Training in progress, step 1300, checkpoint
Browse files- last-checkpoint/config.json +27 -0
- last-checkpoint/generation_config.json +7 -0
- last-checkpoint/model.safetensors +3 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/trainer_state.json +606 -0
- last-checkpoint/training_args.bin +3 -0
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MistralForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"bos_token_id": 1,
|
| 7 |
+
"eos_token_id": 2,
|
| 8 |
+
"head_dim": null,
|
| 9 |
+
"hidden_act": "silu",
|
| 10 |
+
"hidden_size": 256,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 512,
|
| 13 |
+
"max_position_embeddings": 1024,
|
| 14 |
+
"model_type": "mistral",
|
| 15 |
+
"num_attention_heads": 4,
|
| 16 |
+
"num_hidden_layers": 2,
|
| 17 |
+
"num_key_value_heads": 4,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_theta": 10000.0,
|
| 21 |
+
"sliding_window": 256,
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.52.3",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 30000
|
| 27 |
+
}
|
last-checkpoint/generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.52.3"
|
| 7 |
+
}
|
last-checkpoint/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:eeb256257e095023202df2821297a8d339ad1834d968399938dd0c2e0d61ddb0
|
| 3 |
+
size 66690264
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8ef34edc57783b5d2ff75e949910673211e6c4ca385376406335fa83c91745cd
|
| 3 |
+
size 133393631
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4569729d12b4ffcd3c0004d02f4f36833455efd52f16ae44ecb408092e8532c9
|
| 3 |
+
size 13990
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e617b832469a9a84668626e0822faae0a42147d51b13c5698eb75b59b231f4f
|
| 3 |
+
size 1064
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,606 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1000,
|
| 3 |
+
"best_metric": 9.911575317382812,
|
| 4 |
+
"best_model_checkpoint": "./models/v-001/checkpoint-1000",
|
| 5 |
+
"epoch": 41.97826086956522,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1300,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.6521739130434783,
|
| 14 |
+
"grad_norm": 0.5281125903129578,
|
| 15 |
+
"learning_rate": 3.166666666666667e-06,
|
| 16 |
+
"loss": 10.3483,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 1.2934782608695652,
|
| 21 |
+
"grad_norm": 0.6133605241775513,
|
| 22 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 23 |
+
"loss": 10.3417,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.9456521739130435,
|
| 28 |
+
"grad_norm": 0.6125457882881165,
|
| 29 |
+
"learning_rate": 9.833333333333333e-06,
|
| 30 |
+
"loss": 10.3299,
|
| 31 |
+
"step": 60
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 2.5869565217391304,
|
| 35 |
+
"grad_norm": 0.5962333679199219,
|
| 36 |
+
"learning_rate": 1.3166666666666665e-05,
|
| 37 |
+
"loss": 10.3064,
|
| 38 |
+
"step": 80
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 3.2282608695652173,
|
| 42 |
+
"grad_norm": 0.6132860779762268,
|
| 43 |
+
"learning_rate": 1.65e-05,
|
| 44 |
+
"loss": 10.2727,
|
| 45 |
+
"step": 100
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 3.2282608695652173,
|
| 49 |
+
"eval_accuracy": 4.39651795777744e-05,
|
| 50 |
+
"eval_loss": 10.328398704528809,
|
| 51 |
+
"eval_runtime": 17.8533,
|
| 52 |
+
"eval_samples_per_second": 16.916,
|
| 53 |
+
"eval_steps_per_second": 1.064,
|
| 54 |
+
"step": 100
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.880434782608696,
|
| 58 |
+
"grad_norm": 0.5671436190605164,
|
| 59 |
+
"learning_rate": 1.9833333333333335e-05,
|
| 60 |
+
"loss": 10.2083,
|
| 61 |
+
"step": 120
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 4.521739130434782,
|
| 65 |
+
"grad_norm": 0.4685352146625519,
|
| 66 |
+
"learning_rate": 2.3166666666666666e-05,
|
| 67 |
+
"loss": 10.1064,
|
| 68 |
+
"step": 140
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.163043478260869,
|
| 72 |
+
"grad_norm": 0.45952484011650085,
|
| 73 |
+
"learning_rate": 2.6500000000000004e-05,
|
| 74 |
+
"loss": 9.9909,
|
| 75 |
+
"step": 160
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 5.815217391304348,
|
| 79 |
+
"grad_norm": 0.41472023725509644,
|
| 80 |
+
"learning_rate": 2.9833333333333335e-05,
|
| 81 |
+
"loss": 9.8718,
|
| 82 |
+
"step": 180
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 6.456521739130435,
|
| 86 |
+
"grad_norm": 0.43127089738845825,
|
| 87 |
+
"learning_rate": 3.316666666666667e-05,
|
| 88 |
+
"loss": 9.7582,
|
| 89 |
+
"step": 200
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 6.456521739130435,
|
| 93 |
+
"eval_accuracy": 0.0026139297676240417,
|
| 94 |
+
"eval_loss": 10.096575736999512,
|
| 95 |
+
"eval_runtime": 35.2536,
|
| 96 |
+
"eval_samples_per_second": 8.566,
|
| 97 |
+
"eval_steps_per_second": 0.539,
|
| 98 |
+
"step": 200
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.0978260869565215,
|
| 102 |
+
"grad_norm": 0.4042549431324005,
|
| 103 |
+
"learning_rate": 3.65e-05,
|
| 104 |
+
"loss": 9.6266,
|
| 105 |
+
"step": 220
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 7.75,
|
| 109 |
+
"grad_norm": 0.3853429853916168,
|
| 110 |
+
"learning_rate": 3.983333333333333e-05,
|
| 111 |
+
"loss": 9.5189,
|
| 112 |
+
"step": 240
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 8.391304347826088,
|
| 116 |
+
"grad_norm": 0.38628196716308594,
|
| 117 |
+
"learning_rate": 4.316666666666667e-05,
|
| 118 |
+
"loss": 9.4044,
|
| 119 |
+
"step": 260
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 9.032608695652174,
|
| 123 |
+
"grad_norm": 0.43260782957077026,
|
| 124 |
+
"learning_rate": 4.6500000000000005e-05,
|
| 125 |
+
"loss": 9.3052,
|
| 126 |
+
"step": 280
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.684782608695652,
|
| 130 |
+
"grad_norm": 0.4363991916179657,
|
| 131 |
+
"learning_rate": 4.9833333333333336e-05,
|
| 132 |
+
"loss": 9.2052,
|
| 133 |
+
"step": 300
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 9.684782608695652,
|
| 137 |
+
"eval_accuracy": 0.003677087746504768,
|
| 138 |
+
"eval_loss": 9.951318740844727,
|
| 139 |
+
"eval_runtime": 47.0129,
|
| 140 |
+
"eval_samples_per_second": 6.424,
|
| 141 |
+
"eval_steps_per_second": 0.404,
|
| 142 |
+
"step": 300
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 10.326086956521738,
|
| 146 |
+
"grad_norm": 0.49219855666160583,
|
| 147 |
+
"learning_rate": 5.316666666666667e-05,
|
| 148 |
+
"loss": 9.1327,
|
| 149 |
+
"step": 320
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 10.978260869565217,
|
| 153 |
+
"grad_norm": 0.4471158981323242,
|
| 154 |
+
"learning_rate": 5.65e-05,
|
| 155 |
+
"loss": 9.0328,
|
| 156 |
+
"step": 340
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 11.619565217391305,
|
| 160 |
+
"grad_norm": 0.3913232684135437,
|
| 161 |
+
"learning_rate": 5.983333333333334e-05,
|
| 162 |
+
"loss": 8.9514,
|
| 163 |
+
"step": 360
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 12.26086956521739,
|
| 167 |
+
"grad_norm": 1.0882291793823242,
|
| 168 |
+
"learning_rate": 6.316666666666668e-05,
|
| 169 |
+
"loss": 8.8916,
|
| 170 |
+
"step": 380
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 12.91304347826087,
|
| 174 |
+
"grad_norm": 0.6010486483573914,
|
| 175 |
+
"learning_rate": 6.65e-05,
|
| 176 |
+
"loss": 8.8216,
|
| 177 |
+
"step": 400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 12.91304347826087,
|
| 181 |
+
"eval_accuracy": 0.003413296669038122,
|
| 182 |
+
"eval_loss": 9.953831672668457,
|
| 183 |
+
"eval_runtime": 36.8026,
|
| 184 |
+
"eval_samples_per_second": 8.206,
|
| 185 |
+
"eval_steps_per_second": 0.516,
|
| 186 |
+
"step": 400
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 13.554347826086957,
|
| 190 |
+
"grad_norm": 0.49556687474250793,
|
| 191 |
+
"learning_rate": 6.983333333333334e-05,
|
| 192 |
+
"loss": 8.7406,
|
| 193 |
+
"step": 420
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 14.195652173913043,
|
| 197 |
+
"grad_norm": 0.495381623506546,
|
| 198 |
+
"learning_rate": 7.316666666666668e-05,
|
| 199 |
+
"loss": 8.679,
|
| 200 |
+
"step": 440
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 14.847826086956522,
|
| 204 |
+
"grad_norm": 0.6165482401847839,
|
| 205 |
+
"learning_rate": 7.65e-05,
|
| 206 |
+
"loss": 8.5904,
|
| 207 |
+
"step": 460
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 15.48913043478261,
|
| 211 |
+
"grad_norm": 0.5654007792472839,
|
| 212 |
+
"learning_rate": 7.983333333333334e-05,
|
| 213 |
+
"loss": 8.4896,
|
| 214 |
+
"step": 480
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 16.130434782608695,
|
| 218 |
+
"grad_norm": 0.6611935496330261,
|
| 219 |
+
"learning_rate": 8.316666666666666e-05,
|
| 220 |
+
"loss": 8.406,
|
| 221 |
+
"step": 500
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 16.130434782608695,
|
| 225 |
+
"eval_accuracy": 0.0029496638662179554,
|
| 226 |
+
"eval_loss": 9.952414512634277,
|
| 227 |
+
"eval_runtime": 39.649,
|
| 228 |
+
"eval_samples_per_second": 7.617,
|
| 229 |
+
"eval_steps_per_second": 0.479,
|
| 230 |
+
"step": 500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 16.782608695652176,
|
| 234 |
+
"grad_norm": 0.7537912726402283,
|
| 235 |
+
"learning_rate": 8.65e-05,
|
| 236 |
+
"loss": 8.2948,
|
| 237 |
+
"step": 520
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 17.42391304347826,
|
| 241 |
+
"grad_norm": 0.9145230650901794,
|
| 242 |
+
"learning_rate": 8.983333333333334e-05,
|
| 243 |
+
"loss": 8.1992,
|
| 244 |
+
"step": 540
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 18.065217391304348,
|
| 248 |
+
"grad_norm": 0.810655415058136,
|
| 249 |
+
"learning_rate": 9.316666666666666e-05,
|
| 250 |
+
"loss": 8.08,
|
| 251 |
+
"step": 560
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 18.717391304347824,
|
| 255 |
+
"grad_norm": 0.9121057987213135,
|
| 256 |
+
"learning_rate": 9.65e-05,
|
| 257 |
+
"loss": 7.9438,
|
| 258 |
+
"step": 580
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 19.358695652173914,
|
| 262 |
+
"grad_norm": 0.8612993359565735,
|
| 263 |
+
"learning_rate": 9.983333333333334e-05,
|
| 264 |
+
"loss": 7.8326,
|
| 265 |
+
"step": 600
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 19.358695652173914,
|
| 269 |
+
"eval_accuracy": 0.0021423032957897346,
|
| 270 |
+
"eval_loss": 9.945837020874023,
|
| 271 |
+
"eval_runtime": 25.9715,
|
| 272 |
+
"eval_samples_per_second": 11.628,
|
| 273 |
+
"eval_steps_per_second": 0.732,
|
| 274 |
+
"step": 600
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 20.0,
|
| 278 |
+
"grad_norm": 0.8960981369018555,
|
| 279 |
+
"learning_rate": 9.995456138403733e-05,
|
| 280 |
+
"loss": 7.716,
|
| 281 |
+
"step": 620
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 20.652173913043477,
|
| 285 |
+
"grad_norm": 0.9986662268638611,
|
| 286 |
+
"learning_rate": 9.980864681729001e-05,
|
| 287 |
+
"loss": 7.5692,
|
| 288 |
+
"step": 640
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 21.293478260869566,
|
| 292 |
+
"grad_norm": 0.9232766628265381,
|
| 293 |
+
"learning_rate": 9.956242426451834e-05,
|
| 294 |
+
"loss": 7.4208,
|
| 295 |
+
"step": 660
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 21.945652173913043,
|
| 299 |
+
"grad_norm": 0.9031963348388672,
|
| 300 |
+
"learning_rate": 9.921638958517565e-05,
|
| 301 |
+
"loss": 7.3481,
|
| 302 |
+
"step": 680
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 22.58695652173913,
|
| 306 |
+
"grad_norm": 0.9567400813102722,
|
| 307 |
+
"learning_rate": 9.877123964705497e-05,
|
| 308 |
+
"loss": 7.1956,
|
| 309 |
+
"step": 700
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 22.58695652173913,
|
| 313 |
+
"eval_accuracy": 0.001666679989448357,
|
| 314 |
+
"eval_loss": 9.986405372619629,
|
| 315 |
+
"eval_runtime": 57.7614,
|
| 316 |
+
"eval_samples_per_second": 5.228,
|
| 317 |
+
"eval_steps_per_second": 0.329,
|
| 318 |
+
"step": 700
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 23.22826086956522,
|
| 322 |
+
"grad_norm": 1.135650873184204,
|
| 323 |
+
"learning_rate": 9.822787092288991e-05,
|
| 324 |
+
"loss": 7.0604,
|
| 325 |
+
"step": 720
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 23.880434782608695,
|
| 329 |
+
"grad_norm": 0.8771520853042603,
|
| 330 |
+
"learning_rate": 9.758737768497802e-05,
|
| 331 |
+
"loss": 6.9215,
|
| 332 |
+
"step": 740
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 24.52173913043478,
|
| 336 |
+
"grad_norm": 1.0156564712524414,
|
| 337 |
+
"learning_rate": 9.685104980146193e-05,
|
| 338 |
+
"loss": 6.8363,
|
| 339 |
+
"step": 760
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 25.16304347826087,
|
| 343 |
+
"grad_norm": 0.9963734149932861,
|
| 344 |
+
"learning_rate": 9.60203701387066e-05,
|
| 345 |
+
"loss": 6.6577,
|
| 346 |
+
"step": 780
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 25.815217391304348,
|
| 350 |
+
"grad_norm": 0.8723818063735962,
|
| 351 |
+
"learning_rate": 9.509701157500376e-05,
|
| 352 |
+
"loss": 6.5659,
|
| 353 |
+
"step": 800
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 25.815217391304348,
|
| 357 |
+
"eval_accuracy": 0.0014988129401514,
|
| 358 |
+
"eval_loss": 9.925810813903809,
|
| 359 |
+
"eval_runtime": 49.4048,
|
| 360 |
+
"eval_samples_per_second": 6.113,
|
| 361 |
+
"eval_steps_per_second": 0.385,
|
| 362 |
+
"step": 800
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 26.456521739130434,
|
| 366 |
+
"grad_norm": 1.1702412366867065,
|
| 367 |
+
"learning_rate": 9.408283363161774e-05,
|
| 368 |
+
"loss": 6.393,
|
| 369 |
+
"step": 820
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 27.097826086956523,
|
| 373 |
+
"grad_norm": 0.8747526407241821,
|
| 374 |
+
"learning_rate": 9.297987872795705e-05,
|
| 375 |
+
"loss": 6.3074,
|
| 376 |
+
"step": 840
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 27.75,
|
| 380 |
+
"grad_norm": 0.973866879940033,
|
| 381 |
+
"learning_rate": 9.179036806841353e-05,
|
| 382 |
+
"loss": 6.1801,
|
| 383 |
+
"step": 860
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 28.391304347826086,
|
| 387 |
+
"grad_norm": 1.585481882095337,
|
| 388 |
+
"learning_rate": 9.051669716915227e-05,
|
| 389 |
+
"loss": 6.1175,
|
| 390 |
+
"step": 880
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 29.032608695652176,
|
| 394 |
+
"grad_norm": 1.0919766426086426,
|
| 395 |
+
"learning_rate": 8.916143103386093e-05,
|
| 396 |
+
"loss": 5.9719,
|
| 397 |
+
"step": 900
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 29.032608695652176,
|
| 401 |
+
"eval_accuracy": 0.001458844595080696,
|
| 402 |
+
"eval_loss": 9.97097396850586,
|
| 403 |
+
"eval_runtime": 46.643,
|
| 404 |
+
"eval_samples_per_second": 6.475,
|
| 405 |
+
"eval_steps_per_second": 0.407,
|
| 406 |
+
"step": 900
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 29.684782608695652,
|
| 410 |
+
"grad_norm": 1.173614501953125,
|
| 411 |
+
"learning_rate": 8.77272989881736e-05,
|
| 412 |
+
"loss": 5.8702,
|
| 413 |
+
"step": 920
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 30.32608695652174,
|
| 417 |
+
"grad_norm": 1.0350476503372192,
|
| 418 |
+
"learning_rate": 8.621718918317225e-05,
|
| 419 |
+
"loss": 5.704,
|
| 420 |
+
"step": 940
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 30.97826086956522,
|
| 424 |
+
"grad_norm": 1.1128321886062622,
|
| 425 |
+
"learning_rate": 8.463414277903475e-05,
|
| 426 |
+
"loss": 5.6413,
|
| 427 |
+
"step": 960
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 31.619565217391305,
|
| 431 |
+
"grad_norm": 1.2460695505142212,
|
| 432 |
+
"learning_rate": 8.298134782054305e-05,
|
| 433 |
+
"loss": 5.4948,
|
| 434 |
+
"step": 980
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 32.26086956521739,
|
| 438 |
+
"grad_norm": 1.1606298685073853,
|
| 439 |
+
"learning_rate": 8.126213281678526e-05,
|
| 440 |
+
"loss": 5.4031,
|
| 441 |
+
"step": 1000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 32.26086956521739,
|
| 445 |
+
"eval_accuracy": 0.001079145316909008,
|
| 446 |
+
"eval_loss": 9.911575317382812,
|
| 447 |
+
"eval_runtime": 48.65,
|
| 448 |
+
"eval_samples_per_second": 6.208,
|
| 449 |
+
"eval_steps_per_second": 0.391,
|
| 450 |
+
"step": 1000
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 32.91304347826087,
|
| 454 |
+
"grad_norm": 1.0351324081420898,
|
| 455 |
+
"learning_rate": 7.94799600379813e-05,
|
| 456 |
+
"loss": 5.3272,
|
| 457 |
+
"step": 1020
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 33.55434782608695,
|
| 461 |
+
"grad_norm": 1.089340329170227,
|
| 462 |
+
"learning_rate": 7.763841854293145e-05,
|
| 463 |
+
"loss": 5.2996,
|
| 464 |
+
"step": 1040
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 34.19565217391305,
|
| 468 |
+
"grad_norm": 1.3611856698989868,
|
| 469 |
+
"learning_rate": 7.574121695112954e-05,
|
| 470 |
+
"loss": 5.1266,
|
| 471 |
+
"step": 1060
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 34.84782608695652,
|
| 475 |
+
"grad_norm": 1.2501380443572998,
|
| 476 |
+
"learning_rate": 7.379217597409688e-05,
|
| 477 |
+
"loss": 5.0434,
|
| 478 |
+
"step": 1080
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 35.48913043478261,
|
| 482 |
+
"grad_norm": 1.057522177696228,
|
| 483 |
+
"learning_rate": 7.179522072097774e-05,
|
| 484 |
+
"loss": 4.9784,
|
| 485 |
+
"step": 1100
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 35.48913043478261,
|
| 489 |
+
"eval_accuracy": 0.0011830630140928385,
|
| 490 |
+
"eval_loss": 9.981914520263672,
|
| 491 |
+
"eval_runtime": 106.635,
|
| 492 |
+
"eval_samples_per_second": 2.832,
|
| 493 |
+
"eval_steps_per_second": 0.178,
|
| 494 |
+
"step": 1100
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 36.130434782608695,
|
| 498 |
+
"grad_norm": 1.1768474578857422,
|
| 499 |
+
"learning_rate": 6.975437279389181e-05,
|
| 500 |
+
"loss": 4.9012,
|
| 501 |
+
"step": 1120
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.78260869565217,
|
| 505 |
+
"grad_norm": 1.1783802509307861,
|
| 506 |
+
"learning_rate": 6.767374218896286e-05,
|
| 507 |
+
"loss": 4.823,
|
| 508 |
+
"step": 1140
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 37.42391304347826,
|
| 512 |
+
"grad_norm": 1.220082402229309,
|
| 513 |
+
"learning_rate": 6.555751901933342e-05,
|
| 514 |
+
"loss": 4.7149,
|
| 515 |
+
"step": 1160
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 38.06521739130435,
|
| 519 |
+
"grad_norm": 1.3078495264053345,
|
| 520 |
+
"learning_rate": 6.340996507683458e-05,
|
| 521 |
+
"loss": 4.6413,
|
| 522 |
+
"step": 1180
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 38.71739130434783,
|
| 526 |
+
"grad_norm": 1.2146966457366943,
|
| 527 |
+
"learning_rate": 6.123540524930442e-05,
|
| 528 |
+
"loss": 4.6684,
|
| 529 |
+
"step": 1200
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 38.71739130434783,
|
| 533 |
+
"eval_accuracy": 0.0008952909295837696,
|
| 534 |
+
"eval_loss": 10.014237403869629,
|
| 535 |
+
"eval_runtime": 31.4866,
|
| 536 |
+
"eval_samples_per_second": 9.591,
|
| 537 |
+
"eval_steps_per_second": 0.603,
|
| 538 |
+
"step": 1200
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 39.391304347826086,
|
| 542 |
+
"grad_norm": 1.090649962425232,
|
| 543 |
+
"learning_rate": 5.903821881083942e-05,
|
| 544 |
+
"loss": 4.7794,
|
| 545 |
+
"step": 1220
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 40.03260869565217,
|
| 549 |
+
"grad_norm": 1.2051453590393066,
|
| 550 |
+
"learning_rate": 5.682283060251932e-05,
|
| 551 |
+
"loss": 4.4631,
|
| 552 |
+
"step": 1240
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 40.68478260869565,
|
| 556 |
+
"grad_norm": 1.0512608289718628,
|
| 557 |
+
"learning_rate": 5.4593702121365955e-05,
|
| 558 |
+
"loss": 4.4119,
|
| 559 |
+
"step": 1260
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 41.32608695652174,
|
| 563 |
+
"grad_norm": 0.9912136793136597,
|
| 564 |
+
"learning_rate": 5.235532253548213e-05,
|
| 565 |
+
"loss": 4.3377,
|
| 566 |
+
"step": 1280
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 41.97826086956522,
|
| 570 |
+
"grad_norm": 1.0219991207122803,
|
| 571 |
+
"learning_rate": 5.0112199643464376e-05,
|
| 572 |
+
"loss": 4.3184,
|
| 573 |
+
"step": 1300
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 41.97826086956522,
|
| 577 |
+
"eval_accuracy": 0.001019192799302952,
|
| 578 |
+
"eval_loss": 10.048251152038574,
|
| 579 |
+
"eval_runtime": 18.0166,
|
| 580 |
+
"eval_samples_per_second": 16.762,
|
| 581 |
+
"eval_steps_per_second": 1.055,
|
| 582 |
+
"step": 1300
|
| 583 |
+
}
|
| 584 |
+
],
|
| 585 |
+
"logging_steps": 20,
|
| 586 |
+
"max_steps": 2000,
|
| 587 |
+
"num_input_tokens_seen": 0,
|
| 588 |
+
"num_train_epochs": 65,
|
| 589 |
+
"save_steps": 100,
|
| 590 |
+
"stateful_callbacks": {
|
| 591 |
+
"TrainerControl": {
|
| 592 |
+
"args": {
|
| 593 |
+
"should_epoch_stop": false,
|
| 594 |
+
"should_evaluate": false,
|
| 595 |
+
"should_log": false,
|
| 596 |
+
"should_save": true,
|
| 597 |
+
"should_training_stop": false
|
| 598 |
+
},
|
| 599 |
+
"attributes": {}
|
| 600 |
+
}
|
| 601 |
+
},
|
| 602 |
+
"total_flos": 1697662857792000.0,
|
| 603 |
+
"train_batch_size": 8,
|
| 604 |
+
"trial_name": null,
|
| 605 |
+
"trial_params": null
|
| 606 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e652f371cbe27456b95b1d9e9643f9465e54ffa6734a66c5db4740dea53b2296
|
| 3 |
+
size 5368
|