Upload 7 files
Browse files- config.json +56 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +405 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gogamza/kobart-base-v2",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"BartForDialogueGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"author": "Heewon Jeon(madjakarta@gmail.com)",
|
| 12 |
+
"bos_token_id": 1,
|
| 13 |
+
"classif_dropout": 0.1,
|
| 14 |
+
"classifier_dropout": 0.1,
|
| 15 |
+
"d_model": 768,
|
| 16 |
+
"decoder_attention_heads": 16,
|
| 17 |
+
"decoder_ffn_dim": 3072,
|
| 18 |
+
"decoder_layerdrop": 0.0,
|
| 19 |
+
"decoder_layers": 6,
|
| 20 |
+
"decoder_start_token_id": 1,
|
| 21 |
+
"do_blenderbot_90_layernorm": false,
|
| 22 |
+
"dropout": 0.1,
|
| 23 |
+
"encoder_attention_heads": 16,
|
| 24 |
+
"encoder_ffn_dim": 3072,
|
| 25 |
+
"encoder_layerdrop": 0.0,
|
| 26 |
+
"encoder_layers": 6,
|
| 27 |
+
"eos_token_id": 1,
|
| 28 |
+
"extra_pos_embeddings": 2,
|
| 29 |
+
"force_bos_token_to_be_generated": false,
|
| 30 |
+
"forced_eos_token_id": 1,
|
| 31 |
+
"gradient_checkpointing": false,
|
| 32 |
+
"id2label": {
|
| 33 |
+
"0": "NEGATIVE",
|
| 34 |
+
"1": "POSITIVE"
|
| 35 |
+
},
|
| 36 |
+
"init_std": 0.02,
|
| 37 |
+
"is_encoder_decoder": true,
|
| 38 |
+
"kobart_version": 2.0,
|
| 39 |
+
"label2id": {
|
| 40 |
+
"NEGATIVE": 0,
|
| 41 |
+
"POSITIVE": 1
|
| 42 |
+
},
|
| 43 |
+
"max_position_embeddings": 1026,
|
| 44 |
+
"model_type": "bart",
|
| 45 |
+
"normalize_before": false,
|
| 46 |
+
"normalize_embedding": true,
|
| 47 |
+
"num_hidden_layers": 6,
|
| 48 |
+
"pad_token_id": 3,
|
| 49 |
+
"scale_embedding": false,
|
| 50 |
+
"static_position_embeddings": false,
|
| 51 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 52 |
+
"torch_dtype": "float32",
|
| 53 |
+
"transformers_version": "4.17.0",
|
| 54 |
+
"use_cache": true,
|
| 55 |
+
"vocab_size": 30004
|
| 56 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:339fd69214325799a030d941184d800e794df5c4a9e19ba69dc3f408d55220ca
|
| 3 |
+
size 495883396
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "</s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": "<mask>", "additional_special_tokens": ["<agent>", "<user>", "<agent_mem>", "<user_mem>"]}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"additional_special_tokens": ["<agent>", "<user>", "<agent_mem>", "<user_mem>"], "special_tokens_map_file": "/home/mongjin/.cache/huggingface/transformers/a87d2ed77831bb40ce806a97c04126addf5ecc82b3e23ecf916b2a4acdb9c29a.c23d5e62137984cf842a885705037b25b156747d145406702932d5f5d5e7c88e", "name_or_path": "gogamza/kobart-base-v2", "tokenizer_class": "PreTrainedTokenizerFast"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,405 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 27925,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.09,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 3.5105,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.18,
|
| 18 |
+
"learning_rate": 2.945305378304467e-05,
|
| 19 |
+
"loss": 1.6044,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.27,
|
| 24 |
+
"learning_rate": 2.8906107566089334e-05,
|
| 25 |
+
"loss": 1.5561,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.36,
|
| 30 |
+
"learning_rate": 2.8359161349134002e-05,
|
| 31 |
+
"loss": 1.5065,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.45,
|
| 36 |
+
"learning_rate": 2.781221513217867e-05,
|
| 37 |
+
"loss": 1.4607,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.54,
|
| 42 |
+
"learning_rate": 2.726526891522334e-05,
|
| 43 |
+
"loss": 1.4575,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.63,
|
| 48 |
+
"learning_rate": 2.6718322698268003e-05,
|
| 49 |
+
"loss": 1.432,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.72,
|
| 54 |
+
"learning_rate": 2.617137648131267e-05,
|
| 55 |
+
"loss": 1.4269,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.81,
|
| 60 |
+
"learning_rate": 2.562443026435734e-05,
|
| 61 |
+
"loss": 1.4204,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.9,
|
| 66 |
+
"learning_rate": 2.5077484047402008e-05,
|
| 67 |
+
"loss": 1.4037,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.98,
|
| 72 |
+
"learning_rate": 2.4530537830446673e-05,
|
| 73 |
+
"loss": 1.4064,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.0,
|
| 78 |
+
"eval_bleu": 0.8646,
|
| 79 |
+
"eval_gen_len": 16.6,
|
| 80 |
+
"eval_loss": 1.450329303741455,
|
| 81 |
+
"eval_runtime": 368.5927,
|
| 82 |
+
"eval_samples_per_second": 56.358,
|
| 83 |
+
"eval_steps_per_second": 0.882,
|
| 84 |
+
"step": 5585
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.07,
|
| 88 |
+
"learning_rate": 2.398359161349134e-05,
|
| 89 |
+
"loss": 1.3601,
|
| 90 |
+
"step": 6000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.16,
|
| 94 |
+
"learning_rate": 2.343664539653601e-05,
|
| 95 |
+
"loss": 1.3596,
|
| 96 |
+
"step": 6500
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.25,
|
| 100 |
+
"learning_rate": 2.2889699179580678e-05,
|
| 101 |
+
"loss": 1.3655,
|
| 102 |
+
"step": 7000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.34,
|
| 106 |
+
"learning_rate": 2.2342752962625342e-05,
|
| 107 |
+
"loss": 1.326,
|
| 108 |
+
"step": 7500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.43,
|
| 112 |
+
"learning_rate": 2.179580674567001e-05,
|
| 113 |
+
"loss": 1.3327,
|
| 114 |
+
"step": 8000
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.52,
|
| 118 |
+
"learning_rate": 2.124886052871468e-05,
|
| 119 |
+
"loss": 1.3367,
|
| 120 |
+
"step": 8500
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.61,
|
| 124 |
+
"learning_rate": 2.0701914311759347e-05,
|
| 125 |
+
"loss": 1.3305,
|
| 126 |
+
"step": 9000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.7,
|
| 130 |
+
"learning_rate": 2.0154968094804012e-05,
|
| 131 |
+
"loss": 1.3218,
|
| 132 |
+
"step": 9500
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.79,
|
| 136 |
+
"learning_rate": 1.960802187784868e-05,
|
| 137 |
+
"loss": 1.3302,
|
| 138 |
+
"step": 10000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 1.88,
|
| 142 |
+
"learning_rate": 1.906107566089335e-05,
|
| 143 |
+
"loss": 1.3173,
|
| 144 |
+
"step": 10500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.97,
|
| 148 |
+
"learning_rate": 1.8514129443938013e-05,
|
| 149 |
+
"loss": 1.3306,
|
| 150 |
+
"step": 11000
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 2.0,
|
| 154 |
+
"eval_bleu": 0.8302,
|
| 155 |
+
"eval_gen_len": 16.9,
|
| 156 |
+
"eval_loss": 1.422479271888733,
|
| 157 |
+
"eval_runtime": 368.3493,
|
| 158 |
+
"eval_samples_per_second": 56.395,
|
| 159 |
+
"eval_steps_per_second": 0.882,
|
| 160 |
+
"step": 11170
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.06,
|
| 164 |
+
"learning_rate": 1.7967183226982678e-05,
|
| 165 |
+
"loss": 1.2841,
|
| 166 |
+
"step": 11500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 2.15,
|
| 170 |
+
"learning_rate": 1.7420237010027346e-05,
|
| 171 |
+
"loss": 1.287,
|
| 172 |
+
"step": 12000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.24,
|
| 176 |
+
"learning_rate": 1.6873290793072015e-05,
|
| 177 |
+
"loss": 1.2772,
|
| 178 |
+
"step": 12500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 2.33,
|
| 182 |
+
"learning_rate": 1.632634457611668e-05,
|
| 183 |
+
"loss": 1.2802,
|
| 184 |
+
"step": 13000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 2.42,
|
| 188 |
+
"learning_rate": 1.5779398359161348e-05,
|
| 189 |
+
"loss": 1.27,
|
| 190 |
+
"step": 13500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 2.51,
|
| 194 |
+
"learning_rate": 1.5232452142206018e-05,
|
| 195 |
+
"loss": 1.273,
|
| 196 |
+
"step": 14000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.6,
|
| 200 |
+
"learning_rate": 1.4685505925250684e-05,
|
| 201 |
+
"loss": 1.2888,
|
| 202 |
+
"step": 14500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.69,
|
| 206 |
+
"learning_rate": 1.4138559708295352e-05,
|
| 207 |
+
"loss": 1.2767,
|
| 208 |
+
"step": 15000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.78,
|
| 212 |
+
"learning_rate": 1.3591613491340019e-05,
|
| 213 |
+
"loss": 1.2713,
|
| 214 |
+
"step": 15500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 2.86,
|
| 218 |
+
"learning_rate": 1.3044667274384685e-05,
|
| 219 |
+
"loss": 1.253,
|
| 220 |
+
"step": 16000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 2.95,
|
| 224 |
+
"learning_rate": 1.2497721057429352e-05,
|
| 225 |
+
"loss": 1.262,
|
| 226 |
+
"step": 16500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 3.0,
|
| 230 |
+
"eval_bleu": 1.0163,
|
| 231 |
+
"eval_gen_len": 17.1,
|
| 232 |
+
"eval_loss": 1.4149495363235474,
|
| 233 |
+
"eval_runtime": 367.931,
|
| 234 |
+
"eval_samples_per_second": 56.459,
|
| 235 |
+
"eval_steps_per_second": 0.883,
|
| 236 |
+
"step": 16755
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.04,
|
| 240 |
+
"learning_rate": 1.195077484047402e-05,
|
| 241 |
+
"loss": 1.2486,
|
| 242 |
+
"step": 17000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 3.13,
|
| 246 |
+
"learning_rate": 1.1403828623518687e-05,
|
| 247 |
+
"loss": 1.2413,
|
| 248 |
+
"step": 17500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.22,
|
| 252 |
+
"learning_rate": 1.0856882406563355e-05,
|
| 253 |
+
"loss": 1.2174,
|
| 254 |
+
"step": 18000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.31,
|
| 258 |
+
"learning_rate": 1.0309936189608021e-05,
|
| 259 |
+
"loss": 1.2258,
|
| 260 |
+
"step": 18500
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.4,
|
| 264 |
+
"learning_rate": 9.76298997265269e-06,
|
| 265 |
+
"loss": 1.2378,
|
| 266 |
+
"step": 19000
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 3.49,
|
| 270 |
+
"learning_rate": 9.216043755697356e-06,
|
| 271 |
+
"loss": 1.2286,
|
| 272 |
+
"step": 19500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.58,
|
| 276 |
+
"learning_rate": 8.669097538742024e-06,
|
| 277 |
+
"loss": 1.2317,
|
| 278 |
+
"step": 20000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.67,
|
| 282 |
+
"learning_rate": 8.122151321786691e-06,
|
| 283 |
+
"loss": 1.2386,
|
| 284 |
+
"step": 20500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 3.76,
|
| 288 |
+
"learning_rate": 7.575205104831359e-06,
|
| 289 |
+
"loss": 1.22,
|
| 290 |
+
"step": 21000
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 3.85,
|
| 294 |
+
"learning_rate": 7.028258887876026e-06,
|
| 295 |
+
"loss": 1.2252,
|
| 296 |
+
"step": 21500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 3.94,
|
| 300 |
+
"learning_rate": 6.481312670920693e-06,
|
| 301 |
+
"loss": 1.2175,
|
| 302 |
+
"step": 22000
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 4.0,
|
| 306 |
+
"eval_bleu": 0.9445,
|
| 307 |
+
"eval_gen_len": 17.2,
|
| 308 |
+
"eval_loss": 1.416054606437683,
|
| 309 |
+
"eval_runtime": 367.7817,
|
| 310 |
+
"eval_samples_per_second": 56.482,
|
| 311 |
+
"eval_steps_per_second": 0.884,
|
| 312 |
+
"step": 22340
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 4.03,
|
| 316 |
+
"learning_rate": 5.9343664539653605e-06,
|
| 317 |
+
"loss": 1.2158,
|
| 318 |
+
"step": 22500
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 4.12,
|
| 322 |
+
"learning_rate": 5.387420237010028e-06,
|
| 323 |
+
"loss": 1.1916,
|
| 324 |
+
"step": 23000
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 4.21,
|
| 328 |
+
"learning_rate": 4.8404740200546945e-06,
|
| 329 |
+
"loss": 1.2079,
|
| 330 |
+
"step": 23500
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 4.3,
|
| 334 |
+
"learning_rate": 4.293527803099362e-06,
|
| 335 |
+
"loss": 1.2124,
|
| 336 |
+
"step": 24000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 4.39,
|
| 340 |
+
"learning_rate": 3.7465815861440297e-06,
|
| 341 |
+
"loss": 1.2006,
|
| 342 |
+
"step": 24500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 4.48,
|
| 346 |
+
"learning_rate": 3.199635369188696e-06,
|
| 347 |
+
"loss": 1.2121,
|
| 348 |
+
"step": 25000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 4.57,
|
| 352 |
+
"learning_rate": 2.6526891522333636e-06,
|
| 353 |
+
"loss": 1.2022,
|
| 354 |
+
"step": 25500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 4.66,
|
| 358 |
+
"learning_rate": 2.105742935278031e-06,
|
| 359 |
+
"loss": 1.2029,
|
| 360 |
+
"step": 26000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 4.74,
|
| 364 |
+
"learning_rate": 1.5587967183226983e-06,
|
| 365 |
+
"loss": 1.2008,
|
| 366 |
+
"step": 26500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 4.83,
|
| 370 |
+
"learning_rate": 1.0118505013673655e-06,
|
| 371 |
+
"loss": 1.2193,
|
| 372 |
+
"step": 27000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 4.92,
|
| 376 |
+
"learning_rate": 4.6490428441203284e-07,
|
| 377 |
+
"loss": 1.2108,
|
| 378 |
+
"step": 27500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.0,
|
| 382 |
+
"eval_bleu": 1.0114,
|
| 383 |
+
"eval_gen_len": 17.3,
|
| 384 |
+
"eval_loss": 1.4185353517532349,
|
| 385 |
+
"eval_runtime": 367.8873,
|
| 386 |
+
"eval_samples_per_second": 56.466,
|
| 387 |
+
"eval_steps_per_second": 0.883,
|
| 388 |
+
"step": 27925
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 5.0,
|
| 392 |
+
"step": 27925,
|
| 393 |
+
"total_flos": 5.320769336064e+17,
|
| 394 |
+
"train_loss": 1.3384765100436438,
|
| 395 |
+
"train_runtime": 44633.7325,
|
| 396 |
+
"train_samples_per_second": 40.04,
|
| 397 |
+
"train_steps_per_second": 0.626
|
| 398 |
+
}
|
| 399 |
+
],
|
| 400 |
+
"max_steps": 27925,
|
| 401 |
+
"num_train_epochs": 5,
|
| 402 |
+
"total_flos": 5.320769336064e+17,
|
| 403 |
+
"trial_name": null,
|
| 404 |
+
"trial_params": null
|
| 405 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:367bad5f9682264dadbc8c88b4f7f5c52170059a3d5be12dce6a03ef1cd30281
|
| 3 |
+
size 3183
|