Upload folder using huggingface_hub
Browse files- checkpoint-700/config.json +16 -0
- checkpoint-700/generation_config.json +6 -0
- checkpoint-700/model.safetensors +3 -0
- checkpoint-700/optimizer.pt +3 -0
- checkpoint-700/rng_state.pth +3 -0
- checkpoint-700/scaler.pt +3 -0
- checkpoint-700/scheduler.pt +3 -0
- checkpoint-700/trainer_state.json +524 -0
- checkpoint-700/training_args.bin +3 -0
checkpoint-700/config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"FastHyperNet125mForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"dtype": "float32",
|
| 6 |
+
"hidden_size": 768,
|
| 7 |
+
"hyperbolic_dim": 64,
|
| 8 |
+
"intermediate_size": 2048,
|
| 9 |
+
"max_position_embeddings": 512,
|
| 10 |
+
"model_type": "fasthypernet_125m_tagless",
|
| 11 |
+
"num_attention_heads": 12,
|
| 12 |
+
"num_hidden_layers": 16,
|
| 13 |
+
"transformers_version": "5.13.1",
|
| 14 |
+
"use_cache": false,
|
| 15 |
+
"vocab_size": 8192
|
| 16 |
+
}
|
checkpoint-700/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"output_attentions": false,
|
| 4 |
+
"output_hidden_states": false,
|
| 5 |
+
"transformers_version": "5.13.1"
|
| 6 |
+
}
|
checkpoint-700/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f09d56a91235222d5a8fdc19cd3b8284107ec33d09185d956ba28d5362715b2c
|
| 3 |
+
size 543424784
|
checkpoint-700/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5f83929d13b894ae09e210b63b0ae123c3f32d5e37fd8809de9919a2248b6b4
|
| 3 |
+
size 273342083
|
checkpoint-700/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fafcfd9ae4bbeb9903c611e3d4174a60e57a657df13b8c6f00c1c30865328812
|
| 3 |
+
size 14645
|
checkpoint-700/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:06c0a9161d45ab63667aa7b321fcc009bff889a955865636f0cd59493a7685de
|
| 3 |
+
size 1383
|
checkpoint-700/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b96643f2727ebe94509289c3f98dac6cf5be80a32a9bfe9c6ca737ec80846f98
|
| 3 |
+
size 1465
|
checkpoint-700/trainer_state.json
ADDED
|
@@ -0,0 +1,524 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.04855294872461808,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 700,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.0006936135532088297,
|
| 14 |
+
"grad_norm": 80.97769927978516,
|
| 15 |
+
"learning_rate": 2.0833333333333334e-06,
|
| 16 |
+
"loss": 39.45225219726562,
|
| 17 |
+
"step": 10
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.0013872271064176594,
|
| 21 |
+
"grad_norm": 69.19830322265625,
|
| 22 |
+
"learning_rate": 6.25e-06,
|
| 23 |
+
"loss": 38.231222534179686,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.0020808406596264893,
|
| 28 |
+
"grad_norm": 55.03429412841797,
|
| 29 |
+
"learning_rate": 1.0416666666666666e-05,
|
| 30 |
+
"loss": 37.07829284667969,
|
| 31 |
+
"step": 30
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.002774454212835319,
|
| 35 |
+
"grad_norm": 36.03839874267578,
|
| 36 |
+
"learning_rate": 1.4583333333333333e-05,
|
| 37 |
+
"loss": 35.945098876953125,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.0034680677660441487,
|
| 42 |
+
"grad_norm": 48.63792419433594,
|
| 43 |
+
"learning_rate": 1.875e-05,
|
| 44 |
+
"loss": 34.40289306640625,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.004161681319252979,
|
| 49 |
+
"grad_norm": 41.98967361450195,
|
| 50 |
+
"learning_rate": 2.2916666666666667e-05,
|
| 51 |
+
"loss": 33.45003662109375,
|
| 52 |
+
"step": 60
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.004855294872461808,
|
| 56 |
+
"grad_norm": 81.57576751708984,
|
| 57 |
+
"learning_rate": 2.7083333333333335e-05,
|
| 58 |
+
"loss": 32.02000427246094,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.005548908425670638,
|
| 63 |
+
"grad_norm": 60.575645446777344,
|
| 64 |
+
"learning_rate": 3.125e-05,
|
| 65 |
+
"loss": 30.74504699707031,
|
| 66 |
+
"step": 80
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.006242521978879467,
|
| 70 |
+
"grad_norm": 61.149410247802734,
|
| 71 |
+
"learning_rate": 3.541666666666667e-05,
|
| 72 |
+
"loss": 29.49912109375,
|
| 73 |
+
"step": 90
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.006936135532088297,
|
| 77 |
+
"grad_norm": 69.96782684326172,
|
| 78 |
+
"learning_rate": 3.958333333333333e-05,
|
| 79 |
+
"loss": 28.152670288085936,
|
| 80 |
+
"step": 100
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.007629749085297127,
|
| 84 |
+
"grad_norm": 33.35784149169922,
|
| 85 |
+
"learning_rate": 4.375e-05,
|
| 86 |
+
"loss": 27.86654052734375,
|
| 87 |
+
"step": 110
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.008323362638505957,
|
| 91 |
+
"grad_norm": 42.682308197021484,
|
| 92 |
+
"learning_rate": 4.791666666666667e-05,
|
| 93 |
+
"loss": 27.098068237304688,
|
| 94 |
+
"step": 120
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.009016976191714786,
|
| 98 |
+
"grad_norm": 45.40013122558594,
|
| 99 |
+
"learning_rate": 5.208333333333334e-05,
|
| 100 |
+
"loss": 25.768472290039064,
|
| 101 |
+
"step": 130
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.009710589744923616,
|
| 105 |
+
"grad_norm": 36.54875564575195,
|
| 106 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 107 |
+
"loss": 25.419296264648438,
|
| 108 |
+
"step": 140
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.010404203298132445,
|
| 112 |
+
"grad_norm": 54.964447021484375,
|
| 113 |
+
"learning_rate": 6.041666666666667e-05,
|
| 114 |
+
"loss": 24.5228759765625,
|
| 115 |
+
"step": 150
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.011097816851341275,
|
| 119 |
+
"grad_norm": 62.00859451293945,
|
| 120 |
+
"learning_rate": 6.458333333333334e-05,
|
| 121 |
+
"loss": 23.92150115966797,
|
| 122 |
+
"step": 160
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.011791430404550106,
|
| 126 |
+
"grad_norm": 50.303070068359375,
|
| 127 |
+
"learning_rate": 6.875e-05,
|
| 128 |
+
"loss": 23.61223449707031,
|
| 129 |
+
"step": 170
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.012485043957758934,
|
| 133 |
+
"grad_norm": 40.088356018066406,
|
| 134 |
+
"learning_rate": 7.291666666666667e-05,
|
| 135 |
+
"loss": 23.29130401611328,
|
| 136 |
+
"step": 180
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.013178657510967764,
|
| 140 |
+
"grad_norm": 58.86671829223633,
|
| 141 |
+
"learning_rate": 7.708333333333334e-05,
|
| 142 |
+
"loss": 23.32776794433594,
|
| 143 |
+
"step": 190
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.013872271064176595,
|
| 147 |
+
"grad_norm": 38.86407470703125,
|
| 148 |
+
"learning_rate": 8.125000000000001e-05,
|
| 149 |
+
"loss": 23.280059814453125,
|
| 150 |
+
"step": 200
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.014565884617385423,
|
| 154 |
+
"grad_norm": 37.631717681884766,
|
| 155 |
+
"learning_rate": 8.541666666666666e-05,
|
| 156 |
+
"loss": 22.259375,
|
| 157 |
+
"step": 210
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.015259498170594254,
|
| 161 |
+
"grad_norm": 37.38259506225586,
|
| 162 |
+
"learning_rate": 8.958333333333333e-05,
|
| 163 |
+
"loss": 21.507147216796874,
|
| 164 |
+
"step": 220
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.015953111723803082,
|
| 168 |
+
"grad_norm": 42.106143951416016,
|
| 169 |
+
"learning_rate": 9.375e-05,
|
| 170 |
+
"loss": 21.231875610351562,
|
| 171 |
+
"step": 230
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.016646725277011915,
|
| 175 |
+
"grad_norm": 50.364681243896484,
|
| 176 |
+
"learning_rate": 9.791666666666667e-05,
|
| 177 |
+
"loss": 21.234603881835938,
|
| 178 |
+
"step": 240
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.017340338830220743,
|
| 182 |
+
"grad_norm": 40.50483322143555,
|
| 183 |
+
"learning_rate": 0.00010208333333333334,
|
| 184 |
+
"loss": 20.62218017578125,
|
| 185 |
+
"step": 250
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.018033952383429572,
|
| 189 |
+
"grad_norm": 60.68687057495117,
|
| 190 |
+
"learning_rate": 0.00010625,
|
| 191 |
+
"loss": 19.96156463623047,
|
| 192 |
+
"step": 260
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.0187275659366384,
|
| 196 |
+
"grad_norm": 31.746686935424805,
|
| 197 |
+
"learning_rate": 0.00011041666666666666,
|
| 198 |
+
"loss": 19.283287048339844,
|
| 199 |
+
"step": 270
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.019421179489847232,
|
| 203 |
+
"grad_norm": 40.83134841918945,
|
| 204 |
+
"learning_rate": 0.00011458333333333333,
|
| 205 |
+
"loss": 18.95526580810547,
|
| 206 |
+
"step": 280
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.02011479304305606,
|
| 210 |
+
"grad_norm": 27.94510841369629,
|
| 211 |
+
"learning_rate": 0.00011875,
|
| 212 |
+
"loss": 19.398623657226562,
|
| 213 |
+
"step": 290
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.02080840659626489,
|
| 217 |
+
"grad_norm": 43.998260498046875,
|
| 218 |
+
"learning_rate": 0.00012291666666666665,
|
| 219 |
+
"loss": 18.262245178222656,
|
| 220 |
+
"step": 300
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.021502020149473722,
|
| 224 |
+
"grad_norm": 22.61271095275879,
|
| 225 |
+
"learning_rate": 0.00012708333333333332,
|
| 226 |
+
"loss": 18.646043395996095,
|
| 227 |
+
"step": 310
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.02219563370268255,
|
| 231 |
+
"grad_norm": 27.580198287963867,
|
| 232 |
+
"learning_rate": 0.00013125000000000002,
|
| 233 |
+
"loss": 17.519342041015626,
|
| 234 |
+
"step": 320
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.02288924725589138,
|
| 238 |
+
"grad_norm": 36.54242706298828,
|
| 239 |
+
"learning_rate": 0.00013541666666666666,
|
| 240 |
+
"loss": 16.588734436035157,
|
| 241 |
+
"step": 330
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.02358286080910021,
|
| 245 |
+
"grad_norm": 24.73045539855957,
|
| 246 |
+
"learning_rate": 0.00013958333333333333,
|
| 247 |
+
"loss": 16.85819091796875,
|
| 248 |
+
"step": 340
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.02427647436230904,
|
| 252 |
+
"grad_norm": 27.577693939208984,
|
| 253 |
+
"learning_rate": 0.00014375,
|
| 254 |
+
"loss": 16.190365600585938,
|
| 255 |
+
"step": 350
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.02497008791551787,
|
| 259 |
+
"grad_norm": 20.408132553100586,
|
| 260 |
+
"learning_rate": 0.00014791666666666667,
|
| 261 |
+
"loss": 15.682986450195312,
|
| 262 |
+
"step": 360
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.0256637014687267,
|
| 266 |
+
"grad_norm": 28.981483459472656,
|
| 267 |
+
"learning_rate": 0.00015208333333333333,
|
| 268 |
+
"loss": 15.275959777832032,
|
| 269 |
+
"step": 370
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.02635731502193553,
|
| 273 |
+
"grad_norm": 19.978363037109375,
|
| 274 |
+
"learning_rate": 0.00015625,
|
| 275 |
+
"loss": 15.033628845214844,
|
| 276 |
+
"step": 380
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.027050928575144358,
|
| 280 |
+
"grad_norm": 15.66555404663086,
|
| 281 |
+
"learning_rate": 0.00016041666666666667,
|
| 282 |
+
"loss": 14.446699523925782,
|
| 283 |
+
"step": 390
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.02774454212835319,
|
| 287 |
+
"grad_norm": 17.955888748168945,
|
| 288 |
+
"learning_rate": 0.00016458333333333334,
|
| 289 |
+
"loss": 14.196858215332032,
|
| 290 |
+
"step": 400
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.02843815568156202,
|
| 294 |
+
"grad_norm": 16.955345153808594,
|
| 295 |
+
"learning_rate": 0.00016875,
|
| 296 |
+
"loss": 13.838204956054687,
|
| 297 |
+
"step": 410
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.029131769234770847,
|
| 301 |
+
"grad_norm": 14.126781463623047,
|
| 302 |
+
"learning_rate": 0.00017291666666666668,
|
| 303 |
+
"loss": 13.345814514160157,
|
| 304 |
+
"step": 420
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.029825382787979676,
|
| 308 |
+
"grad_norm": 16.29298973083496,
|
| 309 |
+
"learning_rate": 0.00017708333333333335,
|
| 310 |
+
"loss": 13.06510009765625,
|
| 311 |
+
"step": 430
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.030518996341188508,
|
| 315 |
+
"grad_norm": 14.477455139160156,
|
| 316 |
+
"learning_rate": 0.00018125,
|
| 317 |
+
"loss": 12.520622253417969,
|
| 318 |
+
"step": 440
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.031212609894397336,
|
| 322 |
+
"grad_norm": 12.074920654296875,
|
| 323 |
+
"learning_rate": 0.00018541666666666668,
|
| 324 |
+
"loss": 12.411215209960938,
|
| 325 |
+
"step": 450
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.031906223447606165,
|
| 329 |
+
"grad_norm": 15.564898490905762,
|
| 330 |
+
"learning_rate": 0.00018958333333333332,
|
| 331 |
+
"loss": 12.00233154296875,
|
| 332 |
+
"step": 460
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.032599837000814993,
|
| 336 |
+
"grad_norm": 15.621328353881836,
|
| 337 |
+
"learning_rate": 0.00019375000000000002,
|
| 338 |
+
"loss": 12.109133911132812,
|
| 339 |
+
"step": 470
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.03329345055402383,
|
| 343 |
+
"grad_norm": 10.823607444763184,
|
| 344 |
+
"learning_rate": 0.00019791666666666666,
|
| 345 |
+
"loss": 11.684900665283203,
|
| 346 |
+
"step": 480
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.03398706410723266,
|
| 350 |
+
"grad_norm": 14.166363716125488,
|
| 351 |
+
"learning_rate": 0.00020208333333333333,
|
| 352 |
+
"loss": 11.60366439819336,
|
| 353 |
+
"step": 490
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.034680677660441486,
|
| 357 |
+
"grad_norm": 39.65184783935547,
|
| 358 |
+
"learning_rate": 0.00020625,
|
| 359 |
+
"loss": 11.060487365722656,
|
| 360 |
+
"step": 500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.035374291213650315,
|
| 364 |
+
"grad_norm": 12.022449493408203,
|
| 365 |
+
"learning_rate": 0.00021041666666666667,
|
| 366 |
+
"loss": 11.373404693603515,
|
| 367 |
+
"step": 510
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.036067904766859143,
|
| 371 |
+
"grad_norm": 9.625805854797363,
|
| 372 |
+
"learning_rate": 0.00021458333333333334,
|
| 373 |
+
"loss": 11.405331420898438,
|
| 374 |
+
"step": 520
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.03676151832006797,
|
| 378 |
+
"grad_norm": 12.523700714111328,
|
| 379 |
+
"learning_rate": 0.00021875,
|
| 380 |
+
"loss": 10.810578155517579,
|
| 381 |
+
"step": 530
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.0374551318732768,
|
| 385 |
+
"grad_norm": 11.111766815185547,
|
| 386 |
+
"learning_rate": 0.00022291666666666667,
|
| 387 |
+
"loss": 10.427848815917969,
|
| 388 |
+
"step": 540
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.038148745426485636,
|
| 392 |
+
"grad_norm": 9.552438735961914,
|
| 393 |
+
"learning_rate": 0.00022708333333333334,
|
| 394 |
+
"loss": 10.221891784667969,
|
| 395 |
+
"step": 550
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.038842358979694465,
|
| 399 |
+
"grad_norm": 10.854446411132812,
|
| 400 |
+
"learning_rate": 0.00023125,
|
| 401 |
+
"loss": 10.453067779541016,
|
| 402 |
+
"step": 560
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.039535972532903293,
|
| 406 |
+
"grad_norm": 8.054693222045898,
|
| 407 |
+
"learning_rate": 0.00023541666666666668,
|
| 408 |
+
"loss": 9.835127258300782,
|
| 409 |
+
"step": 570
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.04022958608611212,
|
| 413 |
+
"grad_norm": 9.217058181762695,
|
| 414 |
+
"learning_rate": 0.00023958333333333335,
|
| 415 |
+
"loss": 9.983591461181641,
|
| 416 |
+
"step": 580
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.04092319963932095,
|
| 420 |
+
"grad_norm": 7.442865371704102,
|
| 421 |
+
"learning_rate": 0.00024375,
|
| 422 |
+
"loss": 9.785844421386718,
|
| 423 |
+
"step": 590
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.04161681319252978,
|
| 427 |
+
"grad_norm": 7.499183177947998,
|
| 428 |
+
"learning_rate": 0.0002479166666666667,
|
| 429 |
+
"loss": 9.547152709960937,
|
| 430 |
+
"step": 600
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.042310426745738615,
|
| 434 |
+
"grad_norm": 6.999185085296631,
|
| 435 |
+
"learning_rate": 0.0002520833333333333,
|
| 436 |
+
"loss": 9.602791595458985,
|
| 437 |
+
"step": 610
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.043004040298947444,
|
| 441 |
+
"grad_norm": 8.540186882019043,
|
| 442 |
+
"learning_rate": 0.00025624999999999997,
|
| 443 |
+
"loss": 9.688284301757813,
|
| 444 |
+
"step": 620
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.04369765385215627,
|
| 448 |
+
"grad_norm": 7.028185844421387,
|
| 449 |
+
"learning_rate": 0.0002604166666666667,
|
| 450 |
+
"loss": 9.46673355102539,
|
| 451 |
+
"step": 630
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.0443912674053651,
|
| 455 |
+
"grad_norm": 10.2844820022583,
|
| 456 |
+
"learning_rate": 0.00026458333333333336,
|
| 457 |
+
"loss": 9.028829956054688,
|
| 458 |
+
"step": 640
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.04508488095857393,
|
| 462 |
+
"grad_norm": 6.920653820037842,
|
| 463 |
+
"learning_rate": 0.00026875,
|
| 464 |
+
"loss": 9.584944152832032,
|
| 465 |
+
"step": 650
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.04577849451178276,
|
| 469 |
+
"grad_norm": 7.473561763763428,
|
| 470 |
+
"learning_rate": 0.00027291666666666664,
|
| 471 |
+
"loss": 9.417964935302734,
|
| 472 |
+
"step": 660
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.04647210806499159,
|
| 476 |
+
"grad_norm": 6.416537761688232,
|
| 477 |
+
"learning_rate": 0.00027708333333333334,
|
| 478 |
+
"loss": 9.235848236083985,
|
| 479 |
+
"step": 670
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.04716572161820042,
|
| 483 |
+
"grad_norm": 7.264041900634766,
|
| 484 |
+
"learning_rate": 0.00028125000000000003,
|
| 485 |
+
"loss": 9.052194976806641,
|
| 486 |
+
"step": 680
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.04785933517140925,
|
| 490 |
+
"grad_norm": 5.6128692626953125,
|
| 491 |
+
"learning_rate": 0.0002854166666666667,
|
| 492 |
+
"loss": 8.73398208618164,
|
| 493 |
+
"step": 690
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.04855294872461808,
|
| 497 |
+
"grad_norm": 6.392711639404297,
|
| 498 |
+
"learning_rate": 0.00028958333333333337,
|
| 499 |
+
"loss": 8.775264739990234,
|
| 500 |
+
"step": 700
|
| 501 |
+
}
|
| 502 |
+
],
|
| 503 |
+
"logging_steps": 10,
|
| 504 |
+
"max_steps": 100000,
|
| 505 |
+
"num_input_tokens_seen": 0,
|
| 506 |
+
"num_train_epochs": 7,
|
| 507 |
+
"save_steps": 100,
|
| 508 |
+
"stateful_callbacks": {
|
| 509 |
+
"TrainerControl": {
|
| 510 |
+
"args": {
|
| 511 |
+
"should_epoch_stop": false,
|
| 512 |
+
"should_evaluate": false,
|
| 513 |
+
"should_log": false,
|
| 514 |
+
"should_save": true,
|
| 515 |
+
"should_training_stop": false
|
| 516 |
+
},
|
| 517 |
+
"attributes": {}
|
| 518 |
+
}
|
| 519 |
+
},
|
| 520 |
+
"total_flos": 7574703726182400.0,
|
| 521 |
+
"train_batch_size": 8,
|
| 522 |
+
"trial_name": null,
|
| 523 |
+
"trial_params": null
|
| 524 |
+
}
|
checkpoint-700/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:49f3e14a77c081da7c71dab3b3b7c4ec1dede30e28c9fb753d87e3fc83f88ec1
|
| 3 |
+
size 5137
|