prune 10 old checkpoints
Browse files- slots/0/checkpoint-360/config.json +0 -24
- slots/0/checkpoint-360/trainer_state.json +0 -167
- slots/10/checkpoint-399/config.json +0 -24
- slots/10/checkpoint-399/trainer_state.json +0 -174
- slots/12/checkpoint-389/config.json +0 -24
- slots/12/checkpoint-389/trainer_state.json +0 -174
- slots/14/checkpoint-409/config.json +0 -24
- slots/14/checkpoint-409/trainer_state.json +0 -181
- slots/16/checkpoint-470/config.json +0 -24
- slots/16/checkpoint-470/trainer_state.json +0 -202
- slots/18/checkpoint-380/config.json +0 -24
- slots/18/checkpoint-380/trainer_state.json +0 -174
- slots/2/checkpoint-376/config.json +0 -24
- slots/2/checkpoint-376/trainer_state.json +0 -167
- slots/4/checkpoint-340/config.json +0 -24
- slots/4/checkpoint-340/trainer_state.json +0 -160
- slots/6/checkpoint-405/config.json +0 -24
- slots/6/checkpoint-405/trainer_state.json +0 -181
- slots/8/checkpoint-403/config.json +0 -24
- slots/8/checkpoint-403/trainer_state.json +0 -181
slots/0/checkpoint-360/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/0/checkpoint-360/trainer_state.json
DELETED
|
@@ -1,167 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.018842519715949014,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 360,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
}
|
| 145 |
-
],
|
| 146 |
-
"logging_steps": 20,
|
| 147 |
-
"max_steps": 47765,
|
| 148 |
-
"num_input_tokens_seen": 0,
|
| 149 |
-
"num_train_epochs": 3,
|
| 150 |
-
"save_steps": 1000000000,
|
| 151 |
-
"stateful_callbacks": {
|
| 152 |
-
"TrainerControl": {
|
| 153 |
-
"args": {
|
| 154 |
-
"should_epoch_stop": false,
|
| 155 |
-
"should_evaluate": false,
|
| 156 |
-
"should_log": false,
|
| 157 |
-
"should_save": true,
|
| 158 |
-
"should_training_stop": false
|
| 159 |
-
},
|
| 160 |
-
"attributes": {}
|
| 161 |
-
}
|
| 162 |
-
},
|
| 163 |
-
"total_flos": 372965094604800.0,
|
| 164 |
-
"train_batch_size": 1,
|
| 165 |
-
"trial_name": null,
|
| 166 |
-
"trial_params": null
|
| 167 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/10/checkpoint-399/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/10/checkpoint-399/trainer_state.json
DELETED
|
@@ -1,174 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.020883792685176827,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 399,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 52.77220153808594,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 251.9318359375,
|
| 150 |
-
"step": 380
|
| 151 |
-
}
|
| 152 |
-
],
|
| 153 |
-
"logging_steps": 20,
|
| 154 |
-
"max_steps": 47765,
|
| 155 |
-
"num_input_tokens_seen": 0,
|
| 156 |
-
"num_train_epochs": 3,
|
| 157 |
-
"save_steps": 1000000000,
|
| 158 |
-
"stateful_callbacks": {
|
| 159 |
-
"TrainerControl": {
|
| 160 |
-
"args": {
|
| 161 |
-
"should_epoch_stop": false,
|
| 162 |
-
"should_evaluate": false,
|
| 163 |
-
"should_log": false,
|
| 164 |
-
"should_save": true,
|
| 165 |
-
"should_training_stop": false
|
| 166 |
-
},
|
| 167 |
-
"attributes": {}
|
| 168 |
-
}
|
| 169 |
-
},
|
| 170 |
-
"total_flos": 413369646520320.0,
|
| 171 |
-
"train_batch_size": 1,
|
| 172 |
-
"trial_name": null,
|
| 173 |
-
"trial_params": null
|
| 174 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/12/checkpoint-389/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/12/checkpoint-389/trainer_state.json
DELETED
|
@@ -1,174 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.020360389359733798,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 389,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 52.77220153808594,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 251.9318359375,
|
| 150 |
-
"step": 380
|
| 151 |
-
}
|
| 152 |
-
],
|
| 153 |
-
"logging_steps": 20,
|
| 154 |
-
"max_steps": 47765,
|
| 155 |
-
"num_input_tokens_seen": 0,
|
| 156 |
-
"num_train_epochs": 3,
|
| 157 |
-
"save_steps": 1000000000,
|
| 158 |
-
"stateful_callbacks": {
|
| 159 |
-
"TrainerControl": {
|
| 160 |
-
"args": {
|
| 161 |
-
"should_epoch_stop": false,
|
| 162 |
-
"should_evaluate": false,
|
| 163 |
-
"should_log": false,
|
| 164 |
-
"should_save": true,
|
| 165 |
-
"should_training_stop": false
|
| 166 |
-
},
|
| 167 |
-
"attributes": {}
|
| 168 |
-
}
|
| 169 |
-
},
|
| 170 |
-
"total_flos": 403009505003520.0,
|
| 171 |
-
"train_batch_size": 1,
|
| 172 |
-
"trial_name": null,
|
| 173 |
-
"trial_params": null
|
| 174 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/14/checkpoint-409/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/14/checkpoint-409/trainer_state.json
DELETED
|
@@ -1,181 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.021407196010619852,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 409,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 52.77220153808594,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 251.9318359375,
|
| 150 |
-
"step": 380
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 0.02093613301772113,
|
| 154 |
-
"grad_norm": 63.60757827758789,
|
| 155 |
-
"learning_rate": 4.999139182552709e-05,
|
| 156 |
-
"loss": 261.9561767578125,
|
| 157 |
-
"step": 400
|
| 158 |
-
}
|
| 159 |
-
],
|
| 160 |
-
"logging_steps": 20,
|
| 161 |
-
"max_steps": 47765,
|
| 162 |
-
"num_input_tokens_seen": 0,
|
| 163 |
-
"num_train_epochs": 3,
|
| 164 |
-
"save_steps": 1000000000,
|
| 165 |
-
"stateful_callbacks": {
|
| 166 |
-
"TrainerControl": {
|
| 167 |
-
"args": {
|
| 168 |
-
"should_epoch_stop": false,
|
| 169 |
-
"should_evaluate": false,
|
| 170 |
-
"should_log": false,
|
| 171 |
-
"should_save": true,
|
| 172 |
-
"should_training_stop": false
|
| 173 |
-
},
|
| 174 |
-
"attributes": {}
|
| 175 |
-
}
|
| 176 |
-
},
|
| 177 |
-
"total_flos": 423729788037120.0,
|
| 178 |
-
"train_batch_size": 1,
|
| 179 |
-
"trial_name": null,
|
| 180 |
-
"trial_params": null
|
| 181 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/16/checkpoint-470/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/16/checkpoint-470/trainer_state.json
DELETED
|
@@ -1,202 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.024599956295822324,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 470,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 48.20169448852539,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 288.82497336647725,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 68.66603088378906,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 293.87509765625,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 61.69401931762695,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 324.10361328125,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 56.778385162353516,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 274.875537109375,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 71.3816146850586,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 281.357080078125,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 75.08697509765625,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.852783203125,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 53.91341781616211,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 258.546630859375,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 56.39162063598633,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 270.5470703125,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 62.93777084350586,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 252.8383056640625,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 61.51578903198242,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 287.8167724609375,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 62.75235366821289,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 258.673779296875,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 56.2258186340332,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 252.9609375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 54.998130798339844,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 241.9411376953125,
|
| 150 |
-
"step": 380
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 0.02093613301772113,
|
| 154 |
-
"grad_norm": 63.37522888183594,
|
| 155 |
-
"learning_rate": 4.999139182552709e-05,
|
| 156 |
-
"loss": 249.7280517578125,
|
| 157 |
-
"step": 400
|
| 158 |
-
},
|
| 159 |
-
{
|
| 160 |
-
"epoch": 0.021982939668607183,
|
| 161 |
-
"grad_norm": 60.007144927978516,
|
| 162 |
-
"learning_rate": 4.9990507278185374e-05,
|
| 163 |
-
"loss": 236.941015625,
|
| 164 |
-
"step": 420
|
| 165 |
-
},
|
| 166 |
-
{
|
| 167 |
-
"epoch": 0.02302974631949324,
|
| 168 |
-
"grad_norm": 66.06304168701172,
|
| 169 |
-
"learning_rate": 4.998957948790677e-05,
|
| 170 |
-
"loss": 240.55878363715277,
|
| 171 |
-
"step": 440
|
| 172 |
-
},
|
| 173 |
-
{
|
| 174 |
-
"epoch": 0.0240765529703793,
|
| 175 |
-
"grad_norm": 68.9736099243164,
|
| 176 |
-
"learning_rate": 4.998860845629671e-05,
|
| 177 |
-
"loss": 247.988720703125,
|
| 178 |
-
"step": 460
|
| 179 |
-
}
|
| 180 |
-
],
|
| 181 |
-
"logging_steps": 20,
|
| 182 |
-
"max_steps": 47765,
|
| 183 |
-
"num_input_tokens_seen": 0,
|
| 184 |
-
"num_train_epochs": 3,
|
| 185 |
-
"save_steps": 1000000000,
|
| 186 |
-
"stateful_callbacks": {
|
| 187 |
-
"TrainerControl": {
|
| 188 |
-
"args": {
|
| 189 |
-
"should_epoch_stop": false,
|
| 190 |
-
"should_evaluate": false,
|
| 191 |
-
"should_log": false,
|
| 192 |
-
"should_save": true,
|
| 193 |
-
"should_training_stop": false
|
| 194 |
-
},
|
| 195 |
-
"attributes": {}
|
| 196 |
-
}
|
| 197 |
-
},
|
| 198 |
-
"total_flos": 486926651289600.0,
|
| 199 |
-
"train_batch_size": 1,
|
| 200 |
-
"trial_name": null,
|
| 201 |
-
"trial_params": null
|
| 202 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/18/checkpoint-380/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/18/checkpoint-380/trainer_state.json
DELETED
|
@@ -1,174 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.01988932636683507,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 380,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 52.77220153808594,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 251.9318359375,
|
| 150 |
-
"step": 380
|
| 151 |
-
}
|
| 152 |
-
],
|
| 153 |
-
"logging_steps": 20,
|
| 154 |
-
"max_steps": 47765,
|
| 155 |
-
"num_input_tokens_seen": 0,
|
| 156 |
-
"num_train_epochs": 3,
|
| 157 |
-
"save_steps": 1000000000,
|
| 158 |
-
"stateful_callbacks": {
|
| 159 |
-
"TrainerControl": {
|
| 160 |
-
"args": {
|
| 161 |
-
"should_epoch_stop": false,
|
| 162 |
-
"should_evaluate": false,
|
| 163 |
-
"should_log": false,
|
| 164 |
-
"should_save": true,
|
| 165 |
-
"should_training_stop": false
|
| 166 |
-
},
|
| 167 |
-
"attributes": {}
|
| 168 |
-
}
|
| 169 |
-
},
|
| 170 |
-
"total_flos": 393685377638400.0,
|
| 171 |
-
"train_batch_size": 1,
|
| 172 |
-
"trial_name": null,
|
| 173 |
-
"trial_params": null
|
| 174 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/2/checkpoint-376/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/2/checkpoint-376/trainer_state.json
DELETED
|
@@ -1,167 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.01967996503665786,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 376,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
}
|
| 145 |
-
],
|
| 146 |
-
"logging_steps": 20,
|
| 147 |
-
"max_steps": 47765,
|
| 148 |
-
"num_input_tokens_seen": 0,
|
| 149 |
-
"num_train_epochs": 3,
|
| 150 |
-
"save_steps": 1000000000,
|
| 151 |
-
"stateful_callbacks": {
|
| 152 |
-
"TrainerControl": {
|
| 153 |
-
"args": {
|
| 154 |
-
"should_epoch_stop": false,
|
| 155 |
-
"should_evaluate": false,
|
| 156 |
-
"should_log": false,
|
| 157 |
-
"should_save": true,
|
| 158 |
-
"should_training_stop": false
|
| 159 |
-
},
|
| 160 |
-
"attributes": {}
|
| 161 |
-
}
|
| 162 |
-
},
|
| 163 |
-
"total_flos": 389541321031680.0,
|
| 164 |
-
"train_batch_size": 1,
|
| 165 |
-
"trial_name": null,
|
| 166 |
-
"trial_params": null
|
| 167 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/4/checkpoint-340/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/4/checkpoint-340/trainer_state.json
DELETED
|
@@ -1,160 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.01779571306506296,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 340,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
}
|
| 138 |
-
],
|
| 139 |
-
"logging_steps": 20,
|
| 140 |
-
"max_steps": 47765,
|
| 141 |
-
"num_input_tokens_seen": 0,
|
| 142 |
-
"num_train_epochs": 3,
|
| 143 |
-
"save_steps": 1000000000,
|
| 144 |
-
"stateful_callbacks": {
|
| 145 |
-
"TrainerControl": {
|
| 146 |
-
"args": {
|
| 147 |
-
"should_epoch_stop": false,
|
| 148 |
-
"should_evaluate": false,
|
| 149 |
-
"should_log": false,
|
| 150 |
-
"should_save": true,
|
| 151 |
-
"should_training_stop": false
|
| 152 |
-
},
|
| 153 |
-
"attributes": {}
|
| 154 |
-
}
|
| 155 |
-
},
|
| 156 |
-
"total_flos": 352244811571200.0,
|
| 157 |
-
"train_batch_size": 1,
|
| 158 |
-
"trial_name": null,
|
| 159 |
-
"trial_params": null
|
| 160 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/6/checkpoint-405/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/6/checkpoint-405/trainer_state.json
DELETED
|
@@ -1,181 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.021197834680442642,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 405,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 52.77220153808594,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 251.9318359375,
|
| 150 |
-
"step": 380
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 0.02093613301772113,
|
| 154 |
-
"grad_norm": 63.60757827758789,
|
| 155 |
-
"learning_rate": 4.999139182552709e-05,
|
| 156 |
-
"loss": 261.9561767578125,
|
| 157 |
-
"step": 400
|
| 158 |
-
}
|
| 159 |
-
],
|
| 160 |
-
"logging_steps": 20,
|
| 161 |
-
"max_steps": 47765,
|
| 162 |
-
"num_input_tokens_seen": 0,
|
| 163 |
-
"num_train_epochs": 3,
|
| 164 |
-
"save_steps": 1000000000,
|
| 165 |
-
"stateful_callbacks": {
|
| 166 |
-
"TrainerControl": {
|
| 167 |
-
"args": {
|
| 168 |
-
"should_epoch_stop": false,
|
| 169 |
-
"should_evaluate": false,
|
| 170 |
-
"should_log": false,
|
| 171 |
-
"should_save": true,
|
| 172 |
-
"should_training_stop": false
|
| 173 |
-
},
|
| 174 |
-
"attributes": {}
|
| 175 |
-
}
|
| 176 |
-
},
|
| 177 |
-
"total_flos": 419585731430400.0,
|
| 178 |
-
"train_batch_size": 1,
|
| 179 |
-
"trial_name": null,
|
| 180 |
-
"trial_params": null
|
| 181 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/8/checkpoint-403/config.json
DELETED
|
@@ -1,24 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"TwinyForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_dropout": 0.0,
|
| 6 |
-
"dtype": "float32",
|
| 7 |
-
"hidden_dropout": 0.0,
|
| 8 |
-
"hidden_size": 768,
|
| 9 |
-
"initializer_range": 0.02,
|
| 10 |
-
"intermediate_size": 3072,
|
| 11 |
-
"max_position_embeddings": 128,
|
| 12 |
-
"model_type": "twiny",
|
| 13 |
-
"neftune_alpha": 0.0,
|
| 14 |
-
"num_attention_heads": 12,
|
| 15 |
-
"num_hidden_layers": 1,
|
| 16 |
-
"num_key_value_heads": 3,
|
| 17 |
-
"qk_norm": true,
|
| 18 |
-
"rezero_init": 1.0,
|
| 19 |
-
"rms_norm_eps": 1e-06,
|
| 20 |
-
"rope_theta": 10000.0,
|
| 21 |
-
"transformers_version": "5.0.0",
|
| 22 |
-
"use_cache": false,
|
| 23 |
-
"vocab_size": 32000
|
| 24 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
slots/8/checkpoint-403/trainer_state.json
DELETED
|
@@ -1,181 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"best_global_step": null,
|
| 3 |
-
"best_metric": null,
|
| 4 |
-
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.021093154015354037,
|
| 6 |
-
"eval_steps": 500,
|
| 7 |
-
"global_step": 403,
|
| 8 |
-
"is_hyper_param_search": false,
|
| 9 |
-
"is_local_process_zero": true,
|
| 10 |
-
"is_world_process_zero": true,
|
| 11 |
-
"log_history": [
|
| 12 |
-
{
|
| 13 |
-
"epoch": 2.617016627215141e-05,
|
| 14 |
-
"grad_norm": 122.33340454101562,
|
| 15 |
-
"learning_rate": 5e-05,
|
| 16 |
-
"loss": 419.9996032714844,
|
| 17 |
-
"step": 1
|
| 18 |
-
},
|
| 19 |
-
{
|
| 20 |
-
"epoch": 0.0005234033254430282,
|
| 21 |
-
"grad_norm": 90.82410430908203,
|
| 22 |
-
"learning_rate": 4.999999511980259e-05,
|
| 23 |
-
"loss": 428.18197471217104,
|
| 24 |
-
"step": 20
|
| 25 |
-
},
|
| 26 |
-
{
|
| 27 |
-
"epoch": 0.0010468066508860564,
|
| 28 |
-
"grad_norm": 59.52641296386719,
|
| 29 |
-
"learning_rate": 4.999997943828395e-05,
|
| 30 |
-
"loss": 366.3947265625,
|
| 31 |
-
"step": 40
|
| 32 |
-
},
|
| 33 |
-
{
|
| 34 |
-
"epoch": 0.0015702099763290846,
|
| 35 |
-
"grad_norm": 41.61277389526367,
|
| 36 |
-
"learning_rate": 4.999995294193233e-05,
|
| 37 |
-
"loss": 325.6545166015625,
|
| 38 |
-
"step": 60
|
| 39 |
-
},
|
| 40 |
-
{
|
| 41 |
-
"epoch": 0.0020936133017721128,
|
| 42 |
-
"grad_norm": 46.67556381225586,
|
| 43 |
-
"learning_rate": 4.999991563075919e-05,
|
| 44 |
-
"loss": 310.3648193359375,
|
| 45 |
-
"step": 80
|
| 46 |
-
},
|
| 47 |
-
{
|
| 48 |
-
"epoch": 0.002617016627215141,
|
| 49 |
-
"grad_norm": 47.65824890136719,
|
| 50 |
-
"learning_rate": 4.999986750478065e-05,
|
| 51 |
-
"loss": 303.49443359375,
|
| 52 |
-
"step": 100
|
| 53 |
-
},
|
| 54 |
-
{
|
| 55 |
-
"epoch": 0.003140419952658169,
|
| 56 |
-
"grad_norm": 62.748138427734375,
|
| 57 |
-
"learning_rate": 4.999980856401755e-05,
|
| 58 |
-
"loss": 322.389990234375,
|
| 59 |
-
"step": 120
|
| 60 |
-
},
|
| 61 |
-
{
|
| 62 |
-
"epoch": 0.007327646556202395,
|
| 63 |
-
"grad_norm": 50.97892761230469,
|
| 64 |
-
"learning_rate": 4.999895523943923e-05,
|
| 65 |
-
"loss": 290.35420365767044,
|
| 66 |
-
"step": 140
|
| 67 |
-
},
|
| 68 |
-
{
|
| 69 |
-
"epoch": 0.008374453207088451,
|
| 70 |
-
"grad_norm": 71.21157836914062,
|
| 71 |
-
"learning_rate": 4.99986329623214e-05,
|
| 72 |
-
"loss": 286.067724609375,
|
| 73 |
-
"step": 160
|
| 74 |
-
},
|
| 75 |
-
{
|
| 76 |
-
"epoch": 0.009421259857974507,
|
| 77 |
-
"grad_norm": 76.37614440917969,
|
| 78 |
-
"learning_rate": 4.999826742820622e-05,
|
| 79 |
-
"loss": 279.012451171875,
|
| 80 |
-
"step": 180
|
| 81 |
-
},
|
| 82 |
-
{
|
| 83 |
-
"epoch": 0.010468066508860565,
|
| 84 |
-
"grad_norm": 76.57452392578125,
|
| 85 |
-
"learning_rate": 4.999785863772618e-05,
|
| 86 |
-
"loss": 287.710888671875,
|
| 87 |
-
"step": 200
|
| 88 |
-
},
|
| 89 |
-
{
|
| 90 |
-
"epoch": 0.01151487315974662,
|
| 91 |
-
"grad_norm": 55.99504852294922,
|
| 92 |
-
"learning_rate": 4.999740659158865e-05,
|
| 93 |
-
"loss": 283.03916015625,
|
| 94 |
-
"step": 220
|
| 95 |
-
},
|
| 96 |
-
{
|
| 97 |
-
"epoch": 0.012561679810632677,
|
| 98 |
-
"grad_norm": 61.760135650634766,
|
| 99 |
-
"learning_rate": 4.9996911290575844e-05,
|
| 100 |
-
"loss": 263.22587890625,
|
| 101 |
-
"step": 240
|
| 102 |
-
},
|
| 103 |
-
{
|
| 104 |
-
"epoch": 0.013608486461518732,
|
| 105 |
-
"grad_norm": 51.249114990234375,
|
| 106 |
-
"learning_rate": 4.999637273554481e-05,
|
| 107 |
-
"loss": 261.2483642578125,
|
| 108 |
-
"step": 260
|
| 109 |
-
},
|
| 110 |
-
{
|
| 111 |
-
"epoch": 0.01465529311240479,
|
| 112 |
-
"grad_norm": 53.85161590576172,
|
| 113 |
-
"learning_rate": 4.9995790927427446e-05,
|
| 114 |
-
"loss": 257.092822265625,
|
| 115 |
-
"step": 280
|
| 116 |
-
},
|
| 117 |
-
{
|
| 118 |
-
"epoch": 0.015702099763290844,
|
| 119 |
-
"grad_norm": 55.01472473144531,
|
| 120 |
-
"learning_rate": 4.99951658672305e-05,
|
| 121 |
-
"loss": 256.8466796875,
|
| 122 |
-
"step": 300
|
| 123 |
-
},
|
| 124 |
-
{
|
| 125 |
-
"epoch": 0.016748906414176902,
|
| 126 |
-
"grad_norm": 57.187171936035156,
|
| 127 |
-
"learning_rate": 4.9994497556035567e-05,
|
| 128 |
-
"loss": 251.65341796875,
|
| 129 |
-
"step": 320
|
| 130 |
-
},
|
| 131 |
-
{
|
| 132 |
-
"epoch": 0.01779571306506296,
|
| 133 |
-
"grad_norm": 56.99523162841797,
|
| 134 |
-
"learning_rate": 4.9993785994999074e-05,
|
| 135 |
-
"loss": 250.6625244140625,
|
| 136 |
-
"step": 340
|
| 137 |
-
},
|
| 138 |
-
{
|
| 139 |
-
"epoch": 0.018842519715949014,
|
| 140 |
-
"grad_norm": 51.036258697509766,
|
| 141 |
-
"learning_rate": 4.999303118535229e-05,
|
| 142 |
-
"loss": 246.926318359375,
|
| 143 |
-
"step": 360
|
| 144 |
-
},
|
| 145 |
-
{
|
| 146 |
-
"epoch": 0.01988932636683507,
|
| 147 |
-
"grad_norm": 52.77220153808594,
|
| 148 |
-
"learning_rate": 4.9992233128401314e-05,
|
| 149 |
-
"loss": 251.9318359375,
|
| 150 |
-
"step": 380
|
| 151 |
-
},
|
| 152 |
-
{
|
| 153 |
-
"epoch": 0.02093613301772113,
|
| 154 |
-
"grad_norm": 63.60757827758789,
|
| 155 |
-
"learning_rate": 4.999139182552709e-05,
|
| 156 |
-
"loss": 261.9561767578125,
|
| 157 |
-
"step": 400
|
| 158 |
-
}
|
| 159 |
-
],
|
| 160 |
-
"logging_steps": 20,
|
| 161 |
-
"max_steps": 47765,
|
| 162 |
-
"num_input_tokens_seen": 0,
|
| 163 |
-
"num_train_epochs": 3,
|
| 164 |
-
"save_steps": 1000000000,
|
| 165 |
-
"stateful_callbacks": {
|
| 166 |
-
"TrainerControl": {
|
| 167 |
-
"args": {
|
| 168 |
-
"should_epoch_stop": false,
|
| 169 |
-
"should_evaluate": false,
|
| 170 |
-
"should_log": false,
|
| 171 |
-
"should_save": true,
|
| 172 |
-
"should_training_stop": false
|
| 173 |
-
},
|
| 174 |
-
"attributes": {}
|
| 175 |
-
}
|
| 176 |
-
},
|
| 177 |
-
"total_flos": 417513703127040.0,
|
| 178 |
-
"train_batch_size": 1,
|
| 179 |
-
"trial_name": null,
|
| 180 |
-
"trial_params": null
|
| 181 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|