Commit
·
32bace4
1
Parent(s):
ed11604
Upload 11 files
Browse files- config.json +31 -0
- generation_config.json +7 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +107 -0
- spiece.model +3 -0
- tokenizer_config.json +114 -0
- trainer_state.json +1424 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/t5-v1_1-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 2816,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 1024,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "gelu_new",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "gated-gelu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": true,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"num_decoder_layers": 24,
|
| 20 |
+
"num_heads": 16,
|
| 21 |
+
"num_layers": 24,
|
| 22 |
+
"output_past": true,
|
| 23 |
+
"pad_token_id": 0,
|
| 24 |
+
"relative_attention_max_distance": 128,
|
| 25 |
+
"relative_attention_num_buckets": 32,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.31.0.dev0",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 32128
|
| 31 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.31.0.dev0"
|
| 7 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c0521e3876e712d9e0165f89e18f643fbe01696e1b7f7e87c2fd6ccb1a85951b
|
| 3 |
+
size 6265534689
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea59e5912ca9efba0a436ef179dddcaade7b66516b523b3f94dc1d65e951ba86
|
| 3 |
+
size 3132789733
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:17ce29f4284d56efe19cecf6d51fdd5cc9aa5ad5a520785be042184bdb239d3e
|
| 3 |
+
size 14503
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1c00e80dc7ae30693573182ef91f9010b2b37eaa1c48b41e0d5b64763c19520d
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"eos_token": "</s>",
|
| 105 |
+
"pad_token": "<pad>",
|
| 106 |
+
"unk_token": "<unk>"
|
| 107 |
+
}
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
|
| 3 |
+
size 791656
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,114 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<extra_id_0>",
|
| 4 |
+
"<extra_id_1>",
|
| 5 |
+
"<extra_id_2>",
|
| 6 |
+
"<extra_id_3>",
|
| 7 |
+
"<extra_id_4>",
|
| 8 |
+
"<extra_id_5>",
|
| 9 |
+
"<extra_id_6>",
|
| 10 |
+
"<extra_id_7>",
|
| 11 |
+
"<extra_id_8>",
|
| 12 |
+
"<extra_id_9>",
|
| 13 |
+
"<extra_id_10>",
|
| 14 |
+
"<extra_id_11>",
|
| 15 |
+
"<extra_id_12>",
|
| 16 |
+
"<extra_id_13>",
|
| 17 |
+
"<extra_id_14>",
|
| 18 |
+
"<extra_id_15>",
|
| 19 |
+
"<extra_id_16>",
|
| 20 |
+
"<extra_id_17>",
|
| 21 |
+
"<extra_id_18>",
|
| 22 |
+
"<extra_id_19>",
|
| 23 |
+
"<extra_id_20>",
|
| 24 |
+
"<extra_id_21>",
|
| 25 |
+
"<extra_id_22>",
|
| 26 |
+
"<extra_id_23>",
|
| 27 |
+
"<extra_id_24>",
|
| 28 |
+
"<extra_id_25>",
|
| 29 |
+
"<extra_id_26>",
|
| 30 |
+
"<extra_id_27>",
|
| 31 |
+
"<extra_id_28>",
|
| 32 |
+
"<extra_id_29>",
|
| 33 |
+
"<extra_id_30>",
|
| 34 |
+
"<extra_id_31>",
|
| 35 |
+
"<extra_id_32>",
|
| 36 |
+
"<extra_id_33>",
|
| 37 |
+
"<extra_id_34>",
|
| 38 |
+
"<extra_id_35>",
|
| 39 |
+
"<extra_id_36>",
|
| 40 |
+
"<extra_id_37>",
|
| 41 |
+
"<extra_id_38>",
|
| 42 |
+
"<extra_id_39>",
|
| 43 |
+
"<extra_id_40>",
|
| 44 |
+
"<extra_id_41>",
|
| 45 |
+
"<extra_id_42>",
|
| 46 |
+
"<extra_id_43>",
|
| 47 |
+
"<extra_id_44>",
|
| 48 |
+
"<extra_id_45>",
|
| 49 |
+
"<extra_id_46>",
|
| 50 |
+
"<extra_id_47>",
|
| 51 |
+
"<extra_id_48>",
|
| 52 |
+
"<extra_id_49>",
|
| 53 |
+
"<extra_id_50>",
|
| 54 |
+
"<extra_id_51>",
|
| 55 |
+
"<extra_id_52>",
|
| 56 |
+
"<extra_id_53>",
|
| 57 |
+
"<extra_id_54>",
|
| 58 |
+
"<extra_id_55>",
|
| 59 |
+
"<extra_id_56>",
|
| 60 |
+
"<extra_id_57>",
|
| 61 |
+
"<extra_id_58>",
|
| 62 |
+
"<extra_id_59>",
|
| 63 |
+
"<extra_id_60>",
|
| 64 |
+
"<extra_id_61>",
|
| 65 |
+
"<extra_id_62>",
|
| 66 |
+
"<extra_id_63>",
|
| 67 |
+
"<extra_id_64>",
|
| 68 |
+
"<extra_id_65>",
|
| 69 |
+
"<extra_id_66>",
|
| 70 |
+
"<extra_id_67>",
|
| 71 |
+
"<extra_id_68>",
|
| 72 |
+
"<extra_id_69>",
|
| 73 |
+
"<extra_id_70>",
|
| 74 |
+
"<extra_id_71>",
|
| 75 |
+
"<extra_id_72>",
|
| 76 |
+
"<extra_id_73>",
|
| 77 |
+
"<extra_id_74>",
|
| 78 |
+
"<extra_id_75>",
|
| 79 |
+
"<extra_id_76>",
|
| 80 |
+
"<extra_id_77>",
|
| 81 |
+
"<extra_id_78>",
|
| 82 |
+
"<extra_id_79>",
|
| 83 |
+
"<extra_id_80>",
|
| 84 |
+
"<extra_id_81>",
|
| 85 |
+
"<extra_id_82>",
|
| 86 |
+
"<extra_id_83>",
|
| 87 |
+
"<extra_id_84>",
|
| 88 |
+
"<extra_id_85>",
|
| 89 |
+
"<extra_id_86>",
|
| 90 |
+
"<extra_id_87>",
|
| 91 |
+
"<extra_id_88>",
|
| 92 |
+
"<extra_id_89>",
|
| 93 |
+
"<extra_id_90>",
|
| 94 |
+
"<extra_id_91>",
|
| 95 |
+
"<extra_id_92>",
|
| 96 |
+
"<extra_id_93>",
|
| 97 |
+
"<extra_id_94>",
|
| 98 |
+
"<extra_id_95>",
|
| 99 |
+
"<extra_id_96>",
|
| 100 |
+
"<extra_id_97>",
|
| 101 |
+
"<extra_id_98>",
|
| 102 |
+
"<extra_id_99>"
|
| 103 |
+
],
|
| 104 |
+
"clean_up_tokenization_spaces": true,
|
| 105 |
+
"eos_token": "</s>",
|
| 106 |
+
"extra_ids": 100,
|
| 107 |
+
"model_max_length": 512,
|
| 108 |
+
"pad_token": "<pad>",
|
| 109 |
+
"padding_side": "right",
|
| 110 |
+
"sp_model_kwargs": {},
|
| 111 |
+
"tokenizer_class": "T5Tokenizer",
|
| 112 |
+
"truncation_side": "left",
|
| 113 |
+
"unk_token": "<unk>"
|
| 114 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1424 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.00048321104259230196,
|
| 3 |
+
"best_model_checkpoint": "./home/mkapadni/scratch/finetuning_experiments/models/t5_large_finetuned_meetup_d1/checkpoint-22880",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"global_step": 22880,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 4.000000000000001e-06,
|
| 13 |
+
"loss": 60.4275,
|
| 14 |
+
"step": 100
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.04,
|
| 18 |
+
"learning_rate": 8.000000000000001e-06,
|
| 19 |
+
"loss": 57.9583,
|
| 20 |
+
"step": 200
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 1.2e-05,
|
| 25 |
+
"loss": 54.114,
|
| 26 |
+
"step": 300
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 31 |
+
"loss": 46.0159,
|
| 32 |
+
"step": 400
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.11,
|
| 36 |
+
"learning_rate": 2e-05,
|
| 37 |
+
"loss": 38.0285,
|
| 38 |
+
"step": 500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.13,
|
| 42 |
+
"learning_rate": 1.9910634495084898e-05,
|
| 43 |
+
"loss": 34.5486,
|
| 44 |
+
"step": 600
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15,
|
| 48 |
+
"learning_rate": 1.9821268990169794e-05,
|
| 49 |
+
"loss": 32.3095,
|
| 50 |
+
"step": 700
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.17,
|
| 54 |
+
"learning_rate": 1.9731903485254694e-05,
|
| 55 |
+
"loss": 30.5876,
|
| 56 |
+
"step": 800
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.2,
|
| 60 |
+
"learning_rate": 1.964253798033959e-05,
|
| 61 |
+
"loss": 28.5188,
|
| 62 |
+
"step": 900
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.22,
|
| 66 |
+
"learning_rate": 1.9553172475424487e-05,
|
| 67 |
+
"loss": 5.5749,
|
| 68 |
+
"step": 1000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.24,
|
| 72 |
+
"learning_rate": 1.9463806970509383e-05,
|
| 73 |
+
"loss": 0.1655,
|
| 74 |
+
"step": 1100
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.26,
|
| 78 |
+
"learning_rate": 1.9374441465594283e-05,
|
| 79 |
+
"loss": 0.1157,
|
| 80 |
+
"step": 1200
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.28,
|
| 84 |
+
"learning_rate": 1.928507596067918e-05,
|
| 85 |
+
"loss": 0.1249,
|
| 86 |
+
"step": 1300
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.31,
|
| 90 |
+
"learning_rate": 1.9195710455764076e-05,
|
| 91 |
+
"loss": 0.0992,
|
| 92 |
+
"step": 1400
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.33,
|
| 96 |
+
"learning_rate": 1.9106344950848975e-05,
|
| 97 |
+
"loss": 0.0703,
|
| 98 |
+
"step": 1500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.35,
|
| 102 |
+
"learning_rate": 1.901697944593387e-05,
|
| 103 |
+
"loss": 0.0694,
|
| 104 |
+
"step": 1600
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.37,
|
| 108 |
+
"learning_rate": 1.8927613941018768e-05,
|
| 109 |
+
"loss": 0.0567,
|
| 110 |
+
"step": 1700
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.39,
|
| 114 |
+
"learning_rate": 1.8838248436103664e-05,
|
| 115 |
+
"loss": 0.0478,
|
| 116 |
+
"step": 1800
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.42,
|
| 120 |
+
"learning_rate": 1.8748882931188564e-05,
|
| 121 |
+
"loss": 0.0412,
|
| 122 |
+
"step": 1900
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.44,
|
| 126 |
+
"learning_rate": 1.865951742627346e-05,
|
| 127 |
+
"loss": 0.0377,
|
| 128 |
+
"step": 2000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.46,
|
| 132 |
+
"learning_rate": 1.8570151921358357e-05,
|
| 133 |
+
"loss": 0.0324,
|
| 134 |
+
"step": 2100
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.48,
|
| 138 |
+
"learning_rate": 1.8480786416443253e-05,
|
| 139 |
+
"loss": 0.0291,
|
| 140 |
+
"step": 2200
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.5,
|
| 144 |
+
"learning_rate": 1.839142091152815e-05,
|
| 145 |
+
"loss": 0.0243,
|
| 146 |
+
"step": 2300
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.52,
|
| 150 |
+
"learning_rate": 1.830205540661305e-05,
|
| 151 |
+
"loss": 0.0206,
|
| 152 |
+
"step": 2400
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.55,
|
| 156 |
+
"learning_rate": 1.8212689901697946e-05,
|
| 157 |
+
"loss": 0.0207,
|
| 158 |
+
"step": 2500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.57,
|
| 162 |
+
"learning_rate": 1.8123324396782845e-05,
|
| 163 |
+
"loss": 0.0158,
|
| 164 |
+
"step": 2600
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.59,
|
| 168 |
+
"learning_rate": 1.803395889186774e-05,
|
| 169 |
+
"loss": 0.0201,
|
| 170 |
+
"step": 2700
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.61,
|
| 174 |
+
"learning_rate": 1.7944593386952638e-05,
|
| 175 |
+
"loss": 0.0182,
|
| 176 |
+
"step": 2800
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.63,
|
| 180 |
+
"learning_rate": 1.7855227882037534e-05,
|
| 181 |
+
"loss": 0.0131,
|
| 182 |
+
"step": 2900
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.66,
|
| 186 |
+
"learning_rate": 1.776586237712243e-05,
|
| 187 |
+
"loss": 0.0121,
|
| 188 |
+
"step": 3000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.68,
|
| 192 |
+
"learning_rate": 1.767649687220733e-05,
|
| 193 |
+
"loss": 0.0116,
|
| 194 |
+
"step": 3100
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.7,
|
| 198 |
+
"learning_rate": 1.7587131367292227e-05,
|
| 199 |
+
"loss": 0.0132,
|
| 200 |
+
"step": 3200
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.72,
|
| 204 |
+
"learning_rate": 1.7497765862377123e-05,
|
| 205 |
+
"loss": 0.0132,
|
| 206 |
+
"step": 3300
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.74,
|
| 210 |
+
"learning_rate": 1.740840035746202e-05,
|
| 211 |
+
"loss": 0.0105,
|
| 212 |
+
"step": 3400
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.76,
|
| 216 |
+
"learning_rate": 1.731903485254692e-05,
|
| 217 |
+
"loss": 0.0095,
|
| 218 |
+
"step": 3500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.79,
|
| 222 |
+
"learning_rate": 1.7229669347631816e-05,
|
| 223 |
+
"loss": 0.0132,
|
| 224 |
+
"step": 3600
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.81,
|
| 228 |
+
"learning_rate": 1.7140303842716712e-05,
|
| 229 |
+
"loss": 0.0105,
|
| 230 |
+
"step": 3700
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.83,
|
| 234 |
+
"learning_rate": 1.7050938337801612e-05,
|
| 235 |
+
"loss": 0.0103,
|
| 236 |
+
"step": 3800
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.85,
|
| 240 |
+
"learning_rate": 1.6961572832886508e-05,
|
| 241 |
+
"loss": 0.0075,
|
| 242 |
+
"step": 3900
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.87,
|
| 246 |
+
"learning_rate": 1.6872207327971404e-05,
|
| 247 |
+
"loss": 0.01,
|
| 248 |
+
"step": 4000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.9,
|
| 252 |
+
"learning_rate": 1.67828418230563e-05,
|
| 253 |
+
"loss": 0.0106,
|
| 254 |
+
"step": 4100
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.92,
|
| 258 |
+
"learning_rate": 1.66934763181412e-05,
|
| 259 |
+
"loss": 0.0068,
|
| 260 |
+
"step": 4200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.94,
|
| 264 |
+
"learning_rate": 1.6604110813226097e-05,
|
| 265 |
+
"loss": 0.0073,
|
| 266 |
+
"step": 4300
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.96,
|
| 270 |
+
"learning_rate": 1.6514745308310993e-05,
|
| 271 |
+
"loss": 0.007,
|
| 272 |
+
"step": 4400
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.98,
|
| 276 |
+
"learning_rate": 1.642537980339589e-05,
|
| 277 |
+
"loss": 0.0054,
|
| 278 |
+
"step": 4500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.0,
|
| 282 |
+
"eval_loss": 0.0016077366890385747,
|
| 283 |
+
"eval_runtime": 582.2402,
|
| 284 |
+
"eval_samples_per_second": 1.913,
|
| 285 |
+
"eval_steps_per_second": 1.913,
|
| 286 |
+
"step": 4576
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.01,
|
| 290 |
+
"learning_rate": 1.6336014298480786e-05,
|
| 291 |
+
"loss": 0.0069,
|
| 292 |
+
"step": 4600
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.03,
|
| 296 |
+
"learning_rate": 1.6246648793565686e-05,
|
| 297 |
+
"loss": 0.0035,
|
| 298 |
+
"step": 4700
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.05,
|
| 302 |
+
"learning_rate": 1.6157283288650582e-05,
|
| 303 |
+
"loss": 0.0054,
|
| 304 |
+
"step": 4800
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.07,
|
| 308 |
+
"learning_rate": 1.6067917783735482e-05,
|
| 309 |
+
"loss": 0.0054,
|
| 310 |
+
"step": 4900
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.09,
|
| 314 |
+
"learning_rate": 1.5978552278820375e-05,
|
| 315 |
+
"loss": 0.0027,
|
| 316 |
+
"step": 5000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.11,
|
| 320 |
+
"learning_rate": 1.5889186773905275e-05,
|
| 321 |
+
"loss": 0.0047,
|
| 322 |
+
"step": 5100
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.14,
|
| 326 |
+
"learning_rate": 1.579982126899017e-05,
|
| 327 |
+
"loss": 0.0049,
|
| 328 |
+
"step": 5200
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.16,
|
| 332 |
+
"learning_rate": 1.5710455764075067e-05,
|
| 333 |
+
"loss": 0.0033,
|
| 334 |
+
"step": 5300
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.18,
|
| 338 |
+
"learning_rate": 1.5621090259159967e-05,
|
| 339 |
+
"loss": 0.0044,
|
| 340 |
+
"step": 5400
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.2,
|
| 344 |
+
"learning_rate": 1.5531724754244863e-05,
|
| 345 |
+
"loss": 0.0037,
|
| 346 |
+
"step": 5500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.22,
|
| 350 |
+
"learning_rate": 1.544235924932976e-05,
|
| 351 |
+
"loss": 0.0034,
|
| 352 |
+
"step": 5600
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.25,
|
| 356 |
+
"learning_rate": 1.5352993744414656e-05,
|
| 357 |
+
"loss": 0.0042,
|
| 358 |
+
"step": 5700
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.27,
|
| 362 |
+
"learning_rate": 1.5263628239499556e-05,
|
| 363 |
+
"loss": 0.0045,
|
| 364 |
+
"step": 5800
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 1.29,
|
| 368 |
+
"learning_rate": 1.517426273458445e-05,
|
| 369 |
+
"loss": 0.0036,
|
| 370 |
+
"step": 5900
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.31,
|
| 374 |
+
"learning_rate": 1.5084897229669349e-05,
|
| 375 |
+
"loss": 0.004,
|
| 376 |
+
"step": 6000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.33,
|
| 380 |
+
"learning_rate": 1.4995531724754247e-05,
|
| 381 |
+
"loss": 0.0045,
|
| 382 |
+
"step": 6100
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.35,
|
| 386 |
+
"learning_rate": 1.4906166219839145e-05,
|
| 387 |
+
"loss": 0.0031,
|
| 388 |
+
"step": 6200
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.38,
|
| 392 |
+
"learning_rate": 1.481680071492404e-05,
|
| 393 |
+
"loss": 0.0028,
|
| 394 |
+
"step": 6300
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.4,
|
| 398 |
+
"learning_rate": 1.4727435210008937e-05,
|
| 399 |
+
"loss": 0.003,
|
| 400 |
+
"step": 6400
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 1.42,
|
| 404 |
+
"learning_rate": 1.4638069705093835e-05,
|
| 405 |
+
"loss": 0.0035,
|
| 406 |
+
"step": 6500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.44,
|
| 410 |
+
"learning_rate": 1.4548704200178732e-05,
|
| 411 |
+
"loss": 0.0028,
|
| 412 |
+
"step": 6600
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 1.46,
|
| 416 |
+
"learning_rate": 1.445933869526363e-05,
|
| 417 |
+
"loss": 0.0036,
|
| 418 |
+
"step": 6700
|
| 419 |
+
},
|
| 420 |
+
{
|
| 421 |
+
"epoch": 1.49,
|
| 422 |
+
"learning_rate": 1.4369973190348528e-05,
|
| 423 |
+
"loss": 0.0047,
|
| 424 |
+
"step": 6800
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.51,
|
| 428 |
+
"learning_rate": 1.4280607685433422e-05,
|
| 429 |
+
"loss": 0.0016,
|
| 430 |
+
"step": 6900
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.53,
|
| 434 |
+
"learning_rate": 1.419124218051832e-05,
|
| 435 |
+
"loss": 0.0034,
|
| 436 |
+
"step": 7000
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.55,
|
| 440 |
+
"learning_rate": 1.4101876675603219e-05,
|
| 441 |
+
"loss": 0.0037,
|
| 442 |
+
"step": 7100
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 1.57,
|
| 446 |
+
"learning_rate": 1.4012511170688115e-05,
|
| 447 |
+
"loss": 8.1941,
|
| 448 |
+
"step": 7200
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.6,
|
| 452 |
+
"learning_rate": 1.3923145665773013e-05,
|
| 453 |
+
"loss": 28.1153,
|
| 454 |
+
"step": 7300
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 1.62,
|
| 458 |
+
"learning_rate": 1.3833780160857911e-05,
|
| 459 |
+
"loss": 34.0805,
|
| 460 |
+
"step": 7400
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 1.64,
|
| 464 |
+
"learning_rate": 1.3744414655942809e-05,
|
| 465 |
+
"loss": 34.9717,
|
| 466 |
+
"step": 7500
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 1.66,
|
| 470 |
+
"learning_rate": 1.3655049151027704e-05,
|
| 471 |
+
"loss": 33.9775,
|
| 472 |
+
"step": 7600
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 1.68,
|
| 476 |
+
"learning_rate": 1.3565683646112602e-05,
|
| 477 |
+
"loss": 30.7487,
|
| 478 |
+
"step": 7700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.7,
|
| 482 |
+
"learning_rate": 1.34763181411975e-05,
|
| 483 |
+
"loss": 15.3977,
|
| 484 |
+
"step": 7800
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 1.73,
|
| 488 |
+
"learning_rate": 1.3386952636282396e-05,
|
| 489 |
+
"loss": 1.1498,
|
| 490 |
+
"step": 7900
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.75,
|
| 494 |
+
"learning_rate": 1.3297587131367293e-05,
|
| 495 |
+
"loss": 0.1552,
|
| 496 |
+
"step": 8000
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.77,
|
| 500 |
+
"learning_rate": 1.320822162645219e-05,
|
| 501 |
+
"loss": 0.0114,
|
| 502 |
+
"step": 8100
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 1.79,
|
| 506 |
+
"learning_rate": 1.3118856121537087e-05,
|
| 507 |
+
"loss": 0.0909,
|
| 508 |
+
"step": 8200
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.81,
|
| 512 |
+
"learning_rate": 1.3029490616621985e-05,
|
| 513 |
+
"loss": 0.0104,
|
| 514 |
+
"step": 8300
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.84,
|
| 518 |
+
"learning_rate": 1.2940125111706883e-05,
|
| 519 |
+
"loss": 0.0026,
|
| 520 |
+
"step": 8400
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.86,
|
| 524 |
+
"learning_rate": 1.2850759606791778e-05,
|
| 525 |
+
"loss": 0.0028,
|
| 526 |
+
"step": 8500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.88,
|
| 530 |
+
"learning_rate": 1.2761394101876676e-05,
|
| 531 |
+
"loss": 0.0026,
|
| 532 |
+
"step": 8600
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.9,
|
| 536 |
+
"learning_rate": 1.2672028596961574e-05,
|
| 537 |
+
"loss": 0.0026,
|
| 538 |
+
"step": 8700
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.92,
|
| 542 |
+
"learning_rate": 1.2582663092046472e-05,
|
| 543 |
+
"loss": 0.0018,
|
| 544 |
+
"step": 8800
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.94,
|
| 548 |
+
"learning_rate": 1.2493297587131368e-05,
|
| 549 |
+
"loss": 0.0023,
|
| 550 |
+
"step": 8900
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.97,
|
| 554 |
+
"learning_rate": 1.2403932082216266e-05,
|
| 555 |
+
"loss": 0.1979,
|
| 556 |
+
"step": 9000
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 1.99,
|
| 560 |
+
"learning_rate": 1.2314566577301164e-05,
|
| 561 |
+
"loss": 0.0038,
|
| 562 |
+
"step": 9100
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 2.0,
|
| 566 |
+
"eval_loss": 0.0005774666788056493,
|
| 567 |
+
"eval_runtime": 582.3329,
|
| 568 |
+
"eval_samples_per_second": 1.913,
|
| 569 |
+
"eval_steps_per_second": 1.913,
|
| 570 |
+
"step": 9152
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.01,
|
| 574 |
+
"learning_rate": 1.2225201072386059e-05,
|
| 575 |
+
"loss": 0.0013,
|
| 576 |
+
"step": 9200
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.03,
|
| 580 |
+
"learning_rate": 1.2135835567470957e-05,
|
| 581 |
+
"loss": 0.0623,
|
| 582 |
+
"step": 9300
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.05,
|
| 586 |
+
"learning_rate": 1.2046470062555855e-05,
|
| 587 |
+
"loss": 0.0007,
|
| 588 |
+
"step": 9400
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 2.08,
|
| 592 |
+
"learning_rate": 1.1957104557640751e-05,
|
| 593 |
+
"loss": 0.0019,
|
| 594 |
+
"step": 9500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.1,
|
| 598 |
+
"learning_rate": 1.186773905272565e-05,
|
| 599 |
+
"loss": 0.002,
|
| 600 |
+
"step": 9600
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.12,
|
| 604 |
+
"learning_rate": 1.1778373547810546e-05,
|
| 605 |
+
"loss": 0.0012,
|
| 606 |
+
"step": 9700
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.14,
|
| 610 |
+
"learning_rate": 1.1689008042895442e-05,
|
| 611 |
+
"loss": 0.0018,
|
| 612 |
+
"step": 9800
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.16,
|
| 616 |
+
"learning_rate": 1.159964253798034e-05,
|
| 617 |
+
"loss": 0.0016,
|
| 618 |
+
"step": 9900
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.19,
|
| 622 |
+
"learning_rate": 1.1510277033065238e-05,
|
| 623 |
+
"loss": 0.0013,
|
| 624 |
+
"step": 10000
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 2.21,
|
| 628 |
+
"learning_rate": 1.1420911528150136e-05,
|
| 629 |
+
"loss": 0.0011,
|
| 630 |
+
"step": 10100
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 2.23,
|
| 634 |
+
"learning_rate": 1.1331546023235031e-05,
|
| 635 |
+
"loss": 0.0021,
|
| 636 |
+
"step": 10200
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 2.25,
|
| 640 |
+
"learning_rate": 1.1242180518319929e-05,
|
| 641 |
+
"loss": 0.0019,
|
| 642 |
+
"step": 10300
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.27,
|
| 646 |
+
"learning_rate": 1.1152815013404827e-05,
|
| 647 |
+
"loss": 0.0017,
|
| 648 |
+
"step": 10400
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.29,
|
| 652 |
+
"learning_rate": 1.1063449508489723e-05,
|
| 653 |
+
"loss": 0.0013,
|
| 654 |
+
"step": 10500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 2.32,
|
| 658 |
+
"learning_rate": 1.0974084003574621e-05,
|
| 659 |
+
"loss": 0.0015,
|
| 660 |
+
"step": 10600
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 2.34,
|
| 664 |
+
"learning_rate": 1.088471849865952e-05,
|
| 665 |
+
"loss": 0.0021,
|
| 666 |
+
"step": 10700
|
| 667 |
+
},
|
| 668 |
+
{
|
| 669 |
+
"epoch": 2.36,
|
| 670 |
+
"learning_rate": 1.0795352993744414e-05,
|
| 671 |
+
"loss": 0.0012,
|
| 672 |
+
"step": 10800
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 2.38,
|
| 676 |
+
"learning_rate": 1.0705987488829312e-05,
|
| 677 |
+
"loss": 0.001,
|
| 678 |
+
"step": 10900
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.4,
|
| 682 |
+
"learning_rate": 1.061662198391421e-05,
|
| 683 |
+
"loss": 0.0011,
|
| 684 |
+
"step": 11000
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 2.43,
|
| 688 |
+
"learning_rate": 1.0527256478999108e-05,
|
| 689 |
+
"loss": 0.0011,
|
| 690 |
+
"step": 11100
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 2.45,
|
| 694 |
+
"learning_rate": 1.0437890974084005e-05,
|
| 695 |
+
"loss": 0.0024,
|
| 696 |
+
"step": 11200
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 2.47,
|
| 700 |
+
"learning_rate": 1.0348525469168903e-05,
|
| 701 |
+
"loss": 0.0012,
|
| 702 |
+
"step": 11300
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 2.49,
|
| 706 |
+
"learning_rate": 1.0259159964253799e-05,
|
| 707 |
+
"loss": 0.0009,
|
| 708 |
+
"step": 11400
|
| 709 |
+
},
|
| 710 |
+
{
|
| 711 |
+
"epoch": 2.51,
|
| 712 |
+
"learning_rate": 1.0169794459338695e-05,
|
| 713 |
+
"loss": 0.0008,
|
| 714 |
+
"step": 11500
|
| 715 |
+
},
|
| 716 |
+
{
|
| 717 |
+
"epoch": 2.53,
|
| 718 |
+
"learning_rate": 1.0080428954423593e-05,
|
| 719 |
+
"loss": 0.0014,
|
| 720 |
+
"step": 11600
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.56,
|
| 724 |
+
"learning_rate": 9.99106344950849e-06,
|
| 725 |
+
"loss": 0.0021,
|
| 726 |
+
"step": 11700
|
| 727 |
+
},
|
| 728 |
+
{
|
| 729 |
+
"epoch": 2.58,
|
| 730 |
+
"learning_rate": 9.901697944593388e-06,
|
| 731 |
+
"loss": 0.001,
|
| 732 |
+
"step": 11800
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 2.6,
|
| 736 |
+
"learning_rate": 9.812332439678284e-06,
|
| 737 |
+
"loss": 0.0014,
|
| 738 |
+
"step": 11900
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 2.62,
|
| 742 |
+
"learning_rate": 9.722966934763182e-06,
|
| 743 |
+
"loss": 0.0014,
|
| 744 |
+
"step": 12000
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.64,
|
| 748 |
+
"learning_rate": 9.63360142984808e-06,
|
| 749 |
+
"loss": 0.0012,
|
| 750 |
+
"step": 12100
|
| 751 |
+
},
|
| 752 |
+
{
|
| 753 |
+
"epoch": 2.67,
|
| 754 |
+
"learning_rate": 9.544235924932977e-06,
|
| 755 |
+
"loss": 0.001,
|
| 756 |
+
"step": 12200
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 2.69,
|
| 760 |
+
"learning_rate": 9.454870420017875e-06,
|
| 761 |
+
"loss": 0.0004,
|
| 762 |
+
"step": 12300
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.71,
|
| 766 |
+
"learning_rate": 9.365504915102771e-06,
|
| 767 |
+
"loss": 0.0011,
|
| 768 |
+
"step": 12400
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 2.73,
|
| 772 |
+
"learning_rate": 9.276139410187667e-06,
|
| 773 |
+
"loss": 0.0013,
|
| 774 |
+
"step": 12500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.75,
|
| 778 |
+
"learning_rate": 9.186773905272565e-06,
|
| 779 |
+
"loss": 0.0005,
|
| 780 |
+
"step": 12600
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.78,
|
| 784 |
+
"learning_rate": 9.097408400357462e-06,
|
| 785 |
+
"loss": 0.0006,
|
| 786 |
+
"step": 12700
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.8,
|
| 790 |
+
"learning_rate": 9.00804289544236e-06,
|
| 791 |
+
"loss": 0.0005,
|
| 792 |
+
"step": 12800
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.82,
|
| 796 |
+
"learning_rate": 8.918677390527258e-06,
|
| 797 |
+
"loss": 0.001,
|
| 798 |
+
"step": 12900
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 2.84,
|
| 802 |
+
"learning_rate": 8.829311885612154e-06,
|
| 803 |
+
"loss": 0.0013,
|
| 804 |
+
"step": 13000
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 2.86,
|
| 808 |
+
"learning_rate": 8.739946380697052e-06,
|
| 809 |
+
"loss": 0.0013,
|
| 810 |
+
"step": 13100
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 2.88,
|
| 814 |
+
"learning_rate": 8.650580875781949e-06,
|
| 815 |
+
"loss": 0.0007,
|
| 816 |
+
"step": 13200
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.91,
|
| 820 |
+
"learning_rate": 8.561215370866847e-06,
|
| 821 |
+
"loss": 0.0012,
|
| 822 |
+
"step": 13300
|
| 823 |
+
},
|
| 824 |
+
{
|
| 825 |
+
"epoch": 2.93,
|
| 826 |
+
"learning_rate": 8.471849865951743e-06,
|
| 827 |
+
"loss": 0.0008,
|
| 828 |
+
"step": 13400
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.95,
|
| 832 |
+
"learning_rate": 8.38248436103664e-06,
|
| 833 |
+
"loss": 0.0015,
|
| 834 |
+
"step": 13500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.97,
|
| 838 |
+
"learning_rate": 8.293118856121537e-06,
|
| 839 |
+
"loss": 0.0011,
|
| 840 |
+
"step": 13600
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 2.99,
|
| 844 |
+
"learning_rate": 8.203753351206435e-06,
|
| 845 |
+
"loss": 0.001,
|
| 846 |
+
"step": 13700
|
| 847 |
+
},
|
| 848 |
+
{
|
| 849 |
+
"epoch": 3.0,
|
| 850 |
+
"eval_loss": 0.0004990746383555233,
|
| 851 |
+
"eval_runtime": 582.349,
|
| 852 |
+
"eval_samples_per_second": 1.913,
|
| 853 |
+
"eval_steps_per_second": 1.913,
|
| 854 |
+
"step": 13728
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 3.02,
|
| 858 |
+
"learning_rate": 8.114387846291332e-06,
|
| 859 |
+
"loss": 0.0006,
|
| 860 |
+
"step": 13800
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 3.04,
|
| 864 |
+
"learning_rate": 8.02502234137623e-06,
|
| 865 |
+
"loss": 0.0007,
|
| 866 |
+
"step": 13900
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 3.06,
|
| 870 |
+
"learning_rate": 7.935656836461126e-06,
|
| 871 |
+
"loss": 0.0012,
|
| 872 |
+
"step": 14000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.08,
|
| 876 |
+
"learning_rate": 7.846291331546024e-06,
|
| 877 |
+
"loss": 0.0013,
|
| 878 |
+
"step": 14100
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 3.1,
|
| 882 |
+
"learning_rate": 7.75692582663092e-06,
|
| 883 |
+
"loss": 0.0004,
|
| 884 |
+
"step": 14200
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 3.12,
|
| 888 |
+
"learning_rate": 7.667560321715819e-06,
|
| 889 |
+
"loss": 0.0015,
|
| 890 |
+
"step": 14300
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 3.15,
|
| 894 |
+
"learning_rate": 7.578194816800716e-06,
|
| 895 |
+
"loss": 0.0011,
|
| 896 |
+
"step": 14400
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 3.17,
|
| 900 |
+
"learning_rate": 7.488829311885612e-06,
|
| 901 |
+
"loss": 0.0009,
|
| 902 |
+
"step": 14500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 3.19,
|
| 906 |
+
"learning_rate": 7.39946380697051e-06,
|
| 907 |
+
"loss": 0.001,
|
| 908 |
+
"step": 14600
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 3.21,
|
| 912 |
+
"learning_rate": 7.3100983020554075e-06,
|
| 913 |
+
"loss": 0.0004,
|
| 914 |
+
"step": 14700
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 3.23,
|
| 918 |
+
"learning_rate": 7.220732797140304e-06,
|
| 919 |
+
"loss": 0.0013,
|
| 920 |
+
"step": 14800
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 3.26,
|
| 924 |
+
"learning_rate": 7.131367292225202e-06,
|
| 925 |
+
"loss": 0.0007,
|
| 926 |
+
"step": 14900
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 3.28,
|
| 930 |
+
"learning_rate": 7.042001787310099e-06,
|
| 931 |
+
"loss": 0.0007,
|
| 932 |
+
"step": 15000
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 3.3,
|
| 936 |
+
"learning_rate": 6.9526362823949954e-06,
|
| 937 |
+
"loss": 0.0004,
|
| 938 |
+
"step": 15100
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 3.32,
|
| 942 |
+
"learning_rate": 6.8632707774798935e-06,
|
| 943 |
+
"loss": 0.0007,
|
| 944 |
+
"step": 15200
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 3.34,
|
| 948 |
+
"learning_rate": 6.77390527256479e-06,
|
| 949 |
+
"loss": 0.0003,
|
| 950 |
+
"step": 15300
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 3.37,
|
| 954 |
+
"learning_rate": 6.684539767649688e-06,
|
| 955 |
+
"loss": 0.0014,
|
| 956 |
+
"step": 15400
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 3.39,
|
| 960 |
+
"learning_rate": 6.595174262734585e-06,
|
| 961 |
+
"loss": 0.0005,
|
| 962 |
+
"step": 15500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 3.41,
|
| 966 |
+
"learning_rate": 6.5058087578194814e-06,
|
| 967 |
+
"loss": 0.0008,
|
| 968 |
+
"step": 15600
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 3.43,
|
| 972 |
+
"learning_rate": 6.4164432529043795e-06,
|
| 973 |
+
"loss": 0.0009,
|
| 974 |
+
"step": 15700
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 3.45,
|
| 978 |
+
"learning_rate": 6.327077747989277e-06,
|
| 979 |
+
"loss": 0.0004,
|
| 980 |
+
"step": 15800
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 3.47,
|
| 984 |
+
"learning_rate": 6.237712243074175e-06,
|
| 985 |
+
"loss": 0.0008,
|
| 986 |
+
"step": 15900
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 3.5,
|
| 990 |
+
"learning_rate": 6.148346738159071e-06,
|
| 991 |
+
"loss": 0.0009,
|
| 992 |
+
"step": 16000
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 3.52,
|
| 996 |
+
"learning_rate": 6.058981233243968e-06,
|
| 997 |
+
"loss": 0.001,
|
| 998 |
+
"step": 16100
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 3.54,
|
| 1002 |
+
"learning_rate": 5.9696157283288655e-06,
|
| 1003 |
+
"loss": 0.0005,
|
| 1004 |
+
"step": 16200
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 3.56,
|
| 1008 |
+
"learning_rate": 5.880250223413763e-06,
|
| 1009 |
+
"loss": 0.0011,
|
| 1010 |
+
"step": 16300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 3.58,
|
| 1014 |
+
"learning_rate": 5.790884718498661e-06,
|
| 1015 |
+
"loss": 0.0003,
|
| 1016 |
+
"step": 16400
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 3.61,
|
| 1020 |
+
"learning_rate": 5.701519213583557e-06,
|
| 1021 |
+
"loss": 0.0003,
|
| 1022 |
+
"step": 16500
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 3.63,
|
| 1026 |
+
"learning_rate": 5.612153708668454e-06,
|
| 1027 |
+
"loss": 0.0004,
|
| 1028 |
+
"step": 16600
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 3.65,
|
| 1032 |
+
"learning_rate": 5.522788203753352e-06,
|
| 1033 |
+
"loss": 0.0006,
|
| 1034 |
+
"step": 16700
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 3.67,
|
| 1038 |
+
"learning_rate": 5.433422698838249e-06,
|
| 1039 |
+
"loss": 0.0004,
|
| 1040 |
+
"step": 16800
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.69,
|
| 1044 |
+
"learning_rate": 5.344057193923146e-06,
|
| 1045 |
+
"loss": 0.0005,
|
| 1046 |
+
"step": 16900
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 3.72,
|
| 1050 |
+
"learning_rate": 5.254691689008043e-06,
|
| 1051 |
+
"loss": 0.0005,
|
| 1052 |
+
"step": 17000
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.74,
|
| 1056 |
+
"learning_rate": 5.16532618409294e-06,
|
| 1057 |
+
"loss": 0.0006,
|
| 1058 |
+
"step": 17100
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 3.76,
|
| 1062 |
+
"learning_rate": 5.075960679177838e-06,
|
| 1063 |
+
"loss": 0.001,
|
| 1064 |
+
"step": 17200
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.78,
|
| 1068 |
+
"learning_rate": 4.986595174262735e-06,
|
| 1069 |
+
"loss": 0.0002,
|
| 1070 |
+
"step": 17300
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 3.8,
|
| 1074 |
+
"learning_rate": 4.897229669347633e-06,
|
| 1075 |
+
"loss": 0.0006,
|
| 1076 |
+
"step": 17400
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.82,
|
| 1080 |
+
"learning_rate": 4.80786416443253e-06,
|
| 1081 |
+
"loss": 0.0006,
|
| 1082 |
+
"step": 17500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 3.85,
|
| 1086 |
+
"learning_rate": 4.718498659517426e-06,
|
| 1087 |
+
"loss": 0.0006,
|
| 1088 |
+
"step": 17600
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 3.87,
|
| 1092 |
+
"learning_rate": 4.6291331546023235e-06,
|
| 1093 |
+
"loss": 0.0005,
|
| 1094 |
+
"step": 17700
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.89,
|
| 1098 |
+
"learning_rate": 4.5397676496872215e-06,
|
| 1099 |
+
"loss": 0.0007,
|
| 1100 |
+
"step": 17800
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 3.91,
|
| 1104 |
+
"learning_rate": 4.450402144772119e-06,
|
| 1105 |
+
"loss": 0.0004,
|
| 1106 |
+
"step": 17900
|
| 1107 |
+
},
|
| 1108 |
+
{
|
| 1109 |
+
"epoch": 3.93,
|
| 1110 |
+
"learning_rate": 4.361036639857015e-06,
|
| 1111 |
+
"loss": 0.0004,
|
| 1112 |
+
"step": 18000
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 3.96,
|
| 1116 |
+
"learning_rate": 4.271671134941912e-06,
|
| 1117 |
+
"loss": 0.0004,
|
| 1118 |
+
"step": 18100
|
| 1119 |
+
},
|
| 1120 |
+
{
|
| 1121 |
+
"epoch": 3.98,
|
| 1122 |
+
"learning_rate": 4.18230563002681e-06,
|
| 1123 |
+
"loss": 0.0002,
|
| 1124 |
+
"step": 18200
|
| 1125 |
+
},
|
| 1126 |
+
{
|
| 1127 |
+
"epoch": 4.0,
|
| 1128 |
+
"learning_rate": 4.0929401251117075e-06,
|
| 1129 |
+
"loss": 0.0007,
|
| 1130 |
+
"step": 18300
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 4.0,
|
| 1134 |
+
"eval_loss": 0.0004931196453981102,
|
| 1135 |
+
"eval_runtime": 582.3074,
|
| 1136 |
+
"eval_samples_per_second": 1.913,
|
| 1137 |
+
"eval_steps_per_second": 1.913,
|
| 1138 |
+
"step": 18304
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 4.02,
|
| 1142 |
+
"learning_rate": 4.003574620196605e-06,
|
| 1143 |
+
"loss": 0.0007,
|
| 1144 |
+
"step": 18400
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 4.04,
|
| 1148 |
+
"learning_rate": 3.914209115281501e-06,
|
| 1149 |
+
"loss": 0.0005,
|
| 1150 |
+
"step": 18500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 4.06,
|
| 1154 |
+
"learning_rate": 3.824843610366399e-06,
|
| 1155 |
+
"loss": 0.0007,
|
| 1156 |
+
"step": 18600
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 4.09,
|
| 1160 |
+
"learning_rate": 3.7354781054512963e-06,
|
| 1161 |
+
"loss": 0.0003,
|
| 1162 |
+
"step": 18700
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 4.11,
|
| 1166 |
+
"learning_rate": 3.6461126005361935e-06,
|
| 1167 |
+
"loss": 0.0004,
|
| 1168 |
+
"step": 18800
|
| 1169 |
+
},
|
| 1170 |
+
{
|
| 1171 |
+
"epoch": 4.13,
|
| 1172 |
+
"learning_rate": 3.5567470956210903e-06,
|
| 1173 |
+
"loss": 0.0002,
|
| 1174 |
+
"step": 18900
|
| 1175 |
+
},
|
| 1176 |
+
{
|
| 1177 |
+
"epoch": 4.15,
|
| 1178 |
+
"learning_rate": 3.4673815907059875e-06,
|
| 1179 |
+
"loss": 0.0004,
|
| 1180 |
+
"step": 19000
|
| 1181 |
+
},
|
| 1182 |
+
{
|
| 1183 |
+
"epoch": 4.17,
|
| 1184 |
+
"learning_rate": 3.378016085790885e-06,
|
| 1185 |
+
"loss": 0.0003,
|
| 1186 |
+
"step": 19100
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 4.2,
|
| 1190 |
+
"learning_rate": 3.2886505808757823e-06,
|
| 1191 |
+
"loss": 0.0009,
|
| 1192 |
+
"step": 19200
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 4.22,
|
| 1196 |
+
"learning_rate": 3.199285075960679e-06,
|
| 1197 |
+
"loss": 0.0007,
|
| 1198 |
+
"step": 19300
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 4.24,
|
| 1202 |
+
"learning_rate": 3.1099195710455763e-06,
|
| 1203 |
+
"loss": 0.0006,
|
| 1204 |
+
"step": 19400
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 4.26,
|
| 1208 |
+
"learning_rate": 3.020554066130474e-06,
|
| 1209 |
+
"loss": 0.0006,
|
| 1210 |
+
"step": 19500
|
| 1211 |
+
},
|
| 1212 |
+
{
|
| 1213 |
+
"epoch": 4.28,
|
| 1214 |
+
"learning_rate": 2.931188561215371e-06,
|
| 1215 |
+
"loss": 0.0002,
|
| 1216 |
+
"step": 19600
|
| 1217 |
+
},
|
| 1218 |
+
{
|
| 1219 |
+
"epoch": 4.31,
|
| 1220 |
+
"learning_rate": 2.8418230563002683e-06,
|
| 1221 |
+
"loss": 0.0004,
|
| 1222 |
+
"step": 19700
|
| 1223 |
+
},
|
| 1224 |
+
{
|
| 1225 |
+
"epoch": 4.33,
|
| 1226 |
+
"learning_rate": 2.7524575513851655e-06,
|
| 1227 |
+
"loss": 0.0004,
|
| 1228 |
+
"step": 19800
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 4.35,
|
| 1232 |
+
"learning_rate": 2.6630920464700628e-06,
|
| 1233 |
+
"loss": 0.0004,
|
| 1234 |
+
"step": 19900
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 4.37,
|
| 1238 |
+
"learning_rate": 2.57372654155496e-06,
|
| 1239 |
+
"loss": 0.0005,
|
| 1240 |
+
"step": 20000
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 4.39,
|
| 1244 |
+
"learning_rate": 2.484361036639857e-06,
|
| 1245 |
+
"loss": 0.0003,
|
| 1246 |
+
"step": 20100
|
| 1247 |
+
},
|
| 1248 |
+
{
|
| 1249 |
+
"epoch": 4.41,
|
| 1250 |
+
"learning_rate": 2.3949955317247544e-06,
|
| 1251 |
+
"loss": 0.0004,
|
| 1252 |
+
"step": 20200
|
| 1253 |
+
},
|
| 1254 |
+
{
|
| 1255 |
+
"epoch": 4.44,
|
| 1256 |
+
"learning_rate": 2.3056300268096516e-06,
|
| 1257 |
+
"loss": 0.0003,
|
| 1258 |
+
"step": 20300
|
| 1259 |
+
},
|
| 1260 |
+
{
|
| 1261 |
+
"epoch": 4.46,
|
| 1262 |
+
"learning_rate": 2.2162645218945488e-06,
|
| 1263 |
+
"loss": 0.0004,
|
| 1264 |
+
"step": 20400
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 4.48,
|
| 1268 |
+
"learning_rate": 2.1268990169794464e-06,
|
| 1269 |
+
"loss": 0.0005,
|
| 1270 |
+
"step": 20500
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 4.5,
|
| 1274 |
+
"learning_rate": 2.037533512064343e-06,
|
| 1275 |
+
"loss": 0.0003,
|
| 1276 |
+
"step": 20600
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 4.52,
|
| 1280 |
+
"learning_rate": 1.9481680071492408e-06,
|
| 1281 |
+
"loss": 0.0003,
|
| 1282 |
+
"step": 20700
|
| 1283 |
+
},
|
| 1284 |
+
{
|
| 1285 |
+
"epoch": 4.55,
|
| 1286 |
+
"learning_rate": 1.8588025022341378e-06,
|
| 1287 |
+
"loss": 0.0003,
|
| 1288 |
+
"step": 20800
|
| 1289 |
+
},
|
| 1290 |
+
{
|
| 1291 |
+
"epoch": 4.57,
|
| 1292 |
+
"learning_rate": 1.769436997319035e-06,
|
| 1293 |
+
"loss": 0.0003,
|
| 1294 |
+
"step": 20900
|
| 1295 |
+
},
|
| 1296 |
+
{
|
| 1297 |
+
"epoch": 4.59,
|
| 1298 |
+
"learning_rate": 1.6800714924039322e-06,
|
| 1299 |
+
"loss": 0.001,
|
| 1300 |
+
"step": 21000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 4.61,
|
| 1304 |
+
"learning_rate": 1.5907059874888296e-06,
|
| 1305 |
+
"loss": 0.0008,
|
| 1306 |
+
"step": 21100
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 4.63,
|
| 1310 |
+
"learning_rate": 1.5013404825737266e-06,
|
| 1311 |
+
"loss": 0.0002,
|
| 1312 |
+
"step": 21200
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 4.65,
|
| 1316 |
+
"learning_rate": 1.411974977658624e-06,
|
| 1317 |
+
"loss": 0.0005,
|
| 1318 |
+
"step": 21300
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 4.68,
|
| 1322 |
+
"learning_rate": 1.322609472743521e-06,
|
| 1323 |
+
"loss": 0.0004,
|
| 1324 |
+
"step": 21400
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 4.7,
|
| 1328 |
+
"learning_rate": 1.2332439678284184e-06,
|
| 1329 |
+
"loss": 0.0003,
|
| 1330 |
+
"step": 21500
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 4.72,
|
| 1334 |
+
"learning_rate": 1.1438784629133156e-06,
|
| 1335 |
+
"loss": 0.0001,
|
| 1336 |
+
"step": 21600
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 4.74,
|
| 1340 |
+
"learning_rate": 1.0545129579982128e-06,
|
| 1341 |
+
"loss": 0.0006,
|
| 1342 |
+
"step": 21700
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 4.76,
|
| 1346 |
+
"learning_rate": 9.6514745308311e-07,
|
| 1347 |
+
"loss": 0.0004,
|
| 1348 |
+
"step": 21800
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 4.79,
|
| 1352 |
+
"learning_rate": 8.757819481680072e-07,
|
| 1353 |
+
"loss": 0.0002,
|
| 1354 |
+
"step": 21900
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 4.81,
|
| 1358 |
+
"learning_rate": 7.864164432529045e-07,
|
| 1359 |
+
"loss": 0.0004,
|
| 1360 |
+
"step": 22000
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 4.83,
|
| 1364 |
+
"learning_rate": 6.970509383378017e-07,
|
| 1365 |
+
"loss": 0.0001,
|
| 1366 |
+
"step": 22100
|
| 1367 |
+
},
|
| 1368 |
+
{
|
| 1369 |
+
"epoch": 4.85,
|
| 1370 |
+
"learning_rate": 6.076854334226989e-07,
|
| 1371 |
+
"loss": 0.0003,
|
| 1372 |
+
"step": 22200
|
| 1373 |
+
},
|
| 1374 |
+
{
|
| 1375 |
+
"epoch": 4.87,
|
| 1376 |
+
"learning_rate": 5.183199285075961e-07,
|
| 1377 |
+
"loss": 0.0004,
|
| 1378 |
+
"step": 22300
|
| 1379 |
+
},
|
| 1380 |
+
{
|
| 1381 |
+
"epoch": 4.9,
|
| 1382 |
+
"learning_rate": 4.2895442359249334e-07,
|
| 1383 |
+
"loss": 0.0003,
|
| 1384 |
+
"step": 22400
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 4.92,
|
| 1388 |
+
"learning_rate": 3.3958891867739054e-07,
|
| 1389 |
+
"loss": 0.0003,
|
| 1390 |
+
"step": 22500
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 4.94,
|
| 1394 |
+
"learning_rate": 2.502234137622878e-07,
|
| 1395 |
+
"loss": 0.0004,
|
| 1396 |
+
"step": 22600
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 4.96,
|
| 1400 |
+
"learning_rate": 1.60857908847185e-07,
|
| 1401 |
+
"loss": 0.0005,
|
| 1402 |
+
"step": 22700
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 4.98,
|
| 1406 |
+
"learning_rate": 7.149240393208222e-08,
|
| 1407 |
+
"loss": 0.0002,
|
| 1408 |
+
"step": 22800
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 5.0,
|
| 1412 |
+
"eval_loss": 0.00048321104259230196,
|
| 1413 |
+
"eval_runtime": 582.3092,
|
| 1414 |
+
"eval_samples_per_second": 1.913,
|
| 1415 |
+
"eval_steps_per_second": 1.913,
|
| 1416 |
+
"step": 22880
|
| 1417 |
+
}
|
| 1418 |
+
],
|
| 1419 |
+
"max_steps": 22880,
|
| 1420 |
+
"num_train_epochs": 5,
|
| 1421 |
+
"total_flos": 2.1093236286160896e+17,
|
| 1422 |
+
"trial_name": null,
|
| 1423 |
+
"trial_params": null
|
| 1424 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e1492ec39ff0345ea7870432773f0787c0f2e0a4462269355e7d8791e315c6d1
|
| 3 |
+
size 3951
|