Upload 33 files
Browse files- BF16/config.json +44 -0
- BF16/gpt2_model.safetensors +3 -0
- BF16/special_tokens_map.json +6 -0
- BF16/tokenizer.json +0 -0
- BF16/tokenizer_config.json +192 -0
- FP16/config.json +44 -0
- FP16/gpt2_model.safetensors +3 -0
- FP16/special_tokens_map.json +6 -0
- FP16/tokenizer.json +0 -0
- FP16/tokenizer_config.json +192 -0
- FP32/config.json +44 -0
- FP32/gpt2_model.safetensors +3 -0
- FP32/special_tokens_map.json +6 -0
- FP32/tokenizer.json +0 -0
- FP32/tokenizer_config.json +192 -0
- INT2/config.json +44 -0
- INT2/gpt2_model.safetensors +3 -0
- INT2/int2_metadata.json +2333 -0
- INT2/special_tokens_map.json +6 -0
- INT2/tokenizer.json +0 -0
- INT2/tokenizer_config.json +192 -0
- INT4/config.json +44 -0
- INT4/gpt2_model.safetensors +3 -0
- INT4/int4_metadata.json +2333 -0
- INT4/special_tokens_map.json +6 -0
- INT4/tokenizer.json +0 -0
- INT4/tokenizer_config.json +192 -0
- INT8/config.json +44 -0
- INT8/gpt2_model.safetensors +3 -0
- INT8/int8_metadata.json +370 -0
- INT8/special_tokens_map.json +6 -0
- INT8/tokenizer.json +0 -0
- INT8/tokenizer_config.json +192 -0
BF16/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XttsGPT"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"audio_config": {
|
| 8 |
+
"mel_channels": 80,
|
| 9 |
+
"output_sample_rate": 24000,
|
| 10 |
+
"sample_rate": 22050
|
| 11 |
+
},
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "AstraMindAI/xtts2-gpt--gpt_config.XTTSGPTConfig",
|
| 14 |
+
"AutoModelForCausalLM": "AstraMindAI/xtts2-gpt--xtts2_gpt_modeling.XttsGPT",
|
| 15 |
+
"AutoTokenizer": "AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast"
|
| 16 |
+
},
|
| 17 |
+
"decoder_input_dim": 1024,
|
| 18 |
+
"enable_redaction": false,
|
| 19 |
+
"gpt_batch_size": 1,
|
| 20 |
+
"gpt_max_audio_tokens": 605,
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"kv_cache": true,
|
| 24 |
+
"layer_norm_epsilon": 1e-05,
|
| 25 |
+
"max_audio_tokens": 605,
|
| 26 |
+
"max_prompt_tokens": 70,
|
| 27 |
+
"max_text_tokens": 402,
|
| 28 |
+
"model_type": "xtts_gpt",
|
| 29 |
+
"n_inner": 4096,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"num_audio_tokens": 1026,
|
| 32 |
+
"num_hidden_layers": 30,
|
| 33 |
+
"number_text_tokens": 6681,
|
| 34 |
+
"reorder_and_upcast_attn": false,
|
| 35 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 36 |
+
"start_audio_token": 1024,
|
| 37 |
+
"start_text_token": null,
|
| 38 |
+
"stop_audio_token": 1025,
|
| 39 |
+
"stop_text_token": null,
|
| 40 |
+
"transformers_version": "4.46.0",
|
| 41 |
+
"use_masking_gt_prompt_approach": true,
|
| 42 |
+
"use_perceiver_resampler": true,
|
| 43 |
+
"vocab_size": 6681
|
| 44 |
+
}
|
BF16/gpt2_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e23ffef8c95bd7d046155ff003a6eb30f38be57ccae1197f42301520d45252d4
|
| 3 |
+
size 761266620
|
BF16/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[START]",
|
| 3 |
+
"eos_token": "[STOP]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
BF16/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
BF16/tokenizer_config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[STOP]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SPACE]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"259": {
|
| 28 |
+
"content": "[en]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"260": {
|
| 36 |
+
"content": "[de]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"261": {
|
| 44 |
+
"content": "[START]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"262": {
|
| 52 |
+
"content": "[fr]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"267": {
|
| 60 |
+
"content": "[ru]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"284": {
|
| 68 |
+
"content": "[es]",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"285": {
|
| 76 |
+
"content": "[it]",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"286": {
|
| 84 |
+
"content": "[pt]",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"293": {
|
| 92 |
+
"content": "[cs]",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"294": {
|
| 100 |
+
"content": "[pl]",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"295": {
|
| 108 |
+
"content": "[tr]",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"297": {
|
| 116 |
+
"content": "[nl]",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"5022": {
|
| 124 |
+
"content": "[ar]",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"5023": {
|
| 132 |
+
"content": "[zh-cn]",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"5412": {
|
| 140 |
+
"content": "[ja]",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"5753": {
|
| 148 |
+
"content": "[hu]",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"6152": {
|
| 156 |
+
"content": "[ko]",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"6680": {
|
| 164 |
+
"content": "[hi]",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"6681": {
|
| 172 |
+
"content": "[PAD]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"auto_map": {"AutoTokenizer": ["AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast", null]},
|
| 181 |
+
"bos_token": "[START]",
|
| 182 |
+
"clean_up_tokenization_spaces": true,
|
| 183 |
+
"eos_token": "[STOP]",
|
| 184 |
+
"max_length": null,
|
| 185 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 186 |
+
"pad_to_multiple_of": null,
|
| 187 |
+
"pad_token": "[PAD]",
|
| 188 |
+
"pad_token_type_id": 0,
|
| 189 |
+
"padding_side": "right",
|
| 190 |
+
"tokenizer_class": "XTTSTokenizerFast",
|
| 191 |
+
"unk_token": "[UNK]"
|
| 192 |
+
}
|
FP16/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XttsGPT"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"audio_config": {
|
| 8 |
+
"mel_channels": 80,
|
| 9 |
+
"output_sample_rate": 24000,
|
| 10 |
+
"sample_rate": 22050
|
| 11 |
+
},
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "AstraMindAI/xtts2-gpt--gpt_config.XTTSGPTConfig",
|
| 14 |
+
"AutoModelForCausalLM": "AstraMindAI/xtts2-gpt--xtts2_gpt_modeling.XttsGPT",
|
| 15 |
+
"AutoTokenizer": "AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast"
|
| 16 |
+
},
|
| 17 |
+
"decoder_input_dim": 1024,
|
| 18 |
+
"enable_redaction": false,
|
| 19 |
+
"gpt_batch_size": 1,
|
| 20 |
+
"gpt_max_audio_tokens": 605,
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"kv_cache": true,
|
| 24 |
+
"layer_norm_epsilon": 1e-05,
|
| 25 |
+
"max_audio_tokens": 605,
|
| 26 |
+
"max_prompt_tokens": 70,
|
| 27 |
+
"max_text_tokens": 402,
|
| 28 |
+
"model_type": "xtts_gpt",
|
| 29 |
+
"n_inner": 4096,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"num_audio_tokens": 1026,
|
| 32 |
+
"num_hidden_layers": 30,
|
| 33 |
+
"number_text_tokens": 6681,
|
| 34 |
+
"reorder_and_upcast_attn": false,
|
| 35 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 36 |
+
"start_audio_token": 1024,
|
| 37 |
+
"start_text_token": null,
|
| 38 |
+
"stop_audio_token": 1025,
|
| 39 |
+
"stop_text_token": null,
|
| 40 |
+
"transformers_version": "4.46.0",
|
| 41 |
+
"use_masking_gt_prompt_approach": true,
|
| 42 |
+
"use_perceiver_resampler": true,
|
| 43 |
+
"vocab_size": 6681
|
| 44 |
+
}
|
FP16/gpt2_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:61f43220971207d4527f390d2d7a4d702099e3df4525c1a3b2b4aaf378736cb1
|
| 3 |
+
size 761266252
|
FP16/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[START]",
|
| 3 |
+
"eos_token": "[STOP]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
FP16/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
FP16/tokenizer_config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[STOP]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SPACE]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"259": {
|
| 28 |
+
"content": "[en]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"260": {
|
| 36 |
+
"content": "[de]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"261": {
|
| 44 |
+
"content": "[START]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"262": {
|
| 52 |
+
"content": "[fr]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"267": {
|
| 60 |
+
"content": "[ru]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"284": {
|
| 68 |
+
"content": "[es]",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"285": {
|
| 76 |
+
"content": "[it]",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"286": {
|
| 84 |
+
"content": "[pt]",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"293": {
|
| 92 |
+
"content": "[cs]",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"294": {
|
| 100 |
+
"content": "[pl]",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"295": {
|
| 108 |
+
"content": "[tr]",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"297": {
|
| 116 |
+
"content": "[nl]",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"5022": {
|
| 124 |
+
"content": "[ar]",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"5023": {
|
| 132 |
+
"content": "[zh-cn]",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"5412": {
|
| 140 |
+
"content": "[ja]",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"5753": {
|
| 148 |
+
"content": "[hu]",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"6152": {
|
| 156 |
+
"content": "[ko]",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"6680": {
|
| 164 |
+
"content": "[hi]",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"6681": {
|
| 172 |
+
"content": "[PAD]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"auto_map": {"AutoTokenizer": ["AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast", null]},
|
| 181 |
+
"bos_token": "[START]",
|
| 182 |
+
"clean_up_tokenization_spaces": true,
|
| 183 |
+
"eos_token": "[STOP]",
|
| 184 |
+
"max_length": null,
|
| 185 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 186 |
+
"pad_to_multiple_of": null,
|
| 187 |
+
"pad_token": "[PAD]",
|
| 188 |
+
"pad_token_type_id": 0,
|
| 189 |
+
"padding_side": "right",
|
| 190 |
+
"tokenizer_class": "XTTSTokenizerFast",
|
| 191 |
+
"unk_token": "[UNK]"
|
| 192 |
+
}
|
FP32/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XttsGPT"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"audio_config": {
|
| 8 |
+
"mel_channels": 80,
|
| 9 |
+
"output_sample_rate": 24000,
|
| 10 |
+
"sample_rate": 22050
|
| 11 |
+
},
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "AstraMindAI/xtts2-gpt--gpt_config.XTTSGPTConfig",
|
| 14 |
+
"AutoModelForCausalLM": "AstraMindAI/xtts2-gpt--xtts2_gpt_modeling.XttsGPT",
|
| 15 |
+
"AutoTokenizer": "AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast"
|
| 16 |
+
},
|
| 17 |
+
"decoder_input_dim": 1024,
|
| 18 |
+
"enable_redaction": false,
|
| 19 |
+
"gpt_batch_size": 1,
|
| 20 |
+
"gpt_max_audio_tokens": 605,
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"kv_cache": true,
|
| 24 |
+
"layer_norm_epsilon": 1e-05,
|
| 25 |
+
"max_audio_tokens": 605,
|
| 26 |
+
"max_prompt_tokens": 70,
|
| 27 |
+
"max_text_tokens": 402,
|
| 28 |
+
"model_type": "xtts_gpt",
|
| 29 |
+
"n_inner": 4096,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"num_audio_tokens": 1026,
|
| 32 |
+
"num_hidden_layers": 30,
|
| 33 |
+
"number_text_tokens": 6681,
|
| 34 |
+
"reorder_and_upcast_attn": false,
|
| 35 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 36 |
+
"start_audio_token": 1024,
|
| 37 |
+
"start_text_token": null,
|
| 38 |
+
"stop_audio_token": 1025,
|
| 39 |
+
"stop_text_token": null,
|
| 40 |
+
"transformers_version": "4.46.0",
|
| 41 |
+
"use_masking_gt_prompt_approach": true,
|
| 42 |
+
"use_perceiver_resampler": true,
|
| 43 |
+
"vocab_size": 6681
|
| 44 |
+
}
|
FP32/gpt2_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:104d92b2297c243b64d1417bd5cfda015faca0a670e9bc90088eed0e844f8e35
|
| 3 |
+
size 1522497936
|
FP32/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[START]",
|
| 3 |
+
"eos_token": "[STOP]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
FP32/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
FP32/tokenizer_config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[STOP]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SPACE]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"259": {
|
| 28 |
+
"content": "[en]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"260": {
|
| 36 |
+
"content": "[de]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"261": {
|
| 44 |
+
"content": "[START]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"262": {
|
| 52 |
+
"content": "[fr]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"267": {
|
| 60 |
+
"content": "[ru]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"284": {
|
| 68 |
+
"content": "[es]",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"285": {
|
| 76 |
+
"content": "[it]",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"286": {
|
| 84 |
+
"content": "[pt]",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"293": {
|
| 92 |
+
"content": "[cs]",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"294": {
|
| 100 |
+
"content": "[pl]",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"295": {
|
| 108 |
+
"content": "[tr]",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"297": {
|
| 116 |
+
"content": "[nl]",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"5022": {
|
| 124 |
+
"content": "[ar]",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"5023": {
|
| 132 |
+
"content": "[zh-cn]",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"5412": {
|
| 140 |
+
"content": "[ja]",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"5753": {
|
| 148 |
+
"content": "[hu]",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"6152": {
|
| 156 |
+
"content": "[ko]",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"6680": {
|
| 164 |
+
"content": "[hi]",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"6681": {
|
| 172 |
+
"content": "[PAD]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"auto_map": {"AutoTokenizer": ["AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast", null]},
|
| 181 |
+
"bos_token": "[START]",
|
| 182 |
+
"clean_up_tokenization_spaces": true,
|
| 183 |
+
"eos_token": "[STOP]",
|
| 184 |
+
"max_length": null,
|
| 185 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 186 |
+
"pad_to_multiple_of": null,
|
| 187 |
+
"pad_token": "[PAD]",
|
| 188 |
+
"pad_token_type_id": 0,
|
| 189 |
+
"padding_side": "right",
|
| 190 |
+
"tokenizer_class": "XTTSTokenizerFast",
|
| 191 |
+
"unk_token": "[UNK]"
|
| 192 |
+
}
|
INT2/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XttsGPT"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"audio_config": {
|
| 8 |
+
"mel_channels": 80,
|
| 9 |
+
"output_sample_rate": 24000,
|
| 10 |
+
"sample_rate": 22050
|
| 11 |
+
},
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "AstraMindAI/xtts2-gpt--gpt_config.XTTSGPTConfig",
|
| 14 |
+
"AutoModelForCausalLM": "AstraMindAI/xtts2-gpt--xtts2_gpt_modeling.XttsGPT",
|
| 15 |
+
"AutoTokenizer": "AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast"
|
| 16 |
+
},
|
| 17 |
+
"decoder_input_dim": 1024,
|
| 18 |
+
"enable_redaction": false,
|
| 19 |
+
"gpt_batch_size": 1,
|
| 20 |
+
"gpt_max_audio_tokens": 605,
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"kv_cache": true,
|
| 24 |
+
"layer_norm_epsilon": 1e-05,
|
| 25 |
+
"max_audio_tokens": 605,
|
| 26 |
+
"max_prompt_tokens": 70,
|
| 27 |
+
"max_text_tokens": 402,
|
| 28 |
+
"model_type": "xtts_gpt",
|
| 29 |
+
"n_inner": 4096,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"num_audio_tokens": 1026,
|
| 32 |
+
"num_hidden_layers": 30,
|
| 33 |
+
"number_text_tokens": 6681,
|
| 34 |
+
"reorder_and_upcast_attn": false,
|
| 35 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 36 |
+
"start_audio_token": 1024,
|
| 37 |
+
"start_text_token": null,
|
| 38 |
+
"stop_audio_token": 1025,
|
| 39 |
+
"stop_text_token": null,
|
| 40 |
+
"transformers_version": "4.46.0",
|
| 41 |
+
"use_masking_gt_prompt_approach": true,
|
| 42 |
+
"use_perceiver_resampler": true,
|
| 43 |
+
"vocab_size": 6681
|
| 44 |
+
}
|
INT2/gpt2_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dabde08132c71284df3d672aaeac4d57e898d8766f041369641fadfd8db6dea3
|
| 3 |
+
size 95187217
|
INT2/int2_metadata.json
ADDED
|
@@ -0,0 +1,2333 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"final_norm.bias": {
|
| 3 |
+
"scale": 0.6184078454971313,
|
| 4 |
+
"shape": [
|
| 5 |
+
1024
|
| 6 |
+
]
|
| 7 |
+
},
|
| 8 |
+
"final_norm.weight": {
|
| 9 |
+
"scale": 5.523232936859131,
|
| 10 |
+
"shape": [
|
| 11 |
+
1024
|
| 12 |
+
]
|
| 13 |
+
},
|
| 14 |
+
"gpt.h.0.attn.c_attn.bias": {
|
| 15 |
+
"scale": 0.8616526126861572,
|
| 16 |
+
"shape": [
|
| 17 |
+
3072
|
| 18 |
+
]
|
| 19 |
+
},
|
| 20 |
+
"gpt.h.0.attn.c_attn.weight": {
|
| 21 |
+
"scale": 0.7289496064186096,
|
| 22 |
+
"shape": [
|
| 23 |
+
1024,
|
| 24 |
+
3072
|
| 25 |
+
]
|
| 26 |
+
},
|
| 27 |
+
"gpt.h.0.attn.c_proj.bias": {
|
| 28 |
+
"scale": 0.6265745759010315,
|
| 29 |
+
"shape": [
|
| 30 |
+
1024
|
| 31 |
+
]
|
| 32 |
+
},
|
| 33 |
+
"gpt.h.0.attn.c_proj.weight": {
|
| 34 |
+
"scale": 0.8160552382469177,
|
| 35 |
+
"shape": [
|
| 36 |
+
1024,
|
| 37 |
+
1024
|
| 38 |
+
]
|
| 39 |
+
},
|
| 40 |
+
"gpt.h.0.ln_1.bias": {
|
| 41 |
+
"scale": 0.18356449902057648,
|
| 42 |
+
"shape": [
|
| 43 |
+
1024
|
| 44 |
+
]
|
| 45 |
+
},
|
| 46 |
+
"gpt.h.0.ln_1.weight": {
|
| 47 |
+
"scale": 0.24023672938346863,
|
| 48 |
+
"shape": [
|
| 49 |
+
1024
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"gpt.h.0.ln_2.bias": {
|
| 53 |
+
"scale": 1.4397389888763428,
|
| 54 |
+
"shape": [
|
| 55 |
+
1024
|
| 56 |
+
]
|
| 57 |
+
},
|
| 58 |
+
"gpt.h.0.ln_2.weight": {
|
| 59 |
+
"scale": 0.8021197319030762,
|
| 60 |
+
"shape": [
|
| 61 |
+
1024
|
| 62 |
+
]
|
| 63 |
+
},
|
| 64 |
+
"gpt.h.0.mlp.c_fc.bias": {
|
| 65 |
+
"scale": 0.3322474956512451,
|
| 66 |
+
"shape": [
|
| 67 |
+
4096
|
| 68 |
+
]
|
| 69 |
+
},
|
| 70 |
+
"gpt.h.0.mlp.c_fc.weight": {
|
| 71 |
+
"scale": 1.1929363012313843,
|
| 72 |
+
"shape": [
|
| 73 |
+
1024,
|
| 74 |
+
4096
|
| 75 |
+
]
|
| 76 |
+
},
|
| 77 |
+
"gpt.h.0.mlp.c_proj.bias": {
|
| 78 |
+
"scale": 2.0688436031341553,
|
| 79 |
+
"shape": [
|
| 80 |
+
1024
|
| 81 |
+
]
|
| 82 |
+
},
|
| 83 |
+
"gpt.h.0.mlp.c_proj.weight": {
|
| 84 |
+
"scale": 4.6349406242370605,
|
| 85 |
+
"shape": [
|
| 86 |
+
4096,
|
| 87 |
+
1024
|
| 88 |
+
]
|
| 89 |
+
},
|
| 90 |
+
"gpt.h.1.attn.c_attn.bias": {
|
| 91 |
+
"scale": 0.6426463723182678,
|
| 92 |
+
"shape": [
|
| 93 |
+
3072
|
| 94 |
+
]
|
| 95 |
+
},
|
| 96 |
+
"gpt.h.1.attn.c_attn.weight": {
|
| 97 |
+
"scale": 0.43571868538856506,
|
| 98 |
+
"shape": [
|
| 99 |
+
1024,
|
| 100 |
+
3072
|
| 101 |
+
]
|
| 102 |
+
},
|
| 103 |
+
"gpt.h.1.attn.c_proj.bias": {
|
| 104 |
+
"scale": 2.0807175636291504,
|
| 105 |
+
"shape": [
|
| 106 |
+
1024
|
| 107 |
+
]
|
| 108 |
+
},
|
| 109 |
+
"gpt.h.1.attn.c_proj.weight": {
|
| 110 |
+
"scale": 3.195742607116699,
|
| 111 |
+
"shape": [
|
| 112 |
+
1024,
|
| 113 |
+
1024
|
| 114 |
+
]
|
| 115 |
+
},
|
| 116 |
+
"gpt.h.1.ln_1.bias": {
|
| 117 |
+
"scale": 1.403233289718628,
|
| 118 |
+
"shape": [
|
| 119 |
+
1024
|
| 120 |
+
]
|
| 121 |
+
},
|
| 122 |
+
"gpt.h.1.ln_1.weight": {
|
| 123 |
+
"scale": 0.5358895659446716,
|
| 124 |
+
"shape": [
|
| 125 |
+
1024
|
| 126 |
+
]
|
| 127 |
+
},
|
| 128 |
+
"gpt.h.1.ln_2.bias": {
|
| 129 |
+
"scale": 1.513456106185913,
|
| 130 |
+
"shape": [
|
| 131 |
+
1024
|
| 132 |
+
]
|
| 133 |
+
},
|
| 134 |
+
"gpt.h.1.ln_2.weight": {
|
| 135 |
+
"scale": 1.418876051902771,
|
| 136 |
+
"shape": [
|
| 137 |
+
1024
|
| 138 |
+
]
|
| 139 |
+
},
|
| 140 |
+
"gpt.h.1.mlp.c_fc.bias": {
|
| 141 |
+
"scale": 0.185274139046669,
|
| 142 |
+
"shape": [
|
| 143 |
+
4096
|
| 144 |
+
]
|
| 145 |
+
},
|
| 146 |
+
"gpt.h.1.mlp.c_fc.weight": {
|
| 147 |
+
"scale": 0.8499930500984192,
|
| 148 |
+
"shape": [
|
| 149 |
+
1024,
|
| 150 |
+
4096
|
| 151 |
+
]
|
| 152 |
+
},
|
| 153 |
+
"gpt.h.1.mlp.c_proj.bias": {
|
| 154 |
+
"scale": 2.412322998046875,
|
| 155 |
+
"shape": [
|
| 156 |
+
1024
|
| 157 |
+
]
|
| 158 |
+
},
|
| 159 |
+
"gpt.h.1.mlp.c_proj.weight": {
|
| 160 |
+
"scale": 5.190143585205078,
|
| 161 |
+
"shape": [
|
| 162 |
+
4096,
|
| 163 |
+
1024
|
| 164 |
+
]
|
| 165 |
+
},
|
| 166 |
+
"gpt.h.10.attn.c_attn.bias": {
|
| 167 |
+
"scale": 0.30900076031684875,
|
| 168 |
+
"shape": [
|
| 169 |
+
3072
|
| 170 |
+
]
|
| 171 |
+
},
|
| 172 |
+
"gpt.h.10.attn.c_attn.weight": {
|
| 173 |
+
"scale": 0.8287355303764343,
|
| 174 |
+
"shape": [
|
| 175 |
+
1024,
|
| 176 |
+
3072
|
| 177 |
+
]
|
| 178 |
+
},
|
| 179 |
+
"gpt.h.10.attn.c_proj.bias": {
|
| 180 |
+
"scale": 0.3192511796951294,
|
| 181 |
+
"shape": [
|
| 182 |
+
1024
|
| 183 |
+
]
|
| 184 |
+
},
|
| 185 |
+
"gpt.h.10.attn.c_proj.weight": {
|
| 186 |
+
"scale": 0.6080719232559204,
|
| 187 |
+
"shape": [
|
| 188 |
+
1024,
|
| 189 |
+
1024
|
| 190 |
+
]
|
| 191 |
+
},
|
| 192 |
+
"gpt.h.10.ln_1.bias": {
|
| 193 |
+
"scale": 1.0192170143127441,
|
| 194 |
+
"shape": [
|
| 195 |
+
1024
|
| 196 |
+
]
|
| 197 |
+
},
|
| 198 |
+
"gpt.h.10.ln_1.weight": {
|
| 199 |
+
"scale": 1.0667375326156616,
|
| 200 |
+
"shape": [
|
| 201 |
+
1024
|
| 202 |
+
]
|
| 203 |
+
},
|
| 204 |
+
"gpt.h.10.ln_2.bias": {
|
| 205 |
+
"scale": 0.9129897356033325,
|
| 206 |
+
"shape": [
|
| 207 |
+
1024
|
| 208 |
+
]
|
| 209 |
+
},
|
| 210 |
+
"gpt.h.10.ln_2.weight": {
|
| 211 |
+
"scale": 1.7987780570983887,
|
| 212 |
+
"shape": [
|
| 213 |
+
1024
|
| 214 |
+
]
|
| 215 |
+
},
|
| 216 |
+
"gpt.h.10.mlp.c_fc.bias": {
|
| 217 |
+
"scale": 0.20667202770709991,
|
| 218 |
+
"shape": [
|
| 219 |
+
4096
|
| 220 |
+
]
|
| 221 |
+
},
|
| 222 |
+
"gpt.h.10.mlp.c_fc.weight": {
|
| 223 |
+
"scale": 0.6766823530197144,
|
| 224 |
+
"shape": [
|
| 225 |
+
1024,
|
| 226 |
+
4096
|
| 227 |
+
]
|
| 228 |
+
},
|
| 229 |
+
"gpt.h.10.mlp.c_proj.bias": {
|
| 230 |
+
"scale": 0.7202504873275757,
|
| 231 |
+
"shape": [
|
| 232 |
+
1024
|
| 233 |
+
]
|
| 234 |
+
},
|
| 235 |
+
"gpt.h.10.mlp.c_proj.weight": {
|
| 236 |
+
"scale": 2.558100461959839,
|
| 237 |
+
"shape": [
|
| 238 |
+
4096,
|
| 239 |
+
1024
|
| 240 |
+
]
|
| 241 |
+
},
|
| 242 |
+
"gpt.h.11.attn.c_attn.bias": {
|
| 243 |
+
"scale": 0.3394349217414856,
|
| 244 |
+
"shape": [
|
| 245 |
+
3072
|
| 246 |
+
]
|
| 247 |
+
},
|
| 248 |
+
"gpt.h.11.attn.c_attn.weight": {
|
| 249 |
+
"scale": 0.7863844037055969,
|
| 250 |
+
"shape": [
|
| 251 |
+
1024,
|
| 252 |
+
3072
|
| 253 |
+
]
|
| 254 |
+
},
|
| 255 |
+
"gpt.h.11.attn.c_proj.bias": {
|
| 256 |
+
"scale": 0.3575643002986908,
|
| 257 |
+
"shape": [
|
| 258 |
+
1024
|
| 259 |
+
]
|
| 260 |
+
},
|
| 261 |
+
"gpt.h.11.attn.c_proj.weight": {
|
| 262 |
+
"scale": 0.615801990032196,
|
| 263 |
+
"shape": [
|
| 264 |
+
1024,
|
| 265 |
+
1024
|
| 266 |
+
]
|
| 267 |
+
},
|
| 268 |
+
"gpt.h.11.ln_1.bias": {
|
| 269 |
+
"scale": 1.1525065898895264,
|
| 270 |
+
"shape": [
|
| 271 |
+
1024
|
| 272 |
+
]
|
| 273 |
+
},
|
| 274 |
+
"gpt.h.11.ln_1.weight": {
|
| 275 |
+
"scale": 1.129403829574585,
|
| 276 |
+
"shape": [
|
| 277 |
+
1024
|
| 278 |
+
]
|
| 279 |
+
},
|
| 280 |
+
"gpt.h.11.ln_2.bias": {
|
| 281 |
+
"scale": 0.6890860199928284,
|
| 282 |
+
"shape": [
|
| 283 |
+
1024
|
| 284 |
+
]
|
| 285 |
+
},
|
| 286 |
+
"gpt.h.11.ln_2.weight": {
|
| 287 |
+
"scale": 1.56197988986969,
|
| 288 |
+
"shape": [
|
| 289 |
+
1024
|
| 290 |
+
]
|
| 291 |
+
},
|
| 292 |
+
"gpt.h.11.mlp.c_fc.bias": {
|
| 293 |
+
"scale": 0.21424338221549988,
|
| 294 |
+
"shape": [
|
| 295 |
+
4096
|
| 296 |
+
]
|
| 297 |
+
},
|
| 298 |
+
"gpt.h.11.mlp.c_fc.weight": {
|
| 299 |
+
"scale": 0.4732816219329834,
|
| 300 |
+
"shape": [
|
| 301 |
+
1024,
|
| 302 |
+
4096
|
| 303 |
+
]
|
| 304 |
+
},
|
| 305 |
+
"gpt.h.11.mlp.c_proj.bias": {
|
| 306 |
+
"scale": 0.7450451850891113,
|
| 307 |
+
"shape": [
|
| 308 |
+
1024
|
| 309 |
+
]
|
| 310 |
+
},
|
| 311 |
+
"gpt.h.11.mlp.c_proj.weight": {
|
| 312 |
+
"scale": 3.203523635864258,
|
| 313 |
+
"shape": [
|
| 314 |
+
4096,
|
| 315 |
+
1024
|
| 316 |
+
]
|
| 317 |
+
},
|
| 318 |
+
"gpt.h.12.attn.c_attn.bias": {
|
| 319 |
+
"scale": 0.3494969308376312,
|
| 320 |
+
"shape": [
|
| 321 |
+
3072
|
| 322 |
+
]
|
| 323 |
+
},
|
| 324 |
+
"gpt.h.12.attn.c_attn.weight": {
|
| 325 |
+
"scale": 1.0152732133865356,
|
| 326 |
+
"shape": [
|
| 327 |
+
1024,
|
| 328 |
+
3072
|
| 329 |
+
]
|
| 330 |
+
},
|
| 331 |
+
"gpt.h.12.attn.c_proj.bias": {
|
| 332 |
+
"scale": 0.23432914912700653,
|
| 333 |
+
"shape": [
|
| 334 |
+
1024
|
| 335 |
+
]
|
| 336 |
+
},
|
| 337 |
+
"gpt.h.12.attn.c_proj.weight": {
|
| 338 |
+
"scale": 0.7732473611831665,
|
| 339 |
+
"shape": [
|
| 340 |
+
1024,
|
| 341 |
+
1024
|
| 342 |
+
]
|
| 343 |
+
},
|
| 344 |
+
"gpt.h.12.ln_1.bias": {
|
| 345 |
+
"scale": 1.3029590845108032,
|
| 346 |
+
"shape": [
|
| 347 |
+
1024
|
| 348 |
+
]
|
| 349 |
+
},
|
| 350 |
+
"gpt.h.12.ln_1.weight": {
|
| 351 |
+
"scale": 1.1806756258010864,
|
| 352 |
+
"shape": [
|
| 353 |
+
1024
|
| 354 |
+
]
|
| 355 |
+
},
|
| 356 |
+
"gpt.h.12.ln_2.bias": {
|
| 357 |
+
"scale": 0.8905816078186035,
|
| 358 |
+
"shape": [
|
| 359 |
+
1024
|
| 360 |
+
]
|
| 361 |
+
},
|
| 362 |
+
"gpt.h.12.ln_2.weight": {
|
| 363 |
+
"scale": 1.6275115013122559,
|
| 364 |
+
"shape": [
|
| 365 |
+
1024
|
| 366 |
+
]
|
| 367 |
+
},
|
| 368 |
+
"gpt.h.12.mlp.c_fc.bias": {
|
| 369 |
+
"scale": 0.10368738323450089,
|
| 370 |
+
"shape": [
|
| 371 |
+
4096
|
| 372 |
+
]
|
| 373 |
+
},
|
| 374 |
+
"gpt.h.12.mlp.c_fc.weight": {
|
| 375 |
+
"scale": 0.48194029927253723,
|
| 376 |
+
"shape": [
|
| 377 |
+
1024,
|
| 378 |
+
4096
|
| 379 |
+
]
|
| 380 |
+
},
|
| 381 |
+
"gpt.h.12.mlp.c_proj.bias": {
|
| 382 |
+
"scale": 0.8424944281578064,
|
| 383 |
+
"shape": [
|
| 384 |
+
1024
|
| 385 |
+
]
|
| 386 |
+
},
|
| 387 |
+
"gpt.h.12.mlp.c_proj.weight": {
|
| 388 |
+
"scale": 1.9203577041625977,
|
| 389 |
+
"shape": [
|
| 390 |
+
4096,
|
| 391 |
+
1024
|
| 392 |
+
]
|
| 393 |
+
},
|
| 394 |
+
"gpt.h.13.attn.c_attn.bias": {
|
| 395 |
+
"scale": 0.33320215344429016,
|
| 396 |
+
"shape": [
|
| 397 |
+
3072
|
| 398 |
+
]
|
| 399 |
+
},
|
| 400 |
+
"gpt.h.13.attn.c_attn.weight": {
|
| 401 |
+
"scale": 0.6099980473518372,
|
| 402 |
+
"shape": [
|
| 403 |
+
1024,
|
| 404 |
+
3072
|
| 405 |
+
]
|
| 406 |
+
},
|
| 407 |
+
"gpt.h.13.attn.c_proj.bias": {
|
| 408 |
+
"scale": 0.47228217124938965,
|
| 409 |
+
"shape": [
|
| 410 |
+
1024
|
| 411 |
+
]
|
| 412 |
+
},
|
| 413 |
+
"gpt.h.13.attn.c_proj.weight": {
|
| 414 |
+
"scale": 0.8291131258010864,
|
| 415 |
+
"shape": [
|
| 416 |
+
1024,
|
| 417 |
+
1024
|
| 418 |
+
]
|
| 419 |
+
},
|
| 420 |
+
"gpt.h.13.ln_1.bias": {
|
| 421 |
+
"scale": 1.0610405206680298,
|
| 422 |
+
"shape": [
|
| 423 |
+
1024
|
| 424 |
+
]
|
| 425 |
+
},
|
| 426 |
+
"gpt.h.13.ln_1.weight": {
|
| 427 |
+
"scale": 1.1475324630737305,
|
| 428 |
+
"shape": [
|
| 429 |
+
1024
|
| 430 |
+
]
|
| 431 |
+
},
|
| 432 |
+
"gpt.h.13.ln_2.bias": {
|
| 433 |
+
"scale": 0.9145744442939758,
|
| 434 |
+
"shape": [
|
| 435 |
+
1024
|
| 436 |
+
]
|
| 437 |
+
},
|
| 438 |
+
"gpt.h.13.ln_2.weight": {
|
| 439 |
+
"scale": 1.5574055910110474,
|
| 440 |
+
"shape": [
|
| 441 |
+
1024
|
| 442 |
+
]
|
| 443 |
+
},
|
| 444 |
+
"gpt.h.13.mlp.c_fc.bias": {
|
| 445 |
+
"scale": 0.13628917932510376,
|
| 446 |
+
"shape": [
|
| 447 |
+
4096
|
| 448 |
+
]
|
| 449 |
+
},
|
| 450 |
+
"gpt.h.13.mlp.c_fc.weight": {
|
| 451 |
+
"scale": 0.4501611590385437,
|
| 452 |
+
"shape": [
|
| 453 |
+
1024,
|
| 454 |
+
4096
|
| 455 |
+
]
|
| 456 |
+
},
|
| 457 |
+
"gpt.h.13.mlp.c_proj.bias": {
|
| 458 |
+
"scale": 0.7718632817268372,
|
| 459 |
+
"shape": [
|
| 460 |
+
1024
|
| 461 |
+
]
|
| 462 |
+
},
|
| 463 |
+
"gpt.h.13.mlp.c_proj.weight": {
|
| 464 |
+
"scale": 1.7728174924850464,
|
| 465 |
+
"shape": [
|
| 466 |
+
4096,
|
| 467 |
+
1024
|
| 468 |
+
]
|
| 469 |
+
},
|
| 470 |
+
"gpt.h.14.attn.c_attn.bias": {
|
| 471 |
+
"scale": 0.28137704730033875,
|
| 472 |
+
"shape": [
|
| 473 |
+
3072
|
| 474 |
+
]
|
| 475 |
+
},
|
| 476 |
+
"gpt.h.14.attn.c_attn.weight": {
|
| 477 |
+
"scale": 0.9850640892982483,
|
| 478 |
+
"shape": [
|
| 479 |
+
1024,
|
| 480 |
+
3072
|
| 481 |
+
]
|
| 482 |
+
},
|
| 483 |
+
"gpt.h.14.attn.c_proj.bias": {
|
| 484 |
+
"scale": 0.1655244529247284,
|
| 485 |
+
"shape": [
|
| 486 |
+
1024
|
| 487 |
+
]
|
| 488 |
+
},
|
| 489 |
+
"gpt.h.14.attn.c_proj.weight": {
|
| 490 |
+
"scale": 0.7112484574317932,
|
| 491 |
+
"shape": [
|
| 492 |
+
1024,
|
| 493 |
+
1024
|
| 494 |
+
]
|
| 495 |
+
},
|
| 496 |
+
"gpt.h.14.ln_1.bias": {
|
| 497 |
+
"scale": 1.1183710098266602,
|
| 498 |
+
"shape": [
|
| 499 |
+
1024
|
| 500 |
+
]
|
| 501 |
+
},
|
| 502 |
+
"gpt.h.14.ln_1.weight": {
|
| 503 |
+
"scale": 1.135362148284912,
|
| 504 |
+
"shape": [
|
| 505 |
+
1024
|
| 506 |
+
]
|
| 507 |
+
},
|
| 508 |
+
"gpt.h.14.ln_2.bias": {
|
| 509 |
+
"scale": 0.9109926819801331,
|
| 510 |
+
"shape": [
|
| 511 |
+
1024
|
| 512 |
+
]
|
| 513 |
+
},
|
| 514 |
+
"gpt.h.14.ln_2.weight": {
|
| 515 |
+
"scale": 1.4943935871124268,
|
| 516 |
+
"shape": [
|
| 517 |
+
1024
|
| 518 |
+
]
|
| 519 |
+
},
|
| 520 |
+
"gpt.h.14.mlp.c_fc.bias": {
|
| 521 |
+
"scale": 0.1072084978222847,
|
| 522 |
+
"shape": [
|
| 523 |
+
4096
|
| 524 |
+
]
|
| 525 |
+
},
|
| 526 |
+
"gpt.h.14.mlp.c_fc.weight": {
|
| 527 |
+
"scale": 0.5771117806434631,
|
| 528 |
+
"shape": [
|
| 529 |
+
1024,
|
| 530 |
+
4096
|
| 531 |
+
]
|
| 532 |
+
},
|
| 533 |
+
"gpt.h.14.mlp.c_proj.bias": {
|
| 534 |
+
"scale": 0.734090268611908,
|
| 535 |
+
"shape": [
|
| 536 |
+
1024
|
| 537 |
+
]
|
| 538 |
+
},
|
| 539 |
+
"gpt.h.14.mlp.c_proj.weight": {
|
| 540 |
+
"scale": 2.0087053775787354,
|
| 541 |
+
"shape": [
|
| 542 |
+
4096,
|
| 543 |
+
1024
|
| 544 |
+
]
|
| 545 |
+
},
|
| 546 |
+
"gpt.h.15.attn.c_attn.bias": {
|
| 547 |
+
"scale": 0.28571373224258423,
|
| 548 |
+
"shape": [
|
| 549 |
+
3072
|
| 550 |
+
]
|
| 551 |
+
},
|
| 552 |
+
"gpt.h.15.attn.c_attn.weight": {
|
| 553 |
+
"scale": 0.5473542213439941,
|
| 554 |
+
"shape": [
|
| 555 |
+
1024,
|
| 556 |
+
3072
|
| 557 |
+
]
|
| 558 |
+
},
|
| 559 |
+
"gpt.h.15.attn.c_proj.bias": {
|
| 560 |
+
"scale": 0.5944628119468689,
|
| 561 |
+
"shape": [
|
| 562 |
+
1024
|
| 563 |
+
]
|
| 564 |
+
},
|
| 565 |
+
"gpt.h.15.attn.c_proj.weight": {
|
| 566 |
+
"scale": 0.8187956213951111,
|
| 567 |
+
"shape": [
|
| 568 |
+
1024,
|
| 569 |
+
1024
|
| 570 |
+
]
|
| 571 |
+
},
|
| 572 |
+
"gpt.h.15.ln_1.bias": {
|
| 573 |
+
"scale": 1.0116039514541626,
|
| 574 |
+
"shape": [
|
| 575 |
+
1024
|
| 576 |
+
]
|
| 577 |
+
},
|
| 578 |
+
"gpt.h.15.ln_1.weight": {
|
| 579 |
+
"scale": 1.1605725288391113,
|
| 580 |
+
"shape": [
|
| 581 |
+
1024
|
| 582 |
+
]
|
| 583 |
+
},
|
| 584 |
+
"gpt.h.15.ln_2.bias": {
|
| 585 |
+
"scale": 0.7876421809196472,
|
| 586 |
+
"shape": [
|
| 587 |
+
1024
|
| 588 |
+
]
|
| 589 |
+
},
|
| 590 |
+
"gpt.h.15.ln_2.weight": {
|
| 591 |
+
"scale": 1.402604579925537,
|
| 592 |
+
"shape": [
|
| 593 |
+
1024
|
| 594 |
+
]
|
| 595 |
+
},
|
| 596 |
+
"gpt.h.15.mlp.c_fc.bias": {
|
| 597 |
+
"scale": 0.14053300023078918,
|
| 598 |
+
"shape": [
|
| 599 |
+
4096
|
| 600 |
+
]
|
| 601 |
+
},
|
| 602 |
+
"gpt.h.15.mlp.c_fc.weight": {
|
| 603 |
+
"scale": 0.5913069844245911,
|
| 604 |
+
"shape": [
|
| 605 |
+
1024,
|
| 606 |
+
4096
|
| 607 |
+
]
|
| 608 |
+
},
|
| 609 |
+
"gpt.h.15.mlp.c_proj.bias": {
|
| 610 |
+
"scale": 0.68301922082901,
|
| 611 |
+
"shape": [
|
| 612 |
+
1024
|
| 613 |
+
]
|
| 614 |
+
},
|
| 615 |
+
"gpt.h.15.mlp.c_proj.weight": {
|
| 616 |
+
"scale": 2.1755576133728027,
|
| 617 |
+
"shape": [
|
| 618 |
+
4096,
|
| 619 |
+
1024
|
| 620 |
+
]
|
| 621 |
+
},
|
| 622 |
+
"gpt.h.16.attn.c_attn.bias": {
|
| 623 |
+
"scale": 0.332042932510376,
|
| 624 |
+
"shape": [
|
| 625 |
+
3072
|
| 626 |
+
]
|
| 627 |
+
},
|
| 628 |
+
"gpt.h.16.attn.c_attn.weight": {
|
| 629 |
+
"scale": 0.8905439972877502,
|
| 630 |
+
"shape": [
|
| 631 |
+
1024,
|
| 632 |
+
3072
|
| 633 |
+
]
|
| 634 |
+
},
|
| 635 |
+
"gpt.h.16.attn.c_proj.bias": {
|
| 636 |
+
"scale": 0.6724244952201843,
|
| 637 |
+
"shape": [
|
| 638 |
+
1024
|
| 639 |
+
]
|
| 640 |
+
},
|
| 641 |
+
"gpt.h.16.attn.c_proj.weight": {
|
| 642 |
+
"scale": 0.5313841104507446,
|
| 643 |
+
"shape": [
|
| 644 |
+
1024,
|
| 645 |
+
1024
|
| 646 |
+
]
|
| 647 |
+
},
|
| 648 |
+
"gpt.h.16.ln_1.bias": {
|
| 649 |
+
"scale": 1.0996408462524414,
|
| 650 |
+
"shape": [
|
| 651 |
+
1024
|
| 652 |
+
]
|
| 653 |
+
},
|
| 654 |
+
"gpt.h.16.ln_1.weight": {
|
| 655 |
+
"scale": 1.1137791872024536,
|
| 656 |
+
"shape": [
|
| 657 |
+
1024
|
| 658 |
+
]
|
| 659 |
+
},
|
| 660 |
+
"gpt.h.16.ln_2.bias": {
|
| 661 |
+
"scale": 0.8403527736663818,
|
| 662 |
+
"shape": [
|
| 663 |
+
1024
|
| 664 |
+
]
|
| 665 |
+
},
|
| 666 |
+
"gpt.h.16.ln_2.weight": {
|
| 667 |
+
"scale": 1.2900326251983643,
|
| 668 |
+
"shape": [
|
| 669 |
+
1024
|
| 670 |
+
]
|
| 671 |
+
},
|
| 672 |
+
"gpt.h.16.mlp.c_fc.bias": {
|
| 673 |
+
"scale": 0.15540914237499237,
|
| 674 |
+
"shape": [
|
| 675 |
+
4096
|
| 676 |
+
]
|
| 677 |
+
},
|
| 678 |
+
"gpt.h.16.mlp.c_fc.weight": {
|
| 679 |
+
"scale": 0.5606194734573364,
|
| 680 |
+
"shape": [
|
| 681 |
+
1024,
|
| 682 |
+
4096
|
| 683 |
+
]
|
| 684 |
+
},
|
| 685 |
+
"gpt.h.16.mlp.c_proj.bias": {
|
| 686 |
+
"scale": 0.6842435598373413,
|
| 687 |
+
"shape": [
|
| 688 |
+
1024
|
| 689 |
+
]
|
| 690 |
+
},
|
| 691 |
+
"gpt.h.16.mlp.c_proj.weight": {
|
| 692 |
+
"scale": 1.1348406076431274,
|
| 693 |
+
"shape": [
|
| 694 |
+
4096,
|
| 695 |
+
1024
|
| 696 |
+
]
|
| 697 |
+
},
|
| 698 |
+
"gpt.h.17.attn.c_attn.bias": {
|
| 699 |
+
"scale": 0.3092385530471802,
|
| 700 |
+
"shape": [
|
| 701 |
+
3072
|
| 702 |
+
]
|
| 703 |
+
},
|
| 704 |
+
"gpt.h.17.attn.c_attn.weight": {
|
| 705 |
+
"scale": 0.9243433475494385,
|
| 706 |
+
"shape": [
|
| 707 |
+
1024,
|
| 708 |
+
3072
|
| 709 |
+
]
|
| 710 |
+
},
|
| 711 |
+
"gpt.h.17.attn.c_proj.bias": {
|
| 712 |
+
"scale": 0.4436689019203186,
|
| 713 |
+
"shape": [
|
| 714 |
+
1024
|
| 715 |
+
]
|
| 716 |
+
},
|
| 717 |
+
"gpt.h.17.attn.c_proj.weight": {
|
| 718 |
+
"scale": 0.9725503325462341,
|
| 719 |
+
"shape": [
|
| 720 |
+
1024,
|
| 721 |
+
1024
|
| 722 |
+
]
|
| 723 |
+
},
|
| 724 |
+
"gpt.h.17.ln_1.bias": {
|
| 725 |
+
"scale": 1.1076136827468872,
|
| 726 |
+
"shape": [
|
| 727 |
+
1024
|
| 728 |
+
]
|
| 729 |
+
},
|
| 730 |
+
"gpt.h.17.ln_1.weight": {
|
| 731 |
+
"scale": 1.1201558113098145,
|
| 732 |
+
"shape": [
|
| 733 |
+
1024
|
| 734 |
+
]
|
| 735 |
+
},
|
| 736 |
+
"gpt.h.17.ln_2.bias": {
|
| 737 |
+
"scale": 0.9228811860084534,
|
| 738 |
+
"shape": [
|
| 739 |
+
1024
|
| 740 |
+
]
|
| 741 |
+
},
|
| 742 |
+
"gpt.h.17.ln_2.weight": {
|
| 743 |
+
"scale": 1.3084325790405273,
|
| 744 |
+
"shape": [
|
| 745 |
+
1024
|
| 746 |
+
]
|
| 747 |
+
},
|
| 748 |
+
"gpt.h.17.mlp.c_fc.bias": {
|
| 749 |
+
"scale": 0.1736840307712555,
|
| 750 |
+
"shape": [
|
| 751 |
+
4096
|
| 752 |
+
]
|
| 753 |
+
},
|
| 754 |
+
"gpt.h.17.mlp.c_fc.weight": {
|
| 755 |
+
"scale": 0.47977033257484436,
|
| 756 |
+
"shape": [
|
| 757 |
+
1024,
|
| 758 |
+
4096
|
| 759 |
+
]
|
| 760 |
+
},
|
| 761 |
+
"gpt.h.17.mlp.c_proj.bias": {
|
| 762 |
+
"scale": 0.6184632182121277,
|
| 763 |
+
"shape": [
|
| 764 |
+
1024
|
| 765 |
+
]
|
| 766 |
+
},
|
| 767 |
+
"gpt.h.17.mlp.c_proj.weight": {
|
| 768 |
+
"scale": 1.289231538772583,
|
| 769 |
+
"shape": [
|
| 770 |
+
4096,
|
| 771 |
+
1024
|
| 772 |
+
]
|
| 773 |
+
},
|
| 774 |
+
"gpt.h.18.attn.c_attn.bias": {
|
| 775 |
+
"scale": 0.2630675435066223,
|
| 776 |
+
"shape": [
|
| 777 |
+
3072
|
| 778 |
+
]
|
| 779 |
+
},
|
| 780 |
+
"gpt.h.18.attn.c_attn.weight": {
|
| 781 |
+
"scale": 0.8577865958213806,
|
| 782 |
+
"shape": [
|
| 783 |
+
1024,
|
| 784 |
+
3072
|
| 785 |
+
]
|
| 786 |
+
},
|
| 787 |
+
"gpt.h.18.attn.c_proj.bias": {
|
| 788 |
+
"scale": 0.7874951958656311,
|
| 789 |
+
"shape": [
|
| 790 |
+
1024
|
| 791 |
+
]
|
| 792 |
+
},
|
| 793 |
+
"gpt.h.18.attn.c_proj.weight": {
|
| 794 |
+
"scale": 0.6830109357833862,
|
| 795 |
+
"shape": [
|
| 796 |
+
1024,
|
| 797 |
+
1024
|
| 798 |
+
]
|
| 799 |
+
},
|
| 800 |
+
"gpt.h.18.ln_1.bias": {
|
| 801 |
+
"scale": 1.0202974081039429,
|
| 802 |
+
"shape": [
|
| 803 |
+
1024
|
| 804 |
+
]
|
| 805 |
+
},
|
| 806 |
+
"gpt.h.18.ln_1.weight": {
|
| 807 |
+
"scale": 1.112379550933838,
|
| 808 |
+
"shape": [
|
| 809 |
+
1024
|
| 810 |
+
]
|
| 811 |
+
},
|
| 812 |
+
"gpt.h.18.ln_2.bias": {
|
| 813 |
+
"scale": 0.8227484226226807,
|
| 814 |
+
"shape": [
|
| 815 |
+
1024
|
| 816 |
+
]
|
| 817 |
+
},
|
| 818 |
+
"gpt.h.18.ln_2.weight": {
|
| 819 |
+
"scale": 1.2532376050949097,
|
| 820 |
+
"shape": [
|
| 821 |
+
1024
|
| 822 |
+
]
|
| 823 |
+
},
|
| 824 |
+
"gpt.h.18.mlp.c_fc.bias": {
|
| 825 |
+
"scale": 0.1509416699409485,
|
| 826 |
+
"shape": [
|
| 827 |
+
4096
|
| 828 |
+
]
|
| 829 |
+
},
|
| 830 |
+
"gpt.h.18.mlp.c_fc.weight": {
|
| 831 |
+
"scale": 0.47805407643318176,
|
| 832 |
+
"shape": [
|
| 833 |
+
1024,
|
| 834 |
+
4096
|
| 835 |
+
]
|
| 836 |
+
},
|
| 837 |
+
"gpt.h.18.mlp.c_proj.bias": {
|
| 838 |
+
"scale": 0.4773893356323242,
|
| 839 |
+
"shape": [
|
| 840 |
+
1024
|
| 841 |
+
]
|
| 842 |
+
},
|
| 843 |
+
"gpt.h.18.mlp.c_proj.weight": {
|
| 844 |
+
"scale": 0.8970383405685425,
|
| 845 |
+
"shape": [
|
| 846 |
+
4096,
|
| 847 |
+
1024
|
| 848 |
+
]
|
| 849 |
+
},
|
| 850 |
+
"gpt.h.19.attn.c_attn.bias": {
|
| 851 |
+
"scale": 0.32839706540107727,
|
| 852 |
+
"shape": [
|
| 853 |
+
3072
|
| 854 |
+
]
|
| 855 |
+
},
|
| 856 |
+
"gpt.h.19.attn.c_attn.weight": {
|
| 857 |
+
"scale": 0.8178861737251282,
|
| 858 |
+
"shape": [
|
| 859 |
+
1024,
|
| 860 |
+
3072
|
| 861 |
+
]
|
| 862 |
+
},
|
| 863 |
+
"gpt.h.19.attn.c_proj.bias": {
|
| 864 |
+
"scale": 0.41575977206230164,
|
| 865 |
+
"shape": [
|
| 866 |
+
1024
|
| 867 |
+
]
|
| 868 |
+
},
|
| 869 |
+
"gpt.h.19.attn.c_proj.weight": {
|
| 870 |
+
"scale": 0.8766708970069885,
|
| 871 |
+
"shape": [
|
| 872 |
+
1024,
|
| 873 |
+
1024
|
| 874 |
+
]
|
| 875 |
+
},
|
| 876 |
+
"gpt.h.19.ln_1.bias": {
|
| 877 |
+
"scale": 1.034698724746704,
|
| 878 |
+
"shape": [
|
| 879 |
+
1024
|
| 880 |
+
]
|
| 881 |
+
},
|
| 882 |
+
"gpt.h.19.ln_1.weight": {
|
| 883 |
+
"scale": 1.092841625213623,
|
| 884 |
+
"shape": [
|
| 885 |
+
1024
|
| 886 |
+
]
|
| 887 |
+
},
|
| 888 |
+
"gpt.h.19.ln_2.bias": {
|
| 889 |
+
"scale": 0.7605751156806946,
|
| 890 |
+
"shape": [
|
| 891 |
+
1024
|
| 892 |
+
]
|
| 893 |
+
},
|
| 894 |
+
"gpt.h.19.ln_2.weight": {
|
| 895 |
+
"scale": 1.2519071102142334,
|
| 896 |
+
"shape": [
|
| 897 |
+
1024
|
| 898 |
+
]
|
| 899 |
+
},
|
| 900 |
+
"gpt.h.19.mlp.c_fc.bias": {
|
| 901 |
+
"scale": 0.16412009298801422,
|
| 902 |
+
"shape": [
|
| 903 |
+
4096
|
| 904 |
+
]
|
| 905 |
+
},
|
| 906 |
+
"gpt.h.19.mlp.c_fc.weight": {
|
| 907 |
+
"scale": 0.5440301895141602,
|
| 908 |
+
"shape": [
|
| 909 |
+
1024,
|
| 910 |
+
4096
|
| 911 |
+
]
|
| 912 |
+
},
|
| 913 |
+
"gpt.h.19.mlp.c_proj.bias": {
|
| 914 |
+
"scale": 0.41071999073028564,
|
| 915 |
+
"shape": [
|
| 916 |
+
1024
|
| 917 |
+
]
|
| 918 |
+
},
|
| 919 |
+
"gpt.h.19.mlp.c_proj.weight": {
|
| 920 |
+
"scale": 1.0663466453552246,
|
| 921 |
+
"shape": [
|
| 922 |
+
4096,
|
| 923 |
+
1024
|
| 924 |
+
]
|
| 925 |
+
},
|
| 926 |
+
"gpt.h.2.attn.c_attn.bias": {
|
| 927 |
+
"scale": 0.31375154852867126,
|
| 928 |
+
"shape": [
|
| 929 |
+
3072
|
| 930 |
+
]
|
| 931 |
+
},
|
| 932 |
+
"gpt.h.2.attn.c_attn.weight": {
|
| 933 |
+
"scale": 0.43119457364082336,
|
| 934 |
+
"shape": [
|
| 935 |
+
1024,
|
| 936 |
+
3072
|
| 937 |
+
]
|
| 938 |
+
},
|
| 939 |
+
"gpt.h.2.attn.c_proj.bias": {
|
| 940 |
+
"scale": 2.6263744831085205,
|
| 941 |
+
"shape": [
|
| 942 |
+
1024
|
| 943 |
+
]
|
| 944 |
+
},
|
| 945 |
+
"gpt.h.2.attn.c_proj.weight": {
|
| 946 |
+
"scale": 3.013683319091797,
|
| 947 |
+
"shape": [
|
| 948 |
+
1024,
|
| 949 |
+
1024
|
| 950 |
+
]
|
| 951 |
+
},
|
| 952 |
+
"gpt.h.2.ln_1.bias": {
|
| 953 |
+
"scale": 1.1809124946594238,
|
| 954 |
+
"shape": [
|
| 955 |
+
1024
|
| 956 |
+
]
|
| 957 |
+
},
|
| 958 |
+
"gpt.h.2.ln_1.weight": {
|
| 959 |
+
"scale": 0.6806290745735168,
|
| 960 |
+
"shape": [
|
| 961 |
+
1024
|
| 962 |
+
]
|
| 963 |
+
},
|
| 964 |
+
"gpt.h.2.ln_2.bias": {
|
| 965 |
+
"scale": 1.0246816873550415,
|
| 966 |
+
"shape": [
|
| 967 |
+
1024
|
| 968 |
+
]
|
| 969 |
+
},
|
| 970 |
+
"gpt.h.2.ln_2.weight": {
|
| 971 |
+
"scale": 1.704537034034729,
|
| 972 |
+
"shape": [
|
| 973 |
+
1024
|
| 974 |
+
]
|
| 975 |
+
},
|
| 976 |
+
"gpt.h.2.mlp.c_fc.bias": {
|
| 977 |
+
"scale": 0.3829772174358368,
|
| 978 |
+
"shape": [
|
| 979 |
+
4096
|
| 980 |
+
]
|
| 981 |
+
},
|
| 982 |
+
"gpt.h.2.mlp.c_fc.weight": {
|
| 983 |
+
"scale": 1.365037441253662,
|
| 984 |
+
"shape": [
|
| 985 |
+
1024,
|
| 986 |
+
4096
|
| 987 |
+
]
|
| 988 |
+
},
|
| 989 |
+
"gpt.h.2.mlp.c_proj.bias": {
|
| 990 |
+
"scale": 1.9403208494186401,
|
| 991 |
+
"shape": [
|
| 992 |
+
1024
|
| 993 |
+
]
|
| 994 |
+
},
|
| 995 |
+
"gpt.h.2.mlp.c_proj.weight": {
|
| 996 |
+
"scale": 5.108565807342529,
|
| 997 |
+
"shape": [
|
| 998 |
+
4096,
|
| 999 |
+
1024
|
| 1000 |
+
]
|
| 1001 |
+
},
|
| 1002 |
+
"gpt.h.20.attn.c_attn.bias": {
|
| 1003 |
+
"scale": 0.2516125738620758,
|
| 1004 |
+
"shape": [
|
| 1005 |
+
3072
|
| 1006 |
+
]
|
| 1007 |
+
},
|
| 1008 |
+
"gpt.h.20.attn.c_attn.weight": {
|
| 1009 |
+
"scale": 0.870423436164856,
|
| 1010 |
+
"shape": [
|
| 1011 |
+
1024,
|
| 1012 |
+
3072
|
| 1013 |
+
]
|
| 1014 |
+
},
|
| 1015 |
+
"gpt.h.20.attn.c_proj.bias": {
|
| 1016 |
+
"scale": 0.6339138150215149,
|
| 1017 |
+
"shape": [
|
| 1018 |
+
1024
|
| 1019 |
+
]
|
| 1020 |
+
},
|
| 1021 |
+
"gpt.h.20.attn.c_proj.weight": {
|
| 1022 |
+
"scale": 0.7554279565811157,
|
| 1023 |
+
"shape": [
|
| 1024 |
+
1024,
|
| 1025 |
+
1024
|
| 1026 |
+
]
|
| 1027 |
+
},
|
| 1028 |
+
"gpt.h.20.ln_1.bias": {
|
| 1029 |
+
"scale": 0.9357683062553406,
|
| 1030 |
+
"shape": [
|
| 1031 |
+
1024
|
| 1032 |
+
]
|
| 1033 |
+
},
|
| 1034 |
+
"gpt.h.20.ln_1.weight": {
|
| 1035 |
+
"scale": 1.0969212055206299,
|
| 1036 |
+
"shape": [
|
| 1037 |
+
1024
|
| 1038 |
+
]
|
| 1039 |
+
},
|
| 1040 |
+
"gpt.h.20.ln_2.bias": {
|
| 1041 |
+
"scale": 0.7808690071105957,
|
| 1042 |
+
"shape": [
|
| 1043 |
+
1024
|
| 1044 |
+
]
|
| 1045 |
+
},
|
| 1046 |
+
"gpt.h.20.ln_2.weight": {
|
| 1047 |
+
"scale": 1.1708141565322876,
|
| 1048 |
+
"shape": [
|
| 1049 |
+
1024
|
| 1050 |
+
]
|
| 1051 |
+
},
|
| 1052 |
+
"gpt.h.20.mlp.c_fc.bias": {
|
| 1053 |
+
"scale": 0.10677170008420944,
|
| 1054 |
+
"shape": [
|
| 1055 |
+
4096
|
| 1056 |
+
]
|
| 1057 |
+
},
|
| 1058 |
+
"gpt.h.20.mlp.c_fc.weight": {
|
| 1059 |
+
"scale": 0.45480090379714966,
|
| 1060 |
+
"shape": [
|
| 1061 |
+
1024,
|
| 1062 |
+
4096
|
| 1063 |
+
]
|
| 1064 |
+
},
|
| 1065 |
+
"gpt.h.20.mlp.c_proj.bias": {
|
| 1066 |
+
"scale": 0.30032098293304443,
|
| 1067 |
+
"shape": [
|
| 1068 |
+
1024
|
| 1069 |
+
]
|
| 1070 |
+
},
|
| 1071 |
+
"gpt.h.20.mlp.c_proj.weight": {
|
| 1072 |
+
"scale": 1.8839131593704224,
|
| 1073 |
+
"shape": [
|
| 1074 |
+
4096,
|
| 1075 |
+
1024
|
| 1076 |
+
]
|
| 1077 |
+
},
|
| 1078 |
+
"gpt.h.21.attn.c_attn.bias": {
|
| 1079 |
+
"scale": 0.3100431561470032,
|
| 1080 |
+
"shape": [
|
| 1081 |
+
3072
|
| 1082 |
+
]
|
| 1083 |
+
},
|
| 1084 |
+
"gpt.h.21.attn.c_attn.weight": {
|
| 1085 |
+
"scale": 0.7054853439331055,
|
| 1086 |
+
"shape": [
|
| 1087 |
+
1024,
|
| 1088 |
+
3072
|
| 1089 |
+
]
|
| 1090 |
+
},
|
| 1091 |
+
"gpt.h.21.attn.c_proj.bias": {
|
| 1092 |
+
"scale": 0.5535825490951538,
|
| 1093 |
+
"shape": [
|
| 1094 |
+
1024
|
| 1095 |
+
]
|
| 1096 |
+
},
|
| 1097 |
+
"gpt.h.21.attn.c_proj.weight": {
|
| 1098 |
+
"scale": 1.1769016981124878,
|
| 1099 |
+
"shape": [
|
| 1100 |
+
1024,
|
| 1101 |
+
1024
|
| 1102 |
+
]
|
| 1103 |
+
},
|
| 1104 |
+
"gpt.h.21.ln_1.bias": {
|
| 1105 |
+
"scale": 0.8685834407806396,
|
| 1106 |
+
"shape": [
|
| 1107 |
+
1024
|
| 1108 |
+
]
|
| 1109 |
+
},
|
| 1110 |
+
"gpt.h.21.ln_1.weight": {
|
| 1111 |
+
"scale": 1.1164261102676392,
|
| 1112 |
+
"shape": [
|
| 1113 |
+
1024
|
| 1114 |
+
]
|
| 1115 |
+
},
|
| 1116 |
+
"gpt.h.21.ln_2.bias": {
|
| 1117 |
+
"scale": 0.7616092562675476,
|
| 1118 |
+
"shape": [
|
| 1119 |
+
1024
|
| 1120 |
+
]
|
| 1121 |
+
},
|
| 1122 |
+
"gpt.h.21.ln_2.weight": {
|
| 1123 |
+
"scale": 1.1056314706802368,
|
| 1124 |
+
"shape": [
|
| 1125 |
+
1024
|
| 1126 |
+
]
|
| 1127 |
+
},
|
| 1128 |
+
"gpt.h.21.mlp.c_fc.bias": {
|
| 1129 |
+
"scale": 0.15077142417430878,
|
| 1130 |
+
"shape": [
|
| 1131 |
+
4096
|
| 1132 |
+
]
|
| 1133 |
+
},
|
| 1134 |
+
"gpt.h.21.mlp.c_fc.weight": {
|
| 1135 |
+
"scale": 0.4911968410015106,
|
| 1136 |
+
"shape": [
|
| 1137 |
+
1024,
|
| 1138 |
+
4096
|
| 1139 |
+
]
|
| 1140 |
+
},
|
| 1141 |
+
"gpt.h.21.mlp.c_proj.bias": {
|
| 1142 |
+
"scale": 0.1756763458251953,
|
| 1143 |
+
"shape": [
|
| 1144 |
+
1024
|
| 1145 |
+
]
|
| 1146 |
+
},
|
| 1147 |
+
"gpt.h.21.mlp.c_proj.weight": {
|
| 1148 |
+
"scale": 1.4334403276443481,
|
| 1149 |
+
"shape": [
|
| 1150 |
+
4096,
|
| 1151 |
+
1024
|
| 1152 |
+
]
|
| 1153 |
+
},
|
| 1154 |
+
"gpt.h.22.attn.c_attn.bias": {
|
| 1155 |
+
"scale": 0.2804168462753296,
|
| 1156 |
+
"shape": [
|
| 1157 |
+
3072
|
| 1158 |
+
]
|
| 1159 |
+
},
|
| 1160 |
+
"gpt.h.22.attn.c_attn.weight": {
|
| 1161 |
+
"scale": 0.4919489324092865,
|
| 1162 |
+
"shape": [
|
| 1163 |
+
1024,
|
| 1164 |
+
3072
|
| 1165 |
+
]
|
| 1166 |
+
},
|
| 1167 |
+
"gpt.h.22.attn.c_proj.bias": {
|
| 1168 |
+
"scale": 0.4737534523010254,
|
| 1169 |
+
"shape": [
|
| 1170 |
+
1024
|
| 1171 |
+
]
|
| 1172 |
+
},
|
| 1173 |
+
"gpt.h.22.attn.c_proj.weight": {
|
| 1174 |
+
"scale": 0.6300857067108154,
|
| 1175 |
+
"shape": [
|
| 1176 |
+
1024,
|
| 1177 |
+
1024
|
| 1178 |
+
]
|
| 1179 |
+
},
|
| 1180 |
+
"gpt.h.22.ln_1.bias": {
|
| 1181 |
+
"scale": 0.8146415948867798,
|
| 1182 |
+
"shape": [
|
| 1183 |
+
1024
|
| 1184 |
+
]
|
| 1185 |
+
},
|
| 1186 |
+
"gpt.h.22.ln_1.weight": {
|
| 1187 |
+
"scale": 1.0955730676651,
|
| 1188 |
+
"shape": [
|
| 1189 |
+
1024
|
| 1190 |
+
]
|
| 1191 |
+
},
|
| 1192 |
+
"gpt.h.22.ln_2.bias": {
|
| 1193 |
+
"scale": 0.8315455317497253,
|
| 1194 |
+
"shape": [
|
| 1195 |
+
1024
|
| 1196 |
+
]
|
| 1197 |
+
},
|
| 1198 |
+
"gpt.h.22.ln_2.weight": {
|
| 1199 |
+
"scale": 1.116787314414978,
|
| 1200 |
+
"shape": [
|
| 1201 |
+
1024
|
| 1202 |
+
]
|
| 1203 |
+
},
|
| 1204 |
+
"gpt.h.22.mlp.c_fc.bias": {
|
| 1205 |
+
"scale": 0.1203167513012886,
|
| 1206 |
+
"shape": [
|
| 1207 |
+
4096
|
| 1208 |
+
]
|
| 1209 |
+
},
|
| 1210 |
+
"gpt.h.22.mlp.c_fc.weight": {
|
| 1211 |
+
"scale": 0.5276402235031128,
|
| 1212 |
+
"shape": [
|
| 1213 |
+
1024,
|
| 1214 |
+
4096
|
| 1215 |
+
]
|
| 1216 |
+
},
|
| 1217 |
+
"gpt.h.22.mlp.c_proj.bias": {
|
| 1218 |
+
"scale": 0.17383702099323273,
|
| 1219 |
+
"shape": [
|
| 1220 |
+
1024
|
| 1221 |
+
]
|
| 1222 |
+
},
|
| 1223 |
+
"gpt.h.22.mlp.c_proj.weight": {
|
| 1224 |
+
"scale": 1.2244986295700073,
|
| 1225 |
+
"shape": [
|
| 1226 |
+
4096,
|
| 1227 |
+
1024
|
| 1228 |
+
]
|
| 1229 |
+
},
|
| 1230 |
+
"gpt.h.23.attn.c_attn.bias": {
|
| 1231 |
+
"scale": 0.2580159902572632,
|
| 1232 |
+
"shape": [
|
| 1233 |
+
3072
|
| 1234 |
+
]
|
| 1235 |
+
},
|
| 1236 |
+
"gpt.h.23.attn.c_attn.weight": {
|
| 1237 |
+
"scale": 0.6216667890548706,
|
| 1238 |
+
"shape": [
|
| 1239 |
+
1024,
|
| 1240 |
+
3072
|
| 1241 |
+
]
|
| 1242 |
+
},
|
| 1243 |
+
"gpt.h.23.attn.c_proj.bias": {
|
| 1244 |
+
"scale": 0.19547155499458313,
|
| 1245 |
+
"shape": [
|
| 1246 |
+
1024
|
| 1247 |
+
]
|
| 1248 |
+
},
|
| 1249 |
+
"gpt.h.23.attn.c_proj.weight": {
|
| 1250 |
+
"scale": 0.7300105690956116,
|
| 1251 |
+
"shape": [
|
| 1252 |
+
1024,
|
| 1253 |
+
1024
|
| 1254 |
+
]
|
| 1255 |
+
},
|
| 1256 |
+
"gpt.h.23.ln_1.bias": {
|
| 1257 |
+
"scale": 0.7185365557670593,
|
| 1258 |
+
"shape": [
|
| 1259 |
+
1024
|
| 1260 |
+
]
|
| 1261 |
+
},
|
| 1262 |
+
"gpt.h.23.ln_1.weight": {
|
| 1263 |
+
"scale": 1.0195420980453491,
|
| 1264 |
+
"shape": [
|
| 1265 |
+
1024
|
| 1266 |
+
]
|
| 1267 |
+
},
|
| 1268 |
+
"gpt.h.23.ln_2.bias": {
|
| 1269 |
+
"scale": 0.8253968954086304,
|
| 1270 |
+
"shape": [
|
| 1271 |
+
1024
|
| 1272 |
+
]
|
| 1273 |
+
},
|
| 1274 |
+
"gpt.h.23.ln_2.weight": {
|
| 1275 |
+
"scale": 1.0707511901855469,
|
| 1276 |
+
"shape": [
|
| 1277 |
+
1024
|
| 1278 |
+
]
|
| 1279 |
+
},
|
| 1280 |
+
"gpt.h.23.mlp.c_fc.bias": {
|
| 1281 |
+
"scale": 0.12819896638393402,
|
| 1282 |
+
"shape": [
|
| 1283 |
+
4096
|
| 1284 |
+
]
|
| 1285 |
+
},
|
| 1286 |
+
"gpt.h.23.mlp.c_fc.weight": {
|
| 1287 |
+
"scale": 0.4703911542892456,
|
| 1288 |
+
"shape": [
|
| 1289 |
+
1024,
|
| 1290 |
+
4096
|
| 1291 |
+
]
|
| 1292 |
+
},
|
| 1293 |
+
"gpt.h.23.mlp.c_proj.bias": {
|
| 1294 |
+
"scale": 0.21622495353221893,
|
| 1295 |
+
"shape": [
|
| 1296 |
+
1024
|
| 1297 |
+
]
|
| 1298 |
+
},
|
| 1299 |
+
"gpt.h.23.mlp.c_proj.weight": {
|
| 1300 |
+
"scale": 0.9520456790924072,
|
| 1301 |
+
"shape": [
|
| 1302 |
+
4096,
|
| 1303 |
+
1024
|
| 1304 |
+
]
|
| 1305 |
+
},
|
| 1306 |
+
"gpt.h.24.attn.c_attn.bias": {
|
| 1307 |
+
"scale": 0.29281240701675415,
|
| 1308 |
+
"shape": [
|
| 1309 |
+
3072
|
| 1310 |
+
]
|
| 1311 |
+
},
|
| 1312 |
+
"gpt.h.24.attn.c_attn.weight": {
|
| 1313 |
+
"scale": 0.40381887555122375,
|
| 1314 |
+
"shape": [
|
| 1315 |
+
1024,
|
| 1316 |
+
3072
|
| 1317 |
+
]
|
| 1318 |
+
},
|
| 1319 |
+
"gpt.h.24.attn.c_proj.bias": {
|
| 1320 |
+
"scale": 0.20989258587360382,
|
| 1321 |
+
"shape": [
|
| 1322 |
+
1024
|
| 1323 |
+
]
|
| 1324 |
+
},
|
| 1325 |
+
"gpt.h.24.attn.c_proj.weight": {
|
| 1326 |
+
"scale": 0.6699181795120239,
|
| 1327 |
+
"shape": [
|
| 1328 |
+
1024,
|
| 1329 |
+
1024
|
| 1330 |
+
]
|
| 1331 |
+
},
|
| 1332 |
+
"gpt.h.24.ln_1.bias": {
|
| 1333 |
+
"scale": 0.7208132743835449,
|
| 1334 |
+
"shape": [
|
| 1335 |
+
1024
|
| 1336 |
+
]
|
| 1337 |
+
},
|
| 1338 |
+
"gpt.h.24.ln_1.weight": {
|
| 1339 |
+
"scale": 1.151794195175171,
|
| 1340 |
+
"shape": [
|
| 1341 |
+
1024
|
| 1342 |
+
]
|
| 1343 |
+
},
|
| 1344 |
+
"gpt.h.24.ln_2.bias": {
|
| 1345 |
+
"scale": 0.8072383999824524,
|
| 1346 |
+
"shape": [
|
| 1347 |
+
1024
|
| 1348 |
+
]
|
| 1349 |
+
},
|
| 1350 |
+
"gpt.h.24.ln_2.weight": {
|
| 1351 |
+
"scale": 1.1724637746810913,
|
| 1352 |
+
"shape": [
|
| 1353 |
+
1024
|
| 1354 |
+
]
|
| 1355 |
+
},
|
| 1356 |
+
"gpt.h.24.mlp.c_fc.bias": {
|
| 1357 |
+
"scale": 0.1703694760799408,
|
| 1358 |
+
"shape": [
|
| 1359 |
+
4096
|
| 1360 |
+
]
|
| 1361 |
+
},
|
| 1362 |
+
"gpt.h.24.mlp.c_fc.weight": {
|
| 1363 |
+
"scale": 0.39706242084503174,
|
| 1364 |
+
"shape": [
|
| 1365 |
+
1024,
|
| 1366 |
+
4096
|
| 1367 |
+
]
|
| 1368 |
+
},
|
| 1369 |
+
"gpt.h.24.mlp.c_proj.bias": {
|
| 1370 |
+
"scale": 0.21926502883434296,
|
| 1371 |
+
"shape": [
|
| 1372 |
+
1024
|
| 1373 |
+
]
|
| 1374 |
+
},
|
| 1375 |
+
"gpt.h.24.mlp.c_proj.weight": {
|
| 1376 |
+
"scale": 0.5858592987060547,
|
| 1377 |
+
"shape": [
|
| 1378 |
+
4096,
|
| 1379 |
+
1024
|
| 1380 |
+
]
|
| 1381 |
+
},
|
| 1382 |
+
"gpt.h.25.attn.c_attn.bias": {
|
| 1383 |
+
"scale": 0.30726155638694763,
|
| 1384 |
+
"shape": [
|
| 1385 |
+
3072
|
| 1386 |
+
]
|
| 1387 |
+
},
|
| 1388 |
+
"gpt.h.25.attn.c_attn.weight": {
|
| 1389 |
+
"scale": 0.411799818277359,
|
| 1390 |
+
"shape": [
|
| 1391 |
+
1024,
|
| 1392 |
+
3072
|
| 1393 |
+
]
|
| 1394 |
+
},
|
| 1395 |
+
"gpt.h.25.attn.c_proj.bias": {
|
| 1396 |
+
"scale": 0.24399296939373016,
|
| 1397 |
+
"shape": [
|
| 1398 |
+
1024
|
| 1399 |
+
]
|
| 1400 |
+
},
|
| 1401 |
+
"gpt.h.25.attn.c_proj.weight": {
|
| 1402 |
+
"scale": 0.6184800863265991,
|
| 1403 |
+
"shape": [
|
| 1404 |
+
1024,
|
| 1405 |
+
1024
|
| 1406 |
+
]
|
| 1407 |
+
},
|
| 1408 |
+
"gpt.h.25.ln_1.bias": {
|
| 1409 |
+
"scale": 0.650389552116394,
|
| 1410 |
+
"shape": [
|
| 1411 |
+
1024
|
| 1412 |
+
]
|
| 1413 |
+
},
|
| 1414 |
+
"gpt.h.25.ln_1.weight": {
|
| 1415 |
+
"scale": 1.253990650177002,
|
| 1416 |
+
"shape": [
|
| 1417 |
+
1024
|
| 1418 |
+
]
|
| 1419 |
+
},
|
| 1420 |
+
"gpt.h.25.ln_2.bias": {
|
| 1421 |
+
"scale": 0.8235530853271484,
|
| 1422 |
+
"shape": [
|
| 1423 |
+
1024
|
| 1424 |
+
]
|
| 1425 |
+
},
|
| 1426 |
+
"gpt.h.25.ln_2.weight": {
|
| 1427 |
+
"scale": 1.1858876943588257,
|
| 1428 |
+
"shape": [
|
| 1429 |
+
1024
|
| 1430 |
+
]
|
| 1431 |
+
},
|
| 1432 |
+
"gpt.h.25.mlp.c_fc.bias": {
|
| 1433 |
+
"scale": 0.15443487465381622,
|
| 1434 |
+
"shape": [
|
| 1435 |
+
4096
|
| 1436 |
+
]
|
| 1437 |
+
},
|
| 1438 |
+
"gpt.h.25.mlp.c_fc.weight": {
|
| 1439 |
+
"scale": 0.4894944131374359,
|
| 1440 |
+
"shape": [
|
| 1441 |
+
1024,
|
| 1442 |
+
4096
|
| 1443 |
+
]
|
| 1444 |
+
},
|
| 1445 |
+
"gpt.h.25.mlp.c_proj.bias": {
|
| 1446 |
+
"scale": 0.2847861051559448,
|
| 1447 |
+
"shape": [
|
| 1448 |
+
1024
|
| 1449 |
+
]
|
| 1450 |
+
},
|
| 1451 |
+
"gpt.h.25.mlp.c_proj.weight": {
|
| 1452 |
+
"scale": 0.6402336359024048,
|
| 1453 |
+
"shape": [
|
| 1454 |
+
4096,
|
| 1455 |
+
1024
|
| 1456 |
+
]
|
| 1457 |
+
},
|
| 1458 |
+
"gpt.h.26.attn.c_attn.bias": {
|
| 1459 |
+
"scale": 0.5720527768135071,
|
| 1460 |
+
"shape": [
|
| 1461 |
+
3072
|
| 1462 |
+
]
|
| 1463 |
+
},
|
| 1464 |
+
"gpt.h.26.attn.c_attn.weight": {
|
| 1465 |
+
"scale": 0.4341934025287628,
|
| 1466 |
+
"shape": [
|
| 1467 |
+
1024,
|
| 1468 |
+
3072
|
| 1469 |
+
]
|
| 1470 |
+
},
|
| 1471 |
+
"gpt.h.26.attn.c_proj.bias": {
|
| 1472 |
+
"scale": 0.26778459548950195,
|
| 1473 |
+
"shape": [
|
| 1474 |
+
1024
|
| 1475 |
+
]
|
| 1476 |
+
},
|
| 1477 |
+
"gpt.h.26.attn.c_proj.weight": {
|
| 1478 |
+
"scale": 0.7562843561172485,
|
| 1479 |
+
"shape": [
|
| 1480 |
+
1024,
|
| 1481 |
+
1024
|
| 1482 |
+
]
|
| 1483 |
+
},
|
| 1484 |
+
"gpt.h.26.ln_1.bias": {
|
| 1485 |
+
"scale": 0.5681367516517639,
|
| 1486 |
+
"shape": [
|
| 1487 |
+
1024
|
| 1488 |
+
]
|
| 1489 |
+
},
|
| 1490 |
+
"gpt.h.26.ln_1.weight": {
|
| 1491 |
+
"scale": 1.196866750717163,
|
| 1492 |
+
"shape": [
|
| 1493 |
+
1024
|
| 1494 |
+
]
|
| 1495 |
+
},
|
| 1496 |
+
"gpt.h.26.ln_2.bias": {
|
| 1497 |
+
"scale": 0.8213993906974792,
|
| 1498 |
+
"shape": [
|
| 1499 |
+
1024
|
| 1500 |
+
]
|
| 1501 |
+
},
|
| 1502 |
+
"gpt.h.26.ln_2.weight": {
|
| 1503 |
+
"scale": 1.2492419481277466,
|
| 1504 |
+
"shape": [
|
| 1505 |
+
1024
|
| 1506 |
+
]
|
| 1507 |
+
},
|
| 1508 |
+
"gpt.h.26.mlp.c_fc.bias": {
|
| 1509 |
+
"scale": 0.22124537825584412,
|
| 1510 |
+
"shape": [
|
| 1511 |
+
4096
|
| 1512 |
+
]
|
| 1513 |
+
},
|
| 1514 |
+
"gpt.h.26.mlp.c_fc.weight": {
|
| 1515 |
+
"scale": 0.3691442310810089,
|
| 1516 |
+
"shape": [
|
| 1517 |
+
1024,
|
| 1518 |
+
4096
|
| 1519 |
+
]
|
| 1520 |
+
},
|
| 1521 |
+
"gpt.h.26.mlp.c_proj.bias": {
|
| 1522 |
+
"scale": 0.35506486892700195,
|
| 1523 |
+
"shape": [
|
| 1524 |
+
1024
|
| 1525 |
+
]
|
| 1526 |
+
},
|
| 1527 |
+
"gpt.h.26.mlp.c_proj.weight": {
|
| 1528 |
+
"scale": 1.0029659271240234,
|
| 1529 |
+
"shape": [
|
| 1530 |
+
4096,
|
| 1531 |
+
1024
|
| 1532 |
+
]
|
| 1533 |
+
},
|
| 1534 |
+
"gpt.h.27.attn.c_attn.bias": {
|
| 1535 |
+
"scale": 0.3983057737350464,
|
| 1536 |
+
"shape": [
|
| 1537 |
+
3072
|
| 1538 |
+
]
|
| 1539 |
+
},
|
| 1540 |
+
"gpt.h.27.attn.c_attn.weight": {
|
| 1541 |
+
"scale": 0.5026274919509888,
|
| 1542 |
+
"shape": [
|
| 1543 |
+
1024,
|
| 1544 |
+
3072
|
| 1545 |
+
]
|
| 1546 |
+
},
|
| 1547 |
+
"gpt.h.27.attn.c_proj.bias": {
|
| 1548 |
+
"scale": 0.37028607726097107,
|
| 1549 |
+
"shape": [
|
| 1550 |
+
1024
|
| 1551 |
+
]
|
| 1552 |
+
},
|
| 1553 |
+
"gpt.h.27.attn.c_proj.weight": {
|
| 1554 |
+
"scale": 0.6866351962089539,
|
| 1555 |
+
"shape": [
|
| 1556 |
+
1024,
|
| 1557 |
+
1024
|
| 1558 |
+
]
|
| 1559 |
+
},
|
| 1560 |
+
"gpt.h.27.ln_1.bias": {
|
| 1561 |
+
"scale": 0.5788735747337341,
|
| 1562 |
+
"shape": [
|
| 1563 |
+
1024
|
| 1564 |
+
]
|
| 1565 |
+
},
|
| 1566 |
+
"gpt.h.27.ln_1.weight": {
|
| 1567 |
+
"scale": 1.2025846242904663,
|
| 1568 |
+
"shape": [
|
| 1569 |
+
1024
|
| 1570 |
+
]
|
| 1571 |
+
},
|
| 1572 |
+
"gpt.h.27.ln_2.bias": {
|
| 1573 |
+
"scale": 0.8350062370300293,
|
| 1574 |
+
"shape": [
|
| 1575 |
+
1024
|
| 1576 |
+
]
|
| 1577 |
+
},
|
| 1578 |
+
"gpt.h.27.ln_2.weight": {
|
| 1579 |
+
"scale": 1.2651458978652954,
|
| 1580 |
+
"shape": [
|
| 1581 |
+
1024
|
| 1582 |
+
]
|
| 1583 |
+
},
|
| 1584 |
+
"gpt.h.27.mlp.c_fc.bias": {
|
| 1585 |
+
"scale": 0.2119043618440628,
|
| 1586 |
+
"shape": [
|
| 1587 |
+
4096
|
| 1588 |
+
]
|
| 1589 |
+
},
|
| 1590 |
+
"gpt.h.27.mlp.c_fc.weight": {
|
| 1591 |
+
"scale": 0.37953221797943115,
|
| 1592 |
+
"shape": [
|
| 1593 |
+
1024,
|
| 1594 |
+
4096
|
| 1595 |
+
]
|
| 1596 |
+
},
|
| 1597 |
+
"gpt.h.27.mlp.c_proj.bias": {
|
| 1598 |
+
"scale": 0.3048810064792633,
|
| 1599 |
+
"shape": [
|
| 1600 |
+
1024
|
| 1601 |
+
]
|
| 1602 |
+
},
|
| 1603 |
+
"gpt.h.27.mlp.c_proj.weight": {
|
| 1604 |
+
"scale": 1.0571913719177246,
|
| 1605 |
+
"shape": [
|
| 1606 |
+
4096,
|
| 1607 |
+
1024
|
| 1608 |
+
]
|
| 1609 |
+
},
|
| 1610 |
+
"gpt.h.28.attn.c_attn.bias": {
|
| 1611 |
+
"scale": 0.3161656856536865,
|
| 1612 |
+
"shape": [
|
| 1613 |
+
3072
|
| 1614 |
+
]
|
| 1615 |
+
},
|
| 1616 |
+
"gpt.h.28.attn.c_attn.weight": {
|
| 1617 |
+
"scale": 0.39138245582580566,
|
| 1618 |
+
"shape": [
|
| 1619 |
+
1024,
|
| 1620 |
+
3072
|
| 1621 |
+
]
|
| 1622 |
+
},
|
| 1623 |
+
"gpt.h.28.attn.c_proj.bias": {
|
| 1624 |
+
"scale": 0.5387474894523621,
|
| 1625 |
+
"shape": [
|
| 1626 |
+
1024
|
| 1627 |
+
]
|
| 1628 |
+
},
|
| 1629 |
+
"gpt.h.28.attn.c_proj.weight": {
|
| 1630 |
+
"scale": 1.1945445537567139,
|
| 1631 |
+
"shape": [
|
| 1632 |
+
1024,
|
| 1633 |
+
1024
|
| 1634 |
+
]
|
| 1635 |
+
},
|
| 1636 |
+
"gpt.h.28.ln_1.bias": {
|
| 1637 |
+
"scale": 0.6207168102264404,
|
| 1638 |
+
"shape": [
|
| 1639 |
+
1024
|
| 1640 |
+
]
|
| 1641 |
+
},
|
| 1642 |
+
"gpt.h.28.ln_1.weight": {
|
| 1643 |
+
"scale": 1.2662208080291748,
|
| 1644 |
+
"shape": [
|
| 1645 |
+
1024
|
| 1646 |
+
]
|
| 1647 |
+
},
|
| 1648 |
+
"gpt.h.28.ln_2.bias": {
|
| 1649 |
+
"scale": 0.7384198307991028,
|
| 1650 |
+
"shape": [
|
| 1651 |
+
1024
|
| 1652 |
+
]
|
| 1653 |
+
},
|
| 1654 |
+
"gpt.h.28.ln_2.weight": {
|
| 1655 |
+
"scale": 1.3807166814804077,
|
| 1656 |
+
"shape": [
|
| 1657 |
+
1024
|
| 1658 |
+
]
|
| 1659 |
+
},
|
| 1660 |
+
"gpt.h.28.mlp.c_fc.bias": {
|
| 1661 |
+
"scale": 0.25197261571884155,
|
| 1662 |
+
"shape": [
|
| 1663 |
+
4096
|
| 1664 |
+
]
|
| 1665 |
+
},
|
| 1666 |
+
"gpt.h.28.mlp.c_fc.weight": {
|
| 1667 |
+
"scale": 1.0825285911560059,
|
| 1668 |
+
"shape": [
|
| 1669 |
+
1024,
|
| 1670 |
+
4096
|
| 1671 |
+
]
|
| 1672 |
+
},
|
| 1673 |
+
"gpt.h.28.mlp.c_proj.bias": {
|
| 1674 |
+
"scale": 0.7067692875862122,
|
| 1675 |
+
"shape": [
|
| 1676 |
+
1024
|
| 1677 |
+
]
|
| 1678 |
+
},
|
| 1679 |
+
"gpt.h.28.mlp.c_proj.weight": {
|
| 1680 |
+
"scale": 4.0657830238342285,
|
| 1681 |
+
"shape": [
|
| 1682 |
+
4096,
|
| 1683 |
+
1024
|
| 1684 |
+
]
|
| 1685 |
+
},
|
| 1686 |
+
"gpt.h.29.attn.c_attn.bias": {
|
| 1687 |
+
"scale": 0.4292069375514984,
|
| 1688 |
+
"shape": [
|
| 1689 |
+
3072
|
| 1690 |
+
]
|
| 1691 |
+
},
|
| 1692 |
+
"gpt.h.29.attn.c_attn.weight": {
|
| 1693 |
+
"scale": 0.5109118223190308,
|
| 1694 |
+
"shape": [
|
| 1695 |
+
1024,
|
| 1696 |
+
3072
|
| 1697 |
+
]
|
| 1698 |
+
},
|
| 1699 |
+
"gpt.h.29.attn.c_proj.bias": {
|
| 1700 |
+
"scale": 0.23389524221420288,
|
| 1701 |
+
"shape": [
|
| 1702 |
+
1024
|
| 1703 |
+
]
|
| 1704 |
+
},
|
| 1705 |
+
"gpt.h.29.attn.c_proj.weight": {
|
| 1706 |
+
"scale": 2.3270254135131836,
|
| 1707 |
+
"shape": [
|
| 1708 |
+
1024,
|
| 1709 |
+
1024
|
| 1710 |
+
]
|
| 1711 |
+
},
|
| 1712 |
+
"gpt.h.29.ln_1.bias": {
|
| 1713 |
+
"scale": 0.5844658017158508,
|
| 1714 |
+
"shape": [
|
| 1715 |
+
1024
|
| 1716 |
+
]
|
| 1717 |
+
},
|
| 1718 |
+
"gpt.h.29.ln_1.weight": {
|
| 1719 |
+
"scale": 1.2286149263381958,
|
| 1720 |
+
"shape": [
|
| 1721 |
+
1024
|
| 1722 |
+
]
|
| 1723 |
+
},
|
| 1724 |
+
"gpt.h.29.ln_2.bias": {
|
| 1725 |
+
"scale": 0.4225330948829651,
|
| 1726 |
+
"shape": [
|
| 1727 |
+
1024
|
| 1728 |
+
]
|
| 1729 |
+
},
|
| 1730 |
+
"gpt.h.29.ln_2.weight": {
|
| 1731 |
+
"scale": 1.4540377855300903,
|
| 1732 |
+
"shape": [
|
| 1733 |
+
1024
|
| 1734 |
+
]
|
| 1735 |
+
},
|
| 1736 |
+
"gpt.h.29.mlp.c_fc.bias": {
|
| 1737 |
+
"scale": 0.28030163049697876,
|
| 1738 |
+
"shape": [
|
| 1739 |
+
4096
|
| 1740 |
+
]
|
| 1741 |
+
},
|
| 1742 |
+
"gpt.h.29.mlp.c_fc.weight": {
|
| 1743 |
+
"scale": 1.8019921779632568,
|
| 1744 |
+
"shape": [
|
| 1745 |
+
1024,
|
| 1746 |
+
4096
|
| 1747 |
+
]
|
| 1748 |
+
},
|
| 1749 |
+
"gpt.h.29.mlp.c_proj.bias": {
|
| 1750 |
+
"scale": 0.5616198778152466,
|
| 1751 |
+
"shape": [
|
| 1752 |
+
1024
|
| 1753 |
+
]
|
| 1754 |
+
},
|
| 1755 |
+
"gpt.h.29.mlp.c_proj.weight": {
|
| 1756 |
+
"scale": 11.012739181518555,
|
| 1757 |
+
"shape": [
|
| 1758 |
+
4096,
|
| 1759 |
+
1024
|
| 1760 |
+
]
|
| 1761 |
+
},
|
| 1762 |
+
"gpt.h.3.attn.c_attn.bias": {
|
| 1763 |
+
"scale": 0.25713393092155457,
|
| 1764 |
+
"shape": [
|
| 1765 |
+
3072
|
| 1766 |
+
]
|
| 1767 |
+
},
|
| 1768 |
+
"gpt.h.3.attn.c_attn.weight": {
|
| 1769 |
+
"scale": 0.6324517726898193,
|
| 1770 |
+
"shape": [
|
| 1771 |
+
1024,
|
| 1772 |
+
3072
|
| 1773 |
+
]
|
| 1774 |
+
},
|
| 1775 |
+
"gpt.h.3.attn.c_proj.bias": {
|
| 1776 |
+
"scale": 2.667752265930176,
|
| 1777 |
+
"shape": [
|
| 1778 |
+
1024
|
| 1779 |
+
]
|
| 1780 |
+
},
|
| 1781 |
+
"gpt.h.3.attn.c_proj.weight": {
|
| 1782 |
+
"scale": 3.59769344329834,
|
| 1783 |
+
"shape": [
|
| 1784 |
+
1024,
|
| 1785 |
+
1024
|
| 1786 |
+
]
|
| 1787 |
+
},
|
| 1788 |
+
"gpt.h.3.ln_1.bias": {
|
| 1789 |
+
"scale": 1.02127206325531,
|
| 1790 |
+
"shape": [
|
| 1791 |
+
1024
|
| 1792 |
+
]
|
| 1793 |
+
},
|
| 1794 |
+
"gpt.h.3.ln_1.weight": {
|
| 1795 |
+
"scale": 0.8167343139648438,
|
| 1796 |
+
"shape": [
|
| 1797 |
+
1024
|
| 1798 |
+
]
|
| 1799 |
+
},
|
| 1800 |
+
"gpt.h.3.ln_2.bias": {
|
| 1801 |
+
"scale": 0.8380739092826843,
|
| 1802 |
+
"shape": [
|
| 1803 |
+
1024
|
| 1804 |
+
]
|
| 1805 |
+
},
|
| 1806 |
+
"gpt.h.3.ln_2.weight": {
|
| 1807 |
+
"scale": 1.6680330038070679,
|
| 1808 |
+
"shape": [
|
| 1809 |
+
1024
|
| 1810 |
+
]
|
| 1811 |
+
},
|
| 1812 |
+
"gpt.h.3.mlp.c_fc.bias": {
|
| 1813 |
+
"scale": 0.3283645212650299,
|
| 1814 |
+
"shape": [
|
| 1815 |
+
4096
|
| 1816 |
+
]
|
| 1817 |
+
},
|
| 1818 |
+
"gpt.h.3.mlp.c_fc.weight": {
|
| 1819 |
+
"scale": 1.1588833332061768,
|
| 1820 |
+
"shape": [
|
| 1821 |
+
1024,
|
| 1822 |
+
4096
|
| 1823 |
+
]
|
| 1824 |
+
},
|
| 1825 |
+
"gpt.h.3.mlp.c_proj.bias": {
|
| 1826 |
+
"scale": 1.6539766788482666,
|
| 1827 |
+
"shape": [
|
| 1828 |
+
1024
|
| 1829 |
+
]
|
| 1830 |
+
},
|
| 1831 |
+
"gpt.h.3.mlp.c_proj.weight": {
|
| 1832 |
+
"scale": 6.646170139312744,
|
| 1833 |
+
"shape": [
|
| 1834 |
+
4096,
|
| 1835 |
+
1024
|
| 1836 |
+
]
|
| 1837 |
+
},
|
| 1838 |
+
"gpt.h.4.attn.c_attn.bias": {
|
| 1839 |
+
"scale": 0.2780280113220215,
|
| 1840 |
+
"shape": [
|
| 1841 |
+
3072
|
| 1842 |
+
]
|
| 1843 |
+
},
|
| 1844 |
+
"gpt.h.4.attn.c_attn.weight": {
|
| 1845 |
+
"scale": 0.4671952426433563,
|
| 1846 |
+
"shape": [
|
| 1847 |
+
1024,
|
| 1848 |
+
3072
|
| 1849 |
+
]
|
| 1850 |
+
},
|
| 1851 |
+
"gpt.h.4.attn.c_proj.bias": {
|
| 1852 |
+
"scale": 2.2085044384002686,
|
| 1853 |
+
"shape": [
|
| 1854 |
+
1024
|
| 1855 |
+
]
|
| 1856 |
+
},
|
| 1857 |
+
"gpt.h.4.attn.c_proj.weight": {
|
| 1858 |
+
"scale": 2.1969192028045654,
|
| 1859 |
+
"shape": [
|
| 1860 |
+
1024,
|
| 1861 |
+
1024
|
| 1862 |
+
]
|
| 1863 |
+
},
|
| 1864 |
+
"gpt.h.4.ln_1.bias": {
|
| 1865 |
+
"scale": 0.9655510187149048,
|
| 1866 |
+
"shape": [
|
| 1867 |
+
1024
|
| 1868 |
+
]
|
| 1869 |
+
},
|
| 1870 |
+
"gpt.h.4.ln_1.weight": {
|
| 1871 |
+
"scale": 0.9420844316482544,
|
| 1872 |
+
"shape": [
|
| 1873 |
+
1024
|
| 1874 |
+
]
|
| 1875 |
+
},
|
| 1876 |
+
"gpt.h.4.ln_2.bias": {
|
| 1877 |
+
"scale": 0.600337028503418,
|
| 1878 |
+
"shape": [
|
| 1879 |
+
1024
|
| 1880 |
+
]
|
| 1881 |
+
},
|
| 1882 |
+
"gpt.h.4.ln_2.weight": {
|
| 1883 |
+
"scale": 1.6738598346710205,
|
| 1884 |
+
"shape": [
|
| 1885 |
+
1024
|
| 1886 |
+
]
|
| 1887 |
+
},
|
| 1888 |
+
"gpt.h.4.mlp.c_fc.bias": {
|
| 1889 |
+
"scale": 0.11800195276737213,
|
| 1890 |
+
"shape": [
|
| 1891 |
+
4096
|
| 1892 |
+
]
|
| 1893 |
+
},
|
| 1894 |
+
"gpt.h.4.mlp.c_fc.weight": {
|
| 1895 |
+
"scale": 0.6646812558174133,
|
| 1896 |
+
"shape": [
|
| 1897 |
+
1024,
|
| 1898 |
+
4096
|
| 1899 |
+
]
|
| 1900 |
+
},
|
| 1901 |
+
"gpt.h.4.mlp.c_proj.bias": {
|
| 1902 |
+
"scale": 0.99233478307724,
|
| 1903 |
+
"shape": [
|
| 1904 |
+
1024
|
| 1905 |
+
]
|
| 1906 |
+
},
|
| 1907 |
+
"gpt.h.4.mlp.c_proj.weight": {
|
| 1908 |
+
"scale": 7.154963493347168,
|
| 1909 |
+
"shape": [
|
| 1910 |
+
4096,
|
| 1911 |
+
1024
|
| 1912 |
+
]
|
| 1913 |
+
},
|
| 1914 |
+
"gpt.h.5.attn.c_attn.bias": {
|
| 1915 |
+
"scale": 0.2130964696407318,
|
| 1916 |
+
"shape": [
|
| 1917 |
+
3072
|
| 1918 |
+
]
|
| 1919 |
+
},
|
| 1920 |
+
"gpt.h.5.attn.c_attn.weight": {
|
| 1921 |
+
"scale": 0.3897348940372467,
|
| 1922 |
+
"shape": [
|
| 1923 |
+
1024,
|
| 1924 |
+
3072
|
| 1925 |
+
]
|
| 1926 |
+
},
|
| 1927 |
+
"gpt.h.5.attn.c_proj.bias": {
|
| 1928 |
+
"scale": 1.2391482591629028,
|
| 1929 |
+
"shape": [
|
| 1930 |
+
1024
|
| 1931 |
+
]
|
| 1932 |
+
},
|
| 1933 |
+
"gpt.h.5.attn.c_proj.weight": {
|
| 1934 |
+
"scale": 0.9562830328941345,
|
| 1935 |
+
"shape": [
|
| 1936 |
+
1024,
|
| 1937 |
+
1024
|
| 1938 |
+
]
|
| 1939 |
+
},
|
| 1940 |
+
"gpt.h.5.ln_1.bias": {
|
| 1941 |
+
"scale": 0.7676820755004883,
|
| 1942 |
+
"shape": [
|
| 1943 |
+
1024
|
| 1944 |
+
]
|
| 1945 |
+
},
|
| 1946 |
+
"gpt.h.5.ln_1.weight": {
|
| 1947 |
+
"scale": 1.2011964321136475,
|
| 1948 |
+
"shape": [
|
| 1949 |
+
1024
|
| 1950 |
+
]
|
| 1951 |
+
},
|
| 1952 |
+
"gpt.h.5.ln_2.bias": {
|
| 1953 |
+
"scale": 0.4611365795135498,
|
| 1954 |
+
"shape": [
|
| 1955 |
+
1024
|
| 1956 |
+
]
|
| 1957 |
+
},
|
| 1958 |
+
"gpt.h.5.ln_2.weight": {
|
| 1959 |
+
"scale": 1.7594345808029175,
|
| 1960 |
+
"shape": [
|
| 1961 |
+
1024
|
| 1962 |
+
]
|
| 1963 |
+
},
|
| 1964 |
+
"gpt.h.5.mlp.c_fc.bias": {
|
| 1965 |
+
"scale": 0.1486712545156479,
|
| 1966 |
+
"shape": [
|
| 1967 |
+
4096
|
| 1968 |
+
]
|
| 1969 |
+
},
|
| 1970 |
+
"gpt.h.5.mlp.c_fc.weight": {
|
| 1971 |
+
"scale": 0.575945258140564,
|
| 1972 |
+
"shape": [
|
| 1973 |
+
1024,
|
| 1974 |
+
4096
|
| 1975 |
+
]
|
| 1976 |
+
},
|
| 1977 |
+
"gpt.h.5.mlp.c_proj.bias": {
|
| 1978 |
+
"scale": 0.638988196849823,
|
| 1979 |
+
"shape": [
|
| 1980 |
+
1024
|
| 1981 |
+
]
|
| 1982 |
+
},
|
| 1983 |
+
"gpt.h.5.mlp.c_proj.weight": {
|
| 1984 |
+
"scale": 6.2721123695373535,
|
| 1985 |
+
"shape": [
|
| 1986 |
+
4096,
|
| 1987 |
+
1024
|
| 1988 |
+
]
|
| 1989 |
+
},
|
| 1990 |
+
"gpt.h.6.attn.c_attn.bias": {
|
| 1991 |
+
"scale": 0.28897368907928467,
|
| 1992 |
+
"shape": [
|
| 1993 |
+
3072
|
| 1994 |
+
]
|
| 1995 |
+
},
|
| 1996 |
+
"gpt.h.6.attn.c_attn.weight": {
|
| 1997 |
+
"scale": 0.6516239643096924,
|
| 1998 |
+
"shape": [
|
| 1999 |
+
1024,
|
| 2000 |
+
3072
|
| 2001 |
+
]
|
| 2002 |
+
},
|
| 2003 |
+
"gpt.h.6.attn.c_proj.bias": {
|
| 2004 |
+
"scale": 0.6968675255775452,
|
| 2005 |
+
"shape": [
|
| 2006 |
+
1024
|
| 2007 |
+
]
|
| 2008 |
+
},
|
| 2009 |
+
"gpt.h.6.attn.c_proj.weight": {
|
| 2010 |
+
"scale": 0.6980696320533752,
|
| 2011 |
+
"shape": [
|
| 2012 |
+
1024,
|
| 2013 |
+
1024
|
| 2014 |
+
]
|
| 2015 |
+
},
|
| 2016 |
+
"gpt.h.6.ln_1.bias": {
|
| 2017 |
+
"scale": 0.741898238658905,
|
| 2018 |
+
"shape": [
|
| 2019 |
+
1024
|
| 2020 |
+
]
|
| 2021 |
+
},
|
| 2022 |
+
"gpt.h.6.ln_1.weight": {
|
| 2023 |
+
"scale": 1.4237617254257202,
|
| 2024 |
+
"shape": [
|
| 2025 |
+
1024
|
| 2026 |
+
]
|
| 2027 |
+
},
|
| 2028 |
+
"gpt.h.6.ln_2.bias": {
|
| 2029 |
+
"scale": 0.5870022177696228,
|
| 2030 |
+
"shape": [
|
| 2031 |
+
1024
|
| 2032 |
+
]
|
| 2033 |
+
},
|
| 2034 |
+
"gpt.h.6.ln_2.weight": {
|
| 2035 |
+
"scale": 1.7807389497756958,
|
| 2036 |
+
"shape": [
|
| 2037 |
+
1024
|
| 2038 |
+
]
|
| 2039 |
+
},
|
| 2040 |
+
"gpt.h.6.mlp.c_fc.bias": {
|
| 2041 |
+
"scale": 0.18498767912387848,
|
| 2042 |
+
"shape": [
|
| 2043 |
+
4096
|
| 2044 |
+
]
|
| 2045 |
+
},
|
| 2046 |
+
"gpt.h.6.mlp.c_fc.weight": {
|
| 2047 |
+
"scale": 0.8209737539291382,
|
| 2048 |
+
"shape": [
|
| 2049 |
+
1024,
|
| 2050 |
+
4096
|
| 2051 |
+
]
|
| 2052 |
+
},
|
| 2053 |
+
"gpt.h.6.mlp.c_proj.bias": {
|
| 2054 |
+
"scale": 0.5902945399284363,
|
| 2055 |
+
"shape": [
|
| 2056 |
+
1024
|
| 2057 |
+
]
|
| 2058 |
+
},
|
| 2059 |
+
"gpt.h.6.mlp.c_proj.weight": {
|
| 2060 |
+
"scale": 6.16025972366333,
|
| 2061 |
+
"shape": [
|
| 2062 |
+
4096,
|
| 2063 |
+
1024
|
| 2064 |
+
]
|
| 2065 |
+
},
|
| 2066 |
+
"gpt.h.7.attn.c_attn.bias": {
|
| 2067 |
+
"scale": 0.3166765570640564,
|
| 2068 |
+
"shape": [
|
| 2069 |
+
3072
|
| 2070 |
+
]
|
| 2071 |
+
},
|
| 2072 |
+
"gpt.h.7.attn.c_attn.weight": {
|
| 2073 |
+
"scale": 0.6890222430229187,
|
| 2074 |
+
"shape": [
|
| 2075 |
+
1024,
|
| 2076 |
+
3072
|
| 2077 |
+
]
|
| 2078 |
+
},
|
| 2079 |
+
"gpt.h.7.attn.c_proj.bias": {
|
| 2080 |
+
"scale": 0.63676917552948,
|
| 2081 |
+
"shape": [
|
| 2082 |
+
1024
|
| 2083 |
+
]
|
| 2084 |
+
},
|
| 2085 |
+
"gpt.h.7.attn.c_proj.weight": {
|
| 2086 |
+
"scale": 0.659460186958313,
|
| 2087 |
+
"shape": [
|
| 2088 |
+
1024,
|
| 2089 |
+
1024
|
| 2090 |
+
]
|
| 2091 |
+
},
|
| 2092 |
+
"gpt.h.7.ln_1.bias": {
|
| 2093 |
+
"scale": 1.0708993673324585,
|
| 2094 |
+
"shape": [
|
| 2095 |
+
1024
|
| 2096 |
+
]
|
| 2097 |
+
},
|
| 2098 |
+
"gpt.h.7.ln_1.weight": {
|
| 2099 |
+
"scale": 1.055467963218689,
|
| 2100 |
+
"shape": [
|
| 2101 |
+
1024
|
| 2102 |
+
]
|
| 2103 |
+
},
|
| 2104 |
+
"gpt.h.7.ln_2.bias": {
|
| 2105 |
+
"scale": 0.9893343448638916,
|
| 2106 |
+
"shape": [
|
| 2107 |
+
1024
|
| 2108 |
+
]
|
| 2109 |
+
},
|
| 2110 |
+
"gpt.h.7.ln_2.weight": {
|
| 2111 |
+
"scale": 1.8035027980804443,
|
| 2112 |
+
"shape": [
|
| 2113 |
+
1024
|
| 2114 |
+
]
|
| 2115 |
+
},
|
| 2116 |
+
"gpt.h.7.mlp.c_fc.bias": {
|
| 2117 |
+
"scale": 0.2329855114221573,
|
| 2118 |
+
"shape": [
|
| 2119 |
+
4096
|
| 2120 |
+
]
|
| 2121 |
+
},
|
| 2122 |
+
"gpt.h.7.mlp.c_fc.weight": {
|
| 2123 |
+
"scale": 0.4914255440235138,
|
| 2124 |
+
"shape": [
|
| 2125 |
+
1024,
|
| 2126 |
+
4096
|
| 2127 |
+
]
|
| 2128 |
+
},
|
| 2129 |
+
"gpt.h.7.mlp.c_proj.bias": {
|
| 2130 |
+
"scale": 0.7499263882637024,
|
| 2131 |
+
"shape": [
|
| 2132 |
+
1024
|
| 2133 |
+
]
|
| 2134 |
+
},
|
| 2135 |
+
"gpt.h.7.mlp.c_proj.weight": {
|
| 2136 |
+
"scale": 2.9818549156188965,
|
| 2137 |
+
"shape": [
|
| 2138 |
+
4096,
|
| 2139 |
+
1024
|
| 2140 |
+
]
|
| 2141 |
+
},
|
| 2142 |
+
"gpt.h.8.attn.c_attn.bias": {
|
| 2143 |
+
"scale": 0.2652284502983093,
|
| 2144 |
+
"shape": [
|
| 2145 |
+
3072
|
| 2146 |
+
]
|
| 2147 |
+
},
|
| 2148 |
+
"gpt.h.8.attn.c_attn.weight": {
|
| 2149 |
+
"scale": 0.6239627599716187,
|
| 2150 |
+
"shape": [
|
| 2151 |
+
1024,
|
| 2152 |
+
3072
|
| 2153 |
+
]
|
| 2154 |
+
},
|
| 2155 |
+
"gpt.h.8.attn.c_proj.bias": {
|
| 2156 |
+
"scale": 0.6191520094871521,
|
| 2157 |
+
"shape": [
|
| 2158 |
+
1024
|
| 2159 |
+
]
|
| 2160 |
+
},
|
| 2161 |
+
"gpt.h.8.attn.c_proj.weight": {
|
| 2162 |
+
"scale": 0.6664621829986572,
|
| 2163 |
+
"shape": [
|
| 2164 |
+
1024,
|
| 2165 |
+
1024
|
| 2166 |
+
]
|
| 2167 |
+
},
|
| 2168 |
+
"gpt.h.8.ln_1.bias": {
|
| 2169 |
+
"scale": 0.8762368559837341,
|
| 2170 |
+
"shape": [
|
| 2171 |
+
1024
|
| 2172 |
+
]
|
| 2173 |
+
},
|
| 2174 |
+
"gpt.h.8.ln_1.weight": {
|
| 2175 |
+
"scale": 1.153762936592102,
|
| 2176 |
+
"shape": [
|
| 2177 |
+
1024
|
| 2178 |
+
]
|
| 2179 |
+
},
|
| 2180 |
+
"gpt.h.8.ln_2.bias": {
|
| 2181 |
+
"scale": 0.9868759512901306,
|
| 2182 |
+
"shape": [
|
| 2183 |
+
1024
|
| 2184 |
+
]
|
| 2185 |
+
},
|
| 2186 |
+
"gpt.h.8.ln_2.weight": {
|
| 2187 |
+
"scale": 1.7482178211212158,
|
| 2188 |
+
"shape": [
|
| 2189 |
+
1024
|
| 2190 |
+
]
|
| 2191 |
+
},
|
| 2192 |
+
"gpt.h.8.mlp.c_fc.bias": {
|
| 2193 |
+
"scale": 0.18552374839782715,
|
| 2194 |
+
"shape": [
|
| 2195 |
+
4096
|
| 2196 |
+
]
|
| 2197 |
+
},
|
| 2198 |
+
"gpt.h.8.mlp.c_fc.weight": {
|
| 2199 |
+
"scale": 0.47477778792381287,
|
| 2200 |
+
"shape": [
|
| 2201 |
+
1024,
|
| 2202 |
+
4096
|
| 2203 |
+
]
|
| 2204 |
+
},
|
| 2205 |
+
"gpt.h.8.mlp.c_proj.bias": {
|
| 2206 |
+
"scale": 0.6799322962760925,
|
| 2207 |
+
"shape": [
|
| 2208 |
+
1024
|
| 2209 |
+
]
|
| 2210 |
+
},
|
| 2211 |
+
"gpt.h.8.mlp.c_proj.weight": {
|
| 2212 |
+
"scale": 3.0612871646881104,
|
| 2213 |
+
"shape": [
|
| 2214 |
+
4096,
|
| 2215 |
+
1024
|
| 2216 |
+
]
|
| 2217 |
+
},
|
| 2218 |
+
"gpt.h.9.attn.c_attn.bias": {
|
| 2219 |
+
"scale": 0.309771865606308,
|
| 2220 |
+
"shape": [
|
| 2221 |
+
3072
|
| 2222 |
+
]
|
| 2223 |
+
},
|
| 2224 |
+
"gpt.h.9.attn.c_attn.weight": {
|
| 2225 |
+
"scale": 0.8271536231040955,
|
| 2226 |
+
"shape": [
|
| 2227 |
+
1024,
|
| 2228 |
+
3072
|
| 2229 |
+
]
|
| 2230 |
+
},
|
| 2231 |
+
"gpt.h.9.attn.c_proj.bias": {
|
| 2232 |
+
"scale": 0.49563321471214294,
|
| 2233 |
+
"shape": [
|
| 2234 |
+
1024
|
| 2235 |
+
]
|
| 2236 |
+
},
|
| 2237 |
+
"gpt.h.9.attn.c_proj.weight": {
|
| 2238 |
+
"scale": 0.7561616897583008,
|
| 2239 |
+
"shape": [
|
| 2240 |
+
1024,
|
| 2241 |
+
1024
|
| 2242 |
+
]
|
| 2243 |
+
},
|
| 2244 |
+
"gpt.h.9.ln_1.bias": {
|
| 2245 |
+
"scale": 1.0751368999481201,
|
| 2246 |
+
"shape": [
|
| 2247 |
+
1024
|
| 2248 |
+
]
|
| 2249 |
+
},
|
| 2250 |
+
"gpt.h.9.ln_1.weight": {
|
| 2251 |
+
"scale": 1.0969300270080566,
|
| 2252 |
+
"shape": [
|
| 2253 |
+
1024
|
| 2254 |
+
]
|
| 2255 |
+
},
|
| 2256 |
+
"gpt.h.9.ln_2.bias": {
|
| 2257 |
+
"scale": 0.9641826152801514,
|
| 2258 |
+
"shape": [
|
| 2259 |
+
1024
|
| 2260 |
+
]
|
| 2261 |
+
},
|
| 2262 |
+
"gpt.h.9.ln_2.weight": {
|
| 2263 |
+
"scale": 1.803221344947815,
|
| 2264 |
+
"shape": [
|
| 2265 |
+
1024
|
| 2266 |
+
]
|
| 2267 |
+
},
|
| 2268 |
+
"gpt.h.9.mlp.c_fc.bias": {
|
| 2269 |
+
"scale": 0.22008062899112701,
|
| 2270 |
+
"shape": [
|
| 2271 |
+
4096
|
| 2272 |
+
]
|
| 2273 |
+
},
|
| 2274 |
+
"gpt.h.9.mlp.c_fc.weight": {
|
| 2275 |
+
"scale": 0.5351904630661011,
|
| 2276 |
+
"shape": [
|
| 2277 |
+
1024,
|
| 2278 |
+
4096
|
| 2279 |
+
]
|
| 2280 |
+
},
|
| 2281 |
+
"gpt.h.9.mlp.c_proj.bias": {
|
| 2282 |
+
"scale": 0.7559229731559753,
|
| 2283 |
+
"shape": [
|
| 2284 |
+
1024
|
| 2285 |
+
]
|
| 2286 |
+
},
|
| 2287 |
+
"gpt.h.9.mlp.c_proj.weight": {
|
| 2288 |
+
"scale": 3.0584769248962402,
|
| 2289 |
+
"shape": [
|
| 2290 |
+
4096,
|
| 2291 |
+
1024
|
| 2292 |
+
]
|
| 2293 |
+
},
|
| 2294 |
+
"gpt.ln_f.bias": {
|
| 2295 |
+
"scale": 1.4052708148956299,
|
| 2296 |
+
"shape": [
|
| 2297 |
+
1024
|
| 2298 |
+
]
|
| 2299 |
+
},
|
| 2300 |
+
"gpt.ln_f.weight": {
|
| 2301 |
+
"scale": 2.9010915756225586,
|
| 2302 |
+
"shape": [
|
| 2303 |
+
1024
|
| 2304 |
+
]
|
| 2305 |
+
},
|
| 2306 |
+
"gpt.wpe.emb.weight": {
|
| 2307 |
+
"scale": 0.5411291122436523,
|
| 2308 |
+
"shape": [
|
| 2309 |
+
608,
|
| 2310 |
+
1024
|
| 2311 |
+
]
|
| 2312 |
+
},
|
| 2313 |
+
"gpt.wte.weight": {
|
| 2314 |
+
"scale": 0.5614672303199768,
|
| 2315 |
+
"shape": [
|
| 2316 |
+
1026,
|
| 2317 |
+
1024
|
| 2318 |
+
]
|
| 2319 |
+
},
|
| 2320 |
+
"mel_head.bias": {
|
| 2321 |
+
"scale": 0.19914697110652924,
|
| 2322 |
+
"shape": [
|
| 2323 |
+
1026
|
| 2324 |
+
]
|
| 2325 |
+
},
|
| 2326 |
+
"mel_head.weight": {
|
| 2327 |
+
"scale": 0.5308834314346313,
|
| 2328 |
+
"shape": [
|
| 2329 |
+
1026,
|
| 2330 |
+
1024
|
| 2331 |
+
]
|
| 2332 |
+
}
|
| 2333 |
+
}
|
INT2/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[START]",
|
| 3 |
+
"eos_token": "[STOP]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
INT2/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
INT2/tokenizer_config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[STOP]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SPACE]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"259": {
|
| 28 |
+
"content": "[en]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"260": {
|
| 36 |
+
"content": "[de]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"261": {
|
| 44 |
+
"content": "[START]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"262": {
|
| 52 |
+
"content": "[fr]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"267": {
|
| 60 |
+
"content": "[ru]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"284": {
|
| 68 |
+
"content": "[es]",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"285": {
|
| 76 |
+
"content": "[it]",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"286": {
|
| 84 |
+
"content": "[pt]",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"293": {
|
| 92 |
+
"content": "[cs]",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"294": {
|
| 100 |
+
"content": "[pl]",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"295": {
|
| 108 |
+
"content": "[tr]",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"297": {
|
| 116 |
+
"content": "[nl]",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"5022": {
|
| 124 |
+
"content": "[ar]",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"5023": {
|
| 132 |
+
"content": "[zh-cn]",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"5412": {
|
| 140 |
+
"content": "[ja]",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"5753": {
|
| 148 |
+
"content": "[hu]",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"6152": {
|
| 156 |
+
"content": "[ko]",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"6680": {
|
| 164 |
+
"content": "[hi]",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"6681": {
|
| 172 |
+
"content": "[PAD]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"auto_map": {"AutoTokenizer": ["AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast", null]},
|
| 181 |
+
"bos_token": "[START]",
|
| 182 |
+
"clean_up_tokenization_spaces": true,
|
| 183 |
+
"eos_token": "[STOP]",
|
| 184 |
+
"max_length": null,
|
| 185 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 186 |
+
"pad_to_multiple_of": null,
|
| 187 |
+
"pad_token": "[PAD]",
|
| 188 |
+
"pad_token_type_id": 0,
|
| 189 |
+
"padding_side": "right",
|
| 190 |
+
"tokenizer_class": "XTTSTokenizerFast",
|
| 191 |
+
"unk_token": "[UNK]"
|
| 192 |
+
}
|
INT4/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XttsGPT"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"audio_config": {
|
| 8 |
+
"mel_channels": 80,
|
| 9 |
+
"output_sample_rate": 24000,
|
| 10 |
+
"sample_rate": 22050
|
| 11 |
+
},
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "AstraMindAI/xtts2-gpt--gpt_config.XTTSGPTConfig",
|
| 14 |
+
"AutoModelForCausalLM": "AstraMindAI/xtts2-gpt--xtts2_gpt_modeling.XttsGPT",
|
| 15 |
+
"AutoTokenizer": "AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast"
|
| 16 |
+
},
|
| 17 |
+
"decoder_input_dim": 1024,
|
| 18 |
+
"enable_redaction": false,
|
| 19 |
+
"gpt_batch_size": 1,
|
| 20 |
+
"gpt_max_audio_tokens": 605,
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"kv_cache": true,
|
| 24 |
+
"layer_norm_epsilon": 1e-05,
|
| 25 |
+
"max_audio_tokens": 605,
|
| 26 |
+
"max_prompt_tokens": 70,
|
| 27 |
+
"max_text_tokens": 402,
|
| 28 |
+
"model_type": "xtts_gpt",
|
| 29 |
+
"n_inner": 4096,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"num_audio_tokens": 1026,
|
| 32 |
+
"num_hidden_layers": 30,
|
| 33 |
+
"number_text_tokens": 6681,
|
| 34 |
+
"reorder_and_upcast_attn": false,
|
| 35 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 36 |
+
"start_audio_token": 1024,
|
| 37 |
+
"start_text_token": null,
|
| 38 |
+
"stop_audio_token": 1025,
|
| 39 |
+
"stop_text_token": null,
|
| 40 |
+
"transformers_version": "4.46.0",
|
| 41 |
+
"use_masking_gt_prompt_approach": true,
|
| 42 |
+
"use_perceiver_resampler": true,
|
| 43 |
+
"vocab_size": 6681
|
| 44 |
+
}
|
INT4/gpt2_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79002172c227f60838f8de5b1d956856f83e595f33bb03613c3c716a5529f220
|
| 3 |
+
size 190341585
|
INT4/int4_metadata.json
ADDED
|
@@ -0,0 +1,2333 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"final_norm.bias": {
|
| 3 |
+
"scale": 0.08834397792816162,
|
| 4 |
+
"shape": [
|
| 5 |
+
1024
|
| 6 |
+
]
|
| 7 |
+
},
|
| 8 |
+
"final_norm.weight": {
|
| 9 |
+
"scale": 0.7890332937240601,
|
| 10 |
+
"shape": [
|
| 11 |
+
1024
|
| 12 |
+
]
|
| 13 |
+
},
|
| 14 |
+
"gpt.h.0.attn.c_attn.bias": {
|
| 15 |
+
"scale": 0.12309323251247406,
|
| 16 |
+
"shape": [
|
| 17 |
+
3072
|
| 18 |
+
]
|
| 19 |
+
},
|
| 20 |
+
"gpt.h.0.attn.c_attn.weight": {
|
| 21 |
+
"scale": 0.1041356548666954,
|
| 22 |
+
"shape": [
|
| 23 |
+
1024,
|
| 24 |
+
3072
|
| 25 |
+
]
|
| 26 |
+
},
|
| 27 |
+
"gpt.h.0.attn.c_proj.bias": {
|
| 28 |
+
"scale": 0.08951065689325333,
|
| 29 |
+
"shape": [
|
| 30 |
+
1024
|
| 31 |
+
]
|
| 32 |
+
},
|
| 33 |
+
"gpt.h.0.attn.c_proj.weight": {
|
| 34 |
+
"scale": 0.1165793165564537,
|
| 35 |
+
"shape": [
|
| 36 |
+
1024,
|
| 37 |
+
1024
|
| 38 |
+
]
|
| 39 |
+
},
|
| 40 |
+
"gpt.h.0.ln_1.bias": {
|
| 41 |
+
"scale": 0.026223499327898026,
|
| 42 |
+
"shape": [
|
| 43 |
+
1024
|
| 44 |
+
]
|
| 45 |
+
},
|
| 46 |
+
"gpt.h.0.ln_1.weight": {
|
| 47 |
+
"scale": 0.03431953117251396,
|
| 48 |
+
"shape": [
|
| 49 |
+
1024
|
| 50 |
+
]
|
| 51 |
+
},
|
| 52 |
+
"gpt.h.0.ln_2.bias": {
|
| 53 |
+
"scale": 0.20567700266838074,
|
| 54 |
+
"shape": [
|
| 55 |
+
1024
|
| 56 |
+
]
|
| 57 |
+
},
|
| 58 |
+
"gpt.h.0.ln_2.weight": {
|
| 59 |
+
"scale": 0.11458853632211685,
|
| 60 |
+
"shape": [
|
| 61 |
+
1024
|
| 62 |
+
]
|
| 63 |
+
},
|
| 64 |
+
"gpt.h.0.mlp.c_fc.bias": {
|
| 65 |
+
"scale": 0.047463927417993546,
|
| 66 |
+
"shape": [
|
| 67 |
+
4096
|
| 68 |
+
]
|
| 69 |
+
},
|
| 70 |
+
"gpt.h.0.mlp.c_fc.weight": {
|
| 71 |
+
"scale": 0.17041946947574615,
|
| 72 |
+
"shape": [
|
| 73 |
+
1024,
|
| 74 |
+
4096
|
| 75 |
+
]
|
| 76 |
+
},
|
| 77 |
+
"gpt.h.0.mlp.c_proj.bias": {
|
| 78 |
+
"scale": 0.29554909467697144,
|
| 79 |
+
"shape": [
|
| 80 |
+
1024
|
| 81 |
+
]
|
| 82 |
+
},
|
| 83 |
+
"gpt.h.0.mlp.c_proj.weight": {
|
| 84 |
+
"scale": 0.6621343493461609,
|
| 85 |
+
"shape": [
|
| 86 |
+
4096,
|
| 87 |
+
1024
|
| 88 |
+
]
|
| 89 |
+
},
|
| 90 |
+
"gpt.h.1.attn.c_attn.bias": {
|
| 91 |
+
"scale": 0.09180662781000137,
|
| 92 |
+
"shape": [
|
| 93 |
+
3072
|
| 94 |
+
]
|
| 95 |
+
},
|
| 96 |
+
"gpt.h.1.attn.c_attn.weight": {
|
| 97 |
+
"scale": 0.06224552541971207,
|
| 98 |
+
"shape": [
|
| 99 |
+
1024,
|
| 100 |
+
3072
|
| 101 |
+
]
|
| 102 |
+
},
|
| 103 |
+
"gpt.h.1.attn.c_proj.bias": {
|
| 104 |
+
"scale": 0.2972453534603119,
|
| 105 |
+
"shape": [
|
| 106 |
+
1024
|
| 107 |
+
]
|
| 108 |
+
},
|
| 109 |
+
"gpt.h.1.attn.c_proj.weight": {
|
| 110 |
+
"scale": 0.45653465390205383,
|
| 111 |
+
"shape": [
|
| 112 |
+
1024,
|
| 113 |
+
1024
|
| 114 |
+
]
|
| 115 |
+
},
|
| 116 |
+
"gpt.h.1.ln_1.bias": {
|
| 117 |
+
"scale": 0.20046189427375793,
|
| 118 |
+
"shape": [
|
| 119 |
+
1024
|
| 120 |
+
]
|
| 121 |
+
},
|
| 122 |
+
"gpt.h.1.ln_1.weight": {
|
| 123 |
+
"scale": 0.07655565440654755,
|
| 124 |
+
"shape": [
|
| 125 |
+
1024
|
| 126 |
+
]
|
| 127 |
+
},
|
| 128 |
+
"gpt.h.1.ln_2.bias": {
|
| 129 |
+
"scale": 0.21620801091194153,
|
| 130 |
+
"shape": [
|
| 131 |
+
1024
|
| 132 |
+
]
|
| 133 |
+
},
|
| 134 |
+
"gpt.h.1.ln_2.weight": {
|
| 135 |
+
"scale": 0.20269657671451569,
|
| 136 |
+
"shape": [
|
| 137 |
+
1024
|
| 138 |
+
]
|
| 139 |
+
},
|
| 140 |
+
"gpt.h.1.mlp.c_fc.bias": {
|
| 141 |
+
"scale": 0.026467734947800636,
|
| 142 |
+
"shape": [
|
| 143 |
+
4096
|
| 144 |
+
]
|
| 145 |
+
},
|
| 146 |
+
"gpt.h.1.mlp.c_fc.weight": {
|
| 147 |
+
"scale": 0.12142758071422577,
|
| 148 |
+
"shape": [
|
| 149 |
+
1024,
|
| 150 |
+
4096
|
| 151 |
+
]
|
| 152 |
+
},
|
| 153 |
+
"gpt.h.1.mlp.c_proj.bias": {
|
| 154 |
+
"scale": 0.3446175754070282,
|
| 155 |
+
"shape": [
|
| 156 |
+
1024
|
| 157 |
+
]
|
| 158 |
+
},
|
| 159 |
+
"gpt.h.1.mlp.c_proj.weight": {
|
| 160 |
+
"scale": 0.7414490580558777,
|
| 161 |
+
"shape": [
|
| 162 |
+
4096,
|
| 163 |
+
1024
|
| 164 |
+
]
|
| 165 |
+
},
|
| 166 |
+
"gpt.h.10.attn.c_attn.bias": {
|
| 167 |
+
"scale": 0.044142965227365494,
|
| 168 |
+
"shape": [
|
| 169 |
+
3072
|
| 170 |
+
]
|
| 171 |
+
},
|
| 172 |
+
"gpt.h.10.attn.c_attn.weight": {
|
| 173 |
+
"scale": 0.11839079111814499,
|
| 174 |
+
"shape": [
|
| 175 |
+
1024,
|
| 176 |
+
3072
|
| 177 |
+
]
|
| 178 |
+
},
|
| 179 |
+
"gpt.h.10.attn.c_proj.bias": {
|
| 180 |
+
"scale": 0.0456073097884655,
|
| 181 |
+
"shape": [
|
| 182 |
+
1024
|
| 183 |
+
]
|
| 184 |
+
},
|
| 185 |
+
"gpt.h.10.attn.c_proj.weight": {
|
| 186 |
+
"scale": 0.08686741441488266,
|
| 187 |
+
"shape": [
|
| 188 |
+
1024,
|
| 189 |
+
1024
|
| 190 |
+
]
|
| 191 |
+
},
|
| 192 |
+
"gpt.h.10.ln_1.bias": {
|
| 193 |
+
"scale": 0.14560243487358093,
|
| 194 |
+
"shape": [
|
| 195 |
+
1024
|
| 196 |
+
]
|
| 197 |
+
},
|
| 198 |
+
"gpt.h.10.ln_1.weight": {
|
| 199 |
+
"scale": 0.15239107608795166,
|
| 200 |
+
"shape": [
|
| 201 |
+
1024
|
| 202 |
+
]
|
| 203 |
+
},
|
| 204 |
+
"gpt.h.10.ln_2.bias": {
|
| 205 |
+
"scale": 0.13042710721492767,
|
| 206 |
+
"shape": [
|
| 207 |
+
1024
|
| 208 |
+
]
|
| 209 |
+
},
|
| 210 |
+
"gpt.h.10.ln_2.weight": {
|
| 211 |
+
"scale": 0.25696828961372375,
|
| 212 |
+
"shape": [
|
| 213 |
+
1024
|
| 214 |
+
]
|
| 215 |
+
},
|
| 216 |
+
"gpt.h.10.mlp.c_fc.bias": {
|
| 217 |
+
"scale": 0.029524575918912888,
|
| 218 |
+
"shape": [
|
| 219 |
+
4096
|
| 220 |
+
]
|
| 221 |
+
},
|
| 222 |
+
"gpt.h.10.mlp.c_fc.weight": {
|
| 223 |
+
"scale": 0.09666890650987625,
|
| 224 |
+
"shape": [
|
| 225 |
+
1024,
|
| 226 |
+
4096
|
| 227 |
+
]
|
| 228 |
+
},
|
| 229 |
+
"gpt.h.10.mlp.c_proj.bias": {
|
| 230 |
+
"scale": 0.1028929278254509,
|
| 231 |
+
"shape": [
|
| 232 |
+
1024
|
| 233 |
+
]
|
| 234 |
+
},
|
| 235 |
+
"gpt.h.10.mlp.c_proj.weight": {
|
| 236 |
+
"scale": 0.3654429316520691,
|
| 237 |
+
"shape": [
|
| 238 |
+
4096,
|
| 239 |
+
1024
|
| 240 |
+
]
|
| 241 |
+
},
|
| 242 |
+
"gpt.h.11.attn.c_attn.bias": {
|
| 243 |
+
"scale": 0.048490703105926514,
|
| 244 |
+
"shape": [
|
| 245 |
+
3072
|
| 246 |
+
]
|
| 247 |
+
},
|
| 248 |
+
"gpt.h.11.attn.c_attn.weight": {
|
| 249 |
+
"scale": 0.11234062910079956,
|
| 250 |
+
"shape": [
|
| 251 |
+
1024,
|
| 252 |
+
3072
|
| 253 |
+
]
|
| 254 |
+
},
|
| 255 |
+
"gpt.h.11.attn.c_proj.bias": {
|
| 256 |
+
"scale": 0.0510806143283844,
|
| 257 |
+
"shape": [
|
| 258 |
+
1024
|
| 259 |
+
]
|
| 260 |
+
},
|
| 261 |
+
"gpt.h.11.attn.c_proj.weight": {
|
| 262 |
+
"scale": 0.08797170966863632,
|
| 263 |
+
"shape": [
|
| 264 |
+
1024,
|
| 265 |
+
1024
|
| 266 |
+
]
|
| 267 |
+
},
|
| 268 |
+
"gpt.h.11.ln_1.bias": {
|
| 269 |
+
"scale": 0.164643794298172,
|
| 270 |
+
"shape": [
|
| 271 |
+
1024
|
| 272 |
+
]
|
| 273 |
+
},
|
| 274 |
+
"gpt.h.11.ln_1.weight": {
|
| 275 |
+
"scale": 0.16134341061115265,
|
| 276 |
+
"shape": [
|
| 277 |
+
1024
|
| 278 |
+
]
|
| 279 |
+
},
|
| 280 |
+
"gpt.h.11.ln_2.bias": {
|
| 281 |
+
"scale": 0.09844086319208145,
|
| 282 |
+
"shape": [
|
| 283 |
+
1024
|
| 284 |
+
]
|
| 285 |
+
},
|
| 286 |
+
"gpt.h.11.ln_2.weight": {
|
| 287 |
+
"scale": 0.22313998639583588,
|
| 288 |
+
"shape": [
|
| 289 |
+
1024
|
| 290 |
+
]
|
| 291 |
+
},
|
| 292 |
+
"gpt.h.11.mlp.c_fc.bias": {
|
| 293 |
+
"scale": 0.03060619719326496,
|
| 294 |
+
"shape": [
|
| 295 |
+
4096
|
| 296 |
+
]
|
| 297 |
+
},
|
| 298 |
+
"gpt.h.11.mlp.c_fc.weight": {
|
| 299 |
+
"scale": 0.06761165708303452,
|
| 300 |
+
"shape": [
|
| 301 |
+
1024,
|
| 302 |
+
4096
|
| 303 |
+
]
|
| 304 |
+
},
|
| 305 |
+
"gpt.h.11.mlp.c_proj.bias": {
|
| 306 |
+
"scale": 0.10643502324819565,
|
| 307 |
+
"shape": [
|
| 308 |
+
1024
|
| 309 |
+
]
|
| 310 |
+
},
|
| 311 |
+
"gpt.h.11.mlp.c_proj.weight": {
|
| 312 |
+
"scale": 0.4576462209224701,
|
| 313 |
+
"shape": [
|
| 314 |
+
4096,
|
| 315 |
+
1024
|
| 316 |
+
]
|
| 317 |
+
},
|
| 318 |
+
"gpt.h.12.attn.c_attn.bias": {
|
| 319 |
+
"scale": 0.04992813244462013,
|
| 320 |
+
"shape": [
|
| 321 |
+
3072
|
| 322 |
+
]
|
| 323 |
+
},
|
| 324 |
+
"gpt.h.12.attn.c_attn.weight": {
|
| 325 |
+
"scale": 0.14503903687000275,
|
| 326 |
+
"shape": [
|
| 327 |
+
1024,
|
| 328 |
+
3072
|
| 329 |
+
]
|
| 330 |
+
},
|
| 331 |
+
"gpt.h.12.attn.c_proj.bias": {
|
| 332 |
+
"scale": 0.033475592732429504,
|
| 333 |
+
"shape": [
|
| 334 |
+
1024
|
| 335 |
+
]
|
| 336 |
+
},
|
| 337 |
+
"gpt.h.12.attn.c_proj.weight": {
|
| 338 |
+
"scale": 0.11046390980482101,
|
| 339 |
+
"shape": [
|
| 340 |
+
1024,
|
| 341 |
+
1024
|
| 342 |
+
]
|
| 343 |
+
},
|
| 344 |
+
"gpt.h.12.ln_1.bias": {
|
| 345 |
+
"scale": 0.18613700568675995,
|
| 346 |
+
"shape": [
|
| 347 |
+
1024
|
| 348 |
+
]
|
| 349 |
+
},
|
| 350 |
+
"gpt.h.12.ln_1.weight": {
|
| 351 |
+
"scale": 0.16866794228553772,
|
| 352 |
+
"shape": [
|
| 353 |
+
1024
|
| 354 |
+
]
|
| 355 |
+
},
|
| 356 |
+
"gpt.h.12.ln_2.bias": {
|
| 357 |
+
"scale": 0.12722595036029816,
|
| 358 |
+
"shape": [
|
| 359 |
+
1024
|
| 360 |
+
]
|
| 361 |
+
},
|
| 362 |
+
"gpt.h.12.ln_2.weight": {
|
| 363 |
+
"scale": 0.23250164091587067,
|
| 364 |
+
"shape": [
|
| 365 |
+
1024
|
| 366 |
+
]
|
| 367 |
+
},
|
| 368 |
+
"gpt.h.12.mlp.c_fc.bias": {
|
| 369 |
+
"scale": 0.014812483452260494,
|
| 370 |
+
"shape": [
|
| 371 |
+
4096
|
| 372 |
+
]
|
| 373 |
+
},
|
| 374 |
+
"gpt.h.12.mlp.c_fc.weight": {
|
| 375 |
+
"scale": 0.068848617374897,
|
| 376 |
+
"shape": [
|
| 377 |
+
1024,
|
| 378 |
+
4096
|
| 379 |
+
]
|
| 380 |
+
},
|
| 381 |
+
"gpt.h.12.mlp.c_proj.bias": {
|
| 382 |
+
"scale": 0.12035634368658066,
|
| 383 |
+
"shape": [
|
| 384 |
+
1024
|
| 385 |
+
]
|
| 386 |
+
},
|
| 387 |
+
"gpt.h.12.mlp.c_proj.weight": {
|
| 388 |
+
"scale": 0.2743368148803711,
|
| 389 |
+
"shape": [
|
| 390 |
+
4096,
|
| 391 |
+
1024
|
| 392 |
+
]
|
| 393 |
+
},
|
| 394 |
+
"gpt.h.13.attn.c_attn.bias": {
|
| 395 |
+
"scale": 0.04760030657052994,
|
| 396 |
+
"shape": [
|
| 397 |
+
3072
|
| 398 |
+
]
|
| 399 |
+
},
|
| 400 |
+
"gpt.h.13.attn.c_attn.weight": {
|
| 401 |
+
"scale": 0.08714257925748825,
|
| 402 |
+
"shape": [
|
| 403 |
+
1024,
|
| 404 |
+
3072
|
| 405 |
+
]
|
| 406 |
+
},
|
| 407 |
+
"gpt.h.13.attn.c_proj.bias": {
|
| 408 |
+
"scale": 0.06746888160705566,
|
| 409 |
+
"shape": [
|
| 410 |
+
1024
|
| 411 |
+
]
|
| 412 |
+
},
|
| 413 |
+
"gpt.h.13.attn.c_proj.weight": {
|
| 414 |
+
"scale": 0.11844473332166672,
|
| 415 |
+
"shape": [
|
| 416 |
+
1024,
|
| 417 |
+
1024
|
| 418 |
+
]
|
| 419 |
+
},
|
| 420 |
+
"gpt.h.13.ln_1.bias": {
|
| 421 |
+
"scale": 0.15157721936702728,
|
| 422 |
+
"shape": [
|
| 423 |
+
1024
|
| 424 |
+
]
|
| 425 |
+
},
|
| 426 |
+
"gpt.h.13.ln_1.weight": {
|
| 427 |
+
"scale": 0.16393320262432098,
|
| 428 |
+
"shape": [
|
| 429 |
+
1024
|
| 430 |
+
]
|
| 431 |
+
},
|
| 432 |
+
"gpt.h.13.ln_2.bias": {
|
| 433 |
+
"scale": 0.1306534856557846,
|
| 434 |
+
"shape": [
|
| 435 |
+
1024
|
| 436 |
+
]
|
| 437 |
+
},
|
| 438 |
+
"gpt.h.13.ln_2.weight": {
|
| 439 |
+
"scale": 0.22248651087284088,
|
| 440 |
+
"shape": [
|
| 441 |
+
1024
|
| 442 |
+
]
|
| 443 |
+
},
|
| 444 |
+
"gpt.h.13.mlp.c_fc.bias": {
|
| 445 |
+
"scale": 0.019469883292913437,
|
| 446 |
+
"shape": [
|
| 447 |
+
4096
|
| 448 |
+
]
|
| 449 |
+
},
|
| 450 |
+
"gpt.h.13.mlp.c_fc.weight": {
|
| 451 |
+
"scale": 0.06430874019861221,
|
| 452 |
+
"shape": [
|
| 453 |
+
1024,
|
| 454 |
+
4096
|
| 455 |
+
]
|
| 456 |
+
},
|
| 457 |
+
"gpt.h.13.mlp.c_proj.bias": {
|
| 458 |
+
"scale": 0.11026618629693985,
|
| 459 |
+
"shape": [
|
| 460 |
+
1024
|
| 461 |
+
]
|
| 462 |
+
},
|
| 463 |
+
"gpt.h.13.mlp.c_proj.weight": {
|
| 464 |
+
"scale": 0.2532596290111542,
|
| 465 |
+
"shape": [
|
| 466 |
+
4096,
|
| 467 |
+
1024
|
| 468 |
+
]
|
| 469 |
+
},
|
| 470 |
+
"gpt.h.14.attn.c_attn.bias": {
|
| 471 |
+
"scale": 0.04019672051072121,
|
| 472 |
+
"shape": [
|
| 473 |
+
3072
|
| 474 |
+
]
|
| 475 |
+
},
|
| 476 |
+
"gpt.h.14.attn.c_attn.weight": {
|
| 477 |
+
"scale": 0.14072343707084656,
|
| 478 |
+
"shape": [
|
| 479 |
+
1024,
|
| 480 |
+
3072
|
| 481 |
+
]
|
| 482 |
+
},
|
| 483 |
+
"gpt.h.14.attn.c_proj.bias": {
|
| 484 |
+
"scale": 0.02364635095000267,
|
| 485 |
+
"shape": [
|
| 486 |
+
1024
|
| 487 |
+
]
|
| 488 |
+
},
|
| 489 |
+
"gpt.h.14.attn.c_proj.weight": {
|
| 490 |
+
"scale": 0.10160692036151886,
|
| 491 |
+
"shape": [
|
| 492 |
+
1024,
|
| 493 |
+
1024
|
| 494 |
+
]
|
| 495 |
+
},
|
| 496 |
+
"gpt.h.14.ln_1.bias": {
|
| 497 |
+
"scale": 0.159767284989357,
|
| 498 |
+
"shape": [
|
| 499 |
+
1024
|
| 500 |
+
]
|
| 501 |
+
},
|
| 502 |
+
"gpt.h.14.ln_1.weight": {
|
| 503 |
+
"scale": 0.16219459474086761,
|
| 504 |
+
"shape": [
|
| 505 |
+
1024
|
| 506 |
+
]
|
| 507 |
+
},
|
| 508 |
+
"gpt.h.14.ln_2.bias": {
|
| 509 |
+
"scale": 0.13014180958271027,
|
| 510 |
+
"shape": [
|
| 511 |
+
1024
|
| 512 |
+
]
|
| 513 |
+
},
|
| 514 |
+
"gpt.h.14.ln_2.weight": {
|
| 515 |
+
"scale": 0.21348479390144348,
|
| 516 |
+
"shape": [
|
| 517 |
+
1024
|
| 518 |
+
]
|
| 519 |
+
},
|
| 520 |
+
"gpt.h.14.mlp.c_fc.bias": {
|
| 521 |
+
"scale": 0.01531550008803606,
|
| 522 |
+
"shape": [
|
| 523 |
+
4096
|
| 524 |
+
]
|
| 525 |
+
},
|
| 526 |
+
"gpt.h.14.mlp.c_fc.weight": {
|
| 527 |
+
"scale": 0.08244454115629196,
|
| 528 |
+
"shape": [
|
| 529 |
+
1024,
|
| 530 |
+
4096
|
| 531 |
+
]
|
| 532 |
+
},
|
| 533 |
+
"gpt.h.14.mlp.c_proj.bias": {
|
| 534 |
+
"scale": 0.1048700362443924,
|
| 535 |
+
"shape": [
|
| 536 |
+
1024
|
| 537 |
+
]
|
| 538 |
+
},
|
| 539 |
+
"gpt.h.14.mlp.c_proj.weight": {
|
| 540 |
+
"scale": 0.28695791959762573,
|
| 541 |
+
"shape": [
|
| 542 |
+
4096,
|
| 543 |
+
1024
|
| 544 |
+
]
|
| 545 |
+
},
|
| 546 |
+
"gpt.h.15.attn.c_attn.bias": {
|
| 547 |
+
"scale": 0.04081624746322632,
|
| 548 |
+
"shape": [
|
| 549 |
+
3072
|
| 550 |
+
]
|
| 551 |
+
},
|
| 552 |
+
"gpt.h.15.attn.c_attn.weight": {
|
| 553 |
+
"scale": 0.07819346338510513,
|
| 554 |
+
"shape": [
|
| 555 |
+
1024,
|
| 556 |
+
3072
|
| 557 |
+
]
|
| 558 |
+
},
|
| 559 |
+
"gpt.h.15.attn.c_proj.bias": {
|
| 560 |
+
"scale": 0.08492325991392136,
|
| 561 |
+
"shape": [
|
| 562 |
+
1024
|
| 563 |
+
]
|
| 564 |
+
},
|
| 565 |
+
"gpt.h.15.attn.c_proj.weight": {
|
| 566 |
+
"scale": 0.11697079986333847,
|
| 567 |
+
"shape": [
|
| 568 |
+
1024,
|
| 569 |
+
1024
|
| 570 |
+
]
|
| 571 |
+
},
|
| 572 |
+
"gpt.h.15.ln_1.bias": {
|
| 573 |
+
"scale": 0.14451484382152557,
|
| 574 |
+
"shape": [
|
| 575 |
+
1024
|
| 576 |
+
]
|
| 577 |
+
},
|
| 578 |
+
"gpt.h.15.ln_1.weight": {
|
| 579 |
+
"scale": 0.16579607129096985,
|
| 580 |
+
"shape": [
|
| 581 |
+
1024
|
| 582 |
+
]
|
| 583 |
+
},
|
| 584 |
+
"gpt.h.15.ln_2.bias": {
|
| 585 |
+
"scale": 0.11252031475305557,
|
| 586 |
+
"shape": [
|
| 587 |
+
1024
|
| 588 |
+
]
|
| 589 |
+
},
|
| 590 |
+
"gpt.h.15.ln_2.weight": {
|
| 591 |
+
"scale": 0.20037208497524261,
|
| 592 |
+
"shape": [
|
| 593 |
+
1024
|
| 594 |
+
]
|
| 595 |
+
},
|
| 596 |
+
"gpt.h.15.mlp.c_fc.bias": {
|
| 597 |
+
"scale": 0.020076142624020576,
|
| 598 |
+
"shape": [
|
| 599 |
+
4096
|
| 600 |
+
]
|
| 601 |
+
},
|
| 602 |
+
"gpt.h.15.mlp.c_fc.weight": {
|
| 603 |
+
"scale": 0.0844724252820015,
|
| 604 |
+
"shape": [
|
| 605 |
+
1024,
|
| 606 |
+
4096
|
| 607 |
+
]
|
| 608 |
+
},
|
| 609 |
+
"gpt.h.15.mlp.c_proj.bias": {
|
| 610 |
+
"scale": 0.09757417440414429,
|
| 611 |
+
"shape": [
|
| 612 |
+
1024
|
| 613 |
+
]
|
| 614 |
+
},
|
| 615 |
+
"gpt.h.15.mlp.c_proj.weight": {
|
| 616 |
+
"scale": 0.310793936252594,
|
| 617 |
+
"shape": [
|
| 618 |
+
4096,
|
| 619 |
+
1024
|
| 620 |
+
]
|
| 621 |
+
},
|
| 622 |
+
"gpt.h.16.attn.c_attn.bias": {
|
| 623 |
+
"scale": 0.04743470624089241,
|
| 624 |
+
"shape": [
|
| 625 |
+
3072
|
| 626 |
+
]
|
| 627 |
+
},
|
| 628 |
+
"gpt.h.16.attn.c_attn.weight": {
|
| 629 |
+
"scale": 0.12722057104110718,
|
| 630 |
+
"shape": [
|
| 631 |
+
1024,
|
| 632 |
+
3072
|
| 633 |
+
]
|
| 634 |
+
},
|
| 635 |
+
"gpt.h.16.attn.c_proj.bias": {
|
| 636 |
+
"scale": 0.09606064110994339,
|
| 637 |
+
"shape": [
|
| 638 |
+
1024
|
| 639 |
+
]
|
| 640 |
+
},
|
| 641 |
+
"gpt.h.16.attn.c_proj.weight": {
|
| 642 |
+
"scale": 0.07591201364994049,
|
| 643 |
+
"shape": [
|
| 644 |
+
1024,
|
| 645 |
+
1024
|
| 646 |
+
]
|
| 647 |
+
},
|
| 648 |
+
"gpt.h.16.ln_1.bias": {
|
| 649 |
+
"scale": 0.15709154307842255,
|
| 650 |
+
"shape": [
|
| 651 |
+
1024
|
| 652 |
+
]
|
| 653 |
+
},
|
| 654 |
+
"gpt.h.16.ln_1.weight": {
|
| 655 |
+
"scale": 0.15911130607128143,
|
| 656 |
+
"shape": [
|
| 657 |
+
1024
|
| 658 |
+
]
|
| 659 |
+
},
|
| 660 |
+
"gpt.h.16.ln_2.bias": {
|
| 661 |
+
"scale": 0.12005039304494858,
|
| 662 |
+
"shape": [
|
| 663 |
+
1024
|
| 664 |
+
]
|
| 665 |
+
},
|
| 666 |
+
"gpt.h.16.ln_2.weight": {
|
| 667 |
+
"scale": 0.18429037928581238,
|
| 668 |
+
"shape": [
|
| 669 |
+
1024
|
| 670 |
+
]
|
| 671 |
+
},
|
| 672 |
+
"gpt.h.16.mlp.c_fc.bias": {
|
| 673 |
+
"scale": 0.022201305255293846,
|
| 674 |
+
"shape": [
|
| 675 |
+
4096
|
| 676 |
+
]
|
| 677 |
+
},
|
| 678 |
+
"gpt.h.16.mlp.c_fc.weight": {
|
| 679 |
+
"scale": 0.08008849620819092,
|
| 680 |
+
"shape": [
|
| 681 |
+
1024,
|
| 682 |
+
4096
|
| 683 |
+
]
|
| 684 |
+
},
|
| 685 |
+
"gpt.h.16.mlp.c_proj.bias": {
|
| 686 |
+
"scale": 0.09774907678365707,
|
| 687 |
+
"shape": [
|
| 688 |
+
1024
|
| 689 |
+
]
|
| 690 |
+
},
|
| 691 |
+
"gpt.h.16.mlp.c_proj.weight": {
|
| 692 |
+
"scale": 0.16212008893489838,
|
| 693 |
+
"shape": [
|
| 694 |
+
4096,
|
| 695 |
+
1024
|
| 696 |
+
]
|
| 697 |
+
},
|
| 698 |
+
"gpt.h.17.attn.c_attn.bias": {
|
| 699 |
+
"scale": 0.04417693614959717,
|
| 700 |
+
"shape": [
|
| 701 |
+
3072
|
| 702 |
+
]
|
| 703 |
+
},
|
| 704 |
+
"gpt.h.17.attn.c_attn.weight": {
|
| 705 |
+
"scale": 0.1320490539073944,
|
| 706 |
+
"shape": [
|
| 707 |
+
1024,
|
| 708 |
+
3072
|
| 709 |
+
]
|
| 710 |
+
},
|
| 711 |
+
"gpt.h.17.attn.c_proj.bias": {
|
| 712 |
+
"scale": 0.06338126957416534,
|
| 713 |
+
"shape": [
|
| 714 |
+
1024
|
| 715 |
+
]
|
| 716 |
+
},
|
| 717 |
+
"gpt.h.17.attn.c_proj.weight": {
|
| 718 |
+
"scale": 0.13893575966358185,
|
| 719 |
+
"shape": [
|
| 720 |
+
1024,
|
| 721 |
+
1024
|
| 722 |
+
]
|
| 723 |
+
},
|
| 724 |
+
"gpt.h.17.ln_1.bias": {
|
| 725 |
+
"scale": 0.15823052823543549,
|
| 726 |
+
"shape": [
|
| 727 |
+
1024
|
| 728 |
+
]
|
| 729 |
+
},
|
| 730 |
+
"gpt.h.17.ln_1.weight": {
|
| 731 |
+
"scale": 0.16002225875854492,
|
| 732 |
+
"shape": [
|
| 733 |
+
1024
|
| 734 |
+
]
|
| 735 |
+
},
|
| 736 |
+
"gpt.h.17.ln_2.bias": {
|
| 737 |
+
"scale": 0.13184016942977905,
|
| 738 |
+
"shape": [
|
| 739 |
+
1024
|
| 740 |
+
]
|
| 741 |
+
},
|
| 742 |
+
"gpt.h.17.ln_2.weight": {
|
| 743 |
+
"scale": 0.1869189441204071,
|
| 744 |
+
"shape": [
|
| 745 |
+
1024
|
| 746 |
+
]
|
| 747 |
+
},
|
| 748 |
+
"gpt.h.17.mlp.c_fc.bias": {
|
| 749 |
+
"scale": 0.02481200359761715,
|
| 750 |
+
"shape": [
|
| 751 |
+
4096
|
| 752 |
+
]
|
| 753 |
+
},
|
| 754 |
+
"gpt.h.17.mlp.c_fc.weight": {
|
| 755 |
+
"scale": 0.0685386210680008,
|
| 756 |
+
"shape": [
|
| 757 |
+
1024,
|
| 758 |
+
4096
|
| 759 |
+
]
|
| 760 |
+
},
|
| 761 |
+
"gpt.h.17.mlp.c_proj.bias": {
|
| 762 |
+
"scale": 0.08835189044475555,
|
| 763 |
+
"shape": [
|
| 764 |
+
1024
|
| 765 |
+
]
|
| 766 |
+
},
|
| 767 |
+
"gpt.h.17.mlp.c_proj.weight": {
|
| 768 |
+
"scale": 0.18417593836784363,
|
| 769 |
+
"shape": [
|
| 770 |
+
4096,
|
| 771 |
+
1024
|
| 772 |
+
]
|
| 773 |
+
},
|
| 774 |
+
"gpt.h.18.attn.c_attn.bias": {
|
| 775 |
+
"scale": 0.037581078708171844,
|
| 776 |
+
"shape": [
|
| 777 |
+
3072
|
| 778 |
+
]
|
| 779 |
+
},
|
| 780 |
+
"gpt.h.18.attn.c_attn.weight": {
|
| 781 |
+
"scale": 0.12254094332456589,
|
| 782 |
+
"shape": [
|
| 783 |
+
1024,
|
| 784 |
+
3072
|
| 785 |
+
]
|
| 786 |
+
},
|
| 787 |
+
"gpt.h.18.attn.c_proj.bias": {
|
| 788 |
+
"scale": 0.11249931156635284,
|
| 789 |
+
"shape": [
|
| 790 |
+
1024
|
| 791 |
+
]
|
| 792 |
+
},
|
| 793 |
+
"gpt.h.18.attn.c_proj.weight": {
|
| 794 |
+
"scale": 0.09757298976182938,
|
| 795 |
+
"shape": [
|
| 796 |
+
1024,
|
| 797 |
+
1024
|
| 798 |
+
]
|
| 799 |
+
},
|
| 800 |
+
"gpt.h.18.ln_1.bias": {
|
| 801 |
+
"scale": 0.1457567662000656,
|
| 802 |
+
"shape": [
|
| 803 |
+
1024
|
| 804 |
+
]
|
| 805 |
+
},
|
| 806 |
+
"gpt.h.18.ln_1.weight": {
|
| 807 |
+
"scale": 0.15891136229038239,
|
| 808 |
+
"shape": [
|
| 809 |
+
1024
|
| 810 |
+
]
|
| 811 |
+
},
|
| 812 |
+
"gpt.h.18.ln_2.bias": {
|
| 813 |
+
"scale": 0.11753548681735992,
|
| 814 |
+
"shape": [
|
| 815 |
+
1024
|
| 816 |
+
]
|
| 817 |
+
},
|
| 818 |
+
"gpt.h.18.ln_2.weight": {
|
| 819 |
+
"scale": 0.17903394997119904,
|
| 820 |
+
"shape": [
|
| 821 |
+
1024
|
| 822 |
+
]
|
| 823 |
+
},
|
| 824 |
+
"gpt.h.18.mlp.c_fc.bias": {
|
| 825 |
+
"scale": 0.021563095971941948,
|
| 826 |
+
"shape": [
|
| 827 |
+
4096
|
| 828 |
+
]
|
| 829 |
+
},
|
| 830 |
+
"gpt.h.18.mlp.c_fc.weight": {
|
| 831 |
+
"scale": 0.06829343736171722,
|
| 832 |
+
"shape": [
|
| 833 |
+
1024,
|
| 834 |
+
4096
|
| 835 |
+
]
|
| 836 |
+
},
|
| 837 |
+
"gpt.h.18.mlp.c_proj.bias": {
|
| 838 |
+
"scale": 0.06819847971200943,
|
| 839 |
+
"shape": [
|
| 840 |
+
1024
|
| 841 |
+
]
|
| 842 |
+
},
|
| 843 |
+
"gpt.h.18.mlp.c_proj.weight": {
|
| 844 |
+
"scale": 0.12814833223819733,
|
| 845 |
+
"shape": [
|
| 846 |
+
4096,
|
| 847 |
+
1024
|
| 848 |
+
]
|
| 849 |
+
},
|
| 850 |
+
"gpt.h.19.attn.c_attn.bias": {
|
| 851 |
+
"scale": 0.04691386595368385,
|
| 852 |
+
"shape": [
|
| 853 |
+
3072
|
| 854 |
+
]
|
| 855 |
+
},
|
| 856 |
+
"gpt.h.19.attn.c_attn.weight": {
|
| 857 |
+
"scale": 0.11684088408946991,
|
| 858 |
+
"shape": [
|
| 859 |
+
1024,
|
| 860 |
+
3072
|
| 861 |
+
]
|
| 862 |
+
},
|
| 863 |
+
"gpt.h.19.attn.c_proj.bias": {
|
| 864 |
+
"scale": 0.05939425155520439,
|
| 865 |
+
"shape": [
|
| 866 |
+
1024
|
| 867 |
+
]
|
| 868 |
+
},
|
| 869 |
+
"gpt.h.19.attn.c_proj.weight": {
|
| 870 |
+
"scale": 0.12523870170116425,
|
| 871 |
+
"shape": [
|
| 872 |
+
1024,
|
| 873 |
+
1024
|
| 874 |
+
]
|
| 875 |
+
},
|
| 876 |
+
"gpt.h.19.ln_1.bias": {
|
| 877 |
+
"scale": 0.14781410992145538,
|
| 878 |
+
"shape": [
|
| 879 |
+
1024
|
| 880 |
+
]
|
| 881 |
+
},
|
| 882 |
+
"gpt.h.19.ln_1.weight": {
|
| 883 |
+
"scale": 0.15612022578716278,
|
| 884 |
+
"shape": [
|
| 885 |
+
1024
|
| 886 |
+
]
|
| 887 |
+
},
|
| 888 |
+
"gpt.h.19.ln_2.bias": {
|
| 889 |
+
"scale": 0.10865359008312225,
|
| 890 |
+
"shape": [
|
| 891 |
+
1024
|
| 892 |
+
]
|
| 893 |
+
},
|
| 894 |
+
"gpt.h.19.ln_2.weight": {
|
| 895 |
+
"scale": 0.17884387075901031,
|
| 896 |
+
"shape": [
|
| 897 |
+
1024
|
| 898 |
+
]
|
| 899 |
+
},
|
| 900 |
+
"gpt.h.19.mlp.c_fc.bias": {
|
| 901 |
+
"scale": 0.023445727303624153,
|
| 902 |
+
"shape": [
|
| 903 |
+
4096
|
| 904 |
+
]
|
| 905 |
+
},
|
| 906 |
+
"gpt.h.19.mlp.c_fc.weight": {
|
| 907 |
+
"scale": 0.07771860063076019,
|
| 908 |
+
"shape": [
|
| 909 |
+
1024,
|
| 910 |
+
4096
|
| 911 |
+
]
|
| 912 |
+
},
|
| 913 |
+
"gpt.h.19.mlp.c_proj.bias": {
|
| 914 |
+
"scale": 0.05867428332567215,
|
| 915 |
+
"shape": [
|
| 916 |
+
1024
|
| 917 |
+
]
|
| 918 |
+
},
|
| 919 |
+
"gpt.h.19.mlp.c_proj.weight": {
|
| 920 |
+
"scale": 0.1523352414369583,
|
| 921 |
+
"shape": [
|
| 922 |
+
4096,
|
| 923 |
+
1024
|
| 924 |
+
]
|
| 925 |
+
},
|
| 926 |
+
"gpt.h.2.attn.c_attn.bias": {
|
| 927 |
+
"scale": 0.04482164978981018,
|
| 928 |
+
"shape": [
|
| 929 |
+
3072
|
| 930 |
+
]
|
| 931 |
+
},
|
| 932 |
+
"gpt.h.2.attn.c_attn.weight": {
|
| 933 |
+
"scale": 0.06159922480583191,
|
| 934 |
+
"shape": [
|
| 935 |
+
1024,
|
| 936 |
+
3072
|
| 937 |
+
]
|
| 938 |
+
},
|
| 939 |
+
"gpt.h.2.attn.c_proj.bias": {
|
| 940 |
+
"scale": 0.3751963675022125,
|
| 941 |
+
"shape": [
|
| 942 |
+
1024
|
| 943 |
+
]
|
| 944 |
+
},
|
| 945 |
+
"gpt.h.2.attn.c_proj.weight": {
|
| 946 |
+
"scale": 0.4305261969566345,
|
| 947 |
+
"shape": [
|
| 948 |
+
1024,
|
| 949 |
+
1024
|
| 950 |
+
]
|
| 951 |
+
},
|
| 952 |
+
"gpt.h.2.ln_1.bias": {
|
| 953 |
+
"scale": 0.16870178282260895,
|
| 954 |
+
"shape": [
|
| 955 |
+
1024
|
| 956 |
+
]
|
| 957 |
+
},
|
| 958 |
+
"gpt.h.2.ln_1.weight": {
|
| 959 |
+
"scale": 0.09723272174596786,
|
| 960 |
+
"shape": [
|
| 961 |
+
1024
|
| 962 |
+
]
|
| 963 |
+
},
|
| 964 |
+
"gpt.h.2.ln_2.bias": {
|
| 965 |
+
"scale": 0.14638309180736542,
|
| 966 |
+
"shape": [
|
| 967 |
+
1024
|
| 968 |
+
]
|
| 969 |
+
},
|
| 970 |
+
"gpt.h.2.ln_2.weight": {
|
| 971 |
+
"scale": 0.24350528419017792,
|
| 972 |
+
"shape": [
|
| 973 |
+
1024
|
| 974 |
+
]
|
| 975 |
+
},
|
| 976 |
+
"gpt.h.2.mlp.c_fc.bias": {
|
| 977 |
+
"scale": 0.054711032658815384,
|
| 978 |
+
"shape": [
|
| 979 |
+
4096
|
| 980 |
+
]
|
| 981 |
+
},
|
| 982 |
+
"gpt.h.2.mlp.c_fc.weight": {
|
| 983 |
+
"scale": 0.19500534236431122,
|
| 984 |
+
"shape": [
|
| 985 |
+
1024,
|
| 986 |
+
4096
|
| 987 |
+
]
|
| 988 |
+
},
|
| 989 |
+
"gpt.h.2.mlp.c_proj.bias": {
|
| 990 |
+
"scale": 0.2771886885166168,
|
| 991 |
+
"shape": [
|
| 992 |
+
1024
|
| 993 |
+
]
|
| 994 |
+
},
|
| 995 |
+
"gpt.h.2.mlp.c_proj.weight": {
|
| 996 |
+
"scale": 0.7297950983047485,
|
| 997 |
+
"shape": [
|
| 998 |
+
4096,
|
| 999 |
+
1024
|
| 1000 |
+
]
|
| 1001 |
+
},
|
| 1002 |
+
"gpt.h.20.attn.c_attn.bias": {
|
| 1003 |
+
"scale": 0.03594465181231499,
|
| 1004 |
+
"shape": [
|
| 1005 |
+
3072
|
| 1006 |
+
]
|
| 1007 |
+
},
|
| 1008 |
+
"gpt.h.20.attn.c_attn.weight": {
|
| 1009 |
+
"scale": 0.12434620410203934,
|
| 1010 |
+
"shape": [
|
| 1011 |
+
1024,
|
| 1012 |
+
3072
|
| 1013 |
+
]
|
| 1014 |
+
},
|
| 1015 |
+
"gpt.h.20.attn.c_proj.bias": {
|
| 1016 |
+
"scale": 0.09055911749601364,
|
| 1017 |
+
"shape": [
|
| 1018 |
+
1024
|
| 1019 |
+
]
|
| 1020 |
+
},
|
| 1021 |
+
"gpt.h.20.attn.c_proj.weight": {
|
| 1022 |
+
"scale": 0.10791827738285065,
|
| 1023 |
+
"shape": [
|
| 1024 |
+
1024,
|
| 1025 |
+
1024
|
| 1026 |
+
]
|
| 1027 |
+
},
|
| 1028 |
+
"gpt.h.20.ln_1.bias": {
|
| 1029 |
+
"scale": 0.13368119299411774,
|
| 1030 |
+
"shape": [
|
| 1031 |
+
1024
|
| 1032 |
+
]
|
| 1033 |
+
},
|
| 1034 |
+
"gpt.h.20.ln_1.weight": {
|
| 1035 |
+
"scale": 0.15670302510261536,
|
| 1036 |
+
"shape": [
|
| 1037 |
+
1024
|
| 1038 |
+
]
|
| 1039 |
+
},
|
| 1040 |
+
"gpt.h.20.ln_2.bias": {
|
| 1041 |
+
"scale": 0.11155271530151367,
|
| 1042 |
+
"shape": [
|
| 1043 |
+
1024
|
| 1044 |
+
]
|
| 1045 |
+
},
|
| 1046 |
+
"gpt.h.20.ln_2.weight": {
|
| 1047 |
+
"scale": 0.16725917160511017,
|
| 1048 |
+
"shape": [
|
| 1049 |
+
1024
|
| 1050 |
+
]
|
| 1051 |
+
},
|
| 1052 |
+
"gpt.h.20.mlp.c_fc.bias": {
|
| 1053 |
+
"scale": 0.015253099612891674,
|
| 1054 |
+
"shape": [
|
| 1055 |
+
4096
|
| 1056 |
+
]
|
| 1057 |
+
},
|
| 1058 |
+
"gpt.h.20.mlp.c_fc.weight": {
|
| 1059 |
+
"scale": 0.06497155874967575,
|
| 1060 |
+
"shape": [
|
| 1061 |
+
1024,
|
| 1062 |
+
4096
|
| 1063 |
+
]
|
| 1064 |
+
},
|
| 1065 |
+
"gpt.h.20.mlp.c_proj.bias": {
|
| 1066 |
+
"scale": 0.04290299862623215,
|
| 1067 |
+
"shape": [
|
| 1068 |
+
1024
|
| 1069 |
+
]
|
| 1070 |
+
},
|
| 1071 |
+
"gpt.h.20.mlp.c_proj.weight": {
|
| 1072 |
+
"scale": 0.2691304385662079,
|
| 1073 |
+
"shape": [
|
| 1074 |
+
4096,
|
| 1075 |
+
1024
|
| 1076 |
+
]
|
| 1077 |
+
},
|
| 1078 |
+
"gpt.h.21.attn.c_attn.bias": {
|
| 1079 |
+
"scale": 0.04429187998175621,
|
| 1080 |
+
"shape": [
|
| 1081 |
+
3072
|
| 1082 |
+
]
|
| 1083 |
+
},
|
| 1084 |
+
"gpt.h.21.attn.c_attn.weight": {
|
| 1085 |
+
"scale": 0.10078362375497818,
|
| 1086 |
+
"shape": [
|
| 1087 |
+
1024,
|
| 1088 |
+
3072
|
| 1089 |
+
]
|
| 1090 |
+
},
|
| 1091 |
+
"gpt.h.21.attn.c_proj.bias": {
|
| 1092 |
+
"scale": 0.07908321917057037,
|
| 1093 |
+
"shape": [
|
| 1094 |
+
1024
|
| 1095 |
+
]
|
| 1096 |
+
},
|
| 1097 |
+
"gpt.h.21.attn.c_proj.weight": {
|
| 1098 |
+
"scale": 0.1681288182735443,
|
| 1099 |
+
"shape": [
|
| 1100 |
+
1024,
|
| 1101 |
+
1024
|
| 1102 |
+
]
|
| 1103 |
+
},
|
| 1104 |
+
"gpt.h.21.ln_1.bias": {
|
| 1105 |
+
"scale": 0.12408334761857986,
|
| 1106 |
+
"shape": [
|
| 1107 |
+
1024
|
| 1108 |
+
]
|
| 1109 |
+
},
|
| 1110 |
+
"gpt.h.21.ln_1.weight": {
|
| 1111 |
+
"scale": 0.1594894379377365,
|
| 1112 |
+
"shape": [
|
| 1113 |
+
1024
|
| 1114 |
+
]
|
| 1115 |
+
},
|
| 1116 |
+
"gpt.h.21.ln_2.bias": {
|
| 1117 |
+
"scale": 0.10880132019519806,
|
| 1118 |
+
"shape": [
|
| 1119 |
+
1024
|
| 1120 |
+
]
|
| 1121 |
+
},
|
| 1122 |
+
"gpt.h.21.ln_2.weight": {
|
| 1123 |
+
"scale": 0.1579473465681076,
|
| 1124 |
+
"shape": [
|
| 1125 |
+
1024
|
| 1126 |
+
]
|
| 1127 |
+
},
|
| 1128 |
+
"gpt.h.21.mlp.c_fc.bias": {
|
| 1129 |
+
"scale": 0.02153877541422844,
|
| 1130 |
+
"shape": [
|
| 1131 |
+
4096
|
| 1132 |
+
]
|
| 1133 |
+
},
|
| 1134 |
+
"gpt.h.21.mlp.c_fc.weight": {
|
| 1135 |
+
"scale": 0.07017097622156143,
|
| 1136 |
+
"shape": [
|
| 1137 |
+
1024,
|
| 1138 |
+
4096
|
| 1139 |
+
]
|
| 1140 |
+
},
|
| 1141 |
+
"gpt.h.21.mlp.c_proj.bias": {
|
| 1142 |
+
"scale": 0.025096621364355087,
|
| 1143 |
+
"shape": [
|
| 1144 |
+
1024
|
| 1145 |
+
]
|
| 1146 |
+
},
|
| 1147 |
+
"gpt.h.21.mlp.c_proj.weight": {
|
| 1148 |
+
"scale": 0.20477719604969025,
|
| 1149 |
+
"shape": [
|
| 1150 |
+
4096,
|
| 1151 |
+
1024
|
| 1152 |
+
]
|
| 1153 |
+
},
|
| 1154 |
+
"gpt.h.22.attn.c_attn.bias": {
|
| 1155 |
+
"scale": 0.04005954787135124,
|
| 1156 |
+
"shape": [
|
| 1157 |
+
3072
|
| 1158 |
+
]
|
| 1159 |
+
},
|
| 1160 |
+
"gpt.h.22.attn.c_attn.weight": {
|
| 1161 |
+
"scale": 0.07027842104434967,
|
| 1162 |
+
"shape": [
|
| 1163 |
+
1024,
|
| 1164 |
+
3072
|
| 1165 |
+
]
|
| 1166 |
+
},
|
| 1167 |
+
"gpt.h.22.attn.c_proj.bias": {
|
| 1168 |
+
"scale": 0.06767906248569489,
|
| 1169 |
+
"shape": [
|
| 1170 |
+
1024
|
| 1171 |
+
]
|
| 1172 |
+
},
|
| 1173 |
+
"gpt.h.22.attn.c_proj.weight": {
|
| 1174 |
+
"scale": 0.09001224488019943,
|
| 1175 |
+
"shape": [
|
| 1176 |
+
1024,
|
| 1177 |
+
1024
|
| 1178 |
+
]
|
| 1179 |
+
},
|
| 1180 |
+
"gpt.h.22.ln_1.bias": {
|
| 1181 |
+
"scale": 0.11637736856937408,
|
| 1182 |
+
"shape": [
|
| 1183 |
+
1024
|
| 1184 |
+
]
|
| 1185 |
+
},
|
| 1186 |
+
"gpt.h.22.ln_1.weight": {
|
| 1187 |
+
"scale": 0.15651044249534607,
|
| 1188 |
+
"shape": [
|
| 1189 |
+
1024
|
| 1190 |
+
]
|
| 1191 |
+
},
|
| 1192 |
+
"gpt.h.22.ln_2.bias": {
|
| 1193 |
+
"scale": 0.11879222095012665,
|
| 1194 |
+
"shape": [
|
| 1195 |
+
1024
|
| 1196 |
+
]
|
| 1197 |
+
},
|
| 1198 |
+
"gpt.h.22.ln_2.weight": {
|
| 1199 |
+
"scale": 0.1595410406589508,
|
| 1200 |
+
"shape": [
|
| 1201 |
+
1024
|
| 1202 |
+
]
|
| 1203 |
+
},
|
| 1204 |
+
"gpt.h.22.mlp.c_fc.bias": {
|
| 1205 |
+
"scale": 0.01718810759484768,
|
| 1206 |
+
"shape": [
|
| 1207 |
+
4096
|
| 1208 |
+
]
|
| 1209 |
+
},
|
| 1210 |
+
"gpt.h.22.mlp.c_fc.weight": {
|
| 1211 |
+
"scale": 0.07537717372179031,
|
| 1212 |
+
"shape": [
|
| 1213 |
+
1024,
|
| 1214 |
+
4096
|
| 1215 |
+
]
|
| 1216 |
+
},
|
| 1217 |
+
"gpt.h.22.mlp.c_proj.bias": {
|
| 1218 |
+
"scale": 0.024833859875798225,
|
| 1219 |
+
"shape": [
|
| 1220 |
+
1024
|
| 1221 |
+
]
|
| 1222 |
+
},
|
| 1223 |
+
"gpt.h.22.mlp.c_proj.weight": {
|
| 1224 |
+
"scale": 0.17492838203907013,
|
| 1225 |
+
"shape": [
|
| 1226 |
+
4096,
|
| 1227 |
+
1024
|
| 1228 |
+
]
|
| 1229 |
+
},
|
| 1230 |
+
"gpt.h.23.attn.c_attn.bias": {
|
| 1231 |
+
"scale": 0.0368594266474247,
|
| 1232 |
+
"shape": [
|
| 1233 |
+
3072
|
| 1234 |
+
]
|
| 1235 |
+
},
|
| 1236 |
+
"gpt.h.23.attn.c_attn.weight": {
|
| 1237 |
+
"scale": 0.0888095423579216,
|
| 1238 |
+
"shape": [
|
| 1239 |
+
1024,
|
| 1240 |
+
3072
|
| 1241 |
+
]
|
| 1242 |
+
},
|
| 1243 |
+
"gpt.h.23.attn.c_proj.bias": {
|
| 1244 |
+
"scale": 0.02792450785636902,
|
| 1245 |
+
"shape": [
|
| 1246 |
+
1024
|
| 1247 |
+
]
|
| 1248 |
+
},
|
| 1249 |
+
"gpt.h.23.attn.c_proj.weight": {
|
| 1250 |
+
"scale": 0.10428722202777863,
|
| 1251 |
+
"shape": [
|
| 1252 |
+
1024,
|
| 1253 |
+
1024
|
| 1254 |
+
]
|
| 1255 |
+
},
|
| 1256 |
+
"gpt.h.23.ln_1.bias": {
|
| 1257 |
+
"scale": 0.10264807939529419,
|
| 1258 |
+
"shape": [
|
| 1259 |
+
1024
|
| 1260 |
+
]
|
| 1261 |
+
},
|
| 1262 |
+
"gpt.h.23.ln_1.weight": {
|
| 1263 |
+
"scale": 0.14564886689186096,
|
| 1264 |
+
"shape": [
|
| 1265 |
+
1024
|
| 1266 |
+
]
|
| 1267 |
+
},
|
| 1268 |
+
"gpt.h.23.ln_2.bias": {
|
| 1269 |
+
"scale": 0.11791384220123291,
|
| 1270 |
+
"shape": [
|
| 1271 |
+
1024
|
| 1272 |
+
]
|
| 1273 |
+
},
|
| 1274 |
+
"gpt.h.23.ln_2.weight": {
|
| 1275 |
+
"scale": 0.15296445786952972,
|
| 1276 |
+
"shape": [
|
| 1277 |
+
1024
|
| 1278 |
+
]
|
| 1279 |
+
},
|
| 1280 |
+
"gpt.h.23.mlp.c_fc.bias": {
|
| 1281 |
+
"scale": 0.018314138054847717,
|
| 1282 |
+
"shape": [
|
| 1283 |
+
4096
|
| 1284 |
+
]
|
| 1285 |
+
},
|
| 1286 |
+
"gpt.h.23.mlp.c_fc.weight": {
|
| 1287 |
+
"scale": 0.0671987384557724,
|
| 1288 |
+
"shape": [
|
| 1289 |
+
1024,
|
| 1290 |
+
4096
|
| 1291 |
+
]
|
| 1292 |
+
},
|
| 1293 |
+
"gpt.h.23.mlp.c_proj.bias": {
|
| 1294 |
+
"scale": 0.030889278277754784,
|
| 1295 |
+
"shape": [
|
| 1296 |
+
1024
|
| 1297 |
+
]
|
| 1298 |
+
},
|
| 1299 |
+
"gpt.h.23.mlp.c_proj.weight": {
|
| 1300 |
+
"scale": 0.13600651919841766,
|
| 1301 |
+
"shape": [
|
| 1302 |
+
4096,
|
| 1303 |
+
1024
|
| 1304 |
+
]
|
| 1305 |
+
},
|
| 1306 |
+
"gpt.h.24.attn.c_attn.bias": {
|
| 1307 |
+
"scale": 0.04183034226298332,
|
| 1308 |
+
"shape": [
|
| 1309 |
+
3072
|
| 1310 |
+
]
|
| 1311 |
+
},
|
| 1312 |
+
"gpt.h.24.attn.c_attn.weight": {
|
| 1313 |
+
"scale": 0.05768841132521629,
|
| 1314 |
+
"shape": [
|
| 1315 |
+
1024,
|
| 1316 |
+
3072
|
| 1317 |
+
]
|
| 1318 |
+
},
|
| 1319 |
+
"gpt.h.24.attn.c_proj.bias": {
|
| 1320 |
+
"scale": 0.02998465485870838,
|
| 1321 |
+
"shape": [
|
| 1322 |
+
1024
|
| 1323 |
+
]
|
| 1324 |
+
},
|
| 1325 |
+
"gpt.h.24.attn.c_proj.weight": {
|
| 1326 |
+
"scale": 0.09570259600877762,
|
| 1327 |
+
"shape": [
|
| 1328 |
+
1024,
|
| 1329 |
+
1024
|
| 1330 |
+
]
|
| 1331 |
+
},
|
| 1332 |
+
"gpt.h.24.ln_1.bias": {
|
| 1333 |
+
"scale": 0.1029733270406723,
|
| 1334 |
+
"shape": [
|
| 1335 |
+
1024
|
| 1336 |
+
]
|
| 1337 |
+
},
|
| 1338 |
+
"gpt.h.24.ln_1.weight": {
|
| 1339 |
+
"scale": 0.1645420342683792,
|
| 1340 |
+
"shape": [
|
| 1341 |
+
1024
|
| 1342 |
+
]
|
| 1343 |
+
},
|
| 1344 |
+
"gpt.h.24.ln_2.bias": {
|
| 1345 |
+
"scale": 0.11531977355480194,
|
| 1346 |
+
"shape": [
|
| 1347 |
+
1024
|
| 1348 |
+
]
|
| 1349 |
+
},
|
| 1350 |
+
"gpt.h.24.ln_2.weight": {
|
| 1351 |
+
"scale": 0.16749481856822968,
|
| 1352 |
+
"shape": [
|
| 1353 |
+
1024
|
| 1354 |
+
]
|
| 1355 |
+
},
|
| 1356 |
+
"gpt.h.24.mlp.c_fc.bias": {
|
| 1357 |
+
"scale": 0.02433849684894085,
|
| 1358 |
+
"shape": [
|
| 1359 |
+
4096
|
| 1360 |
+
]
|
| 1361 |
+
},
|
| 1362 |
+
"gpt.h.24.mlp.c_fc.weight": {
|
| 1363 |
+
"scale": 0.056723203510046005,
|
| 1364 |
+
"shape": [
|
| 1365 |
+
1024,
|
| 1366 |
+
4096
|
| 1367 |
+
]
|
| 1368 |
+
},
|
| 1369 |
+
"gpt.h.24.mlp.c_proj.bias": {
|
| 1370 |
+
"scale": 0.03132357448339462,
|
| 1371 |
+
"shape": [
|
| 1372 |
+
1024
|
| 1373 |
+
]
|
| 1374 |
+
},
|
| 1375 |
+
"gpt.h.24.mlp.c_proj.weight": {
|
| 1376 |
+
"scale": 0.08369418233633041,
|
| 1377 |
+
"shape": [
|
| 1378 |
+
4096,
|
| 1379 |
+
1024
|
| 1380 |
+
]
|
| 1381 |
+
},
|
| 1382 |
+
"gpt.h.25.attn.c_attn.bias": {
|
| 1383 |
+
"scale": 0.043894506990909576,
|
| 1384 |
+
"shape": [
|
| 1385 |
+
3072
|
| 1386 |
+
]
|
| 1387 |
+
},
|
| 1388 |
+
"gpt.h.25.attn.c_attn.weight": {
|
| 1389 |
+
"scale": 0.05882854387164116,
|
| 1390 |
+
"shape": [
|
| 1391 |
+
1024,
|
| 1392 |
+
3072
|
| 1393 |
+
]
|
| 1394 |
+
},
|
| 1395 |
+
"gpt.h.25.attn.c_proj.bias": {
|
| 1396 |
+
"scale": 0.03485613688826561,
|
| 1397 |
+
"shape": [
|
| 1398 |
+
1024
|
| 1399 |
+
]
|
| 1400 |
+
},
|
| 1401 |
+
"gpt.h.25.attn.c_proj.weight": {
|
| 1402 |
+
"scale": 0.08835429698228836,
|
| 1403 |
+
"shape": [
|
| 1404 |
+
1024,
|
| 1405 |
+
1024
|
| 1406 |
+
]
|
| 1407 |
+
},
|
| 1408 |
+
"gpt.h.25.ln_1.bias": {
|
| 1409 |
+
"scale": 0.09291279315948486,
|
| 1410 |
+
"shape": [
|
| 1411 |
+
1024
|
| 1412 |
+
]
|
| 1413 |
+
},
|
| 1414 |
+
"gpt.h.25.ln_1.weight": {
|
| 1415 |
+
"scale": 0.17914152145385742,
|
| 1416 |
+
"shape": [
|
| 1417 |
+
1024
|
| 1418 |
+
]
|
| 1419 |
+
},
|
| 1420 |
+
"gpt.h.25.ln_2.bias": {
|
| 1421 |
+
"scale": 0.11765044182538986,
|
| 1422 |
+
"shape": [
|
| 1423 |
+
1024
|
| 1424 |
+
]
|
| 1425 |
+
},
|
| 1426 |
+
"gpt.h.25.ln_2.weight": {
|
| 1427 |
+
"scale": 0.1694125235080719,
|
| 1428 |
+
"shape": [
|
| 1429 |
+
1024
|
| 1430 |
+
]
|
| 1431 |
+
},
|
| 1432 |
+
"gpt.h.25.mlp.c_fc.bias": {
|
| 1433 |
+
"scale": 0.02206212468445301,
|
| 1434 |
+
"shape": [
|
| 1435 |
+
4096
|
| 1436 |
+
]
|
| 1437 |
+
},
|
| 1438 |
+
"gpt.h.25.mlp.c_fc.weight": {
|
| 1439 |
+
"scale": 0.06992777436971664,
|
| 1440 |
+
"shape": [
|
| 1441 |
+
1024,
|
| 1442 |
+
4096
|
| 1443 |
+
]
|
| 1444 |
+
},
|
| 1445 |
+
"gpt.h.25.mlp.c_proj.bias": {
|
| 1446 |
+
"scale": 0.04068372771143913,
|
| 1447 |
+
"shape": [
|
| 1448 |
+
1024
|
| 1449 |
+
]
|
| 1450 |
+
},
|
| 1451 |
+
"gpt.h.25.mlp.c_proj.weight": {
|
| 1452 |
+
"scale": 0.09146194905042648,
|
| 1453 |
+
"shape": [
|
| 1454 |
+
4096,
|
| 1455 |
+
1024
|
| 1456 |
+
]
|
| 1457 |
+
},
|
| 1458 |
+
"gpt.h.26.attn.c_attn.bias": {
|
| 1459 |
+
"scale": 0.08172182738780975,
|
| 1460 |
+
"shape": [
|
| 1461 |
+
3072
|
| 1462 |
+
]
|
| 1463 |
+
},
|
| 1464 |
+
"gpt.h.26.attn.c_attn.weight": {
|
| 1465 |
+
"scale": 0.06202762946486473,
|
| 1466 |
+
"shape": [
|
| 1467 |
+
1024,
|
| 1468 |
+
3072
|
| 1469 |
+
]
|
| 1470 |
+
},
|
| 1471 |
+
"gpt.h.26.attn.c_proj.bias": {
|
| 1472 |
+
"scale": 0.03825494274497032,
|
| 1473 |
+
"shape": [
|
| 1474 |
+
1024
|
| 1475 |
+
]
|
| 1476 |
+
},
|
| 1477 |
+
"gpt.h.26.attn.c_proj.weight": {
|
| 1478 |
+
"scale": 0.10804062336683273,
|
| 1479 |
+
"shape": [
|
| 1480 |
+
1024,
|
| 1481 |
+
1024
|
| 1482 |
+
]
|
| 1483 |
+
},
|
| 1484 |
+
"gpt.h.26.ln_1.bias": {
|
| 1485 |
+
"scale": 0.08116239309310913,
|
| 1486 |
+
"shape": [
|
| 1487 |
+
1024
|
| 1488 |
+
]
|
| 1489 |
+
},
|
| 1490 |
+
"gpt.h.26.ln_1.weight": {
|
| 1491 |
+
"scale": 0.17098096013069153,
|
| 1492 |
+
"shape": [
|
| 1493 |
+
1024
|
| 1494 |
+
]
|
| 1495 |
+
},
|
| 1496 |
+
"gpt.h.26.ln_2.bias": {
|
| 1497 |
+
"scale": 0.11734277009963989,
|
| 1498 |
+
"shape": [
|
| 1499 |
+
1024
|
| 1500 |
+
]
|
| 1501 |
+
},
|
| 1502 |
+
"gpt.h.26.ln_2.weight": {
|
| 1503 |
+
"scale": 0.1784631311893463,
|
| 1504 |
+
"shape": [
|
| 1505 |
+
1024
|
| 1506 |
+
]
|
| 1507 |
+
},
|
| 1508 |
+
"gpt.h.26.mlp.c_fc.bias": {
|
| 1509 |
+
"scale": 0.031606484204530716,
|
| 1510 |
+
"shape": [
|
| 1511 |
+
4096
|
| 1512 |
+
]
|
| 1513 |
+
},
|
| 1514 |
+
"gpt.h.26.mlp.c_fc.weight": {
|
| 1515 |
+
"scale": 0.05273488909006119,
|
| 1516 |
+
"shape": [
|
| 1517 |
+
1024,
|
| 1518 |
+
4096
|
| 1519 |
+
]
|
| 1520 |
+
},
|
| 1521 |
+
"gpt.h.26.mlp.c_proj.bias": {
|
| 1522 |
+
"scale": 0.05072355270385742,
|
| 1523 |
+
"shape": [
|
| 1524 |
+
1024
|
| 1525 |
+
]
|
| 1526 |
+
},
|
| 1527 |
+
"gpt.h.26.mlp.c_proj.weight": {
|
| 1528 |
+
"scale": 0.14328084886074066,
|
| 1529 |
+
"shape": [
|
| 1530 |
+
4096,
|
| 1531 |
+
1024
|
| 1532 |
+
]
|
| 1533 |
+
},
|
| 1534 |
+
"gpt.h.27.attn.c_attn.bias": {
|
| 1535 |
+
"scale": 0.05690082535147667,
|
| 1536 |
+
"shape": [
|
| 1537 |
+
3072
|
| 1538 |
+
]
|
| 1539 |
+
},
|
| 1540 |
+
"gpt.h.27.attn.c_attn.weight": {
|
| 1541 |
+
"scale": 0.07180392742156982,
|
| 1542 |
+
"shape": [
|
| 1543 |
+
1024,
|
| 1544 |
+
3072
|
| 1545 |
+
]
|
| 1546 |
+
},
|
| 1547 |
+
"gpt.h.27.attn.c_proj.bias": {
|
| 1548 |
+
"scale": 0.05289801210165024,
|
| 1549 |
+
"shape": [
|
| 1550 |
+
1024
|
| 1551 |
+
]
|
| 1552 |
+
},
|
| 1553 |
+
"gpt.h.27.attn.c_proj.weight": {
|
| 1554 |
+
"scale": 0.0980907455086708,
|
| 1555 |
+
"shape": [
|
| 1556 |
+
1024,
|
| 1557 |
+
1024
|
| 1558 |
+
]
|
| 1559 |
+
},
|
| 1560 |
+
"gpt.h.27.ln_1.bias": {
|
| 1561 |
+
"scale": 0.08269622176885605,
|
| 1562 |
+
"shape": [
|
| 1563 |
+
1024
|
| 1564 |
+
]
|
| 1565 |
+
},
|
| 1566 |
+
"gpt.h.27.ln_1.weight": {
|
| 1567 |
+
"scale": 0.17179779708385468,
|
| 1568 |
+
"shape": [
|
| 1569 |
+
1024
|
| 1570 |
+
]
|
| 1571 |
+
},
|
| 1572 |
+
"gpt.h.27.ln_2.bias": {
|
| 1573 |
+
"scale": 0.11928660422563553,
|
| 1574 |
+
"shape": [
|
| 1575 |
+
1024
|
| 1576 |
+
]
|
| 1577 |
+
},
|
| 1578 |
+
"gpt.h.27.ln_2.weight": {
|
| 1579 |
+
"scale": 0.18073512613773346,
|
| 1580 |
+
"shape": [
|
| 1581 |
+
1024
|
| 1582 |
+
]
|
| 1583 |
+
},
|
| 1584 |
+
"gpt.h.27.mlp.c_fc.bias": {
|
| 1585 |
+
"scale": 0.030272051692008972,
|
| 1586 |
+
"shape": [
|
| 1587 |
+
4096
|
| 1588 |
+
]
|
| 1589 |
+
},
|
| 1590 |
+
"gpt.h.27.mlp.c_fc.weight": {
|
| 1591 |
+
"scale": 0.05421888828277588,
|
| 1592 |
+
"shape": [
|
| 1593 |
+
1024,
|
| 1594 |
+
4096
|
| 1595 |
+
]
|
| 1596 |
+
},
|
| 1597 |
+
"gpt.h.27.mlp.c_proj.bias": {
|
| 1598 |
+
"scale": 0.04355442896485329,
|
| 1599 |
+
"shape": [
|
| 1600 |
+
1024
|
| 1601 |
+
]
|
| 1602 |
+
},
|
| 1603 |
+
"gpt.h.27.mlp.c_proj.weight": {
|
| 1604 |
+
"scale": 0.15102733671665192,
|
| 1605 |
+
"shape": [
|
| 1606 |
+
4096,
|
| 1607 |
+
1024
|
| 1608 |
+
]
|
| 1609 |
+
},
|
| 1610 |
+
"gpt.h.28.attn.c_attn.bias": {
|
| 1611 |
+
"scale": 0.04516652598977089,
|
| 1612 |
+
"shape": [
|
| 1613 |
+
3072
|
| 1614 |
+
]
|
| 1615 |
+
},
|
| 1616 |
+
"gpt.h.28.attn.c_attn.weight": {
|
| 1617 |
+
"scale": 0.05591177940368652,
|
| 1618 |
+
"shape": [
|
| 1619 |
+
1024,
|
| 1620 |
+
3072
|
| 1621 |
+
]
|
| 1622 |
+
},
|
| 1623 |
+
"gpt.h.28.attn.c_proj.bias": {
|
| 1624 |
+
"scale": 0.07696392387151718,
|
| 1625 |
+
"shape": [
|
| 1626 |
+
1024
|
| 1627 |
+
]
|
| 1628 |
+
},
|
| 1629 |
+
"gpt.h.28.attn.c_proj.weight": {
|
| 1630 |
+
"scale": 0.1706492155790329,
|
| 1631 |
+
"shape": [
|
| 1632 |
+
1024,
|
| 1633 |
+
1024
|
| 1634 |
+
]
|
| 1635 |
+
},
|
| 1636 |
+
"gpt.h.28.ln_1.bias": {
|
| 1637 |
+
"scale": 0.08867383003234863,
|
| 1638 |
+
"shape": [
|
| 1639 |
+
1024
|
| 1640 |
+
]
|
| 1641 |
+
},
|
| 1642 |
+
"gpt.h.28.ln_1.weight": {
|
| 1643 |
+
"scale": 0.18088868260383606,
|
| 1644 |
+
"shape": [
|
| 1645 |
+
1024
|
| 1646 |
+
]
|
| 1647 |
+
},
|
| 1648 |
+
"gpt.h.28.ln_2.bias": {
|
| 1649 |
+
"scale": 0.10548854619264603,
|
| 1650 |
+
"shape": [
|
| 1651 |
+
1024
|
| 1652 |
+
]
|
| 1653 |
+
},
|
| 1654 |
+
"gpt.h.28.ln_2.weight": {
|
| 1655 |
+
"scale": 0.19724524021148682,
|
| 1656 |
+
"shape": [
|
| 1657 |
+
1024
|
| 1658 |
+
]
|
| 1659 |
+
},
|
| 1660 |
+
"gpt.h.28.mlp.c_fc.bias": {
|
| 1661 |
+
"scale": 0.03599608689546585,
|
| 1662 |
+
"shape": [
|
| 1663 |
+
4096
|
| 1664 |
+
]
|
| 1665 |
+
},
|
| 1666 |
+
"gpt.h.28.mlp.c_fc.weight": {
|
| 1667 |
+
"scale": 0.15464694797992706,
|
| 1668 |
+
"shape": [
|
| 1669 |
+
1024,
|
| 1670 |
+
4096
|
| 1671 |
+
]
|
| 1672 |
+
},
|
| 1673 |
+
"gpt.h.28.mlp.c_proj.bias": {
|
| 1674 |
+
"scale": 0.10096704214811325,
|
| 1675 |
+
"shape": [
|
| 1676 |
+
1024
|
| 1677 |
+
]
|
| 1678 |
+
},
|
| 1679 |
+
"gpt.h.28.mlp.c_proj.weight": {
|
| 1680 |
+
"scale": 0.5808261632919312,
|
| 1681 |
+
"shape": [
|
| 1682 |
+
4096,
|
| 1683 |
+
1024
|
| 1684 |
+
]
|
| 1685 |
+
},
|
| 1686 |
+
"gpt.h.29.attn.c_attn.bias": {
|
| 1687 |
+
"scale": 0.061315275728702545,
|
| 1688 |
+
"shape": [
|
| 1689 |
+
3072
|
| 1690 |
+
]
|
| 1691 |
+
},
|
| 1692 |
+
"gpt.h.29.attn.c_attn.weight": {
|
| 1693 |
+
"scale": 0.072987399995327,
|
| 1694 |
+
"shape": [
|
| 1695 |
+
1024,
|
| 1696 |
+
3072
|
| 1697 |
+
]
|
| 1698 |
+
},
|
| 1699 |
+
"gpt.h.29.attn.c_proj.bias": {
|
| 1700 |
+
"scale": 0.0334136076271534,
|
| 1701 |
+
"shape": [
|
| 1702 |
+
1024
|
| 1703 |
+
]
|
| 1704 |
+
},
|
| 1705 |
+
"gpt.h.29.attn.c_proj.weight": {
|
| 1706 |
+
"scale": 0.33243221044540405,
|
| 1707 |
+
"shape": [
|
| 1708 |
+
1024,
|
| 1709 |
+
1024
|
| 1710 |
+
]
|
| 1711 |
+
},
|
| 1712 |
+
"gpt.h.29.ln_1.bias": {
|
| 1713 |
+
"scale": 0.0834951177239418,
|
| 1714 |
+
"shape": [
|
| 1715 |
+
1024
|
| 1716 |
+
]
|
| 1717 |
+
},
|
| 1718 |
+
"gpt.h.29.ln_1.weight": {
|
| 1719 |
+
"scale": 0.17551641166210175,
|
| 1720 |
+
"shape": [
|
| 1721 |
+
1024
|
| 1722 |
+
]
|
| 1723 |
+
},
|
| 1724 |
+
"gpt.h.29.ln_2.bias": {
|
| 1725 |
+
"scale": 0.060361869633197784,
|
| 1726 |
+
"shape": [
|
| 1727 |
+
1024
|
| 1728 |
+
]
|
| 1729 |
+
},
|
| 1730 |
+
"gpt.h.29.ln_2.weight": {
|
| 1731 |
+
"scale": 0.20771968364715576,
|
| 1732 |
+
"shape": [
|
| 1733 |
+
1024
|
| 1734 |
+
]
|
| 1735 |
+
},
|
| 1736 |
+
"gpt.h.29.mlp.c_fc.bias": {
|
| 1737 |
+
"scale": 0.04004308953881264,
|
| 1738 |
+
"shape": [
|
| 1739 |
+
4096
|
| 1740 |
+
]
|
| 1741 |
+
},
|
| 1742 |
+
"gpt.h.29.mlp.c_fc.weight": {
|
| 1743 |
+
"scale": 0.257427453994751,
|
| 1744 |
+
"shape": [
|
| 1745 |
+
1024,
|
| 1746 |
+
4096
|
| 1747 |
+
]
|
| 1748 |
+
},
|
| 1749 |
+
"gpt.h.29.mlp.c_proj.bias": {
|
| 1750 |
+
"scale": 0.08023141324520111,
|
| 1751 |
+
"shape": [
|
| 1752 |
+
1024
|
| 1753 |
+
]
|
| 1754 |
+
},
|
| 1755 |
+
"gpt.h.29.mlp.c_proj.weight": {
|
| 1756 |
+
"scale": 1.5732485055923462,
|
| 1757 |
+
"shape": [
|
| 1758 |
+
4096,
|
| 1759 |
+
1024
|
| 1760 |
+
]
|
| 1761 |
+
},
|
| 1762 |
+
"gpt.h.3.attn.c_attn.bias": {
|
| 1763 |
+
"scale": 0.036733418703079224,
|
| 1764 |
+
"shape": [
|
| 1765 |
+
3072
|
| 1766 |
+
]
|
| 1767 |
+
},
|
| 1768 |
+
"gpt.h.3.attn.c_attn.weight": {
|
| 1769 |
+
"scale": 0.09035025537014008,
|
| 1770 |
+
"shape": [
|
| 1771 |
+
1024,
|
| 1772 |
+
3072
|
| 1773 |
+
]
|
| 1774 |
+
},
|
| 1775 |
+
"gpt.h.3.attn.c_proj.bias": {
|
| 1776 |
+
"scale": 0.38110747933387756,
|
| 1777 |
+
"shape": [
|
| 1778 |
+
1024
|
| 1779 |
+
]
|
| 1780 |
+
},
|
| 1781 |
+
"gpt.h.3.attn.c_proj.weight": {
|
| 1782 |
+
"scale": 0.5139561891555786,
|
| 1783 |
+
"shape": [
|
| 1784 |
+
1024,
|
| 1785 |
+
1024
|
| 1786 |
+
]
|
| 1787 |
+
},
|
| 1788 |
+
"gpt.h.3.ln_1.bias": {
|
| 1789 |
+
"scale": 0.14589600265026093,
|
| 1790 |
+
"shape": [
|
| 1791 |
+
1024
|
| 1792 |
+
]
|
| 1793 |
+
},
|
| 1794 |
+
"gpt.h.3.ln_1.weight": {
|
| 1795 |
+
"scale": 0.11667633056640625,
|
| 1796 |
+
"shape": [
|
| 1797 |
+
1024
|
| 1798 |
+
]
|
| 1799 |
+
},
|
| 1800 |
+
"gpt.h.3.ln_2.bias": {
|
| 1801 |
+
"scale": 0.11972484737634659,
|
| 1802 |
+
"shape": [
|
| 1803 |
+
1024
|
| 1804 |
+
]
|
| 1805 |
+
},
|
| 1806 |
+
"gpt.h.3.ln_2.weight": {
|
| 1807 |
+
"scale": 0.2382904291152954,
|
| 1808 |
+
"shape": [
|
| 1809 |
+
1024
|
| 1810 |
+
]
|
| 1811 |
+
},
|
| 1812 |
+
"gpt.h.3.mlp.c_fc.bias": {
|
| 1813 |
+
"scale": 0.04690921679139137,
|
| 1814 |
+
"shape": [
|
| 1815 |
+
4096
|
| 1816 |
+
]
|
| 1817 |
+
},
|
| 1818 |
+
"gpt.h.3.mlp.c_fc.weight": {
|
| 1819 |
+
"scale": 0.16555476188659668,
|
| 1820 |
+
"shape": [
|
| 1821 |
+
1024,
|
| 1822 |
+
4096
|
| 1823 |
+
]
|
| 1824 |
+
},
|
| 1825 |
+
"gpt.h.3.mlp.c_proj.bias": {
|
| 1826 |
+
"scale": 0.2362823784351349,
|
| 1827 |
+
"shape": [
|
| 1828 |
+
1024
|
| 1829 |
+
]
|
| 1830 |
+
},
|
| 1831 |
+
"gpt.h.3.mlp.c_proj.weight": {
|
| 1832 |
+
"scale": 0.9494528770446777,
|
| 1833 |
+
"shape": [
|
| 1834 |
+
4096,
|
| 1835 |
+
1024
|
| 1836 |
+
]
|
| 1837 |
+
},
|
| 1838 |
+
"gpt.h.4.attn.c_attn.bias": {
|
| 1839 |
+
"scale": 0.03971828892827034,
|
| 1840 |
+
"shape": [
|
| 1841 |
+
3072
|
| 1842 |
+
]
|
| 1843 |
+
},
|
| 1844 |
+
"gpt.h.4.attn.c_attn.weight": {
|
| 1845 |
+
"scale": 0.0667421743273735,
|
| 1846 |
+
"shape": [
|
| 1847 |
+
1024,
|
| 1848 |
+
3072
|
| 1849 |
+
]
|
| 1850 |
+
},
|
| 1851 |
+
"gpt.h.4.attn.c_proj.bias": {
|
| 1852 |
+
"scale": 0.3155006468296051,
|
| 1853 |
+
"shape": [
|
| 1854 |
+
1024
|
| 1855 |
+
]
|
| 1856 |
+
},
|
| 1857 |
+
"gpt.h.4.attn.c_proj.weight": {
|
| 1858 |
+
"scale": 0.31384560465812683,
|
| 1859 |
+
"shape": [
|
| 1860 |
+
1024,
|
| 1861 |
+
1024
|
| 1862 |
+
]
|
| 1863 |
+
},
|
| 1864 |
+
"gpt.h.4.ln_1.bias": {
|
| 1865 |
+
"scale": 0.13793586194515228,
|
| 1866 |
+
"shape": [
|
| 1867 |
+
1024
|
| 1868 |
+
]
|
| 1869 |
+
},
|
| 1870 |
+
"gpt.h.4.ln_1.weight": {
|
| 1871 |
+
"scale": 0.1345834881067276,
|
| 1872 |
+
"shape": [
|
| 1873 |
+
1024
|
| 1874 |
+
]
|
| 1875 |
+
},
|
| 1876 |
+
"gpt.h.4.ln_2.bias": {
|
| 1877 |
+
"scale": 0.08576243370771408,
|
| 1878 |
+
"shape": [
|
| 1879 |
+
1024
|
| 1880 |
+
]
|
| 1881 |
+
},
|
| 1882 |
+
"gpt.h.4.ln_2.weight": {
|
| 1883 |
+
"scale": 0.23912283778190613,
|
| 1884 |
+
"shape": [
|
| 1885 |
+
1024
|
| 1886 |
+
]
|
| 1887 |
+
},
|
| 1888 |
+
"gpt.h.4.mlp.c_fc.bias": {
|
| 1889 |
+
"scale": 0.016857421025633812,
|
| 1890 |
+
"shape": [
|
| 1891 |
+
4096
|
| 1892 |
+
]
|
| 1893 |
+
},
|
| 1894 |
+
"gpt.h.4.mlp.c_fc.weight": {
|
| 1895 |
+
"scale": 0.0949544683098793,
|
| 1896 |
+
"shape": [
|
| 1897 |
+
1024,
|
| 1898 |
+
4096
|
| 1899 |
+
]
|
| 1900 |
+
},
|
| 1901 |
+
"gpt.h.4.mlp.c_proj.bias": {
|
| 1902 |
+
"scale": 0.14176210761070251,
|
| 1903 |
+
"shape": [
|
| 1904 |
+
1024
|
| 1905 |
+
]
|
| 1906 |
+
},
|
| 1907 |
+
"gpt.h.4.mlp.c_proj.weight": {
|
| 1908 |
+
"scale": 1.0221376419067383,
|
| 1909 |
+
"shape": [
|
| 1910 |
+
4096,
|
| 1911 |
+
1024
|
| 1912 |
+
]
|
| 1913 |
+
},
|
| 1914 |
+
"gpt.h.5.attn.c_attn.bias": {
|
| 1915 |
+
"scale": 0.03044235333800316,
|
| 1916 |
+
"shape": [
|
| 1917 |
+
3072
|
| 1918 |
+
]
|
| 1919 |
+
},
|
| 1920 |
+
"gpt.h.5.attn.c_attn.weight": {
|
| 1921 |
+
"scale": 0.0556764118373394,
|
| 1922 |
+
"shape": [
|
| 1923 |
+
1024,
|
| 1924 |
+
3072
|
| 1925 |
+
]
|
| 1926 |
+
},
|
| 1927 |
+
"gpt.h.5.attn.c_proj.bias": {
|
| 1928 |
+
"scale": 0.17702117562294006,
|
| 1929 |
+
"shape": [
|
| 1930 |
+
1024
|
| 1931 |
+
]
|
| 1932 |
+
},
|
| 1933 |
+
"gpt.h.5.attn.c_proj.weight": {
|
| 1934 |
+
"scale": 0.13661186397075653,
|
| 1935 |
+
"shape": [
|
| 1936 |
+
1024,
|
| 1937 |
+
1024
|
| 1938 |
+
]
|
| 1939 |
+
},
|
| 1940 |
+
"gpt.h.5.ln_1.bias": {
|
| 1941 |
+
"scale": 0.10966886579990387,
|
| 1942 |
+
"shape": [
|
| 1943 |
+
1024
|
| 1944 |
+
]
|
| 1945 |
+
},
|
| 1946 |
+
"gpt.h.5.ln_1.weight": {
|
| 1947 |
+
"scale": 0.17159949243068695,
|
| 1948 |
+
"shape": [
|
| 1949 |
+
1024
|
| 1950 |
+
]
|
| 1951 |
+
},
|
| 1952 |
+
"gpt.h.5.ln_2.bias": {
|
| 1953 |
+
"scale": 0.06587665528059006,
|
| 1954 |
+
"shape": [
|
| 1955 |
+
1024
|
| 1956 |
+
]
|
| 1957 |
+
},
|
| 1958 |
+
"gpt.h.5.ln_2.weight": {
|
| 1959 |
+
"scale": 0.2513478100299835,
|
| 1960 |
+
"shape": [
|
| 1961 |
+
1024
|
| 1962 |
+
]
|
| 1963 |
+
},
|
| 1964 |
+
"gpt.h.5.mlp.c_fc.bias": {
|
| 1965 |
+
"scale": 0.021238749846816063,
|
| 1966 |
+
"shape": [
|
| 1967 |
+
4096
|
| 1968 |
+
]
|
| 1969 |
+
},
|
| 1970 |
+
"gpt.h.5.mlp.c_fc.weight": {
|
| 1971 |
+
"scale": 0.08227789402008057,
|
| 1972 |
+
"shape": [
|
| 1973 |
+
1024,
|
| 1974 |
+
4096
|
| 1975 |
+
]
|
| 1976 |
+
},
|
| 1977 |
+
"gpt.h.5.mlp.c_proj.bias": {
|
| 1978 |
+
"scale": 0.09128402918577194,
|
| 1979 |
+
"shape": [
|
| 1980 |
+
1024
|
| 1981 |
+
]
|
| 1982 |
+
},
|
| 1983 |
+
"gpt.h.5.mlp.c_proj.weight": {
|
| 1984 |
+
"scale": 0.8960160613059998,
|
| 1985 |
+
"shape": [
|
| 1986 |
+
4096,
|
| 1987 |
+
1024
|
| 1988 |
+
]
|
| 1989 |
+
},
|
| 1990 |
+
"gpt.h.6.attn.c_attn.bias": {
|
| 1991 |
+
"scale": 0.04128195717930794,
|
| 1992 |
+
"shape": [
|
| 1993 |
+
3072
|
| 1994 |
+
]
|
| 1995 |
+
},
|
| 1996 |
+
"gpt.h.6.attn.c_attn.weight": {
|
| 1997 |
+
"scale": 0.09308914095163345,
|
| 1998 |
+
"shape": [
|
| 1999 |
+
1024,
|
| 2000 |
+
3072
|
| 2001 |
+
]
|
| 2002 |
+
},
|
| 2003 |
+
"gpt.h.6.attn.c_proj.bias": {
|
| 2004 |
+
"scale": 0.09955250471830368,
|
| 2005 |
+
"shape": [
|
| 2006 |
+
1024
|
| 2007 |
+
]
|
| 2008 |
+
},
|
| 2009 |
+
"gpt.h.6.attn.c_proj.weight": {
|
| 2010 |
+
"scale": 0.09972423315048218,
|
| 2011 |
+
"shape": [
|
| 2012 |
+
1024,
|
| 2013 |
+
1024
|
| 2014 |
+
]
|
| 2015 |
+
},
|
| 2016 |
+
"gpt.h.6.ln_1.bias": {
|
| 2017 |
+
"scale": 0.10598546266555786,
|
| 2018 |
+
"shape": [
|
| 2019 |
+
1024
|
| 2020 |
+
]
|
| 2021 |
+
},
|
| 2022 |
+
"gpt.h.6.ln_1.weight": {
|
| 2023 |
+
"scale": 0.20339453220367432,
|
| 2024 |
+
"shape": [
|
| 2025 |
+
1024
|
| 2026 |
+
]
|
| 2027 |
+
},
|
| 2028 |
+
"gpt.h.6.ln_2.bias": {
|
| 2029 |
+
"scale": 0.083857461810112,
|
| 2030 |
+
"shape": [
|
| 2031 |
+
1024
|
| 2032 |
+
]
|
| 2033 |
+
},
|
| 2034 |
+
"gpt.h.6.ln_2.weight": {
|
| 2035 |
+
"scale": 0.25439128279685974,
|
| 2036 |
+
"shape": [
|
| 2037 |
+
1024
|
| 2038 |
+
]
|
| 2039 |
+
},
|
| 2040 |
+
"gpt.h.6.mlp.c_fc.bias": {
|
| 2041 |
+
"scale": 0.026426810771226883,
|
| 2042 |
+
"shape": [
|
| 2043 |
+
4096
|
| 2044 |
+
]
|
| 2045 |
+
},
|
| 2046 |
+
"gpt.h.6.mlp.c_fc.weight": {
|
| 2047 |
+
"scale": 0.1172819659113884,
|
| 2048 |
+
"shape": [
|
| 2049 |
+
1024,
|
| 2050 |
+
4096
|
| 2051 |
+
]
|
| 2052 |
+
},
|
| 2053 |
+
"gpt.h.6.mlp.c_proj.bias": {
|
| 2054 |
+
"scale": 0.08432779461145401,
|
| 2055 |
+
"shape": [
|
| 2056 |
+
1024
|
| 2057 |
+
]
|
| 2058 |
+
},
|
| 2059 |
+
"gpt.h.6.mlp.c_proj.weight": {
|
| 2060 |
+
"scale": 0.8800371289253235,
|
| 2061 |
+
"shape": [
|
| 2062 |
+
4096,
|
| 2063 |
+
1024
|
| 2064 |
+
]
|
| 2065 |
+
},
|
| 2066 |
+
"gpt.h.7.attn.c_attn.bias": {
|
| 2067 |
+
"scale": 0.04523950815200806,
|
| 2068 |
+
"shape": [
|
| 2069 |
+
3072
|
| 2070 |
+
]
|
| 2071 |
+
},
|
| 2072 |
+
"gpt.h.7.attn.c_attn.weight": {
|
| 2073 |
+
"scale": 0.09843175113201141,
|
| 2074 |
+
"shape": [
|
| 2075 |
+
1024,
|
| 2076 |
+
3072
|
| 2077 |
+
]
|
| 2078 |
+
},
|
| 2079 |
+
"gpt.h.7.attn.c_proj.bias": {
|
| 2080 |
+
"scale": 0.09096702188253403,
|
| 2081 |
+
"shape": [
|
| 2082 |
+
1024
|
| 2083 |
+
]
|
| 2084 |
+
},
|
| 2085 |
+
"gpt.h.7.attn.c_proj.weight": {
|
| 2086 |
+
"scale": 0.09420859813690186,
|
| 2087 |
+
"shape": [
|
| 2088 |
+
1024,
|
| 2089 |
+
1024
|
| 2090 |
+
]
|
| 2091 |
+
},
|
| 2092 |
+
"gpt.h.7.ln_1.bias": {
|
| 2093 |
+
"scale": 0.152985617518425,
|
| 2094 |
+
"shape": [
|
| 2095 |
+
1024
|
| 2096 |
+
]
|
| 2097 |
+
},
|
| 2098 |
+
"gpt.h.7.ln_1.weight": {
|
| 2099 |
+
"scale": 0.15078113973140717,
|
| 2100 |
+
"shape": [
|
| 2101 |
+
1024
|
| 2102 |
+
]
|
| 2103 |
+
},
|
| 2104 |
+
"gpt.h.7.ln_2.bias": {
|
| 2105 |
+
"scale": 0.1413334757089615,
|
| 2106 |
+
"shape": [
|
| 2107 |
+
1024
|
| 2108 |
+
]
|
| 2109 |
+
},
|
| 2110 |
+
"gpt.h.7.ln_2.weight": {
|
| 2111 |
+
"scale": 0.2576432526111603,
|
| 2112 |
+
"shape": [
|
| 2113 |
+
1024
|
| 2114 |
+
]
|
| 2115 |
+
},
|
| 2116 |
+
"gpt.h.7.mlp.c_fc.bias": {
|
| 2117 |
+
"scale": 0.033283643424510956,
|
| 2118 |
+
"shape": [
|
| 2119 |
+
4096
|
| 2120 |
+
]
|
| 2121 |
+
},
|
| 2122 |
+
"gpt.h.7.mlp.c_fc.weight": {
|
| 2123 |
+
"scale": 0.07020364701747894,
|
| 2124 |
+
"shape": [
|
| 2125 |
+
1024,
|
| 2126 |
+
4096
|
| 2127 |
+
]
|
| 2128 |
+
},
|
| 2129 |
+
"gpt.h.7.mlp.c_proj.bias": {
|
| 2130 |
+
"scale": 0.10713233798742294,
|
| 2131 |
+
"shape": [
|
| 2132 |
+
1024
|
| 2133 |
+
]
|
| 2134 |
+
},
|
| 2135 |
+
"gpt.h.7.mlp.c_proj.weight": {
|
| 2136 |
+
"scale": 0.42597928643226624,
|
| 2137 |
+
"shape": [
|
| 2138 |
+
4096,
|
| 2139 |
+
1024
|
| 2140 |
+
]
|
| 2141 |
+
},
|
| 2142 |
+
"gpt.h.8.attn.c_attn.bias": {
|
| 2143 |
+
"scale": 0.03788977861404419,
|
| 2144 |
+
"shape": [
|
| 2145 |
+
3072
|
| 2146 |
+
]
|
| 2147 |
+
},
|
| 2148 |
+
"gpt.h.8.attn.c_attn.weight": {
|
| 2149 |
+
"scale": 0.08913753926753998,
|
| 2150 |
+
"shape": [
|
| 2151 |
+
1024,
|
| 2152 |
+
3072
|
| 2153 |
+
]
|
| 2154 |
+
},
|
| 2155 |
+
"gpt.h.8.attn.c_proj.bias": {
|
| 2156 |
+
"scale": 0.08845029026269913,
|
| 2157 |
+
"shape": [
|
| 2158 |
+
1024
|
| 2159 |
+
]
|
| 2160 |
+
},
|
| 2161 |
+
"gpt.h.8.attn.c_proj.weight": {
|
| 2162 |
+
"scale": 0.09520888328552246,
|
| 2163 |
+
"shape": [
|
| 2164 |
+
1024,
|
| 2165 |
+
1024
|
| 2166 |
+
]
|
| 2167 |
+
},
|
| 2168 |
+
"gpt.h.8.ln_1.bias": {
|
| 2169 |
+
"scale": 0.12517669796943665,
|
| 2170 |
+
"shape": [
|
| 2171 |
+
1024
|
| 2172 |
+
]
|
| 2173 |
+
},
|
| 2174 |
+
"gpt.h.8.ln_1.weight": {
|
| 2175 |
+
"scale": 0.1648232787847519,
|
| 2176 |
+
"shape": [
|
| 2177 |
+
1024
|
| 2178 |
+
]
|
| 2179 |
+
},
|
| 2180 |
+
"gpt.h.8.ln_2.bias": {
|
| 2181 |
+
"scale": 0.14098228514194489,
|
| 2182 |
+
"shape": [
|
| 2183 |
+
1024
|
| 2184 |
+
]
|
| 2185 |
+
},
|
| 2186 |
+
"gpt.h.8.ln_2.weight": {
|
| 2187 |
+
"scale": 0.24974539875984192,
|
| 2188 |
+
"shape": [
|
| 2189 |
+
1024
|
| 2190 |
+
]
|
| 2191 |
+
},
|
| 2192 |
+
"gpt.h.8.mlp.c_fc.bias": {
|
| 2193 |
+
"scale": 0.026503393426537514,
|
| 2194 |
+
"shape": [
|
| 2195 |
+
4096
|
| 2196 |
+
]
|
| 2197 |
+
},
|
| 2198 |
+
"gpt.h.8.mlp.c_fc.weight": {
|
| 2199 |
+
"scale": 0.06782539933919907,
|
| 2200 |
+
"shape": [
|
| 2201 |
+
1024,
|
| 2202 |
+
4096
|
| 2203 |
+
]
|
| 2204 |
+
},
|
| 2205 |
+
"gpt.h.8.mlp.c_proj.bias": {
|
| 2206 |
+
"scale": 0.09713318198919296,
|
| 2207 |
+
"shape": [
|
| 2208 |
+
1024
|
| 2209 |
+
]
|
| 2210 |
+
},
|
| 2211 |
+
"gpt.h.8.mlp.c_proj.weight": {
|
| 2212 |
+
"scale": 0.43732672929763794,
|
| 2213 |
+
"shape": [
|
| 2214 |
+
4096,
|
| 2215 |
+
1024
|
| 2216 |
+
]
|
| 2217 |
+
},
|
| 2218 |
+
"gpt.h.9.attn.c_attn.bias": {
|
| 2219 |
+
"scale": 0.04425312206149101,
|
| 2220 |
+
"shape": [
|
| 2221 |
+
3072
|
| 2222 |
+
]
|
| 2223 |
+
},
|
| 2224 |
+
"gpt.h.9.attn.c_attn.weight": {
|
| 2225 |
+
"scale": 0.1181648001074791,
|
| 2226 |
+
"shape": [
|
| 2227 |
+
1024,
|
| 2228 |
+
3072
|
| 2229 |
+
]
|
| 2230 |
+
},
|
| 2231 |
+
"gpt.h.9.attn.c_proj.bias": {
|
| 2232 |
+
"scale": 0.07080474495887756,
|
| 2233 |
+
"shape": [
|
| 2234 |
+
1024
|
| 2235 |
+
]
|
| 2236 |
+
},
|
| 2237 |
+
"gpt.h.9.attn.c_proj.weight": {
|
| 2238 |
+
"scale": 0.10802309960126877,
|
| 2239 |
+
"shape": [
|
| 2240 |
+
1024,
|
| 2241 |
+
1024
|
| 2242 |
+
]
|
| 2243 |
+
},
|
| 2244 |
+
"gpt.h.9.ln_1.bias": {
|
| 2245 |
+
"scale": 0.15359099209308624,
|
| 2246 |
+
"shape": [
|
| 2247 |
+
1024
|
| 2248 |
+
]
|
| 2249 |
+
},
|
| 2250 |
+
"gpt.h.9.ln_1.weight": {
|
| 2251 |
+
"scale": 0.15670429170131683,
|
| 2252 |
+
"shape": [
|
| 2253 |
+
1024
|
| 2254 |
+
]
|
| 2255 |
+
},
|
| 2256 |
+
"gpt.h.9.ln_2.bias": {
|
| 2257 |
+
"scale": 0.1377403736114502,
|
| 2258 |
+
"shape": [
|
| 2259 |
+
1024
|
| 2260 |
+
]
|
| 2261 |
+
},
|
| 2262 |
+
"gpt.h.9.ln_2.weight": {
|
| 2263 |
+
"scale": 0.2576030492782593,
|
| 2264 |
+
"shape": [
|
| 2265 |
+
1024
|
| 2266 |
+
]
|
| 2267 |
+
},
|
| 2268 |
+
"gpt.h.9.mlp.c_fc.bias": {
|
| 2269 |
+
"scale": 0.031440090388059616,
|
| 2270 |
+
"shape": [
|
| 2271 |
+
4096
|
| 2272 |
+
]
|
| 2273 |
+
},
|
| 2274 |
+
"gpt.h.9.mlp.c_fc.weight": {
|
| 2275 |
+
"scale": 0.07645577937364578,
|
| 2276 |
+
"shape": [
|
| 2277 |
+
1024,
|
| 2278 |
+
4096
|
| 2279 |
+
]
|
| 2280 |
+
},
|
| 2281 |
+
"gpt.h.9.mlp.c_proj.bias": {
|
| 2282 |
+
"scale": 0.10798899829387665,
|
| 2283 |
+
"shape": [
|
| 2284 |
+
1024
|
| 2285 |
+
]
|
| 2286 |
+
},
|
| 2287 |
+
"gpt.h.9.mlp.c_proj.weight": {
|
| 2288 |
+
"scale": 0.4369252622127533,
|
| 2289 |
+
"shape": [
|
| 2290 |
+
4096,
|
| 2291 |
+
1024
|
| 2292 |
+
]
|
| 2293 |
+
},
|
| 2294 |
+
"gpt.ln_f.bias": {
|
| 2295 |
+
"scale": 0.20075297355651855,
|
| 2296 |
+
"shape": [
|
| 2297 |
+
1024
|
| 2298 |
+
]
|
| 2299 |
+
},
|
| 2300 |
+
"gpt.ln_f.weight": {
|
| 2301 |
+
"scale": 0.41444164514541626,
|
| 2302 |
+
"shape": [
|
| 2303 |
+
1024
|
| 2304 |
+
]
|
| 2305 |
+
},
|
| 2306 |
+
"gpt.wpe.emb.weight": {
|
| 2307 |
+
"scale": 0.0773041620850563,
|
| 2308 |
+
"shape": [
|
| 2309 |
+
608,
|
| 2310 |
+
1024
|
| 2311 |
+
]
|
| 2312 |
+
},
|
| 2313 |
+
"gpt.wte.weight": {
|
| 2314 |
+
"scale": 0.08020960539579391,
|
| 2315 |
+
"shape": [
|
| 2316 |
+
1026,
|
| 2317 |
+
1024
|
| 2318 |
+
]
|
| 2319 |
+
},
|
| 2320 |
+
"mel_head.bias": {
|
| 2321 |
+
"scale": 0.028449567034840584,
|
| 2322 |
+
"shape": [
|
| 2323 |
+
1026
|
| 2324 |
+
]
|
| 2325 |
+
},
|
| 2326 |
+
"mel_head.weight": {
|
| 2327 |
+
"scale": 0.07584048807621002,
|
| 2328 |
+
"shape": [
|
| 2329 |
+
1026,
|
| 2330 |
+
1024
|
| 2331 |
+
]
|
| 2332 |
+
}
|
| 2333 |
+
}
|
INT4/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[START]",
|
| 3 |
+
"eos_token": "[STOP]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
INT4/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
INT4/tokenizer_config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[STOP]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SPACE]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"259": {
|
| 28 |
+
"content": "[en]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"260": {
|
| 36 |
+
"content": "[de]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"261": {
|
| 44 |
+
"content": "[START]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"262": {
|
| 52 |
+
"content": "[fr]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"267": {
|
| 60 |
+
"content": "[ru]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"284": {
|
| 68 |
+
"content": "[es]",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"285": {
|
| 76 |
+
"content": "[it]",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"286": {
|
| 84 |
+
"content": "[pt]",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"293": {
|
| 92 |
+
"content": "[cs]",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"294": {
|
| 100 |
+
"content": "[pl]",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"295": {
|
| 108 |
+
"content": "[tr]",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"297": {
|
| 116 |
+
"content": "[nl]",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"5022": {
|
| 124 |
+
"content": "[ar]",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"5023": {
|
| 132 |
+
"content": "[zh-cn]",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"5412": {
|
| 140 |
+
"content": "[ja]",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"5753": {
|
| 148 |
+
"content": "[hu]",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"6152": {
|
| 156 |
+
"content": "[ko]",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"6680": {
|
| 164 |
+
"content": "[hi]",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"6681": {
|
| 172 |
+
"content": "[PAD]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"auto_map": {"AutoTokenizer": ["AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast", null]},
|
| 181 |
+
"bos_token": "[START]",
|
| 182 |
+
"clean_up_tokenization_spaces": true,
|
| 183 |
+
"eos_token": "[STOP]",
|
| 184 |
+
"max_length": null,
|
| 185 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 186 |
+
"pad_to_multiple_of": null,
|
| 187 |
+
"pad_token": "[PAD]",
|
| 188 |
+
"pad_token_type_id": 0,
|
| 189 |
+
"padding_side": "right",
|
| 190 |
+
"tokenizer_class": "XTTSTokenizerFast",
|
| 191 |
+
"unk_token": "[UNK]"
|
| 192 |
+
}
|
INT8/config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XttsGPT"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"audio_config": {
|
| 8 |
+
"mel_channels": 80,
|
| 9 |
+
"output_sample_rate": 24000,
|
| 10 |
+
"sample_rate": 22050
|
| 11 |
+
},
|
| 12 |
+
"auto_map": {
|
| 13 |
+
"AutoConfig": "AstraMindAI/xtts2-gpt--gpt_config.XTTSGPTConfig",
|
| 14 |
+
"AutoModelForCausalLM": "AstraMindAI/xtts2-gpt--xtts2_gpt_modeling.XttsGPT",
|
| 15 |
+
"AutoTokenizer": "AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast"
|
| 16 |
+
},
|
| 17 |
+
"decoder_input_dim": 1024,
|
| 18 |
+
"enable_redaction": false,
|
| 19 |
+
"gpt_batch_size": 1,
|
| 20 |
+
"gpt_max_audio_tokens": 605,
|
| 21 |
+
"hidden_size": 1024,
|
| 22 |
+
"initializer_range": 0.02,
|
| 23 |
+
"kv_cache": true,
|
| 24 |
+
"layer_norm_epsilon": 1e-05,
|
| 25 |
+
"max_audio_tokens": 605,
|
| 26 |
+
"max_prompt_tokens": 70,
|
| 27 |
+
"max_text_tokens": 402,
|
| 28 |
+
"model_type": "xtts_gpt",
|
| 29 |
+
"n_inner": 4096,
|
| 30 |
+
"num_attention_heads": 16,
|
| 31 |
+
"num_audio_tokens": 1026,
|
| 32 |
+
"num_hidden_layers": 30,
|
| 33 |
+
"number_text_tokens": 6681,
|
| 34 |
+
"reorder_and_upcast_attn": false,
|
| 35 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 36 |
+
"start_audio_token": 1024,
|
| 37 |
+
"start_text_token": null,
|
| 38 |
+
"stop_audio_token": 1025,
|
| 39 |
+
"stop_text_token": null,
|
| 40 |
+
"transformers_version": "4.46.0",
|
| 41 |
+
"use_masking_gt_prompt_approach": true,
|
| 42 |
+
"use_perceiver_resampler": true,
|
| 43 |
+
"vocab_size": 6681
|
| 44 |
+
}
|
INT8/gpt2_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a53c20858ef96c20b980b195de7f3611dda7003921b22845d89be2562dffeea2
|
| 3 |
+
size 380650106
|
INT8/int8_metadata.json
ADDED
|
@@ -0,0 +1,370 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"final_norm.bias": 0.004869353026151657,
|
| 3 |
+
"final_norm.weight": 0.04349002242088318,
|
| 4 |
+
"gpt.h.0.attn.c_attn.bias": 0.006784666329622269,
|
| 5 |
+
"gpt.h.0.attn.c_attn.weight": 0.005739760585129261,
|
| 6 |
+
"gpt.h.0.attn.c_proj.bias": 0.004933658055961132,
|
| 7 |
+
"gpt.h.0.attn.c_proj.weight": 0.006425631698220968,
|
| 8 |
+
"gpt.h.0.ln_1.bias": 0.0014453897019848228,
|
| 9 |
+
"gpt.h.0.ln_1.weight": 0.001891627791337669,
|
| 10 |
+
"gpt.h.0.ln_2.bias": 0.01133652776479721,
|
| 11 |
+
"gpt.h.0.ln_2.weight": 0.006315903272479773,
|
| 12 |
+
"gpt.h.0.mlp.c_fc.bias": 0.0026161218993365765,
|
| 13 |
+
"gpt.h.0.mlp.c_fc.weight": 0.009393199346959591,
|
| 14 |
+
"gpt.h.0.mlp.c_proj.bias": 0.016290107741951942,
|
| 15 |
+
"gpt.h.0.mlp.c_proj.weight": 0.036495596170425415,
|
| 16 |
+
"gpt.h.1.attn.c_attn.bias": 0.005060207564383745,
|
| 17 |
+
"gpt.h.1.attn.c_attn.weight": 0.003430855693295598,
|
| 18 |
+
"gpt.h.1.attn.c_proj.bias": 0.01638360321521759,
|
| 19 |
+
"gpt.h.1.attn.c_proj.weight": 0.025163328275084496,
|
| 20 |
+
"gpt.h.1.ln_1.bias": 0.011049080640077591,
|
| 21 |
+
"gpt.h.1.ln_1.weight": 0.004219602793455124,
|
| 22 |
+
"gpt.h.1.ln_2.bias": 0.011916977353394032,
|
| 23 |
+
"gpt.h.1.ln_2.weight": 0.011172252707183361,
|
| 24 |
+
"gpt.h.1.mlp.c_fc.bias": 0.0014588515041396022,
|
| 25 |
+
"gpt.h.1.mlp.c_fc.weight": 0.006692858878523111,
|
| 26 |
+
"gpt.h.1.mlp.c_proj.bias": 0.018994668498635292,
|
| 27 |
+
"gpt.h.1.mlp.c_proj.weight": 0.04086727276444435,
|
| 28 |
+
"gpt.h.10.attn.c_attn.bias": 0.0024330769665539265,
|
| 29 |
+
"gpt.h.10.attn.c_attn.weight": 0.006525476463139057,
|
| 30 |
+
"gpt.h.10.attn.c_proj.bias": 0.0025137888733297586,
|
| 31 |
+
"gpt.h.10.attn.c_proj.weight": 0.004787968005985022,
|
| 32 |
+
"gpt.h.10.ln_1.bias": 0.008025330491364002,
|
| 33 |
+
"gpt.h.10.ln_1.weight": 0.008399507962167263,
|
| 34 |
+
"gpt.h.10.ln_2.bias": 0.007188895717263222,
|
| 35 |
+
"gpt.h.10.ln_2.weight": 0.01416360680013895,
|
| 36 |
+
"gpt.h.10.mlp.c_fc.bias": 0.0016273388173431158,
|
| 37 |
+
"gpt.h.10.mlp.c_fc.weight": 0.005328207276761532,
|
| 38 |
+
"gpt.h.10.mlp.c_proj.bias": 0.005671263672411442,
|
| 39 |
+
"gpt.h.10.mlp.c_proj.weight": 0.02014252357184887,
|
| 40 |
+
"gpt.h.11.attn.c_attn.bias": 0.002672715811058879,
|
| 41 |
+
"gpt.h.11.attn.c_attn.weight": 0.0061920033767819405,
|
| 42 |
+
"gpt.h.11.attn.c_proj.bias": 0.0028154668398201466,
|
| 43 |
+
"gpt.h.11.attn.c_proj.weight": 0.004848834592849016,
|
| 44 |
+
"gpt.h.11.ln_1.bias": 0.009074854664504528,
|
| 45 |
+
"gpt.h.11.ln_1.weight": 0.008892943151295185,
|
| 46 |
+
"gpt.h.11.ln_2.bias": 0.0054258741438388824,
|
| 47 |
+
"gpt.h.11.ln_2.weight": 0.01229905430227518,
|
| 48 |
+
"gpt.h.11.mlp.c_fc.bias": 0.00168695580214262,
|
| 49 |
+
"gpt.h.11.mlp.c_fc.weight": 0.003726626979187131,
|
| 50 |
+
"gpt.h.11.mlp.c_proj.bias": 0.005866497289389372,
|
| 51 |
+
"gpt.h.11.mlp.c_proj.weight": 0.02522459626197815,
|
| 52 |
+
"gpt.h.12.attn.c_attn.bias": 0.0027519443538039923,
|
| 53 |
+
"gpt.h.12.attn.c_attn.weight": 0.007994277402758598,
|
| 54 |
+
"gpt.h.12.attn.c_proj.bias": 0.001845111371949315,
|
| 55 |
+
"gpt.h.12.attn.c_proj.weight": 0.006088561844080687,
|
| 56 |
+
"gpt.h.12.ln_1.bias": 0.010259520262479782,
|
| 57 |
+
"gpt.h.12.ln_1.weight": 0.00929665844887495,
|
| 58 |
+
"gpt.h.12.ln_2.bias": 0.007012453395873308,
|
| 59 |
+
"gpt.h.12.ln_2.weight": 0.012815050780773163,
|
| 60 |
+
"gpt.h.12.mlp.c_fc.bias": 0.0008164360770024359,
|
| 61 |
+
"gpt.h.12.mlp.c_fc.weight": 0.0037948056124150753,
|
| 62 |
+
"gpt.h.12.mlp.c_proj.bias": 0.006633814424276352,
|
| 63 |
+
"gpt.h.12.mlp.c_proj.weight": 0.01512092724442482,
|
| 64 |
+
"gpt.h.13.attn.c_attn.bias": 0.0026236390694975853,
|
| 65 |
+
"gpt.h.13.attn.c_attn.weight": 0.004803134128451347,
|
| 66 |
+
"gpt.h.13.attn.c_proj.bias": 0.00371875730343163,
|
| 67 |
+
"gpt.h.13.attn.c_proj.weight": 0.006528449710458517,
|
| 68 |
+
"gpt.h.13.ln_1.bias": 0.008354649879038334,
|
| 69 |
+
"gpt.h.13.ln_1.weight": 0.009035688824951649,
|
| 70 |
+
"gpt.h.13.ln_2.bias": 0.00720137357711792,
|
| 71 |
+
"gpt.h.13.ln_2.weight": 0.012263036333024502,
|
| 72 |
+
"gpt.h.13.mlp.c_fc.bias": 0.0010731430957093835,
|
| 73 |
+
"gpt.h.13.mlp.c_fc.weight": 0.0035445760004222393,
|
| 74 |
+
"gpt.h.13.mlp.c_proj.bias": 0.006077663507312536,
|
| 75 |
+
"gpt.h.13.mlp.c_proj.weight": 0.013959192670881748,
|
| 76 |
+
"gpt.h.14.attn.c_attn.bias": 0.0022155672777444124,
|
| 77 |
+
"gpt.h.14.attn.c_attn.weight": 0.007756410166621208,
|
| 78 |
+
"gpt.h.14.attn.c_proj.bias": 0.0013033421710133553,
|
| 79 |
+
"gpt.h.14.attn.c_proj.weight": 0.005600381642580032,
|
| 80 |
+
"gpt.h.14.ln_1.bias": 0.008806071244180202,
|
| 81 |
+
"gpt.h.14.ln_1.weight": 0.008939859457314014,
|
| 82 |
+
"gpt.h.14.ln_2.bias": 0.007173170801252127,
|
| 83 |
+
"gpt.h.14.ln_2.weight": 0.011766878888010979,
|
| 84 |
+
"gpt.h.14.mlp.c_fc.bias": 0.000844161375425756,
|
| 85 |
+
"gpt.h.14.mlp.c_fc.weight": 0.0045441873371601105,
|
| 86 |
+
"gpt.h.14.mlp.c_proj.bias": 0.005780238192528486,
|
| 87 |
+
"gpt.h.14.mlp.c_proj.weight": 0.01581657864153385,
|
| 88 |
+
"gpt.h.15.attn.c_attn.bias": 0.0022497144527733326,
|
| 89 |
+
"gpt.h.15.attn.c_attn.weight": 0.004309875890612602,
|
| 90 |
+
"gpt.h.15.attn.c_proj.bias": 0.004680809564888477,
|
| 91 |
+
"gpt.h.15.attn.c_proj.weight": 0.006447209510952234,
|
| 92 |
+
"gpt.h.15.ln_1.bias": 0.007965384982526302,
|
| 93 |
+
"gpt.h.15.ln_1.weight": 0.00913836620748043,
|
| 94 |
+
"gpt.h.15.ln_2.bias": 0.0062019070610404015,
|
| 95 |
+
"gpt.h.15.ln_2.weight": 0.01104413066059351,
|
| 96 |
+
"gpt.h.15.mlp.c_fc.bias": 0.0011065590661019087,
|
| 97 |
+
"gpt.h.15.mlp.c_fc.weight": 0.0046559604816138744,
|
| 98 |
+
"gpt.h.15.mlp.c_proj.bias": 0.005378104280680418,
|
| 99 |
+
"gpt.h.15.mlp.c_proj.weight": 0.017130374908447266,
|
| 100 |
+
"gpt.h.16.attn.c_attn.bias": 0.002614511176943779,
|
| 101 |
+
"gpt.h.16.attn.c_attn.weight": 0.0070121572352945805,
|
| 102 |
+
"gpt.h.16.attn.c_proj.bias": 0.0052946810610592365,
|
| 103 |
+
"gpt.h.16.attn.c_proj.weight": 0.004184126853942871,
|
| 104 |
+
"gpt.h.16.ln_1.bias": 0.00865858979523182,
|
| 105 |
+
"gpt.h.16.ln_1.weight": 0.008769914507865906,
|
| 106 |
+
"gpt.h.16.ln_2.bias": 0.0066169509664177895,
|
| 107 |
+
"gpt.h.16.ln_2.weight": 0.010157736949622631,
|
| 108 |
+
"gpt.h.16.mlp.c_fc.bias": 0.0012236940674483776,
|
| 109 |
+
"gpt.h.16.mlp.c_fc.weight": 0.004414326511323452,
|
| 110 |
+
"gpt.h.16.mlp.c_proj.bias": 0.005387744400650263,
|
| 111 |
+
"gpt.h.16.mlp.c_proj.weight": 0.008935753256082535,
|
| 112 |
+
"gpt.h.17.attn.c_attn.bias": 0.0024349491577595472,
|
| 113 |
+
"gpt.h.17.attn.c_attn.weight": 0.007278294302523136,
|
| 114 |
+
"gpt.h.17.attn.c_proj.bias": 0.0034934559371322393,
|
| 115 |
+
"gpt.h.17.attn.c_proj.weight": 0.007657876703888178,
|
| 116 |
+
"gpt.h.17.ln_1.bias": 0.008721367456018925,
|
| 117 |
+
"gpt.h.17.ln_1.weight": 0.00882012490183115,
|
| 118 |
+
"gpt.h.17.ln_2.bias": 0.007266780827194452,
|
| 119 |
+
"gpt.h.17.ln_2.weight": 0.010302619077265263,
|
| 120 |
+
"gpt.h.17.mlp.c_fc.bias": 0.0013675907393917441,
|
| 121 |
+
"gpt.h.17.mlp.c_fc.weight": 0.0037777191027998924,
|
| 122 |
+
"gpt.h.17.mlp.c_proj.bias": 0.0048697893507778645,
|
| 123 |
+
"gpt.h.17.mlp.c_proj.weight": 0.01015142910182476,
|
| 124 |
+
"gpt.h.18.attn.c_attn.bias": 0.002071398077532649,
|
| 125 |
+
"gpt.h.18.attn.c_attn.weight": 0.006754225119948387,
|
| 126 |
+
"gpt.h.18.attn.c_proj.bias": 0.006200749427080154,
|
| 127 |
+
"gpt.h.18.attn.c_proj.weight": 0.005378039088100195,
|
| 128 |
+
"gpt.h.18.ln_1.bias": 0.008033838123083115,
|
| 129 |
+
"gpt.h.18.ln_1.weight": 0.00875889416784048,
|
| 130 |
+
"gpt.h.18.ln_2.bias": 0.006478333845734596,
|
| 131 |
+
"gpt.h.18.ln_2.weight": 0.009868012741208076,
|
| 132 |
+
"gpt.h.18.mlp.c_fc.bias": 0.0011885170824825764,
|
| 133 |
+
"gpt.h.18.mlp.c_fc.weight": 0.003764205379411578,
|
| 134 |
+
"gpt.h.18.mlp.c_proj.bias": 0.0037589711137115955,
|
| 135 |
+
"gpt.h.18.mlp.c_proj.weight": 0.00706329382956028,
|
| 136 |
+
"gpt.h.19.attn.c_attn.bias": 0.002585803624242544,
|
| 137 |
+
"gpt.h.19.attn.c_attn.weight": 0.006440048571676016,
|
| 138 |
+
"gpt.h.19.attn.c_proj.bias": 0.003273698966950178,
|
| 139 |
+
"gpt.h.19.attn.c_proj.weight": 0.006902920547872782,
|
| 140 |
+
"gpt.h.19.ln_1.bias": 0.008147234097123146,
|
| 141 |
+
"gpt.h.19.ln_1.weight": 0.008605051785707474,
|
| 142 |
+
"gpt.h.19.ln_2.bias": 0.0059887804090976715,
|
| 143 |
+
"gpt.h.19.ln_2.weight": 0.009857536293566227,
|
| 144 |
+
"gpt.h.19.mlp.c_fc.bias": 0.0012922842288389802,
|
| 145 |
+
"gpt.h.19.mlp.c_fc.weight": 0.004283702466636896,
|
| 146 |
+
"gpt.h.19.mlp.c_proj.bias": 0.0032340157777071,
|
| 147 |
+
"gpt.h.19.mlp.c_proj.weight": 0.008396429941058159,
|
| 148 |
+
"gpt.h.2.attn.c_attn.bias": 0.0024704847019165754,
|
| 149 |
+
"gpt.h.2.attn.c_attn.weight": 0.0033952328376471996,
|
| 150 |
+
"gpt.h.2.attn.c_proj.bias": 0.02068011462688446,
|
| 151 |
+
"gpt.h.2.attn.c_proj.weight": 0.02372979000210762,
|
| 152 |
+
"gpt.h.2.ln_1.bias": 0.009298523887991905,
|
| 153 |
+
"gpt.h.2.ln_1.weight": 0.005359284114092588,
|
| 154 |
+
"gpt.h.2.ln_2.bias": 0.008068359456956387,
|
| 155 |
+
"gpt.h.2.ln_2.weight": 0.013421551324427128,
|
| 156 |
+
"gpt.h.2.mlp.c_fc.bias": 0.0030155687127262354,
|
| 157 |
+
"gpt.h.2.mlp.c_fc.weight": 0.01074832584708929,
|
| 158 |
+
"gpt.h.2.mlp.c_proj.bias": 0.015278116799890995,
|
| 159 |
+
"gpt.h.2.mlp.c_proj.weight": 0.04022492840886116,
|
| 160 |
+
"gpt.h.20.attn.c_attn.bias": 0.001981201348826289,
|
| 161 |
+
"gpt.h.20.attn.c_attn.weight": 0.006853727623820305,
|
| 162 |
+
"gpt.h.20.attn.c_proj.bias": 0.004991447553038597,
|
| 163 |
+
"gpt.h.20.attn.c_proj.weight": 0.005948251578956842,
|
| 164 |
+
"gpt.h.20.ln_1.bias": 0.007368254475295544,
|
| 165 |
+
"gpt.h.20.ln_1.weight": 0.00863717496395111,
|
| 166 |
+
"gpt.h.20.ln_2.bias": 0.006148574873805046,
|
| 167 |
+
"gpt.h.20.ln_2.weight": 0.009219009429216385,
|
| 168 |
+
"gpt.h.20.mlp.c_fc.bias": 0.0008407220593653619,
|
| 169 |
+
"gpt.h.20.mlp.c_fc.weight": 0.003581109456717968,
|
| 170 |
+
"gpt.h.20.mlp.c_proj.bias": 0.0023647320922464132,
|
| 171 |
+
"gpt.h.20.mlp.c_proj.weight": 0.014833961613476276,
|
| 172 |
+
"gpt.h.21.attn.c_attn.bias": 0.0024412847124040127,
|
| 173 |
+
"gpt.h.21.attn.c_attn.weight": 0.005555002484470606,
|
| 174 |
+
"gpt.h.21.attn.c_proj.bias": 0.004358917940407991,
|
| 175 |
+
"gpt.h.21.attn.c_proj.weight": 0.009266942739486694,
|
| 176 |
+
"gpt.h.21.ln_1.bias": 0.006839239504188299,
|
| 177 |
+
"gpt.h.21.ln_1.weight": 0.008790756575763226,
|
| 178 |
+
"gpt.h.21.ln_2.bias": 0.005996923428028822,
|
| 179 |
+
"gpt.h.21.ln_2.weight": 0.008705759420990944,
|
| 180 |
+
"gpt.h.21.mlp.c_fc.bias": 0.0011871765600517392,
|
| 181 |
+
"gpt.h.21.mlp.c_fc.weight": 0.0038676916155964136,
|
| 182 |
+
"gpt.h.21.mlp.c_proj.bias": 0.0013832782860845327,
|
| 183 |
+
"gpt.h.21.mlp.c_proj.weight": 0.011286932043731213,
|
| 184 |
+
"gpt.h.22.attn.c_attn.bias": 0.0022080065682530403,
|
| 185 |
+
"gpt.h.22.attn.c_attn.weight": 0.00387361366301775,
|
| 186 |
+
"gpt.h.22.attn.c_proj.bias": 0.0037303422577679157,
|
| 187 |
+
"gpt.h.22.attn.c_proj.weight": 0.00496130483224988,
|
| 188 |
+
"gpt.h.22.ln_1.bias": 0.006414500530809164,
|
| 189 |
+
"gpt.h.22.ln_1.weight": 0.008626559749245644,
|
| 190 |
+
"gpt.h.22.ln_2.bias": 0.006547602824866772,
|
| 191 |
+
"gpt.h.22.ln_2.weight": 0.008793600834906101,
|
| 192 |
+
"gpt.h.22.mlp.c_fc.bias": 0.0009473760146647692,
|
| 193 |
+
"gpt.h.22.mlp.c_fc.weight": 0.00415464723482728,
|
| 194 |
+
"gpt.h.22.mlp.c_proj.bias": 0.0013687954051420093,
|
| 195 |
+
"gpt.h.22.mlp.c_proj.weight": 0.009641721844673157,
|
| 196 |
+
"gpt.h.23.attn.c_attn.bias": 0.002031621988862753,
|
| 197 |
+
"gpt.h.23.attn.c_attn.weight": 0.004895014222711325,
|
| 198 |
+
"gpt.h.23.attn.c_proj.bias": 0.0015391460619866848,
|
| 199 |
+
"gpt.h.23.attn.c_proj.weight": 0.005748114548623562,
|
| 200 |
+
"gpt.h.23.ln_1.bias": 0.005657768342643976,
|
| 201 |
+
"gpt.h.23.ln_1.weight": 0.00802789069712162,
|
| 202 |
+
"gpt.h.23.ln_2.bias": 0.006499188020825386,
|
| 203 |
+
"gpt.h.23.ln_2.weight": 0.008431111462414265,
|
| 204 |
+
"gpt.h.23.mlp.c_fc.bias": 0.0010094406316056848,
|
| 205 |
+
"gpt.h.23.mlp.c_fc.weight": 0.003703867318108678,
|
| 206 |
+
"gpt.h.23.mlp.c_proj.bias": 0.0017025587148964405,
|
| 207 |
+
"gpt.h.23.mlp.c_proj.weight": 0.0074964226223528385,
|
| 208 |
+
"gpt.h.24.attn.c_attn.bias": 0.002305609406903386,
|
| 209 |
+
"gpt.h.24.attn.c_attn.weight": 0.0031796761322766542,
|
| 210 |
+
"gpt.h.24.attn.c_proj.bias": 0.001652697566896677,
|
| 211 |
+
"gpt.h.24.attn.c_proj.weight": 0.0052749463357031345,
|
| 212 |
+
"gpt.h.24.ln_1.bias": 0.005675694905221462,
|
| 213 |
+
"gpt.h.24.ln_1.weight": 0.009069245308637619,
|
| 214 |
+
"gpt.h.24.ln_2.bias": 0.006356207653880119,
|
| 215 |
+
"gpt.h.24.ln_2.weight": 0.009231998585164547,
|
| 216 |
+
"gpt.h.24.mlp.c_fc.bias": 0.001341491937637329,
|
| 217 |
+
"gpt.h.24.mlp.c_fc.weight": 0.003126475727185607,
|
| 218 |
+
"gpt.h.24.mlp.c_proj.bias": 0.0017264962662011385,
|
| 219 |
+
"gpt.h.24.mlp.c_proj.weight": 0.004613065160810947,
|
| 220 |
+
"gpt.h.25.attn.c_attn.bias": 0.002419382333755493,
|
| 221 |
+
"gpt.h.25.attn.c_attn.weight": 0.003242518287152052,
|
| 222 |
+
"gpt.h.25.attn.c_proj.bias": 0.0019212045008316636,
|
| 223 |
+
"gpt.h.25.attn.c_proj.weight": 0.004869922064244747,
|
| 224 |
+
"gpt.h.25.ln_1.bias": 0.005121177528053522,
|
| 225 |
+
"gpt.h.25.ln_1.weight": 0.009873942472040653,
|
| 226 |
+
"gpt.h.25.ln_2.bias": 0.006484670098870993,
|
| 227 |
+
"gpt.h.25.ln_2.weight": 0.009337698109447956,
|
| 228 |
+
"gpt.h.25.mlp.c_fc.bias": 0.001216022646985948,
|
| 229 |
+
"gpt.h.25.mlp.c_fc.weight": 0.0038542866241186857,
|
| 230 |
+
"gpt.h.25.mlp.c_proj.bias": 0.002242410322651267,
|
| 231 |
+
"gpt.h.25.mlp.c_proj.weight": 0.005041209515184164,
|
| 232 |
+
"gpt.h.26.attn.c_attn.bias": 0.00450435234233737,
|
| 233 |
+
"gpt.h.26.attn.c_attn.weight": 0.0034188455902040005,
|
| 234 |
+
"gpt.h.26.attn.c_proj.bias": 0.002108540153130889,
|
| 235 |
+
"gpt.h.26.attn.c_proj.weight": 0.005954994820058346,
|
| 236 |
+
"gpt.h.26.ln_1.bias": 0.004473517648875713,
|
| 237 |
+
"gpt.h.26.ln_1.weight": 0.009424147196114063,
|
| 238 |
+
"gpt.h.26.ln_2.bias": 0.006467711646109819,
|
| 239 |
+
"gpt.h.26.ln_2.weight": 0.009836550801992416,
|
| 240 |
+
"gpt.h.26.mlp.c_fc.bias": 0.001742089632898569,
|
| 241 |
+
"gpt.h.26.mlp.c_fc.weight": 0.0029066475108265877,
|
| 242 |
+
"gpt.h.26.mlp.c_proj.bias": 0.002795786364004016,
|
| 243 |
+
"gpt.h.26.mlp.c_proj.weight": 0.00789736956357956,
|
| 244 |
+
"gpt.h.27.attn.c_attn.bias": 0.0031362660229206085,
|
| 245 |
+
"gpt.h.27.attn.c_attn.weight": 0.003957696724683046,
|
| 246 |
+
"gpt.h.27.attn.c_proj.bias": 0.0029156384989619255,
|
| 247 |
+
"gpt.h.27.attn.c_proj.weight": 0.005406576208770275,
|
| 248 |
+
"gpt.h.27.ln_1.bias": 0.004558059852570295,
|
| 249 |
+
"gpt.h.27.ln_1.weight": 0.0094691701233387,
|
| 250 |
+
"gpt.h.27.ln_2.bias": 0.006574852392077446,
|
| 251 |
+
"gpt.h.27.ln_2.weight": 0.009961778298020363,
|
| 252 |
+
"gpt.h.27.mlp.c_fc.bias": 0.0016685383161529899,
|
| 253 |
+
"gpt.h.27.mlp.c_fc.weight": 0.0029884425457566977,
|
| 254 |
+
"gpt.h.27.mlp.c_proj.bias": 0.002400637837126851,
|
| 255 |
+
"gpt.h.27.mlp.c_proj.weight": 0.008324341848492622,
|
| 256 |
+
"gpt.h.28.attn.c_attn.bias": 0.0024894936941564083,
|
| 257 |
+
"gpt.h.28.attn.c_attn.weight": 0.003081751521676779,
|
| 258 |
+
"gpt.h.28.attn.c_proj.bias": 0.004242106340825558,
|
| 259 |
+
"gpt.h.28.attn.c_proj.weight": 0.009405862540006638,
|
| 260 |
+
"gpt.h.28.ln_1.bias": 0.004887533839792013,
|
| 261 |
+
"gpt.h.28.ln_1.weight": 0.009970242157578468,
|
| 262 |
+
"gpt.h.28.ln_2.bias": 0.005814329255372286,
|
| 263 |
+
"gpt.h.28.ln_2.weight": 0.010871784761548042,
|
| 264 |
+
"gpt.h.28.mlp.c_fc.bias": 0.0019840362947434187,
|
| 265 |
+
"gpt.h.28.mlp.c_fc.weight": 0.008523846976459026,
|
| 266 |
+
"gpt.h.28.mlp.c_proj.bias": 0.005565112456679344,
|
| 267 |
+
"gpt.h.28.mlp.c_proj.weight": 0.032014038413763046,
|
| 268 |
+
"gpt.h.29.attn.c_attn.bias": 0.00337958219461143,
|
| 269 |
+
"gpt.h.29.attn.c_attn.weight": 0.004022927954792976,
|
| 270 |
+
"gpt.h.29.attn.c_proj.bias": 0.0018416948150843382,
|
| 271 |
+
"gpt.h.29.attn.c_proj.weight": 0.018323034048080444,
|
| 272 |
+
"gpt.h.29.ln_1.bias": 0.004602092783898115,
|
| 273 |
+
"gpt.h.29.ln_1.weight": 0.009674133732914925,
|
| 274 |
+
"gpt.h.29.ln_2.bias": 0.003327032318338752,
|
| 275 |
+
"gpt.h.29.ln_2.weight": 0.01144911628216505,
|
| 276 |
+
"gpt.h.29.mlp.c_fc.bias": 0.002207099460065365,
|
| 277 |
+
"gpt.h.29.mlp.c_fc.weight": 0.014188914559781551,
|
| 278 |
+
"gpt.h.29.mlp.c_proj.bias": 0.00442220363765955,
|
| 279 |
+
"gpt.h.29.mlp.c_proj.weight": 0.0867144837975502,
|
| 280 |
+
"gpt.h.3.attn.c_attn.bias": 0.002024676650762558,
|
| 281 |
+
"gpt.h.3.attn.c_attn.weight": 0.004979935009032488,
|
| 282 |
+
"gpt.h.3.attn.c_proj.bias": 0.021005922928452492,
|
| 283 |
+
"gpt.h.3.attn.c_proj.weight": 0.028328293934464455,
|
| 284 |
+
"gpt.h.3.ln_1.bias": 0.008041512221097946,
|
| 285 |
+
"gpt.h.3.ln_1.weight": 0.006430978886783123,
|
| 286 |
+
"gpt.h.3.ln_2.bias": 0.006599007174372673,
|
| 287 |
+
"gpt.h.3.ln_2.weight": 0.013134118169546127,
|
| 288 |
+
"gpt.h.3.mlp.c_fc.bias": 0.002585547510534525,
|
| 289 |
+
"gpt.h.3.mlp.c_fc.weight": 0.009125065989792347,
|
| 290 |
+
"gpt.h.3.mlp.c_proj.bias": 0.01302343886345625,
|
| 291 |
+
"gpt.h.3.mlp.c_proj.weight": 0.05233204737305641,
|
| 292 |
+
"gpt.h.4.attn.c_attn.bias": 0.002189196879044175,
|
| 293 |
+
"gpt.h.4.attn.c_attn.weight": 0.003678702749311924,
|
| 294 |
+
"gpt.h.4.attn.c_proj.bias": 0.017389798536896706,
|
| 295 |
+
"gpt.h.4.attn.c_proj.weight": 0.01729857549071312,
|
| 296 |
+
"gpt.h.4.ln_1.bias": 0.007602763827890158,
|
| 297 |
+
"gpt.h.4.ln_1.weight": 0.0074179875664412975,
|
| 298 |
+
"gpt.h.4.ln_2.bias": 0.004727063234895468,
|
| 299 |
+
"gpt.h.4.ln_2.weight": 0.013179998844861984,
|
| 300 |
+
"gpt.h.4.mlp.c_fc.bias": 0.0009291492169722915,
|
| 301 |
+
"gpt.h.4.mlp.c_fc.weight": 0.005233710631728172,
|
| 302 |
+
"gpt.h.4.mlp.c_proj.bias": 0.007813659496605396,
|
| 303 |
+
"gpt.h.4.mlp.c_proj.weight": 0.056338295340538025,
|
| 304 |
+
"gpt.h.5.attn.c_attn.bias": 0.0016779249999672174,
|
| 305 |
+
"gpt.h.5.attn.c_attn.weight": 0.0030687786638736725,
|
| 306 |
+
"gpt.h.5.attn.c_proj.bias": 0.009757072664797306,
|
| 307 |
+
"gpt.h.5.attn.c_proj.weight": 0.007529787719249725,
|
| 308 |
+
"gpt.h.5.ln_1.bias": 0.006044740788638592,
|
| 309 |
+
"gpt.h.5.ln_1.weight": 0.009458239190280437,
|
| 310 |
+
"gpt.h.5.ln_2.bias": 0.0036309966817498207,
|
| 311 |
+
"gpt.h.5.ln_2.weight": 0.013853815384209156,
|
| 312 |
+
"gpt.h.5.mlp.c_fc.bias": 0.0011706397635862231,
|
| 313 |
+
"gpt.h.5.mlp.c_fc.weight": 0.004535002168267965,
|
| 314 |
+
"gpt.h.5.mlp.c_proj.bias": 0.00503140315413475,
|
| 315 |
+
"gpt.h.5.mlp.c_proj.weight": 0.04938670992851257,
|
| 316 |
+
"gpt.h.6.attn.c_attn.bias": 0.0022753833327442408,
|
| 317 |
+
"gpt.h.6.attn.c_attn.weight": 0.0051308972761034966,
|
| 318 |
+
"gpt.h.6.attn.c_proj.bias": 0.005487145856022835,
|
| 319 |
+
"gpt.h.6.attn.c_proj.weight": 0.005496611353009939,
|
| 320 |
+
"gpt.h.6.ln_1.bias": 0.0058417185209691525,
|
| 321 |
+
"gpt.h.6.ln_1.weight": 0.011210721917450428,
|
| 322 |
+
"gpt.h.6.ln_2.bias": 0.004622064530849457,
|
| 323 |
+
"gpt.h.6.ln_2.weight": 0.01402156613767147,
|
| 324 |
+
"gpt.h.6.mlp.c_fc.bias": 0.0014565958408638835,
|
| 325 |
+
"gpt.h.6.mlp.c_fc.weight": 0.006464360281825066,
|
| 326 |
+
"gpt.h.6.mlp.c_proj.bias": 0.004647988360375166,
|
| 327 |
+
"gpt.h.6.mlp.c_proj.weight": 0.048505980521440506,
|
| 328 |
+
"gpt.h.7.attn.c_attn.bias": 0.002493516309186816,
|
| 329 |
+
"gpt.h.7.attn.c_attn.weight": 0.005425372160971165,
|
| 330 |
+
"gpt.h.7.attn.c_proj.bias": 0.005013930611312389,
|
| 331 |
+
"gpt.h.7.attn.c_proj.weight": 0.0051925997249782085,
|
| 332 |
+
"gpt.h.7.ln_1.bias": 0.008432278409600258,
|
| 333 |
+
"gpt.h.7.ln_1.weight": 0.0083107715472579,
|
| 334 |
+
"gpt.h.7.ln_2.bias": 0.007790034171193838,
|
| 335 |
+
"gpt.h.7.ln_2.weight": 0.01420080941170454,
|
| 336 |
+
"gpt.h.7.mlp.c_fc.bias": 0.0018345315475016832,
|
| 337 |
+
"gpt.h.7.mlp.c_fc.weight": 0.0038694925606250763,
|
| 338 |
+
"gpt.h.7.mlp.c_proj.bias": 0.005904932040721178,
|
| 339 |
+
"gpt.h.7.mlp.c_proj.weight": 0.023479172959923744,
|
| 340 |
+
"gpt.h.8.attn.c_attn.bias": 0.00208841310814023,
|
| 341 |
+
"gpt.h.8.attn.c_attn.weight": 0.004913092590868473,
|
| 342 |
+
"gpt.h.8.attn.c_proj.bias": 0.00487521244212985,
|
| 343 |
+
"gpt.h.8.attn.c_proj.weight": 0.0052477335557341576,
|
| 344 |
+
"gpt.h.8.ln_1.bias": 0.006899502594023943,
|
| 345 |
+
"gpt.h.8.ln_1.weight": 0.009084747172892094,
|
| 346 |
+
"gpt.h.8.ln_2.bias": 0.007770676631480455,
|
| 347 |
+
"gpt.h.8.ln_2.weight": 0.013765494339168072,
|
| 348 |
+
"gpt.h.8.mlp.c_fc.bias": 0.0014608169440180063,
|
| 349 |
+
"gpt.h.8.mlp.c_fc.weight": 0.0037384077440947294,
|
| 350 |
+
"gpt.h.8.mlp.c_proj.bias": 0.005353797692805529,
|
| 351 |
+
"gpt.h.8.mlp.c_proj.weight": 0.02410462312400341,
|
| 352 |
+
"gpt.h.9.attn.c_attn.bias": 0.0024391484912484884,
|
| 353 |
+
"gpt.h.9.attn.c_attn.weight": 0.0065130204893648624,
|
| 354 |
+
"gpt.h.9.attn.c_proj.bias": 0.003902623662725091,
|
| 355 |
+
"gpt.h.9.attn.c_proj.weight": 0.0059540290385484695,
|
| 356 |
+
"gpt.h.9.ln_1.bias": 0.008465644903481007,
|
| 357 |
+
"gpt.h.9.ln_1.weight": 0.008637243881821632,
|
| 358 |
+
"gpt.h.9.ln_2.bias": 0.007591988891363144,
|
| 359 |
+
"gpt.h.9.ln_2.weight": 0.014198592863976955,
|
| 360 |
+
"gpt.h.9.mlp.c_fc.bias": 0.0017329183174297214,
|
| 361 |
+
"gpt.h.9.mlp.c_fc.weight": 0.004214098211377859,
|
| 362 |
+
"gpt.h.9.mlp.c_proj.bias": 0.005952149163931608,
|
| 363 |
+
"gpt.h.9.mlp.c_proj.weight": 0.024082494899630547,
|
| 364 |
+
"gpt.ln_f.bias": 0.011065124534070492,
|
| 365 |
+
"gpt.ln_f.weight": 0.022843239828944206,
|
| 366 |
+
"gpt.wpe.emb.weight": 0.004260858986526728,
|
| 367 |
+
"gpt.wte.weight": 0.0044210017658770084,
|
| 368 |
+
"mel_head.bias": 0.0015680863289162517,
|
| 369 |
+
"mel_head.weight": 0.004180184565484524
|
| 370 |
+
}
|
INT8/special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[START]",
|
| 3 |
+
"eos_token": "[STOP]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"unk_token": "[UNK]"
|
| 6 |
+
}
|
INT8/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
INT8/tokenizer_config.json
ADDED
|
@@ -0,0 +1,192 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[STOP]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "[SPACE]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"259": {
|
| 28 |
+
"content": "[en]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"260": {
|
| 36 |
+
"content": "[de]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"261": {
|
| 44 |
+
"content": "[START]",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
},
|
| 51 |
+
"262": {
|
| 52 |
+
"content": "[fr]",
|
| 53 |
+
"lstrip": false,
|
| 54 |
+
"normalized": false,
|
| 55 |
+
"rstrip": false,
|
| 56 |
+
"single_word": false,
|
| 57 |
+
"special": true
|
| 58 |
+
},
|
| 59 |
+
"267": {
|
| 60 |
+
"content": "[ru]",
|
| 61 |
+
"lstrip": false,
|
| 62 |
+
"normalized": false,
|
| 63 |
+
"rstrip": false,
|
| 64 |
+
"single_word": false,
|
| 65 |
+
"special": true
|
| 66 |
+
},
|
| 67 |
+
"284": {
|
| 68 |
+
"content": "[es]",
|
| 69 |
+
"lstrip": false,
|
| 70 |
+
"normalized": false,
|
| 71 |
+
"rstrip": false,
|
| 72 |
+
"single_word": false,
|
| 73 |
+
"special": true
|
| 74 |
+
},
|
| 75 |
+
"285": {
|
| 76 |
+
"content": "[it]",
|
| 77 |
+
"lstrip": false,
|
| 78 |
+
"normalized": false,
|
| 79 |
+
"rstrip": false,
|
| 80 |
+
"single_word": false,
|
| 81 |
+
"special": true
|
| 82 |
+
},
|
| 83 |
+
"286": {
|
| 84 |
+
"content": "[pt]",
|
| 85 |
+
"lstrip": false,
|
| 86 |
+
"normalized": false,
|
| 87 |
+
"rstrip": false,
|
| 88 |
+
"single_word": false,
|
| 89 |
+
"special": true
|
| 90 |
+
},
|
| 91 |
+
"293": {
|
| 92 |
+
"content": "[cs]",
|
| 93 |
+
"lstrip": false,
|
| 94 |
+
"normalized": false,
|
| 95 |
+
"rstrip": false,
|
| 96 |
+
"single_word": false,
|
| 97 |
+
"special": true
|
| 98 |
+
},
|
| 99 |
+
"294": {
|
| 100 |
+
"content": "[pl]",
|
| 101 |
+
"lstrip": false,
|
| 102 |
+
"normalized": false,
|
| 103 |
+
"rstrip": false,
|
| 104 |
+
"single_word": false,
|
| 105 |
+
"special": true
|
| 106 |
+
},
|
| 107 |
+
"295": {
|
| 108 |
+
"content": "[tr]",
|
| 109 |
+
"lstrip": false,
|
| 110 |
+
"normalized": false,
|
| 111 |
+
"rstrip": false,
|
| 112 |
+
"single_word": false,
|
| 113 |
+
"special": true
|
| 114 |
+
},
|
| 115 |
+
"297": {
|
| 116 |
+
"content": "[nl]",
|
| 117 |
+
"lstrip": false,
|
| 118 |
+
"normalized": false,
|
| 119 |
+
"rstrip": false,
|
| 120 |
+
"single_word": false,
|
| 121 |
+
"special": true
|
| 122 |
+
},
|
| 123 |
+
"5022": {
|
| 124 |
+
"content": "[ar]",
|
| 125 |
+
"lstrip": false,
|
| 126 |
+
"normalized": false,
|
| 127 |
+
"rstrip": false,
|
| 128 |
+
"single_word": false,
|
| 129 |
+
"special": true
|
| 130 |
+
},
|
| 131 |
+
"5023": {
|
| 132 |
+
"content": "[zh-cn]",
|
| 133 |
+
"lstrip": false,
|
| 134 |
+
"normalized": false,
|
| 135 |
+
"rstrip": false,
|
| 136 |
+
"single_word": false,
|
| 137 |
+
"special": true
|
| 138 |
+
},
|
| 139 |
+
"5412": {
|
| 140 |
+
"content": "[ja]",
|
| 141 |
+
"lstrip": false,
|
| 142 |
+
"normalized": false,
|
| 143 |
+
"rstrip": false,
|
| 144 |
+
"single_word": false,
|
| 145 |
+
"special": true
|
| 146 |
+
},
|
| 147 |
+
"5753": {
|
| 148 |
+
"content": "[hu]",
|
| 149 |
+
"lstrip": false,
|
| 150 |
+
"normalized": false,
|
| 151 |
+
"rstrip": false,
|
| 152 |
+
"single_word": false,
|
| 153 |
+
"special": true
|
| 154 |
+
},
|
| 155 |
+
"6152": {
|
| 156 |
+
"content": "[ko]",
|
| 157 |
+
"lstrip": false,
|
| 158 |
+
"normalized": false,
|
| 159 |
+
"rstrip": false,
|
| 160 |
+
"single_word": false,
|
| 161 |
+
"special": true
|
| 162 |
+
},
|
| 163 |
+
"6680": {
|
| 164 |
+
"content": "[hi]",
|
| 165 |
+
"lstrip": false,
|
| 166 |
+
"normalized": false,
|
| 167 |
+
"rstrip": false,
|
| 168 |
+
"single_word": false,
|
| 169 |
+
"special": true
|
| 170 |
+
},
|
| 171 |
+
"6681": {
|
| 172 |
+
"content": "[PAD]",
|
| 173 |
+
"lstrip": false,
|
| 174 |
+
"normalized": false,
|
| 175 |
+
"rstrip": false,
|
| 176 |
+
"single_word": false,
|
| 177 |
+
"special": true
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"auto_map": {"AutoTokenizer": ["AstraMindAI/xtts2-gpt--tokenizer.XTTSTokenizerFast", null]},
|
| 181 |
+
"bos_token": "[START]",
|
| 182 |
+
"clean_up_tokenization_spaces": true,
|
| 183 |
+
"eos_token": "[STOP]",
|
| 184 |
+
"max_length": null,
|
| 185 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 186 |
+
"pad_to_multiple_of": null,
|
| 187 |
+
"pad_token": "[PAD]",
|
| 188 |
+
"pad_token_type_id": 0,
|
| 189 |
+
"padding_side": "right",
|
| 190 |
+
"tokenizer_class": "XTTSTokenizerFast",
|
| 191 |
+
"unk_token": "[UNK]"
|
| 192 |
+
}
|