.weight": "model-00006-of-00007.safetensors",
484
+ "model.layers.3.mlp.experts.23.down_proj.weight": "model-00002-of-00007.safetensors",
485
+ "model.layers.3.mlp.experts.23.gate_proj.weight": "model-00002-of-00007.safetensors",
486
+ "model.layers.3.mlp.experts.23.up_proj.weight": "model-00002-of-00007.safetensors",
487
+ "model.layers.3.mlp.experts.230.down_proj.weight": "model-00006-of-00007.safetensors",
488
+ "model.layers.3.mlp.experts.230.gate_proj.weight": "model-00006-of-00007.safetensors",
489
+ "model.layers.3.mlp.experts.230.up_proj.weight": "model-00006-of-00007.safetensors",
490
+ "model.layers.3.mlp.experts.231.down_proj.weight": "model-00006-of-00007.safetensors",
491
+ "model.layers.3.mlp.experts.231.gate_proj.weight": "model-00006-of-00007.safetensors",
492
+ "model.layers.3.mlp.experts.231.up_proj.weight": "model-00006-of-00007.safetensors",
493
+ "model.layers.3.mlp.experts.232.down_proj.weight": "model-00006-of-00007.safetensors",
494
+ "model.layers.3.mlp.experts.232.gate_proj.weight": "model-00006-of-00007.safetensors",
495
+ "model.layers.3.mlp.experts.232.up_proj.weight": "model-00006-of-00007.safetensors",
496
+ "model.layers.3.mlp.experts.233.down_proj.weight": "model-00006-of-00007.safetensors",
497
+ "model.layers.3.mlp.experts.233.gate_proj.weight": "model-00006-of-00007.safetensors",
498
+ "model.layers.3.mlp.experts.233.up_proj.weight": "model-00006-of-00007.safetensors",
499
+ "model.layers.3.mlp.experts.234.down_proj.weight": "model-00006-of-00007.safetensors",
500
+ "model.layers.3.mlp.experts.234.gate_proj.weight": "model-00006-of-00007.safetensors",
501
+ "model.layers.3.mlp.experts.234.up_proj.weight": "model-00006-of-00007.safetensors",
502
+ "model.layers.3.mlp.experts.235.down_proj.weight": "model-00006-of-00007.safetensors",
503
+ "model.layers.3.mlp.experts.235.gate_proj.weight": "model-00006-of-00007.safetensors",
504
+ "model.layers.3.mlp.experts.235.up_proj.weight": "model-00006-of-00007.safetensors",
505
+ "model.layers.3.mlp.experts.236.down_proj.weight": "model-00006-of-00007.safetensors",
506
+ "model.layers.3.mlp.experts.236.gate_proj.weight": "model-00006-of-00007.safetensors",
507
+ "model.layers.3.mlp.experts.236.up_proj.weight": "model-00006-of-00007.safetensors",
508
+ "model.layers.3.mlp.experts.237.down_proj.weight": "model-00006-of-00007.safetensors",
509
+ "model.layers.3.mlp.experts.237.gate_proj.weight": "model-00006-of-00007.safetensors",
510
+ "model.layers.3.mlp.experts.237.up_proj.weight": "model-00006-of-00007.safetensors",
511
+ "model.layers.3.mlp.experts.238.down_proj.weight": "model-00006-of-00007.safetensors",
512
+ "model.layers.3.mlp.experts.238.gate_proj.weight": "model-00006-of-00007.safetensors",
513
+ "model.layers.3.mlp.experts.238.up_proj.weight": "model-00006-of-00007.safetensors",
514
+ "model.layers.3.mlp.experts.239.down_proj.weight": "model-00006-of-00007.safetensors",
515
+ "model.layers.3.mlp.experts.239.gate_proj.weight": "model-00006-of-00007.safetensors",
516
+ "model.layers.3.mlp.experts.239.up_proj.weight": "model-00006-of-00007.safetensors",
517
+ "model.layers.3.mlp.experts.24.down_proj.weight": "model-00002-of-00007.safetensors",
518
+ "model.layers.3.mlp.experts.24.gate_proj.weight": "model-00002-of-00007.safetensors",
519
+ "model.layers.3.mlp.experts.24.up_proj.weight": "model-00002-of-00007.safetensors",
520
+ "model.layers.3.mlp.experts.240.down_proj.weight": "model-00006-of-00007.safetensors",
521
+ "model.layers.3.mlp.experts.240.gate_proj.weight": "model-00006-of-00007.safetensors",
522
+ "model.layers.3.mlp.experts.240.up_proj.weight": "model-00006-of-00007.safetensors",
523
+ "model.layers.3.mlp.experts.241.down_proj.weight": "model-00006-of-00007.safetensors",
524
+ "model.layers.3.mlp.experts.241.gate_proj.weight": "model-00006-of-00007.safetensors",
525
+ "model.layers.3.mlp.experts.241.up_proj.weight": "model-00006-of-00007.safetensors",
526
+ "model.layers.3.mlp.experts.242.down_proj.weight": "model-00006-of-00007.safetensors",
527
+ "model.layers.3.mlp.experts.242.gate_proj.weight": "model-00006-of-00007.safetensors",
528
+ "model.layers.3.mlp.experts.242.up_proj.weight": "model-00006-of-00007.safetensors",
529
+ "model.layers.3.mlp.experts.243.down_proj.weight": "model-00006-of-00007.safetensors",
530
+ "model.layers.3.mlp.experts.243.gate_proj.weight": "model-00006-of-00007.safetensors",
531
+ "model.layers.3.mlp.experts.243.up_proj.weight": "model-00006-of-00007.safetensors",
532
+ "model.layers.3.mlp.experts.244.down_proj.weight": "model-00006-of-00007.safetensors",
533
+ "model.layers.3.mlp.experts.244.gate_proj.weight": "model-00006-of-00007.safetensors",
534
+ "model.layers.3.mlp.experts.244.up_proj.weight": "model-00006-of-00007.safetensors",
535
+ "model.layers.3.mlp.experts.245.down_proj.weight": "model-00006-of-00007.safetensors",
536
+ "model.layers.3.mlp.experts.245.gate_proj.weight": "model-00006-of-00007.safetensors",
537
+ "model.layers.3.mlp.experts.245.up_proj.weight": "model-00006-of-00007.safetensors",
538
+ "model.layers.3.mlp.experts.246.down_proj.weight": "model-00006-of-00007.safetensors",
539
+ "model.layers.3.mlp.experts.246.gate_proj.weight": "model-00006-of-00007.safetensors",
540
+ "model.layers.3.mlp.experts.246.up_proj.weight": "model-00006-of-00007.safetensors",
541
+ "model.layers.3.mlp.experts.247.down_proj.weight": "model-00006-of-00007.safetensors",
542
+ "model.layers.3.mlp.experts.247.gate_proj.weight": "model-00006-of-00007.safetensors",
543
+ "model.layers.3.mlp.experts.247.up_proj.weight": "model-00006-of-00007.safetensors",
544
+ "model.layers.3.mlp.experts.248.down_proj.weight": "model-00006-of-00007.safetensors",
545
+ "model.layers.3.mlp.experts.248.gate_proj.weight": "model-00006-of-00007.safetensors",
546
+ "model.layers.3.mlp.experts.248.up_proj.weight": "model-00006-of-00007.safetensors",
547
+ "model.layers.3.mlp.experts.249.down_proj.weight": "model-00006-of-00007.safetensors",
548
+ "model.layers.3.mlp.experts.249.gate_proj.weight": "model-00006-of-00007.safetensors",
549
+ "model.layers.3.mlp.experts.249.up_proj.weight": "model-00006-of-00007.safetensors",
550
+ "model.layers.3.mlp.experts.25.down_proj.weight": "model-00002-of-00007.safetensors",
551
+ "model.layers.3.mlp.experts.25.gate_proj.weight": "model-00002-of-00007.safetensors",
552
+ "model.layers.3.mlp.experts.25.up_proj.weight": "model-00002-of-00007.safetensors",
553
+ "model.layers.3.mlp.experts.250.down_proj.weight": "model-00006-of-00007.safetensors",
554
+ "model.layers.3.mlp.experts.250.gate_proj.weight": "model-00006-of-00007.safetensors",
555
+ "model.layers.3.mlp.experts.250.up_proj.weight": "model-00006-of-00007.safetensors",
556
+ "model.layers.3.mlp.experts.251.down_proj.weight": "model-00006-of-00007.safetensors",
557
+ "model.layers.3.mlp.experts.251.gate_proj.weight": "model-00006-of-00007.safetensors",
558
+ "model.layers.3.mlp.experts.251.up_proj.weight": "model-00006-of-00007.safetensors",
559
+ "model.layers.3.mlp.experts.252.down_proj.weight": "model-00006-of-00007.safetensors",
560
+ "model.layers.3.mlp.experts.252.gate_proj.weight": "model-00006-of-00007.safetensors",
561
+ "model.layers.3.mlp.experts.252.up_proj.weight": "model-00006-of-00007.safetensors",
562
+ "model.layers.3.mlp.experts.253.down_proj.weight": "model-00006-of-00007.safetensors",
563
+ "model.layers.3.mlp.experts.253.gate_proj.weight": "model-00006-of-00007.safetensors",
564
+ "model.layers.3.mlp.experts.253.up_proj.weight": "model-00006-of-00007.safetensors",
565
+ "model.layers.3.mlp.experts.254.down_proj.weight": "model-00006-of-00007.safetensors",
566
+ "model.layers.3.mlp.experts.254.gate_proj.weight": "model-00006-of-00007.safetensors",
567
+ "model.layers.3.mlp.experts.254.up_proj.weight": "model-00006-of-00007.safetensors",
568
+ "model.layers.3.mlp.experts.255.down_proj.weight": "model-00006-of-00007.safetensors",
569
+ "model.layers.3.mlp.experts.255.gate_proj.weight": "model-00006-of-00007.safetensors",
570
+ "model.layers.3.mlp.experts.255.up_proj.weight": "model-00006-of-00007.safetensors",
571
+ "model.layers.3.mlp.experts.26.down_proj.weight": "model-00002-of-00007.safetensors",
572
+ "model.layers.3.mlp.experts.26.gate_proj.weight": "model-00002-of-00007.safetensors",
573
+ "model.layers.3.mlp.experts.26.up_proj.weight": "model-00002-of-00007.safetensors",
574
+ "model.layers.3.mlp.experts.27.down_proj.weight": "model-00002-of-00007.safetensors",
575
+ "model.layers.3.mlp.experts.27.gate_proj.weight": "model-00002-of-00007.safetensors",
576
+ "model.layers.3.mlp.experts.27.up_proj.weight": "model-00002-of-00007.safetensors",
577
+ "model.layers.3.mlp.experts.28.down_proj.weight": "model-00002-of-00007.safetensors",
578
+ "model.layers.3.mlp.experts.28.gate_proj.weight": "model-00002-of-00007.safetensors",
579
+ "model.layers.3.mlp.experts.28.up_proj.weight": "model-00002-of-00007.safetensors",
580
+ "model.layers.3.mlp.experts.29.down_proj.weight": "model-00002-of-00007.safetensors",
581
+ "model.layers.3.mlp.experts.29.gate_proj.weight": "model-00002-of-00007.safetensors",
582
+ "model.layers.3.mlp.experts.29.up_proj.weight": "model-00002-of-00007.safetensors",
583
+ "model.layers.3.mlp.experts.3.down_proj.weight": "model-00002-of-00007.safetensors",
584
+ "model.layers.3.mlp.experts.3.gate_proj.weight": "model-00002-of-00007.safetensors",
585
+ "model.layers.3.mlp.experts.3.up_proj.weight": "model-00002-of-00007.safetensors",
586
+ "model.layers.3.mlp.experts.30.down_proj.weight": "model-00002-of-00007.safetensors",
587
+ "model.layers.3.mlp.experts.30.gate_proj.weight": "model-00002-of-00007.safetensors",
588
+ "model.layers.3.mlp.experts.30.up_proj.weight": "model-00002-of-00007.safetensors",
589
+ "model.layers.3.mlp.experts.31.down_proj.weight": "model-00002-of-00007.safetensors",
590
+ "model.layers.3.mlp.experts.31.gate_proj.weight": "model-00002-of-00007.safetensors",
591
+ "model.layers.3.mlp.experts.31.up_proj.weight": "model-00002-of-00007.safetensors",
592
+ "model.layers.3.mlp.experts.32.down_proj.weight": "model-00002-of-00007.safetensors",
593
+ "model.layers.3.mlp.experts.32.gate_proj.weight": "model-00002-of-00007.safetensors",
594
+ "model.layers.3.mlp.experts.32.up_proj.weight": "model-00002-of-00007.safetensors",
595
+ "model.layers.3.mlp.experts.33.down_proj.weight": "model-00002-of-00007.safetensors",
596
+ "model.layers.3.mlp.experts.33.gate_proj.weight": "model-00002-of-00007.safetensors",
597
+ "model.layers.3.mlp.experts.33.up_proj.weight": "model-00002-of-00007.safetensors",
598
+ "model.layers.3.mlp.experts.34.down_proj.weight": "model-00002-of-00007.safetensors",
599
+ "model.layers.3.mlp.experts.34.gate_proj.weight": "model-00002-of-00007.safetensors",
600
+ "model.layers.3.mlp.experts.34.up_proj.weight": "model-00002-of-00007.safetensors",
601
+ "model.layers.3.mlp.experts.35.down_proj.weight": "model-00002-of-00007.safetensors",
602
+ "model.layers.3.mlp.experts.35.gate_proj.weight": "model-00002-of-00007.safetensors",
603
+ "model.layers.3.mlp.experts.35.up_proj.weight": "model-00002-of-00007.safetensors",
604
+ "model.layers.3.mlp.experts.36.down_proj.weight": "model-00002-of-00007.safetensors",
605
+ "model.layers.3.mlp.experts.36.gate_proj.weight": "model-00002-of-00007.safetensors",
606
+ "model.layers.3.mlp.experts.36.up_proj.weight": "model-00002-of-00007.safetensors",
607
+ "model.layers.3.mlp.experts.37.down_proj.weight": "model-00002-of-00007.safetensors",
608
+ "model.layers.3.mlp.experts.37.gate_proj.weight": "model-00002-of-00007.safetensors",
609
+ "model.layers.3.mlp.experts.37.up_proj.weight": "model-00002-of-00007.safetensors",
610
+ "model.layers.3.mlp.experts.38.down_proj.weight": "model-00002-of-00007.safetensors",
611
+ "model.layers.3.mlp.experts.38.gate_proj.weight": "model-00002-of-00007.safetensors",
612
+ "model.layers.3.mlp.experts.38.up_proj.weight": "model-00002-of-00007.safetensors",
613
+ "model.layers.3.mlp.experts.39.down_proj.weight": "model-00002-of-00007.safetensors",
614
+ "model.layers.3.mlp.experts.39.gate_proj.weight": "model-00002-of-00007.safetensors",
615
+ "model.layers.3.mlp.experts.39.up_proj.weight": "model-00002-of-00007.safetensors",
616
+ "model.layers.3.mlp.experts.4.down_proj.weight": "model-00002-of-00007.safetensors",
617
+ "model.layers.3.mlp.experts.4.gate_proj.weight": "model-00002-of-00007.safetensors",
618
+ "model.layers.3.mlp.experts.4.up_proj.weight": "model-00002-of-00007.safetensors",
619
+ "model.layers.3.mlp.experts.40.down_proj.weight": "model-00002-of-00007.safetensors",
620
+ "model.layers.3.mlp.experts.40.gate_proj.weight": "model-00002-of-00007.safetensors",
621
+ "model.layers.3.mlp.experts.40.up_proj.weight": "model-00002-of-00007.safetensors",
622
+ "model.layers.3.mlp.experts.41.down_proj.weight": "model-00002-of-00007.safetensors",
623
+ "model.layers.3.mlp.experts.41.gate_proj.weight": "model-00002-of-00007.safetensors",
624
+ "model.layers.3.mlp.experts.41.up_proj.weight": "model-00002-of-00007.safetensors",
625
+ "model.layers.3.mlp.experts.42.down_proj.weight": "model-00002-of-00007.safetensors",
626
+ "model.layers.3.mlp.experts.42.gate_proj.weight": "model-00002-of-00007.safetensors",
627
+ "model.layers.3.mlp.experts.42.up_proj.weight": "model-00002-of-00007.safetensors",
628
+ "model.layers.3.mlp.experts.43.down_proj.weight": "model-00002-of-00007.safetensors",
629
+ "model.layers.3.mlp.experts.43.gate_proj.weight": "model-00002-of-00007.safetensors",
630
+ "model.layers.3.mlp.experts.43.up_proj.weight": "model-00002-of-00007.safetensors",
631
+ "model.layers.3.mlp.experts.44.down_proj.weight": "model-00002-of-00007.safetensors",
632
+ "model.layers.3.mlp.experts.44.gate_proj.weight": "model-00002-of-00007.safetensors",
633
+ "model.layers.3.mlp.experts.44.up_proj.weight": "model-00002-of-00007.safetensors",
634
+ "model.layers.3.mlp.experts.45.down_proj.weight": "model-00002-of-00007.safetensors",
635
+ "model.layers.3.mlp.experts.45.gate_proj.weight": "model-00002-of-00007.safetensors",
636
+ "model.layers.3.mlp.experts.45.up_proj.weight": "model-00002-of-00007.safetensors",
637
+ "model.layers.3.mlp.experts.46.down_proj.weight": "model-00003-of-00007.safetensors",
638
+ "model.layers.3.mlp.experts.46.gate_proj.weight": "model-00002-of-00007.safetensors",
639
+ "model.layers.3.mlp.experts.46.up_proj.weight": "model-00003-of-00007.safetensors",
640
+ "model.layers.3.mlp.experts.47.down_proj.weight": "model-00003-of-00007.safetensors",
641
+ "model.layers.3.mlp.experts.47.gate_proj.weight": "model-00003-of-00007.safetensors",
642
+ "model.layers.3.mlp.experts.47.up_proj.weight": "model-00003-of-00007.safetensors",
643
+ "model.layers.3.mlp.experts.48.down_proj.weight": "model-00003-of-00007.safetensors",
644
+ "model.layers.3.mlp.experts.48.gate_proj.weight": "model-00003-of-00007.safetensors",
645
+ "model.layers.3.mlp.experts.48.up_proj.weight": "model-00003-of-00007.safetensors",
646
+ "model.layers.3.mlp.experts.49.down_proj.weight": "model-00003-of-00007.safetensors",
647
+ "model.layers.3.mlp.experts.49.gate_proj.weight": "model-00003-of-00007.safetensors",
648
+ "model.layers.3.mlp.experts.49.up_proj.weight": "model-00003-of-00007.safetensors",
649
+ "model.layers.3.mlp.experts.5.down_proj.weight": "model-00002-of-00007.safetensors",
650
+ "model.layers.3.mlp.experts.5.gate_proj.weight": "model-00002-of-00007.safetensors",
651
+ "model.layers.3.mlp.experts.5.up_proj.weight": "model-00002-of-00007.safetensors",
652
+ "model.layers.3.mlp.experts.50.down_proj.weight": "model-00003-of-00007.safetensors",
653
+ "model.layers.3.mlp.experts.50.gate_proj.weight": "model-00003-of-00007.safetensors",
654
+ "model.layers.3.mlp.experts.50.up_proj.weight": "model-00003-of-00007.safetensors",
655
+ "model.layers.3.mlp.experts.51.down_proj.weight": "model-00003-of-00007.safetensors",
656
+ "model.layers.3.mlp.experts.51.gate_proj.weight": "model-00003-of-00007.safetensors",
657
+ "model.layers.3.mlp.experts.51.up_proj.weight": "model-00003-of-00007.safetensors",
658
+ "model.layers.3.mlp.experts.52.down_proj.weight": "model-00003-of-00007.safetensors",
659
+ "model.layers.3.mlp.experts.52.gate_proj.weight": "model-00003-of-00007.safetensors",
660
+ "model.layers.3.mlp.experts.52.up_proj.weight": "model-00003-of-00007.safetensors",
661
+ "model.layers.3.mlp.experts.53.down_proj.weight": "model-00003-of-00007.safetensors",
662
+ "model.layers.3.mlp.experts.53.gate_proj.weight": "model-00003-of-00007.safetensors",
663
+ "model.layers.3.mlp.experts.53.up_proj.weight": "model-00003-of-00007.safetensors",
664
+ "model.layers.3.mlp.experts.54.down_proj.weight": "model-00003-of-00007.safetensors",
665
+ "model.layers.3.mlp.experts.54.gate_proj.weight": "model-00003-of-00007.safetensors",
666
+ "model.layers.3.mlp.experts.54.up_proj.weight": "model-00003-of-00007.safetensors",
667
+ "model.layers.3.mlp.experts.55.down_proj.weight": "model-00003-of-00007.safetensors",
668
+ "model.layers.3.mlp.experts.55.gate_proj.weight": "model-00003-of-00007.safetensors",
669
+ "model.layers.3.mlp.experts.55.up_proj.weight": "model-00003-of-00007.safetensors",
670
+ "model.layers.3.mlp.experts.56.down_proj.weight": "model-00003-of-00007.safetensors",
671
+ "model.layers.3.mlp.experts.56.gate_proj.weight": "model-00003-of-00007.safetensors",
672
+ "model.layers.3.mlp.experts.56.up_proj.weight": "model-00003-of-00007.safetensors",
673
+ "model.layers.3.mlp.experts.57.down_proj.weight": "model-00003-of-00007.safetensors",
674
+ "model.layers.3.mlp.experts.57.gate_proj.weight": "model-00003-of-00007.safetensors",
675
+ "model.layers.3.mlp.experts.57.up_proj.weight": "model-00003-of-00007.safetensors",
676
+ "model.layers.3.mlp.experts.58.down_proj.weight": "model-00003-of-00007.safetensors",
677
+ "model.layers.3.mlp.experts.58.gate_proj.weight": "model-00003-of-00007.safetensors",
678
+ "model.layers.3.mlp.experts.58.up_proj.weight": "model-00003-of-00007.safetensors",
679
+ "model.layers.3.mlp.experts.59.down_proj.weight": "model-00003-of-00007.safetensors",
680
+ "model.layers.3.mlp.experts.59.gate_proj.weight": "model-00003-of-00007.safetensors",
681
+ "model.layers.3.mlp.experts.59.up_proj.weight": "model-00003-of-00007.safetensors",
682
+ "model.layers.3.mlp.experts.6.down_proj.weight": "model-00002-of-00007.safetensors",
683
+ "model.layers.3.mlp.experts.6.gate_proj.weight": "model-00002-of-00007.safetensors",
684
+ "model.layers.3.mlp.experts.6.up_proj.weight": "model-00002-of-00007.safetensors",
685
+ "model.layers.3.mlp.experts.60.down_proj.weight": "model-00003-of-00007.safetensors",
686
+ "model.layers.3.mlp.experts.60.gate_proj.weight": "model-00003-of-00007.safetensors",
687
+ "model.layers.3.mlp.experts.60.up_proj.weight": "model-00003-of-00007.safetensors",
688
+ "model.layers.3.mlp.experts.61.down_proj.weight": "model-00003-of-00007.safetensors",
689
+ "model.layers.3.mlp.experts.61.gate_proj.weight": "model-00003-of-00007.safetensors",
690
+ "model.layers.3.mlp.experts.61.up_proj.weight": "model-00003-of-00007.safetensors",
691
+ "model.layers.3.mlp.experts.62.down_proj.weight": "model-00003-of-00007.safetensors",
692
+ "model.layers.3.mlp.experts.62.gate_proj.weight": "model-00003-of-00007.safetensors",
693
+ "model.layers.3.mlp.experts.62.up_proj.weight": "model-00003-of-00007.safetensors",
694
+ "model.layers.3.mlp.experts.63.down_proj.weight": "model-00003-of-00007.safetensors",
695
+ "model.layers.3.mlp.experts.63.gate_proj.weight": "model-00003-of-00007.safetensors",
696
+ "model.layers.3.mlp.experts.63.up_proj.weight": "model-00003-of-00007.safetensors",
697
+ "model.layers.3.mlp.experts.64.down_proj.weight": "model-00003-of-00007.safetensors",
698
+ "model.layers.3.mlp.experts.64.gate_proj.weight": "model-00003-of-00007.safetensors",
699
+ "model.layers.3.mlp.experts.64.up_proj.weight": "model-00003-of-00007.safetensors",
700
+ "model.layers.3.mlp.experts.65.down_proj.weight": "model-00003-of-00007.safetensors",
701
+ "model.layers.3.mlp.experts.65.gate_proj.weight": "model-00003-of-00007.safetensors",
702
+ "model.layers.3.mlp.experts.65.up_proj.weight": "model-00003-of-00007.safetensors",
703
+ "model.layers.3.mlp.experts.66.down_proj.weight": "model-00003-of-00007.safetensors",
704
+ "model.layers.3.mlp.experts.66.gate_proj.weight": "model-00003-of-00007.safetensors",
705
+ "model.layers.3.mlp.experts.66.up_proj.weight": "model-00003-of-00007.safetensors",
706
+ "model.layers.3.mlp.experts.67.down_proj.weight": "model-00003-of-00007.safetensors",
707
+ "model.layers.3.mlp.experts.67.gate_proj.weight": "model-00003-of-00007.safetensors",
708
+ "model.layers.3.mlp.experts.67.up_proj.weight": "model-00003-of-00007.safetensors",
709
+ "model.layers.3.mlp.experts.68.down_proj.weight": "model-00003-of-00007.safetensors",
710
+ "model.layers.3.mlp.experts.68.gate_proj.weight": "model-00003-of-00007.safetensors",
711
+ "model.layers.3.mlp.experts.68.up_proj.weight": "model-00003-of-00007.safetensors",
712
+ "model.layers.3.mlp.experts.69.down_proj.weight": "model-00003-of-00007.safetensors",
713
+ "model.layers.3.mlp.experts.69.gate_proj.weight": "model-00003-of-00007.safetensors",
714
+ "model.layers.3.mlp.experts.69.up_proj.weight": "model-00003-of-00007.safetensors",
715
+ "model.layers.3.mlp.experts.7.down_proj.weight": "model-00002-of-00007.safetensors",
716
+ "model.layers.3.mlp.experts.7.gate_proj.weight": "model-00002-of-00007.safetensors",
717
+ "model.layers.3.mlp.experts.7.up_proj.weight": "model-00002-of-00007.safetensors",
718
+ "model.layers.3.mlp.experts.70.down_proj.weight": "model-00003-of-00007.safetensors",
719
+ "model.layers.3.mlp.experts.70.gate_proj.weight": "model-00003-of-00007.safetensors",
720
+ "model.layers.3.mlp.experts.70.up_proj.weight": "model-00003-of-00007.safetensors",
721
+ "model.layers.3.mlp.experts.71.down_proj.weight": "model-00003-of-00007.safetensors",
722
+ "model.layers.3.mlp.experts.71.gate_proj.weight": "model-00003-of-00007.safetensors",
723
+ "model.layers.3.mlp.experts.71.up_proj.weight": "model-00003-of-00007.safetensors",
724
+ "model.layers.3.mlp.experts.72.down_proj.weight": "model-00003-of-00007.safetensors",
725
+ "model.layers.3.mlp.experts.72.gate_proj.weight": "model-00003-of-00007.safetensors",
726
+ "model.layers.3.mlp.experts.72.up_proj.weight": "model-00003-of-00007.safetensors",
727
+ "model.layers.3.mlp.experts.73.down_proj.weight": "model-00003-of-00007.safetensors",
728
+ "model.layers.3.mlp.experts.73.gate_proj.weight": "model-00003-of-00007.safetensors",
729
+ "model.layers.3.mlp.experts.73.up_proj.weight": "model-00003-of-00007.safetensors",
730
+ "model.layers.3.mlp.experts.74.down_proj.weight": "model-00003-of-00007.safetensors",
731
+ "model.layers.3.mlp.experts.74.gate_proj.weight": "model-00003-of-00007.safetensors",
732
+ "model.layers.3.mlp.experts.74.up_proj.weight": "model-00003-of-00007.safetensors",
733
+ "model.layers.3.mlp.experts.75.down_proj.weight": "model-00003-of-00007.safetensors",
734
+ "model.layers.3.mlp.experts.75.gate_proj.weight": "model-00003-of-00007.safetensors",
735
+ "model.layers.3.mlp.experts.75.up_proj.weight": "model-00003-of-00007.safetensors",
736
+ "model.layers.3.mlp.experts.76.down_proj.weight": "model-00003-of-00007.safetensors",
737
+ "model.layers.3.mlp.experts.76.gate_proj.weight": "model-00003-of-00007.safetensors",
738
+ "model.layers.3.mlp.experts.76.up_proj.weight": "model-00003-of-00007.safetensors",
739
+ "model.layers.3.mlp.experts.77.down_proj.weight": "model-00003-of-00007.safetensors",
740
+ "model.layers.3.mlp.experts.77.gate_proj.weight": "model-00003-of-00007.safetensors",
741
+ "model.layers.3.mlp.experts.77.up_proj.weight": "model-00003-of-00007.safetensors",
742
+ "model.layers.3.mlp.experts.78.down_proj.weight": "model-00003-of-00007.safetensors",
743
+ "model.layers.3.mlp.experts.78.gate_proj.weight": "model-00003-of-00007.safetensors",
744
+ "model.layers.3.mlp.experts.78.up_proj.weight": "model-00003-of-00007.safetensors",
745
+ "model.layers.3.mlp.experts.79.down_proj.weight": "model-00003-of-00007.safetensors",
746
+ "model.layers.3.mlp.experts.79.gate_proj.weight": "model-00003-of-00007.safetensors",
747
+ "model.layers.3.mlp.experts.79.up_proj.weight": "model-00003-of-00007.safetensors",
748
+ "model.layers.3.mlp.experts.8.down_proj.weight": "model-00002-of-00007.safetensors",
749
+ "model.layers.3.mlp.experts.8.gate_proj.weight": "model-00002-of-00007.safetensors",
750
+ "model.layers.3.mlp.experts.8.up_proj.weight": "model-00002-of-00007.safetensors",
751
+ "model.layers.3.mlp.experts.80.down_proj.weight": "model-00003-of-00007.safetensors",
752
+ "model.layers.3.mlp.experts.80.gate_proj.weight": "model-00003-of-00007.safetensors",
753
+ "model.layers.3.mlp.experts.80.up_proj.weight": "model-00003-of-00007.safetensors",
754
+ "model.layers.3.mlp.experts.81.down_proj.weight": "model-00003-of-00007.safetensors",
755
+ "model.layers.3.mlp.experts.81.gate_proj.weight": "model-00003-of-00007.safetensors",
756
+ "model.layers.3.mlp.experts.81.up_proj.weight": "model-00003-of-00007.safetensors",
757
+ "model.layers.3.mlp.experts.82.down_proj.weight": "model-00003-of-00007.safetensors",
758
+ "model.layers.3.mlp.experts.82.gate_proj.weight": "model-00003-of-00007.safetensors",
759
+ "model.layers.3.mlp.experts.82.up_proj.weight": "model-00003-of-00007.safetensors",
760
+ "model.layers.3.mlp.experts.83.down_proj.weight": "model-00003-of-00007.safetensors",
761
+ "model.layers.3.mlp.experts.83.gate_proj.weight": "model-00003-of-00007.safetensors",
762
+ "model.layers.3.mlp.experts.83.up_proj.weight": "model-00003-of-00007.safetensors",
763
+ "model.layers.3.mlp.experts.84.down_proj.weight": "model-00003-of-00007.safetensors",
764
+ "model.layers.3.mlp.experts.84.gate_proj.weight": "model-00003-of-00007.safetensors",
765
+ "model.layers.3.mlp.experts.84.up_proj.weight": "model-00003-of-00007.safetensors",
766
+ "model.layers.3.mlp.experts.85.down_proj.weight": "model-00003-of-00007.safetensors",
767
+ "model.layers.3.mlp.experts.85.gate_proj.weight": "model-00003-of-00007.safetensors",
768
+ "model.layers.3.mlp.experts.85.up_proj.weight": "model-00003-of-00007.safetensors",
769
+ "model.layers.3.mlp.experts.86.down_proj.weight": "model-00003-of-00007.safetensors",
770
+ "model.layers.3.mlp.experts.86.gate_proj.weight": "model-00003-of-00007.safetensors",
771
+ "model.layers.3.mlp.experts.86.up_proj.weight": "model-00003-of-00007.safetensors",
772
+ "model.layers.3.mlp.experts.87.down_proj.weight": "model-00003-of-00007.safetensors",
773
+ "model.layers.3.mlp.experts.87.gate_proj.weight": "model-00003-of-00007.safetensors",
774
+ "model.layers.3.mlp.experts.87.up_proj.weight": "model-00003-of-00007.safetensors",
775
+ "model.layers.3.mlp.experts.88.down_proj.weight": "model-00003-of-00007.safetensors",
776
+ "model.layers.3.mlp.experts.88.gate_proj.weight": "model-00003-of-00007.safetensors",
777
+ "model.layers.3.mlp.experts.88.up_proj.weight": "model-00003-of-00007.safetensors",
778
+ "model.layers.3.mlp.experts.89.down_proj.weight": "model-00003-of-00007.safetensors",
779
+ "model.layers.3.mlp.experts.89.gate_proj.weight": "model-00003-of-00007.safetensors",
780
+ "model.layers.3.mlp.experts.89.up_proj.weight": "model-00003-of-00007.safetensors",
781
+ "model.layers.3.mlp.experts.9.down_proj.weight": "model-00002-of-00007.safetensors",
782
+ "model.layers.3.mlp.experts.9.gate_proj.weight": "model-00002-of-00007.safetensors",
783
+ "model.layers.3.mlp.experts.9.up_proj.weight": "model-00002-of-00007.safetensors",
784
+ "model.layers.3.mlp.experts.90.down_proj.weight": "model-00003-of-00007.safetensors",
785
+ "model.layers.3.mlp.experts.90.gate_proj.weight": "model-00003-of-00007.safetensors",
786
+ "model.layers.3.mlp.experts.90.up_proj.weight": "model-00003-of-00007.safetensors",
787
+ "model.layers.3.mlp.experts.91.down_proj.weight": "model-00003-of-00007.safetensors",
788
+ "model.layers.3.mlp.experts.91.gate_proj.weight": "model-00003-of-00007.safetensors",
789
+ "model.layers.3.mlp.experts.91.up_proj.weight": "model-00003-of-00007.safetensors",
790
+ "model.layers.3.mlp.experts.92.down_proj.weight": "model-00003-of-00007.safetensors",
791
+ "model.layers.3.mlp.experts.92.gate_proj.weight": "model-00003-of-00007.safetensors",
792
+ "model.layers.3.mlp.experts.92.up_proj.weight": "model-00003-of-00007.safetensors",
793
+ "model.layers.3.mlp.experts.93.down_proj.weight": "model-00003-of-00007.safetensors",
794
+ "model.layers.3.mlp.experts.93.gate_proj.weight": "model-00003-of-00007.safetensors",
795
+ "model.layers.3.mlp.experts.93.up_proj.weight": "model-00003-of-00007.safetensors",
796
+ "model.layers.3.mlp.experts.94.down_proj.weight": "model-00003-of-00007.safetensors",
797
+ "model.layers.3.mlp.experts.94.gate_proj.weight": "model-00003-of-00007.safetensors",
798
+ "model.layers.3.mlp.experts.94.up_proj.weight": "model-00003-of-00007.safetensors",
799
+ "model.layers.3.mlp.experts.95.down_proj.weight": "model-00003-of-00007.safetensors",
800
+ "model.layers.3.mlp.experts.95.gate_proj.weight": "model-00003-of-00007.safetensors",
801
+ "model.layers.3.mlp.experts.95.up_proj.weight": "model-00003-of-00007.safetensors",
802
+ "model.layers.3.mlp.experts.96.down_proj.weight": "model-00003-of-00007.safetensors",
803
+ "model.layers.3.mlp.experts.96.gate_proj.weight": "model-00003-of-00007.safetensors",
804
+ "model.layers.3.mlp.experts.96.up_proj.weight": "model-00003-of-00007.safetensors",
805
+ "model.layers.3.mlp.experts.97.down_proj.weight": "model-00003-of-00007.safetensors",
806
+ "model.layers.3.mlp.experts.97.gate_proj.weight": "model-00003-of-00007.safetensors",
807
+ "model.layers.3.mlp.experts.97.up_proj.weight": "model-00003-of-00007.safetensors",
808
+ "model.layers.3.mlp.experts.98.down_proj.weight": "model-00003-of-00007.safetensors",
809
+ "model.layers.3.mlp.experts.98.gate_proj.weight": "model-00003-of-00007.safetensors",
810
+ "model.layers.3.mlp.experts.98.up_proj.weight": "model-00003-of-00007.safetensors",
811
+ "model.layers.3.mlp.experts.99.down_proj.weight": "model-00003-of-00007.safetensors",
812
+ "model.layers.3.mlp.experts.99.gate_proj.weight": "model-00003-of-00007.safetensors",
813
+ "model.layers.3.mlp.experts.99.up_proj.weight": "model-00003-of-00007.safetensors",
814
+ "model.layers.3.mlp.gate.e_score_correction_bias": "model-00006-of-00007.safetensors",
815
+ "model.layers.3.mlp.gate.weight": "model-00006-of-00007.safetensors",
816
+ "model.layers.3.mlp.shared_experts.down_proj.weight": "model-00006-of-00007.safetensors",
817
+ "model.layers.3.mlp.shared_experts.gate_proj.weight": "model-00006-of-00007.safetensors",
818
+ "model.layers.3.mlp.shared_experts.up_proj.weight": "model-00006-of-00007.safetensors",
819
+ "model.layers.3.post_attention_layernorm.weight": "model-00006-of-00007.safetensors",
820
+ "model.layers.3.self_attn.kv_a_layernorm.weight": "model-00002-of-00007.safetensors",
821
+ "model.layers.3.self_attn.kv_a_proj_with_mqa.weight": "model-00002-of-00007.safetensors",
822
+ "model.layers.3.self_attn.kv_b_proj.weight": "model-00002-of-00007.safetensors",
823
+ "model.layers.3.self_attn.o_proj.weight": "model-00002-of-00007.safetensors",
824
+ "model.layers.3.self_attn.q_a_layernorm.weight": "model-00002-of-00007.safetensors",
825
+ "model.layers.3.self_attn.q_a_proj.weight": "model-00002-of-00007.safetensors",
826
+ "model.layers.3.self_attn.q_b_proj.weight": "model-00002-of-00007.safetensors",
827
+ "model.norm.weight": "model-00006-of-00007.safetensors"
828
+ }
829
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|begin▁of▁sentence|>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|end▁of▁sentence|>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<|end▁of▁sentence|>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff