Upload folder using huggingface_hub
Browse files- config.json +28 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +561 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "FacebookAI/xlm-roberta-large",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"XLMRobertaForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"bos_token_id": 0,
|
| 8 |
+
"classifier_dropout": null,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-05,
|
| 16 |
+
"max_position_embeddings": 514,
|
| 17 |
+
"model_type": "xlm-roberta",
|
| 18 |
+
"num_attention_heads": 16,
|
| 19 |
+
"num_hidden_layers": 24,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"position_embedding_type": "absolute",
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.48.0",
|
| 25 |
+
"type_vocab_size": 1,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 250002
|
| 28 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0353fad2ef29b49f1308e7a9a870616fa310ddee171204b465189d94848aa6f8
|
| 3 |
+
size 2239618672
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ed9f4bff3d0768c91f785e05335294b516bab85a8890ed272d375074d5bee6a8
|
| 3 |
+
size 4479472785
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a50e9cb5ffe6b61fcaacf8a8b8825c880a87fdde4b3b810a440b76998c770c95
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fe56ca14f7fb09ab9f2892706755ed252c9eb9cedd4a2e20dc6c5662b347f21
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,561 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9842267397484589,
|
| 3 |
+
"best_model_checkpoint": "/home/alexanders/Workspaces/malexandersalazar/xlm-roberta-large-cls-toxicity/data/models/xlm-roberta-large-cls-toxicity/config_5/checkpoint-28610",
|
| 4 |
+
"epoch": 9.999737922599808,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 28610,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1747182667947934,
|
| 13 |
+
"grad_norm": 12.17603588104248,
|
| 14 |
+
"learning_rate": 6.934153502027121e-07,
|
| 15 |
+
"loss": 0.7119,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.3494365335895868,
|
| 20 |
+
"grad_norm": 13.519432067871094,
|
| 21 |
+
"learning_rate": 1.3910247448622957e-06,
|
| 22 |
+
"loss": 0.3407,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.5241548003843802,
|
| 27 |
+
"grad_norm": 11.993674278259277,
|
| 28 |
+
"learning_rate": 2.090032154340836e-06,
|
| 29 |
+
"loss": 0.1725,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.6988730671791736,
|
| 34 |
+
"grad_norm": 13.306364059448242,
|
| 35 |
+
"learning_rate": 2.789039563819377e-06,
|
| 36 |
+
"loss": 0.1498,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.8735913339739669,
|
| 41 |
+
"grad_norm": 13.724279403686523,
|
| 42 |
+
"learning_rate": 3.488046973297917e-06,
|
| 43 |
+
"loss": 0.1368,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.9997379225998078,
|
| 48 |
+
"eval_accuracy": 0.9578529000698812,
|
| 49 |
+
"eval_f1": 0.9580195344026164,
|
| 50 |
+
"eval_loss": 0.14260601997375488,
|
| 51 |
+
"eval_precision": 0.9585943807196626,
|
| 52 |
+
"eval_recall": 0.9578529000698812,
|
| 53 |
+
"eval_runtime": 68.6764,
|
| 54 |
+
"eval_samples_per_second": 333.389,
|
| 55 |
+
"eval_steps_per_second": 20.837,
|
| 56 |
+
"step": 2861
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.0485716781689525,
|
| 60 |
+
"grad_norm": 15.129600524902344,
|
| 61 |
+
"learning_rate": 4.187054382776457e-06,
|
| 62 |
+
"loss": 0.1225,
|
| 63 |
+
"step": 3000
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.223289944963746,
|
| 67 |
+
"grad_norm": 18.516948699951172,
|
| 68 |
+
"learning_rate": 4.884663777436041e-06,
|
| 69 |
+
"loss": 0.111,
|
| 70 |
+
"step": 3500
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 1.3980082117585393,
|
| 74 |
+
"grad_norm": 3.170327663421631,
|
| 75 |
+
"learning_rate": 5.583671186914582e-06,
|
| 76 |
+
"loss": 0.109,
|
| 77 |
+
"step": 4000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.5727264785533328,
|
| 81 |
+
"grad_norm": 9.642043113708496,
|
| 82 |
+
"learning_rate": 6.281280581574165e-06,
|
| 83 |
+
"loss": 0.1045,
|
| 84 |
+
"step": 4500
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.747444745348126,
|
| 88 |
+
"grad_norm": 12.047298431396484,
|
| 89 |
+
"learning_rate": 6.980287991052706e-06,
|
| 90 |
+
"loss": 0.0981,
|
| 91 |
+
"step": 5000
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 1.9221630121429194,
|
| 95 |
+
"grad_norm": 11.591704368591309,
|
| 96 |
+
"learning_rate": 7.679295400531246e-06,
|
| 97 |
+
"loss": 0.0983,
|
| 98 |
+
"step": 5500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.9997379225998078,
|
| 102 |
+
"eval_accuracy": 0.9699074074074074,
|
| 103 |
+
"eval_f1": 0.9698689694737291,
|
| 104 |
+
"eval_loss": 0.09155124425888062,
|
| 105 |
+
"eval_precision": 0.9698681669937823,
|
| 106 |
+
"eval_recall": 0.9699074074074074,
|
| 107 |
+
"eval_runtime": 68.6299,
|
| 108 |
+
"eval_samples_per_second": 333.616,
|
| 109 |
+
"eval_steps_per_second": 20.851,
|
| 110 |
+
"step": 5722
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.097143356337905,
|
| 114 |
+
"grad_norm": 7.699872970581055,
|
| 115 |
+
"learning_rate": 8.378302810009785e-06,
|
| 116 |
+
"loss": 0.0818,
|
| 117 |
+
"step": 6000
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 2.2718616231326987,
|
| 121 |
+
"grad_norm": 2.9679672718048096,
|
| 122 |
+
"learning_rate": 9.077310219488328e-06,
|
| 123 |
+
"loss": 0.078,
|
| 124 |
+
"step": 6500
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 2.446579889927492,
|
| 128 |
+
"grad_norm": 5.794170379638672,
|
| 129 |
+
"learning_rate": 9.776317628966868e-06,
|
| 130 |
+
"loss": 0.0769,
|
| 131 |
+
"step": 7000
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 2.6212981567222853,
|
| 135 |
+
"grad_norm": 7.152242183685303,
|
| 136 |
+
"learning_rate": 9.842009600596542e-06,
|
| 137 |
+
"loss": 0.0719,
|
| 138 |
+
"step": 7500
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.7960164235170786,
|
| 142 |
+
"grad_norm": 3.8340728282928467,
|
| 143 |
+
"learning_rate": 9.608985412685837e-06,
|
| 144 |
+
"loss": 0.0708,
|
| 145 |
+
"step": 8000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 2.970734690311872,
|
| 149 |
+
"grad_norm": 13.128349304199219,
|
| 150 |
+
"learning_rate": 9.375961224775132e-06,
|
| 151 |
+
"loss": 0.0748,
|
| 152 |
+
"step": 8500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.9997379225998078,
|
| 156 |
+
"eval_accuracy": 0.9731394129979035,
|
| 157 |
+
"eval_f1": 0.9731494346588756,
|
| 158 |
+
"eval_loss": 0.09937231242656708,
|
| 159 |
+
"eval_precision": 0.9731628120366744,
|
| 160 |
+
"eval_recall": 0.9731394129979035,
|
| 161 |
+
"eval_runtime": 68.6513,
|
| 162 |
+
"eval_samples_per_second": 333.512,
|
| 163 |
+
"eval_steps_per_second": 20.844,
|
| 164 |
+
"step": 8583
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.145715034506858,
|
| 168 |
+
"grad_norm": 20.52392578125,
|
| 169 |
+
"learning_rate": 9.142937036864427e-06,
|
| 170 |
+
"loss": 0.0504,
|
| 171 |
+
"step": 9000
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 3.320433301301651,
|
| 175 |
+
"grad_norm": 33.67174530029297,
|
| 176 |
+
"learning_rate": 8.909912848953722e-06,
|
| 177 |
+
"loss": 0.049,
|
| 178 |
+
"step": 9500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 3.4951515680964444,
|
| 182 |
+
"grad_norm": 0.8490325212478638,
|
| 183 |
+
"learning_rate": 8.676888661043017e-06,
|
| 184 |
+
"loss": 0.0464,
|
| 185 |
+
"step": 10000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 3.6698698348912377,
|
| 189 |
+
"grad_norm": 13.017010688781738,
|
| 190 |
+
"learning_rate": 8.443864473132312e-06,
|
| 191 |
+
"loss": 0.0463,
|
| 192 |
+
"step": 10500
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 3.844588101686031,
|
| 196 |
+
"grad_norm": 0.6549157500267029,
|
| 197 |
+
"learning_rate": 8.210840285221606e-06,
|
| 198 |
+
"loss": 0.049,
|
| 199 |
+
"step": 11000
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 3.9997379225998078,
|
| 203 |
+
"eval_accuracy": 0.9785552061495457,
|
| 204 |
+
"eval_f1": 0.9785386610066998,
|
| 205 |
+
"eval_loss": 0.0886269360780716,
|
| 206 |
+
"eval_precision": 0.9785361579407956,
|
| 207 |
+
"eval_recall": 0.9785552061495457,
|
| 208 |
+
"eval_runtime": 68.6459,
|
| 209 |
+
"eval_samples_per_second": 333.538,
|
| 210 |
+
"eval_steps_per_second": 20.846,
|
| 211 |
+
"step": 11444
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 4.019568445881017,
|
| 215 |
+
"grad_norm": 40.50947952270508,
|
| 216 |
+
"learning_rate": 7.977816097310901e-06,
|
| 217 |
+
"loss": 0.0454,
|
| 218 |
+
"step": 11500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.19428671267581,
|
| 222 |
+
"grad_norm": 1.1528542041778564,
|
| 223 |
+
"learning_rate": 7.744791909400196e-06,
|
| 224 |
+
"loss": 0.029,
|
| 225 |
+
"step": 12000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 4.369004979470604,
|
| 229 |
+
"grad_norm": 42.926326751708984,
|
| 230 |
+
"learning_rate": 7.512233769865313e-06,
|
| 231 |
+
"loss": 0.0313,
|
| 232 |
+
"step": 12500
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 4.543723246265397,
|
| 236 |
+
"grad_norm": 13.328317642211914,
|
| 237 |
+
"learning_rate": 7.2792095819546074e-06,
|
| 238 |
+
"loss": 0.0326,
|
| 239 |
+
"step": 13000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 4.71844151306019,
|
| 243 |
+
"grad_norm": 0.08160242438316345,
|
| 244 |
+
"learning_rate": 7.0461853940439014e-06,
|
| 245 |
+
"loss": 0.0293,
|
| 246 |
+
"step": 13500
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 4.893159779854984,
|
| 250 |
+
"grad_norm": 21.219942092895508,
|
| 251 |
+
"learning_rate": 6.813161206133197e-06,
|
| 252 |
+
"loss": 0.0281,
|
| 253 |
+
"step": 14000
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 4.999737922599808,
|
| 257 |
+
"eval_accuracy": 0.9791229909154437,
|
| 258 |
+
"eval_f1": 0.9791175700337718,
|
| 259 |
+
"eval_loss": 0.12131603062152863,
|
| 260 |
+
"eval_precision": 0.9791137606071209,
|
| 261 |
+
"eval_recall": 0.9791229909154437,
|
| 262 |
+
"eval_runtime": 68.6286,
|
| 263 |
+
"eval_samples_per_second": 333.622,
|
| 264 |
+
"eval_steps_per_second": 20.851,
|
| 265 |
+
"step": 14305
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 5.068140124049969,
|
| 269 |
+
"grad_norm": 8.010854721069336,
|
| 270 |
+
"learning_rate": 6.580137018222492e-06,
|
| 271 |
+
"loss": 0.0292,
|
| 272 |
+
"step": 14500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 5.242858390844763,
|
| 276 |
+
"grad_norm": 4.4188337326049805,
|
| 277 |
+
"learning_rate": 6.347112830311787e-06,
|
| 278 |
+
"loss": 0.0215,
|
| 279 |
+
"step": 15000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 5.4175766576395565,
|
| 283 |
+
"grad_norm": 65.55230712890625,
|
| 284 |
+
"learning_rate": 6.1145546907769025e-06,
|
| 285 |
+
"loss": 0.021,
|
| 286 |
+
"step": 15500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 5.592294924434349,
|
| 290 |
+
"grad_norm": 0.20221485197544098,
|
| 291 |
+
"learning_rate": 5.881530502866198e-06,
|
| 292 |
+
"loss": 0.0225,
|
| 293 |
+
"step": 16000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 5.767013191229143,
|
| 297 |
+
"grad_norm": 65.68180084228516,
|
| 298 |
+
"learning_rate": 5.648506314955493e-06,
|
| 299 |
+
"loss": 0.0199,
|
| 300 |
+
"step": 16500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 5.941731458023936,
|
| 304 |
+
"grad_norm": 0.33296695351600647,
|
| 305 |
+
"learning_rate": 5.415482127044788e-06,
|
| 306 |
+
"loss": 0.0211,
|
| 307 |
+
"step": 17000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 5.999737922599808,
|
| 311 |
+
"eval_accuracy": 0.9804769392033543,
|
| 312 |
+
"eval_f1": 0.98049806714974,
|
| 313 |
+
"eval_loss": 0.10157082974910736,
|
| 314 |
+
"eval_precision": 0.9805481970936862,
|
| 315 |
+
"eval_recall": 0.9804769392033543,
|
| 316 |
+
"eval_runtime": 68.666,
|
| 317 |
+
"eval_samples_per_second": 333.44,
|
| 318 |
+
"eval_steps_per_second": 20.84,
|
| 319 |
+
"step": 17166
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 6.116711802218922,
|
| 323 |
+
"grad_norm": 0.5054177045822144,
|
| 324 |
+
"learning_rate": 5.182457939134083e-06,
|
| 325 |
+
"loss": 0.0149,
|
| 326 |
+
"step": 17500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 6.291430069013716,
|
| 330 |
+
"grad_norm": 0.08331651240587234,
|
| 331 |
+
"learning_rate": 4.9494337512233775e-06,
|
| 332 |
+
"loss": 0.0141,
|
| 333 |
+
"step": 18000
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 6.466148335808509,
|
| 337 |
+
"grad_norm": 0.04097016155719757,
|
| 338 |
+
"learning_rate": 4.716409563312672e-06,
|
| 339 |
+
"loss": 0.0107,
|
| 340 |
+
"step": 18500
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 6.640866602603302,
|
| 344 |
+
"grad_norm": 0.009394422173500061,
|
| 345 |
+
"learning_rate": 4.483385375401967e-06,
|
| 346 |
+
"loss": 0.0142,
|
| 347 |
+
"step": 19000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 6.815584869398096,
|
| 351 |
+
"grad_norm": 55.671607971191406,
|
| 352 |
+
"learning_rate": 4.250361187491262e-06,
|
| 353 |
+
"loss": 0.0112,
|
| 354 |
+
"step": 19500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 6.990303136192889,
|
| 358 |
+
"grad_norm": 37.646053314208984,
|
| 359 |
+
"learning_rate": 4.018269096332199e-06,
|
| 360 |
+
"loss": 0.0143,
|
| 361 |
+
"step": 20000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 6.999737922599808,
|
| 365 |
+
"eval_accuracy": 0.9821366177498253,
|
| 366 |
+
"eval_f1": 0.9821391542473246,
|
| 367 |
+
"eval_loss": 0.12774688005447388,
|
| 368 |
+
"eval_precision": 0.9821421791956497,
|
| 369 |
+
"eval_recall": 0.9821366177498253,
|
| 370 |
+
"eval_runtime": 68.6239,
|
| 371 |
+
"eval_samples_per_second": 333.645,
|
| 372 |
+
"eval_steps_per_second": 20.853,
|
| 373 |
+
"step": 20027
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 7.165283480387875,
|
| 377 |
+
"grad_norm": 4.926445007324219,
|
| 378 |
+
"learning_rate": 3.7852449084214942e-06,
|
| 379 |
+
"loss": 0.0079,
|
| 380 |
+
"step": 20500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 7.340001747182668,
|
| 384 |
+
"grad_norm": 8.786434173583984,
|
| 385 |
+
"learning_rate": 3.5522207205107895e-06,
|
| 386 |
+
"loss": 0.0089,
|
| 387 |
+
"step": 21000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 7.5147200139774615,
|
| 391 |
+
"grad_norm": 0.1356939673423767,
|
| 392 |
+
"learning_rate": 3.319196532600084e-06,
|
| 393 |
+
"loss": 0.0074,
|
| 394 |
+
"step": 21500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 7.689438280772254,
|
| 398 |
+
"grad_norm": 16.782020568847656,
|
| 399 |
+
"learning_rate": 3.0861723446893788e-06,
|
| 400 |
+
"loss": 0.0096,
|
| 401 |
+
"step": 22000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 7.864156547567048,
|
| 405 |
+
"grad_norm": 0.0004071469884365797,
|
| 406 |
+
"learning_rate": 2.853148156778674e-06,
|
| 407 |
+
"loss": 0.007,
|
| 408 |
+
"step": 22500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 7.999737922599808,
|
| 412 |
+
"eval_accuracy": 0.9819619147449337,
|
| 413 |
+
"eval_f1": 0.9819917414068494,
|
| 414 |
+
"eval_loss": 0.14060455560684204,
|
| 415 |
+
"eval_precision": 0.9820904217450387,
|
| 416 |
+
"eval_recall": 0.9819619147449337,
|
| 417 |
+
"eval_runtime": 68.6348,
|
| 418 |
+
"eval_samples_per_second": 333.592,
|
| 419 |
+
"eval_steps_per_second": 20.849,
|
| 420 |
+
"step": 22888
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 8.039136891762034,
|
| 424 |
+
"grad_norm": 0.003943814896047115,
|
| 425 |
+
"learning_rate": 2.6201239688679685e-06,
|
| 426 |
+
"loss": 0.008,
|
| 427 |
+
"step": 23000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 8.213855158556827,
|
| 431 |
+
"grad_norm": 12.356912612915039,
|
| 432 |
+
"learning_rate": 2.3870997809572637e-06,
|
| 433 |
+
"loss": 0.0051,
|
| 434 |
+
"step": 23500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 8.38857342535162,
|
| 438 |
+
"grad_norm": 0.018336663022637367,
|
| 439 |
+
"learning_rate": 2.1540755930465586e-06,
|
| 440 |
+
"loss": 0.0053,
|
| 441 |
+
"step": 24000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 8.563291692146414,
|
| 445 |
+
"grad_norm": 0.031037895008921623,
|
| 446 |
+
"learning_rate": 1.9210514051358534e-06,
|
| 447 |
+
"loss": 0.0048,
|
| 448 |
+
"step": 24500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 8.738009958941207,
|
| 452 |
+
"grad_norm": 0.0018096828134730458,
|
| 453 |
+
"learning_rate": 1.6884932656009695e-06,
|
| 454 |
+
"loss": 0.0041,
|
| 455 |
+
"step": 25000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 8.912728225736,
|
| 459 |
+
"grad_norm": 0.006106176879256964,
|
| 460 |
+
"learning_rate": 1.455935126066086e-06,
|
| 461 |
+
"loss": 0.0043,
|
| 462 |
+
"step": 25500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 8.999737922599808,
|
| 466 |
+
"eval_accuracy": 0.9835779175401816,
|
| 467 |
+
"eval_f1": 0.9835697887213877,
|
| 468 |
+
"eval_loss": 0.15101368725299835,
|
| 469 |
+
"eval_precision": 0.9835675135482098,
|
| 470 |
+
"eval_recall": 0.9835779175401816,
|
| 471 |
+
"eval_runtime": 68.619,
|
| 472 |
+
"eval_samples_per_second": 333.668,
|
| 473 |
+
"eval_steps_per_second": 20.854,
|
| 474 |
+
"step": 25749
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 9.087708569930987,
|
| 478 |
+
"grad_norm": 0.00021469616331160069,
|
| 479 |
+
"learning_rate": 1.2229109381553806e-06,
|
| 480 |
+
"loss": 0.0036,
|
| 481 |
+
"step": 26000
|
| 482 |
+
},
|
| 483 |
+
{
|
| 484 |
+
"epoch": 9.26242683672578,
|
| 485 |
+
"grad_norm": 0.008365228772163391,
|
| 486 |
+
"learning_rate": 9.898867502446755e-07,
|
| 487 |
+
"loss": 0.0027,
|
| 488 |
+
"step": 26500
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 9.437145103520573,
|
| 492 |
+
"grad_norm": 0.0006792581407353282,
|
| 493 |
+
"learning_rate": 7.568625623339704e-07,
|
| 494 |
+
"loss": 0.0028,
|
| 495 |
+
"step": 27000
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 9.611863370315366,
|
| 499 |
+
"grad_norm": 0.004307614639401436,
|
| 500 |
+
"learning_rate": 5.238383744232652e-07,
|
| 501 |
+
"loss": 0.0021,
|
| 502 |
+
"step": 27500
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 9.78658163711016,
|
| 506 |
+
"grad_norm": 0.0005787264672107995,
|
| 507 |
+
"learning_rate": 2.9081418651256006e-07,
|
| 508 |
+
"loss": 0.0031,
|
| 509 |
+
"step": 28000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 9.961299903904953,
|
| 513 |
+
"grad_norm": 0.008150073699653149,
|
| 514 |
+
"learning_rate": 5.7789998601854876e-08,
|
| 515 |
+
"loss": 0.0019,
|
| 516 |
+
"step": 28500
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 9.999737922599808,
|
| 520 |
+
"eval_accuracy": 0.9842330538085255,
|
| 521 |
+
"eval_f1": 0.9842291652151357,
|
| 522 |
+
"eval_loss": 0.14871150255203247,
|
| 523 |
+
"eval_precision": 0.9842267397484589,
|
| 524 |
+
"eval_recall": 0.9842330538085255,
|
| 525 |
+
"eval_runtime": 68.5176,
|
| 526 |
+
"eval_samples_per_second": 334.162,
|
| 527 |
+
"eval_steps_per_second": 20.885,
|
| 528 |
+
"step": 28610
|
| 529 |
+
}
|
| 530 |
+
],
|
| 531 |
+
"logging_steps": 500,
|
| 532 |
+
"max_steps": 28610,
|
| 533 |
+
"num_input_tokens_seen": 0,
|
| 534 |
+
"num_train_epochs": 10,
|
| 535 |
+
"save_steps": 500,
|
| 536 |
+
"stateful_callbacks": {
|
| 537 |
+
"EarlyStoppingCallback": {
|
| 538 |
+
"args": {
|
| 539 |
+
"early_stopping_patience": 3,
|
| 540 |
+
"early_stopping_threshold": 0.001
|
| 541 |
+
},
|
| 542 |
+
"attributes": {
|
| 543 |
+
"early_stopping_patience_counter": 1
|
| 544 |
+
}
|
| 545 |
+
},
|
| 546 |
+
"TrainerControl": {
|
| 547 |
+
"args": {
|
| 548 |
+
"should_epoch_stop": false,
|
| 549 |
+
"should_evaluate": false,
|
| 550 |
+
"should_log": false,
|
| 551 |
+
"should_save": true,
|
| 552 |
+
"should_training_stop": true
|
| 553 |
+
},
|
| 554 |
+
"attributes": {}
|
| 555 |
+
}
|
| 556 |
+
},
|
| 557 |
+
"total_flos": 1.7066719997407396e+18,
|
| 558 |
+
"train_batch_size": 16,
|
| 559 |
+
"trial_name": null,
|
| 560 |
+
"trial_params": null
|
| 561 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0f628c5e446ae54cdc2d20f466a5d879e7f13d5508ffaa08829c41eea5d0fde3
|
| 3 |
+
size 5560
|