Commit
•
ec393f3
1
Parent(s):
525fe58
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 +397 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "data/models/Mixtral-8x7B-v0.3-dna",
|
| 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": 2688,
|
| 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": "4.41.0",
|
| 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": "4.41.0"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:78eac1f41dbf5338f72a78f0ce36503839b3a0d7ad58e4a829ee3976cdc0f382
|
| 3 |
+
size 843210512
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d245e05e72192c132e0f2edb6fdcae0c578c890f0fe912f17ec7b0bba2d38cc3
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:91c60f5573968c0ff71f3b190220d6ef5d099fd98edf20ed90e201815de74a76
|
| 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 +1,397 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 5.346867561340332,
|
| 3 |
+
"best_model_checkpoint": "./results/models/mistral-dna/checkpoint-22675",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 22675,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11025358324145534,
|
| 13 |
+
"grad_norm": 0.142578125,
|
| 14 |
+
"learning_rate": 0.0003991179713340684,
|
| 15 |
+
"loss": 6.6131,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.2205071664829107,
|
| 20 |
+
"grad_norm": 0.1298828125,
|
| 21 |
+
"learning_rate": 0.00039823594266813673,
|
| 22 |
+
"loss": 6.0212,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.33076074972436603,
|
| 27 |
+
"grad_norm": 0.12255859375,
|
| 28 |
+
"learning_rate": 0.00039735391400220506,
|
| 29 |
+
"loss": 5.8931,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.4410143329658214,
|
| 34 |
+
"grad_norm": 0.11572265625,
|
| 35 |
+
"learning_rate": 0.00039647188533627344,
|
| 36 |
+
"loss": 5.822,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.5512679162072768,
|
| 41 |
+
"grad_norm": 0.125,
|
| 42 |
+
"learning_rate": 0.0003955898566703418,
|
| 43 |
+
"loss": 5.7678,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.6615214994487321,
|
| 48 |
+
"grad_norm": 0.119140625,
|
| 49 |
+
"learning_rate": 0.00039470782800441015,
|
| 50 |
+
"loss": 5.7204,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.7717750826901875,
|
| 55 |
+
"grad_norm": 0.12255859375,
|
| 56 |
+
"learning_rate": 0.00039382579933847854,
|
| 57 |
+
"loss": 5.7025,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.8820286659316428,
|
| 62 |
+
"grad_norm": 0.1376953125,
|
| 63 |
+
"learning_rate": 0.00039294377067254687,
|
| 64 |
+
"loss": 5.6733,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.9922822491730982,
|
| 69 |
+
"grad_norm": 0.1328125,
|
| 70 |
+
"learning_rate": 0.00039206174200661525,
|
| 71 |
+
"loss": 5.6491,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.0,
|
| 76 |
+
"eval_loss": 5.551197528839111,
|
| 77 |
+
"eval_runtime": 3.3539,
|
| 78 |
+
"eval_samples_per_second": 86.765,
|
| 79 |
+
"eval_steps_per_second": 1.491,
|
| 80 |
+
"step": 4535
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.1025358324145536,
|
| 84 |
+
"grad_norm": 0.126953125,
|
| 85 |
+
"learning_rate": 0.0003911797133406836,
|
| 86 |
+
"loss": 5.6279,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 1.2127894156560088,
|
| 91 |
+
"grad_norm": 0.1435546875,
|
| 92 |
+
"learning_rate": 0.00039029768467475196,
|
| 93 |
+
"loss": 5.6068,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 1.3230429988974641,
|
| 98 |
+
"grad_norm": 0.1455078125,
|
| 99 |
+
"learning_rate": 0.0003894156560088203,
|
| 100 |
+
"loss": 5.6068,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.4332965821389196,
|
| 105 |
+
"grad_norm": 0.130859375,
|
| 106 |
+
"learning_rate": 0.00038853362734288867,
|
| 107 |
+
"loss": 5.5658,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.543550165380375,
|
| 112 |
+
"grad_norm": 0.1318359375,
|
| 113 |
+
"learning_rate": 0.000387651598676957,
|
| 114 |
+
"loss": 5.5665,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.6538037486218302,
|
| 119 |
+
"grad_norm": 0.138671875,
|
| 120 |
+
"learning_rate": 0.0003867695700110254,
|
| 121 |
+
"loss": 5.5565,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.7640573318632855,
|
| 126 |
+
"grad_norm": 0.1376953125,
|
| 127 |
+
"learning_rate": 0.00038588754134509377,
|
| 128 |
+
"loss": 5.5477,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 1.8743109151047408,
|
| 133 |
+
"grad_norm": 0.1298828125,
|
| 134 |
+
"learning_rate": 0.0003850055126791621,
|
| 135 |
+
"loss": 5.5295,
|
| 136 |
+
"step": 8500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 1.9845644983461963,
|
| 140 |
+
"grad_norm": 0.140625,
|
| 141 |
+
"learning_rate": 0.0003841234840132304,
|
| 142 |
+
"loss": 5.5208,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 2.0,
|
| 147 |
+
"eval_loss": 5.451062202453613,
|
| 148 |
+
"eval_runtime": 3.2969,
|
| 149 |
+
"eval_samples_per_second": 88.264,
|
| 150 |
+
"eval_steps_per_second": 1.517,
|
| 151 |
+
"step": 9070
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 2.0948180815876514,
|
| 155 |
+
"grad_norm": 0.138671875,
|
| 156 |
+
"learning_rate": 0.0003832414553472988,
|
| 157 |
+
"loss": 5.5017,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.205071664829107,
|
| 162 |
+
"grad_norm": 0.142578125,
|
| 163 |
+
"learning_rate": 0.0003823594266813672,
|
| 164 |
+
"loss": 5.4918,
|
| 165 |
+
"step": 10000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.3153252480705624,
|
| 169 |
+
"grad_norm": 0.1357421875,
|
| 170 |
+
"learning_rate": 0.0003814773980154355,
|
| 171 |
+
"loss": 5.4806,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.4255788313120177,
|
| 176 |
+
"grad_norm": 0.13671875,
|
| 177 |
+
"learning_rate": 0.00038059536934950385,
|
| 178 |
+
"loss": 5.4877,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.535832414553473,
|
| 183 |
+
"grad_norm": 0.138671875,
|
| 184 |
+
"learning_rate": 0.00037971334068357223,
|
| 185 |
+
"loss": 5.4696,
|
| 186 |
+
"step": 11500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.6460859977949283,
|
| 190 |
+
"grad_norm": 0.15625,
|
| 191 |
+
"learning_rate": 0.0003788313120176406,
|
| 192 |
+
"loss": 5.4781,
|
| 193 |
+
"step": 12000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.7563395810363835,
|
| 197 |
+
"grad_norm": 0.140625,
|
| 198 |
+
"learning_rate": 0.00037794928335170894,
|
| 199 |
+
"loss": 5.4748,
|
| 200 |
+
"step": 12500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 2.8665931642778393,
|
| 204 |
+
"grad_norm": 0.1591796875,
|
| 205 |
+
"learning_rate": 0.0003770672546857773,
|
| 206 |
+
"loss": 5.4608,
|
| 207 |
+
"step": 13000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.9768467475192946,
|
| 211 |
+
"grad_norm": 0.1923828125,
|
| 212 |
+
"learning_rate": 0.00037618522601984565,
|
| 213 |
+
"loss": 5.453,
|
| 214 |
+
"step": 13500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 3.0,
|
| 218 |
+
"eval_loss": 5.399576663970947,
|
| 219 |
+
"eval_runtime": 3.4053,
|
| 220 |
+
"eval_samples_per_second": 85.455,
|
| 221 |
+
"eval_steps_per_second": 1.468,
|
| 222 |
+
"step": 13605
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.08710033076075,
|
| 226 |
+
"grad_norm": 0.1865234375,
|
| 227 |
+
"learning_rate": 0.00037530319735391404,
|
| 228 |
+
"loss": 5.4278,
|
| 229 |
+
"step": 14000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 3.197353914002205,
|
| 233 |
+
"grad_norm": 0.158203125,
|
| 234 |
+
"learning_rate": 0.00037442116868798236,
|
| 235 |
+
"loss": 5.4226,
|
| 236 |
+
"step": 14500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.3076074972436604,
|
| 240 |
+
"grad_norm": 0.1484375,
|
| 241 |
+
"learning_rate": 0.00037353914002205075,
|
| 242 |
+
"loss": 5.4334,
|
| 243 |
+
"step": 15000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.4178610804851157,
|
| 247 |
+
"grad_norm": 0.162109375,
|
| 248 |
+
"learning_rate": 0.0003726571113561191,
|
| 249 |
+
"loss": 5.4321,
|
| 250 |
+
"step": 15500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.528114663726571,
|
| 254 |
+
"grad_norm": 0.1572265625,
|
| 255 |
+
"learning_rate": 0.00037177508269018746,
|
| 256 |
+
"loss": 5.4277,
|
| 257 |
+
"step": 16000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.6383682469680263,
|
| 261 |
+
"grad_norm": 0.1591796875,
|
| 262 |
+
"learning_rate": 0.0003708930540242558,
|
| 263 |
+
"loss": 5.4193,
|
| 264 |
+
"step": 16500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.7486218302094816,
|
| 268 |
+
"grad_norm": 0.1611328125,
|
| 269 |
+
"learning_rate": 0.00037001102535832417,
|
| 270 |
+
"loss": 5.4226,
|
| 271 |
+
"step": 17000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 3.8588754134509373,
|
| 275 |
+
"grad_norm": 0.154296875,
|
| 276 |
+
"learning_rate": 0.00036912899669239255,
|
| 277 |
+
"loss": 5.4214,
|
| 278 |
+
"step": 17500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 3.9691289966923926,
|
| 282 |
+
"grad_norm": 0.1533203125,
|
| 283 |
+
"learning_rate": 0.0003682469680264609,
|
| 284 |
+
"loss": 5.4218,
|
| 285 |
+
"step": 18000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.0,
|
| 289 |
+
"eval_loss": 5.367885112762451,
|
| 290 |
+
"eval_runtime": 3.6662,
|
| 291 |
+
"eval_samples_per_second": 79.373,
|
| 292 |
+
"eval_steps_per_second": 1.364,
|
| 293 |
+
"step": 18140
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 4.0793825799338475,
|
| 297 |
+
"grad_norm": 0.1748046875,
|
| 298 |
+
"learning_rate": 0.0003673649393605292,
|
| 299 |
+
"loss": 5.4067,
|
| 300 |
+
"step": 18500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 4.189636163175303,
|
| 304 |
+
"grad_norm": 0.1572265625,
|
| 305 |
+
"learning_rate": 0.0003664829106945976,
|
| 306 |
+
"loss": 5.3963,
|
| 307 |
+
"step": 19000
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 4.299889746416759,
|
| 311 |
+
"grad_norm": 0.158203125,
|
| 312 |
+
"learning_rate": 0.000365600882028666,
|
| 313 |
+
"loss": 5.3929,
|
| 314 |
+
"step": 19500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.410143329658214,
|
| 318 |
+
"grad_norm": 0.1630859375,
|
| 319 |
+
"learning_rate": 0.0003647188533627343,
|
| 320 |
+
"loss": 5.3888,
|
| 321 |
+
"step": 20000
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.5203969128996695,
|
| 325 |
+
"grad_norm": 0.16015625,
|
| 326 |
+
"learning_rate": 0.00036383682469680263,
|
| 327 |
+
"loss": 5.3982,
|
| 328 |
+
"step": 20500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 4.630650496141125,
|
| 332 |
+
"grad_norm": 0.173828125,
|
| 333 |
+
"learning_rate": 0.000362954796030871,
|
| 334 |
+
"loss": 5.3754,
|
| 335 |
+
"step": 21000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 4.74090407938258,
|
| 339 |
+
"grad_norm": 0.1591796875,
|
| 340 |
+
"learning_rate": 0.0003620727673649394,
|
| 341 |
+
"loss": 5.3854,
|
| 342 |
+
"step": 21500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 4.851157662624035,
|
| 346 |
+
"grad_norm": 0.1748046875,
|
| 347 |
+
"learning_rate": 0.00036119073869900773,
|
| 348 |
+
"loss": 5.3876,
|
| 349 |
+
"step": 22000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.961411245865491,
|
| 353 |
+
"grad_norm": 0.1767578125,
|
| 354 |
+
"learning_rate": 0.0003603087100330761,
|
| 355 |
+
"loss": 5.3899,
|
| 356 |
+
"step": 22500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 5.0,
|
| 360 |
+
"eval_loss": 5.346867561340332,
|
| 361 |
+
"eval_runtime": 3.7974,
|
| 362 |
+
"eval_samples_per_second": 76.63,
|
| 363 |
+
"eval_steps_per_second": 1.317,
|
| 364 |
+
"step": 22675
|
| 365 |
+
}
|
| 366 |
+
],
|
| 367 |
+
"logging_steps": 500,
|
| 368 |
+
"max_steps": 226750,
|
| 369 |
+
"num_input_tokens_seen": 0,
|
| 370 |
+
"num_train_epochs": 50,
|
| 371 |
+
"save_steps": 500,
|
| 372 |
+
"stateful_callbacks": {
|
| 373 |
+
"EarlyStoppingCallback": {
|
| 374 |
+
"args": {
|
| 375 |
+
"early_stopping_patience": 3,
|
| 376 |
+
"early_stopping_threshold": 0.0
|
| 377 |
+
},
|
| 378 |
+
"attributes": {
|
| 379 |
+
"early_stopping_patience_counter": 0
|
| 380 |
+
}
|
| 381 |
+
},
|
| 382 |
+
"TrainerControl": {
|
| 383 |
+
"args": {
|
| 384 |
+
"should_epoch_stop": false,
|
| 385 |
+
"should_evaluate": false,
|
| 386 |
+
"should_log": false,
|
| 387 |
+
"should_save": true,
|
| 388 |
+
"should_training_stop": false
|
| 389 |
+
},
|
| 390 |
+
"attributes": {}
|
| 391 |
+
}
|
| 392 |
+
},
|
| 393 |
+
"total_flos": 7.976338069294799e+18,
|
| 394 |
+
"train_batch_size": 64,
|
| 395 |
+
"trial_name": null,
|
| 396 |
+
"trial_params": null
|
| 397 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d1a27c36f0c14e112e39bb643e41fa635841153000658e93c47f76465b9c6c0
|
| 3 |
+
size 5048
|