MAGAer13 Mizukiluke commited on
Commit
0d071fe
1 Parent(s): d5bcf41

Upload 43 files (#1)

Browse files

- Upload 43 files (236ae1f25afd559e6b8684311b00388e4b3e7d7f)


Co-authored-by: Luke <[email protected]>

README.md ADDED
@@ -0,0 +1,52 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tasks:
3
+
4
+ - multimodal-dialogue
5
+
6
+ studios:
7
+
8
+ - damo/mPLUG-Owl
9
+
10
+ model-type:
11
+
12
+ - mplug-owl2
13
+
14
+ domain:
15
+
16
+ - multi-modal
17
+
18
+ frameworks:
19
+
20
+ - pytorch
21
+
22
+ backbone:
23
+
24
+ - transformer
25
+
26
+ containers:
27
+
28
+ license: apache-2.0
29
+
30
+ language:
31
+
32
+ - en
33
+
34
+ tags:
35
+
36
+ - transformer
37
+ - mPLUG
38
+ - Multimodal
39
+ - ChatGPT
40
+ - GPT
41
+ - Alibaba
42
+
43
+ ---
44
+
45
+ # mPLUG-Owl2介绍
46
+
47
+ mPLUG-Owl2是一种面向多模态语言模型的模块化的训练范式。其能学习与语言空间相适应的视觉知识,并支持在多模态场景(支持图片、文本输入)下进行多轮对话。它涌现多图关系理解,场景文本理解和基于视觉的文档理解等能力。
48
+
49
+ ## 模型描述
50
+
51
+ mPLUG-Owl2基于mPLUG-2模块化的思想,通过多阶段分别训练模型的视觉底座与语言模型,使其视觉知识能与预训练语言模型紧密协作,达到了显著优于主流多模态语言模型的效果。
52
+ ![Training paradigm and model overview](assets/mplug_owl2_radar.png)
assets/mplug_owl2_radar.png ADDED
config.json ADDED
@@ -0,0 +1,176 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "attention_bias": false,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "hidden_act": "silu",
6
+ "hidden_size": 4096,
7
+ "initializer_range": 0.02,
8
+ "intermediate_size": 11008,
9
+ "max_position_embeddings": 2048,
10
+ "model_type": "mplug_owl2",
11
+ "num_attention_heads": 32,
12
+ "num_hidden_layers": 32,
13
+ "num_key_value_heads": 32,
14
+ "pretraining_tp": 1,
15
+ "rms_norm_eps": 1e-06,
16
+ "rope_scaling": null,
17
+ "rope_theta": 10000.0,
18
+ "tie_word_embeddings": false,
19
+ "transformers_version": "4.28.1",
20
+ "use_cache": true,
21
+ "visual_config": {
22
+ "visual_abstractor": {
23
+ "_name_or_path": "",
24
+ "add_cross_attention": false,
25
+ "architectures": null,
26
+ "attention_probs_dropout_prob": 0.0,
27
+ "bad_words_ids": null,
28
+ "begin_suppress_tokens": null,
29
+ "bos_token_id": null,
30
+ "chunk_size_feed_forward": 0,
31
+ "cross_attention_hidden_size": null,
32
+ "decoder_start_token_id": null,
33
+ "diversity_penalty": 0.0,
34
+ "do_sample": false,
35
+ "early_stopping": false,
36
+ "encoder_hidden_size": 1024,
37
+ "encoder_no_repeat_ngram_size": 0,
38
+ "eos_token_id": null,
39
+ "exponential_decay_length_penalty": null,
40
+ "finetuning_task": null,
41
+ "forced_bos_token_id": null,
42
+ "forced_eos_token_id": null,
43
+ "grid_size": 32,
44
+ "hidden_size": 1024,
45
+ "id2label": {
46
+ "0": "LABEL_0",
47
+ "1": "LABEL_1"
48
+ },
49
+ "initializer_range": 0.02,
50
+ "intermediate_size": 2816,
51
+ "is_decoder": false,
52
+ "is_encoder_decoder": false,
53
+ "label2id": {
54
+ "LABEL_0": 0,
55
+ "LABEL_1": 1
56
+ },
57
+ "layer_norm_eps": 1e-06,
58
+ "length_penalty": 1.0,
59
+ "max_length": 20,
60
+ "min_length": 0,
61
+ "model_type": "mplug_owl_visual_abstract",
62
+ "no_repeat_ngram_size": 0,
63
+ "num_attention_heads": 16,
64
+ "num_beam_groups": 1,
65
+ "num_beams": 1,
66
+ "num_hidden_layers": 6,
67
+ "num_learnable_queries": 64,
68
+ "num_return_sequences": 1,
69
+ "output_attentions": false,
70
+ "output_hidden_states": false,
71
+ "output_scores": false,
72
+ "pad_token_id": null,
73
+ "prefix": null,
74
+ "problem_type": null,
75
+ "pruned_heads": {},
76
+ "remove_invalid_values": false,
77
+ "repetition_penalty": 1.0,
78
+ "return_dict": true,
79
+ "return_dict_in_generate": false,
80
+ "sep_token_id": null,
81
+ "suppress_tokens": null,
82
+ "task_specific_params": null,
83
+ "temperature": 1.0,
84
+ "tf_legacy_loss": false,
85
+ "tie_encoder_decoder": false,
86
+ "tie_word_embeddings": true,
87
+ "tokenizer_class": null,
88
+ "top_k": 50,
89
+ "top_p": 1.0,
90
+ "torch_dtype": null,
91
+ "torchscript": false,
92
+ "transformers_version": "4.28.1",
93
+ "typical_p": 1.0,
94
+ "use_bfloat16": false
95
+ },
96
+ "visual_model": {
97
+ "_name_or_path": "",
98
+ "add_cross_attention": false,
99
+ "architectures": null,
100
+ "attention_dropout": 0.0,
101
+ "bad_words_ids": null,
102
+ "begin_suppress_tokens": null,
103
+ "bos_token_id": null,
104
+ "chunk_size_feed_forward": 0,
105
+ "cross_attention_hidden_size": null,
106
+ "decoder_start_token_id": null,
107
+ "diversity_penalty": 0.0,
108
+ "do_sample": false,
109
+ "early_stopping": false,
110
+ "encoder_no_repeat_ngram_size": 0,
111
+ "eos_token_id": null,
112
+ "exponential_decay_length_penalty": null,
113
+ "finetuning_task": null,
114
+ "forced_bos_token_id": null,
115
+ "forced_eos_token_id": null,
116
+ "hidden_act": "quick_gelu",
117
+ "hidden_size": 1024,
118
+ "id2label": {
119
+ "0": "LABEL_0",
120
+ "1": "LABEL_1"
121
+ },
122
+ "image_size": 448,
123
+ "initializer_factor": 1.0,
124
+ "initializer_range": 0.02,
125
+ "intermediate_size": 4096,
126
+ "is_decoder": false,
127
+ "is_encoder_decoder": false,
128
+ "label2id": {
129
+ "LABEL_0": 0,
130
+ "LABEL_1": 1
131
+ },
132
+ "layer_norm_eps": 1e-06,
133
+ "length_penalty": 1.0,
134
+ "max_length": 20,
135
+ "min_length": 0,
136
+ "model_type": "mplug_owl_vision_model",
137
+ "no_repeat_ngram_size": 0,
138
+ "num_attention_heads": 16,
139
+ "num_beam_groups": 1,
140
+ "num_beams": 1,
141
+ "num_channels": 3,
142
+ "num_hidden_layers": 24,
143
+ "num_return_sequences": 1,
144
+ "output_attentions": false,
145
+ "output_hidden_states": false,
146
+ "output_scores": false,
147
+ "pad_token_id": null,
148
+ "patch_size": 14,
149
+ "prefix": null,
150
+ "problem_type": null,
151
+ "projection_dim": 768,
152
+ "pruned_heads": {},
153
+ "remove_invalid_values": false,
154
+ "repetition_penalty": 1.0,
155
+ "return_dict": true,
156
+ "return_dict_in_generate": false,
157
+ "sep_token_id": null,
158
+ "suppress_tokens": null,
159
+ "task_specific_params": null,
160
+ "temperature": 1.0,
161
+ "tf_legacy_loss": false,
162
+ "tie_encoder_decoder": false,
163
+ "tie_word_embeddings": true,
164
+ "tokenizer_class": null,
165
+ "top_k": 50,
166
+ "top_p": 1.0,
167
+ "torch_dtype": null,
168
+ "torchscript": false,
169
+ "transformers_version": "4.28.1",
170
+ "typical_p": 1.0,
171
+ "use_bfloat16": false,
172
+ "use_flash_attn": false
173
+ }
174
+ },
175
+ "vocab_size": 32000
176
+ }
configuration.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"framework":"Pytorch","task":"multimodal-dialogue"}
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "eos_token_id": 2,
4
+ "max_length": 4096,
5
+ "pad_token_id": 0,
6
+ "temperature": 0.9,
7
+ "top_p": 0.6,
8
+ "transformers_version": "4.31.0"
9
+ }
preprocessor_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "crop_size": 448,
3
+ "do_center_crop": true,
4
+ "do_normalize": true,
5
+ "do_resize": true,
6
+ "feature_extractor_type": "CLIPFeatureExtractor",
7
+ "image_mean": [
8
+ 0.48145466,
9
+ 0.4578275,
10
+ 0.40821073
11
+ ],
12
+ "image_std": [
13
+ 0.26862954,
14
+ 0.26130258,
15
+ 0.27577711
16
+ ],
17
+ "resample": 3,
18
+ "size": 448
19
+ }
20
+
pytorch_model-1-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3eda171a965e0f4f98941f0a45b2322d9f046f687205d0242fa6f45eefad04bc
3
+ size 471896315
pytorch_model-10-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:963430967add27b3cd84b55cc5d52a11fa1676450c790be813044fc59c43c3c5
3
+ size 471896315
pytorch_model-11-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:182c23f36504e850c910a373a01372c52f0151f461e531d24a2b6e33a5b0b05a
3
+ size 471896315
pytorch_model-12-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7abefa632c598f0eff7dc0b0bcba31debdd02850f65afc775e1b35b2b756004f
3
+ size 471896315
pytorch_model-13-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bbb00fc9a0375bd56e3817c25f950c1fbb9fd37da340e6a1e9ae19d856cdb34
3
+ size 471896315
pytorch_model-14-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1abf014e4e312136130bedfb0a77a623df76809dce37602ac072e655db353672
3
+ size 471896315
pytorch_model-15-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c843bda69c0324a1ea56b41b99098a358c4925e9211847a446a29d52d95196cf
3
+ size 471896315
pytorch_model-16-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32afd10b95a8f5871a6f9adc420772fb4288eec8cab80811b90e553aa08a1096
3
+ size 471896315
pytorch_model-17-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b2bcfbf9937a670b675f2a3cade55d4c23b1ee6342c2e330a4f107c882db0fe8
3
+ size 471896315
pytorch_model-18-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:220b47b918bca9b929fda0598d941101efba77c2be23d6f694b5369c12d7ad5e
3
+ size 471896315
pytorch_model-19-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7f4d830c36a82131725760e5a8e1535cbdba217aeb5ed1ff5dab801ece4bd6c
3
+ size 471896315
pytorch_model-2-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:07ebd22419c925cd4528d3182fa976694da76116d2896399f392134a4ffc4f60
3
+ size 471896315
pytorch_model-20-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf47d1d758293babcb2cfd6242c1ff3fa179bc1e27f93efb83a4f1ed285886f3
3
+ size 471896315
pytorch_model-21-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ecdc86f5b55942bd703518a82edc1b99b146e797cb911e6e1b9f392731a32a7
3
+ size 471896315
pytorch_model-22-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01b61398968246dad32f9a79729bb8b17ff7fbbc54df6df06d5a9c4a767f12e0
3
+ size 471896315
pytorch_model-23-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94014e3c353d9e3a52afd6a9009a63fcfc9445545ca36ee73c3dfd69c024a3f4
3
+ size 471896315
pytorch_model-24-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b73c4c64ab429f85e92f2bf03f28c7450e56823bcdf5f3596ba4fcf63c86bd96
3
+ size 471896315
pytorch_model-25-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b05ec589fe7d22e270b7a11fd18974610a65db598aee16633a9f16a156b26701
3
+ size 471896315
pytorch_model-26-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0053e94bfb2a49d60b9b6af5e3337df017cc284a8d3f3c3c724ac972dc42a92
3
+ size 471896315
pytorch_model-27-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33d49791d532ca66d65002aaa0aa0f7fc52708ee28503804d845ffeb77c7842b
3
+ size 471896315
pytorch_model-28-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:306cb650c62110e8b2e170659280c8d37569141611fa8ea69fe0c504bcac69c0
3
+ size 471896315
pytorch_model-29-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51abf23024b4b6d219cbeb907a314faf4de60700936d7a879943d537335c9131
3
+ size 471896315
pytorch_model-3-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1d4c4fbb39a54e3a8bf79643ab7e74b80231c3f52504421ff8bca44dc9abb36
3
+ size 471896315
pytorch_model-30-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a2bccfdf1a1771248aef72d1f99babedfe6ef37a04a12787d08ef4318940c11
3
+ size 471896315
pytorch_model-31-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4c897a84a550da60e6ab2315b9c62e7faaa0e762b8d29ecfb567d9d81c2b439
3
+ size 471896315
pytorch_model-32-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:18ad53e6f4961533b809a24e834496218c9b94d06c106b2799fcf4fe0d749f58
3
+ size 471896315
pytorch_model-33-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c5d2ac6af938bedd67cd5c53ef78a9d32401971e96d88ebf81168af56828258
3
+ size 1295321821
pytorch_model-4-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c4bfa35eb4a736c02e3d82d6e265b6eceebdaf012b81d02c64bcd2077e54bd4
3
+ size 471896315
pytorch_model-5-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f26905a37688e3a1676f9a0f45ed325042691a336cc46cc0fd3c5e4aadfe62cc
3
+ size 471896315
pytorch_model-6-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67857b8c240e5a4e6b6486f1ef86b8b86d48c014d39020a9364761bc286ddbbd
3
+ size 471896315
pytorch_model-7-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b74098fe2053bbce4507160f4549cfde263c42246b9b15f87a2535e4f3979fd4
3
+ size 471896315
pytorch_model-8-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:256484eef34f0326ca6af84922a66e1106a184da183d9a2b8ccf122f16009b8f
3
+ size 471896315
pytorch_model-9-of-33.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69c1424bc245b2882cb8d4f8cc33d8b13ee4619b08d0a5b004911f1a42432778
3
+ size 471896315
pytorch_model.bin.index.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"weight_map": {"model.layers.0.self_attn.q_proj.weight": "pytorch_model-1-of-33.bin", "model.layers.0.self_attn.k_proj.multiway.0.weight": "pytorch_model-1-of-33.bin", "model.layers.0.self_attn.v_proj.multiway.0.weight": "pytorch_model-1-of-33.bin", "model.layers.0.self_attn.k_proj.multiway.1.weight": "pytorch_model-1-of-33.bin", "model.layers.0.self_attn.v_proj.multiway.1.weight": "pytorch_model-1-of-33.bin", "model.layers.0.self_attn.o_proj.weight": "pytorch_model-1-of-33.bin", "model.layers.0.mlp.gate_proj.weight": "pytorch_model-1-of-33.bin", "model.layers.0.mlp.down_proj.weight": "pytorch_model-1-of-33.bin", "model.layers.0.mlp.up_proj.weight": "pytorch_model-1-of-33.bin", "model.layers.0.input_layernorm.multiway.0.weight": "pytorch_model-1-of-33.bin", "model.layers.0.post_attention_layernorm.multiway.0.weight": "pytorch_model-1-of-33.bin", "model.layers.0.input_layernorm.multiway.1.weight": "pytorch_model-1-of-33.bin", "model.layers.0.post_attention_layernorm.multiway.1.weight": "pytorch_model-1-of-33.bin", "model.layers.0.self_attn.rotary_emb.inv_freq": "pytorch_model-1-of-33.bin", "model.layers.1.self_attn.q_proj.weight": "pytorch_model-2-of-33.bin", "model.layers.1.self_attn.k_proj.multiway.0.weight": "pytorch_model-2-of-33.bin", "model.layers.1.self_attn.v_proj.multiway.0.weight": "pytorch_model-2-of-33.bin", "model.layers.1.self_attn.k_proj.multiway.1.weight": "pytorch_model-2-of-33.bin", "model.layers.1.self_attn.v_proj.multiway.1.weight": "pytorch_model-2-of-33.bin", "model.layers.1.self_attn.o_proj.weight": "pytorch_model-2-of-33.bin", "model.layers.1.mlp.gate_proj.weight": "pytorch_model-2-of-33.bin", "model.layers.1.mlp.down_proj.weight": "pytorch_model-2-of-33.bin", "model.layers.1.mlp.up_proj.weight": "pytorch_model-2-of-33.bin", "model.layers.1.input_layernorm.multiway.0.weight": "pytorch_model-2-of-33.bin", "model.layers.1.post_attention_layernorm.multiway.0.weight": "pytorch_model-2-of-33.bin", "model.layers.1.input_layernorm.multiway.1.weight": "pytorch_model-2-of-33.bin", "model.layers.1.post_attention_layernorm.multiway.1.weight": "pytorch_model-2-of-33.bin", "model.layers.1.self_attn.rotary_emb.inv_freq": "pytorch_model-2-of-33.bin", "model.layers.2.self_attn.q_proj.weight": "pytorch_model-3-of-33.bin", "model.layers.2.self_attn.k_proj.multiway.0.weight": "pytorch_model-3-of-33.bin", "model.layers.2.self_attn.v_proj.multiway.0.weight": "pytorch_model-3-of-33.bin", "model.layers.2.self_attn.k_proj.multiway.1.weight": "pytorch_model-3-of-33.bin", "model.layers.2.self_attn.v_proj.multiway.1.weight": "pytorch_model-3-of-33.bin", "model.layers.2.self_attn.o_proj.weight": "pytorch_model-3-of-33.bin", "model.layers.2.mlp.gate_proj.weight": "pytorch_model-3-of-33.bin", "model.layers.2.mlp.down_proj.weight": "pytorch_model-3-of-33.bin", "model.layers.2.mlp.up_proj.weight": "pytorch_model-3-of-33.bin", "model.layers.2.input_layernorm.multiway.0.weight": "pytorch_model-3-of-33.bin", "model.layers.2.post_attention_layernorm.multiway.0.weight": "pytorch_model-3-of-33.bin", "model.layers.2.input_layernorm.multiway.1.weight": "pytorch_model-3-of-33.bin", "model.layers.2.post_attention_layernorm.multiway.1.weight": "pytorch_model-3-of-33.bin", "model.layers.2.self_attn.rotary_emb.inv_freq": "pytorch_model-3-of-33.bin", "model.layers.3.self_attn.q_proj.weight": "pytorch_model-4-of-33.bin", "model.layers.3.self_attn.k_proj.multiway.0.weight": "pytorch_model-4-of-33.bin", "model.layers.3.self_attn.v_proj.multiway.0.weight": "pytorch_model-4-of-33.bin", "model.layers.3.self_attn.k_proj.multiway.1.weight": "pytorch_model-4-of-33.bin", "model.layers.3.self_attn.v_proj.multiway.1.weight": "pytorch_model-4-of-33.bin", "model.layers.3.self_attn.o_proj.weight": "pytorch_model-4-of-33.bin", "model.layers.3.mlp.gate_proj.weight": "pytorch_model-4-of-33.bin", "model.layers.3.mlp.down_proj.weight": "pytorch_model-4-of-33.bin", "model.layers.3.mlp.up_proj.weight": "pytorch_model-4-of-33.bin", "model.layers.3.input_layernorm.multiway.0.weight": "pytorch_model-4-of-33.bin", "model.layers.3.post_attention_layernorm.multiway.0.weight": "pytorch_model-4-of-33.bin", "model.layers.3.input_layernorm.multiway.1.weight": "pytorch_model-4-of-33.bin", "model.layers.3.post_attention_layernorm.multiway.1.weight": "pytorch_model-4-of-33.bin", "model.layers.3.self_attn.rotary_emb.inv_freq": "pytorch_model-4-of-33.bin", "model.layers.4.self_attn.q_proj.weight": "pytorch_model-5-of-33.bin", "model.layers.4.self_attn.k_proj.multiway.0.weight": "pytorch_model-5-of-33.bin", "model.layers.4.self_attn.v_proj.multiway.0.weight": "pytorch_model-5-of-33.bin", "model.layers.4.self_attn.k_proj.multiway.1.weight": "pytorch_model-5-of-33.bin", "model.layers.4.self_attn.v_proj.multiway.1.weight": "pytorch_model-5-of-33.bin", "model.layers.4.self_attn.o_proj.weight": "pytorch_model-5-of-33.bin", "model.layers.4.mlp.gate_proj.weight": "pytorch_model-5-of-33.bin", "model.layers.4.mlp.down_proj.weight": "pytorch_model-5-of-33.bin", "model.layers.4.mlp.up_proj.weight": "pytorch_model-5-of-33.bin", "model.layers.4.input_layernorm.multiway.0.weight": "pytorch_model-5-of-33.bin", "model.layers.4.post_attention_layernorm.multiway.0.weight": "pytorch_model-5-of-33.bin", "model.layers.4.input_layernorm.multiway.1.weight": "pytorch_model-5-of-33.bin", "model.layers.4.post_attention_layernorm.multiway.1.weight": "pytorch_model-5-of-33.bin", "model.layers.4.self_attn.rotary_emb.inv_freq": "pytorch_model-5-of-33.bin", "model.layers.5.self_attn.q_proj.weight": "pytorch_model-6-of-33.bin", "model.layers.5.self_attn.k_proj.multiway.0.weight": "pytorch_model-6-of-33.bin", "model.layers.5.self_attn.v_proj.multiway.0.weight": "pytorch_model-6-of-33.bin", "model.layers.5.self_attn.k_proj.multiway.1.weight": "pytorch_model-6-of-33.bin", "model.layers.5.self_attn.v_proj.multiway.1.weight": "pytorch_model-6-of-33.bin", "model.layers.5.self_attn.o_proj.weight": "pytorch_model-6-of-33.bin", "model.layers.5.mlp.gate_proj.weight": "pytorch_model-6-of-33.bin", "model.layers.5.mlp.down_proj.weight": "pytorch_model-6-of-33.bin", "model.layers.5.mlp.up_proj.weight": "pytorch_model-6-of-33.bin", "model.layers.5.input_layernorm.multiway.0.weight": "pytorch_model-6-of-33.bin", "model.layers.5.post_attention_layernorm.multiway.0.weight": "pytorch_model-6-of-33.bin", "model.layers.5.input_layernorm.multiway.1.weight": "pytorch_model-6-of-33.bin", "model.layers.5.post_attention_layernorm.multiway.1.weight": "pytorch_model-6-of-33.bin", "model.layers.5.self_attn.rotary_emb.inv_freq": "pytorch_model-6-of-33.bin", "model.layers.6.self_attn.q_proj.weight": "pytorch_model-7-of-33.bin", "model.layers.6.self_attn.k_proj.multiway.0.weight": "pytorch_model-7-of-33.bin", "model.layers.6.self_attn.v_proj.multiway.0.weight": "pytorch_model-7-of-33.bin", "model.layers.6.self_attn.k_proj.multiway.1.weight": "pytorch_model-7-of-33.bin", "model.layers.6.self_attn.v_proj.multiway.1.weight": "pytorch_model-7-of-33.bin", "model.layers.6.self_attn.o_proj.weight": "pytorch_model-7-of-33.bin", "model.layers.6.mlp.gate_proj.weight": "pytorch_model-7-of-33.bin", "model.layers.6.mlp.down_proj.weight": "pytorch_model-7-of-33.bin", "model.layers.6.mlp.up_proj.weight": "pytorch_model-7-of-33.bin", "model.layers.6.input_layernorm.multiway.0.weight": "pytorch_model-7-of-33.bin", "model.layers.6.post_attention_layernorm.multiway.0.weight": "pytorch_model-7-of-33.bin", "model.layers.6.input_layernorm.multiway.1.weight": "pytorch_model-7-of-33.bin", "model.layers.6.post_attention_layernorm.multiway.1.weight": "pytorch_model-7-of-33.bin", "model.layers.6.self_attn.rotary_emb.inv_freq": "pytorch_model-7-of-33.bin", "model.layers.7.self_attn.q_proj.weight": "pytorch_model-8-of-33.bin", "model.layers.7.self_attn.k_proj.multiway.0.weight": "pytorch_model-8-of-33.bin", "model.layers.7.self_attn.v_proj.multiway.0.weight": "pytorch_model-8-of-33.bin", "model.layers.7.self_attn.k_proj.multiway.1.weight": "pytorch_model-8-of-33.bin", "model.layers.7.self_attn.v_proj.multiway.1.weight": "pytorch_model-8-of-33.bin", "model.layers.7.self_attn.o_proj.weight": "pytorch_model-8-of-33.bin", "model.layers.7.mlp.gate_proj.weight": "pytorch_model-8-of-33.bin", "model.layers.7.mlp.down_proj.weight": "pytorch_model-8-of-33.bin", "model.layers.7.mlp.up_proj.weight": "pytorch_model-8-of-33.bin", "model.layers.7.input_layernorm.multiway.0.weight": "pytorch_model-8-of-33.bin", "model.layers.7.post_attention_layernorm.multiway.0.weight": "pytorch_model-8-of-33.bin", "model.layers.7.input_layernorm.multiway.1.weight": "pytorch_model-8-of-33.bin", "model.layers.7.post_attention_layernorm.multiway.1.weight": "pytorch_model-8-of-33.bin", "model.layers.7.self_attn.rotary_emb.inv_freq": "pytorch_model-8-of-33.bin", "model.layers.8.self_attn.q_proj.weight": "pytorch_model-9-of-33.bin", "model.layers.8.self_attn.k_proj.multiway.0.weight": "pytorch_model-9-of-33.bin", "model.layers.8.self_attn.v_proj.multiway.0.weight": "pytorch_model-9-of-33.bin", "model.layers.8.self_attn.k_proj.multiway.1.weight": "pytorch_model-9-of-33.bin", "model.layers.8.self_attn.v_proj.multiway.1.weight": "pytorch_model-9-of-33.bin", "model.layers.8.self_attn.o_proj.weight": "pytorch_model-9-of-33.bin", "model.layers.8.mlp.gate_proj.weight": "pytorch_model-9-of-33.bin", "model.layers.8.mlp.down_proj.weight": "pytorch_model-9-of-33.bin", "model.layers.8.mlp.up_proj.weight": "pytorch_model-9-of-33.bin", "model.layers.8.input_layernorm.multiway.0.weight": "pytorch_model-9-of-33.bin", "model.layers.8.post_attention_layernorm.multiway.0.weight": "pytorch_model-9-of-33.bin", "model.layers.8.input_layernorm.multiway.1.weight": "pytorch_model-9-of-33.bin", "model.layers.8.post_attention_layernorm.multiway.1.weight": "pytorch_model-9-of-33.bin", "model.layers.8.self_attn.rotary_emb.inv_freq": "pytorch_model-9-of-33.bin", "model.layers.9.self_attn.q_proj.weight": "pytorch_model-10-of-33.bin", "model.layers.9.self_attn.k_proj.multiway.0.weight": "pytorch_model-10-of-33.bin", "model.layers.9.self_attn.v_proj.multiway.0.weight": "pytorch_model-10-of-33.bin", "model.layers.9.self_attn.k_proj.multiway.1.weight": "pytorch_model-10-of-33.bin", "model.layers.9.self_attn.v_proj.multiway.1.weight": "pytorch_model-10-of-33.bin", "model.layers.9.self_attn.o_proj.weight": "pytorch_model-10-of-33.bin", "model.layers.9.mlp.gate_proj.weight": "pytorch_model-10-of-33.bin", "model.layers.9.mlp.down_proj.weight": "pytorch_model-10-of-33.bin", "model.layers.9.mlp.up_proj.weight": "pytorch_model-10-of-33.bin", "model.layers.9.input_layernorm.multiway.0.weight": "pytorch_model-10-of-33.bin", "model.layers.9.post_attention_layernorm.multiway.0.weight": "pytorch_model-10-of-33.bin", "model.layers.9.input_layernorm.multiway.1.weight": "pytorch_model-10-of-33.bin", "model.layers.9.post_attention_layernorm.multiway.1.weight": "pytorch_model-10-of-33.bin", "model.layers.9.self_attn.rotary_emb.inv_freq": "pytorch_model-10-of-33.bin", "model.layers.10.self_attn.q_proj.weight": "pytorch_model-11-of-33.bin", "model.layers.10.self_attn.k_proj.multiway.0.weight": "pytorch_model-11-of-33.bin", "model.layers.10.self_attn.v_proj.multiway.0.weight": "pytorch_model-11-of-33.bin", "model.layers.10.self_attn.k_proj.multiway.1.weight": "pytorch_model-11-of-33.bin", "model.layers.10.self_attn.v_proj.multiway.1.weight": "pytorch_model-11-of-33.bin", "model.layers.10.self_attn.o_proj.weight": "pytorch_model-11-of-33.bin", "model.layers.10.mlp.gate_proj.weight": "pytorch_model-11-of-33.bin", "model.layers.10.mlp.down_proj.weight": "pytorch_model-11-of-33.bin", "model.layers.10.mlp.up_proj.weight": "pytorch_model-11-of-33.bin", "model.layers.10.input_layernorm.multiway.0.weight": "pytorch_model-11-of-33.bin", "model.layers.10.post_attention_layernorm.multiway.0.weight": "pytorch_model-11-of-33.bin", "model.layers.10.input_layernorm.multiway.1.weight": "pytorch_model-11-of-33.bin", "model.layers.10.post_attention_layernorm.multiway.1.weight": "pytorch_model-11-of-33.bin", "model.layers.10.self_attn.rotary_emb.inv_freq": "pytorch_model-11-of-33.bin", "model.layers.11.self_attn.q_proj.weight": "pytorch_model-12-of-33.bin", "model.layers.11.self_attn.k_proj.multiway.0.weight": "pytorch_model-12-of-33.bin", "model.layers.11.self_attn.v_proj.multiway.0.weight": "pytorch_model-12-of-33.bin", "model.layers.11.self_attn.k_proj.multiway.1.weight": "pytorch_model-12-of-33.bin", "model.layers.11.self_attn.v_proj.multiway.1.weight": "pytorch_model-12-of-33.bin", "model.layers.11.self_attn.o_proj.weight": "pytorch_model-12-of-33.bin", "model.layers.11.mlp.gate_proj.weight": "pytorch_model-12-of-33.bin", "model.layers.11.mlp.down_proj.weight": "pytorch_model-12-of-33.bin", "model.layers.11.mlp.up_proj.weight": "pytorch_model-12-of-33.bin", "model.layers.11.input_layernorm.multiway.0.weight": "pytorch_model-12-of-33.bin", "model.layers.11.post_attention_layernorm.multiway.0.weight": "pytorch_model-12-of-33.bin", "model.layers.11.input_layernorm.multiway.1.weight": "pytorch_model-12-of-33.bin", "model.layers.11.post_attention_layernorm.multiway.1.weight": "pytorch_model-12-of-33.bin", "model.layers.11.self_attn.rotary_emb.inv_freq": "pytorch_model-12-of-33.bin", "model.layers.12.self_attn.q_proj.weight": "pytorch_model-13-of-33.bin", "model.layers.12.self_attn.k_proj.multiway.0.weight": "pytorch_model-13-of-33.bin", "model.layers.12.self_attn.v_proj.multiway.0.weight": "pytorch_model-13-of-33.bin", "model.layers.12.self_attn.k_proj.multiway.1.weight": "pytorch_model-13-of-33.bin", "model.layers.12.self_attn.v_proj.multiway.1.weight": "pytorch_model-13-of-33.bin", "model.layers.12.self_attn.o_proj.weight": "pytorch_model-13-of-33.bin", "model.layers.12.mlp.gate_proj.weight": "pytorch_model-13-of-33.bin", "model.layers.12.mlp.down_proj.weight": "pytorch_model-13-of-33.bin", "model.layers.12.mlp.up_proj.weight": "pytorch_model-13-of-33.bin", "model.layers.12.input_layernorm.multiway.0.weight": "pytorch_model-13-of-33.bin", "model.layers.12.post_attention_layernorm.multiway.0.weight": "pytorch_model-13-of-33.bin", "model.layers.12.input_layernorm.multiway.1.weight": "pytorch_model-13-of-33.bin", "model.layers.12.post_attention_layernorm.multiway.1.weight": "pytorch_model-13-of-33.bin", "model.layers.12.self_attn.rotary_emb.inv_freq": "pytorch_model-13-of-33.bin", "model.layers.13.self_attn.q_proj.weight": "pytorch_model-14-of-33.bin", "model.layers.13.self_attn.k_proj.multiway.0.weight": "pytorch_model-14-of-33.bin", "model.layers.13.self_attn.v_proj.multiway.0.weight": "pytorch_model-14-of-33.bin", "model.layers.13.self_attn.k_proj.multiway.1.weight": "pytorch_model-14-of-33.bin", "model.layers.13.self_attn.v_proj.multiway.1.weight": "pytorch_model-14-of-33.bin", "model.layers.13.self_attn.o_proj.weight": "pytorch_model-14-of-33.bin", "model.layers.13.mlp.gate_proj.weight": "pytorch_model-14-of-33.bin", "model.layers.13.mlp.down_proj.weight": "pytorch_model-14-of-33.bin", "model.layers.13.mlp.up_proj.weight": "pytorch_model-14-of-33.bin", "model.layers.13.input_layernorm.multiway.0.weight": "pytorch_model-14-of-33.bin", "model.layers.13.post_attention_layernorm.multiway.0.weight": "pytorch_model-14-of-33.bin", "model.layers.13.input_layernorm.multiway.1.weight": "pytorch_model-14-of-33.bin", "model.layers.13.post_attention_layernorm.multiway.1.weight": "pytorch_model-14-of-33.bin", "model.layers.13.self_attn.rotary_emb.inv_freq": "pytorch_model-14-of-33.bin", "model.layers.14.self_attn.q_proj.weight": "pytorch_model-15-of-33.bin", "model.layers.14.self_attn.k_proj.multiway.0.weight": "pytorch_model-15-of-33.bin", "model.layers.14.self_attn.v_proj.multiway.0.weight": "pytorch_model-15-of-33.bin", "model.layers.14.self_attn.k_proj.multiway.1.weight": "pytorch_model-15-of-33.bin", "model.layers.14.self_attn.v_proj.multiway.1.weight": "pytorch_model-15-of-33.bin", "model.layers.14.self_attn.o_proj.weight": "pytorch_model-15-of-33.bin", "model.layers.14.mlp.gate_proj.weight": "pytorch_model-15-of-33.bin", "model.layers.14.mlp.down_proj.weight": "pytorch_model-15-of-33.bin", "model.layers.14.mlp.up_proj.weight": "pytorch_model-15-of-33.bin", "model.layers.14.input_layernorm.multiway.0.weight": "pytorch_model-15-of-33.bin", "model.layers.14.post_attention_layernorm.multiway.0.weight": "pytorch_model-15-of-33.bin", "model.layers.14.input_layernorm.multiway.1.weight": "pytorch_model-15-of-33.bin", "model.layers.14.post_attention_layernorm.multiway.1.weight": "pytorch_model-15-of-33.bin", "model.layers.14.self_attn.rotary_emb.inv_freq": "pytorch_model-15-of-33.bin", "model.layers.15.self_attn.q_proj.weight": "pytorch_model-16-of-33.bin", "model.layers.15.self_attn.k_proj.multiway.0.weight": "pytorch_model-16-of-33.bin", "model.layers.15.self_attn.v_proj.multiway.0.weight": "pytorch_model-16-of-33.bin", "model.layers.15.self_attn.k_proj.multiway.1.weight": "pytorch_model-16-of-33.bin", "model.layers.15.self_attn.v_proj.multiway.1.weight": "pytorch_model-16-of-33.bin", "model.layers.15.self_attn.o_proj.weight": "pytorch_model-16-of-33.bin", "model.layers.15.mlp.gate_proj.weight": "pytorch_model-16-of-33.bin", "model.layers.15.mlp.down_proj.weight": "pytorch_model-16-of-33.bin", "model.layers.15.mlp.up_proj.weight": "pytorch_model-16-of-33.bin", "model.layers.15.input_layernorm.multiway.0.weight": "pytorch_model-16-of-33.bin", "model.layers.15.post_attention_layernorm.multiway.0.weight": "pytorch_model-16-of-33.bin", "model.layers.15.input_layernorm.multiway.1.weight": "pytorch_model-16-of-33.bin", "model.layers.15.post_attention_layernorm.multiway.1.weight": "pytorch_model-16-of-33.bin", "model.layers.15.self_attn.rotary_emb.inv_freq": "pytorch_model-16-of-33.bin", "model.layers.16.self_attn.q_proj.weight": "pytorch_model-17-of-33.bin", "model.layers.16.self_attn.k_proj.multiway.0.weight": "pytorch_model-17-of-33.bin", "model.layers.16.self_attn.v_proj.multiway.0.weight": "pytorch_model-17-of-33.bin", "model.layers.16.self_attn.k_proj.multiway.1.weight": "pytorch_model-17-of-33.bin", "model.layers.16.self_attn.v_proj.multiway.1.weight": "pytorch_model-17-of-33.bin", "model.layers.16.self_attn.o_proj.weight": "pytorch_model-17-of-33.bin", "model.layers.16.mlp.gate_proj.weight": "pytorch_model-17-of-33.bin", "model.layers.16.mlp.down_proj.weight": "pytorch_model-17-of-33.bin", "model.layers.16.mlp.up_proj.weight": "pytorch_model-17-of-33.bin", "model.layers.16.input_layernorm.multiway.0.weight": "pytorch_model-17-of-33.bin", "model.layers.16.post_attention_layernorm.multiway.0.weight": "pytorch_model-17-of-33.bin", "model.layers.16.input_layernorm.multiway.1.weight": "pytorch_model-17-of-33.bin", "model.layers.16.post_attention_layernorm.multiway.1.weight": "pytorch_model-17-of-33.bin", "model.layers.16.self_attn.rotary_emb.inv_freq": "pytorch_model-17-of-33.bin", "model.layers.17.self_attn.q_proj.weight": "pytorch_model-18-of-33.bin", "model.layers.17.self_attn.k_proj.multiway.0.weight": "pytorch_model-18-of-33.bin", "model.layers.17.self_attn.v_proj.multiway.0.weight": "pytorch_model-18-of-33.bin", "model.layers.17.self_attn.k_proj.multiway.1.weight": "pytorch_model-18-of-33.bin", "model.layers.17.self_attn.v_proj.multiway.1.weight": "pytorch_model-18-of-33.bin", "model.layers.17.self_attn.o_proj.weight": "pytorch_model-18-of-33.bin", "model.layers.17.mlp.gate_proj.weight": "pytorch_model-18-of-33.bin", "model.layers.17.mlp.down_proj.weight": "pytorch_model-18-of-33.bin", "model.layers.17.mlp.up_proj.weight": "pytorch_model-18-of-33.bin", "model.layers.17.input_layernorm.multiway.0.weight": "pytorch_model-18-of-33.bin", "model.layers.17.post_attention_layernorm.multiway.0.weight": "pytorch_model-18-of-33.bin", "model.layers.17.input_layernorm.multiway.1.weight": "pytorch_model-18-of-33.bin", "model.layers.17.post_attention_layernorm.multiway.1.weight": "pytorch_model-18-of-33.bin", "model.layers.17.self_attn.rotary_emb.inv_freq": "pytorch_model-18-of-33.bin", "model.layers.18.self_attn.q_proj.weight": "pytorch_model-19-of-33.bin", "model.layers.18.self_attn.k_proj.multiway.0.weight": "pytorch_model-19-of-33.bin", "model.layers.18.self_attn.v_proj.multiway.0.weight": "pytorch_model-19-of-33.bin", "model.layers.18.self_attn.k_proj.multiway.1.weight": "pytorch_model-19-of-33.bin", "model.layers.18.self_attn.v_proj.multiway.1.weight": "pytorch_model-19-of-33.bin", "model.layers.18.self_attn.o_proj.weight": "pytorch_model-19-of-33.bin", "model.layers.18.mlp.gate_proj.weight": "pytorch_model-19-of-33.bin", "model.layers.18.mlp.down_proj.weight": "pytorch_model-19-of-33.bin", "model.layers.18.mlp.up_proj.weight": "pytorch_model-19-of-33.bin", "model.layers.18.input_layernorm.multiway.0.weight": "pytorch_model-19-of-33.bin", "model.layers.18.post_attention_layernorm.multiway.0.weight": "pytorch_model-19-of-33.bin", "model.layers.18.input_layernorm.multiway.1.weight": "pytorch_model-19-of-33.bin", "model.layers.18.post_attention_layernorm.multiway.1.weight": "pytorch_model-19-of-33.bin", "model.layers.18.self_attn.rotary_emb.inv_freq": "pytorch_model-19-of-33.bin", "model.layers.19.self_attn.q_proj.weight": "pytorch_model-20-of-33.bin", "model.layers.19.self_attn.k_proj.multiway.0.weight": "pytorch_model-20-of-33.bin", "model.layers.19.self_attn.v_proj.multiway.0.weight": "pytorch_model-20-of-33.bin", "model.layers.19.self_attn.k_proj.multiway.1.weight": "pytorch_model-20-of-33.bin", "model.layers.19.self_attn.v_proj.multiway.1.weight": "pytorch_model-20-of-33.bin", "model.layers.19.self_attn.o_proj.weight": "pytorch_model-20-of-33.bin", "model.layers.19.mlp.gate_proj.weight": "pytorch_model-20-of-33.bin", "model.layers.19.mlp.down_proj.weight": "pytorch_model-20-of-33.bin", "model.layers.19.mlp.up_proj.weight": "pytorch_model-20-of-33.bin", "model.layers.19.input_layernorm.multiway.0.weight": "pytorch_model-20-of-33.bin", "model.layers.19.post_attention_layernorm.multiway.0.weight": "pytorch_model-20-of-33.bin", "model.layers.19.input_layernorm.multiway.1.weight": "pytorch_model-20-of-33.bin", "model.layers.19.post_attention_layernorm.multiway.1.weight": "pytorch_model-20-of-33.bin", "model.layers.19.self_attn.rotary_emb.inv_freq": "pytorch_model-20-of-33.bin", "model.layers.20.self_attn.q_proj.weight": "pytorch_model-21-of-33.bin", "model.layers.20.self_attn.k_proj.multiway.0.weight": "pytorch_model-21-of-33.bin", "model.layers.20.self_attn.v_proj.multiway.0.weight": "pytorch_model-21-of-33.bin", "model.layers.20.self_attn.k_proj.multiway.1.weight": "pytorch_model-21-of-33.bin", "model.layers.20.self_attn.v_proj.multiway.1.weight": "pytorch_model-21-of-33.bin", "model.layers.20.self_attn.o_proj.weight": "pytorch_model-21-of-33.bin", "model.layers.20.mlp.gate_proj.weight": "pytorch_model-21-of-33.bin", "model.layers.20.mlp.down_proj.weight": "pytorch_model-21-of-33.bin", "model.layers.20.mlp.up_proj.weight": "pytorch_model-21-of-33.bin", "model.layers.20.input_layernorm.multiway.0.weight": "pytorch_model-21-of-33.bin", "model.layers.20.post_attention_layernorm.multiway.0.weight": "pytorch_model-21-of-33.bin", "model.layers.20.input_layernorm.multiway.1.weight": "pytorch_model-21-of-33.bin", "model.layers.20.post_attention_layernorm.multiway.1.weight": "pytorch_model-21-of-33.bin", "model.layers.20.self_attn.rotary_emb.inv_freq": "pytorch_model-21-of-33.bin", "model.layers.21.self_attn.q_proj.weight": "pytorch_model-22-of-33.bin", "model.layers.21.self_attn.k_proj.multiway.0.weight": "pytorch_model-22-of-33.bin", "model.layers.21.self_attn.v_proj.multiway.0.weight": "pytorch_model-22-of-33.bin", "model.layers.21.self_attn.k_proj.multiway.1.weight": "pytorch_model-22-of-33.bin", "model.layers.21.self_attn.v_proj.multiway.1.weight": "pytorch_model-22-of-33.bin", "model.layers.21.self_attn.o_proj.weight": "pytorch_model-22-of-33.bin", "model.layers.21.mlp.gate_proj.weight": "pytorch_model-22-of-33.bin", "model.layers.21.mlp.down_proj.weight": "pytorch_model-22-of-33.bin", "model.layers.21.mlp.up_proj.weight": "pytorch_model-22-of-33.bin", "model.layers.21.input_layernorm.multiway.0.weight": "pytorch_model-22-of-33.bin", "model.layers.21.post_attention_layernorm.multiway.0.weight": "pytorch_model-22-of-33.bin", "model.layers.21.input_layernorm.multiway.1.weight": "pytorch_model-22-of-33.bin", "model.layers.21.post_attention_layernorm.multiway.1.weight": "pytorch_model-22-of-33.bin", "model.layers.21.self_attn.rotary_emb.inv_freq": "pytorch_model-22-of-33.bin", "model.layers.22.self_attn.q_proj.weight": "pytorch_model-23-of-33.bin", "model.layers.22.self_attn.k_proj.multiway.0.weight": "pytorch_model-23-of-33.bin", "model.layers.22.self_attn.v_proj.multiway.0.weight": "pytorch_model-23-of-33.bin", "model.layers.22.self_attn.k_proj.multiway.1.weight": "pytorch_model-23-of-33.bin", "model.layers.22.self_attn.v_proj.multiway.1.weight": "pytorch_model-23-of-33.bin", "model.layers.22.self_attn.o_proj.weight": "pytorch_model-23-of-33.bin", "model.layers.22.mlp.gate_proj.weight": "pytorch_model-23-of-33.bin", "model.layers.22.mlp.down_proj.weight": "pytorch_model-23-of-33.bin", "model.layers.22.mlp.up_proj.weight": "pytorch_model-23-of-33.bin", "model.layers.22.input_layernorm.multiway.0.weight": "pytorch_model-23-of-33.bin", "model.layers.22.post_attention_layernorm.multiway.0.weight": "pytorch_model-23-of-33.bin", "model.layers.22.input_layernorm.multiway.1.weight": "pytorch_model-23-of-33.bin", "model.layers.22.post_attention_layernorm.multiway.1.weight": "pytorch_model-23-of-33.bin", "model.layers.22.self_attn.rotary_emb.inv_freq": "pytorch_model-23-of-33.bin", "model.layers.23.self_attn.q_proj.weight": "pytorch_model-24-of-33.bin", "model.layers.23.self_attn.k_proj.multiway.0.weight": "pytorch_model-24-of-33.bin", "model.layers.23.self_attn.v_proj.multiway.0.weight": "pytorch_model-24-of-33.bin", "model.layers.23.self_attn.k_proj.multiway.1.weight": "pytorch_model-24-of-33.bin", "model.layers.23.self_attn.v_proj.multiway.1.weight": "pytorch_model-24-of-33.bin", "model.layers.23.self_attn.o_proj.weight": "pytorch_model-24-of-33.bin", "model.layers.23.mlp.gate_proj.weight": "pytorch_model-24-of-33.bin", "model.layers.23.mlp.down_proj.weight": "pytorch_model-24-of-33.bin", "model.layers.23.mlp.up_proj.weight": "pytorch_model-24-of-33.bin", "model.layers.23.input_layernorm.multiway.0.weight": "pytorch_model-24-of-33.bin", "model.layers.23.post_attention_layernorm.multiway.0.weight": "pytorch_model-24-of-33.bin", "model.layers.23.input_layernorm.multiway.1.weight": "pytorch_model-24-of-33.bin", "model.layers.23.post_attention_layernorm.multiway.1.weight": "pytorch_model-24-of-33.bin", "model.layers.23.self_attn.rotary_emb.inv_freq": "pytorch_model-24-of-33.bin", "model.layers.24.self_attn.q_proj.weight": "pytorch_model-25-of-33.bin", "model.layers.24.self_attn.k_proj.multiway.0.weight": "pytorch_model-25-of-33.bin", "model.layers.24.self_attn.v_proj.multiway.0.weight": "pytorch_model-25-of-33.bin", "model.layers.24.self_attn.k_proj.multiway.1.weight": "pytorch_model-25-of-33.bin", "model.layers.24.self_attn.v_proj.multiway.1.weight": "pytorch_model-25-of-33.bin", "model.layers.24.self_attn.o_proj.weight": "pytorch_model-25-of-33.bin", "model.layers.24.mlp.gate_proj.weight": "pytorch_model-25-of-33.bin", "model.layers.24.mlp.down_proj.weight": "pytorch_model-25-of-33.bin", "model.layers.24.mlp.up_proj.weight": "pytorch_model-25-of-33.bin", "model.layers.24.input_layernorm.multiway.0.weight": "pytorch_model-25-of-33.bin", "model.layers.24.post_attention_layernorm.multiway.0.weight": "pytorch_model-25-of-33.bin", "model.layers.24.input_layernorm.multiway.1.weight": "pytorch_model-25-of-33.bin", "model.layers.24.post_attention_layernorm.multiway.1.weight": "pytorch_model-25-of-33.bin", "model.layers.24.self_attn.rotary_emb.inv_freq": "pytorch_model-25-of-33.bin", "model.layers.25.self_attn.q_proj.weight": "pytorch_model-26-of-33.bin", "model.layers.25.self_attn.k_proj.multiway.0.weight": "pytorch_model-26-of-33.bin", "model.layers.25.self_attn.v_proj.multiway.0.weight": "pytorch_model-26-of-33.bin", "model.layers.25.self_attn.k_proj.multiway.1.weight": "pytorch_model-26-of-33.bin", "model.layers.25.self_attn.v_proj.multiway.1.weight": "pytorch_model-26-of-33.bin", "model.layers.25.self_attn.o_proj.weight": "pytorch_model-26-of-33.bin", "model.layers.25.mlp.gate_proj.weight": "pytorch_model-26-of-33.bin", "model.layers.25.mlp.down_proj.weight": "pytorch_model-26-of-33.bin", "model.layers.25.mlp.up_proj.weight": "pytorch_model-26-of-33.bin", "model.layers.25.input_layernorm.multiway.0.weight": "pytorch_model-26-of-33.bin", "model.layers.25.post_attention_layernorm.multiway.0.weight": "pytorch_model-26-of-33.bin", "model.layers.25.input_layernorm.multiway.1.weight": "pytorch_model-26-of-33.bin", "model.layers.25.post_attention_layernorm.multiway.1.weight": "pytorch_model-26-of-33.bin", "model.layers.25.self_attn.rotary_emb.inv_freq": "pytorch_model-26-of-33.bin", "model.layers.26.self_attn.q_proj.weight": "pytorch_model-27-of-33.bin", "model.layers.26.self_attn.k_proj.multiway.0.weight": "pytorch_model-27-of-33.bin", "model.layers.26.self_attn.v_proj.multiway.0.weight": "pytorch_model-27-of-33.bin", "model.layers.26.self_attn.k_proj.multiway.1.weight": "pytorch_model-27-of-33.bin", "model.layers.26.self_attn.v_proj.multiway.1.weight": "pytorch_model-27-of-33.bin", "model.layers.26.self_attn.o_proj.weight": "pytorch_model-27-of-33.bin", "model.layers.26.mlp.gate_proj.weight": "pytorch_model-27-of-33.bin", "model.layers.26.mlp.down_proj.weight": "pytorch_model-27-of-33.bin", "model.layers.26.mlp.up_proj.weight": "pytorch_model-27-of-33.bin", "model.layers.26.input_layernorm.multiway.0.weight": "pytorch_model-27-of-33.bin", "model.layers.26.post_attention_layernorm.multiway.0.weight": "pytorch_model-27-of-33.bin", "model.layers.26.input_layernorm.multiway.1.weight": "pytorch_model-27-of-33.bin", "model.layers.26.post_attention_layernorm.multiway.1.weight": "pytorch_model-27-of-33.bin", "model.layers.26.self_attn.rotary_emb.inv_freq": "pytorch_model-27-of-33.bin", "model.layers.27.self_attn.q_proj.weight": "pytorch_model-28-of-33.bin", "model.layers.27.self_attn.k_proj.multiway.0.weight": "pytorch_model-28-of-33.bin", "model.layers.27.self_attn.v_proj.multiway.0.weight": "pytorch_model-28-of-33.bin", "model.layers.27.self_attn.k_proj.multiway.1.weight": "pytorch_model-28-of-33.bin", "model.layers.27.self_attn.v_proj.multiway.1.weight": "pytorch_model-28-of-33.bin", "model.layers.27.self_attn.o_proj.weight": "pytorch_model-28-of-33.bin", "model.layers.27.mlp.gate_proj.weight": "pytorch_model-28-of-33.bin", "model.layers.27.mlp.down_proj.weight": "pytorch_model-28-of-33.bin", "model.layers.27.mlp.up_proj.weight": "pytorch_model-28-of-33.bin", "model.layers.27.input_layernorm.multiway.0.weight": "pytorch_model-28-of-33.bin", "model.layers.27.post_attention_layernorm.multiway.0.weight": "pytorch_model-28-of-33.bin", "model.layers.27.input_layernorm.multiway.1.weight": "pytorch_model-28-of-33.bin", "model.layers.27.post_attention_layernorm.multiway.1.weight": "pytorch_model-28-of-33.bin", "model.layers.27.self_attn.rotary_emb.inv_freq": "pytorch_model-28-of-33.bin", "model.layers.28.self_attn.q_proj.weight": "pytorch_model-29-of-33.bin", "model.layers.28.self_attn.k_proj.multiway.0.weight": "pytorch_model-29-of-33.bin", "model.layers.28.self_attn.v_proj.multiway.0.weight": "pytorch_model-29-of-33.bin", "model.layers.28.self_attn.k_proj.multiway.1.weight": "pytorch_model-29-of-33.bin", "model.layers.28.self_attn.v_proj.multiway.1.weight": "pytorch_model-29-of-33.bin", "model.layers.28.self_attn.o_proj.weight": "pytorch_model-29-of-33.bin", "model.layers.28.mlp.gate_proj.weight": "pytorch_model-29-of-33.bin", "model.layers.28.mlp.down_proj.weight": "pytorch_model-29-of-33.bin", "model.layers.28.mlp.up_proj.weight": "pytorch_model-29-of-33.bin", "model.layers.28.input_layernorm.multiway.0.weight": "pytorch_model-29-of-33.bin", "model.layers.28.post_attention_layernorm.multiway.0.weight": "pytorch_model-29-of-33.bin", "model.layers.28.input_layernorm.multiway.1.weight": "pytorch_model-29-of-33.bin", "model.layers.28.post_attention_layernorm.multiway.1.weight": "pytorch_model-29-of-33.bin", "model.layers.28.self_attn.rotary_emb.inv_freq": "pytorch_model-29-of-33.bin", "model.layers.29.self_attn.q_proj.weight": "pytorch_model-30-of-33.bin", "model.layers.29.self_attn.k_proj.multiway.0.weight": "pytorch_model-30-of-33.bin", "model.layers.29.self_attn.v_proj.multiway.0.weight": "pytorch_model-30-of-33.bin", "model.layers.29.self_attn.k_proj.multiway.1.weight": "pytorch_model-30-of-33.bin", "model.layers.29.self_attn.v_proj.multiway.1.weight": "pytorch_model-30-of-33.bin", "model.layers.29.self_attn.o_proj.weight": "pytorch_model-30-of-33.bin", "model.layers.29.mlp.gate_proj.weight": "pytorch_model-30-of-33.bin", "model.layers.29.mlp.down_proj.weight": "pytorch_model-30-of-33.bin", "model.layers.29.mlp.up_proj.weight": "pytorch_model-30-of-33.bin", "model.layers.29.input_layernorm.multiway.0.weight": "pytorch_model-30-of-33.bin", "model.layers.29.post_attention_layernorm.multiway.0.weight": "pytorch_model-30-of-33.bin", "model.layers.29.input_layernorm.multiway.1.weight": "pytorch_model-30-of-33.bin", "model.layers.29.post_attention_layernorm.multiway.1.weight": "pytorch_model-30-of-33.bin", "model.layers.29.self_attn.rotary_emb.inv_freq": "pytorch_model-30-of-33.bin", "model.layers.30.self_attn.q_proj.weight": "pytorch_model-31-of-33.bin", "model.layers.30.self_attn.k_proj.multiway.0.weight": "pytorch_model-31-of-33.bin", "model.layers.30.self_attn.v_proj.multiway.0.weight": "pytorch_model-31-of-33.bin", "model.layers.30.self_attn.k_proj.multiway.1.weight": "pytorch_model-31-of-33.bin", "model.layers.30.self_attn.v_proj.multiway.1.weight": "pytorch_model-31-of-33.bin", "model.layers.30.self_attn.o_proj.weight": "pytorch_model-31-of-33.bin", "model.layers.30.mlp.gate_proj.weight": "pytorch_model-31-of-33.bin", "model.layers.30.mlp.down_proj.weight": "pytorch_model-31-of-33.bin", "model.layers.30.mlp.up_proj.weight": "pytorch_model-31-of-33.bin", "model.layers.30.input_layernorm.multiway.0.weight": "pytorch_model-31-of-33.bin", "model.layers.30.post_attention_layernorm.multiway.0.weight": "pytorch_model-31-of-33.bin", "model.layers.30.input_layernorm.multiway.1.weight": "pytorch_model-31-of-33.bin", "model.layers.30.post_attention_layernorm.multiway.1.weight": "pytorch_model-31-of-33.bin", "model.layers.30.self_attn.rotary_emb.inv_freq": "pytorch_model-31-of-33.bin", "model.layers.31.self_attn.q_proj.weight": "pytorch_model-32-of-33.bin", "model.layers.31.self_attn.k_proj.multiway.0.weight": "pytorch_model-32-of-33.bin", "model.layers.31.self_attn.v_proj.multiway.0.weight": "pytorch_model-32-of-33.bin", "model.layers.31.self_attn.k_proj.multiway.1.weight": "pytorch_model-32-of-33.bin", "model.layers.31.self_attn.v_proj.multiway.1.weight": "pytorch_model-32-of-33.bin", "model.layers.31.self_attn.o_proj.weight": "pytorch_model-32-of-33.bin", "model.layers.31.mlp.gate_proj.weight": "pytorch_model-32-of-33.bin", "model.layers.31.mlp.down_proj.weight": "pytorch_model-32-of-33.bin", "model.layers.31.mlp.up_proj.weight": "pytorch_model-32-of-33.bin", "model.layers.31.input_layernorm.multiway.0.weight": "pytorch_model-32-of-33.bin", "model.layers.31.post_attention_layernorm.multiway.0.weight": "pytorch_model-32-of-33.bin", "model.layers.31.input_layernorm.multiway.1.weight": "pytorch_model-32-of-33.bin", "model.layers.31.post_attention_layernorm.multiway.1.weight": "pytorch_model-32-of-33.bin", "model.layers.31.self_attn.rotary_emb.inv_freq": "pytorch_model-32-of-33.bin", "model.embed_tokens.weight": "pytorch_model-33-of-33.bin", "model.norm.weight": "pytorch_model-33-of-33.bin", "lm_head.weight": "pytorch_model-33-of-33.bin", "model.vision_model.embeddings.cls_token": "pytorch_model-33-of-33.bin", "model.vision_model.embeddings.patch_embed.weight": "pytorch_model-33-of-33.bin", "model.vision_model.embeddings.position_embedding": "pytorch_model-33-of-33.bin", "model.vision_model.embeddings.pre_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.embeddings.pre_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.post_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.post_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.0.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.1.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.2.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.3.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.4.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.5.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.6.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.7.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.8.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.9.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.10.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.11.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.12.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.13.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.14.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.15.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.16.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.17.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.18.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.19.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.20.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.21.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.22.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.input_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.input_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.mlp.fc1.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.mlp.fc1.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.mlp.fc2.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.mlp.fc2.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.post_attention_layernorm.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.post_attention_layernorm.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.self_attn.dense.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.self_attn.dense.weight": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.self_attn.query_key_value.bias": "pytorch_model-33-of-33.bin", "model.vision_model.encoder.layers.23.self_attn.query_key_value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.query_embeds": "pytorch_model-33-of-33.bin", "model.visual_abstractor.visual_fc.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.visual_fc.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.vit_eos": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.attention.key.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.attention.key.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.attention.query.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.attention.query.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.attention.value.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.attention.value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.norm1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.norm1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.normk.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.normk.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.ffn_ln.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.ffn_ln.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.w1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.w1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.w2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.w2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.w3.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.mlp.w3.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.norm2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.norm2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.out_proj.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.0.crossattention.output.out_proj.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.attention.key.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.attention.key.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.attention.query.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.attention.query.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.attention.value.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.attention.value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.norm1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.norm1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.normk.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.normk.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.ffn_ln.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.ffn_ln.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.w1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.w1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.w2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.w2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.w3.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.mlp.w3.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.norm2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.norm2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.out_proj.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.1.crossattention.output.out_proj.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.attention.key.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.attention.key.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.attention.query.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.attention.query.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.attention.value.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.attention.value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.norm1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.norm1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.normk.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.normk.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.ffn_ln.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.ffn_ln.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.w1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.w1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.w2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.w2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.w3.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.mlp.w3.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.norm2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.norm2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.out_proj.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.2.crossattention.output.out_proj.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.attention.key.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.attention.key.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.attention.query.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.attention.query.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.attention.value.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.attention.value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.norm1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.norm1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.normk.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.normk.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.ffn_ln.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.ffn_ln.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.w1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.w1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.w2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.w2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.w3.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.mlp.w3.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.norm2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.norm2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.out_proj.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.3.crossattention.output.out_proj.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.attention.key.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.attention.key.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.attention.query.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.attention.query.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.attention.value.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.attention.value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.norm1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.norm1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.normk.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.normk.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.ffn_ln.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.ffn_ln.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.w1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.w1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.w2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.w2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.w3.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.mlp.w3.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.norm2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.norm2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.out_proj.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.4.crossattention.output.out_proj.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.attention.key.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.attention.key.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.attention.query.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.attention.query.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.attention.value.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.attention.value.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.norm1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.norm1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.normk.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.normk.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.ffn_ln.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.ffn_ln.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.w1.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.w1.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.w2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.w2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.w3.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.mlp.w3.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.norm2.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.norm2.weight": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.out_proj.bias": "pytorch_model-33-of-33.bin", "model.visual_abstractor.encoder.layers.5.crossattention.output.out_proj.weight": "pytorch_model-33-of-33.bin"}, "metadata": {"total_size": 16395722752}}
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "bos_token": {
5
+ "__type": "AddedToken",
6
+ "content": "<s>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "clean_up_tokenization_spaces": false,
13
+ "eos_token": {
14
+ "__type": "AddedToken",
15
+ "content": "</s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false
20
+ },
21
+ "legacy": false,
22
+ "model_max_length": 4096,
23
+ "pad_token": null,
24
+ "padding_side": "right",
25
+ "sp_model_kwargs": {},
26
+ "tokenizer_class": "LlamaTokenizer",
27
+ "unk_token": {
28
+ "__type": "AddedToken",
29
+ "content": "<unk>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false
34
+ }
35
+ }