Upload folder using huggingface_hub
Browse files- checkpoint-1012/config.json +61 -0
- checkpoint-1012/generation_config.json +4 -0
- checkpoint-1012/model.safetensors +3 -0
- checkpoint-1012/optimizer.pt +3 -0
- checkpoint-1012/rng_state.pth +3 -0
- checkpoint-1012/scheduler.pt +3 -0
- checkpoint-1012/trainer_state.json +200 -0
- checkpoint-1012/training_args.bin +3 -0
- checkpoint-2024/config.json +61 -0
- checkpoint-2024/generation_config.json +4 -0
- checkpoint-2024/model.safetensors +3 -0
- checkpoint-2024/optimizer.pt +3 -0
- checkpoint-2024/rng_state.pth +3 -0
- checkpoint-2024/scheduler.pt +3 -0
- checkpoint-2024/trainer_state.json +372 -0
- checkpoint-2024/training_args.bin +3 -0
- config.json +61 -61
- generation_config.json +4 -4
- model.safetensors +1 -1
- tb_logs/events.out.tfevents.1760423030.luyao1.2719331.0 +3 -0
- training_args.bin +2 -2
checkpoint-1012/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "time_moe_50m",
|
| 3 |
+
"apply_aux_loss": true,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"TimeMoeForPrediction"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "Maple728/TimeMoE-50M--configuration_time_moe.TimeMoeConfig",
|
| 10 |
+
"AutoModelForCausalLM": "Maple728/TimeMoE-50M--modeling_time_moe.TimeMoeForPrediction"
|
| 11 |
+
},
|
| 12 |
+
"channel_configs": [
|
| 13 |
+
[
|
| 14 |
+
63,
|
| 15 |
+
1,
|
| 16 |
+
1
|
| 17 |
+
],
|
| 18 |
+
[
|
| 19 |
+
6,
|
| 20 |
+
1,
|
| 21 |
+
4
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
5,
|
| 25 |
+
1,
|
| 26 |
+
1
|
| 27 |
+
],
|
| 28 |
+
[
|
| 29 |
+
5,
|
| 30 |
+
1,
|
| 31 |
+
1
|
| 32 |
+
]
|
| 33 |
+
],
|
| 34 |
+
"embedding_hidden_size": 128,
|
| 35 |
+
"hidden_act": "silu",
|
| 36 |
+
"hidden_size": 384,
|
| 37 |
+
"horizon_lengths": [
|
| 38 |
+
1,
|
| 39 |
+
8,
|
| 40 |
+
32,
|
| 41 |
+
64
|
| 42 |
+
],
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"input_size": 42,
|
| 45 |
+
"intermediate_size": 1536,
|
| 46 |
+
"max_position_embeddings": 4096,
|
| 47 |
+
"model_type": "time_moe",
|
| 48 |
+
"num_attention_heads": 12,
|
| 49 |
+
"num_experts": 8,
|
| 50 |
+
"num_experts_per_tok": 2,
|
| 51 |
+
"num_hidden_layers": 12,
|
| 52 |
+
"num_key_value_heads": 12,
|
| 53 |
+
"rms_norm_eps": 1e-06,
|
| 54 |
+
"rope_theta": 10000,
|
| 55 |
+
"router_aux_loss_factor": 0.02,
|
| 56 |
+
"tie_word_embeddings": false,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.40.1",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"use_dense": false
|
| 61 |
+
}
|
checkpoint-1012/generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.40.1"
|
| 4 |
+
}
|
checkpoint-1012/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6b88928e1afca14f30bdd89c066721d058297e625931a890a86e98f678b84e7
|
| 3 |
+
size 484301192
|
checkpoint-1012/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7e79974ab08ca0fccbfb17cd4c57676c930d13a46445285f20686faf9d0409b
|
| 3 |
+
size 968924920
|
checkpoint-1012/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5cbe4833c475023cd0b15013740d375640598d423ffe268731c6a5fe33bce683
|
| 3 |
+
size 14645
|
checkpoint-1012/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:db999be284ba0abf1c609173979ea3f83c7c9b8cfb47a4d480f93d361d849447
|
| 3 |
+
size 1465
|
checkpoint-1012/trainer_state.json
ADDED
|
@@ -0,0 +1,200 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9995061728395062,
|
| 5 |
+
"eval_steps": 250,
|
| 6 |
+
"global_step": 1012,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0009876543209876543,
|
| 13 |
+
"grad_norm": 1.221229910850525,
|
| 14 |
+
"learning_rate": 9.99999698845987e-05,
|
| 15 |
+
"loss": 0.4907,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04938271604938271,
|
| 20 |
+
"grad_norm": 0.05035819858312607,
|
| 21 |
+
"learning_rate": 9.99247492630919e-05,
|
| 22 |
+
"loss": 0.3903,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09876543209876543,
|
| 27 |
+
"grad_norm": 0.14429445564746857,
|
| 28 |
+
"learning_rate": 9.969945006624004e-05,
|
| 29 |
+
"loss": 0.3593,
|
| 30 |
+
"step": 100
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14814814814814814,
|
| 34 |
+
"grad_norm": 0.15117663145065308,
|
| 35 |
+
"learning_rate": 9.932545872389141e-05,
|
| 36 |
+
"loss": 0.3437,
|
| 37 |
+
"step": 150
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.19753086419753085,
|
| 41 |
+
"grad_norm": 0.19646206498146057,
|
| 42 |
+
"learning_rate": 9.880502668597476e-05,
|
| 43 |
+
"loss": 0.3354,
|
| 44 |
+
"step": 200
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24691358024691357,
|
| 48 |
+
"grad_norm": 0.25732266902923584,
|
| 49 |
+
"learning_rate": 9.814128698403919e-05,
|
| 50 |
+
"loss": 0.3283,
|
| 51 |
+
"step": 250
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.24691358024691357,
|
| 55 |
+
"eval_loss": 0.3421019911766052,
|
| 56 |
+
"eval_runtime": 137.4213,
|
| 57 |
+
"eval_samples_per_second": 686.211,
|
| 58 |
+
"eval_steps_per_second": 21.445,
|
| 59 |
+
"step": 250
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.2962962962962963,
|
| 63 |
+
"grad_norm": 0.12474808096885681,
|
| 64 |
+
"learning_rate": 9.733823537021955e-05,
|
| 65 |
+
"loss": 0.3266,
|
| 66 |
+
"step": 300
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.345679012345679,
|
| 70 |
+
"grad_norm": 0.12876740097999573,
|
| 71 |
+
"learning_rate": 9.640070626257307e-05,
|
| 72 |
+
"loss": 0.3259,
|
| 73 |
+
"step": 350
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.3950617283950617,
|
| 77 |
+
"grad_norm": 0.20461653172969818,
|
| 78 |
+
"learning_rate": 9.533434364159762e-05,
|
| 79 |
+
"loss": 0.3248,
|
| 80 |
+
"step": 400
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.4444444444444444,
|
| 84 |
+
"grad_norm": 0.22041602432727814,
|
| 85 |
+
"learning_rate": 9.414556707313633e-05,
|
| 86 |
+
"loss": 0.3219,
|
| 87 |
+
"step": 450
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.49382716049382713,
|
| 91 |
+
"grad_norm": 0.27202293276786804,
|
| 92 |
+
"learning_rate": 9.284153306221289e-05,
|
| 93 |
+
"loss": 0.3178,
|
| 94 |
+
"step": 500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.49382716049382713,
|
| 98 |
+
"eval_loss": 0.33308419585227966,
|
| 99 |
+
"eval_runtime": 140.6684,
|
| 100 |
+
"eval_samples_per_second": 670.371,
|
| 101 |
+
"eval_steps_per_second": 20.95,
|
| 102 |
+
"step": 500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.5432098765432098,
|
| 106 |
+
"grad_norm": 0.1661626547574997,
|
| 107 |
+
"learning_rate": 9.143009197044932e-05,
|
| 108 |
+
"loss": 0.3211,
|
| 109 |
+
"step": 550
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5925925925925926,
|
| 113 |
+
"grad_norm": 0.14871230721473694,
|
| 114 |
+
"learning_rate": 8.991974075642621e-05,
|
| 115 |
+
"loss": 0.3207,
|
| 116 |
+
"step": 600
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.6419753086419753,
|
| 120 |
+
"grad_norm": 0.17097660899162292,
|
| 121 |
+
"learning_rate": 8.831957182349119e-05,
|
| 122 |
+
"loss": 0.3185,
|
| 123 |
+
"step": 650
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.691358024691358,
|
| 127 |
+
"grad_norm": 0.1582462042570114,
|
| 128 |
+
"learning_rate": 8.663921828295474e-05,
|
| 129 |
+
"loss": 0.3173,
|
| 130 |
+
"step": 700
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.7407407407407407,
|
| 134 |
+
"grad_norm": 0.25067687034606934,
|
| 135 |
+
"learning_rate": 8.488879596219216e-05,
|
| 136 |
+
"loss": 0.3191,
|
| 137 |
+
"step": 750
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.7407407407407407,
|
| 141 |
+
"eval_loss": 0.33356761932373047,
|
| 142 |
+
"eval_runtime": 145.0159,
|
| 143 |
+
"eval_samples_per_second": 650.274,
|
| 144 |
+
"eval_steps_per_second": 20.322,
|
| 145 |
+
"step": 750
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.7901234567901234,
|
| 149 |
+
"grad_norm": 0.1577720046043396,
|
| 150 |
+
"learning_rate": 8.307884250676648e-05,
|
| 151 |
+
"loss": 0.3153,
|
| 152 |
+
"step": 800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.8395061728395061,
|
| 156 |
+
"grad_norm": 0.166303813457489,
|
| 157 |
+
"learning_rate": 8.122025394318091e-05,
|
| 158 |
+
"loss": 0.3166,
|
| 159 |
+
"step": 850
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.8888888888888888,
|
| 163 |
+
"grad_norm": 0.17451900243759155,
|
| 164 |
+
"learning_rate": 7.932421908415695e-05,
|
| 165 |
+
"loss": 0.3156,
|
| 166 |
+
"step": 900
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.9382716049382716,
|
| 170 |
+
"grad_norm": 0.18235546350479126,
|
| 171 |
+
"learning_rate": 7.740215217132219e-05,
|
| 172 |
+
"loss": 0.3155,
|
| 173 |
+
"step": 950
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.9876543209876543,
|
| 177 |
+
"grad_norm": 0.22571995854377747,
|
| 178 |
+
"learning_rate": 7.546562416080285e-05,
|
| 179 |
+
"loss": 0.3185,
|
| 180 |
+
"step": 1000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.9876543209876543,
|
| 184 |
+
"eval_loss": 0.33335721492767334,
|
| 185 |
+
"eval_runtime": 144.7926,
|
| 186 |
+
"eval_samples_per_second": 651.276,
|
| 187 |
+
"eval_steps_per_second": 20.353,
|
| 188 |
+
"step": 1000
|
| 189 |
+
}
|
| 190 |
+
],
|
| 191 |
+
"logging_steps": 50,
|
| 192 |
+
"max_steps": 2024,
|
| 193 |
+
"num_input_tokens_seen": 0,
|
| 194 |
+
"num_train_epochs": 2,
|
| 195 |
+
"save_steps": 500,
|
| 196 |
+
"total_flos": 5930576289792000.0,
|
| 197 |
+
"train_batch_size": 16,
|
| 198 |
+
"trial_name": null,
|
| 199 |
+
"trial_params": null
|
| 200 |
+
}
|
checkpoint-1012/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8c0e5ae1d16b61cb0bbe6e1913a36d352f5b976b4d77e04f450f90733900582
|
| 3 |
+
size 5521
|
checkpoint-2024/config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "time_moe_50m",
|
| 3 |
+
"apply_aux_loss": true,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"TimeMoeForPrediction"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "Maple728/TimeMoE-50M--configuration_time_moe.TimeMoeConfig",
|
| 10 |
+
"AutoModelForCausalLM": "Maple728/TimeMoE-50M--modeling_time_moe.TimeMoeForPrediction"
|
| 11 |
+
},
|
| 12 |
+
"channel_configs": [
|
| 13 |
+
[
|
| 14 |
+
63,
|
| 15 |
+
1,
|
| 16 |
+
1
|
| 17 |
+
],
|
| 18 |
+
[
|
| 19 |
+
6,
|
| 20 |
+
1,
|
| 21 |
+
4
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
5,
|
| 25 |
+
1,
|
| 26 |
+
1
|
| 27 |
+
],
|
| 28 |
+
[
|
| 29 |
+
5,
|
| 30 |
+
1,
|
| 31 |
+
1
|
| 32 |
+
]
|
| 33 |
+
],
|
| 34 |
+
"embedding_hidden_size": 128,
|
| 35 |
+
"hidden_act": "silu",
|
| 36 |
+
"hidden_size": 384,
|
| 37 |
+
"horizon_lengths": [
|
| 38 |
+
1,
|
| 39 |
+
8,
|
| 40 |
+
32,
|
| 41 |
+
64
|
| 42 |
+
],
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"input_size": 42,
|
| 45 |
+
"intermediate_size": 1536,
|
| 46 |
+
"max_position_embeddings": 4096,
|
| 47 |
+
"model_type": "time_moe",
|
| 48 |
+
"num_attention_heads": 12,
|
| 49 |
+
"num_experts": 8,
|
| 50 |
+
"num_experts_per_tok": 2,
|
| 51 |
+
"num_hidden_layers": 12,
|
| 52 |
+
"num_key_value_heads": 12,
|
| 53 |
+
"rms_norm_eps": 1e-06,
|
| 54 |
+
"rope_theta": 10000,
|
| 55 |
+
"router_aux_loss_factor": 0.02,
|
| 56 |
+
"tie_word_embeddings": false,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.40.1",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"use_dense": false
|
| 61 |
+
}
|
checkpoint-2024/generation_config.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.40.1"
|
| 4 |
+
}
|
checkpoint-2024/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b84220e3bd1ccdeaecfa9308850b05ee17de6761e6f83d061265fd0c1623007
|
| 3 |
+
size 484301192
|
checkpoint-2024/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3ef94c5e1a4f5e32ac50ddac26d38fd3f3641155ebfc964f641ab2f2c49ab201
|
| 3 |
+
size 968924920
|
checkpoint-2024/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67a6938a14783e6890baf297362eece64c8df71219ffc210b560d0a690a4c99f
|
| 3 |
+
size 14645
|
checkpoint-2024/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89efdf2689bb6790e533b3e51ce5908662bf122670c4e379468efd264d7232ac
|
| 3 |
+
size 1465
|
checkpoint-2024/trainer_state.json
ADDED
|
@@ -0,0 +1,372 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9990123456790123,
|
| 5 |
+
"eval_steps": 250,
|
| 6 |
+
"global_step": 2024,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0009876543209876543,
|
| 13 |
+
"grad_norm": 1.221229910850525,
|
| 14 |
+
"learning_rate": 9.99999698845987e-05,
|
| 15 |
+
"loss": 0.4907,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.04938271604938271,
|
| 20 |
+
"grad_norm": 0.05035819858312607,
|
| 21 |
+
"learning_rate": 9.99247492630919e-05,
|
| 22 |
+
"loss": 0.3903,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09876543209876543,
|
| 27 |
+
"grad_norm": 0.14429445564746857,
|
| 28 |
+
"learning_rate": 9.969945006624004e-05,
|
| 29 |
+
"loss": 0.3593,
|
| 30 |
+
"step": 100
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.14814814814814814,
|
| 34 |
+
"grad_norm": 0.15117663145065308,
|
| 35 |
+
"learning_rate": 9.932545872389141e-05,
|
| 36 |
+
"loss": 0.3437,
|
| 37 |
+
"step": 150
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.19753086419753085,
|
| 41 |
+
"grad_norm": 0.19646206498146057,
|
| 42 |
+
"learning_rate": 9.880502668597476e-05,
|
| 43 |
+
"loss": 0.3354,
|
| 44 |
+
"step": 200
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.24691358024691357,
|
| 48 |
+
"grad_norm": 0.25732266902923584,
|
| 49 |
+
"learning_rate": 9.814128698403919e-05,
|
| 50 |
+
"loss": 0.3283,
|
| 51 |
+
"step": 250
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.24691358024691357,
|
| 55 |
+
"eval_loss": 0.3421019911766052,
|
| 56 |
+
"eval_runtime": 137.4213,
|
| 57 |
+
"eval_samples_per_second": 686.211,
|
| 58 |
+
"eval_steps_per_second": 21.445,
|
| 59 |
+
"step": 250
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.2962962962962963,
|
| 63 |
+
"grad_norm": 0.12474808096885681,
|
| 64 |
+
"learning_rate": 9.733823537021955e-05,
|
| 65 |
+
"loss": 0.3266,
|
| 66 |
+
"step": 300
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.345679012345679,
|
| 70 |
+
"grad_norm": 0.12876740097999573,
|
| 71 |
+
"learning_rate": 9.640070626257307e-05,
|
| 72 |
+
"loss": 0.3259,
|
| 73 |
+
"step": 350
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.3950617283950617,
|
| 77 |
+
"grad_norm": 0.20461653172969818,
|
| 78 |
+
"learning_rate": 9.533434364159762e-05,
|
| 79 |
+
"loss": 0.3248,
|
| 80 |
+
"step": 400
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.4444444444444444,
|
| 84 |
+
"grad_norm": 0.22041602432727814,
|
| 85 |
+
"learning_rate": 9.414556707313633e-05,
|
| 86 |
+
"loss": 0.3219,
|
| 87 |
+
"step": 450
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.49382716049382713,
|
| 91 |
+
"grad_norm": 0.27202293276786804,
|
| 92 |
+
"learning_rate": 9.284153306221289e-05,
|
| 93 |
+
"loss": 0.3178,
|
| 94 |
+
"step": 500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.49382716049382713,
|
| 98 |
+
"eval_loss": 0.33308419585227966,
|
| 99 |
+
"eval_runtime": 140.6684,
|
| 100 |
+
"eval_samples_per_second": 670.371,
|
| 101 |
+
"eval_steps_per_second": 20.95,
|
| 102 |
+
"step": 500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.5432098765432098,
|
| 106 |
+
"grad_norm": 0.1661626547574997,
|
| 107 |
+
"learning_rate": 9.143009197044932e-05,
|
| 108 |
+
"loss": 0.3211,
|
| 109 |
+
"step": 550
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5925925925925926,
|
| 113 |
+
"grad_norm": 0.14871230721473694,
|
| 114 |
+
"learning_rate": 8.991974075642621e-05,
|
| 115 |
+
"loss": 0.3207,
|
| 116 |
+
"step": 600
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.6419753086419753,
|
| 120 |
+
"grad_norm": 0.17097660899162292,
|
| 121 |
+
"learning_rate": 8.831957182349119e-05,
|
| 122 |
+
"loss": 0.3185,
|
| 123 |
+
"step": 650
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.691358024691358,
|
| 127 |
+
"grad_norm": 0.1582462042570114,
|
| 128 |
+
"learning_rate": 8.663921828295474e-05,
|
| 129 |
+
"loss": 0.3173,
|
| 130 |
+
"step": 700
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.7407407407407407,
|
| 134 |
+
"grad_norm": 0.25067687034606934,
|
| 135 |
+
"learning_rate": 8.488879596219216e-05,
|
| 136 |
+
"loss": 0.3191,
|
| 137 |
+
"step": 750
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.7407407407407407,
|
| 141 |
+
"eval_loss": 0.33356761932373047,
|
| 142 |
+
"eval_runtime": 145.0159,
|
| 143 |
+
"eval_samples_per_second": 650.274,
|
| 144 |
+
"eval_steps_per_second": 20.322,
|
| 145 |
+
"step": 750
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.7901234567901234,
|
| 149 |
+
"grad_norm": 0.1577720046043396,
|
| 150 |
+
"learning_rate": 8.307884250676648e-05,
|
| 151 |
+
"loss": 0.3153,
|
| 152 |
+
"step": 800
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.8395061728395061,
|
| 156 |
+
"grad_norm": 0.166303813457489,
|
| 157 |
+
"learning_rate": 8.122025394318091e-05,
|
| 158 |
+
"loss": 0.3166,
|
| 159 |
+
"step": 850
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.8888888888888888,
|
| 163 |
+
"grad_norm": 0.17451900243759155,
|
| 164 |
+
"learning_rate": 7.932421908415695e-05,
|
| 165 |
+
"loss": 0.3156,
|
| 166 |
+
"step": 900
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.9382716049382716,
|
| 170 |
+
"grad_norm": 0.18235546350479126,
|
| 171 |
+
"learning_rate": 7.740215217132219e-05,
|
| 172 |
+
"loss": 0.3155,
|
| 173 |
+
"step": 950
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.9876543209876543,
|
| 177 |
+
"grad_norm": 0.22571995854377747,
|
| 178 |
+
"learning_rate": 7.546562416080285e-05,
|
| 179 |
+
"loss": 0.3185,
|
| 180 |
+
"step": 1000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.9876543209876543,
|
| 184 |
+
"eval_loss": 0.33335721492767334,
|
| 185 |
+
"eval_runtime": 144.7926,
|
| 186 |
+
"eval_samples_per_second": 651.276,
|
| 187 |
+
"eval_steps_per_second": 20.353,
|
| 188 |
+
"step": 1000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.037037037037037,
|
| 192 |
+
"grad_norm": 0.3201523721218109,
|
| 193 |
+
"learning_rate": 7.35262930653857e-05,
|
| 194 |
+
"loss": 0.3183,
|
| 195 |
+
"step": 1050
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 1.0864197530864197,
|
| 199 |
+
"grad_norm": 0.3311742842197418,
|
| 200 |
+
"learning_rate": 7.159583377259385e-05,
|
| 201 |
+
"loss": 0.3172,
|
| 202 |
+
"step": 1100
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.1358024691358024,
|
| 206 |
+
"grad_norm": 0.2752295434474945,
|
| 207 |
+
"learning_rate": 6.968586776117559e-05,
|
| 208 |
+
"loss": 0.3166,
|
| 209 |
+
"step": 1150
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.1851851851851851,
|
| 213 |
+
"grad_norm": 0.33451607823371887,
|
| 214 |
+
"learning_rate": 6.780789313911721e-05,
|
| 215 |
+
"loss": 0.3166,
|
| 216 |
+
"step": 1200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.2345679012345678,
|
| 220 |
+
"grad_norm": 0.12275319546461105,
|
| 221 |
+
"learning_rate": 6.597321542435513e-05,
|
| 222 |
+
"loss": 0.3155,
|
| 223 |
+
"step": 1250
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.2345679012345678,
|
| 227 |
+
"eval_loss": 0.33012306690216064,
|
| 228 |
+
"eval_runtime": 140.4999,
|
| 229 |
+
"eval_samples_per_second": 671.175,
|
| 230 |
+
"eval_steps_per_second": 20.975,
|
| 231 |
+
"step": 1250
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.2839506172839505,
|
| 235 |
+
"grad_norm": 0.2870579659938812,
|
| 236 |
+
"learning_rate": 6.419287948489126e-05,
|
| 237 |
+
"loss": 0.3154,
|
| 238 |
+
"step": 1300
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.3333333333333333,
|
| 242 |
+
"grad_norm": 0.08633468300104141,
|
| 243 |
+
"learning_rate": 6.247760304803671e-05,
|
| 244 |
+
"loss": 0.3138,
|
| 245 |
+
"step": 1350
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.382716049382716,
|
| 249 |
+
"grad_norm": 0.29040753841400146,
|
| 250 |
+
"learning_rate": 6.083771217906143e-05,
|
| 251 |
+
"loss": 0.3101,
|
| 252 |
+
"step": 1400
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.4320987654320987,
|
| 256 |
+
"grad_norm": 0.2062658816576004,
|
| 257 |
+
"learning_rate": 5.928307911767258e-05,
|
| 258 |
+
"loss": 0.3143,
|
| 259 |
+
"step": 1450
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 1.4814814814814814,
|
| 263 |
+
"grad_norm": 0.23233149945735931,
|
| 264 |
+
"learning_rate": 5.7823062846549435e-05,
|
| 265 |
+
"loss": 0.3152,
|
| 266 |
+
"step": 1500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 1.4814814814814814,
|
| 270 |
+
"eval_loss": 0.33253857493400574,
|
| 271 |
+
"eval_runtime": 139.0406,
|
| 272 |
+
"eval_samples_per_second": 678.219,
|
| 273 |
+
"eval_steps_per_second": 21.195,
|
| 274 |
+
"step": 1500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.5308641975308643,
|
| 278 |
+
"grad_norm": 0.1579607129096985,
|
| 279 |
+
"learning_rate": 5.646645274971599e-05,
|
| 280 |
+
"loss": 0.3171,
|
| 281 |
+
"step": 1550
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.5802469135802468,
|
| 285 |
+
"grad_norm": 0.38029882311820984,
|
| 286 |
+
"learning_rate": 5.522141569993096e-05,
|
| 287 |
+
"loss": 0.3148,
|
| 288 |
+
"step": 1600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.6296296296296298,
|
| 292 |
+
"grad_norm": 0.26864030957221985,
|
| 293 |
+
"learning_rate": 5.409544689363224e-05,
|
| 294 |
+
"loss": 0.3146,
|
| 295 |
+
"step": 1650
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.6790123456790123,
|
| 299 |
+
"grad_norm": 0.41325536370277405,
|
| 300 |
+
"learning_rate": 5.309532472941261e-05,
|
| 301 |
+
"loss": 0.3135,
|
| 302 |
+
"step": 1700
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.7283950617283952,
|
| 306 |
+
"grad_norm": 0.26135507225990295,
|
| 307 |
+
"learning_rate": 5.222707000166053e-05,
|
| 308 |
+
"loss": 0.3127,
|
| 309 |
+
"step": 1750
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 1.7283950617283952,
|
| 313 |
+
"eval_loss": 0.3314434289932251,
|
| 314 |
+
"eval_runtime": 137.4334,
|
| 315 |
+
"eval_samples_per_second": 686.151,
|
| 316 |
+
"eval_steps_per_second": 21.443,
|
| 317 |
+
"step": 1750
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.7777777777777777,
|
| 321 |
+
"grad_norm": 0.35287410020828247,
|
| 322 |
+
"learning_rate": 5.1495909655022966e-05,
|
| 323 |
+
"loss": 0.3127,
|
| 324 |
+
"step": 1800
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.8271604938271606,
|
| 328 |
+
"grad_norm": 0.2525063157081604,
|
| 329 |
+
"learning_rate": 5.090624531789053e-05,
|
| 330 |
+
"loss": 0.3116,
|
| 331 |
+
"step": 1850
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.876543209876543,
|
| 335 |
+
"grad_norm": 0.5191920399665833,
|
| 336 |
+
"learning_rate": 5.0461626804335035e-05,
|
| 337 |
+
"loss": 0.312,
|
| 338 |
+
"step": 1900
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.925925925925926,
|
| 342 |
+
"grad_norm": 0.3505001664161682,
|
| 343 |
+
"learning_rate": 5.0164730744019504e-05,
|
| 344 |
+
"loss": 0.3109,
|
| 345 |
+
"step": 1950
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.9753086419753085,
|
| 349 |
+
"grad_norm": 0.2468252331018448,
|
| 350 |
+
"learning_rate": 5.001734446872988e-05,
|
| 351 |
+
"loss": 0.3124,
|
| 352 |
+
"step": 2000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.9753086419753085,
|
| 356 |
+
"eval_loss": 0.32719919085502625,
|
| 357 |
+
"eval_runtime": 138.3829,
|
| 358 |
+
"eval_samples_per_second": 681.443,
|
| 359 |
+
"eval_steps_per_second": 21.296,
|
| 360 |
+
"step": 2000
|
| 361 |
+
}
|
| 362 |
+
],
|
| 363 |
+
"logging_steps": 50,
|
| 364 |
+
"max_steps": 2024,
|
| 365 |
+
"num_input_tokens_seen": 0,
|
| 366 |
+
"num_train_epochs": 2,
|
| 367 |
+
"save_steps": 500,
|
| 368 |
+
"total_flos": 1.185529522028544e+16,
|
| 369 |
+
"train_batch_size": 16,
|
| 370 |
+
"trial_name": null,
|
| 371 |
+
"trial_params": null
|
| 372 |
+
}
|
checkpoint-2024/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8c0e5ae1d16b61cb0bbe6e1913a36d352f5b976b4d77e04f450f90733900582
|
| 3 |
+
size 5521
|
config.json
CHANGED
|
@@ -1,61 +1,61 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_name_or_path": "time_moe_50m",
|
| 3 |
-
"apply_aux_loss": true,
|
| 4 |
-
"architectures": [
|
| 5 |
-
"TimeMoeForPrediction"
|
| 6 |
-
],
|
| 7 |
-
"attention_dropout": 0.0,
|
| 8 |
-
"auto_map": {
|
| 9 |
-
"AutoConfig": "Maple728/TimeMoE-50M--configuration_time_moe.TimeMoeConfig",
|
| 10 |
-
"AutoModelForCausalLM": "Maple728/TimeMoE-50M--modeling_time_moe.TimeMoeForPrediction"
|
| 11 |
-
},
|
| 12 |
-
"channel_configs": [
|
| 13 |
-
[
|
| 14 |
-
63,
|
| 15 |
-
1,
|
| 16 |
-
1
|
| 17 |
-
],
|
| 18 |
-
[
|
| 19 |
-
6,
|
| 20 |
-
1,
|
| 21 |
-
4
|
| 22 |
-
],
|
| 23 |
-
[
|
| 24 |
-
5,
|
| 25 |
-
1,
|
| 26 |
-
1
|
| 27 |
-
],
|
| 28 |
-
[
|
| 29 |
-
5,
|
| 30 |
-
1,
|
| 31 |
-
1
|
| 32 |
-
]
|
| 33 |
-
],
|
| 34 |
-
"embedding_hidden_size": 128,
|
| 35 |
-
"hidden_act": "silu",
|
| 36 |
-
"hidden_size": 384,
|
| 37 |
-
"horizon_lengths": [
|
| 38 |
-
1,
|
| 39 |
-
8,
|
| 40 |
-
32,
|
| 41 |
-
64
|
| 42 |
-
],
|
| 43 |
-
"initializer_range": 0.02,
|
| 44 |
-
"input_size": 42,
|
| 45 |
-
"intermediate_size": 1536,
|
| 46 |
-
"max_position_embeddings": 4096,
|
| 47 |
-
"model_type": "time_moe",
|
| 48 |
-
"num_attention_heads": 12,
|
| 49 |
-
"num_experts": 8,
|
| 50 |
-
"num_experts_per_tok": 2,
|
| 51 |
-
"num_hidden_layers": 12,
|
| 52 |
-
"num_key_value_heads": 12,
|
| 53 |
-
"rms_norm_eps": 1e-06,
|
| 54 |
-
"rope_theta": 10000,
|
| 55 |
-
"router_aux_loss_factor": 0.02,
|
| 56 |
-
"tie_word_embeddings": false,
|
| 57 |
-
"torch_dtype": "float32",
|
| 58 |
-
"transformers_version": "4.40.1",
|
| 59 |
-
"use_cache": true,
|
| 60 |
-
"use_dense": false
|
| 61 |
-
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "time_moe_50m",
|
| 3 |
+
"apply_aux_loss": true,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"TimeMoeForPrediction"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"auto_map": {
|
| 9 |
+
"AutoConfig": "Maple728/TimeMoE-50M--configuration_time_moe.TimeMoeConfig",
|
| 10 |
+
"AutoModelForCausalLM": "Maple728/TimeMoE-50M--modeling_time_moe.TimeMoeForPrediction"
|
| 11 |
+
},
|
| 12 |
+
"channel_configs": [
|
| 13 |
+
[
|
| 14 |
+
63,
|
| 15 |
+
1,
|
| 16 |
+
1
|
| 17 |
+
],
|
| 18 |
+
[
|
| 19 |
+
6,
|
| 20 |
+
1,
|
| 21 |
+
4
|
| 22 |
+
],
|
| 23 |
+
[
|
| 24 |
+
5,
|
| 25 |
+
1,
|
| 26 |
+
1
|
| 27 |
+
],
|
| 28 |
+
[
|
| 29 |
+
5,
|
| 30 |
+
1,
|
| 31 |
+
1
|
| 32 |
+
]
|
| 33 |
+
],
|
| 34 |
+
"embedding_hidden_size": 128,
|
| 35 |
+
"hidden_act": "silu",
|
| 36 |
+
"hidden_size": 384,
|
| 37 |
+
"horizon_lengths": [
|
| 38 |
+
1,
|
| 39 |
+
8,
|
| 40 |
+
32,
|
| 41 |
+
64
|
| 42 |
+
],
|
| 43 |
+
"initializer_range": 0.02,
|
| 44 |
+
"input_size": 42,
|
| 45 |
+
"intermediate_size": 1536,
|
| 46 |
+
"max_position_embeddings": 4096,
|
| 47 |
+
"model_type": "time_moe",
|
| 48 |
+
"num_attention_heads": 12,
|
| 49 |
+
"num_experts": 8,
|
| 50 |
+
"num_experts_per_tok": 2,
|
| 51 |
+
"num_hidden_layers": 12,
|
| 52 |
+
"num_key_value_heads": 12,
|
| 53 |
+
"rms_norm_eps": 1e-06,
|
| 54 |
+
"rope_theta": 10000,
|
| 55 |
+
"router_aux_loss_factor": 0.02,
|
| 56 |
+
"tie_word_embeddings": false,
|
| 57 |
+
"torch_dtype": "float32",
|
| 58 |
+
"transformers_version": "4.40.1",
|
| 59 |
+
"use_cache": true,
|
| 60 |
+
"use_dense": false
|
| 61 |
+
}
|
generation_config.json
CHANGED
|
@@ -1,4 +1,4 @@
|
|
| 1 |
-
{
|
| 2 |
-
"_from_model_config": true,
|
| 3 |
-
"transformers_version": "4.40.1"
|
| 4 |
-
}
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"transformers_version": "4.40.1"
|
| 4 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 484301192
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5b84220e3bd1ccdeaecfa9308850b05ee17de6761e6f83d061265fd0c1623007
|
| 3 |
size 484301192
|
tb_logs/events.out.tfevents.1760423030.luyao1.2719331.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cb25d9dda7f93a6f276b76a0b21b6c59b62fd6c2a847b11fab0dca71de14368d
|
| 3 |
+
size 16427
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8c0e5ae1d16b61cb0bbe6e1913a36d352f5b976b4d77e04f450f90733900582
|
| 3 |
+
size 5521
|