digitous commited on
Commit
8704505
1 Parent(s): d36bfe9

Upload 8 files

Browse files
added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"<|extratoken_14|>": 50270, "<|extratoken_121|>": 50377, "<|extratoken_3|>": 50259, "<|extratoken_25|>": 50281, "<|extratoken_101|>": 50357, "<|extratoken_138|>": 50394, "<|extratoken_10|>": 50266, "<|extratoken_21|>": 50277, "<|extratoken_32|>": 50288, "<|extratoken_46|>": 50302, "<|extratoken_22|>": 50278, "<|extratoken_40|>": 50296, "<|extratoken_96|>": 50352, "<|extratoken_92|>": 50348, "<|extratoken_95|>": 50351, "<|extratoken_141|>": 50397, "<|extratoken_78|>": 50334, "<|extratoken_86|>": 50342, "<|extratoken_56|>": 50312, "<|extratoken_124|>": 50380, "<|extratoken_127|>": 50383, "<|extratoken_122|>": 50378, "<|extratoken_123|>": 50379, "<|extratoken_111|>": 50367, "<|extratoken_93|>": 50349, "<|extratoken_130|>": 50386, "<|extratoken_113|>": 50369, "<|extratoken_50|>": 50306, "<|extratoken_97|>": 50353, "<|extratoken_1|>": 50257, "<|extratoken_55|>": 50311, "<|extratoken_34|>": 50290, "<|extratoken_143|>": 50399, "<|extratoken_62|>": 50318, "<|extratoken_74|>": 50330, "<|extratoken_136|>": 50392, "<|extratoken_117|>": 50373, "<|extratoken_38|>": 50294, "<|extratoken_120|>": 50376, "<|extratoken_39|>": 50295, "<|extratoken_65|>": 50321, "<|extratoken_29|>": 50285, "<|extratoken_104|>": 50360, "<|extratoken_13|>": 50269, "<|extratoken_5|>": 50261, "<|extratoken_107|>": 50363, "<|extratoken_19|>": 50275, "<|extratoken_84|>": 50340, "<|extratoken_77|>": 50333, "<|extratoken_135|>": 50391, "<|extratoken_24|>": 50280, "<|extratoken_134|>": 50390, "<|extratoken_15|>": 50271, "<|extratoken_67|>": 50323, "<|extratoken_89|>": 50345, "<|extratoken_2|>": 50258, "<|extratoken_73|>": 50329, "<|extratoken_129|>": 50385, "<|extratoken_126|>": 50382, "<|extratoken_30|>": 50286, "<|extratoken_41|>": 50297, "<|extratoken_28|>": 50284, "<|extratoken_114|>": 50370, "<|extratoken_128|>": 50384, "<|extratoken_118|>": 50374, "<|extratoken_131|>": 50387, "<|extratoken_68|>": 50324, "<|extratoken_125|>": 50381, "<|extratoken_103|>": 50359, "<|extratoken_8|>": 50264, "<|extratoken_64|>": 50320, "<|extratoken_52|>": 50308, "<|extratoken_45|>": 50301, "<|extratoken_43|>": 50299, "<|extratoken_18|>": 50274, "<|extratoken_139|>": 50395, "<|extratoken_85|>": 50341, "<|extratoken_88|>": 50344, "<|extratoken_63|>": 50319, "<|extratoken_4|>": 50260, "<|extratoken_48|>": 50304, "<|extratoken_112|>": 50368, "<|extratoken_17|>": 50273, "<|extratoken_49|>": 50305, "<|extratoken_108|>": 50364, "<|extratoken_110|>": 50366, "<|extratoken_42|>": 50298, "<|extratoken_70|>": 50326, "<|extratoken_6|>": 50262, "<|extratoken_35|>": 50291, "<|extratoken_23|>": 50279, "<|extratoken_66|>": 50322, "<|extratoken_60|>": 50316, "<|extratoken_71|>": 50327, "<|extratoken_51|>": 50307, "<|extratoken_133|>": 50389, "<|extratoken_20|>": 50276, "<|extratoken_76|>": 50332, "<|extratoken_81|>": 50337, "<|extratoken_142|>": 50398, "<|extratoken_116|>": 50372, "<|extratoken_57|>": 50313, "<|extratoken_75|>": 50331, "<|extratoken_37|>": 50293, "<|extratoken_33|>": 50289, "<|extratoken_16|>": 50272, "<|extratoken_61|>": 50317, "<|extratoken_7|>": 50263, "<|extratoken_12|>": 50268, "<|extratoken_36|>": 50292, "<|extratoken_80|>": 50336, "<|extratoken_98|>": 50354, "<|extratoken_105|>": 50361, "<|extratoken_91|>": 50347, "<|extratoken_53|>": 50309, "<|extratoken_137|>": 50393, "<|extratoken_9|>": 50265, "<|extratoken_79|>": 50335, "<|extratoken_83|>": 50339, "<|extratoken_109|>": 50365, "<|extratoken_99|>": 50355, "<|extratoken_140|>": 50396, "<|extratoken_72|>": 50328, "<|extratoken_11|>": 50267, "<|extratoken_94|>": 50350, "<|extratoken_26|>": 50282, "<|extratoken_59|>": 50315, "<|extratoken_106|>": 50362, "<|extratoken_115|>": 50371, "<|extratoken_58|>": 50314, "<|extratoken_90|>": 50346, "<|extratoken_31|>": 50287, "<|extratoken_102|>": 50358, "<|extratoken_47|>": 50303, "<|extratoken_100|>": 50356, "<|extratoken_82|>": 50338, "<|extratoken_44|>": 50300, "<|extratoken_69|>": 50325, "<|extratoken_54|>": 50310, "<|extratoken_132|>": 50388, "<|extratoken_27|>": 50283, "<|extratoken_87|>": 50343, "<|extratoken_119|>": 50375}
config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Javalion-R",
3
+ "activation_function": "gelu_new",
4
+ "antemplate": "[Genre: <|>]",
5
+ "architectures": [
6
+ "GPTJForCausalLM"
7
+ ],
8
+ "attn_pdrop": 0.0,
9
+ "bos_token_id": 50256,
10
+ "embd_pdrop": 0.0,
11
+ "eos_token_id": 50256,
12
+ "gradient_checkpointing": false,
13
+ "initializer_range": 0.02,
14
+ "layer_norm_epsilon": 1e-05,
15
+ "model_type": "gptj",
16
+ "n_embd": 4096,
17
+ "n_head": 16,
18
+ "n_inner": null,
19
+ "n_layer": 28,
20
+ "n_positions": 2048,
21
+ "resid_pdrop": 0.0,
22
+ "rotary_dim": 64,
23
+ "scale_attn_weights": true,
24
+ "summary_activation": null,
25
+ "summary_first_dropout": 0.1,
26
+ "summary_proj_to_labels": true,
27
+ "summary_type": "cls_index",
28
+ "summary_use_proj": true,
29
+ "task_specific_params": {
30
+ "text-generation": {
31
+ "do_sample": true,
32
+ "max_length": 50,
33
+ "temperature": 1.0
34
+ }
35
+ },
36
+ "tie_word_embeddings": false,
37
+ "tokenizer_class": "GPT2Tokenizer",
38
+ "torch_dtype": "float16",
39
+ "transformers_version": "4.24.0",
40
+ "use_cache": true,
41
+ "vocab_size": 50400
42
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin.index.json ADDED
@@ -0,0 +1,348 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 12219206136
4
+ },
5
+ "weight_map": {
6
+ "lm_head.bias": "pytorch_model-00006-of-00006.bin",
7
+ "lm_head.weight": "pytorch_model-00006-of-00006.bin",
8
+ "transformer.h.0.attn.bias": "pytorch_model-00001-of-00006.bin",
9
+ "transformer.h.0.attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
10
+ "transformer.h.0.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
11
+ "transformer.h.0.attn.out_proj.weight": "pytorch_model-00001-of-00006.bin",
12
+ "transformer.h.0.attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
13
+ "transformer.h.0.attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
14
+ "transformer.h.0.ln_1.bias": "pytorch_model-00001-of-00006.bin",
15
+ "transformer.h.0.ln_1.weight": "pytorch_model-00001-of-00006.bin",
16
+ "transformer.h.0.mlp.fc_in.bias": "pytorch_model-00001-of-00006.bin",
17
+ "transformer.h.0.mlp.fc_in.weight": "pytorch_model-00001-of-00006.bin",
18
+ "transformer.h.0.mlp.fc_out.bias": "pytorch_model-00001-of-00006.bin",
19
+ "transformer.h.0.mlp.fc_out.weight": "pytorch_model-00001-of-00006.bin",
20
+ "transformer.h.1.attn.bias": "pytorch_model-00001-of-00006.bin",
21
+ "transformer.h.1.attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
22
+ "transformer.h.1.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
23
+ "transformer.h.1.attn.out_proj.weight": "pytorch_model-00001-of-00006.bin",
24
+ "transformer.h.1.attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
25
+ "transformer.h.1.attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
26
+ "transformer.h.1.ln_1.bias": "pytorch_model-00001-of-00006.bin",
27
+ "transformer.h.1.ln_1.weight": "pytorch_model-00001-of-00006.bin",
28
+ "transformer.h.1.mlp.fc_in.bias": "pytorch_model-00001-of-00006.bin",
29
+ "transformer.h.1.mlp.fc_in.weight": "pytorch_model-00001-of-00006.bin",
30
+ "transformer.h.1.mlp.fc_out.bias": "pytorch_model-00001-of-00006.bin",
31
+ "transformer.h.1.mlp.fc_out.weight": "pytorch_model-00001-of-00006.bin",
32
+ "transformer.h.10.attn.bias": "pytorch_model-00003-of-00006.bin",
33
+ "transformer.h.10.attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
34
+ "transformer.h.10.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
35
+ "transformer.h.10.attn.out_proj.weight": "pytorch_model-00003-of-00006.bin",
36
+ "transformer.h.10.attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
37
+ "transformer.h.10.attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
38
+ "transformer.h.10.ln_1.bias": "pytorch_model-00003-of-00006.bin",
39
+ "transformer.h.10.ln_1.weight": "pytorch_model-00003-of-00006.bin",
40
+ "transformer.h.10.mlp.fc_in.bias": "pytorch_model-00003-of-00006.bin",
41
+ "transformer.h.10.mlp.fc_in.weight": "pytorch_model-00003-of-00006.bin",
42
+ "transformer.h.10.mlp.fc_out.bias": "pytorch_model-00003-of-00006.bin",
43
+ "transformer.h.10.mlp.fc_out.weight": "pytorch_model-00003-of-00006.bin",
44
+ "transformer.h.11.attn.bias": "pytorch_model-00003-of-00006.bin",
45
+ "transformer.h.11.attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
46
+ "transformer.h.11.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
47
+ "transformer.h.11.attn.out_proj.weight": "pytorch_model-00003-of-00006.bin",
48
+ "transformer.h.11.attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
49
+ "transformer.h.11.attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
50
+ "transformer.h.11.ln_1.bias": "pytorch_model-00003-of-00006.bin",
51
+ "transformer.h.11.ln_1.weight": "pytorch_model-00003-of-00006.bin",
52
+ "transformer.h.11.mlp.fc_in.bias": "pytorch_model-00003-of-00006.bin",
53
+ "transformer.h.11.mlp.fc_in.weight": "pytorch_model-00003-of-00006.bin",
54
+ "transformer.h.11.mlp.fc_out.bias": "pytorch_model-00003-of-00006.bin",
55
+ "transformer.h.11.mlp.fc_out.weight": "pytorch_model-00003-of-00006.bin",
56
+ "transformer.h.12.attn.bias": "pytorch_model-00003-of-00006.bin",
57
+ "transformer.h.12.attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
58
+ "transformer.h.12.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
59
+ "transformer.h.12.attn.out_proj.weight": "pytorch_model-00003-of-00006.bin",
60
+ "transformer.h.12.attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
61
+ "transformer.h.12.attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
62
+ "transformer.h.12.ln_1.bias": "pytorch_model-00003-of-00006.bin",
63
+ "transformer.h.12.ln_1.weight": "pytorch_model-00003-of-00006.bin",
64
+ "transformer.h.12.mlp.fc_in.bias": "pytorch_model-00003-of-00006.bin",
65
+ "transformer.h.12.mlp.fc_in.weight": "pytorch_model-00003-of-00006.bin",
66
+ "transformer.h.12.mlp.fc_out.bias": "pytorch_model-00003-of-00006.bin",
67
+ "transformer.h.12.mlp.fc_out.weight": "pytorch_model-00003-of-00006.bin",
68
+ "transformer.h.13.attn.bias": "pytorch_model-00003-of-00006.bin",
69
+ "transformer.h.13.attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
70
+ "transformer.h.13.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
71
+ "transformer.h.13.attn.out_proj.weight": "pytorch_model-00003-of-00006.bin",
72
+ "transformer.h.13.attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
73
+ "transformer.h.13.attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
74
+ "transformer.h.13.ln_1.bias": "pytorch_model-00003-of-00006.bin",
75
+ "transformer.h.13.ln_1.weight": "pytorch_model-00003-of-00006.bin",
76
+ "transformer.h.13.mlp.fc_in.bias": "pytorch_model-00003-of-00006.bin",
77
+ "transformer.h.13.mlp.fc_in.weight": "pytorch_model-00003-of-00006.bin",
78
+ "transformer.h.13.mlp.fc_out.bias": "pytorch_model-00003-of-00006.bin",
79
+ "transformer.h.13.mlp.fc_out.weight": "pytorch_model-00003-of-00006.bin",
80
+ "transformer.h.14.attn.bias": "pytorch_model-00003-of-00006.bin",
81
+ "transformer.h.14.attn.k_proj.weight": "pytorch_model-00003-of-00006.bin",
82
+ "transformer.h.14.attn.masked_bias": "pytorch_model-00003-of-00006.bin",
83
+ "transformer.h.14.attn.out_proj.weight": "pytorch_model-00004-of-00006.bin",
84
+ "transformer.h.14.attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
85
+ "transformer.h.14.attn.v_proj.weight": "pytorch_model-00003-of-00006.bin",
86
+ "transformer.h.14.ln_1.bias": "pytorch_model-00003-of-00006.bin",
87
+ "transformer.h.14.ln_1.weight": "pytorch_model-00003-of-00006.bin",
88
+ "transformer.h.14.mlp.fc_in.bias": "pytorch_model-00004-of-00006.bin",
89
+ "transformer.h.14.mlp.fc_in.weight": "pytorch_model-00004-of-00006.bin",
90
+ "transformer.h.14.mlp.fc_out.bias": "pytorch_model-00004-of-00006.bin",
91
+ "transformer.h.14.mlp.fc_out.weight": "pytorch_model-00004-of-00006.bin",
92
+ "transformer.h.15.attn.bias": "pytorch_model-00004-of-00006.bin",
93
+ "transformer.h.15.attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
94
+ "transformer.h.15.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
95
+ "transformer.h.15.attn.out_proj.weight": "pytorch_model-00004-of-00006.bin",
96
+ "transformer.h.15.attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
97
+ "transformer.h.15.attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
98
+ "transformer.h.15.ln_1.bias": "pytorch_model-00004-of-00006.bin",
99
+ "transformer.h.15.ln_1.weight": "pytorch_model-00004-of-00006.bin",
100
+ "transformer.h.15.mlp.fc_in.bias": "pytorch_model-00004-of-00006.bin",
101
+ "transformer.h.15.mlp.fc_in.weight": "pytorch_model-00004-of-00006.bin",
102
+ "transformer.h.15.mlp.fc_out.bias": "pytorch_model-00004-of-00006.bin",
103
+ "transformer.h.15.mlp.fc_out.weight": "pytorch_model-00004-of-00006.bin",
104
+ "transformer.h.16.attn.bias": "pytorch_model-00004-of-00006.bin",
105
+ "transformer.h.16.attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
106
+ "transformer.h.16.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
107
+ "transformer.h.16.attn.out_proj.weight": "pytorch_model-00004-of-00006.bin",
108
+ "transformer.h.16.attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
109
+ "transformer.h.16.attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
110
+ "transformer.h.16.ln_1.bias": "pytorch_model-00004-of-00006.bin",
111
+ "transformer.h.16.ln_1.weight": "pytorch_model-00004-of-00006.bin",
112
+ "transformer.h.16.mlp.fc_in.bias": "pytorch_model-00004-of-00006.bin",
113
+ "transformer.h.16.mlp.fc_in.weight": "pytorch_model-00004-of-00006.bin",
114
+ "transformer.h.16.mlp.fc_out.bias": "pytorch_model-00004-of-00006.bin",
115
+ "transformer.h.16.mlp.fc_out.weight": "pytorch_model-00004-of-00006.bin",
116
+ "transformer.h.17.attn.bias": "pytorch_model-00004-of-00006.bin",
117
+ "transformer.h.17.attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
118
+ "transformer.h.17.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
119
+ "transformer.h.17.attn.out_proj.weight": "pytorch_model-00004-of-00006.bin",
120
+ "transformer.h.17.attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
121
+ "transformer.h.17.attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
122
+ "transformer.h.17.ln_1.bias": "pytorch_model-00004-of-00006.bin",
123
+ "transformer.h.17.ln_1.weight": "pytorch_model-00004-of-00006.bin",
124
+ "transformer.h.17.mlp.fc_in.bias": "pytorch_model-00004-of-00006.bin",
125
+ "transformer.h.17.mlp.fc_in.weight": "pytorch_model-00004-of-00006.bin",
126
+ "transformer.h.17.mlp.fc_out.bias": "pytorch_model-00004-of-00006.bin",
127
+ "transformer.h.17.mlp.fc_out.weight": "pytorch_model-00004-of-00006.bin",
128
+ "transformer.h.18.attn.bias": "pytorch_model-00004-of-00006.bin",
129
+ "transformer.h.18.attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
130
+ "transformer.h.18.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
131
+ "transformer.h.18.attn.out_proj.weight": "pytorch_model-00004-of-00006.bin",
132
+ "transformer.h.18.attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
133
+ "transformer.h.18.attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
134
+ "transformer.h.18.ln_1.bias": "pytorch_model-00004-of-00006.bin",
135
+ "transformer.h.18.ln_1.weight": "pytorch_model-00004-of-00006.bin",
136
+ "transformer.h.18.mlp.fc_in.bias": "pytorch_model-00004-of-00006.bin",
137
+ "transformer.h.18.mlp.fc_in.weight": "pytorch_model-00004-of-00006.bin",
138
+ "transformer.h.18.mlp.fc_out.bias": "pytorch_model-00004-of-00006.bin",
139
+ "transformer.h.18.mlp.fc_out.weight": "pytorch_model-00004-of-00006.bin",
140
+ "transformer.h.19.attn.bias": "pytorch_model-00004-of-00006.bin",
141
+ "transformer.h.19.attn.k_proj.weight": "pytorch_model-00004-of-00006.bin",
142
+ "transformer.h.19.attn.masked_bias": "pytorch_model-00004-of-00006.bin",
143
+ "transformer.h.19.attn.out_proj.weight": "pytorch_model-00004-of-00006.bin",
144
+ "transformer.h.19.attn.q_proj.weight": "pytorch_model-00004-of-00006.bin",
145
+ "transformer.h.19.attn.v_proj.weight": "pytorch_model-00004-of-00006.bin",
146
+ "transformer.h.19.ln_1.bias": "pytorch_model-00004-of-00006.bin",
147
+ "transformer.h.19.ln_1.weight": "pytorch_model-00004-of-00006.bin",
148
+ "transformer.h.19.mlp.fc_in.bias": "pytorch_model-00005-of-00006.bin",
149
+ "transformer.h.19.mlp.fc_in.weight": "pytorch_model-00005-of-00006.bin",
150
+ "transformer.h.19.mlp.fc_out.bias": "pytorch_model-00005-of-00006.bin",
151
+ "transformer.h.19.mlp.fc_out.weight": "pytorch_model-00005-of-00006.bin",
152
+ "transformer.h.2.attn.bias": "pytorch_model-00001-of-00006.bin",
153
+ "transformer.h.2.attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
154
+ "transformer.h.2.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
155
+ "transformer.h.2.attn.out_proj.weight": "pytorch_model-00001-of-00006.bin",
156
+ "transformer.h.2.attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
157
+ "transformer.h.2.attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
158
+ "transformer.h.2.ln_1.bias": "pytorch_model-00001-of-00006.bin",
159
+ "transformer.h.2.ln_1.weight": "pytorch_model-00001-of-00006.bin",
160
+ "transformer.h.2.mlp.fc_in.bias": "pytorch_model-00001-of-00006.bin",
161
+ "transformer.h.2.mlp.fc_in.weight": "pytorch_model-00001-of-00006.bin",
162
+ "transformer.h.2.mlp.fc_out.bias": "pytorch_model-00001-of-00006.bin",
163
+ "transformer.h.2.mlp.fc_out.weight": "pytorch_model-00001-of-00006.bin",
164
+ "transformer.h.20.attn.bias": "pytorch_model-00005-of-00006.bin",
165
+ "transformer.h.20.attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
166
+ "transformer.h.20.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
167
+ "transformer.h.20.attn.out_proj.weight": "pytorch_model-00005-of-00006.bin",
168
+ "transformer.h.20.attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
169
+ "transformer.h.20.attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
170
+ "transformer.h.20.ln_1.bias": "pytorch_model-00005-of-00006.bin",
171
+ "transformer.h.20.ln_1.weight": "pytorch_model-00005-of-00006.bin",
172
+ "transformer.h.20.mlp.fc_in.bias": "pytorch_model-00005-of-00006.bin",
173
+ "transformer.h.20.mlp.fc_in.weight": "pytorch_model-00005-of-00006.bin",
174
+ "transformer.h.20.mlp.fc_out.bias": "pytorch_model-00005-of-00006.bin",
175
+ "transformer.h.20.mlp.fc_out.weight": "pytorch_model-00005-of-00006.bin",
176
+ "transformer.h.21.attn.bias": "pytorch_model-00005-of-00006.bin",
177
+ "transformer.h.21.attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
178
+ "transformer.h.21.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
179
+ "transformer.h.21.attn.out_proj.weight": "pytorch_model-00005-of-00006.bin",
180
+ "transformer.h.21.attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
181
+ "transformer.h.21.attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
182
+ "transformer.h.21.ln_1.bias": "pytorch_model-00005-of-00006.bin",
183
+ "transformer.h.21.ln_1.weight": "pytorch_model-00005-of-00006.bin",
184
+ "transformer.h.21.mlp.fc_in.bias": "pytorch_model-00005-of-00006.bin",
185
+ "transformer.h.21.mlp.fc_in.weight": "pytorch_model-00005-of-00006.bin",
186
+ "transformer.h.21.mlp.fc_out.bias": "pytorch_model-00005-of-00006.bin",
187
+ "transformer.h.21.mlp.fc_out.weight": "pytorch_model-00005-of-00006.bin",
188
+ "transformer.h.22.attn.bias": "pytorch_model-00005-of-00006.bin",
189
+ "transformer.h.22.attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
190
+ "transformer.h.22.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
191
+ "transformer.h.22.attn.out_proj.weight": "pytorch_model-00005-of-00006.bin",
192
+ "transformer.h.22.attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
193
+ "transformer.h.22.attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
194
+ "transformer.h.22.ln_1.bias": "pytorch_model-00005-of-00006.bin",
195
+ "transformer.h.22.ln_1.weight": "pytorch_model-00005-of-00006.bin",
196
+ "transformer.h.22.mlp.fc_in.bias": "pytorch_model-00005-of-00006.bin",
197
+ "transformer.h.22.mlp.fc_in.weight": "pytorch_model-00005-of-00006.bin",
198
+ "transformer.h.22.mlp.fc_out.bias": "pytorch_model-00005-of-00006.bin",
199
+ "transformer.h.22.mlp.fc_out.weight": "pytorch_model-00005-of-00006.bin",
200
+ "transformer.h.23.attn.bias": "pytorch_model-00005-of-00006.bin",
201
+ "transformer.h.23.attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
202
+ "transformer.h.23.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
203
+ "transformer.h.23.attn.out_proj.weight": "pytorch_model-00005-of-00006.bin",
204
+ "transformer.h.23.attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
205
+ "transformer.h.23.attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
206
+ "transformer.h.23.ln_1.bias": "pytorch_model-00005-of-00006.bin",
207
+ "transformer.h.23.ln_1.weight": "pytorch_model-00005-of-00006.bin",
208
+ "transformer.h.23.mlp.fc_in.bias": "pytorch_model-00005-of-00006.bin",
209
+ "transformer.h.23.mlp.fc_in.weight": "pytorch_model-00005-of-00006.bin",
210
+ "transformer.h.23.mlp.fc_out.bias": "pytorch_model-00005-of-00006.bin",
211
+ "transformer.h.23.mlp.fc_out.weight": "pytorch_model-00005-of-00006.bin",
212
+ "transformer.h.24.attn.bias": "pytorch_model-00005-of-00006.bin",
213
+ "transformer.h.24.attn.k_proj.weight": "pytorch_model-00005-of-00006.bin",
214
+ "transformer.h.24.attn.masked_bias": "pytorch_model-00005-of-00006.bin",
215
+ "transformer.h.24.attn.out_proj.weight": "pytorch_model-00005-of-00006.bin",
216
+ "transformer.h.24.attn.q_proj.weight": "pytorch_model-00005-of-00006.bin",
217
+ "transformer.h.24.attn.v_proj.weight": "pytorch_model-00005-of-00006.bin",
218
+ "transformer.h.24.ln_1.bias": "pytorch_model-00005-of-00006.bin",
219
+ "transformer.h.24.ln_1.weight": "pytorch_model-00005-of-00006.bin",
220
+ "transformer.h.24.mlp.fc_in.bias": "pytorch_model-00006-of-00006.bin",
221
+ "transformer.h.24.mlp.fc_in.weight": "pytorch_model-00006-of-00006.bin",
222
+ "transformer.h.24.mlp.fc_out.bias": "pytorch_model-00006-of-00006.bin",
223
+ "transformer.h.24.mlp.fc_out.weight": "pytorch_model-00006-of-00006.bin",
224
+ "transformer.h.25.attn.bias": "pytorch_model-00006-of-00006.bin",
225
+ "transformer.h.25.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin",
226
+ "transformer.h.25.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
227
+ "transformer.h.25.attn.out_proj.weight": "pytorch_model-00006-of-00006.bin",
228
+ "transformer.h.25.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin",
229
+ "transformer.h.25.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin",
230
+ "transformer.h.25.ln_1.bias": "pytorch_model-00006-of-00006.bin",
231
+ "transformer.h.25.ln_1.weight": "pytorch_model-00006-of-00006.bin",
232
+ "transformer.h.25.mlp.fc_in.bias": "pytorch_model-00006-of-00006.bin",
233
+ "transformer.h.25.mlp.fc_in.weight": "pytorch_model-00006-of-00006.bin",
234
+ "transformer.h.25.mlp.fc_out.bias": "pytorch_model-00006-of-00006.bin",
235
+ "transformer.h.25.mlp.fc_out.weight": "pytorch_model-00006-of-00006.bin",
236
+ "transformer.h.26.attn.bias": "pytorch_model-00006-of-00006.bin",
237
+ "transformer.h.26.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin",
238
+ "transformer.h.26.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
239
+ "transformer.h.26.attn.out_proj.weight": "pytorch_model-00006-of-00006.bin",
240
+ "transformer.h.26.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin",
241
+ "transformer.h.26.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin",
242
+ "transformer.h.26.ln_1.bias": "pytorch_model-00006-of-00006.bin",
243
+ "transformer.h.26.ln_1.weight": "pytorch_model-00006-of-00006.bin",
244
+ "transformer.h.26.mlp.fc_in.bias": "pytorch_model-00006-of-00006.bin",
245
+ "transformer.h.26.mlp.fc_in.weight": "pytorch_model-00006-of-00006.bin",
246
+ "transformer.h.26.mlp.fc_out.bias": "pytorch_model-00006-of-00006.bin",
247
+ "transformer.h.26.mlp.fc_out.weight": "pytorch_model-00006-of-00006.bin",
248
+ "transformer.h.27.attn.bias": "pytorch_model-00006-of-00006.bin",
249
+ "transformer.h.27.attn.k_proj.weight": "pytorch_model-00006-of-00006.bin",
250
+ "transformer.h.27.attn.masked_bias": "pytorch_model-00006-of-00006.bin",
251
+ "transformer.h.27.attn.out_proj.weight": "pytorch_model-00006-of-00006.bin",
252
+ "transformer.h.27.attn.q_proj.weight": "pytorch_model-00006-of-00006.bin",
253
+ "transformer.h.27.attn.v_proj.weight": "pytorch_model-00006-of-00006.bin",
254
+ "transformer.h.27.ln_1.bias": "pytorch_model-00006-of-00006.bin",
255
+ "transformer.h.27.ln_1.weight": "pytorch_model-00006-of-00006.bin",
256
+ "transformer.h.27.mlp.fc_in.bias": "pytorch_model-00006-of-00006.bin",
257
+ "transformer.h.27.mlp.fc_in.weight": "pytorch_model-00006-of-00006.bin",
258
+ "transformer.h.27.mlp.fc_out.bias": "pytorch_model-00006-of-00006.bin",
259
+ "transformer.h.27.mlp.fc_out.weight": "pytorch_model-00006-of-00006.bin",
260
+ "transformer.h.3.attn.bias": "pytorch_model-00001-of-00006.bin",
261
+ "transformer.h.3.attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
262
+ "transformer.h.3.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
263
+ "transformer.h.3.attn.out_proj.weight": "pytorch_model-00001-of-00006.bin",
264
+ "transformer.h.3.attn.q_proj.weight": "pytorch_model-00001-of-00006.bin",
265
+ "transformer.h.3.attn.v_proj.weight": "pytorch_model-00001-of-00006.bin",
266
+ "transformer.h.3.ln_1.bias": "pytorch_model-00001-of-00006.bin",
267
+ "transformer.h.3.ln_1.weight": "pytorch_model-00001-of-00006.bin",
268
+ "transformer.h.3.mlp.fc_in.bias": "pytorch_model-00001-of-00006.bin",
269
+ "transformer.h.3.mlp.fc_in.weight": "pytorch_model-00001-of-00006.bin",
270
+ "transformer.h.3.mlp.fc_out.bias": "pytorch_model-00001-of-00006.bin",
271
+ "transformer.h.3.mlp.fc_out.weight": "pytorch_model-00001-of-00006.bin",
272
+ "transformer.h.4.attn.bias": "pytorch_model-00001-of-00006.bin",
273
+ "transformer.h.4.attn.k_proj.weight": "pytorch_model-00001-of-00006.bin",
274
+ "transformer.h.4.attn.masked_bias": "pytorch_model-00001-of-00006.bin",
275
+ "transformer.h.4.attn.out_proj.weight": "pytorch_model-00002-of-00006.bin",
276
+ "transformer.h.4.attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
277
+ "transformer.h.4.attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
278
+ "transformer.h.4.ln_1.bias": "pytorch_model-00001-of-00006.bin",
279
+ "transformer.h.4.ln_1.weight": "pytorch_model-00001-of-00006.bin",
280
+ "transformer.h.4.mlp.fc_in.bias": "pytorch_model-00002-of-00006.bin",
281
+ "transformer.h.4.mlp.fc_in.weight": "pytorch_model-00002-of-00006.bin",
282
+ "transformer.h.4.mlp.fc_out.bias": "pytorch_model-00002-of-00006.bin",
283
+ "transformer.h.4.mlp.fc_out.weight": "pytorch_model-00002-of-00006.bin",
284
+ "transformer.h.5.attn.bias": "pytorch_model-00002-of-00006.bin",
285
+ "transformer.h.5.attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
286
+ "transformer.h.5.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
287
+ "transformer.h.5.attn.out_proj.weight": "pytorch_model-00002-of-00006.bin",
288
+ "transformer.h.5.attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
289
+ "transformer.h.5.attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
290
+ "transformer.h.5.ln_1.bias": "pytorch_model-00002-of-00006.bin",
291
+ "transformer.h.5.ln_1.weight": "pytorch_model-00002-of-00006.bin",
292
+ "transformer.h.5.mlp.fc_in.bias": "pytorch_model-00002-of-00006.bin",
293
+ "transformer.h.5.mlp.fc_in.weight": "pytorch_model-00002-of-00006.bin",
294
+ "transformer.h.5.mlp.fc_out.bias": "pytorch_model-00002-of-00006.bin",
295
+ "transformer.h.5.mlp.fc_out.weight": "pytorch_model-00002-of-00006.bin",
296
+ "transformer.h.6.attn.bias": "pytorch_model-00002-of-00006.bin",
297
+ "transformer.h.6.attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
298
+ "transformer.h.6.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
299
+ "transformer.h.6.attn.out_proj.weight": "pytorch_model-00002-of-00006.bin",
300
+ "transformer.h.6.attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
301
+ "transformer.h.6.attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
302
+ "transformer.h.6.ln_1.bias": "pytorch_model-00002-of-00006.bin",
303
+ "transformer.h.6.ln_1.weight": "pytorch_model-00002-of-00006.bin",
304
+ "transformer.h.6.mlp.fc_in.bias": "pytorch_model-00002-of-00006.bin",
305
+ "transformer.h.6.mlp.fc_in.weight": "pytorch_model-00002-of-00006.bin",
306
+ "transformer.h.6.mlp.fc_out.bias": "pytorch_model-00002-of-00006.bin",
307
+ "transformer.h.6.mlp.fc_out.weight": "pytorch_model-00002-of-00006.bin",
308
+ "transformer.h.7.attn.bias": "pytorch_model-00002-of-00006.bin",
309
+ "transformer.h.7.attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
310
+ "transformer.h.7.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
311
+ "transformer.h.7.attn.out_proj.weight": "pytorch_model-00002-of-00006.bin",
312
+ "transformer.h.7.attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
313
+ "transformer.h.7.attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
314
+ "transformer.h.7.ln_1.bias": "pytorch_model-00002-of-00006.bin",
315
+ "transformer.h.7.ln_1.weight": "pytorch_model-00002-of-00006.bin",
316
+ "transformer.h.7.mlp.fc_in.bias": "pytorch_model-00002-of-00006.bin",
317
+ "transformer.h.7.mlp.fc_in.weight": "pytorch_model-00002-of-00006.bin",
318
+ "transformer.h.7.mlp.fc_out.bias": "pytorch_model-00002-of-00006.bin",
319
+ "transformer.h.7.mlp.fc_out.weight": "pytorch_model-00002-of-00006.bin",
320
+ "transformer.h.8.attn.bias": "pytorch_model-00002-of-00006.bin",
321
+ "transformer.h.8.attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
322
+ "transformer.h.8.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
323
+ "transformer.h.8.attn.out_proj.weight": "pytorch_model-00002-of-00006.bin",
324
+ "transformer.h.8.attn.q_proj.weight": "pytorch_model-00002-of-00006.bin",
325
+ "transformer.h.8.attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
326
+ "transformer.h.8.ln_1.bias": "pytorch_model-00002-of-00006.bin",
327
+ "transformer.h.8.ln_1.weight": "pytorch_model-00002-of-00006.bin",
328
+ "transformer.h.8.mlp.fc_in.bias": "pytorch_model-00002-of-00006.bin",
329
+ "transformer.h.8.mlp.fc_in.weight": "pytorch_model-00002-of-00006.bin",
330
+ "transformer.h.8.mlp.fc_out.bias": "pytorch_model-00002-of-00006.bin",
331
+ "transformer.h.8.mlp.fc_out.weight": "pytorch_model-00002-of-00006.bin",
332
+ "transformer.h.9.attn.bias": "pytorch_model-00002-of-00006.bin",
333
+ "transformer.h.9.attn.k_proj.weight": "pytorch_model-00002-of-00006.bin",
334
+ "transformer.h.9.attn.masked_bias": "pytorch_model-00002-of-00006.bin",
335
+ "transformer.h.9.attn.out_proj.weight": "pytorch_model-00003-of-00006.bin",
336
+ "transformer.h.9.attn.q_proj.weight": "pytorch_model-00003-of-00006.bin",
337
+ "transformer.h.9.attn.v_proj.weight": "pytorch_model-00002-of-00006.bin",
338
+ "transformer.h.9.ln_1.bias": "pytorch_model-00002-of-00006.bin",
339
+ "transformer.h.9.ln_1.weight": "pytorch_model-00002-of-00006.bin",
340
+ "transformer.h.9.mlp.fc_in.bias": "pytorch_model-00003-of-00006.bin",
341
+ "transformer.h.9.mlp.fc_in.weight": "pytorch_model-00003-of-00006.bin",
342
+ "transformer.h.9.mlp.fc_out.bias": "pytorch_model-00003-of-00006.bin",
343
+ "transformer.h.9.mlp.fc_out.weight": "pytorch_model-00003-of-00006.bin",
344
+ "transformer.ln_f.bias": "pytorch_model-00006-of-00006.bin",
345
+ "transformer.ln_f.weight": "pytorch_model-00006-of-00006.bin",
346
+ "transformer.wte.weight": "pytorch_model-00001-of-00006.bin"
347
+ }
348
+ }
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "bos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "add_prefix_space": false, "errors": "replace", "model_max_length": 2048, "special_tokens_map_file": null, "name_or_path": "gpt-j-6B", "from_slow": true, "tokenizer_class": "GPT2Tokenizer"}
vocab.json ADDED
The diff for this file is too large to render. See raw diff