Upload 9 files
Browse files- config.json +31 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- trainer_state.json +440 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "../MistralModels/models/Mixtral-8x7B-v0.1-small-4096",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MixtralForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 768,
|
| 13 |
+
"max_position_embeddings": 512,
|
| 14 |
+
"model_type": "mixtral",
|
| 15 |
+
"num_attention_heads": 8,
|
| 16 |
+
"num_experts_per_tok": 1,
|
| 17 |
+
"num_hidden_layers": 8,
|
| 18 |
+
"num_key_value_heads": 8,
|
| 19 |
+
"num_local_experts": 8,
|
| 20 |
+
"output_router_logits": false,
|
| 21 |
+
"rms_norm_eps": 1e-05,
|
| 22 |
+
"rope_theta": 1000000.0,
|
| 23 |
+
"router_aux_loss_coef": 0.02,
|
| 24 |
+
"router_jitter_noise": 0.0,
|
| 25 |
+
"sliding_window": null,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
+
"torch_dtype": "bfloat16",
|
| 28 |
+
"transformers_version": "4.43.3",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 4096
|
| 31 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"transformers_version": "4.43.3"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea39442c1430e42ceede156f881b4666f09a448763c5f76db920e377f1423356
|
| 3 |
+
size 276979168
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e3e5d946241df2516b06d7074d8779088eae7607173ad780df56583910a9589b
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:40e3b79060164408b0824b8881020cfa154eba2896f4e49f58b61e6fe830a547
|
| 3 |
+
size 1064
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"tokenizer_class": "PreTrainedTokenizerFast", "unk_token": "[UNK]", "cls_token": "[CLS]", "sep_token": "[SEP]", "pad_token": "[PAD]", "mask_token": "[MASK]"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,440 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 6.83237886428833,
|
| 3 |
+
"best_model_checkpoint": "./results/models/checkpoint-25230",
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 25230,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11890606420927467,
|
| 13 |
+
"grad_norm": 0.4375,
|
| 14 |
+
"learning_rate": 0.003990487514863258,
|
| 15 |
+
"loss": 6.8648,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.23781212841854935,
|
| 20 |
+
"grad_norm": 3.78125,
|
| 21 |
+
"learning_rate": 0.003980975029726516,
|
| 22 |
+
"loss": 6.8372,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.356718192627824,
|
| 27 |
+
"grad_norm": 3.21875,
|
| 28 |
+
"learning_rate": 0.003971462544589774,
|
| 29 |
+
"loss": 6.872,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4756242568370987,
|
| 34 |
+
"grad_norm": 1.8359375,
|
| 35 |
+
"learning_rate": 0.0039619500594530324,
|
| 36 |
+
"loss": 6.8835,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5945303210463734,
|
| 41 |
+
"grad_norm": 15.25,
|
| 42 |
+
"learning_rate": 0.00395243757431629,
|
| 43 |
+
"loss": 6.8784,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.713436385255648,
|
| 48 |
+
"grad_norm": 2.328125,
|
| 49 |
+
"learning_rate": 0.003942925089179548,
|
| 50 |
+
"loss": 6.8964,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.8323424494649228,
|
| 55 |
+
"grad_norm": 17.0,
|
| 56 |
+
"learning_rate": 0.003933412604042806,
|
| 57 |
+
"loss": 6.8779,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.9512485136741974,
|
| 62 |
+
"grad_norm": 0.98828125,
|
| 63 |
+
"learning_rate": 0.003923900118906064,
|
| 64 |
+
"loss": 6.8788,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.0,
|
| 69 |
+
"eval_loss": 6.88191556930542,
|
| 70 |
+
"eval_runtime": 26.2747,
|
| 71 |
+
"eval_samples_per_second": 76.119,
|
| 72 |
+
"eval_steps_per_second": 0.609,
|
| 73 |
+
"step": 4205
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.070154577883472,
|
| 77 |
+
"grad_norm": 5.1875,
|
| 78 |
+
"learning_rate": 0.003914387633769323,
|
| 79 |
+
"loss": 6.8778,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.1890606420927468,
|
| 84 |
+
"grad_norm": 2.296875,
|
| 85 |
+
"learning_rate": 0.00390487514863258,
|
| 86 |
+
"loss": 6.8945,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.3079667063020213,
|
| 91 |
+
"grad_norm": 2.625,
|
| 92 |
+
"learning_rate": 0.0038953626634958384,
|
| 93 |
+
"loss": 6.8862,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.426872770511296,
|
| 98 |
+
"grad_norm": 0.71875,
|
| 99 |
+
"learning_rate": 0.0038858501783590963,
|
| 100 |
+
"loss": 6.8846,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.5457788347205708,
|
| 105 |
+
"grad_norm": 1.3046875,
|
| 106 |
+
"learning_rate": 0.0038763376932223546,
|
| 107 |
+
"loss": 6.8757,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.6646848989298455,
|
| 112 |
+
"grad_norm": 1.0078125,
|
| 113 |
+
"learning_rate": 0.0038668252080856125,
|
| 114 |
+
"loss": 6.8802,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.7835909631391202,
|
| 119 |
+
"grad_norm": 1.5625,
|
| 120 |
+
"learning_rate": 0.0038573127229488704,
|
| 121 |
+
"loss": 6.8663,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.9024970273483948,
|
| 126 |
+
"grad_norm": 12.75,
|
| 127 |
+
"learning_rate": 0.0038478002378121287,
|
| 128 |
+
"loss": 6.8694,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 2.0,
|
| 133 |
+
"eval_loss": 6.880330562591553,
|
| 134 |
+
"eval_runtime": 23.7249,
|
| 135 |
+
"eval_samples_per_second": 84.299,
|
| 136 |
+
"eval_steps_per_second": 0.674,
|
| 137 |
+
"step": 8410
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 2.0214030915576693,
|
| 141 |
+
"grad_norm": 14.375,
|
| 142 |
+
"learning_rate": 0.003838287752675386,
|
| 143 |
+
"loss": 6.8764,
|
| 144 |
+
"step": 8500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.140309155766944,
|
| 148 |
+
"grad_norm": 5.09375,
|
| 149 |
+
"learning_rate": 0.0038287752675386444,
|
| 150 |
+
"loss": 6.8742,
|
| 151 |
+
"step": 9000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.2592152199762188,
|
| 155 |
+
"grad_norm": 17.125,
|
| 156 |
+
"learning_rate": 0.0038192627824019027,
|
| 157 |
+
"loss": 6.8829,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.3781212841854935,
|
| 162 |
+
"grad_norm": 2.171875,
|
| 163 |
+
"learning_rate": 0.0038097502972651606,
|
| 164 |
+
"loss": 6.8883,
|
| 165 |
+
"step": 10000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.4970273483947683,
|
| 169 |
+
"grad_norm": 1.0,
|
| 170 |
+
"learning_rate": 0.0038002378121284185,
|
| 171 |
+
"loss": 6.8851,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.6159334126040426,
|
| 176 |
+
"grad_norm": 0.84375,
|
| 177 |
+
"learning_rate": 0.003790725326991677,
|
| 178 |
+
"loss": 6.8849,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.7348394768133177,
|
| 183 |
+
"grad_norm": 2.0625,
|
| 184 |
+
"learning_rate": 0.0037812128418549347,
|
| 185 |
+
"loss": 6.8784,
|
| 186 |
+
"step": 11500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.853745541022592,
|
| 190 |
+
"grad_norm": 5.78125,
|
| 191 |
+
"learning_rate": 0.003771700356718193,
|
| 192 |
+
"loss": 6.8769,
|
| 193 |
+
"step": 12000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.972651605231867,
|
| 197 |
+
"grad_norm": 5.125,
|
| 198 |
+
"learning_rate": 0.0037621878715814504,
|
| 199 |
+
"loss": 6.876,
|
| 200 |
+
"step": 12500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.0,
|
| 204 |
+
"eval_loss": 6.865922451019287,
|
| 205 |
+
"eval_runtime": 20.8083,
|
| 206 |
+
"eval_samples_per_second": 96.115,
|
| 207 |
+
"eval_steps_per_second": 0.769,
|
| 208 |
+
"step": 12615
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 3.0915576694411415,
|
| 212 |
+
"grad_norm": 3.8125,
|
| 213 |
+
"learning_rate": 0.0037526753864447087,
|
| 214 |
+
"loss": 6.8658,
|
| 215 |
+
"step": 13000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 3.2104637336504163,
|
| 219 |
+
"grad_norm": 6.21875,
|
| 220 |
+
"learning_rate": 0.003743162901307967,
|
| 221 |
+
"loss": 6.8673,
|
| 222 |
+
"step": 13500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.329369797859691,
|
| 226 |
+
"grad_norm": 3.71875,
|
| 227 |
+
"learning_rate": 0.003733650416171225,
|
| 228 |
+
"loss": 6.8629,
|
| 229 |
+
"step": 14000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.4482758620689653,
|
| 233 |
+
"grad_norm": 4.5,
|
| 234 |
+
"learning_rate": 0.0037241379310344828,
|
| 235 |
+
"loss": 6.866,
|
| 236 |
+
"step": 14500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.56718192627824,
|
| 240 |
+
"grad_norm": 2.546875,
|
| 241 |
+
"learning_rate": 0.003714625445897741,
|
| 242 |
+
"loss": 6.8627,
|
| 243 |
+
"step": 15000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.686087990487515,
|
| 247 |
+
"grad_norm": 7.0625,
|
| 248 |
+
"learning_rate": 0.003705112960760999,
|
| 249 |
+
"loss": 6.859,
|
| 250 |
+
"step": 15500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.8049940546967895,
|
| 254 |
+
"grad_norm": 42.0,
|
| 255 |
+
"learning_rate": 0.003695600475624257,
|
| 256 |
+
"loss": 6.852,
|
| 257 |
+
"step": 16000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.9239001189060643,
|
| 261 |
+
"grad_norm": 24.75,
|
| 262 |
+
"learning_rate": 0.0036860879904875147,
|
| 263 |
+
"loss": 6.8541,
|
| 264 |
+
"step": 16500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 4.0,
|
| 268 |
+
"eval_loss": 6.84573221206665,
|
| 269 |
+
"eval_runtime": 22.0613,
|
| 270 |
+
"eval_samples_per_second": 90.656,
|
| 271 |
+
"eval_steps_per_second": 0.725,
|
| 272 |
+
"step": 16820
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.042806183115339,
|
| 276 |
+
"grad_norm": 19.75,
|
| 277 |
+
"learning_rate": 0.003676575505350773,
|
| 278 |
+
"loss": 6.85,
|
| 279 |
+
"step": 17000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.161712247324614,
|
| 283 |
+
"grad_norm": 18.0,
|
| 284 |
+
"learning_rate": 0.0036670630202140313,
|
| 285 |
+
"loss": 6.8548,
|
| 286 |
+
"step": 17500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 4.280618311533888,
|
| 290 |
+
"grad_norm": 1.7421875,
|
| 291 |
+
"learning_rate": 0.0036575505350772888,
|
| 292 |
+
"loss": 6.8497,
|
| 293 |
+
"step": 18000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.399524375743163,
|
| 297 |
+
"grad_norm": 11.25,
|
| 298 |
+
"learning_rate": 0.003648038049940547,
|
| 299 |
+
"loss": 6.8469,
|
| 300 |
+
"step": 18500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.5184304399524375,
|
| 304 |
+
"grad_norm": 181.0,
|
| 305 |
+
"learning_rate": 0.003638525564803805,
|
| 306 |
+
"loss": 6.8481,
|
| 307 |
+
"step": 19000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.637336504161713,
|
| 311 |
+
"grad_norm": 13.4375,
|
| 312 |
+
"learning_rate": 0.0036290130796670633,
|
| 313 |
+
"loss": 6.854,
|
| 314 |
+
"step": 19500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.756242568370987,
|
| 318 |
+
"grad_norm": 19.75,
|
| 319 |
+
"learning_rate": 0.003619500594530321,
|
| 320 |
+
"loss": 6.848,
|
| 321 |
+
"step": 20000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.875148632580261,
|
| 325 |
+
"grad_norm": 3.71875,
|
| 326 |
+
"learning_rate": 0.003609988109393579,
|
| 327 |
+
"loss": 6.8439,
|
| 328 |
+
"step": 20500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 4.9940546967895365,
|
| 332 |
+
"grad_norm": 3.78125,
|
| 333 |
+
"learning_rate": 0.0036004756242568373,
|
| 334 |
+
"loss": 6.8516,
|
| 335 |
+
"step": 21000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 5.0,
|
| 339 |
+
"eval_loss": 6.845646381378174,
|
| 340 |
+
"eval_runtime": 24.3842,
|
| 341 |
+
"eval_samples_per_second": 82.02,
|
| 342 |
+
"eval_steps_per_second": 0.656,
|
| 343 |
+
"step": 21025
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 5.112960760998811,
|
| 347 |
+
"grad_norm": 43.0,
|
| 348 |
+
"learning_rate": 0.003590963139120095,
|
| 349 |
+
"loss": 6.8426,
|
| 350 |
+
"step": 21500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.231866825208086,
|
| 354 |
+
"grad_norm": 14.5,
|
| 355 |
+
"learning_rate": 0.003581450653983353,
|
| 356 |
+
"loss": 6.8388,
|
| 357 |
+
"step": 22000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 5.35077288941736,
|
| 361 |
+
"grad_norm": 23.25,
|
| 362 |
+
"learning_rate": 0.0035719381688466114,
|
| 363 |
+
"loss": 6.843,
|
| 364 |
+
"step": 22500
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 5.469678953626635,
|
| 368 |
+
"grad_norm": 3.546875,
|
| 369 |
+
"learning_rate": 0.0035624256837098692,
|
| 370 |
+
"loss": 6.8391,
|
| 371 |
+
"step": 23000
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 5.58858501783591,
|
| 375 |
+
"grad_norm": 4.03125,
|
| 376 |
+
"learning_rate": 0.003552913198573127,
|
| 377 |
+
"loss": 6.838,
|
| 378 |
+
"step": 23500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.707491082045184,
|
| 382 |
+
"grad_norm": 2.953125,
|
| 383 |
+
"learning_rate": 0.0035434007134363854,
|
| 384 |
+
"loss": 6.8373,
|
| 385 |
+
"step": 24000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 5.826397146254459,
|
| 389 |
+
"grad_norm": 16.25,
|
| 390 |
+
"learning_rate": 0.0035338882282996433,
|
| 391 |
+
"loss": 6.8372,
|
| 392 |
+
"step": 24500
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 5.945303210463734,
|
| 396 |
+
"grad_norm": 49.0,
|
| 397 |
+
"learning_rate": 0.0035243757431629016,
|
| 398 |
+
"loss": 6.8363,
|
| 399 |
+
"step": 25000
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 6.0,
|
| 403 |
+
"eval_loss": 6.83237886428833,
|
| 404 |
+
"eval_runtime": 24.8165,
|
| 405 |
+
"eval_samples_per_second": 80.591,
|
| 406 |
+
"eval_steps_per_second": 0.645,
|
| 407 |
+
"step": 25230
|
| 408 |
+
}
|
| 409 |
+
],
|
| 410 |
+
"logging_steps": 500,
|
| 411 |
+
"max_steps": 210250,
|
| 412 |
+
"num_input_tokens_seen": 0,
|
| 413 |
+
"num_train_epochs": 50,
|
| 414 |
+
"save_steps": 500,
|
| 415 |
+
"stateful_callbacks": {
|
| 416 |
+
"EarlyStoppingCallback": {
|
| 417 |
+
"args": {
|
| 418 |
+
"early_stopping_patience": 3,
|
| 419 |
+
"early_stopping_threshold": 0.0
|
| 420 |
+
},
|
| 421 |
+
"attributes": {
|
| 422 |
+
"early_stopping_patience_counter": 0
|
| 423 |
+
}
|
| 424 |
+
},
|
| 425 |
+
"TrainerControl": {
|
| 426 |
+
"args": {
|
| 427 |
+
"should_epoch_stop": false,
|
| 428 |
+
"should_evaluate": false,
|
| 429 |
+
"should_log": false,
|
| 430 |
+
"should_save": true,
|
| 431 |
+
"should_training_stop": false
|
| 432 |
+
},
|
| 433 |
+
"attributes": {}
|
| 434 |
+
}
|
| 435 |
+
},
|
| 436 |
+
"total_flos": 2.337350063694092e+18,
|
| 437 |
+
"train_batch_size": 128,
|
| 438 |
+
"trial_name": null,
|
| 439 |
+
"trial_params": null
|
| 440 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66c96da9fdd0257947b5effe2dcc43a3152589b5790be503e89e0f9e61204eb2
|
| 3 |
+
size 5112
|