Upload with huggingface_hub
Browse files- config.json +256 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +753 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,256 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "facebook/hubert-base-ls960",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"apply_spec_augment": true,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"HubertForSequenceClassification"
|
| 7 |
+
],
|
| 8 |
+
"attention_dropout": 0.1,
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
+
"classifier_proj_size": 256,
|
| 11 |
+
"conv_bias": false,
|
| 12 |
+
"conv_dim": [
|
| 13 |
+
512,
|
| 14 |
+
512,
|
| 15 |
+
512,
|
| 16 |
+
512,
|
| 17 |
+
512,
|
| 18 |
+
512,
|
| 19 |
+
512
|
| 20 |
+
],
|
| 21 |
+
"conv_kernel": [
|
| 22 |
+
10,
|
| 23 |
+
3,
|
| 24 |
+
3,
|
| 25 |
+
3,
|
| 26 |
+
3,
|
| 27 |
+
2,
|
| 28 |
+
2
|
| 29 |
+
],
|
| 30 |
+
"conv_stride": [
|
| 31 |
+
5,
|
| 32 |
+
2,
|
| 33 |
+
2,
|
| 34 |
+
2,
|
| 35 |
+
2,
|
| 36 |
+
2,
|
| 37 |
+
2
|
| 38 |
+
],
|
| 39 |
+
"ctc_loss_reduction": "sum",
|
| 40 |
+
"ctc_zero_infinity": false,
|
| 41 |
+
"do_stable_layer_norm": false,
|
| 42 |
+
"eos_token_id": 2,
|
| 43 |
+
"feat_extract_activation": "gelu",
|
| 44 |
+
"feat_extract_dropout": 0.0,
|
| 45 |
+
"feat_extract_norm": "group",
|
| 46 |
+
"feat_proj_dropout": 0.1,
|
| 47 |
+
"feat_proj_layer_norm": true,
|
| 48 |
+
"final_dropout": 0.1,
|
| 49 |
+
"gradient_checkpointing": false,
|
| 50 |
+
"hidden_act": "gelu",
|
| 51 |
+
"hidden_dropout": 0.1,
|
| 52 |
+
"hidden_dropout_prob": 0.1,
|
| 53 |
+
"hidden_size": 768,
|
| 54 |
+
"id2label": {
|
| 55 |
+
"0": "calendar_set",
|
| 56 |
+
"1": "wavs/audio_volume_mute",
|
| 57 |
+
"10": "iot_coffee",
|
| 58 |
+
"11": "calendar_remove",
|
| 59 |
+
"12": "email_sendemail",
|
| 60 |
+
"13": "general_joke",
|
| 61 |
+
"14": "lists_query",
|
| 62 |
+
"15": "iot_cleaning",
|
| 63 |
+
"16": "social_query",
|
| 64 |
+
"17": "cooking_recipe",
|
| 65 |
+
"18": "play_game",
|
| 66 |
+
"19": "weather_query",
|
| 67 |
+
"2": "qa_stock",
|
| 68 |
+
"20": "iot_hue_lightoff",
|
| 69 |
+
"21": "qa_factoid",
|
| 70 |
+
"22": "play_wavs/audiobook",
|
| 71 |
+
"23": "news_query",
|
| 72 |
+
"24": "qa_maths",
|
| 73 |
+
"25": "email_query",
|
| 74 |
+
"26": "recommendation_movies",
|
| 75 |
+
"27": "transport_traffic",
|
| 76 |
+
"28": "takeaway_order",
|
| 77 |
+
"29": "wavs/audio_volume_other",
|
| 78 |
+
"3": "social_post",
|
| 79 |
+
"30": "datetime_convert",
|
| 80 |
+
"31": "music",
|
| 81 |
+
"32": "recommendation_locations",
|
| 82 |
+
"33": "recommendation_events",
|
| 83 |
+
"34": "qa_currency",
|
| 84 |
+
"35": "lists_createoradd",
|
| 85 |
+
"36": "datetime_query",
|
| 86 |
+
"37": "transport_ticket",
|
| 87 |
+
"38": "takeaway_query",
|
| 88 |
+
"39": "general_greet",
|
| 89 |
+
"4": "play_radio",
|
| 90 |
+
"40": "qa_definition",
|
| 91 |
+
"41": "play_podcasts",
|
| 92 |
+
"42": "transport_taxi",
|
| 93 |
+
"43": "alarm_remove",
|
| 94 |
+
"44": "iot_hue_lightchange",
|
| 95 |
+
"45": "email_querycontact",
|
| 96 |
+
"46": "iot_hue_lightdim",
|
| 97 |
+
"47": "alarm_set",
|
| 98 |
+
"48": "iot_hue_lightup",
|
| 99 |
+
"49": "transport_query",
|
| 100 |
+
"5": "calendar_query",
|
| 101 |
+
"50": "iot_wemo_on",
|
| 102 |
+
"51": "music_likeness",
|
| 103 |
+
"52": "alarm_query",
|
| 104 |
+
"53": "music_dislikeness",
|
| 105 |
+
"54": "lists_remove",
|
| 106 |
+
"55": "iot_hue_lighton",
|
| 107 |
+
"56": "wavs/audio_volume_down",
|
| 108 |
+
"57": "factoid",
|
| 109 |
+
"58": "iot_wemo_off",
|
| 110 |
+
"59": "query",
|
| 111 |
+
"6": "music_query",
|
| 112 |
+
"60": "set",
|
| 113 |
+
"61": "quirky",
|
| 114 |
+
"62": "email_addcontact",
|
| 115 |
+
"63": "music_settings",
|
| 116 |
+
"64": "joke",
|
| 117 |
+
"65": "podcasts",
|
| 118 |
+
"66": "game",
|
| 119 |
+
"67": "coffee",
|
| 120 |
+
"68": "radio",
|
| 121 |
+
"69": "post",
|
| 122 |
+
"7": "general_quirky",
|
| 123 |
+
"70": "convert",
|
| 124 |
+
"71": "remove",
|
| 125 |
+
"72": "greet",
|
| 126 |
+
"73": "cooking_query",
|
| 127 |
+
"74": "sendemail",
|
| 128 |
+
"75": "traffic",
|
| 129 |
+
"76": "hue_lightup",
|
| 130 |
+
"77": "hue_lightoff",
|
| 131 |
+
"78": "currency",
|
| 132 |
+
"79": "wemo_off",
|
| 133 |
+
"8": "play_music",
|
| 134 |
+
"80": "hue_lightdim",
|
| 135 |
+
"81": "createoradd",
|
| 136 |
+
"82": "ticket",
|
| 137 |
+
"83": "volume_other",
|
| 138 |
+
"84": "cleaning",
|
| 139 |
+
"85": "querycontact",
|
| 140 |
+
"86": "wemo_on",
|
| 141 |
+
"87": "addcontact",
|
| 142 |
+
"9": "wavs/audio_volume_up"
|
| 143 |
+
},
|
| 144 |
+
"initializer_range": 0.02,
|
| 145 |
+
"intermediate_size": 3072,
|
| 146 |
+
"label2id": {
|
| 147 |
+
"addcontact": "87",
|
| 148 |
+
"alarm_query": "52",
|
| 149 |
+
"alarm_remove": "43",
|
| 150 |
+
"alarm_set": "47",
|
| 151 |
+
"calendar_query": "5",
|
| 152 |
+
"calendar_remove": "11",
|
| 153 |
+
"calendar_set": "0",
|
| 154 |
+
"cleaning": "84",
|
| 155 |
+
"coffee": "67",
|
| 156 |
+
"convert": "70",
|
| 157 |
+
"cooking_query": "73",
|
| 158 |
+
"cooking_recipe": "17",
|
| 159 |
+
"createoradd": "81",
|
| 160 |
+
"currency": "78",
|
| 161 |
+
"datetime_convert": "30",
|
| 162 |
+
"datetime_query": "36",
|
| 163 |
+
"email_addcontact": "62",
|
| 164 |
+
"email_query": "25",
|
| 165 |
+
"email_querycontact": "45",
|
| 166 |
+
"email_sendemail": "12",
|
| 167 |
+
"factoid": "57",
|
| 168 |
+
"game": "66",
|
| 169 |
+
"general_greet": "39",
|
| 170 |
+
"general_joke": "13",
|
| 171 |
+
"general_quirky": "7",
|
| 172 |
+
"greet": "72",
|
| 173 |
+
"hue_lightdim": "80",
|
| 174 |
+
"hue_lightoff": "77",
|
| 175 |
+
"hue_lightup": "76",
|
| 176 |
+
"iot_cleaning": "15",
|
| 177 |
+
"iot_coffee": "10",
|
| 178 |
+
"iot_hue_lightchange": "44",
|
| 179 |
+
"iot_hue_lightdim": "46",
|
| 180 |
+
"iot_hue_lightoff": "20",
|
| 181 |
+
"iot_hue_lighton": "55",
|
| 182 |
+
"iot_hue_lightup": "48",
|
| 183 |
+
"iot_wemo_off": "58",
|
| 184 |
+
"iot_wemo_on": "50",
|
| 185 |
+
"joke": "64",
|
| 186 |
+
"lists_createoradd": "35",
|
| 187 |
+
"lists_query": "14",
|
| 188 |
+
"lists_remove": "54",
|
| 189 |
+
"music": "31",
|
| 190 |
+
"music_dislikeness": "53",
|
| 191 |
+
"music_likeness": "51",
|
| 192 |
+
"music_query": "6",
|
| 193 |
+
"music_settings": "63",
|
| 194 |
+
"news_query": "23",
|
| 195 |
+
"play_game": "18",
|
| 196 |
+
"play_music": "8",
|
| 197 |
+
"play_podcasts": "41",
|
| 198 |
+
"play_radio": "4",
|
| 199 |
+
"play_wavs/audiobook": "22",
|
| 200 |
+
"podcasts": "65",
|
| 201 |
+
"post": "69",
|
| 202 |
+
"qa_currency": "34",
|
| 203 |
+
"qa_definition": "40",
|
| 204 |
+
"qa_factoid": "21",
|
| 205 |
+
"qa_maths": "24",
|
| 206 |
+
"qa_stock": "2",
|
| 207 |
+
"query": "59",
|
| 208 |
+
"querycontact": "85",
|
| 209 |
+
"quirky": "61",
|
| 210 |
+
"radio": "68",
|
| 211 |
+
"recommendation_events": "33",
|
| 212 |
+
"recommendation_locations": "32",
|
| 213 |
+
"recommendation_movies": "26",
|
| 214 |
+
"remove": "71",
|
| 215 |
+
"sendemail": "74",
|
| 216 |
+
"set": "60",
|
| 217 |
+
"social_post": "3",
|
| 218 |
+
"social_query": "16",
|
| 219 |
+
"takeaway_order": "28",
|
| 220 |
+
"takeaway_query": "38",
|
| 221 |
+
"ticket": "82",
|
| 222 |
+
"traffic": "75",
|
| 223 |
+
"transport_query": "49",
|
| 224 |
+
"transport_taxi": "42",
|
| 225 |
+
"transport_ticket": "37",
|
| 226 |
+
"transport_traffic": "27",
|
| 227 |
+
"volume_other": "83",
|
| 228 |
+
"wavs/audio_volume_down": "56",
|
| 229 |
+
"wavs/audio_volume_mute": "1",
|
| 230 |
+
"wavs/audio_volume_other": "29",
|
| 231 |
+
"wavs/audio_volume_up": "9",
|
| 232 |
+
"weather_query": "19",
|
| 233 |
+
"wemo_off": "79",
|
| 234 |
+
"wemo_on": "86"
|
| 235 |
+
},
|
| 236 |
+
"layer_norm_eps": 1e-05,
|
| 237 |
+
"layerdrop": 0.1,
|
| 238 |
+
"mask_feature_length": 10,
|
| 239 |
+
"mask_feature_min_masks": 0,
|
| 240 |
+
"mask_feature_prob": 0.0,
|
| 241 |
+
"mask_time_length": 10,
|
| 242 |
+
"mask_time_min_masks": 2,
|
| 243 |
+
"mask_time_prob": 0.05,
|
| 244 |
+
"model_type": "hubert",
|
| 245 |
+
"num_attention_heads": 12,
|
| 246 |
+
"num_conv_pos_embedding_groups": 16,
|
| 247 |
+
"num_conv_pos_embeddings": 128,
|
| 248 |
+
"num_feat_extract_layers": 7,
|
| 249 |
+
"num_hidden_layers": 12,
|
| 250 |
+
"pad_token_id": 0,
|
| 251 |
+
"tokenizer_class": "Wav2Vec2CTCTokenizer",
|
| 252 |
+
"torch_dtype": "float32",
|
| 253 |
+
"transformers_version": "4.45.2",
|
| 254 |
+
"use_weighted_layer_sum": false,
|
| 255 |
+
"vocab_size": 32
|
| 256 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4f8f088026e0ed9c4a766ffa8cfebbf5418c9b55a58455067897422968d702df
|
| 3 |
+
size 378390360
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fdda0cdc825b178a534dde33d849821546c8ef806dbb252eaff655aaa1159d2
|
| 3 |
+
size 756909370
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3f016aa6153d0b240518607641b991e11dfa02624ec267663efc236ac743d288
|
| 3 |
+
size 14308
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:417c32b40ce266eae8ab814d0af12409a7770b12ed2c5afd1b1ffbf9fa27b42f
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,753 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8415273271774395,
|
| 3 |
+
"best_model_checkpoint": "results_retain/facebook/hubert-base-ls960/42/checkpoint-30000",
|
| 4 |
+
"epoch": 69.20415224913495,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 30000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.1534025374855825,
|
| 13 |
+
"grad_norm": 2.9732823371887207,
|
| 14 |
+
"learning_rate": 8.333333333333333e-05,
|
| 15 |
+
"loss": 3.9827,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 2.306805074971165,
|
| 20 |
+
"grad_norm": 3.6868040561676025,
|
| 21 |
+
"learning_rate": 0.00016666666666666666,
|
| 22 |
+
"loss": 3.3559,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.306805074971165,
|
| 27 |
+
"eval_accuracy": 0.33953082106313953,
|
| 28 |
+
"eval_f1_macro": 0.12928496278744922,
|
| 29 |
+
"eval_loss": 2.5730652809143066,
|
| 30 |
+
"eval_runtime": 35.334,
|
| 31 |
+
"eval_samples_per_second": 226.807,
|
| 32 |
+
"eval_steps_per_second": 7.104,
|
| 33 |
+
"step": 1000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 3.4602076124567476,
|
| 37 |
+
"grad_norm": 7.014188766479492,
|
| 38 |
+
"learning_rate": 0.00025,
|
| 39 |
+
"loss": 2.1949,
|
| 40 |
+
"step": 1500
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 4.61361014994233,
|
| 44 |
+
"grad_norm": 6.496627330780029,
|
| 45 |
+
"learning_rate": 0.0003333333333333333,
|
| 46 |
+
"loss": 1.6389,
|
| 47 |
+
"step": 2000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 4.61361014994233,
|
| 51 |
+
"eval_accuracy": 0.6416271524831545,
|
| 52 |
+
"eval_f1_macro": 0.4299797469182877,
|
| 53 |
+
"eval_loss": 1.4779495000839233,
|
| 54 |
+
"eval_runtime": 35.3694,
|
| 55 |
+
"eval_samples_per_second": 226.58,
|
| 56 |
+
"eval_steps_per_second": 7.097,
|
| 57 |
+
"step": 2000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 5.767012687427912,
|
| 61 |
+
"grad_norm": 7.616945266723633,
|
| 62 |
+
"learning_rate": 0.0004166666666666667,
|
| 63 |
+
"loss": 1.4162,
|
| 64 |
+
"step": 2500
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 6.920415224913495,
|
| 68 |
+
"grad_norm": 8.488947868347168,
|
| 69 |
+
"learning_rate": 0.0005,
|
| 70 |
+
"loss": 1.3587,
|
| 71 |
+
"step": 3000
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 6.920415224913495,
|
| 75 |
+
"eval_accuracy": 0.6574744197654105,
|
| 76 |
+
"eval_f1_macro": 0.4595491108495356,
|
| 77 |
+
"eval_loss": 1.427338719367981,
|
| 78 |
+
"eval_runtime": 72.7464,
|
| 79 |
+
"eval_samples_per_second": 110.164,
|
| 80 |
+
"eval_steps_per_second": 3.45,
|
| 81 |
+
"step": 3000
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 8.073817762399077,
|
| 85 |
+
"grad_norm": 8.156586647033691,
|
| 86 |
+
"learning_rate": 0.0004907407407407408,
|
| 87 |
+
"loss": 1.3027,
|
| 88 |
+
"step": 3500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 9.22722029988466,
|
| 92 |
+
"grad_norm": 6.934875965118408,
|
| 93 |
+
"learning_rate": 0.00048148148148148144,
|
| 94 |
+
"loss": 1.1695,
|
| 95 |
+
"step": 4000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 9.22722029988466,
|
| 99 |
+
"eval_accuracy": 0.6961567257299726,
|
| 100 |
+
"eval_f1_macro": 0.5353451961331095,
|
| 101 |
+
"eval_loss": 1.2435524463653564,
|
| 102 |
+
"eval_runtime": 70.2906,
|
| 103 |
+
"eval_samples_per_second": 114.012,
|
| 104 |
+
"eval_steps_per_second": 3.571,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 10.380622837370241,
|
| 109 |
+
"grad_norm": 7.151013374328613,
|
| 110 |
+
"learning_rate": 0.00047222222222222224,
|
| 111 |
+
"loss": 1.0695,
|
| 112 |
+
"step": 4500
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 11.534025374855824,
|
| 116 |
+
"grad_norm": 7.771185874938965,
|
| 117 |
+
"learning_rate": 0.000462962962962963,
|
| 118 |
+
"loss": 0.9787,
|
| 119 |
+
"step": 5000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 11.534025374855824,
|
| 123 |
+
"eval_accuracy": 0.6901672073870726,
|
| 124 |
+
"eval_f1_macro": 0.5134333803516367,
|
| 125 |
+
"eval_loss": 1.3313419818878174,
|
| 126 |
+
"eval_runtime": 71.0472,
|
| 127 |
+
"eval_samples_per_second": 112.798,
|
| 128 |
+
"eval_steps_per_second": 3.533,
|
| 129 |
+
"step": 5000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 12.687427912341407,
|
| 133 |
+
"grad_norm": 6.754736423492432,
|
| 134 |
+
"learning_rate": 0.0004537037037037037,
|
| 135 |
+
"loss": 0.9056,
|
| 136 |
+
"step": 5500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 13.84083044982699,
|
| 140 |
+
"grad_norm": 6.322958946228027,
|
| 141 |
+
"learning_rate": 0.0004444444444444444,
|
| 142 |
+
"loss": 0.836,
|
| 143 |
+
"step": 6000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 13.84083044982699,
|
| 147 |
+
"eval_accuracy": 0.7138757174943848,
|
| 148 |
+
"eval_f1_macro": 0.5729793747297807,
|
| 149 |
+
"eval_loss": 1.2415224313735962,
|
| 150 |
+
"eval_runtime": 73.4693,
|
| 151 |
+
"eval_samples_per_second": 109.08,
|
| 152 |
+
"eval_steps_per_second": 3.416,
|
| 153 |
+
"step": 6000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 14.994232987312571,
|
| 157 |
+
"grad_norm": 7.2826619148254395,
|
| 158 |
+
"learning_rate": 0.0004351851851851852,
|
| 159 |
+
"loss": 0.7867,
|
| 160 |
+
"step": 6500
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 16.147635524798154,
|
| 164 |
+
"grad_norm": 5.9969482421875,
|
| 165 |
+
"learning_rate": 0.00042592592592592595,
|
| 166 |
+
"loss": 0.7135,
|
| 167 |
+
"step": 7000
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 16.147635524798154,
|
| 171 |
+
"eval_accuracy": 0.7389568255552783,
|
| 172 |
+
"eval_f1_macro": 0.5793519253003285,
|
| 173 |
+
"eval_loss": 1.1902633905410767,
|
| 174 |
+
"eval_runtime": 73.1754,
|
| 175 |
+
"eval_samples_per_second": 109.518,
|
| 176 |
+
"eval_steps_per_second": 3.43,
|
| 177 |
+
"step": 7000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 17.301038062283737,
|
| 181 |
+
"grad_norm": 6.838934421539307,
|
| 182 |
+
"learning_rate": 0.0004166666666666667,
|
| 183 |
+
"loss": 0.6719,
|
| 184 |
+
"step": 7500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 18.45444059976932,
|
| 188 |
+
"grad_norm": 7.784801006317139,
|
| 189 |
+
"learning_rate": 0.0004074074074074074,
|
| 190 |
+
"loss": 0.6009,
|
| 191 |
+
"step": 8000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 18.45444059976932,
|
| 195 |
+
"eval_accuracy": 0.7414524581981532,
|
| 196 |
+
"eval_f1_macro": 0.6081990369390977,
|
| 197 |
+
"eval_loss": 1.2159614562988281,
|
| 198 |
+
"eval_runtime": 55.5881,
|
| 199 |
+
"eval_samples_per_second": 144.168,
|
| 200 |
+
"eval_steps_per_second": 4.515,
|
| 201 |
+
"step": 8000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 19.607843137254903,
|
| 205 |
+
"grad_norm": 6.171660423278809,
|
| 206 |
+
"learning_rate": 0.0003981481481481481,
|
| 207 |
+
"loss": 0.5756,
|
| 208 |
+
"step": 8500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 20.761245674740483,
|
| 212 |
+
"grad_norm": 7.959474563598633,
|
| 213 |
+
"learning_rate": 0.0003888888888888889,
|
| 214 |
+
"loss": 0.5355,
|
| 215 |
+
"step": 9000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 20.761245674740483,
|
| 219 |
+
"eval_accuracy": 0.7543049663089593,
|
| 220 |
+
"eval_f1_macro": 0.5947199785680519,
|
| 221 |
+
"eval_loss": 1.1460059881210327,
|
| 222 |
+
"eval_runtime": 55.0321,
|
| 223 |
+
"eval_samples_per_second": 145.624,
|
| 224 |
+
"eval_steps_per_second": 4.561,
|
| 225 |
+
"step": 9000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 21.914648212226066,
|
| 229 |
+
"grad_norm": 7.287164211273193,
|
| 230 |
+
"learning_rate": 0.00037962962962962966,
|
| 231 |
+
"loss": 0.5046,
|
| 232 |
+
"step": 9500
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 23.06805074971165,
|
| 236 |
+
"grad_norm": 6.368403434753418,
|
| 237 |
+
"learning_rate": 0.00037037037037037035,
|
| 238 |
+
"loss": 0.4737,
|
| 239 |
+
"step": 10000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 23.06805074971165,
|
| 243 |
+
"eval_accuracy": 0.759920139755428,
|
| 244 |
+
"eval_f1_macro": 0.6214937740706044,
|
| 245 |
+
"eval_loss": 1.1644535064697266,
|
| 246 |
+
"eval_runtime": 54.1187,
|
| 247 |
+
"eval_samples_per_second": 148.082,
|
| 248 |
+
"eval_steps_per_second": 4.638,
|
| 249 |
+
"step": 10000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 24.22145328719723,
|
| 253 |
+
"grad_norm": 9.294144630432129,
|
| 254 |
+
"learning_rate": 0.0003611111111111111,
|
| 255 |
+
"loss": 0.4349,
|
| 256 |
+
"step": 10500
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 25.374855824682815,
|
| 260 |
+
"grad_norm": 7.4235310554504395,
|
| 261 |
+
"learning_rate": 0.0003518518518518519,
|
| 262 |
+
"loss": 0.4352,
|
| 263 |
+
"step": 11000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 25.374855824682815,
|
| 267 |
+
"eval_accuracy": 0.7545545295732469,
|
| 268 |
+
"eval_f1_macro": 0.5917892398903293,
|
| 269 |
+
"eval_loss": 1.213472843170166,
|
| 270 |
+
"eval_runtime": 54.2317,
|
| 271 |
+
"eval_samples_per_second": 147.773,
|
| 272 |
+
"eval_steps_per_second": 4.628,
|
| 273 |
+
"step": 11000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 26.528258362168398,
|
| 277 |
+
"grad_norm": 5.937560558319092,
|
| 278 |
+
"learning_rate": 0.00034259259259259263,
|
| 279 |
+
"loss": 0.4017,
|
| 280 |
+
"step": 11500
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 27.68166089965398,
|
| 284 |
+
"grad_norm": 6.036593914031982,
|
| 285 |
+
"learning_rate": 0.0003333333333333333,
|
| 286 |
+
"loss": 0.3652,
|
| 287 |
+
"step": 12000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 27.68166089965398,
|
| 291 |
+
"eval_accuracy": 0.7732717743948091,
|
| 292 |
+
"eval_f1_macro": 0.6375373960767734,
|
| 293 |
+
"eval_loss": 1.1644330024719238,
|
| 294 |
+
"eval_runtime": 75.4416,
|
| 295 |
+
"eval_samples_per_second": 106.228,
|
| 296 |
+
"eval_steps_per_second": 3.327,
|
| 297 |
+
"step": 12000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 28.83506343713956,
|
| 301 |
+
"grad_norm": 6.821892738342285,
|
| 302 |
+
"learning_rate": 0.00032407407407407406,
|
| 303 |
+
"loss": 0.3443,
|
| 304 |
+
"step": 12500
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 29.988465974625143,
|
| 308 |
+
"grad_norm": 4.1507463455200195,
|
| 309 |
+
"learning_rate": 0.0003148148148148148,
|
| 310 |
+
"loss": 0.3246,
|
| 311 |
+
"step": 13000
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 29.988465974625143,
|
| 315 |
+
"eval_accuracy": 0.7776391315198403,
|
| 316 |
+
"eval_f1_macro": 0.6181477901694947,
|
| 317 |
+
"eval_loss": 1.143282175064087,
|
| 318 |
+
"eval_runtime": 75.3981,
|
| 319 |
+
"eval_samples_per_second": 106.289,
|
| 320 |
+
"eval_steps_per_second": 3.329,
|
| 321 |
+
"step": 13000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 31.141868512110726,
|
| 325 |
+
"grad_norm": 7.311563491821289,
|
| 326 |
+
"learning_rate": 0.0003055555555555556,
|
| 327 |
+
"loss": 0.3082,
|
| 328 |
+
"step": 13500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 32.29527104959631,
|
| 332 |
+
"grad_norm": 2.214399576187134,
|
| 333 |
+
"learning_rate": 0.0002962962962962963,
|
| 334 |
+
"loss": 0.2876,
|
| 335 |
+
"step": 14000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 32.29527104959631,
|
| 339 |
+
"eval_accuracy": 0.7700274519590716,
|
| 340 |
+
"eval_f1_macro": 0.6278465966595438,
|
| 341 |
+
"eval_loss": 1.2212963104248047,
|
| 342 |
+
"eval_runtime": 73.2438,
|
| 343 |
+
"eval_samples_per_second": 109.415,
|
| 344 |
+
"eval_steps_per_second": 3.427,
|
| 345 |
+
"step": 14000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 33.44867358708189,
|
| 349 |
+
"grad_norm": 5.876758575439453,
|
| 350 |
+
"learning_rate": 0.00028703703703703703,
|
| 351 |
+
"loss": 0.2722,
|
| 352 |
+
"step": 14500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 34.602076124567475,
|
| 356 |
+
"grad_norm": 3.34192156791687,
|
| 357 |
+
"learning_rate": 0.0002777777777777778,
|
| 358 |
+
"loss": 0.2539,
|
| 359 |
+
"step": 15000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 34.602076124567475,
|
| 363 |
+
"eval_accuracy": 0.7858747192413277,
|
| 364 |
+
"eval_f1_macro": 0.6310309906248334,
|
| 365 |
+
"eval_loss": 1.1600251197814941,
|
| 366 |
+
"eval_runtime": 55.43,
|
| 367 |
+
"eval_samples_per_second": 144.579,
|
| 368 |
+
"eval_steps_per_second": 4.528,
|
| 369 |
+
"step": 15000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 35.75547866205306,
|
| 373 |
+
"grad_norm": 4.611924648284912,
|
| 374 |
+
"learning_rate": 0.0002685185185185186,
|
| 375 |
+
"loss": 0.2428,
|
| 376 |
+
"step": 15500
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 36.90888119953864,
|
| 380 |
+
"grad_norm": 3.3283474445343018,
|
| 381 |
+
"learning_rate": 0.00025925925925925926,
|
| 382 |
+
"loss": 0.2322,
|
| 383 |
+
"step": 16000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 36.90888119953864,
|
| 387 |
+
"eval_accuracy": 0.7816321437484403,
|
| 388 |
+
"eval_f1_macro": 0.6319203799590871,
|
| 389 |
+
"eval_loss": 1.135780930519104,
|
| 390 |
+
"eval_runtime": 63.9861,
|
| 391 |
+
"eval_samples_per_second": 125.246,
|
| 392 |
+
"eval_steps_per_second": 3.923,
|
| 393 |
+
"step": 16000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 38.062283737024224,
|
| 397 |
+
"grad_norm": 6.641352653503418,
|
| 398 |
+
"learning_rate": 0.00025,
|
| 399 |
+
"loss": 0.2146,
|
| 400 |
+
"step": 16500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 39.21568627450981,
|
| 404 |
+
"grad_norm": 6.247890949249268,
|
| 405 |
+
"learning_rate": 0.00024074074074074072,
|
| 406 |
+
"loss": 0.2003,
|
| 407 |
+
"step": 17000
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 39.21568627450981,
|
| 411 |
+
"eval_accuracy": 0.7962315947092587,
|
| 412 |
+
"eval_f1_macro": 0.6542244286445125,
|
| 413 |
+
"eval_loss": 1.150564432144165,
|
| 414 |
+
"eval_runtime": 69.2204,
|
| 415 |
+
"eval_samples_per_second": 115.775,
|
| 416 |
+
"eval_steps_per_second": 3.626,
|
| 417 |
+
"step": 17000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 40.36908881199539,
|
| 421 |
+
"grad_norm": 4.363713264465332,
|
| 422 |
+
"learning_rate": 0.0002314814814814815,
|
| 423 |
+
"loss": 0.1947,
|
| 424 |
+
"step": 17500
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 41.522491349480966,
|
| 428 |
+
"grad_norm": 3.4260287284851074,
|
| 429 |
+
"learning_rate": 0.0002222222222222222,
|
| 430 |
+
"loss": 0.1794,
|
| 431 |
+
"step": 18000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 41.522491349480966,
|
| 435 |
+
"eval_accuracy": 0.7979785375592713,
|
| 436 |
+
"eval_f1_macro": 0.6796792585107833,
|
| 437 |
+
"eval_loss": 1.1864490509033203,
|
| 438 |
+
"eval_runtime": 71.1001,
|
| 439 |
+
"eval_samples_per_second": 112.714,
|
| 440 |
+
"eval_steps_per_second": 3.53,
|
| 441 |
+
"step": 18000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 42.67589388696655,
|
| 445 |
+
"grad_norm": 2.606008291244507,
|
| 446 |
+
"learning_rate": 0.00021296296296296298,
|
| 447 |
+
"loss": 0.1689,
|
| 448 |
+
"step": 18500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 43.82929642445213,
|
| 452 |
+
"grad_norm": 4.665687084197998,
|
| 453 |
+
"learning_rate": 0.0002037037037037037,
|
| 454 |
+
"loss": 0.1645,
|
| 455 |
+
"step": 19000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 43.82929642445213,
|
| 459 |
+
"eval_accuracy": 0.80059895183429,
|
| 460 |
+
"eval_f1_macro": 0.6667890419585701,
|
| 461 |
+
"eval_loss": 1.2014110088348389,
|
| 462 |
+
"eval_runtime": 74.7112,
|
| 463 |
+
"eval_samples_per_second": 107.266,
|
| 464 |
+
"eval_steps_per_second": 3.36,
|
| 465 |
+
"step": 19000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 44.982698961937714,
|
| 469 |
+
"grad_norm": 3.0378100872039795,
|
| 470 |
+
"learning_rate": 0.00019444444444444446,
|
| 471 |
+
"loss": 0.1602,
|
| 472 |
+
"step": 19500
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 46.1361014994233,
|
| 476 |
+
"grad_norm": 5.274627685546875,
|
| 477 |
+
"learning_rate": 0.00018518518518518518,
|
| 478 |
+
"loss": 0.144,
|
| 479 |
+
"step": 20000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 46.1361014994233,
|
| 483 |
+
"eval_accuracy": 0.7989767906164212,
|
| 484 |
+
"eval_f1_macro": 0.6582157335341974,
|
| 485 |
+
"eval_loss": 1.1411352157592773,
|
| 486 |
+
"eval_runtime": 73.2065,
|
| 487 |
+
"eval_samples_per_second": 109.471,
|
| 488 |
+
"eval_steps_per_second": 3.429,
|
| 489 |
+
"step": 20000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 47.28950403690888,
|
| 493 |
+
"grad_norm": 2.336925745010376,
|
| 494 |
+
"learning_rate": 0.00017592592592592595,
|
| 495 |
+
"loss": 0.1368,
|
| 496 |
+
"step": 20500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 48.44290657439446,
|
| 500 |
+
"grad_norm": 2.7309417724609375,
|
| 501 |
+
"learning_rate": 0.00016666666666666666,
|
| 502 |
+
"loss": 0.1298,
|
| 503 |
+
"step": 21000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 48.44290657439446,
|
| 507 |
+
"eval_accuracy": 0.8064636885450461,
|
| 508 |
+
"eval_f1_macro": 0.6782237618476237,
|
| 509 |
+
"eval_loss": 1.1389836072921753,
|
| 510 |
+
"eval_runtime": 73.3032,
|
| 511 |
+
"eval_samples_per_second": 109.327,
|
| 512 |
+
"eval_steps_per_second": 3.424,
|
| 513 |
+
"step": 21000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 49.596309111880046,
|
| 517 |
+
"grad_norm": 2.79067325592041,
|
| 518 |
+
"learning_rate": 0.0001574074074074074,
|
| 519 |
+
"loss": 0.1206,
|
| 520 |
+
"step": 21500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 50.74971164936563,
|
| 524 |
+
"grad_norm": 4.826747417449951,
|
| 525 |
+
"learning_rate": 0.00014814814814814815,
|
| 526 |
+
"loss": 0.1175,
|
| 527 |
+
"step": 22000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 50.74971164936563,
|
| 531 |
+
"eval_accuracy": 0.8068380334414774,
|
| 532 |
+
"eval_f1_macro": 0.6700916407139905,
|
| 533 |
+
"eval_loss": 1.2090946435928345,
|
| 534 |
+
"eval_runtime": 73.8226,
|
| 535 |
+
"eval_samples_per_second": 108.557,
|
| 536 |
+
"eval_steps_per_second": 3.4,
|
| 537 |
+
"step": 22000
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 51.90311418685121,
|
| 541 |
+
"grad_norm": 3.403858184814453,
|
| 542 |
+
"learning_rate": 0.0001388888888888889,
|
| 543 |
+
"loss": 0.1021,
|
| 544 |
+
"step": 22500
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 53.056516724336795,
|
| 548 |
+
"grad_norm": 5.1802496910095215,
|
| 549 |
+
"learning_rate": 0.00012962962962962963,
|
| 550 |
+
"loss": 0.0977,
|
| 551 |
+
"step": 23000
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 53.056516724336795,
|
| 555 |
+
"eval_accuracy": 0.8149488395308211,
|
| 556 |
+
"eval_f1_macro": 0.682806558028361,
|
| 557 |
+
"eval_loss": 1.1759377717971802,
|
| 558 |
+
"eval_runtime": 73.8391,
|
| 559 |
+
"eval_samples_per_second": 108.533,
|
| 560 |
+
"eval_steps_per_second": 3.399,
|
| 561 |
+
"step": 23000
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 54.20991926182238,
|
| 565 |
+
"grad_norm": 1.122316837310791,
|
| 566 |
+
"learning_rate": 0.00012037037037037036,
|
| 567 |
+
"loss": 0.0912,
|
| 568 |
+
"step": 23500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 55.36332179930796,
|
| 572 |
+
"grad_norm": 1.1100833415985107,
|
| 573 |
+
"learning_rate": 0.0001111111111111111,
|
| 574 |
+
"loss": 0.0823,
|
| 575 |
+
"step": 24000
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 55.36332179930796,
|
| 579 |
+
"eval_accuracy": 0.8166957823808335,
|
| 580 |
+
"eval_f1_macro": 0.7045678569443168,
|
| 581 |
+
"eval_loss": 1.2304565906524658,
|
| 582 |
+
"eval_runtime": 74.7299,
|
| 583 |
+
"eval_samples_per_second": 107.24,
|
| 584 |
+
"eval_steps_per_second": 3.359,
|
| 585 |
+
"step": 24000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 56.516724336793544,
|
| 589 |
+
"grad_norm": 4.60992956161499,
|
| 590 |
+
"learning_rate": 0.00010185185185185185,
|
| 591 |
+
"loss": 0.0873,
|
| 592 |
+
"step": 24500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 57.67012687427912,
|
| 596 |
+
"grad_norm": 5.945472240447998,
|
| 597 |
+
"learning_rate": 9.259259259259259e-05,
|
| 598 |
+
"loss": 0.0767,
|
| 599 |
+
"step": 25000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 57.67012687427912,
|
| 603 |
+
"eval_accuracy": 0.8238083354130272,
|
| 604 |
+
"eval_f1_macro": 0.6889311414034964,
|
| 605 |
+
"eval_loss": 1.231188416481018,
|
| 606 |
+
"eval_runtime": 72.9471,
|
| 607 |
+
"eval_samples_per_second": 109.86,
|
| 608 |
+
"eval_steps_per_second": 3.441,
|
| 609 |
+
"step": 25000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 58.8235294117647,
|
| 613 |
+
"grad_norm": 6.51999044418335,
|
| 614 |
+
"learning_rate": 8.333333333333333e-05,
|
| 615 |
+
"loss": 0.0667,
|
| 616 |
+
"step": 25500
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 59.976931949250286,
|
| 620 |
+
"grad_norm": 3.6006715297698975,
|
| 621 |
+
"learning_rate": 7.407407407407407e-05,
|
| 622 |
+
"loss": 0.066,
|
| 623 |
+
"step": 26000
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 59.976931949250286,
|
| 627 |
+
"eval_accuracy": 0.8235587721487397,
|
| 628 |
+
"eval_f1_macro": 0.7127282615425515,
|
| 629 |
+
"eval_loss": 1.212782621383667,
|
| 630 |
+
"eval_runtime": 70.4845,
|
| 631 |
+
"eval_samples_per_second": 113.699,
|
| 632 |
+
"eval_steps_per_second": 3.561,
|
| 633 |
+
"step": 26000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 61.13033448673587,
|
| 637 |
+
"grad_norm": 1.102469563484192,
|
| 638 |
+
"learning_rate": 6.481481481481482e-05,
|
| 639 |
+
"loss": 0.0601,
|
| 640 |
+
"step": 26500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 62.28373702422145,
|
| 644 |
+
"grad_norm": 3.476552724838257,
|
| 645 |
+
"learning_rate": 5.555555555555555e-05,
|
| 646 |
+
"loss": 0.0493,
|
| 647 |
+
"step": 27000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 62.28373702422145,
|
| 651 |
+
"eval_accuracy": 0.8310456700773646,
|
| 652 |
+
"eval_f1_macro": 0.7115209260308665,
|
| 653 |
+
"eval_loss": 1.15741765499115,
|
| 654 |
+
"eval_runtime": 74.8395,
|
| 655 |
+
"eval_samples_per_second": 107.083,
|
| 656 |
+
"eval_steps_per_second": 3.354,
|
| 657 |
+
"step": 27000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 63.437139561707035,
|
| 661 |
+
"grad_norm": 0.26378124952316284,
|
| 662 |
+
"learning_rate": 4.6296296296296294e-05,
|
| 663 |
+
"loss": 0.0527,
|
| 664 |
+
"step": 27500
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 64.59054209919262,
|
| 668 |
+
"grad_norm": 1.6174193620681763,
|
| 669 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 670 |
+
"loss": 0.0479,
|
| 671 |
+
"step": 28000
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 64.59054209919262,
|
| 675 |
+
"eval_accuracy": 0.836785625155977,
|
| 676 |
+
"eval_f1_macro": 0.7171903480510493,
|
| 677 |
+
"eval_loss": 1.1416091918945312,
|
| 678 |
+
"eval_runtime": 72.9274,
|
| 679 |
+
"eval_samples_per_second": 109.89,
|
| 680 |
+
"eval_steps_per_second": 3.442,
|
| 681 |
+
"step": 28000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 65.7439446366782,
|
| 685 |
+
"grad_norm": 1.4499250650405884,
|
| 686 |
+
"learning_rate": 2.7777777777777776e-05,
|
| 687 |
+
"loss": 0.0453,
|
| 688 |
+
"step": 28500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 66.89734717416378,
|
| 692 |
+
"grad_norm": 3.988093614578247,
|
| 693 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 694 |
+
"loss": 0.0389,
|
| 695 |
+
"step": 29000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 66.89734717416378,
|
| 699 |
+
"eval_accuracy": 0.8370351884202646,
|
| 700 |
+
"eval_f1_macro": 0.7212408780468642,
|
| 701 |
+
"eval_loss": 1.1253269910812378,
|
| 702 |
+
"eval_runtime": 53.9632,
|
| 703 |
+
"eval_samples_per_second": 148.509,
|
| 704 |
+
"eval_steps_per_second": 4.651,
|
| 705 |
+
"step": 29000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 68.05074971164936,
|
| 709 |
+
"grad_norm": 2.909609317779541,
|
| 710 |
+
"learning_rate": 9.259259259259259e-06,
|
| 711 |
+
"loss": 0.0433,
|
| 712 |
+
"step": 29500
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 69.20415224913495,
|
| 716 |
+
"grad_norm": 2.1668026447296143,
|
| 717 |
+
"learning_rate": 0.0,
|
| 718 |
+
"loss": 0.0343,
|
| 719 |
+
"step": 30000
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 69.20415224913495,
|
| 723 |
+
"eval_accuracy": 0.8415273271774395,
|
| 724 |
+
"eval_f1_macro": 0.7162084790077747,
|
| 725 |
+
"eval_loss": 1.1328068971633911,
|
| 726 |
+
"eval_runtime": 73.3281,
|
| 727 |
+
"eval_samples_per_second": 109.29,
|
| 728 |
+
"eval_steps_per_second": 3.423,
|
| 729 |
+
"step": 30000
|
| 730 |
+
}
|
| 731 |
+
],
|
| 732 |
+
"logging_steps": 500,
|
| 733 |
+
"max_steps": 30000,
|
| 734 |
+
"num_input_tokens_seen": 0,
|
| 735 |
+
"num_train_epochs": 70,
|
| 736 |
+
"save_steps": 1000,
|
| 737 |
+
"stateful_callbacks": {
|
| 738 |
+
"TrainerControl": {
|
| 739 |
+
"args": {
|
| 740 |
+
"should_epoch_stop": false,
|
| 741 |
+
"should_evaluate": false,
|
| 742 |
+
"should_log": false,
|
| 743 |
+
"should_save": true,
|
| 744 |
+
"should_training_stop": true
|
| 745 |
+
},
|
| 746 |
+
"attributes": {}
|
| 747 |
+
}
|
| 748 |
+
},
|
| 749 |
+
"total_flos": 1.7427529644770302e+20,
|
| 750 |
+
"train_batch_size": 32,
|
| 751 |
+
"trial_name": null,
|
| 752 |
+
"trial_params": null
|
| 753 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f96202ca49dc23aed34401bf9cf6226f15477f09ec142821d1849858c25c8f66
|
| 3 |
+
size 5240
|