Commit
•
76923c1
1
Parent(s):
221c151
c62718f534f20ee7ab44eaebd42ee2c0bb8c460d909ecda815845ba3c9441623
Browse files- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +709 -0
- training_args.bin +3 -0
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:264db4985a059cf9f7defc109c9da516f7ae349492077cf506df5e4f0791f2e5
|
| 3 |
+
size 379087640
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a5d904c94dddd20981ad675b5471b4d770b65b10337780330598d7b656ac63f
|
| 3 |
+
size 3152668
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87cf6a4bbc04caf76a5487daaff4aac7a0bcb9a198a568d8941ea716cf385afa
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c6f96762648505b015243f25d9c34a5d7b09dd1128a3d863de9924e8e2724e56
|
| 3 |
+
size 1064
|
trainer_state.json
ADDED
|
@@ -0,0 +1,709 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.1978042721748352,
|
| 3 |
+
"best_model_checkpoint": "/p/scratch/ccstdl/kalyan1/finetuned-cosine-loss/checkpoint-430",
|
| 4 |
+
"epoch": 0.043,
|
| 5 |
+
"eval_steps": 10,
|
| 6 |
+
"global_step": 430,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.001,
|
| 13 |
+
"grad_norm": 1.646112322807312,
|
| 14 |
+
"learning_rate": 5.000000000000001e-07,
|
| 15 |
+
"loss": 0.6464,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.001,
|
| 20 |
+
"eval_cos_sim": 0.30497944355010986,
|
| 21 |
+
"eval_loss": 0.6448461413383484,
|
| 22 |
+
"eval_runtime": 205.7229,
|
| 23 |
+
"eval_samples_per_second": 4.861,
|
| 24 |
+
"eval_steps_per_second": 0.156,
|
| 25 |
+
"step": 10
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 0.002,
|
| 29 |
+
"grad_norm": 1.769713282585144,
|
| 30 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 31 |
+
"loss": 0.6442,
|
| 32 |
+
"step": 20
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.002,
|
| 36 |
+
"eval_cos_sim": 0.30667245388031006,
|
| 37 |
+
"eval_loss": 0.6431530714035034,
|
| 38 |
+
"eval_runtime": 205.6457,
|
| 39 |
+
"eval_samples_per_second": 4.863,
|
| 40 |
+
"eval_steps_per_second": 0.156,
|
| 41 |
+
"step": 20
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 0.003,
|
| 45 |
+
"grad_norm": 1.794607400894165,
|
| 46 |
+
"learning_rate": 1.5e-06,
|
| 47 |
+
"loss": 0.6231,
|
| 48 |
+
"step": 30
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 0.003,
|
| 52 |
+
"eval_cos_sim": 0.30951327085494995,
|
| 53 |
+
"eval_loss": 0.640312135219574,
|
| 54 |
+
"eval_runtime": 204.7069,
|
| 55 |
+
"eval_samples_per_second": 4.885,
|
| 56 |
+
"eval_steps_per_second": 0.156,
|
| 57 |
+
"step": 30
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.004,
|
| 61 |
+
"grad_norm": 1.7727755308151245,
|
| 62 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 63 |
+
"loss": 0.6291,
|
| 64 |
+
"step": 40
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.004,
|
| 68 |
+
"eval_cos_sim": 0.3135722279548645,
|
| 69 |
+
"eval_loss": 0.6362530589103699,
|
| 70 |
+
"eval_runtime": 204.4705,
|
| 71 |
+
"eval_samples_per_second": 4.891,
|
| 72 |
+
"eval_steps_per_second": 0.157,
|
| 73 |
+
"step": 40
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.005,
|
| 77 |
+
"grad_norm": 1.492561936378479,
|
| 78 |
+
"learning_rate": 2.5e-06,
|
| 79 |
+
"loss": 0.619,
|
| 80 |
+
"step": 50
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.005,
|
| 84 |
+
"eval_cos_sim": 0.31877169013023376,
|
| 85 |
+
"eval_loss": 0.6310535669326782,
|
| 86 |
+
"eval_runtime": 205.9799,
|
| 87 |
+
"eval_samples_per_second": 4.855,
|
| 88 |
+
"eval_steps_per_second": 0.155,
|
| 89 |
+
"step": 50
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.006,
|
| 93 |
+
"grad_norm": 1.761650800704956,
|
| 94 |
+
"learning_rate": 3e-06,
|
| 95 |
+
"loss": 0.619,
|
| 96 |
+
"step": 60
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 0.006,
|
| 100 |
+
"eval_cos_sim": 0.32503262162208557,
|
| 101 |
+
"eval_loss": 0.624792218208313,
|
| 102 |
+
"eval_runtime": 206.3179,
|
| 103 |
+
"eval_samples_per_second": 4.847,
|
| 104 |
+
"eval_steps_per_second": 0.155,
|
| 105 |
+
"step": 60
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 0.007,
|
| 109 |
+
"grad_norm": 1.7647501230239868,
|
| 110 |
+
"learning_rate": 3.5000000000000004e-06,
|
| 111 |
+
"loss": 0.6151,
|
| 112 |
+
"step": 70
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.007,
|
| 116 |
+
"eval_cos_sim": 0.33216750621795654,
|
| 117 |
+
"eval_loss": 0.6176563501358032,
|
| 118 |
+
"eval_runtime": 208.1576,
|
| 119 |
+
"eval_samples_per_second": 4.804,
|
| 120 |
+
"eval_steps_per_second": 0.154,
|
| 121 |
+
"step": 70
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.008,
|
| 125 |
+
"grad_norm": 1.762241005897522,
|
| 126 |
+
"learning_rate": 4.000000000000001e-06,
|
| 127 |
+
"loss": 0.6257,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.008,
|
| 132 |
+
"eval_cos_sim": 0.3403419852256775,
|
| 133 |
+
"eval_loss": 0.6094820499420166,
|
| 134 |
+
"eval_runtime": 205.6449,
|
| 135 |
+
"eval_samples_per_second": 4.863,
|
| 136 |
+
"eval_steps_per_second": 0.156,
|
| 137 |
+
"step": 80
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.009,
|
| 141 |
+
"grad_norm": 1.4978479146957397,
|
| 142 |
+
"learning_rate": 4.5e-06,
|
| 143 |
+
"loss": 0.5978,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.009,
|
| 148 |
+
"eval_cos_sim": 0.34952041506767273,
|
| 149 |
+
"eval_loss": 0.6003041863441467,
|
| 150 |
+
"eval_runtime": 205.9245,
|
| 151 |
+
"eval_samples_per_second": 4.856,
|
| 152 |
+
"eval_steps_per_second": 0.155,
|
| 153 |
+
"step": 90
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.01,
|
| 157 |
+
"grad_norm": 1.4880496263504028,
|
| 158 |
+
"learning_rate": 5e-06,
|
| 159 |
+
"loss": 0.5816,
|
| 160 |
+
"step": 100
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.01,
|
| 164 |
+
"eval_cos_sim": 0.35966718196868896,
|
| 165 |
+
"eval_loss": 0.590160071849823,
|
| 166 |
+
"eval_runtime": 206.7863,
|
| 167 |
+
"eval_samples_per_second": 4.836,
|
| 168 |
+
"eval_steps_per_second": 0.155,
|
| 169 |
+
"step": 100
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 0.011,
|
| 173 |
+
"grad_norm": 1.4825042486190796,
|
| 174 |
+
"learning_rate": 5.500000000000001e-06,
|
| 175 |
+
"loss": 0.5922,
|
| 176 |
+
"step": 110
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.011,
|
| 180 |
+
"eval_cos_sim": 0.37048399448394775,
|
| 181 |
+
"eval_loss": 0.5793455243110657,
|
| 182 |
+
"eval_runtime": 206.0982,
|
| 183 |
+
"eval_samples_per_second": 4.852,
|
| 184 |
+
"eval_steps_per_second": 0.155,
|
| 185 |
+
"step": 110
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.012,
|
| 189 |
+
"grad_norm": 1.761700987815857,
|
| 190 |
+
"learning_rate": 6e-06,
|
| 191 |
+
"loss": 0.5767,
|
| 192 |
+
"step": 120
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.012,
|
| 196 |
+
"eval_cos_sim": 0.38159415125846863,
|
| 197 |
+
"eval_loss": 0.5682348608970642,
|
| 198 |
+
"eval_runtime": 204.1513,
|
| 199 |
+
"eval_samples_per_second": 4.898,
|
| 200 |
+
"eval_steps_per_second": 0.157,
|
| 201 |
+
"step": 120
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.013,
|
| 205 |
+
"grad_norm": 1.7859400510787964,
|
| 206 |
+
"learning_rate": 6.5000000000000004e-06,
|
| 207 |
+
"loss": 0.5625,
|
| 208 |
+
"step": 130
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.013,
|
| 212 |
+
"eval_cos_sim": 0.39334964752197266,
|
| 213 |
+
"eval_loss": 0.5564795732498169,
|
| 214 |
+
"eval_runtime": 205.8548,
|
| 215 |
+
"eval_samples_per_second": 4.858,
|
| 216 |
+
"eval_steps_per_second": 0.155,
|
| 217 |
+
"step": 130
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.014,
|
| 221 |
+
"grad_norm": 1.4824175834655762,
|
| 222 |
+
"learning_rate": 7.000000000000001e-06,
|
| 223 |
+
"loss": 0.5348,
|
| 224 |
+
"step": 140
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.014,
|
| 228 |
+
"eval_cos_sim": 0.4060880243778229,
|
| 229 |
+
"eval_loss": 0.5437397956848145,
|
| 230 |
+
"eval_runtime": 205.6798,
|
| 231 |
+
"eval_samples_per_second": 4.862,
|
| 232 |
+
"eval_steps_per_second": 0.156,
|
| 233 |
+
"step": 140
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.015,
|
| 237 |
+
"grad_norm": 1.453766107559204,
|
| 238 |
+
"learning_rate": 7.5e-06,
|
| 239 |
+
"loss": 0.5359,
|
| 240 |
+
"step": 150
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.015,
|
| 244 |
+
"eval_cos_sim": 0.4195743799209595,
|
| 245 |
+
"eval_loss": 0.5302525758743286,
|
| 246 |
+
"eval_runtime": 206.2792,
|
| 247 |
+
"eval_samples_per_second": 4.848,
|
| 248 |
+
"eval_steps_per_second": 0.155,
|
| 249 |
+
"step": 150
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.016,
|
| 253 |
+
"grad_norm": 1.7521625757217407,
|
| 254 |
+
"learning_rate": 8.000000000000001e-06,
|
| 255 |
+
"loss": 0.5202,
|
| 256 |
+
"step": 160
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 0.016,
|
| 260 |
+
"eval_cos_sim": 0.43342718482017517,
|
| 261 |
+
"eval_loss": 0.516399621963501,
|
| 262 |
+
"eval_runtime": 206.1506,
|
| 263 |
+
"eval_samples_per_second": 4.851,
|
| 264 |
+
"eval_steps_per_second": 0.155,
|
| 265 |
+
"step": 160
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.017,
|
| 269 |
+
"grad_norm": 1.4833202362060547,
|
| 270 |
+
"learning_rate": 8.500000000000002e-06,
|
| 271 |
+
"loss": 0.5133,
|
| 272 |
+
"step": 170
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.017,
|
| 276 |
+
"eval_cos_sim": 0.44808587431907654,
|
| 277 |
+
"eval_loss": 0.5017415285110474,
|
| 278 |
+
"eval_runtime": 205.6637,
|
| 279 |
+
"eval_samples_per_second": 4.862,
|
| 280 |
+
"eval_steps_per_second": 0.156,
|
| 281 |
+
"step": 170
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 0.018,
|
| 285 |
+
"grad_norm": 1.7655549049377441,
|
| 286 |
+
"learning_rate": 9e-06,
|
| 287 |
+
"loss": 0.5066,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 0.018,
|
| 292 |
+
"eval_cos_sim": 0.46300601959228516,
|
| 293 |
+
"eval_loss": 0.4868185222148895,
|
| 294 |
+
"eval_runtime": 206.8425,
|
| 295 |
+
"eval_samples_per_second": 4.835,
|
| 296 |
+
"eval_steps_per_second": 0.155,
|
| 297 |
+
"step": 180
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.019,
|
| 301 |
+
"grad_norm": 1.6160492897033691,
|
| 302 |
+
"learning_rate": 9.5e-06,
|
| 303 |
+
"loss": 0.4772,
|
| 304 |
+
"step": 190
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.019,
|
| 308 |
+
"eval_cos_sim": 0.4783403277397156,
|
| 309 |
+
"eval_loss": 0.4714851975440979,
|
| 310 |
+
"eval_runtime": 205.1447,
|
| 311 |
+
"eval_samples_per_second": 4.875,
|
| 312 |
+
"eval_steps_per_second": 0.156,
|
| 313 |
+
"step": 190
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.02,
|
| 317 |
+
"grad_norm": 1.7988218069076538,
|
| 318 |
+
"learning_rate": 1e-05,
|
| 319 |
+
"loss": 0.4706,
|
| 320 |
+
"step": 200
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.02,
|
| 324 |
+
"eval_cos_sim": 0.49314171075820923,
|
| 325 |
+
"eval_loss": 0.45668038725852966,
|
| 326 |
+
"eval_runtime": 206.3434,
|
| 327 |
+
"eval_samples_per_second": 4.846,
|
| 328 |
+
"eval_steps_per_second": 0.155,
|
| 329 |
+
"step": 200
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 0.021,
|
| 333 |
+
"grad_norm": 1.7824742794036865,
|
| 334 |
+
"learning_rate": 1.05e-05,
|
| 335 |
+
"loss": 0.4531,
|
| 336 |
+
"step": 210
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 0.021,
|
| 340 |
+
"eval_cos_sim": 0.5080329179763794,
|
| 341 |
+
"eval_loss": 0.44177642464637756,
|
| 342 |
+
"eval_runtime": 207.391,
|
| 343 |
+
"eval_samples_per_second": 4.822,
|
| 344 |
+
"eval_steps_per_second": 0.154,
|
| 345 |
+
"step": 210
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.022,
|
| 349 |
+
"grad_norm": 1.800609827041626,
|
| 350 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 351 |
+
"loss": 0.4324,
|
| 352 |
+
"step": 220
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.022,
|
| 356 |
+
"eval_cos_sim": 0.5240386724472046,
|
| 357 |
+
"eval_loss": 0.42576006054878235,
|
| 358 |
+
"eval_runtime": 207.013,
|
| 359 |
+
"eval_samples_per_second": 4.831,
|
| 360 |
+
"eval_steps_per_second": 0.155,
|
| 361 |
+
"step": 220
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.023,
|
| 365 |
+
"grad_norm": 1.4796422719955444,
|
| 366 |
+
"learning_rate": 1.1500000000000002e-05,
|
| 367 |
+
"loss": 0.4139,
|
| 368 |
+
"step": 230
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.023,
|
| 372 |
+
"eval_cos_sim": 0.5389044880867004,
|
| 373 |
+
"eval_loss": 0.410887211561203,
|
| 374 |
+
"eval_runtime": 205.4474,
|
| 375 |
+
"eval_samples_per_second": 4.867,
|
| 376 |
+
"eval_steps_per_second": 0.156,
|
| 377 |
+
"step": 230
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.024,
|
| 381 |
+
"grad_norm": 1.513137936592102,
|
| 382 |
+
"learning_rate": 1.2e-05,
|
| 383 |
+
"loss": 0.3916,
|
| 384 |
+
"step": 240
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.024,
|
| 388 |
+
"eval_cos_sim": 0.5540903806686401,
|
| 389 |
+
"eval_loss": 0.39568817615509033,
|
| 390 |
+
"eval_runtime": 206.197,
|
| 391 |
+
"eval_samples_per_second": 4.85,
|
| 392 |
+
"eval_steps_per_second": 0.155,
|
| 393 |
+
"step": 240
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.025,
|
| 397 |
+
"grad_norm": 1.4907095432281494,
|
| 398 |
+
"learning_rate": 1.25e-05,
|
| 399 |
+
"loss": 0.3867,
|
| 400 |
+
"step": 250
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 0.025,
|
| 404 |
+
"eval_cos_sim": 0.5701205730438232,
|
| 405 |
+
"eval_loss": 0.37963634729385376,
|
| 406 |
+
"eval_runtime": 206.8619,
|
| 407 |
+
"eval_samples_per_second": 4.834,
|
| 408 |
+
"eval_steps_per_second": 0.155,
|
| 409 |
+
"step": 250
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.026,
|
| 413 |
+
"grad_norm": 1.5080854892730713,
|
| 414 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 415 |
+
"loss": 0.3893,
|
| 416 |
+
"step": 260
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.026,
|
| 420 |
+
"eval_cos_sim": 0.5862915515899658,
|
| 421 |
+
"eval_loss": 0.36344093084335327,
|
| 422 |
+
"eval_runtime": 206.4905,
|
| 423 |
+
"eval_samples_per_second": 4.843,
|
| 424 |
+
"eval_steps_per_second": 0.155,
|
| 425 |
+
"step": 260
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 0.027,
|
| 429 |
+
"grad_norm": 1.738050103187561,
|
| 430 |
+
"learning_rate": 1.3500000000000001e-05,
|
| 431 |
+
"loss": 0.3718,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 0.027,
|
| 436 |
+
"eval_cos_sim": 0.5994535088539124,
|
| 437 |
+
"eval_loss": 0.3502576947212219,
|
| 438 |
+
"eval_runtime": 205.2709,
|
| 439 |
+
"eval_samples_per_second": 4.872,
|
| 440 |
+
"eval_steps_per_second": 0.156,
|
| 441 |
+
"step": 270
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.028,
|
| 445 |
+
"grad_norm": 1.7529724836349487,
|
| 446 |
+
"learning_rate": 1.4000000000000001e-05,
|
| 447 |
+
"loss": 0.3364,
|
| 448 |
+
"step": 280
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.028,
|
| 452 |
+
"eval_cos_sim": 0.613076388835907,
|
| 453 |
+
"eval_loss": 0.33661308884620667,
|
| 454 |
+
"eval_runtime": 205.299,
|
| 455 |
+
"eval_samples_per_second": 4.871,
|
| 456 |
+
"eval_steps_per_second": 0.156,
|
| 457 |
+
"step": 280
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.029,
|
| 461 |
+
"grad_norm": 1.753875732421875,
|
| 462 |
+
"learning_rate": 1.45e-05,
|
| 463 |
+
"loss": 0.3149,
|
| 464 |
+
"step": 290
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.029,
|
| 468 |
+
"eval_cos_sim": 0.6267997622489929,
|
| 469 |
+
"eval_loss": 0.32285839319229126,
|
| 470 |
+
"eval_runtime": 205.7429,
|
| 471 |
+
"eval_samples_per_second": 4.86,
|
| 472 |
+
"eval_steps_per_second": 0.156,
|
| 473 |
+
"step": 290
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 0.03,
|
| 477 |
+
"grad_norm": 1.4430509805679321,
|
| 478 |
+
"learning_rate": 1.5e-05,
|
| 479 |
+
"loss": 0.3346,
|
| 480 |
+
"step": 300
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 0.03,
|
| 484 |
+
"eval_cos_sim": 0.6370284557342529,
|
| 485 |
+
"eval_loss": 0.31259986758232117,
|
| 486 |
+
"eval_runtime": 205.8712,
|
| 487 |
+
"eval_samples_per_second": 4.857,
|
| 488 |
+
"eval_steps_per_second": 0.155,
|
| 489 |
+
"step": 300
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.031,
|
| 493 |
+
"grad_norm": 1.6388583183288574,
|
| 494 |
+
"learning_rate": 1.55e-05,
|
| 495 |
+
"loss": 0.2906,
|
| 496 |
+
"step": 310
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 0.031,
|
| 500 |
+
"eval_cos_sim": 0.6471716165542603,
|
| 501 |
+
"eval_loss": 0.3024389445781708,
|
| 502 |
+
"eval_runtime": 205.8913,
|
| 503 |
+
"eval_samples_per_second": 4.857,
|
| 504 |
+
"eval_steps_per_second": 0.155,
|
| 505 |
+
"step": 310
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.032,
|
| 509 |
+
"grad_norm": 1.6151902675628662,
|
| 510 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 511 |
+
"loss": 0.2823,
|
| 512 |
+
"step": 320
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.032,
|
| 516 |
+
"eval_cos_sim": 0.658047616481781,
|
| 517 |
+
"eval_loss": 0.29153889417648315,
|
| 518 |
+
"eval_runtime": 205.5626,
|
| 519 |
+
"eval_samples_per_second": 4.865,
|
| 520 |
+
"eval_steps_per_second": 0.156,
|
| 521 |
+
"step": 320
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.033,
|
| 525 |
+
"grad_norm": 1.3532074689865112,
|
| 526 |
+
"learning_rate": 1.65e-05,
|
| 527 |
+
"loss": 0.2806,
|
| 528 |
+
"step": 330
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.033,
|
| 532 |
+
"eval_cos_sim": 0.6677139401435852,
|
| 533 |
+
"eval_loss": 0.28184273838996887,
|
| 534 |
+
"eval_runtime": 205.9081,
|
| 535 |
+
"eval_samples_per_second": 4.857,
|
| 536 |
+
"eval_steps_per_second": 0.155,
|
| 537 |
+
"step": 330
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.034,
|
| 541 |
+
"grad_norm": 1.5439000129699707,
|
| 542 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 543 |
+
"loss": 0.2604,
|
| 544 |
+
"step": 340
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.034,
|
| 548 |
+
"eval_cos_sim": 0.6758972406387329,
|
| 549 |
+
"eval_loss": 0.2736455500125885,
|
| 550 |
+
"eval_runtime": 206.6691,
|
| 551 |
+
"eval_samples_per_second": 4.839,
|
| 552 |
+
"eval_steps_per_second": 0.155,
|
| 553 |
+
"step": 340
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 0.035,
|
| 557 |
+
"grad_norm": 1.4628266096115112,
|
| 558 |
+
"learning_rate": 1.75e-05,
|
| 559 |
+
"loss": 0.2488,
|
| 560 |
+
"step": 350
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.035,
|
| 564 |
+
"eval_cos_sim": 0.6875510215759277,
|
| 565 |
+
"eval_loss": 0.2619880139827728,
|
| 566 |
+
"eval_runtime": 205.4389,
|
| 567 |
+
"eval_samples_per_second": 4.868,
|
| 568 |
+
"eval_steps_per_second": 0.156,
|
| 569 |
+
"step": 350
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.036,
|
| 573 |
+
"grad_norm": 1.2534619569778442,
|
| 574 |
+
"learning_rate": 1.8e-05,
|
| 575 |
+
"loss": 0.2421,
|
| 576 |
+
"step": 360
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.036,
|
| 580 |
+
"eval_cos_sim": 0.6983117461204529,
|
| 581 |
+
"eval_loss": 0.2512022852897644,
|
| 582 |
+
"eval_runtime": 205.4173,
|
| 583 |
+
"eval_samples_per_second": 4.868,
|
| 584 |
+
"eval_steps_per_second": 0.156,
|
| 585 |
+
"step": 360
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.037,
|
| 589 |
+
"grad_norm": 1.3977046012878418,
|
| 590 |
+
"learning_rate": 1.85e-05,
|
| 591 |
+
"loss": 0.2185,
|
| 592 |
+
"step": 370
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 0.037,
|
| 596 |
+
"eval_cos_sim": 0.7063655257225037,
|
| 597 |
+
"eval_loss": 0.2431405633687973,
|
| 598 |
+
"eval_runtime": 206.1942,
|
| 599 |
+
"eval_samples_per_second": 4.85,
|
| 600 |
+
"eval_steps_per_second": 0.155,
|
| 601 |
+
"step": 370
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 0.038,
|
| 605 |
+
"grad_norm": 1.193185567855835,
|
| 606 |
+
"learning_rate": 1.9e-05,
|
| 607 |
+
"loss": 0.237,
|
| 608 |
+
"step": 380
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.038,
|
| 612 |
+
"eval_cos_sim": 0.7140612602233887,
|
| 613 |
+
"eval_loss": 0.23545415699481964,
|
| 614 |
+
"eval_runtime": 204.7566,
|
| 615 |
+
"eval_samples_per_second": 4.884,
|
| 616 |
+
"eval_steps_per_second": 0.156,
|
| 617 |
+
"step": 380
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.039,
|
| 621 |
+
"grad_norm": 1.2560912370681763,
|
| 622 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 623 |
+
"loss": 0.244,
|
| 624 |
+
"step": 390
|
| 625 |
+
},
|
| 626 |
+
{
|
| 627 |
+
"epoch": 0.039,
|
| 628 |
+
"eval_cos_sim": 0.7227933406829834,
|
| 629 |
+
"eval_loss": 0.22671909630298615,
|
| 630 |
+
"eval_runtime": 205.2004,
|
| 631 |
+
"eval_samples_per_second": 4.873,
|
| 632 |
+
"eval_steps_per_second": 0.156,
|
| 633 |
+
"step": 390
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.04,
|
| 637 |
+
"grad_norm": 1.1640348434448242,
|
| 638 |
+
"learning_rate": 2e-05,
|
| 639 |
+
"loss": 0.2261,
|
| 640 |
+
"step": 400
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.04,
|
| 644 |
+
"eval_cos_sim": 0.7307066917419434,
|
| 645 |
+
"eval_loss": 0.21881377696990967,
|
| 646 |
+
"eval_runtime": 206.284,
|
| 647 |
+
"eval_samples_per_second": 4.848,
|
| 648 |
+
"eval_steps_per_second": 0.155,
|
| 649 |
+
"step": 400
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 0.041,
|
| 653 |
+
"grad_norm": 1.0957697629928589,
|
| 654 |
+
"learning_rate": 2.05e-05,
|
| 655 |
+
"loss": 0.204,
|
| 656 |
+
"step": 410
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.041,
|
| 660 |
+
"eval_cos_sim": 0.7365487813949585,
|
| 661 |
+
"eval_loss": 0.2129872441291809,
|
| 662 |
+
"eval_runtime": 205.7321,
|
| 663 |
+
"eval_samples_per_second": 4.861,
|
| 664 |
+
"eval_steps_per_second": 0.156,
|
| 665 |
+
"step": 410
|
| 666 |
+
},
|
| 667 |
+
{
|
| 668 |
+
"epoch": 0.042,
|
| 669 |
+
"grad_norm": 1.025030493736267,
|
| 670 |
+
"learning_rate": 2.1e-05,
|
| 671 |
+
"loss": 0.2039,
|
| 672 |
+
"step": 420
|
| 673 |
+
},
|
| 674 |
+
{
|
| 675 |
+
"epoch": 0.042,
|
| 676 |
+
"eval_cos_sim": 0.7446093559265137,
|
| 677 |
+
"eval_loss": 0.20492298901081085,
|
| 678 |
+
"eval_runtime": 206.2353,
|
| 679 |
+
"eval_samples_per_second": 4.849,
|
| 680 |
+
"eval_steps_per_second": 0.155,
|
| 681 |
+
"step": 420
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.043,
|
| 685 |
+
"grad_norm": 0.9591805934906006,
|
| 686 |
+
"learning_rate": 2.15e-05,
|
| 687 |
+
"loss": 0.2197,
|
| 688 |
+
"step": 430
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.043,
|
| 692 |
+
"eval_cos_sim": 0.751737117767334,
|
| 693 |
+
"eval_loss": 0.1978042721748352,
|
| 694 |
+
"eval_runtime": 206.0346,
|
| 695 |
+
"eval_samples_per_second": 4.854,
|
| 696 |
+
"eval_steps_per_second": 0.155,
|
| 697 |
+
"step": 430
|
| 698 |
+
}
|
| 699 |
+
],
|
| 700 |
+
"logging_steps": 10,
|
| 701 |
+
"max_steps": 10000,
|
| 702 |
+
"num_input_tokens_seen": 0,
|
| 703 |
+
"num_train_epochs": 9223372036854775807,
|
| 704 |
+
"save_steps": 10,
|
| 705 |
+
"total_flos": 0.0,
|
| 706 |
+
"train_batch_size": 140,
|
| 707 |
+
"trial_name": null,
|
| 708 |
+
"trial_params": null
|
| 709 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:820f28d1edcf77c869c57ddd43701a33ecdd10e592a609c522807ecb5cbb11c2
|
| 3 |
+
size 5048
|