init upload. Wait for empirical validation check
Browse files- .gitattributes +10 -0
- checkpoint-10000/config.json +60 -0
- checkpoint-10000/generation_config.json +7 -0
- checkpoint-10000/optimizer.pt +3 -0
- checkpoint-10000/pytorch_model.bin +3 -0
- checkpoint-10000/rng_state.pth +3 -0
- checkpoint-10000/scheduler.pt +3 -0
- checkpoint-10000/trainer_state.json +276 -0
- checkpoint-10000/training_args.bin +3 -0
- checkpoint-15000/config.json +60 -0
- checkpoint-15000/generation_config.json +7 -0
- checkpoint-15000/optimizer.pt +3 -0
- checkpoint-15000/pytorch_model.bin +3 -0
- checkpoint-15000/rng_state.pth +3 -0
- checkpoint-15000/scheduler.pt +3 -0
- checkpoint-15000/trainer_state.json +406 -0
- checkpoint-15000/training_args.bin +3 -0
- checkpoint-20000/config.json +60 -0
- checkpoint-20000/generation_config.json +7 -0
- checkpoint-20000/optimizer.pt +3 -0
- checkpoint-20000/pytorch_model.bin +3 -0
- checkpoint-20000/rng_state.pth +3 -0
- checkpoint-20000/scheduler.pt +3 -0
- checkpoint-20000/trainer_state.json +536 -0
- checkpoint-20000/training_args.bin +3 -0
- checkpoint-5000/config.json +60 -0
- checkpoint-5000/generation_config.json +7 -0
- checkpoint-5000/optimizer.pt +3 -0
- checkpoint-5000/pytorch_model.bin +3 -0
- checkpoint-5000/rng_state.pth +3 -0
- checkpoint-5000/scheduler.pt +3 -0
- checkpoint-5000/trainer_state.json +146 -0
- checkpoint-5000/training_args.bin +3 -0
- config.json +60 -0
- generation_config.json +7 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +276 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,13 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
training_args.bin filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
checkpoint-5000/training_args.bin filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
checkpoint-10000/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
checkpoint-10000/training_args.bin filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
checkpoint-15000/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
|
| 42 |
+
checkpoint-15000/training_args.bin filter=lfs diff=lfs merge=lfs -text
|
| 43 |
+
checkpoint-20000/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
|
| 44 |
+
checkpoint-20000/training_args.bin filter=lfs diff=lfs merge=lfs -text
|
| 45 |
+
checkpoint-5000/pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
|
checkpoint-10000/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DylanJHJ/fidt5-base-nq",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"FiDT5"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 3072,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "relu",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": false,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 12,
|
| 21 |
+
"num_heads": 12,
|
| 22 |
+
"num_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"relative_attention_max_distance": 128,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.31.0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
checkpoint-10000/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"
|
| 7 |
+
}
|
checkpoint-10000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7227b58627e2ec7c2b03f84198efb31d97df1f089b013769ee54e3e128ff1f2
|
| 3 |
+
size 2185605
|
checkpoint-10000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f649cf96d88f5b744d984a87b6dd3f16c60f4f793edaec4685cdf62dda3fe6b0
|
| 3 |
+
size 920014805
|
checkpoint-10000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d263b9835f87e7764463235361192c1e14372603a00d3f5f6edba42d93057da
|
| 3 |
+
size 14575
|
checkpoint-10000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:22b05183e4bc4eb4b1775c568247f66ae6e1c1616d7ca41e2cea3e991bebc2e1
|
| 3 |
+
size 627
|
checkpoint-10000/trainer_state.json
ADDED
|
@@ -0,0 +1,276 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.2356020942408374,
|
| 5 |
+
"global_step": 10000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.26,
|
| 12 |
+
"learning_rate": 6.25e-06,
|
| 13 |
+
"loss": 7.3722,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.26,
|
| 18 |
+
"eval_runtime": 4.9957,
|
| 19 |
+
"eval_samples_per_second": 20.017,
|
| 20 |
+
"eval_steps_per_second": 10.009,
|
| 21 |
+
"step": 500
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.52,
|
| 25 |
+
"learning_rate": 9.895833333333334e-06,
|
| 26 |
+
"loss": 1.5052,
|
| 27 |
+
"step": 1000
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.52,
|
| 31 |
+
"eval_runtime": 5.5484,
|
| 32 |
+
"eval_samples_per_second": 18.023,
|
| 33 |
+
"eval_steps_per_second": 9.012,
|
| 34 |
+
"step": 1000
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.79,
|
| 38 |
+
"learning_rate": 9.635416666666668e-06,
|
| 39 |
+
"loss": 1.3479,
|
| 40 |
+
"step": 1500
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.79,
|
| 44 |
+
"eval_runtime": 5.4316,
|
| 45 |
+
"eval_samples_per_second": 18.411,
|
| 46 |
+
"eval_steps_per_second": 9.205,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.05,
|
| 51 |
+
"learning_rate": 9.375000000000001e-06,
|
| 52 |
+
"loss": 1.3177,
|
| 53 |
+
"step": 2000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"eval_runtime": 5.4246,
|
| 58 |
+
"eval_samples_per_second": 18.435,
|
| 59 |
+
"eval_steps_per_second": 9.217,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.31,
|
| 64 |
+
"learning_rate": 9.114583333333334e-06,
|
| 65 |
+
"loss": 1.2962,
|
| 66 |
+
"step": 2500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.31,
|
| 70 |
+
"eval_runtime": 5.0933,
|
| 71 |
+
"eval_samples_per_second": 19.634,
|
| 72 |
+
"eval_steps_per_second": 9.817,
|
| 73 |
+
"step": 2500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.57,
|
| 77 |
+
"learning_rate": 8.854166666666667e-06,
|
| 78 |
+
"loss": 1.2824,
|
| 79 |
+
"step": 3000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.57,
|
| 83 |
+
"eval_runtime": 6.1894,
|
| 84 |
+
"eval_samples_per_second": 16.157,
|
| 85 |
+
"eval_steps_per_second": 8.078,
|
| 86 |
+
"step": 3000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.83,
|
| 90 |
+
"learning_rate": 8.59375e-06,
|
| 91 |
+
"loss": 1.2784,
|
| 92 |
+
"step": 3500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.83,
|
| 96 |
+
"eval_runtime": 4.7573,
|
| 97 |
+
"eval_samples_per_second": 21.02,
|
| 98 |
+
"eval_steps_per_second": 10.51,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.09,
|
| 103 |
+
"learning_rate": 8.333333333333334e-06,
|
| 104 |
+
"loss": 1.2655,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.09,
|
| 109 |
+
"eval_runtime": 5.5685,
|
| 110 |
+
"eval_samples_per_second": 17.958,
|
| 111 |
+
"eval_steps_per_second": 8.979,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.36,
|
| 116 |
+
"learning_rate": 8.072916666666667e-06,
|
| 117 |
+
"loss": 1.2553,
|
| 118 |
+
"step": 4500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.36,
|
| 122 |
+
"eval_runtime": 6.732,
|
| 123 |
+
"eval_samples_per_second": 14.854,
|
| 124 |
+
"eval_steps_per_second": 7.427,
|
| 125 |
+
"step": 4500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.62,
|
| 129 |
+
"learning_rate": 7.8125e-06,
|
| 130 |
+
"loss": 1.2627,
|
| 131 |
+
"step": 5000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.62,
|
| 135 |
+
"eval_runtime": 4.5523,
|
| 136 |
+
"eval_samples_per_second": 21.967,
|
| 137 |
+
"eval_steps_per_second": 10.983,
|
| 138 |
+
"step": 5000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.88,
|
| 142 |
+
"learning_rate": 7.552083333333334e-06,
|
| 143 |
+
"loss": 1.2358,
|
| 144 |
+
"step": 5500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.88,
|
| 148 |
+
"eval_runtime": 4.906,
|
| 149 |
+
"eval_samples_per_second": 20.383,
|
| 150 |
+
"eval_steps_per_second": 10.192,
|
| 151 |
+
"step": 5500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 3.14,
|
| 155 |
+
"learning_rate": 7.291666666666667e-06,
|
| 156 |
+
"loss": 1.2495,
|
| 157 |
+
"step": 6000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 3.14,
|
| 161 |
+
"eval_runtime": 5.2379,
|
| 162 |
+
"eval_samples_per_second": 19.092,
|
| 163 |
+
"eval_steps_per_second": 9.546,
|
| 164 |
+
"step": 6000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.4,
|
| 168 |
+
"learning_rate": 7.031250000000001e-06,
|
| 169 |
+
"loss": 1.234,
|
| 170 |
+
"step": 6500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.4,
|
| 174 |
+
"eval_runtime": 5.2459,
|
| 175 |
+
"eval_samples_per_second": 19.062,
|
| 176 |
+
"eval_steps_per_second": 9.531,
|
| 177 |
+
"step": 6500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.66,
|
| 181 |
+
"learning_rate": 6.770833333333334e-06,
|
| 182 |
+
"loss": 1.2228,
|
| 183 |
+
"step": 7000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.66,
|
| 187 |
+
"eval_runtime": 5.186,
|
| 188 |
+
"eval_samples_per_second": 19.283,
|
| 189 |
+
"eval_steps_per_second": 9.641,
|
| 190 |
+
"step": 7000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.93,
|
| 194 |
+
"learning_rate": 6.510416666666667e-06,
|
| 195 |
+
"loss": 1.2403,
|
| 196 |
+
"step": 7500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 3.93,
|
| 200 |
+
"eval_runtime": 5.9443,
|
| 201 |
+
"eval_samples_per_second": 16.823,
|
| 202 |
+
"eval_steps_per_second": 8.411,
|
| 203 |
+
"step": 7500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.19,
|
| 207 |
+
"learning_rate": 6.25e-06,
|
| 208 |
+
"loss": 1.2254,
|
| 209 |
+
"step": 8000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.19,
|
| 213 |
+
"eval_runtime": 6.1004,
|
| 214 |
+
"eval_samples_per_second": 16.392,
|
| 215 |
+
"eval_steps_per_second": 8.196,
|
| 216 |
+
"step": 8000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.45,
|
| 220 |
+
"learning_rate": 5.989583333333334e-06,
|
| 221 |
+
"loss": 1.2203,
|
| 222 |
+
"step": 8500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 4.45,
|
| 226 |
+
"eval_runtime": 4.7155,
|
| 227 |
+
"eval_samples_per_second": 21.207,
|
| 228 |
+
"eval_steps_per_second": 10.603,
|
| 229 |
+
"step": 8500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 4.71,
|
| 233 |
+
"learning_rate": 5.729166666666667e-06,
|
| 234 |
+
"loss": 1.2198,
|
| 235 |
+
"step": 9000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 4.71,
|
| 239 |
+
"eval_runtime": 5.2917,
|
| 240 |
+
"eval_samples_per_second": 18.897,
|
| 241 |
+
"eval_steps_per_second": 9.449,
|
| 242 |
+
"step": 9000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.97,
|
| 246 |
+
"learning_rate": 5.468750000000001e-06,
|
| 247 |
+
"loss": 1.2405,
|
| 248 |
+
"step": 9500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.97,
|
| 252 |
+
"eval_runtime": 5.6681,
|
| 253 |
+
"eval_samples_per_second": 17.643,
|
| 254 |
+
"eval_steps_per_second": 8.821,
|
| 255 |
+
"step": 9500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 5.24,
|
| 259 |
+
"learning_rate": 5.208333333333334e-06,
|
| 260 |
+
"loss": 1.2209,
|
| 261 |
+
"step": 10000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 5.24,
|
| 265 |
+
"eval_runtime": 4.7064,
|
| 266 |
+
"eval_samples_per_second": 21.247,
|
| 267 |
+
"eval_steps_per_second": 10.624,
|
| 268 |
+
"step": 10000
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"max_steps": 20000,
|
| 272 |
+
"num_train_epochs": 11,
|
| 273 |
+
"total_flos": 8.36711025084334e+16,
|
| 274 |
+
"trial_name": null,
|
| 275 |
+
"trial_params": null
|
| 276 |
+
}
|
checkpoint-10000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc04dcf4f9e174fdddb9f15c5a6a5d1a48675326eb949a1f635ccc394c1df49d
|
| 3 |
+
size 4027
|
checkpoint-15000/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DylanJHJ/fidt5-base-nq",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"FiDT5"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 3072,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "relu",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": false,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 12,
|
| 21 |
+
"num_heads": 12,
|
| 22 |
+
"num_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"relative_attention_max_distance": 128,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.31.0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
checkpoint-15000/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"
|
| 7 |
+
}
|
checkpoint-15000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:892db98c8d4ee0ae6df2ee4bc9bac20cd21e0c8cf609548ef9d7943d874e58f9
|
| 3 |
+
size 2185605
|
checkpoint-15000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f493aa30c80172a509de63146718cecbe206ae1a242a3ed82218dcfabcd5275e
|
| 3 |
+
size 920014805
|
checkpoint-15000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c62ccd4a92db71c74f014d9b7995d5e599f49acca73c427dbdf507ac0e3a3362
|
| 3 |
+
size 14575
|
checkpoint-15000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:436010ee5b1eb29da4e773e08b26b0ddc8aeccbc10f62b877fa56c54a5d78e17
|
| 3 |
+
size 627
|
checkpoint-15000/trainer_state.json
ADDED
|
@@ -0,0 +1,406 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 7.853403141361256,
|
| 5 |
+
"global_step": 15000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.26,
|
| 12 |
+
"learning_rate": 6.25e-06,
|
| 13 |
+
"loss": 7.3722,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.26,
|
| 18 |
+
"eval_runtime": 4.9957,
|
| 19 |
+
"eval_samples_per_second": 20.017,
|
| 20 |
+
"eval_steps_per_second": 10.009,
|
| 21 |
+
"step": 500
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.52,
|
| 25 |
+
"learning_rate": 9.895833333333334e-06,
|
| 26 |
+
"loss": 1.5052,
|
| 27 |
+
"step": 1000
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.52,
|
| 31 |
+
"eval_runtime": 5.5484,
|
| 32 |
+
"eval_samples_per_second": 18.023,
|
| 33 |
+
"eval_steps_per_second": 9.012,
|
| 34 |
+
"step": 1000
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.79,
|
| 38 |
+
"learning_rate": 9.635416666666668e-06,
|
| 39 |
+
"loss": 1.3479,
|
| 40 |
+
"step": 1500
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.79,
|
| 44 |
+
"eval_runtime": 5.4316,
|
| 45 |
+
"eval_samples_per_second": 18.411,
|
| 46 |
+
"eval_steps_per_second": 9.205,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.05,
|
| 51 |
+
"learning_rate": 9.375000000000001e-06,
|
| 52 |
+
"loss": 1.3177,
|
| 53 |
+
"step": 2000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"eval_runtime": 5.4246,
|
| 58 |
+
"eval_samples_per_second": 18.435,
|
| 59 |
+
"eval_steps_per_second": 9.217,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.31,
|
| 64 |
+
"learning_rate": 9.114583333333334e-06,
|
| 65 |
+
"loss": 1.2962,
|
| 66 |
+
"step": 2500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.31,
|
| 70 |
+
"eval_runtime": 5.0933,
|
| 71 |
+
"eval_samples_per_second": 19.634,
|
| 72 |
+
"eval_steps_per_second": 9.817,
|
| 73 |
+
"step": 2500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.57,
|
| 77 |
+
"learning_rate": 8.854166666666667e-06,
|
| 78 |
+
"loss": 1.2824,
|
| 79 |
+
"step": 3000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.57,
|
| 83 |
+
"eval_runtime": 6.1894,
|
| 84 |
+
"eval_samples_per_second": 16.157,
|
| 85 |
+
"eval_steps_per_second": 8.078,
|
| 86 |
+
"step": 3000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.83,
|
| 90 |
+
"learning_rate": 8.59375e-06,
|
| 91 |
+
"loss": 1.2784,
|
| 92 |
+
"step": 3500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.83,
|
| 96 |
+
"eval_runtime": 4.7573,
|
| 97 |
+
"eval_samples_per_second": 21.02,
|
| 98 |
+
"eval_steps_per_second": 10.51,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.09,
|
| 103 |
+
"learning_rate": 8.333333333333334e-06,
|
| 104 |
+
"loss": 1.2655,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.09,
|
| 109 |
+
"eval_runtime": 5.5685,
|
| 110 |
+
"eval_samples_per_second": 17.958,
|
| 111 |
+
"eval_steps_per_second": 8.979,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.36,
|
| 116 |
+
"learning_rate": 8.072916666666667e-06,
|
| 117 |
+
"loss": 1.2553,
|
| 118 |
+
"step": 4500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.36,
|
| 122 |
+
"eval_runtime": 6.732,
|
| 123 |
+
"eval_samples_per_second": 14.854,
|
| 124 |
+
"eval_steps_per_second": 7.427,
|
| 125 |
+
"step": 4500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.62,
|
| 129 |
+
"learning_rate": 7.8125e-06,
|
| 130 |
+
"loss": 1.2627,
|
| 131 |
+
"step": 5000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.62,
|
| 135 |
+
"eval_runtime": 4.5523,
|
| 136 |
+
"eval_samples_per_second": 21.967,
|
| 137 |
+
"eval_steps_per_second": 10.983,
|
| 138 |
+
"step": 5000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.88,
|
| 142 |
+
"learning_rate": 7.552083333333334e-06,
|
| 143 |
+
"loss": 1.2358,
|
| 144 |
+
"step": 5500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.88,
|
| 148 |
+
"eval_runtime": 4.906,
|
| 149 |
+
"eval_samples_per_second": 20.383,
|
| 150 |
+
"eval_steps_per_second": 10.192,
|
| 151 |
+
"step": 5500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 3.14,
|
| 155 |
+
"learning_rate": 7.291666666666667e-06,
|
| 156 |
+
"loss": 1.2495,
|
| 157 |
+
"step": 6000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 3.14,
|
| 161 |
+
"eval_runtime": 5.2379,
|
| 162 |
+
"eval_samples_per_second": 19.092,
|
| 163 |
+
"eval_steps_per_second": 9.546,
|
| 164 |
+
"step": 6000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.4,
|
| 168 |
+
"learning_rate": 7.031250000000001e-06,
|
| 169 |
+
"loss": 1.234,
|
| 170 |
+
"step": 6500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.4,
|
| 174 |
+
"eval_runtime": 5.2459,
|
| 175 |
+
"eval_samples_per_second": 19.062,
|
| 176 |
+
"eval_steps_per_second": 9.531,
|
| 177 |
+
"step": 6500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.66,
|
| 181 |
+
"learning_rate": 6.770833333333334e-06,
|
| 182 |
+
"loss": 1.2228,
|
| 183 |
+
"step": 7000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.66,
|
| 187 |
+
"eval_runtime": 5.186,
|
| 188 |
+
"eval_samples_per_second": 19.283,
|
| 189 |
+
"eval_steps_per_second": 9.641,
|
| 190 |
+
"step": 7000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.93,
|
| 194 |
+
"learning_rate": 6.510416666666667e-06,
|
| 195 |
+
"loss": 1.2403,
|
| 196 |
+
"step": 7500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 3.93,
|
| 200 |
+
"eval_runtime": 5.9443,
|
| 201 |
+
"eval_samples_per_second": 16.823,
|
| 202 |
+
"eval_steps_per_second": 8.411,
|
| 203 |
+
"step": 7500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.19,
|
| 207 |
+
"learning_rate": 6.25e-06,
|
| 208 |
+
"loss": 1.2254,
|
| 209 |
+
"step": 8000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.19,
|
| 213 |
+
"eval_runtime": 6.1004,
|
| 214 |
+
"eval_samples_per_second": 16.392,
|
| 215 |
+
"eval_steps_per_second": 8.196,
|
| 216 |
+
"step": 8000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.45,
|
| 220 |
+
"learning_rate": 5.989583333333334e-06,
|
| 221 |
+
"loss": 1.2203,
|
| 222 |
+
"step": 8500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 4.45,
|
| 226 |
+
"eval_runtime": 4.7155,
|
| 227 |
+
"eval_samples_per_second": 21.207,
|
| 228 |
+
"eval_steps_per_second": 10.603,
|
| 229 |
+
"step": 8500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 4.71,
|
| 233 |
+
"learning_rate": 5.729166666666667e-06,
|
| 234 |
+
"loss": 1.2198,
|
| 235 |
+
"step": 9000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 4.71,
|
| 239 |
+
"eval_runtime": 5.2917,
|
| 240 |
+
"eval_samples_per_second": 18.897,
|
| 241 |
+
"eval_steps_per_second": 9.449,
|
| 242 |
+
"step": 9000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.97,
|
| 246 |
+
"learning_rate": 5.468750000000001e-06,
|
| 247 |
+
"loss": 1.2405,
|
| 248 |
+
"step": 9500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.97,
|
| 252 |
+
"eval_runtime": 5.6681,
|
| 253 |
+
"eval_samples_per_second": 17.643,
|
| 254 |
+
"eval_steps_per_second": 8.821,
|
| 255 |
+
"step": 9500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 5.24,
|
| 259 |
+
"learning_rate": 5.208333333333334e-06,
|
| 260 |
+
"loss": 1.2209,
|
| 261 |
+
"step": 10000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 5.24,
|
| 265 |
+
"eval_runtime": 4.7064,
|
| 266 |
+
"eval_samples_per_second": 21.247,
|
| 267 |
+
"eval_steps_per_second": 10.624,
|
| 268 |
+
"step": 10000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 5.5,
|
| 272 |
+
"learning_rate": 4.947916666666667e-06,
|
| 273 |
+
"loss": 1.232,
|
| 274 |
+
"step": 10500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 5.5,
|
| 278 |
+
"eval_runtime": 5.1725,
|
| 279 |
+
"eval_samples_per_second": 19.333,
|
| 280 |
+
"eval_steps_per_second": 9.667,
|
| 281 |
+
"step": 10500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 5.76,
|
| 285 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 286 |
+
"loss": 1.2078,
|
| 287 |
+
"step": 11000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 5.76,
|
| 291 |
+
"eval_runtime": 4.845,
|
| 292 |
+
"eval_samples_per_second": 20.64,
|
| 293 |
+
"eval_steps_per_second": 10.32,
|
| 294 |
+
"step": 11000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.02,
|
| 298 |
+
"learning_rate": 4.427083333333334e-06,
|
| 299 |
+
"loss": 1.2066,
|
| 300 |
+
"step": 11500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 6.02,
|
| 304 |
+
"eval_runtime": 5.6538,
|
| 305 |
+
"eval_samples_per_second": 17.687,
|
| 306 |
+
"eval_steps_per_second": 8.844,
|
| 307 |
+
"step": 11500
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 6.28,
|
| 311 |
+
"learning_rate": 4.166666666666667e-06,
|
| 312 |
+
"loss": 1.2265,
|
| 313 |
+
"step": 12000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 6.28,
|
| 317 |
+
"eval_runtime": 5.06,
|
| 318 |
+
"eval_samples_per_second": 19.763,
|
| 319 |
+
"eval_steps_per_second": 9.881,
|
| 320 |
+
"step": 12000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 6.54,
|
| 324 |
+
"learning_rate": 3.90625e-06,
|
| 325 |
+
"loss": 1.2113,
|
| 326 |
+
"step": 12500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 6.54,
|
| 330 |
+
"eval_runtime": 5.8625,
|
| 331 |
+
"eval_samples_per_second": 17.057,
|
| 332 |
+
"eval_steps_per_second": 8.529,
|
| 333 |
+
"step": 12500
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.81,
|
| 337 |
+
"learning_rate": 3.6458333333333333e-06,
|
| 338 |
+
"loss": 1.1924,
|
| 339 |
+
"step": 13000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 6.81,
|
| 343 |
+
"eval_runtime": 5.4916,
|
| 344 |
+
"eval_samples_per_second": 18.21,
|
| 345 |
+
"eval_steps_per_second": 9.105,
|
| 346 |
+
"step": 13000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 7.07,
|
| 350 |
+
"learning_rate": 3.385416666666667e-06,
|
| 351 |
+
"loss": 1.2023,
|
| 352 |
+
"step": 13500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 7.07,
|
| 356 |
+
"eval_runtime": 6.0325,
|
| 357 |
+
"eval_samples_per_second": 16.577,
|
| 358 |
+
"eval_steps_per_second": 8.288,
|
| 359 |
+
"step": 13500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 7.33,
|
| 363 |
+
"learning_rate": 3.125e-06,
|
| 364 |
+
"loss": 1.2005,
|
| 365 |
+
"step": 14000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 7.33,
|
| 369 |
+
"eval_runtime": 5.7214,
|
| 370 |
+
"eval_samples_per_second": 17.478,
|
| 371 |
+
"eval_steps_per_second": 8.739,
|
| 372 |
+
"step": 14000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 7.59,
|
| 376 |
+
"learning_rate": 2.8645833333333334e-06,
|
| 377 |
+
"loss": 1.2046,
|
| 378 |
+
"step": 14500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 7.59,
|
| 382 |
+
"eval_runtime": 5.7834,
|
| 383 |
+
"eval_samples_per_second": 17.291,
|
| 384 |
+
"eval_steps_per_second": 8.645,
|
| 385 |
+
"step": 14500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 7.85,
|
| 389 |
+
"learning_rate": 2.604166666666667e-06,
|
| 390 |
+
"loss": 1.2132,
|
| 391 |
+
"step": 15000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 7.85,
|
| 395 |
+
"eval_runtime": 6.0375,
|
| 396 |
+
"eval_samples_per_second": 16.563,
|
| 397 |
+
"eval_steps_per_second": 8.282,
|
| 398 |
+
"step": 15000
|
| 399 |
+
}
|
| 400 |
+
],
|
| 401 |
+
"max_steps": 20000,
|
| 402 |
+
"num_train_epochs": 11,
|
| 403 |
+
"total_flos": 1.2550929482376806e+17,
|
| 404 |
+
"trial_name": null,
|
| 405 |
+
"trial_params": null
|
| 406 |
+
}
|
checkpoint-15000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc04dcf4f9e174fdddb9f15c5a6a5d1a48675326eb949a1f635ccc394c1df49d
|
| 3 |
+
size 4027
|
checkpoint-20000/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DylanJHJ/fidt5-base-nq",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"FiDT5"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 3072,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "relu",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": false,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 12,
|
| 21 |
+
"num_heads": 12,
|
| 22 |
+
"num_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"relative_attention_max_distance": 128,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.31.0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
checkpoint-20000/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"
|
| 7 |
+
}
|
checkpoint-20000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6735c6881d9febec83e204b59018fa868ce2c53b23c63a0bd68ee264fe7025e6
|
| 3 |
+
size 2185605
|
checkpoint-20000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fe1eb1ac6b3902c1fe0318d175ddc27dc19b571f25f851753099731f444753f
|
| 3 |
+
size 920014805
|
checkpoint-20000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:02995c93c03dece1f975efe3310ca55415e3a996abf8c0f8f9623825dc94b0ba
|
| 3 |
+
size 14575
|
checkpoint-20000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4988e214c5cc967d210df55aa0e70a2b5b8f8378da4b9ef201c9550cd719cc73
|
| 3 |
+
size 627
|
checkpoint-20000/trainer_state.json
ADDED
|
@@ -0,0 +1,536 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.471204188481675,
|
| 5 |
+
"global_step": 20000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.26,
|
| 12 |
+
"learning_rate": 6.25e-06,
|
| 13 |
+
"loss": 7.3722,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.26,
|
| 18 |
+
"eval_runtime": 4.9957,
|
| 19 |
+
"eval_samples_per_second": 20.017,
|
| 20 |
+
"eval_steps_per_second": 10.009,
|
| 21 |
+
"step": 500
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.52,
|
| 25 |
+
"learning_rate": 9.895833333333334e-06,
|
| 26 |
+
"loss": 1.5052,
|
| 27 |
+
"step": 1000
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.52,
|
| 31 |
+
"eval_runtime": 5.5484,
|
| 32 |
+
"eval_samples_per_second": 18.023,
|
| 33 |
+
"eval_steps_per_second": 9.012,
|
| 34 |
+
"step": 1000
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.79,
|
| 38 |
+
"learning_rate": 9.635416666666668e-06,
|
| 39 |
+
"loss": 1.3479,
|
| 40 |
+
"step": 1500
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.79,
|
| 44 |
+
"eval_runtime": 5.4316,
|
| 45 |
+
"eval_samples_per_second": 18.411,
|
| 46 |
+
"eval_steps_per_second": 9.205,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.05,
|
| 51 |
+
"learning_rate": 9.375000000000001e-06,
|
| 52 |
+
"loss": 1.3177,
|
| 53 |
+
"step": 2000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"eval_runtime": 5.4246,
|
| 58 |
+
"eval_samples_per_second": 18.435,
|
| 59 |
+
"eval_steps_per_second": 9.217,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.31,
|
| 64 |
+
"learning_rate": 9.114583333333334e-06,
|
| 65 |
+
"loss": 1.2962,
|
| 66 |
+
"step": 2500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.31,
|
| 70 |
+
"eval_runtime": 5.0933,
|
| 71 |
+
"eval_samples_per_second": 19.634,
|
| 72 |
+
"eval_steps_per_second": 9.817,
|
| 73 |
+
"step": 2500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.57,
|
| 77 |
+
"learning_rate": 8.854166666666667e-06,
|
| 78 |
+
"loss": 1.2824,
|
| 79 |
+
"step": 3000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.57,
|
| 83 |
+
"eval_runtime": 6.1894,
|
| 84 |
+
"eval_samples_per_second": 16.157,
|
| 85 |
+
"eval_steps_per_second": 8.078,
|
| 86 |
+
"step": 3000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.83,
|
| 90 |
+
"learning_rate": 8.59375e-06,
|
| 91 |
+
"loss": 1.2784,
|
| 92 |
+
"step": 3500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.83,
|
| 96 |
+
"eval_runtime": 4.7573,
|
| 97 |
+
"eval_samples_per_second": 21.02,
|
| 98 |
+
"eval_steps_per_second": 10.51,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.09,
|
| 103 |
+
"learning_rate": 8.333333333333334e-06,
|
| 104 |
+
"loss": 1.2655,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.09,
|
| 109 |
+
"eval_runtime": 5.5685,
|
| 110 |
+
"eval_samples_per_second": 17.958,
|
| 111 |
+
"eval_steps_per_second": 8.979,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.36,
|
| 116 |
+
"learning_rate": 8.072916666666667e-06,
|
| 117 |
+
"loss": 1.2553,
|
| 118 |
+
"step": 4500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.36,
|
| 122 |
+
"eval_runtime": 6.732,
|
| 123 |
+
"eval_samples_per_second": 14.854,
|
| 124 |
+
"eval_steps_per_second": 7.427,
|
| 125 |
+
"step": 4500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.62,
|
| 129 |
+
"learning_rate": 7.8125e-06,
|
| 130 |
+
"loss": 1.2627,
|
| 131 |
+
"step": 5000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.62,
|
| 135 |
+
"eval_runtime": 4.5523,
|
| 136 |
+
"eval_samples_per_second": 21.967,
|
| 137 |
+
"eval_steps_per_second": 10.983,
|
| 138 |
+
"step": 5000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.88,
|
| 142 |
+
"learning_rate": 7.552083333333334e-06,
|
| 143 |
+
"loss": 1.2358,
|
| 144 |
+
"step": 5500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.88,
|
| 148 |
+
"eval_runtime": 4.906,
|
| 149 |
+
"eval_samples_per_second": 20.383,
|
| 150 |
+
"eval_steps_per_second": 10.192,
|
| 151 |
+
"step": 5500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 3.14,
|
| 155 |
+
"learning_rate": 7.291666666666667e-06,
|
| 156 |
+
"loss": 1.2495,
|
| 157 |
+
"step": 6000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 3.14,
|
| 161 |
+
"eval_runtime": 5.2379,
|
| 162 |
+
"eval_samples_per_second": 19.092,
|
| 163 |
+
"eval_steps_per_second": 9.546,
|
| 164 |
+
"step": 6000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.4,
|
| 168 |
+
"learning_rate": 7.031250000000001e-06,
|
| 169 |
+
"loss": 1.234,
|
| 170 |
+
"step": 6500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.4,
|
| 174 |
+
"eval_runtime": 5.2459,
|
| 175 |
+
"eval_samples_per_second": 19.062,
|
| 176 |
+
"eval_steps_per_second": 9.531,
|
| 177 |
+
"step": 6500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.66,
|
| 181 |
+
"learning_rate": 6.770833333333334e-06,
|
| 182 |
+
"loss": 1.2228,
|
| 183 |
+
"step": 7000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.66,
|
| 187 |
+
"eval_runtime": 5.186,
|
| 188 |
+
"eval_samples_per_second": 19.283,
|
| 189 |
+
"eval_steps_per_second": 9.641,
|
| 190 |
+
"step": 7000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.93,
|
| 194 |
+
"learning_rate": 6.510416666666667e-06,
|
| 195 |
+
"loss": 1.2403,
|
| 196 |
+
"step": 7500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 3.93,
|
| 200 |
+
"eval_runtime": 5.9443,
|
| 201 |
+
"eval_samples_per_second": 16.823,
|
| 202 |
+
"eval_steps_per_second": 8.411,
|
| 203 |
+
"step": 7500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.19,
|
| 207 |
+
"learning_rate": 6.25e-06,
|
| 208 |
+
"loss": 1.2254,
|
| 209 |
+
"step": 8000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.19,
|
| 213 |
+
"eval_runtime": 6.1004,
|
| 214 |
+
"eval_samples_per_second": 16.392,
|
| 215 |
+
"eval_steps_per_second": 8.196,
|
| 216 |
+
"step": 8000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.45,
|
| 220 |
+
"learning_rate": 5.989583333333334e-06,
|
| 221 |
+
"loss": 1.2203,
|
| 222 |
+
"step": 8500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 4.45,
|
| 226 |
+
"eval_runtime": 4.7155,
|
| 227 |
+
"eval_samples_per_second": 21.207,
|
| 228 |
+
"eval_steps_per_second": 10.603,
|
| 229 |
+
"step": 8500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 4.71,
|
| 233 |
+
"learning_rate": 5.729166666666667e-06,
|
| 234 |
+
"loss": 1.2198,
|
| 235 |
+
"step": 9000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 4.71,
|
| 239 |
+
"eval_runtime": 5.2917,
|
| 240 |
+
"eval_samples_per_second": 18.897,
|
| 241 |
+
"eval_steps_per_second": 9.449,
|
| 242 |
+
"step": 9000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.97,
|
| 246 |
+
"learning_rate": 5.468750000000001e-06,
|
| 247 |
+
"loss": 1.2405,
|
| 248 |
+
"step": 9500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.97,
|
| 252 |
+
"eval_runtime": 5.6681,
|
| 253 |
+
"eval_samples_per_second": 17.643,
|
| 254 |
+
"eval_steps_per_second": 8.821,
|
| 255 |
+
"step": 9500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 5.24,
|
| 259 |
+
"learning_rate": 5.208333333333334e-06,
|
| 260 |
+
"loss": 1.2209,
|
| 261 |
+
"step": 10000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 5.24,
|
| 265 |
+
"eval_runtime": 4.7064,
|
| 266 |
+
"eval_samples_per_second": 21.247,
|
| 267 |
+
"eval_steps_per_second": 10.624,
|
| 268 |
+
"step": 10000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 5.5,
|
| 272 |
+
"learning_rate": 4.947916666666667e-06,
|
| 273 |
+
"loss": 1.232,
|
| 274 |
+
"step": 10500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 5.5,
|
| 278 |
+
"eval_runtime": 5.1725,
|
| 279 |
+
"eval_samples_per_second": 19.333,
|
| 280 |
+
"eval_steps_per_second": 9.667,
|
| 281 |
+
"step": 10500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 5.76,
|
| 285 |
+
"learning_rate": 4.6875000000000004e-06,
|
| 286 |
+
"loss": 1.2078,
|
| 287 |
+
"step": 11000
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 5.76,
|
| 291 |
+
"eval_runtime": 4.845,
|
| 292 |
+
"eval_samples_per_second": 20.64,
|
| 293 |
+
"eval_steps_per_second": 10.32,
|
| 294 |
+
"step": 11000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 6.02,
|
| 298 |
+
"learning_rate": 4.427083333333334e-06,
|
| 299 |
+
"loss": 1.2066,
|
| 300 |
+
"step": 11500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 6.02,
|
| 304 |
+
"eval_runtime": 5.6538,
|
| 305 |
+
"eval_samples_per_second": 17.687,
|
| 306 |
+
"eval_steps_per_second": 8.844,
|
| 307 |
+
"step": 11500
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 6.28,
|
| 311 |
+
"learning_rate": 4.166666666666667e-06,
|
| 312 |
+
"loss": 1.2265,
|
| 313 |
+
"step": 12000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 6.28,
|
| 317 |
+
"eval_runtime": 5.06,
|
| 318 |
+
"eval_samples_per_second": 19.763,
|
| 319 |
+
"eval_steps_per_second": 9.881,
|
| 320 |
+
"step": 12000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 6.54,
|
| 324 |
+
"learning_rate": 3.90625e-06,
|
| 325 |
+
"loss": 1.2113,
|
| 326 |
+
"step": 12500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 6.54,
|
| 330 |
+
"eval_runtime": 5.8625,
|
| 331 |
+
"eval_samples_per_second": 17.057,
|
| 332 |
+
"eval_steps_per_second": 8.529,
|
| 333 |
+
"step": 12500
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.81,
|
| 337 |
+
"learning_rate": 3.6458333333333333e-06,
|
| 338 |
+
"loss": 1.1924,
|
| 339 |
+
"step": 13000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 6.81,
|
| 343 |
+
"eval_runtime": 5.4916,
|
| 344 |
+
"eval_samples_per_second": 18.21,
|
| 345 |
+
"eval_steps_per_second": 9.105,
|
| 346 |
+
"step": 13000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 7.07,
|
| 350 |
+
"learning_rate": 3.385416666666667e-06,
|
| 351 |
+
"loss": 1.2023,
|
| 352 |
+
"step": 13500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 7.07,
|
| 356 |
+
"eval_runtime": 6.0325,
|
| 357 |
+
"eval_samples_per_second": 16.577,
|
| 358 |
+
"eval_steps_per_second": 8.288,
|
| 359 |
+
"step": 13500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 7.33,
|
| 363 |
+
"learning_rate": 3.125e-06,
|
| 364 |
+
"loss": 1.2005,
|
| 365 |
+
"step": 14000
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 7.33,
|
| 369 |
+
"eval_runtime": 5.7214,
|
| 370 |
+
"eval_samples_per_second": 17.478,
|
| 371 |
+
"eval_steps_per_second": 8.739,
|
| 372 |
+
"step": 14000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 7.59,
|
| 376 |
+
"learning_rate": 2.8645833333333334e-06,
|
| 377 |
+
"loss": 1.2046,
|
| 378 |
+
"step": 14500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 7.59,
|
| 382 |
+
"eval_runtime": 5.7834,
|
| 383 |
+
"eval_samples_per_second": 17.291,
|
| 384 |
+
"eval_steps_per_second": 8.645,
|
| 385 |
+
"step": 14500
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 7.85,
|
| 389 |
+
"learning_rate": 2.604166666666667e-06,
|
| 390 |
+
"loss": 1.2132,
|
| 391 |
+
"step": 15000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 7.85,
|
| 395 |
+
"eval_runtime": 6.0375,
|
| 396 |
+
"eval_samples_per_second": 16.563,
|
| 397 |
+
"eval_steps_per_second": 8.282,
|
| 398 |
+
"step": 15000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 8.12,
|
| 402 |
+
"learning_rate": 2.3437500000000002e-06,
|
| 403 |
+
"loss": 1.2044,
|
| 404 |
+
"step": 15500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 8.12,
|
| 408 |
+
"eval_runtime": 5.1804,
|
| 409 |
+
"eval_samples_per_second": 19.304,
|
| 410 |
+
"eval_steps_per_second": 9.652,
|
| 411 |
+
"step": 15500
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 8.38,
|
| 415 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 416 |
+
"loss": 1.1986,
|
| 417 |
+
"step": 16000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 8.38,
|
| 421 |
+
"eval_runtime": 7.2562,
|
| 422 |
+
"eval_samples_per_second": 13.781,
|
| 423 |
+
"eval_steps_per_second": 6.891,
|
| 424 |
+
"step": 16000
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 8.64,
|
| 428 |
+
"learning_rate": 1.8229166666666666e-06,
|
| 429 |
+
"loss": 1.2057,
|
| 430 |
+
"step": 16500
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 8.64,
|
| 434 |
+
"eval_runtime": 4.8338,
|
| 435 |
+
"eval_samples_per_second": 20.688,
|
| 436 |
+
"eval_steps_per_second": 10.344,
|
| 437 |
+
"step": 16500
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 8.9,
|
| 441 |
+
"learning_rate": 1.5625e-06,
|
| 442 |
+
"loss": 1.1947,
|
| 443 |
+
"step": 17000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 8.9,
|
| 447 |
+
"eval_runtime": 5.2971,
|
| 448 |
+
"eval_samples_per_second": 18.878,
|
| 449 |
+
"eval_steps_per_second": 9.439,
|
| 450 |
+
"step": 17000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 9.16,
|
| 454 |
+
"learning_rate": 1.3020833333333335e-06,
|
| 455 |
+
"loss": 1.215,
|
| 456 |
+
"step": 17500
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 9.16,
|
| 460 |
+
"eval_runtime": 5.1412,
|
| 461 |
+
"eval_samples_per_second": 19.451,
|
| 462 |
+
"eval_steps_per_second": 9.725,
|
| 463 |
+
"step": 17500
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 9.42,
|
| 467 |
+
"learning_rate": 1.0416666666666667e-06,
|
| 468 |
+
"loss": 1.191,
|
| 469 |
+
"step": 18000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 9.42,
|
| 473 |
+
"eval_runtime": 4.8814,
|
| 474 |
+
"eval_samples_per_second": 20.486,
|
| 475 |
+
"eval_steps_per_second": 10.243,
|
| 476 |
+
"step": 18000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 9.69,
|
| 480 |
+
"learning_rate": 7.8125e-07,
|
| 481 |
+
"loss": 1.2013,
|
| 482 |
+
"step": 18500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 9.69,
|
| 486 |
+
"eval_runtime": 5.6388,
|
| 487 |
+
"eval_samples_per_second": 17.734,
|
| 488 |
+
"eval_steps_per_second": 8.867,
|
| 489 |
+
"step": 18500
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 9.95,
|
| 493 |
+
"learning_rate": 5.208333333333334e-07,
|
| 494 |
+
"loss": 1.1838,
|
| 495 |
+
"step": 19000
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 9.95,
|
| 499 |
+
"eval_runtime": 8.057,
|
| 500 |
+
"eval_samples_per_second": 12.412,
|
| 501 |
+
"eval_steps_per_second": 6.206,
|
| 502 |
+
"step": 19000
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 10.21,
|
| 506 |
+
"learning_rate": 2.604166666666667e-07,
|
| 507 |
+
"loss": 1.2097,
|
| 508 |
+
"step": 19500
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 10.21,
|
| 512 |
+
"eval_runtime": 8.8555,
|
| 513 |
+
"eval_samples_per_second": 11.292,
|
| 514 |
+
"eval_steps_per_second": 5.646,
|
| 515 |
+
"step": 19500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 10.47,
|
| 519 |
+
"learning_rate": 0.0,
|
| 520 |
+
"loss": 1.1946,
|
| 521 |
+
"step": 20000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 10.47,
|
| 525 |
+
"eval_runtime": 7.1071,
|
| 526 |
+
"eval_samples_per_second": 14.07,
|
| 527 |
+
"eval_steps_per_second": 7.035,
|
| 528 |
+
"step": 20000
|
| 529 |
+
}
|
| 530 |
+
],
|
| 531 |
+
"max_steps": 20000,
|
| 532 |
+
"num_train_epochs": 11,
|
| 533 |
+
"total_flos": 1.673264670519214e+17,
|
| 534 |
+
"trial_name": null,
|
| 535 |
+
"trial_params": null
|
| 536 |
+
}
|
checkpoint-20000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc04dcf4f9e174fdddb9f15c5a6a5d1a48675326eb949a1f635ccc394c1df49d
|
| 3 |
+
size 4027
|
checkpoint-5000/config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DylanJHJ/fidt5-base-nq",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"FiDT5"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 3072,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "relu",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": false,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 12,
|
| 21 |
+
"num_heads": 12,
|
| 22 |
+
"num_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"relative_attention_max_distance": 128,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.31.0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
checkpoint-5000/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"
|
| 7 |
+
}
|
checkpoint-5000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:47256390eb282777a19b92e9dbfa4181ee99b170cde9ccf6185b56c137601224
|
| 3 |
+
size 2185605
|
checkpoint-5000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d5cb4c0d4a798ebe2cde775fe749d3ce17e4a878136759d6c2617155e3537fe1
|
| 3 |
+
size 920014805
|
checkpoint-5000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d53a2f4681622aae25dc22553522f990452a18d4cef7c5c519874f42d8aa5e4
|
| 3 |
+
size 14575
|
checkpoint-5000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d5fd8140d414539cb6314c208198cdf00a9bfbf24f3c6465f7f9fa454bb6f9c
|
| 3 |
+
size 627
|
checkpoint-5000/trainer_state.json
ADDED
|
@@ -0,0 +1,146 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.6178010471204187,
|
| 5 |
+
"global_step": 5000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.26,
|
| 12 |
+
"learning_rate": 6.25e-06,
|
| 13 |
+
"loss": 7.3722,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.26,
|
| 18 |
+
"eval_runtime": 4.9957,
|
| 19 |
+
"eval_samples_per_second": 20.017,
|
| 20 |
+
"eval_steps_per_second": 10.009,
|
| 21 |
+
"step": 500
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.52,
|
| 25 |
+
"learning_rate": 9.895833333333334e-06,
|
| 26 |
+
"loss": 1.5052,
|
| 27 |
+
"step": 1000
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.52,
|
| 31 |
+
"eval_runtime": 5.5484,
|
| 32 |
+
"eval_samples_per_second": 18.023,
|
| 33 |
+
"eval_steps_per_second": 9.012,
|
| 34 |
+
"step": 1000
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.79,
|
| 38 |
+
"learning_rate": 9.635416666666668e-06,
|
| 39 |
+
"loss": 1.3479,
|
| 40 |
+
"step": 1500
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.79,
|
| 44 |
+
"eval_runtime": 5.4316,
|
| 45 |
+
"eval_samples_per_second": 18.411,
|
| 46 |
+
"eval_steps_per_second": 9.205,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.05,
|
| 51 |
+
"learning_rate": 9.375000000000001e-06,
|
| 52 |
+
"loss": 1.3177,
|
| 53 |
+
"step": 2000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"eval_runtime": 5.4246,
|
| 58 |
+
"eval_samples_per_second": 18.435,
|
| 59 |
+
"eval_steps_per_second": 9.217,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.31,
|
| 64 |
+
"learning_rate": 9.114583333333334e-06,
|
| 65 |
+
"loss": 1.2962,
|
| 66 |
+
"step": 2500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.31,
|
| 70 |
+
"eval_runtime": 5.0933,
|
| 71 |
+
"eval_samples_per_second": 19.634,
|
| 72 |
+
"eval_steps_per_second": 9.817,
|
| 73 |
+
"step": 2500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.57,
|
| 77 |
+
"learning_rate": 8.854166666666667e-06,
|
| 78 |
+
"loss": 1.2824,
|
| 79 |
+
"step": 3000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.57,
|
| 83 |
+
"eval_runtime": 6.1894,
|
| 84 |
+
"eval_samples_per_second": 16.157,
|
| 85 |
+
"eval_steps_per_second": 8.078,
|
| 86 |
+
"step": 3000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.83,
|
| 90 |
+
"learning_rate": 8.59375e-06,
|
| 91 |
+
"loss": 1.2784,
|
| 92 |
+
"step": 3500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.83,
|
| 96 |
+
"eval_runtime": 4.7573,
|
| 97 |
+
"eval_samples_per_second": 21.02,
|
| 98 |
+
"eval_steps_per_second": 10.51,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.09,
|
| 103 |
+
"learning_rate": 8.333333333333334e-06,
|
| 104 |
+
"loss": 1.2655,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.09,
|
| 109 |
+
"eval_runtime": 5.5685,
|
| 110 |
+
"eval_samples_per_second": 17.958,
|
| 111 |
+
"eval_steps_per_second": 8.979,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.36,
|
| 116 |
+
"learning_rate": 8.072916666666667e-06,
|
| 117 |
+
"loss": 1.2553,
|
| 118 |
+
"step": 4500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.36,
|
| 122 |
+
"eval_runtime": 6.732,
|
| 123 |
+
"eval_samples_per_second": 14.854,
|
| 124 |
+
"eval_steps_per_second": 7.427,
|
| 125 |
+
"step": 4500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.62,
|
| 129 |
+
"learning_rate": 7.8125e-06,
|
| 130 |
+
"loss": 1.2627,
|
| 131 |
+
"step": 5000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.62,
|
| 135 |
+
"eval_runtime": 4.5523,
|
| 136 |
+
"eval_samples_per_second": 21.967,
|
| 137 |
+
"eval_steps_per_second": 10.983,
|
| 138 |
+
"step": 5000
|
| 139 |
+
}
|
| 140 |
+
],
|
| 141 |
+
"max_steps": 20000,
|
| 142 |
+
"num_train_epochs": 11,
|
| 143 |
+
"total_flos": 4.184046875234304e+16,
|
| 144 |
+
"trial_name": null,
|
| 145 |
+
"trial_params": null
|
| 146 |
+
}
|
checkpoint-5000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc04dcf4f9e174fdddb9f15c5a6a5d1a48675326eb949a1f635ccc394c1df49d
|
| 3 |
+
size 4027
|
config.json
ADDED
|
@@ -0,0 +1,60 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "DylanJHJ/fidt5-base-nq",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"FiDT5"
|
| 5 |
+
],
|
| 6 |
+
"d_ff": 3072,
|
| 7 |
+
"d_kv": 64,
|
| 8 |
+
"d_model": 768,
|
| 9 |
+
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "relu",
|
| 11 |
+
"dropout_rate": 0.1,
|
| 12 |
+
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "relu",
|
| 14 |
+
"initializer_factor": 1.0,
|
| 15 |
+
"is_encoder_decoder": true,
|
| 16 |
+
"is_gated_act": false,
|
| 17 |
+
"layer_norm_epsilon": 1e-06,
|
| 18 |
+
"model_type": "t5",
|
| 19 |
+
"n_positions": 512,
|
| 20 |
+
"num_decoder_layers": 12,
|
| 21 |
+
"num_heads": 12,
|
| 22 |
+
"num_layers": 12,
|
| 23 |
+
"output_past": true,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"relative_attention_max_distance": 128,
|
| 26 |
+
"relative_attention_num_buckets": 32,
|
| 27 |
+
"task_specific_params": {
|
| 28 |
+
"summarization": {
|
| 29 |
+
"early_stopping": true,
|
| 30 |
+
"length_penalty": 2.0,
|
| 31 |
+
"max_length": 200,
|
| 32 |
+
"min_length": 30,
|
| 33 |
+
"no_repeat_ngram_size": 3,
|
| 34 |
+
"num_beams": 4,
|
| 35 |
+
"prefix": "summarize: "
|
| 36 |
+
},
|
| 37 |
+
"translation_en_to_de": {
|
| 38 |
+
"early_stopping": true,
|
| 39 |
+
"max_length": 300,
|
| 40 |
+
"num_beams": 4,
|
| 41 |
+
"prefix": "translate English to German: "
|
| 42 |
+
},
|
| 43 |
+
"translation_en_to_fr": {
|
| 44 |
+
"early_stopping": true,
|
| 45 |
+
"max_length": 300,
|
| 46 |
+
"num_beams": 4,
|
| 47 |
+
"prefix": "translate English to French: "
|
| 48 |
+
},
|
| 49 |
+
"translation_en_to_ro": {
|
| 50 |
+
"early_stopping": true,
|
| 51 |
+
"max_length": 300,
|
| 52 |
+
"num_beams": 4,
|
| 53 |
+
"prefix": "translate English to Romanian: "
|
| 54 |
+
}
|
| 55 |
+
},
|
| 56 |
+
"torch_dtype": "float32",
|
| 57 |
+
"transformers_version": "4.31.0",
|
| 58 |
+
"use_cache": true,
|
| 59 |
+
"vocab_size": 32128
|
| 60 |
+
}
|
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"
|
| 7 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c7227b58627e2ec7c2b03f84198efb31d97df1f089b013769ee54e3e128ff1f2
|
| 3 |
+
size 2185605
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f649cf96d88f5b744d984a87b6dd3f16c60f4f793edaec4685cdf62dda3fe6b0
|
| 3 |
+
size 920014805
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d263b9835f87e7764463235361192c1e14372603a00d3f5f6edba42d93057da
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:22b05183e4bc4eb4b1775c568247f66ae6e1c1616d7ca41e2cea3e991bebc2e1
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,276 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.2356020942408374,
|
| 5 |
+
"global_step": 10000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.26,
|
| 12 |
+
"learning_rate": 6.25e-06,
|
| 13 |
+
"loss": 7.3722,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.26,
|
| 18 |
+
"eval_runtime": 4.9957,
|
| 19 |
+
"eval_samples_per_second": 20.017,
|
| 20 |
+
"eval_steps_per_second": 10.009,
|
| 21 |
+
"step": 500
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.52,
|
| 25 |
+
"learning_rate": 9.895833333333334e-06,
|
| 26 |
+
"loss": 1.5052,
|
| 27 |
+
"step": 1000
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.52,
|
| 31 |
+
"eval_runtime": 5.5484,
|
| 32 |
+
"eval_samples_per_second": 18.023,
|
| 33 |
+
"eval_steps_per_second": 9.012,
|
| 34 |
+
"step": 1000
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.79,
|
| 38 |
+
"learning_rate": 9.635416666666668e-06,
|
| 39 |
+
"loss": 1.3479,
|
| 40 |
+
"step": 1500
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 0.79,
|
| 44 |
+
"eval_runtime": 5.4316,
|
| 45 |
+
"eval_samples_per_second": 18.411,
|
| 46 |
+
"eval_steps_per_second": 9.205,
|
| 47 |
+
"step": 1500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.05,
|
| 51 |
+
"learning_rate": 9.375000000000001e-06,
|
| 52 |
+
"loss": 1.3177,
|
| 53 |
+
"step": 2000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"eval_runtime": 5.4246,
|
| 58 |
+
"eval_samples_per_second": 18.435,
|
| 59 |
+
"eval_steps_per_second": 9.217,
|
| 60 |
+
"step": 2000
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.31,
|
| 64 |
+
"learning_rate": 9.114583333333334e-06,
|
| 65 |
+
"loss": 1.2962,
|
| 66 |
+
"step": 2500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.31,
|
| 70 |
+
"eval_runtime": 5.0933,
|
| 71 |
+
"eval_samples_per_second": 19.634,
|
| 72 |
+
"eval_steps_per_second": 9.817,
|
| 73 |
+
"step": 2500
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.57,
|
| 77 |
+
"learning_rate": 8.854166666666667e-06,
|
| 78 |
+
"loss": 1.2824,
|
| 79 |
+
"step": 3000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 1.57,
|
| 83 |
+
"eval_runtime": 6.1894,
|
| 84 |
+
"eval_samples_per_second": 16.157,
|
| 85 |
+
"eval_steps_per_second": 8.078,
|
| 86 |
+
"step": 3000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.83,
|
| 90 |
+
"learning_rate": 8.59375e-06,
|
| 91 |
+
"loss": 1.2784,
|
| 92 |
+
"step": 3500
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.83,
|
| 96 |
+
"eval_runtime": 4.7573,
|
| 97 |
+
"eval_samples_per_second": 21.02,
|
| 98 |
+
"eval_steps_per_second": 10.51,
|
| 99 |
+
"step": 3500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.09,
|
| 103 |
+
"learning_rate": 8.333333333333334e-06,
|
| 104 |
+
"loss": 1.2655,
|
| 105 |
+
"step": 4000
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.09,
|
| 109 |
+
"eval_runtime": 5.5685,
|
| 110 |
+
"eval_samples_per_second": 17.958,
|
| 111 |
+
"eval_steps_per_second": 8.979,
|
| 112 |
+
"step": 4000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 2.36,
|
| 116 |
+
"learning_rate": 8.072916666666667e-06,
|
| 117 |
+
"loss": 1.2553,
|
| 118 |
+
"step": 4500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 2.36,
|
| 122 |
+
"eval_runtime": 6.732,
|
| 123 |
+
"eval_samples_per_second": 14.854,
|
| 124 |
+
"eval_steps_per_second": 7.427,
|
| 125 |
+
"step": 4500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.62,
|
| 129 |
+
"learning_rate": 7.8125e-06,
|
| 130 |
+
"loss": 1.2627,
|
| 131 |
+
"step": 5000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.62,
|
| 135 |
+
"eval_runtime": 4.5523,
|
| 136 |
+
"eval_samples_per_second": 21.967,
|
| 137 |
+
"eval_steps_per_second": 10.983,
|
| 138 |
+
"step": 5000
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.88,
|
| 142 |
+
"learning_rate": 7.552083333333334e-06,
|
| 143 |
+
"loss": 1.2358,
|
| 144 |
+
"step": 5500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.88,
|
| 148 |
+
"eval_runtime": 4.906,
|
| 149 |
+
"eval_samples_per_second": 20.383,
|
| 150 |
+
"eval_steps_per_second": 10.192,
|
| 151 |
+
"step": 5500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 3.14,
|
| 155 |
+
"learning_rate": 7.291666666666667e-06,
|
| 156 |
+
"loss": 1.2495,
|
| 157 |
+
"step": 6000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 3.14,
|
| 161 |
+
"eval_runtime": 5.2379,
|
| 162 |
+
"eval_samples_per_second": 19.092,
|
| 163 |
+
"eval_steps_per_second": 9.546,
|
| 164 |
+
"step": 6000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.4,
|
| 168 |
+
"learning_rate": 7.031250000000001e-06,
|
| 169 |
+
"loss": 1.234,
|
| 170 |
+
"step": 6500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.4,
|
| 174 |
+
"eval_runtime": 5.2459,
|
| 175 |
+
"eval_samples_per_second": 19.062,
|
| 176 |
+
"eval_steps_per_second": 9.531,
|
| 177 |
+
"step": 6500
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 3.66,
|
| 181 |
+
"learning_rate": 6.770833333333334e-06,
|
| 182 |
+
"loss": 1.2228,
|
| 183 |
+
"step": 7000
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 3.66,
|
| 187 |
+
"eval_runtime": 5.186,
|
| 188 |
+
"eval_samples_per_second": 19.283,
|
| 189 |
+
"eval_steps_per_second": 9.641,
|
| 190 |
+
"step": 7000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.93,
|
| 194 |
+
"learning_rate": 6.510416666666667e-06,
|
| 195 |
+
"loss": 1.2403,
|
| 196 |
+
"step": 7500
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 3.93,
|
| 200 |
+
"eval_runtime": 5.9443,
|
| 201 |
+
"eval_samples_per_second": 16.823,
|
| 202 |
+
"eval_steps_per_second": 8.411,
|
| 203 |
+
"step": 7500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 4.19,
|
| 207 |
+
"learning_rate": 6.25e-06,
|
| 208 |
+
"loss": 1.2254,
|
| 209 |
+
"step": 8000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 4.19,
|
| 213 |
+
"eval_runtime": 6.1004,
|
| 214 |
+
"eval_samples_per_second": 16.392,
|
| 215 |
+
"eval_steps_per_second": 8.196,
|
| 216 |
+
"step": 8000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 4.45,
|
| 220 |
+
"learning_rate": 5.989583333333334e-06,
|
| 221 |
+
"loss": 1.2203,
|
| 222 |
+
"step": 8500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 4.45,
|
| 226 |
+
"eval_runtime": 4.7155,
|
| 227 |
+
"eval_samples_per_second": 21.207,
|
| 228 |
+
"eval_steps_per_second": 10.603,
|
| 229 |
+
"step": 8500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 4.71,
|
| 233 |
+
"learning_rate": 5.729166666666667e-06,
|
| 234 |
+
"loss": 1.2198,
|
| 235 |
+
"step": 9000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 4.71,
|
| 239 |
+
"eval_runtime": 5.2917,
|
| 240 |
+
"eval_samples_per_second": 18.897,
|
| 241 |
+
"eval_steps_per_second": 9.449,
|
| 242 |
+
"step": 9000
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.97,
|
| 246 |
+
"learning_rate": 5.468750000000001e-06,
|
| 247 |
+
"loss": 1.2405,
|
| 248 |
+
"step": 9500
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 4.97,
|
| 252 |
+
"eval_runtime": 5.6681,
|
| 253 |
+
"eval_samples_per_second": 17.643,
|
| 254 |
+
"eval_steps_per_second": 8.821,
|
| 255 |
+
"step": 9500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 5.24,
|
| 259 |
+
"learning_rate": 5.208333333333334e-06,
|
| 260 |
+
"loss": 1.2209,
|
| 261 |
+
"step": 10000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 5.24,
|
| 265 |
+
"eval_runtime": 4.7064,
|
| 266 |
+
"eval_samples_per_second": 21.247,
|
| 267 |
+
"eval_steps_per_second": 10.624,
|
| 268 |
+
"step": 10000
|
| 269 |
+
}
|
| 270 |
+
],
|
| 271 |
+
"max_steps": 20000,
|
| 272 |
+
"num_train_epochs": 11,
|
| 273 |
+
"total_flos": 8.36711025084334e+16,
|
| 274 |
+
"trial_name": null,
|
| 275 |
+
"trial_params": null
|
| 276 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fc04dcf4f9e174fdddb9f15c5a6a5d1a48675326eb949a1f635ccc394c1df49d
|
| 3 |
+
size 4027
|