Upload model EncoderDecoder ver 2 (#2)
Browse files- Upload model EncoderDecoder ver 2 (5c9c12f991700e488402bea5bd5fd5202fdfc17d)
Co-authored-by: Luu Thien Viet Cuong <[email protected]>
- EncoderDecoder_2/config.json +180 -0
- EncoderDecoder_2/generation_config.json +8 -0
- EncoderDecoder_2/model.safetensors +3 -0
- EncoderDecoder_2/optimizer.pt +3 -0
- EncoderDecoder_2/rng_state.pth +3 -0
- EncoderDecoder_2/scheduler.pt +3 -0
- EncoderDecoder_2/trainer_state.json +1129 -0
- EncoderDecoder_2/training_args.bin +3 -0
EncoderDecoder_2/config.json
ADDED
|
@@ -0,0 +1,180 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"EncoderDecoderModel"
|
| 4 |
+
],
|
| 5 |
+
"decoder": {
|
| 6 |
+
"_attn_implementation_autoset": true,
|
| 7 |
+
"_name_or_path": "vinai/bartpho-word",
|
| 8 |
+
"activation_dropout": 0.0,
|
| 9 |
+
"activation_function": "gelu",
|
| 10 |
+
"add_cross_attention": true,
|
| 11 |
+
"architectures": [
|
| 12 |
+
"MBartModel"
|
| 13 |
+
],
|
| 14 |
+
"attention_dropout": 0.0,
|
| 15 |
+
"bad_words_ids": null,
|
| 16 |
+
"begin_suppress_tokens": null,
|
| 17 |
+
"bos_token_id": 0,
|
| 18 |
+
"chunk_size_feed_forward": 0,
|
| 19 |
+
"classifier_dropout": 0.0,
|
| 20 |
+
"cross_attention_hidden_size": null,
|
| 21 |
+
"d_model": 1024,
|
| 22 |
+
"decoder_attention_heads": 16,
|
| 23 |
+
"decoder_ffn_dim": 4096,
|
| 24 |
+
"decoder_layerdrop": 0.0,
|
| 25 |
+
"decoder_layers": 12,
|
| 26 |
+
"decoder_start_token_id": 2,
|
| 27 |
+
"diversity_penalty": 0.0,
|
| 28 |
+
"do_sample": false,
|
| 29 |
+
"dropout": 0.1,
|
| 30 |
+
"early_stopping": false,
|
| 31 |
+
"encoder_attention_heads": 16,
|
| 32 |
+
"encoder_ffn_dim": 4096,
|
| 33 |
+
"encoder_layerdrop": 0.0,
|
| 34 |
+
"encoder_layers": 12,
|
| 35 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 36 |
+
"eos_token_id": 2,
|
| 37 |
+
"exponential_decay_length_penalty": null,
|
| 38 |
+
"finetuning_task": null,
|
| 39 |
+
"forced_bos_token_id": null,
|
| 40 |
+
"forced_eos_token_id": 2,
|
| 41 |
+
"gradient_checkpointing": false,
|
| 42 |
+
"id2label": {
|
| 43 |
+
"0": "LABEL_0",
|
| 44 |
+
"1": "LABEL_1"
|
| 45 |
+
},
|
| 46 |
+
"init_std": 0.02,
|
| 47 |
+
"is_decoder": true,
|
| 48 |
+
"is_encoder_decoder": false,
|
| 49 |
+
"label2id": {
|
| 50 |
+
"LABEL_0": 0,
|
| 51 |
+
"LABEL_1": 1
|
| 52 |
+
},
|
| 53 |
+
"length_penalty": 1.0,
|
| 54 |
+
"max_length": 20,
|
| 55 |
+
"max_position_embeddings": 1024,
|
| 56 |
+
"min_length": 0,
|
| 57 |
+
"model_type": "mbart",
|
| 58 |
+
"no_repeat_ngram_size": 0,
|
| 59 |
+
"num_beam_groups": 1,
|
| 60 |
+
"num_beams": 1,
|
| 61 |
+
"num_hidden_layers": 12,
|
| 62 |
+
"num_return_sequences": 1,
|
| 63 |
+
"output_attentions": false,
|
| 64 |
+
"output_hidden_states": false,
|
| 65 |
+
"output_scores": false,
|
| 66 |
+
"pad_token_id": 1,
|
| 67 |
+
"prefix": null,
|
| 68 |
+
"problem_type": null,
|
| 69 |
+
"pruned_heads": {},
|
| 70 |
+
"remove_invalid_values": false,
|
| 71 |
+
"repetition_penalty": 1.0,
|
| 72 |
+
"return_dict": true,
|
| 73 |
+
"return_dict_in_generate": false,
|
| 74 |
+
"scale_embedding": false,
|
| 75 |
+
"sep_token_id": null,
|
| 76 |
+
"suppress_tokens": null,
|
| 77 |
+
"task_specific_params": null,
|
| 78 |
+
"temperature": 1.0,
|
| 79 |
+
"tf_legacy_loss": false,
|
| 80 |
+
"tie_encoder_decoder": false,
|
| 81 |
+
"tie_word_embeddings": true,
|
| 82 |
+
"tokenizer_class": "PhobertTokenizer",
|
| 83 |
+
"top_k": 50,
|
| 84 |
+
"top_p": 1.0,
|
| 85 |
+
"torch_dtype": "float32",
|
| 86 |
+
"torchscript": false,
|
| 87 |
+
"typical_p": 1.0,
|
| 88 |
+
"use_bfloat16": false,
|
| 89 |
+
"use_cache": true,
|
| 90 |
+
"vocab_size": 64001
|
| 91 |
+
},
|
| 92 |
+
"decoder_start_token_id": 0,
|
| 93 |
+
"encoder": {
|
| 94 |
+
"_attn_implementation_autoset": true,
|
| 95 |
+
"_name_or_path": "bert-base-uncased",
|
| 96 |
+
"add_cross_attention": false,
|
| 97 |
+
"architectures": [
|
| 98 |
+
"BertForMaskedLM"
|
| 99 |
+
],
|
| 100 |
+
"attention_probs_dropout_prob": 0.1,
|
| 101 |
+
"bad_words_ids": null,
|
| 102 |
+
"begin_suppress_tokens": null,
|
| 103 |
+
"bos_token_id": null,
|
| 104 |
+
"chunk_size_feed_forward": 0,
|
| 105 |
+
"classifier_dropout": null,
|
| 106 |
+
"cross_attention_hidden_size": null,
|
| 107 |
+
"decoder_start_token_id": null,
|
| 108 |
+
"diversity_penalty": 0.0,
|
| 109 |
+
"do_sample": false,
|
| 110 |
+
"early_stopping": false,
|
| 111 |
+
"encoder_no_repeat_ngram_size": 0,
|
| 112 |
+
"eos_token_id": null,
|
| 113 |
+
"exponential_decay_length_penalty": null,
|
| 114 |
+
"finetuning_task": null,
|
| 115 |
+
"forced_bos_token_id": null,
|
| 116 |
+
"forced_eos_token_id": null,
|
| 117 |
+
"gradient_checkpointing": false,
|
| 118 |
+
"hidden_act": "gelu",
|
| 119 |
+
"hidden_dropout_prob": 0.1,
|
| 120 |
+
"hidden_size": 768,
|
| 121 |
+
"id2label": {
|
| 122 |
+
"0": "LABEL_0",
|
| 123 |
+
"1": "LABEL_1"
|
| 124 |
+
},
|
| 125 |
+
"initializer_range": 0.02,
|
| 126 |
+
"intermediate_size": 3072,
|
| 127 |
+
"is_decoder": false,
|
| 128 |
+
"is_encoder_decoder": false,
|
| 129 |
+
"label2id": {
|
| 130 |
+
"LABEL_0": 0,
|
| 131 |
+
"LABEL_1": 1
|
| 132 |
+
},
|
| 133 |
+
"layer_norm_eps": 1e-12,
|
| 134 |
+
"length_penalty": 1.0,
|
| 135 |
+
"max_length": 20,
|
| 136 |
+
"max_position_embeddings": 512,
|
| 137 |
+
"min_length": 0,
|
| 138 |
+
"model_type": "bert",
|
| 139 |
+
"no_repeat_ngram_size": 0,
|
| 140 |
+
"num_attention_heads": 12,
|
| 141 |
+
"num_beam_groups": 1,
|
| 142 |
+
"num_beams": 1,
|
| 143 |
+
"num_hidden_layers": 12,
|
| 144 |
+
"num_return_sequences": 1,
|
| 145 |
+
"output_attentions": false,
|
| 146 |
+
"output_hidden_states": false,
|
| 147 |
+
"output_scores": false,
|
| 148 |
+
"pad_token_id": 0,
|
| 149 |
+
"position_embedding_type": "absolute",
|
| 150 |
+
"prefix": null,
|
| 151 |
+
"problem_type": null,
|
| 152 |
+
"pruned_heads": {},
|
| 153 |
+
"remove_invalid_values": false,
|
| 154 |
+
"repetition_penalty": 1.0,
|
| 155 |
+
"return_dict": true,
|
| 156 |
+
"return_dict_in_generate": false,
|
| 157 |
+
"sep_token_id": null,
|
| 158 |
+
"suppress_tokens": null,
|
| 159 |
+
"task_specific_params": null,
|
| 160 |
+
"temperature": 1.0,
|
| 161 |
+
"tf_legacy_loss": false,
|
| 162 |
+
"tie_encoder_decoder": false,
|
| 163 |
+
"tie_word_embeddings": true,
|
| 164 |
+
"tokenizer_class": null,
|
| 165 |
+
"top_k": 50,
|
| 166 |
+
"top_p": 1.0,
|
| 167 |
+
"torch_dtype": null,
|
| 168 |
+
"torchscript": false,
|
| 169 |
+
"type_vocab_size": 2,
|
| 170 |
+
"typical_p": 1.0,
|
| 171 |
+
"use_bfloat16": false,
|
| 172 |
+
"use_cache": true,
|
| 173 |
+
"vocab_size": 30522
|
| 174 |
+
},
|
| 175 |
+
"is_encoder_decoder": true,
|
| 176 |
+
"model_type": "encoder-decoder",
|
| 177 |
+
"pad_token_id": 1,
|
| 178 |
+
"torch_dtype": "float32",
|
| 179 |
+
"transformers_version": "4.46.3"
|
| 180 |
+
}
|
EncoderDecoder_2/generation_config.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 0,
|
| 3 |
+
"decoder_start_token_id": 2,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"forced_eos_token_id": 2,
|
| 6 |
+
"pad_token_id": 1,
|
| 7 |
+
"transformers_version": "4.46.3"
|
| 8 |
+
}
|
EncoderDecoder_2/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:20401cd34bff715e2f2121bd9dba9fcf621f6f22244a55caf78cca8c98b01d2c
|
| 3 |
+
size 1513750720
|
EncoderDecoder_2/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41b55c87460f3cb5562ee7274533cc83826cf1cc27a4feba0058e29ac56c8f8f
|
| 3 |
+
size 3023080673
|
EncoderDecoder_2/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69153d8f55706625fe10d34a6dd0d524330bc218a68d499c88960deadde52508
|
| 3 |
+
size 14244
|
EncoderDecoder_2/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0300317eca28e38257934485432442b24c7c4f46591a7f21b90a6bb7eb59348
|
| 3 |
+
size 1064
|
EncoderDecoder_2/trainer_state.json
ADDED
|
@@ -0,0 +1,1129 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 29.806659505907625,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 55500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.26852846401718583,
|
| 13 |
+
"grad_norm": 5.849491596221924,
|
| 14 |
+
"learning_rate": 4.955245255997136e-05,
|
| 15 |
+
"loss": 3.8311,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5370569280343717,
|
| 20 |
+
"grad_norm": 5.955765247344971,
|
| 21 |
+
"learning_rate": 4.9104905119942717e-05,
|
| 22 |
+
"loss": 3.0732,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.8055853920515574,
|
| 27 |
+
"grad_norm": 6.657278537750244,
|
| 28 |
+
"learning_rate": 4.8657357679914074e-05,
|
| 29 |
+
"loss": 2.7161,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.0,
|
| 34 |
+
"eval_loss": 2.284536838531494,
|
| 35 |
+
"eval_rouge2_fmeasure": 0.3144,
|
| 36 |
+
"eval_rouge2_precision": 0.3144,
|
| 37 |
+
"eval_rouge2_recall": 0.3144,
|
| 38 |
+
"eval_runtime": 108.4134,
|
| 39 |
+
"eval_samples_per_second": 44.155,
|
| 40 |
+
"eval_steps_per_second": 2.767,
|
| 41 |
+
"step": 1862
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.0741138560687433,
|
| 45 |
+
"grad_norm": 6.813873767852783,
|
| 46 |
+
"learning_rate": 4.820981023988543e-05,
|
| 47 |
+
"loss": 2.3742,
|
| 48 |
+
"step": 2000
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.342642320085929,
|
| 52 |
+
"grad_norm": 6.493252754211426,
|
| 53 |
+
"learning_rate": 4.776226279985679e-05,
|
| 54 |
+
"loss": 1.9193,
|
| 55 |
+
"step": 2500
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 1.6111707841031149,
|
| 59 |
+
"grad_norm": 7.747920989990234,
|
| 60 |
+
"learning_rate": 4.7314715359828145e-05,
|
| 61 |
+
"loss": 1.9001,
|
| 62 |
+
"step": 3000
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.8796992481203008,
|
| 66 |
+
"grad_norm": 6.069449424743652,
|
| 67 |
+
"learning_rate": 4.6867167919799495e-05,
|
| 68 |
+
"loss": 1.8676,
|
| 69 |
+
"step": 3500
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.0,
|
| 73 |
+
"eval_loss": 2.0997378826141357,
|
| 74 |
+
"eval_rouge2_fmeasure": 0.352,
|
| 75 |
+
"eval_rouge2_precision": 0.352,
|
| 76 |
+
"eval_rouge2_recall": 0.352,
|
| 77 |
+
"eval_runtime": 110.0345,
|
| 78 |
+
"eval_samples_per_second": 43.505,
|
| 79 |
+
"eval_steps_per_second": 2.726,
|
| 80 |
+
"step": 3724
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.1482277121374866,
|
| 84 |
+
"grad_norm": 6.031770706176758,
|
| 85 |
+
"learning_rate": 4.641962047977086e-05,
|
| 86 |
+
"loss": 1.511,
|
| 87 |
+
"step": 4000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 2.4167561761546725,
|
| 91 |
+
"grad_norm": 6.697205543518066,
|
| 92 |
+
"learning_rate": 4.5972073039742216e-05,
|
| 93 |
+
"loss": 1.2827,
|
| 94 |
+
"step": 4500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 2.685284640171858,
|
| 98 |
+
"grad_norm": 7.0569071769714355,
|
| 99 |
+
"learning_rate": 4.552452559971357e-05,
|
| 100 |
+
"loss": 1.3136,
|
| 101 |
+
"step": 5000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.9538131041890443,
|
| 105 |
+
"grad_norm": 6.0844526290893555,
|
| 106 |
+
"learning_rate": 4.507697815968493e-05,
|
| 107 |
+
"loss": 1.3387,
|
| 108 |
+
"step": 5500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 3.0,
|
| 112 |
+
"eval_loss": 2.111859083175659,
|
| 113 |
+
"eval_rouge2_fmeasure": 0.3545,
|
| 114 |
+
"eval_rouge2_precision": 0.3545,
|
| 115 |
+
"eval_rouge2_recall": 0.3545,
|
| 116 |
+
"eval_runtime": 108.6339,
|
| 117 |
+
"eval_samples_per_second": 44.065,
|
| 118 |
+
"eval_steps_per_second": 2.762,
|
| 119 |
+
"step": 5586
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 3.2223415682062297,
|
| 123 |
+
"grad_norm": 5.514146327972412,
|
| 124 |
+
"learning_rate": 4.462943071965629e-05,
|
| 125 |
+
"loss": 0.9052,
|
| 126 |
+
"step": 6000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 3.4908700322234156,
|
| 130 |
+
"grad_norm": 6.395975112915039,
|
| 131 |
+
"learning_rate": 4.4181883279627644e-05,
|
| 132 |
+
"loss": 0.866,
|
| 133 |
+
"step": 6500
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 3.7593984962406015,
|
| 137 |
+
"grad_norm": 6.1064348220825195,
|
| 138 |
+
"learning_rate": 4.3734335839599e-05,
|
| 139 |
+
"loss": 0.8991,
|
| 140 |
+
"step": 7000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.0,
|
| 144 |
+
"eval_loss": 2.1967225074768066,
|
| 145 |
+
"eval_rouge2_fmeasure": 0.3531,
|
| 146 |
+
"eval_rouge2_precision": 0.3531,
|
| 147 |
+
"eval_rouge2_recall": 0.3531,
|
| 148 |
+
"eval_runtime": 108.9474,
|
| 149 |
+
"eval_samples_per_second": 43.939,
|
| 150 |
+
"eval_steps_per_second": 2.754,
|
| 151 |
+
"step": 7448
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 4.027926960257787,
|
| 155 |
+
"grad_norm": 5.5337700843811035,
|
| 156 |
+
"learning_rate": 4.328678839957035e-05,
|
| 157 |
+
"loss": 0.89,
|
| 158 |
+
"step": 7500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.296455424274973,
|
| 162 |
+
"grad_norm": 5.39340353012085,
|
| 163 |
+
"learning_rate": 4.2839240959541715e-05,
|
| 164 |
+
"loss": 0.539,
|
| 165 |
+
"step": 8000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.564983888292159,
|
| 169 |
+
"grad_norm": 6.1959967613220215,
|
| 170 |
+
"learning_rate": 4.239169351951307e-05,
|
| 171 |
+
"loss": 0.5854,
|
| 172 |
+
"step": 8500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 4.833512352309345,
|
| 176 |
+
"grad_norm": 6.185136795043945,
|
| 177 |
+
"learning_rate": 4.194414607948442e-05,
|
| 178 |
+
"loss": 0.6131,
|
| 179 |
+
"step": 9000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 5.0,
|
| 183 |
+
"eval_loss": 2.305931806564331,
|
| 184 |
+
"eval_rouge2_fmeasure": 0.3538,
|
| 185 |
+
"eval_rouge2_precision": 0.3538,
|
| 186 |
+
"eval_rouge2_recall": 0.3538,
|
| 187 |
+
"eval_runtime": 108.7292,
|
| 188 |
+
"eval_samples_per_second": 44.027,
|
| 189 |
+
"eval_steps_per_second": 2.759,
|
| 190 |
+
"step": 9310
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 5.1020408163265305,
|
| 194 |
+
"grad_norm": 3.842778205871582,
|
| 195 |
+
"learning_rate": 4.149659863945579e-05,
|
| 196 |
+
"loss": 0.5237,
|
| 197 |
+
"step": 9500
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 5.370569280343717,
|
| 201 |
+
"grad_norm": 4.404533863067627,
|
| 202 |
+
"learning_rate": 4.1049051199427144e-05,
|
| 203 |
+
"loss": 0.3687,
|
| 204 |
+
"step": 10000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 5.639097744360902,
|
| 208 |
+
"grad_norm": 5.195127487182617,
|
| 209 |
+
"learning_rate": 4.0601503759398494e-05,
|
| 210 |
+
"loss": 0.3907,
|
| 211 |
+
"step": 10500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 5.907626208378088,
|
| 215 |
+
"grad_norm": 5.097487926483154,
|
| 216 |
+
"learning_rate": 4.015395631936986e-05,
|
| 217 |
+
"loss": 0.4204,
|
| 218 |
+
"step": 11000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.0,
|
| 222 |
+
"eval_loss": 2.4349613189697266,
|
| 223 |
+
"eval_rouge2_fmeasure": 0.3489,
|
| 224 |
+
"eval_rouge2_precision": 0.3489,
|
| 225 |
+
"eval_rouge2_recall": 0.3489,
|
| 226 |
+
"eval_runtime": 108.5656,
|
| 227 |
+
"eval_samples_per_second": 44.093,
|
| 228 |
+
"eval_steps_per_second": 2.763,
|
| 229 |
+
"step": 11172
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 6.176154672395274,
|
| 233 |
+
"grad_norm": 4.235614776611328,
|
| 234 |
+
"learning_rate": 3.970640887934121e-05,
|
| 235 |
+
"loss": 0.3066,
|
| 236 |
+
"step": 11500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 6.4446831364124595,
|
| 240 |
+
"grad_norm": 4.217000961303711,
|
| 241 |
+
"learning_rate": 3.9258861439312565e-05,
|
| 242 |
+
"loss": 0.2646,
|
| 243 |
+
"step": 12000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 6.713211600429646,
|
| 247 |
+
"grad_norm": 3.1819193363189697,
|
| 248 |
+
"learning_rate": 3.881131399928393e-05,
|
| 249 |
+
"loss": 0.2797,
|
| 250 |
+
"step": 12500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 6.981740064446831,
|
| 254 |
+
"grad_norm": 4.56110954284668,
|
| 255 |
+
"learning_rate": 3.836376655925528e-05,
|
| 256 |
+
"loss": 0.2933,
|
| 257 |
+
"step": 13000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 7.0,
|
| 261 |
+
"eval_loss": 2.5125792026519775,
|
| 262 |
+
"eval_rouge2_fmeasure": 0.3476,
|
| 263 |
+
"eval_rouge2_precision": 0.3476,
|
| 264 |
+
"eval_rouge2_recall": 0.3476,
|
| 265 |
+
"eval_runtime": 108.9982,
|
| 266 |
+
"eval_samples_per_second": 43.918,
|
| 267 |
+
"eval_steps_per_second": 2.752,
|
| 268 |
+
"step": 13034
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.250268528464018,
|
| 272 |
+
"grad_norm": 3.549121856689453,
|
| 273 |
+
"learning_rate": 3.791621911922664e-05,
|
| 274 |
+
"loss": 0.1865,
|
| 275 |
+
"step": 13500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.518796992481203,
|
| 279 |
+
"grad_norm": 3.6021695137023926,
|
| 280 |
+
"learning_rate": 3.7468671679198e-05,
|
| 281 |
+
"loss": 0.194,
|
| 282 |
+
"step": 14000
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.7873254564983885,
|
| 286 |
+
"grad_norm": 4.42069149017334,
|
| 287 |
+
"learning_rate": 3.702112423916935e-05,
|
| 288 |
+
"loss": 0.2065,
|
| 289 |
+
"step": 14500
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 8.0,
|
| 293 |
+
"eval_loss": 2.6083967685699463,
|
| 294 |
+
"eval_rouge2_fmeasure": 0.3468,
|
| 295 |
+
"eval_rouge2_precision": 0.3468,
|
| 296 |
+
"eval_rouge2_recall": 0.3468,
|
| 297 |
+
"eval_runtime": 108.8525,
|
| 298 |
+
"eval_samples_per_second": 43.977,
|
| 299 |
+
"eval_steps_per_second": 2.756,
|
| 300 |
+
"step": 14896
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 8.055853920515574,
|
| 304 |
+
"grad_norm": 3.3351027965545654,
|
| 305 |
+
"learning_rate": 3.6573576799140714e-05,
|
| 306 |
+
"loss": 0.1976,
|
| 307 |
+
"step": 15000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 8.324382384532761,
|
| 311 |
+
"grad_norm": 4.0378007888793945,
|
| 312 |
+
"learning_rate": 3.6126029359112065e-05,
|
| 313 |
+
"loss": 0.1395,
|
| 314 |
+
"step": 15500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 8.592910848549947,
|
| 318 |
+
"grad_norm": 4.384516716003418,
|
| 319 |
+
"learning_rate": 3.567848191908342e-05,
|
| 320 |
+
"loss": 0.1485,
|
| 321 |
+
"step": 16000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 8.861439312567132,
|
| 325 |
+
"grad_norm": 3.7494637966156006,
|
| 326 |
+
"learning_rate": 3.5230934479054786e-05,
|
| 327 |
+
"loss": 0.1555,
|
| 328 |
+
"step": 16500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 9.0,
|
| 332 |
+
"eval_loss": 2.697331428527832,
|
| 333 |
+
"eval_rouge2_fmeasure": 0.345,
|
| 334 |
+
"eval_rouge2_precision": 0.345,
|
| 335 |
+
"eval_rouge2_recall": 0.345,
|
| 336 |
+
"eval_runtime": 108.5316,
|
| 337 |
+
"eval_samples_per_second": 44.107,
|
| 338 |
+
"eval_steps_per_second": 2.764,
|
| 339 |
+
"step": 16758
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 9.129967776584317,
|
| 343 |
+
"grad_norm": 2.9512407779693604,
|
| 344 |
+
"learning_rate": 3.4783387039026136e-05,
|
| 345 |
+
"loss": 0.1341,
|
| 346 |
+
"step": 17000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 9.398496240601503,
|
| 350 |
+
"grad_norm": 2.6564815044403076,
|
| 351 |
+
"learning_rate": 3.433583959899749e-05,
|
| 352 |
+
"loss": 0.1092,
|
| 353 |
+
"step": 17500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 9.66702470461869,
|
| 357 |
+
"grad_norm": 2.2893385887145996,
|
| 358 |
+
"learning_rate": 3.388829215896886e-05,
|
| 359 |
+
"loss": 0.1169,
|
| 360 |
+
"step": 18000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 9.935553168635876,
|
| 364 |
+
"grad_norm": 4.070418834686279,
|
| 365 |
+
"learning_rate": 3.344074471894021e-05,
|
| 366 |
+
"loss": 0.1207,
|
| 367 |
+
"step": 18500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 10.0,
|
| 371 |
+
"eval_loss": 2.7456772327423096,
|
| 372 |
+
"eval_rouge2_fmeasure": 0.3462,
|
| 373 |
+
"eval_rouge2_precision": 0.3462,
|
| 374 |
+
"eval_rouge2_recall": 0.3462,
|
| 375 |
+
"eval_runtime": 108.495,
|
| 376 |
+
"eval_samples_per_second": 44.122,
|
| 377 |
+
"eval_steps_per_second": 2.765,
|
| 378 |
+
"step": 18620
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 10.204081632653061,
|
| 382 |
+
"grad_norm": 2.9593915939331055,
|
| 383 |
+
"learning_rate": 3.2993197278911564e-05,
|
| 384 |
+
"loss": 0.0937,
|
| 385 |
+
"step": 19000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 10.472610096670246,
|
| 389 |
+
"grad_norm": 3.2143216133117676,
|
| 390 |
+
"learning_rate": 3.254564983888292e-05,
|
| 391 |
+
"loss": 0.0905,
|
| 392 |
+
"step": 19500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 10.741138560687434,
|
| 396 |
+
"grad_norm": 3.0030126571655273,
|
| 397 |
+
"learning_rate": 3.209810239885428e-05,
|
| 398 |
+
"loss": 0.0942,
|
| 399 |
+
"step": 20000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 11.0,
|
| 403 |
+
"eval_loss": 2.8160953521728516,
|
| 404 |
+
"eval_rouge2_fmeasure": 0.3516,
|
| 405 |
+
"eval_rouge2_precision": 0.3516,
|
| 406 |
+
"eval_rouge2_recall": 0.3516,
|
| 407 |
+
"eval_runtime": 108.5954,
|
| 408 |
+
"eval_samples_per_second": 44.081,
|
| 409 |
+
"eval_steps_per_second": 2.763,
|
| 410 |
+
"step": 20482
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 11.009667024704619,
|
| 414 |
+
"grad_norm": 2.6557607650756836,
|
| 415 |
+
"learning_rate": 3.1650554958825635e-05,
|
| 416 |
+
"loss": 0.0966,
|
| 417 |
+
"step": 20500
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 11.278195488721805,
|
| 421 |
+
"grad_norm": 2.9481875896453857,
|
| 422 |
+
"learning_rate": 3.120300751879699e-05,
|
| 423 |
+
"loss": 0.0701,
|
| 424 |
+
"step": 21000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 11.54672395273899,
|
| 428 |
+
"grad_norm": 2.3334264755249023,
|
| 429 |
+
"learning_rate": 3.075546007876835e-05,
|
| 430 |
+
"loss": 0.0733,
|
| 431 |
+
"step": 21500
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 11.815252416756175,
|
| 435 |
+
"grad_norm": 2.348362684249878,
|
| 436 |
+
"learning_rate": 3.030791263873971e-05,
|
| 437 |
+
"loss": 0.0793,
|
| 438 |
+
"step": 22000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 12.0,
|
| 442 |
+
"eval_loss": 2.861933708190918,
|
| 443 |
+
"eval_rouge2_fmeasure": 0.351,
|
| 444 |
+
"eval_rouge2_precision": 0.351,
|
| 445 |
+
"eval_rouge2_recall": 0.351,
|
| 446 |
+
"eval_runtime": 109.2797,
|
| 447 |
+
"eval_samples_per_second": 43.805,
|
| 448 |
+
"eval_steps_per_second": 2.745,
|
| 449 |
+
"step": 22344
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 12.083780880773363,
|
| 453 |
+
"grad_norm": 3.213848352432251,
|
| 454 |
+
"learning_rate": 2.9860365198711067e-05,
|
| 455 |
+
"loss": 0.0714,
|
| 456 |
+
"step": 22500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 12.352309344790548,
|
| 460 |
+
"grad_norm": 2.3654093742370605,
|
| 461 |
+
"learning_rate": 2.941281775868242e-05,
|
| 462 |
+
"loss": 0.0585,
|
| 463 |
+
"step": 23000
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 12.620837808807734,
|
| 467 |
+
"grad_norm": 3.389200448989868,
|
| 468 |
+
"learning_rate": 2.896527031865378e-05,
|
| 469 |
+
"loss": 0.0616,
|
| 470 |
+
"step": 23500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 12.889366272824919,
|
| 474 |
+
"grad_norm": 2.646812915802002,
|
| 475 |
+
"learning_rate": 2.8517722878625135e-05,
|
| 476 |
+
"loss": 0.0651,
|
| 477 |
+
"step": 24000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 13.0,
|
| 481 |
+
"eval_loss": 2.9142892360687256,
|
| 482 |
+
"eval_rouge2_fmeasure": 0.3507,
|
| 483 |
+
"eval_rouge2_precision": 0.3507,
|
| 484 |
+
"eval_rouge2_recall": 0.3507,
|
| 485 |
+
"eval_runtime": 109.4829,
|
| 486 |
+
"eval_samples_per_second": 43.724,
|
| 487 |
+
"eval_steps_per_second": 2.74,
|
| 488 |
+
"step": 24206
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 13.157894736842104,
|
| 492 |
+
"grad_norm": 1.8852452039718628,
|
| 493 |
+
"learning_rate": 2.8070175438596492e-05,
|
| 494 |
+
"loss": 0.0552,
|
| 495 |
+
"step": 24500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 13.426423200859292,
|
| 499 |
+
"grad_norm": 3.123187780380249,
|
| 500 |
+
"learning_rate": 2.7622627998567852e-05,
|
| 501 |
+
"loss": 0.0503,
|
| 502 |
+
"step": 25000
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 13.694951664876477,
|
| 506 |
+
"grad_norm": 3.341310739517212,
|
| 507 |
+
"learning_rate": 2.7175080558539206e-05,
|
| 508 |
+
"loss": 0.0527,
|
| 509 |
+
"step": 25500
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 13.963480128893663,
|
| 513 |
+
"grad_norm": 1.7806404829025269,
|
| 514 |
+
"learning_rate": 2.6727533118510563e-05,
|
| 515 |
+
"loss": 0.0532,
|
| 516 |
+
"step": 26000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 14.0,
|
| 520 |
+
"eval_loss": 2.962496042251587,
|
| 521 |
+
"eval_rouge2_fmeasure": 0.3549,
|
| 522 |
+
"eval_rouge2_precision": 0.3549,
|
| 523 |
+
"eval_rouge2_recall": 0.3549,
|
| 524 |
+
"eval_runtime": 108.8902,
|
| 525 |
+
"eval_samples_per_second": 43.962,
|
| 526 |
+
"eval_steps_per_second": 2.755,
|
| 527 |
+
"step": 26068
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 14.232008592910848,
|
| 531 |
+
"grad_norm": 1.865723967552185,
|
| 532 |
+
"learning_rate": 2.6279985678481924e-05,
|
| 533 |
+
"loss": 0.044,
|
| 534 |
+
"step": 26500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 14.500537056928035,
|
| 538 |
+
"grad_norm": 2.5559322834014893,
|
| 539 |
+
"learning_rate": 2.5832438238453277e-05,
|
| 540 |
+
"loss": 0.0422,
|
| 541 |
+
"step": 27000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 14.76906552094522,
|
| 545 |
+
"grad_norm": 2.9102725982666016,
|
| 546 |
+
"learning_rate": 2.538489079842463e-05,
|
| 547 |
+
"loss": 0.0428,
|
| 548 |
+
"step": 27500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 15.0,
|
| 552 |
+
"eval_loss": 2.9846532344818115,
|
| 553 |
+
"eval_rouge2_fmeasure": 0.351,
|
| 554 |
+
"eval_rouge2_precision": 0.351,
|
| 555 |
+
"eval_rouge2_recall": 0.351,
|
| 556 |
+
"eval_runtime": 108.8716,
|
| 557 |
+
"eval_samples_per_second": 43.969,
|
| 558 |
+
"eval_steps_per_second": 2.756,
|
| 559 |
+
"step": 27930
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 15.037593984962406,
|
| 563 |
+
"grad_norm": 1.985103726387024,
|
| 564 |
+
"learning_rate": 2.493734335839599e-05,
|
| 565 |
+
"loss": 0.0461,
|
| 566 |
+
"step": 28000
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 15.306122448979592,
|
| 570 |
+
"grad_norm": 2.463585138320923,
|
| 571 |
+
"learning_rate": 2.448979591836735e-05,
|
| 572 |
+
"loss": 0.0366,
|
| 573 |
+
"step": 28500
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 15.574650912996777,
|
| 577 |
+
"grad_norm": 1.7706794738769531,
|
| 578 |
+
"learning_rate": 2.4042248478338705e-05,
|
| 579 |
+
"loss": 0.0357,
|
| 580 |
+
"step": 29000
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"epoch": 15.843179377013964,
|
| 584 |
+
"grad_norm": 1.8243809938430786,
|
| 585 |
+
"learning_rate": 2.3594701038310063e-05,
|
| 586 |
+
"loss": 0.0361,
|
| 587 |
+
"step": 29500
|
| 588 |
+
},
|
| 589 |
+
{
|
| 590 |
+
"epoch": 16.0,
|
| 591 |
+
"eval_loss": 3.0359654426574707,
|
| 592 |
+
"eval_rouge2_fmeasure": 0.3542,
|
| 593 |
+
"eval_rouge2_precision": 0.3542,
|
| 594 |
+
"eval_rouge2_recall": 0.3542,
|
| 595 |
+
"eval_runtime": 108.0792,
|
| 596 |
+
"eval_samples_per_second": 44.292,
|
| 597 |
+
"eval_steps_per_second": 2.776,
|
| 598 |
+
"step": 29792
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 16.111707841031148,
|
| 602 |
+
"grad_norm": 1.5588297843933105,
|
| 603 |
+
"learning_rate": 2.314715359828142e-05,
|
| 604 |
+
"loss": 0.0335,
|
| 605 |
+
"step": 30000
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 16.380236305048335,
|
| 609 |
+
"grad_norm": 2.016108274459839,
|
| 610 |
+
"learning_rate": 2.2699606158252777e-05,
|
| 611 |
+
"loss": 0.0294,
|
| 612 |
+
"step": 30500
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 16.648764769065522,
|
| 616 |
+
"grad_norm": 2.35304594039917,
|
| 617 |
+
"learning_rate": 2.2252058718224134e-05,
|
| 618 |
+
"loss": 0.0317,
|
| 619 |
+
"step": 31000
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 16.917293233082706,
|
| 623 |
+
"grad_norm": 1.6884231567382812,
|
| 624 |
+
"learning_rate": 2.1804511278195487e-05,
|
| 625 |
+
"loss": 0.0311,
|
| 626 |
+
"step": 31500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 17.0,
|
| 630 |
+
"eval_loss": 3.0791501998901367,
|
| 631 |
+
"eval_rouge2_fmeasure": 0.3557,
|
| 632 |
+
"eval_rouge2_precision": 0.3557,
|
| 633 |
+
"eval_rouge2_recall": 0.3557,
|
| 634 |
+
"eval_runtime": 108.4852,
|
| 635 |
+
"eval_samples_per_second": 44.126,
|
| 636 |
+
"eval_steps_per_second": 2.765,
|
| 637 |
+
"step": 31654
|
| 638 |
+
},
|
| 639 |
+
{
|
| 640 |
+
"epoch": 17.185821697099893,
|
| 641 |
+
"grad_norm": 2.347106695175171,
|
| 642 |
+
"learning_rate": 2.1356963838166848e-05,
|
| 643 |
+
"loss": 0.026,
|
| 644 |
+
"step": 32000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 17.454350161117077,
|
| 648 |
+
"grad_norm": 3.1584882736206055,
|
| 649 |
+
"learning_rate": 2.0909416398138205e-05,
|
| 650 |
+
"loss": 0.0265,
|
| 651 |
+
"step": 32500
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 17.722878625134264,
|
| 655 |
+
"grad_norm": 1.6716469526290894,
|
| 656 |
+
"learning_rate": 2.0461868958109562e-05,
|
| 657 |
+
"loss": 0.0251,
|
| 658 |
+
"step": 33000
|
| 659 |
+
},
|
| 660 |
+
{
|
| 661 |
+
"epoch": 17.99140708915145,
|
| 662 |
+
"grad_norm": 1.3004461526870728,
|
| 663 |
+
"learning_rate": 2.0014321518080916e-05,
|
| 664 |
+
"loss": 0.0255,
|
| 665 |
+
"step": 33500
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 18.0,
|
| 669 |
+
"eval_loss": 3.0882503986358643,
|
| 670 |
+
"eval_rouge2_fmeasure": 0.353,
|
| 671 |
+
"eval_rouge2_precision": 0.353,
|
| 672 |
+
"eval_rouge2_recall": 0.353,
|
| 673 |
+
"eval_runtime": 108.6282,
|
| 674 |
+
"eval_samples_per_second": 44.068,
|
| 675 |
+
"eval_steps_per_second": 2.762,
|
| 676 |
+
"step": 33516
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 18.259935553168635,
|
| 680 |
+
"grad_norm": 0.8307426571846008,
|
| 681 |
+
"learning_rate": 1.9566774078052276e-05,
|
| 682 |
+
"loss": 0.0218,
|
| 683 |
+
"step": 34000
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 18.528464017185822,
|
| 687 |
+
"grad_norm": 1.7560513019561768,
|
| 688 |
+
"learning_rate": 1.9119226638023633e-05,
|
| 689 |
+
"loss": 0.0209,
|
| 690 |
+
"step": 34500
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 18.796992481203006,
|
| 694 |
+
"grad_norm": 1.8823086023330688,
|
| 695 |
+
"learning_rate": 1.8671679197994987e-05,
|
| 696 |
+
"loss": 0.0217,
|
| 697 |
+
"step": 35000
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 19.0,
|
| 701 |
+
"eval_loss": 3.1189284324645996,
|
| 702 |
+
"eval_rouge2_fmeasure": 0.3525,
|
| 703 |
+
"eval_rouge2_precision": 0.3525,
|
| 704 |
+
"eval_rouge2_recall": 0.3525,
|
| 705 |
+
"eval_runtime": 108.8718,
|
| 706 |
+
"eval_samples_per_second": 43.969,
|
| 707 |
+
"eval_steps_per_second": 2.756,
|
| 708 |
+
"step": 35378
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 19.065520945220193,
|
| 712 |
+
"grad_norm": 2.705395460128784,
|
| 713 |
+
"learning_rate": 1.8224131757966344e-05,
|
| 714 |
+
"loss": 0.0211,
|
| 715 |
+
"step": 35500
|
| 716 |
+
},
|
| 717 |
+
{
|
| 718 |
+
"epoch": 19.33404940923738,
|
| 719 |
+
"grad_norm": 0.777070164680481,
|
| 720 |
+
"learning_rate": 1.7776584317937704e-05,
|
| 721 |
+
"loss": 0.0171,
|
| 722 |
+
"step": 36000
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 19.602577873254564,
|
| 726 |
+
"grad_norm": 0.8922987580299377,
|
| 727 |
+
"learning_rate": 1.7329036877909058e-05,
|
| 728 |
+
"loss": 0.0194,
|
| 729 |
+
"step": 36500
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 19.87110633727175,
|
| 733 |
+
"grad_norm": 1.2800770998001099,
|
| 734 |
+
"learning_rate": 1.6881489437880415e-05,
|
| 735 |
+
"loss": 0.0182,
|
| 736 |
+
"step": 37000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 20.0,
|
| 740 |
+
"eval_loss": 3.1534199714660645,
|
| 741 |
+
"eval_rouge2_fmeasure": 0.3553,
|
| 742 |
+
"eval_rouge2_precision": 0.3553,
|
| 743 |
+
"eval_rouge2_recall": 0.3553,
|
| 744 |
+
"eval_runtime": 108.9674,
|
| 745 |
+
"eval_samples_per_second": 43.931,
|
| 746 |
+
"eval_steps_per_second": 2.753,
|
| 747 |
+
"step": 37240
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 20.13963480128894,
|
| 751 |
+
"grad_norm": 0.7617535591125488,
|
| 752 |
+
"learning_rate": 1.6433941997851772e-05,
|
| 753 |
+
"loss": 0.0159,
|
| 754 |
+
"step": 37500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 20.408163265306122,
|
| 758 |
+
"grad_norm": 1.3193833827972412,
|
| 759 |
+
"learning_rate": 1.5986394557823133e-05,
|
| 760 |
+
"loss": 0.0141,
|
| 761 |
+
"step": 38000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 20.67669172932331,
|
| 765 |
+
"grad_norm": 0.7571066617965698,
|
| 766 |
+
"learning_rate": 1.5538847117794486e-05,
|
| 767 |
+
"loss": 0.0151,
|
| 768 |
+
"step": 38500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 20.945220193340493,
|
| 772 |
+
"grad_norm": 0.3401932418346405,
|
| 773 |
+
"learning_rate": 1.5091299677765843e-05,
|
| 774 |
+
"loss": 0.0155,
|
| 775 |
+
"step": 39000
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 21.0,
|
| 779 |
+
"eval_loss": 3.1899983882904053,
|
| 780 |
+
"eval_rouge2_fmeasure": 0.3573,
|
| 781 |
+
"eval_rouge2_precision": 0.3573,
|
| 782 |
+
"eval_rouge2_recall": 0.3573,
|
| 783 |
+
"eval_runtime": 108.6672,
|
| 784 |
+
"eval_samples_per_second": 44.052,
|
| 785 |
+
"eval_steps_per_second": 2.761,
|
| 786 |
+
"step": 39102
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 21.21374865735768,
|
| 790 |
+
"grad_norm": 0.29092851281166077,
|
| 791 |
+
"learning_rate": 1.4643752237737202e-05,
|
| 792 |
+
"loss": 0.0122,
|
| 793 |
+
"step": 39500
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 21.482277121374867,
|
| 797 |
+
"grad_norm": 1.4362056255340576,
|
| 798 |
+
"learning_rate": 1.4196204797708556e-05,
|
| 799 |
+
"loss": 0.0122,
|
| 800 |
+
"step": 40000
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 21.75080558539205,
|
| 804 |
+
"grad_norm": 6.062518119812012,
|
| 805 |
+
"learning_rate": 1.3748657357679915e-05,
|
| 806 |
+
"loss": 0.0119,
|
| 807 |
+
"step": 40500
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 22.0,
|
| 811 |
+
"eval_loss": 3.1969234943389893,
|
| 812 |
+
"eval_rouge2_fmeasure": 0.3561,
|
| 813 |
+
"eval_rouge2_precision": 0.3561,
|
| 814 |
+
"eval_rouge2_recall": 0.3561,
|
| 815 |
+
"eval_runtime": 108.579,
|
| 816 |
+
"eval_samples_per_second": 44.088,
|
| 817 |
+
"eval_steps_per_second": 2.763,
|
| 818 |
+
"step": 40964
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 22.019334049409238,
|
| 822 |
+
"grad_norm": 2.686310052871704,
|
| 823 |
+
"learning_rate": 1.3301109917651272e-05,
|
| 824 |
+
"loss": 0.0126,
|
| 825 |
+
"step": 41000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 22.287862513426422,
|
| 829 |
+
"grad_norm": 1.607986569404602,
|
| 830 |
+
"learning_rate": 1.285356247762263e-05,
|
| 831 |
+
"loss": 0.0104,
|
| 832 |
+
"step": 41500
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 22.55639097744361,
|
| 836 |
+
"grad_norm": 0.5264362692832947,
|
| 837 |
+
"learning_rate": 1.2406015037593984e-05,
|
| 838 |
+
"loss": 0.0099,
|
| 839 |
+
"step": 42000
|
| 840 |
+
},
|
| 841 |
+
{
|
| 842 |
+
"epoch": 22.824919441460796,
|
| 843 |
+
"grad_norm": 0.3918701112270355,
|
| 844 |
+
"learning_rate": 1.1958467597565343e-05,
|
| 845 |
+
"loss": 0.0103,
|
| 846 |
+
"step": 42500
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 23.0,
|
| 850 |
+
"eval_loss": 3.2080607414245605,
|
| 851 |
+
"eval_rouge2_fmeasure": 0.3579,
|
| 852 |
+
"eval_rouge2_precision": 0.3579,
|
| 853 |
+
"eval_rouge2_recall": 0.3579,
|
| 854 |
+
"eval_runtime": 108.8035,
|
| 855 |
+
"eval_samples_per_second": 43.997,
|
| 856 |
+
"eval_steps_per_second": 2.757,
|
| 857 |
+
"step": 42826
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 23.09344790547798,
|
| 861 |
+
"grad_norm": 0.4341621696949005,
|
| 862 |
+
"learning_rate": 1.1510920157536698e-05,
|
| 863 |
+
"loss": 0.0094,
|
| 864 |
+
"step": 43000
|
| 865 |
+
},
|
| 866 |
+
{
|
| 867 |
+
"epoch": 23.361976369495167,
|
| 868 |
+
"grad_norm": 1.674943208694458,
|
| 869 |
+
"learning_rate": 1.1063372717508057e-05,
|
| 870 |
+
"loss": 0.008,
|
| 871 |
+
"step": 43500
|
| 872 |
+
},
|
| 873 |
+
{
|
| 874 |
+
"epoch": 23.63050483351235,
|
| 875 |
+
"grad_norm": 0.439281165599823,
|
| 876 |
+
"learning_rate": 1.0615825277479412e-05,
|
| 877 |
+
"loss": 0.0087,
|
| 878 |
+
"step": 44000
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 23.899033297529538,
|
| 882 |
+
"grad_norm": 0.629629909992218,
|
| 883 |
+
"learning_rate": 1.0168277837450771e-05,
|
| 884 |
+
"loss": 0.0085,
|
| 885 |
+
"step": 44500
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 24.0,
|
| 889 |
+
"eval_loss": 3.2334721088409424,
|
| 890 |
+
"eval_rouge2_fmeasure": 0.358,
|
| 891 |
+
"eval_rouge2_precision": 0.358,
|
| 892 |
+
"eval_rouge2_recall": 0.358,
|
| 893 |
+
"eval_runtime": 108.5696,
|
| 894 |
+
"eval_samples_per_second": 44.092,
|
| 895 |
+
"eval_steps_per_second": 2.763,
|
| 896 |
+
"step": 44688
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 24.167561761546725,
|
| 900 |
+
"grad_norm": 0.44458189606666565,
|
| 901 |
+
"learning_rate": 9.720730397422126e-06,
|
| 902 |
+
"loss": 0.0073,
|
| 903 |
+
"step": 45000
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 24.43609022556391,
|
| 907 |
+
"grad_norm": 0.6769450902938843,
|
| 908 |
+
"learning_rate": 9.273182957393484e-06,
|
| 909 |
+
"loss": 0.0066,
|
| 910 |
+
"step": 45500
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 24.704618689581096,
|
| 914 |
+
"grad_norm": 0.3089818060398102,
|
| 915 |
+
"learning_rate": 8.82563551736484e-06,
|
| 916 |
+
"loss": 0.0065,
|
| 917 |
+
"step": 46000
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 24.97314715359828,
|
| 921 |
+
"grad_norm": 0.40338027477264404,
|
| 922 |
+
"learning_rate": 8.378088077336198e-06,
|
| 923 |
+
"loss": 0.0067,
|
| 924 |
+
"step": 46500
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 25.0,
|
| 928 |
+
"eval_loss": 3.242875576019287,
|
| 929 |
+
"eval_rouge2_fmeasure": 0.3599,
|
| 930 |
+
"eval_rouge2_precision": 0.3599,
|
| 931 |
+
"eval_rouge2_recall": 0.3599,
|
| 932 |
+
"eval_runtime": 109.3379,
|
| 933 |
+
"eval_samples_per_second": 43.782,
|
| 934 |
+
"eval_steps_per_second": 2.744,
|
| 935 |
+
"step": 46550
|
| 936 |
+
},
|
| 937 |
+
{
|
| 938 |
+
"epoch": 25.241675617615467,
|
| 939 |
+
"grad_norm": 0.41913506388664246,
|
| 940 |
+
"learning_rate": 7.930540637307555e-06,
|
| 941 |
+
"loss": 0.0057,
|
| 942 |
+
"step": 47000
|
| 943 |
+
},
|
| 944 |
+
{
|
| 945 |
+
"epoch": 25.510204081632654,
|
| 946 |
+
"grad_norm": 0.20852769911289215,
|
| 947 |
+
"learning_rate": 7.482993197278912e-06,
|
| 948 |
+
"loss": 0.0054,
|
| 949 |
+
"step": 47500
|
| 950 |
+
},
|
| 951 |
+
{
|
| 952 |
+
"epoch": 25.778732545649838,
|
| 953 |
+
"grad_norm": 0.31291133165359497,
|
| 954 |
+
"learning_rate": 7.035445757250268e-06,
|
| 955 |
+
"loss": 0.0053,
|
| 956 |
+
"step": 48000
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 26.0,
|
| 960 |
+
"eval_loss": 3.256943464279175,
|
| 961 |
+
"eval_rouge2_fmeasure": 0.3586,
|
| 962 |
+
"eval_rouge2_precision": 0.3586,
|
| 963 |
+
"eval_rouge2_recall": 0.3586,
|
| 964 |
+
"eval_runtime": 109.1749,
|
| 965 |
+
"eval_samples_per_second": 43.847,
|
| 966 |
+
"eval_steps_per_second": 2.748,
|
| 967 |
+
"step": 48412
|
| 968 |
+
},
|
| 969 |
+
{
|
| 970 |
+
"epoch": 26.047261009667025,
|
| 971 |
+
"grad_norm": 0.19362983107566833,
|
| 972 |
+
"learning_rate": 6.587898317221626e-06,
|
| 973 |
+
"loss": 0.0052,
|
| 974 |
+
"step": 48500
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 26.31578947368421,
|
| 978 |
+
"grad_norm": 0.7348465323448181,
|
| 979 |
+
"learning_rate": 6.140350877192982e-06,
|
| 980 |
+
"loss": 0.0044,
|
| 981 |
+
"step": 49000
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 26.584317937701396,
|
| 985 |
+
"grad_norm": 0.08919904381036758,
|
| 986 |
+
"learning_rate": 5.692803437164339e-06,
|
| 987 |
+
"loss": 0.0048,
|
| 988 |
+
"step": 49500
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 26.852846401718583,
|
| 992 |
+
"grad_norm": 0.22906944155693054,
|
| 993 |
+
"learning_rate": 5.245255997135696e-06,
|
| 994 |
+
"loss": 0.0043,
|
| 995 |
+
"step": 50000
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 27.0,
|
| 999 |
+
"eval_loss": 3.264575719833374,
|
| 1000 |
+
"eval_rouge2_fmeasure": 0.3604,
|
| 1001 |
+
"eval_rouge2_precision": 0.3604,
|
| 1002 |
+
"eval_rouge2_recall": 0.3604,
|
| 1003 |
+
"eval_runtime": 108.6063,
|
| 1004 |
+
"eval_samples_per_second": 44.077,
|
| 1005 |
+
"eval_steps_per_second": 2.762,
|
| 1006 |
+
"step": 50274
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 27.121374865735767,
|
| 1010 |
+
"grad_norm": 0.2078796625137329,
|
| 1011 |
+
"learning_rate": 4.797708557107053e-06,
|
| 1012 |
+
"loss": 0.0037,
|
| 1013 |
+
"step": 50500
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 27.389903329752954,
|
| 1017 |
+
"grad_norm": 0.6463788151741028,
|
| 1018 |
+
"learning_rate": 4.3501611170784104e-06,
|
| 1019 |
+
"loss": 0.0034,
|
| 1020 |
+
"step": 51000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 27.65843179377014,
|
| 1024 |
+
"grad_norm": 0.19699470698833466,
|
| 1025 |
+
"learning_rate": 3.9026136770497675e-06,
|
| 1026 |
+
"loss": 0.0032,
|
| 1027 |
+
"step": 51500
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 27.926960257787325,
|
| 1031 |
+
"grad_norm": 0.08220793306827545,
|
| 1032 |
+
"learning_rate": 3.455066237021124e-06,
|
| 1033 |
+
"loss": 0.0032,
|
| 1034 |
+
"step": 52000
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 28.0,
|
| 1038 |
+
"eval_loss": 3.2737011909484863,
|
| 1039 |
+
"eval_rouge2_fmeasure": 0.3623,
|
| 1040 |
+
"eval_rouge2_precision": 0.3623,
|
| 1041 |
+
"eval_rouge2_recall": 0.3623,
|
| 1042 |
+
"eval_runtime": 108.5841,
|
| 1043 |
+
"eval_samples_per_second": 44.086,
|
| 1044 |
+
"eval_steps_per_second": 2.763,
|
| 1045 |
+
"step": 52136
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 28.195488721804512,
|
| 1049 |
+
"grad_norm": 0.09868729114532471,
|
| 1050 |
+
"learning_rate": 3.007518796992481e-06,
|
| 1051 |
+
"loss": 0.0026,
|
| 1052 |
+
"step": 52500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 28.464017185821696,
|
| 1056 |
+
"grad_norm": 0.12712427973747253,
|
| 1057 |
+
"learning_rate": 2.5599713569638383e-06,
|
| 1058 |
+
"loss": 0.0025,
|
| 1059 |
+
"step": 53000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 28.732545649838883,
|
| 1063 |
+
"grad_norm": 0.12487781047821045,
|
| 1064 |
+
"learning_rate": 2.1124239169351953e-06,
|
| 1065 |
+
"loss": 0.0027,
|
| 1066 |
+
"step": 53500
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 29.0,
|
| 1070 |
+
"eval_loss": 3.2774112224578857,
|
| 1071 |
+
"eval_rouge2_fmeasure": 0.3629,
|
| 1072 |
+
"eval_rouge2_precision": 0.3629,
|
| 1073 |
+
"eval_rouge2_recall": 0.3629,
|
| 1074 |
+
"eval_runtime": 108.3776,
|
| 1075 |
+
"eval_samples_per_second": 44.17,
|
| 1076 |
+
"eval_steps_per_second": 2.768,
|
| 1077 |
+
"step": 53998
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 29.00107411385607,
|
| 1081 |
+
"grad_norm": 0.15216058492660522,
|
| 1082 |
+
"learning_rate": 1.664876476906552e-06,
|
| 1083 |
+
"loss": 0.0028,
|
| 1084 |
+
"step": 54000
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 29.269602577873254,
|
| 1088 |
+
"grad_norm": 0.1937492936849594,
|
| 1089 |
+
"learning_rate": 1.217329036877909e-06,
|
| 1090 |
+
"loss": 0.0025,
|
| 1091 |
+
"step": 54500
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 29.53813104189044,
|
| 1095 |
+
"grad_norm": 0.08752380311489105,
|
| 1096 |
+
"learning_rate": 7.697815968492661e-07,
|
| 1097 |
+
"loss": 0.002,
|
| 1098 |
+
"step": 55000
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 29.806659505907625,
|
| 1102 |
+
"grad_norm": 0.07473881542682648,
|
| 1103 |
+
"learning_rate": 3.22234156820623e-07,
|
| 1104 |
+
"loss": 0.002,
|
| 1105 |
+
"step": 55500
|
| 1106 |
+
}
|
| 1107 |
+
],
|
| 1108 |
+
"logging_steps": 500,
|
| 1109 |
+
"max_steps": 55860,
|
| 1110 |
+
"num_input_tokens_seen": 0,
|
| 1111 |
+
"num_train_epochs": 30,
|
| 1112 |
+
"save_steps": 500,
|
| 1113 |
+
"stateful_callbacks": {
|
| 1114 |
+
"TrainerControl": {
|
| 1115 |
+
"args": {
|
| 1116 |
+
"should_epoch_stop": false,
|
| 1117 |
+
"should_evaluate": false,
|
| 1118 |
+
"should_log": false,
|
| 1119 |
+
"should_save": true,
|
| 1120 |
+
"should_training_stop": false
|
| 1121 |
+
},
|
| 1122 |
+
"attributes": {}
|
| 1123 |
+
}
|
| 1124 |
+
},
|
| 1125 |
+
"total_flos": 9.816339201776026e+16,
|
| 1126 |
+
"train_batch_size": 16,
|
| 1127 |
+
"trial_name": null,
|
| 1128 |
+
"trial_params": null
|
| 1129 |
+
}
|
EncoderDecoder_2/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:73f16296db0ed34a2fc06867f280c290bb1db767db7e399585cc5d9e108d0438
|
| 3 |
+
size 5368
|