mistral-lp2-org_org_a
Browse files- README.md +6 -21
- adapter_config.json +3 -3
- adapter_model.safetensors +1 -1
- trainer_state.json +329 -0
- training_args.bin +1 -1
README.md
CHANGED
|
@@ -16,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss:
|
| 20 |
-
- F1 Micro: 0.
|
| 21 |
-
- F1 Macro: 0.
|
| 22 |
-
- F1 Weighted: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
@@ -44,28 +44,13 @@ The following hyperparameters were used during training:
|
|
| 44 |
- seed: 42
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
-
- training_steps:
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | F1 Weighted |
|
| 52 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:--------:|:-----------:|
|
| 53 |
-
|
|
| 54 |
-
| 1.4371 | 0.0308 | 50 | 1.2747 | 0.6089 | 0.5857 | 0.6052 |
|
| 55 |
-
| 1.2556 | 0.0462 | 75 | 1.1545 | 0.6304 | 0.6036 | 0.6240 |
|
| 56 |
-
| 1.2415 | 0.0615 | 100 | 1.0691 | 0.6320 | 0.6132 | 0.6301 |
|
| 57 |
-
| 0.9864 | 0.0769 | 125 | 1.0264 | 0.6399 | 0.6278 | 0.6411 |
|
| 58 |
-
| 1.0647 | 0.0923 | 150 | 0.9918 | 0.6510 | 0.6266 | 0.6455 |
|
| 59 |
-
| 0.9849 | 0.1077 | 175 | 0.9679 | 0.6576 | 0.6317 | 0.6511 |
|
| 60 |
-
| 1.0067 | 0.1231 | 200 | 0.9383 | 0.6501 | 0.6384 | 0.6513 |
|
| 61 |
-
| 0.8928 | 0.1385 | 225 | 0.9243 | 0.6620 | 0.6405 | 0.6579 |
|
| 62 |
-
| 0.9858 | 0.1538 | 250 | 0.9132 | 0.6627 | 0.6405 | 0.6582 |
|
| 63 |
-
| 0.9085 | 0.1692 | 275 | 0.9011 | 0.6575 | 0.6446 | 0.6581 |
|
| 64 |
-
| 1.0059 | 0.1846 | 300 | 0.9018 | 0.6686 | 0.6436 | 0.6623 |
|
| 65 |
-
| 0.8939 | 0.2 | 325 | 0.8928 | 0.6682 | 0.6448 | 0.6629 |
|
| 66 |
-
| 0.864 | 0.2154 | 350 | 0.8833 | 0.6622 | 0.6478 | 0.6619 |
|
| 67 |
-
| 0.9499 | 0.2308 | 375 | 0.8837 | 0.6585 | 0.6463 | 0.6593 |
|
| 68 |
-
| 0.9721 | 0.2462 | 400 | 0.8808 | 0.6615 | 0.6476 | 0.6615 |
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.6260
|
| 20 |
+
- F1 Micro: 0.5230
|
| 21 |
+
- F1 Macro: 0.5155
|
| 22 |
+
- F1 Weighted: 0.5274
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 44 |
- seed: 42
|
| 45 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
- lr_scheduler_type: linear
|
| 47 |
+
- training_steps: 25
|
| 48 |
|
| 49 |
### Training results
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss | F1 Micro | F1 Macro | F1 Weighted |
|
| 52 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:--------:|:-----------:|
|
| 53 |
+
| 1.8454 | 0.0154 | 25 | 1.6260 | 0.5230 | 0.5155 | 0.5274 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 54 |
|
| 55 |
|
| 56 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -20,10 +20,10 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"k_proj",
|
| 24 |
"q_proj",
|
| 25 |
-
"
|
| 26 |
-
"o_proj"
|
|
|
|
| 27 |
],
|
| 28 |
"task_type": "SEQ_CLS",
|
| 29 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
| 23 |
"q_proj",
|
| 24 |
+
"k_proj",
|
| 25 |
+
"o_proj",
|
| 26 |
+
"v_proj"
|
| 27 |
],
|
| 28 |
"task_type": "SEQ_CLS",
|
| 29 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 578881968
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c436c187cbabace0ab368aea37ad0c4711742d5d7324ad76d9c8544cf59472b4
|
| 3 |
size 578881968
|
trainer_state.json
ADDED
|
@@ -0,0 +1,329 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8807787895202637,
|
| 3 |
+
"best_model_checkpoint": "org_org_a/org_aug_a/checkpoint-400",
|
| 4 |
+
"epoch": 0.24615384615384617,
|
| 5 |
+
"eval_steps": 25,
|
| 6 |
+
"global_step": 400,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.015384615384615385,
|
| 13 |
+
"grad_norm": 99.7617416381836,
|
| 14 |
+
"learning_rate": 9.375e-05,
|
| 15 |
+
"loss": 2.2195,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.015384615384615385,
|
| 20 |
+
"eval_f1_macro": 0.5521299126846324,
|
| 21 |
+
"eval_f1_micro": 0.5691194856224325,
|
| 22 |
+
"eval_f1_weighted": 0.569423288164649,
|
| 23 |
+
"eval_loss": 1.520703673362732,
|
| 24 |
+
"eval_runtime": 1286.7359,
|
| 25 |
+
"eval_samples_per_second": 8.703,
|
| 26 |
+
"eval_steps_per_second": 0.272,
|
| 27 |
+
"step": 25
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.03076923076923077,
|
| 31 |
+
"grad_norm": 40.433555603027344,
|
| 32 |
+
"learning_rate": 8.75e-05,
|
| 33 |
+
"loss": 1.4371,
|
| 34 |
+
"step": 50
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.03076923076923077,
|
| 38 |
+
"eval_f1_macro": 0.5857301862067549,
|
| 39 |
+
"eval_f1_micro": 0.6089480264332917,
|
| 40 |
+
"eval_f1_weighted": 0.6051732662908408,
|
| 41 |
+
"eval_loss": 1.2746953964233398,
|
| 42 |
+
"eval_runtime": 1302.368,
|
| 43 |
+
"eval_samples_per_second": 8.598,
|
| 44 |
+
"eval_steps_per_second": 0.269,
|
| 45 |
+
"step": 50
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.046153846153846156,
|
| 49 |
+
"grad_norm": 50.680335998535156,
|
| 50 |
+
"learning_rate": 8.125000000000001e-05,
|
| 51 |
+
"loss": 1.2556,
|
| 52 |
+
"step": 75
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.046153846153846156,
|
| 56 |
+
"eval_f1_macro": 0.6036315452406847,
|
| 57 |
+
"eval_f1_micro": 0.6303804250759064,
|
| 58 |
+
"eval_f1_weighted": 0.6240449220829647,
|
| 59 |
+
"eval_loss": 1.1545159816741943,
|
| 60 |
+
"eval_runtime": 1308.8563,
|
| 61 |
+
"eval_samples_per_second": 8.556,
|
| 62 |
+
"eval_steps_per_second": 0.267,
|
| 63 |
+
"step": 75
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.06153846153846154,
|
| 67 |
+
"grad_norm": 24.990671157836914,
|
| 68 |
+
"learning_rate": 7.500000000000001e-05,
|
| 69 |
+
"loss": 1.2415,
|
| 70 |
+
"step": 100
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.06153846153846154,
|
| 74 |
+
"eval_f1_macro": 0.6131831448419165,
|
| 75 |
+
"eval_f1_micro": 0.6319878549741025,
|
| 76 |
+
"eval_f1_weighted": 0.6300913896752308,
|
| 77 |
+
"eval_loss": 1.0690622329711914,
|
| 78 |
+
"eval_runtime": 1302.0106,
|
| 79 |
+
"eval_samples_per_second": 8.601,
|
| 80 |
+
"eval_steps_per_second": 0.269,
|
| 81 |
+
"step": 100
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 0.07692307692307693,
|
| 85 |
+
"grad_norm": 20.941816329956055,
|
| 86 |
+
"learning_rate": 6.875e-05,
|
| 87 |
+
"loss": 0.9864,
|
| 88 |
+
"step": 125
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.07692307692307693,
|
| 92 |
+
"eval_f1_macro": 0.627800207605976,
|
| 93 |
+
"eval_f1_micro": 0.6399357028040722,
|
| 94 |
+
"eval_f1_weighted": 0.6411240450638264,
|
| 95 |
+
"eval_loss": 1.0263742208480835,
|
| 96 |
+
"eval_runtime": 1310.3949,
|
| 97 |
+
"eval_samples_per_second": 8.546,
|
| 98 |
+
"eval_steps_per_second": 0.267,
|
| 99 |
+
"step": 125
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 0.09230769230769231,
|
| 103 |
+
"grad_norm": 17.243314743041992,
|
| 104 |
+
"learning_rate": 6.25e-05,
|
| 105 |
+
"loss": 1.0647,
|
| 106 |
+
"step": 150
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.09230769230769231,
|
| 110 |
+
"eval_f1_macro": 0.6265827051287185,
|
| 111 |
+
"eval_f1_micro": 0.6510091087694231,
|
| 112 |
+
"eval_f1_weighted": 0.6455165794591529,
|
| 113 |
+
"eval_loss": 0.9917964339256287,
|
| 114 |
+
"eval_runtime": 1292.6103,
|
| 115 |
+
"eval_samples_per_second": 8.663,
|
| 116 |
+
"eval_steps_per_second": 0.271,
|
| 117 |
+
"step": 150
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 0.1076923076923077,
|
| 121 |
+
"grad_norm": 20.114173889160156,
|
| 122 |
+
"learning_rate": 5.6250000000000005e-05,
|
| 123 |
+
"loss": 0.9849,
|
| 124 |
+
"step": 175
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.1076923076923077,
|
| 128 |
+
"eval_f1_macro": 0.6317476736951155,
|
| 129 |
+
"eval_f1_micro": 0.6576174316842294,
|
| 130 |
+
"eval_f1_weighted": 0.6510976948325254,
|
| 131 |
+
"eval_loss": 0.9679338932037354,
|
| 132 |
+
"eval_runtime": 1305.0812,
|
| 133 |
+
"eval_samples_per_second": 8.58,
|
| 134 |
+
"eval_steps_per_second": 0.268,
|
| 135 |
+
"step": 175
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.12307692307692308,
|
| 139 |
+
"grad_norm": 39.2221565246582,
|
| 140 |
+
"learning_rate": 5e-05,
|
| 141 |
+
"loss": 1.0067,
|
| 142 |
+
"step": 200
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.12307692307692308,
|
| 146 |
+
"eval_f1_macro": 0.6383959350585475,
|
| 147 |
+
"eval_f1_micro": 0.6501160921593142,
|
| 148 |
+
"eval_f1_weighted": 0.6513020604373679,
|
| 149 |
+
"eval_loss": 0.9382981061935425,
|
| 150 |
+
"eval_runtime": 1283.0843,
|
| 151 |
+
"eval_samples_per_second": 8.727,
|
| 152 |
+
"eval_steps_per_second": 0.273,
|
| 153 |
+
"step": 200
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.13846153846153847,
|
| 157 |
+
"grad_norm": 26.992185592651367,
|
| 158 |
+
"learning_rate": 4.375e-05,
|
| 159 |
+
"loss": 0.8928,
|
| 160 |
+
"step": 225
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.13846153846153847,
|
| 164 |
+
"eval_f1_macro": 0.640450740779414,
|
| 165 |
+
"eval_f1_micro": 0.6619932130737631,
|
| 166 |
+
"eval_f1_weighted": 0.6578984928748007,
|
| 167 |
+
"eval_loss": 0.9242791533470154,
|
| 168 |
+
"eval_runtime": 1276.3685,
|
| 169 |
+
"eval_samples_per_second": 8.773,
|
| 170 |
+
"eval_steps_per_second": 0.274,
|
| 171 |
+
"step": 225
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.15384615384615385,
|
| 175 |
+
"grad_norm": 71.61570739746094,
|
| 176 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 177 |
+
"loss": 0.9858,
|
| 178 |
+
"step": 250
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.15384615384615385,
|
| 182 |
+
"eval_f1_macro": 0.640469116958249,
|
| 183 |
+
"eval_f1_micro": 0.6627076263618503,
|
| 184 |
+
"eval_f1_weighted": 0.6581960436641718,
|
| 185 |
+
"eval_loss": 0.9131789803504944,
|
| 186 |
+
"eval_runtime": 1285.3671,
|
| 187 |
+
"eval_samples_per_second": 8.712,
|
| 188 |
+
"eval_steps_per_second": 0.272,
|
| 189 |
+
"step": 250
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.16923076923076924,
|
| 193 |
+
"grad_norm": 51.381019592285156,
|
| 194 |
+
"learning_rate": 3.125e-05,
|
| 195 |
+
"loss": 0.9085,
|
| 196 |
+
"step": 275
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.16923076923076924,
|
| 200 |
+
"eval_f1_macro": 0.6446344740224741,
|
| 201 |
+
"eval_f1_micro": 0.6575281300232184,
|
| 202 |
+
"eval_f1_weighted": 0.6580540316041209,
|
| 203 |
+
"eval_loss": 0.9010853171348572,
|
| 204 |
+
"eval_runtime": 1307.6049,
|
| 205 |
+
"eval_samples_per_second": 8.564,
|
| 206 |
+
"eval_steps_per_second": 0.268,
|
| 207 |
+
"step": 275
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.18461538461538463,
|
| 211 |
+
"grad_norm": 25.47317886352539,
|
| 212 |
+
"learning_rate": 2.5e-05,
|
| 213 |
+
"loss": 1.0059,
|
| 214 |
+
"step": 300
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.18461538461538463,
|
| 218 |
+
"eval_f1_macro": 0.6435798595814134,
|
| 219 |
+
"eval_f1_micro": 0.6686015359885694,
|
| 220 |
+
"eval_f1_weighted": 0.662301841928527,
|
| 221 |
+
"eval_loss": 0.9018191695213318,
|
| 222 |
+
"eval_runtime": 1330.8319,
|
| 223 |
+
"eval_samples_per_second": 8.414,
|
| 224 |
+
"eval_steps_per_second": 0.263,
|
| 225 |
+
"step": 300
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.2,
|
| 229 |
+
"grad_norm": 16.64508628845215,
|
| 230 |
+
"learning_rate": 1.8750000000000002e-05,
|
| 231 |
+
"loss": 0.8939,
|
| 232 |
+
"step": 325
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 0.2,
|
| 236 |
+
"eval_f1_macro": 0.6448244249080287,
|
| 237 |
+
"eval_f1_micro": 0.6681550276835149,
|
| 238 |
+
"eval_f1_weighted": 0.6628710932570111,
|
| 239 |
+
"eval_loss": 0.892798125743866,
|
| 240 |
+
"eval_runtime": 1305.0822,
|
| 241 |
+
"eval_samples_per_second": 8.58,
|
| 242 |
+
"eval_steps_per_second": 0.268,
|
| 243 |
+
"step": 325
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.2153846153846154,
|
| 247 |
+
"grad_norm": 32.9316520690918,
|
| 248 |
+
"learning_rate": 1.25e-05,
|
| 249 |
+
"loss": 0.864,
|
| 250 |
+
"step": 350
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.2153846153846154,
|
| 254 |
+
"eval_f1_macro": 0.6477608498148076,
|
| 255 |
+
"eval_f1_micro": 0.6621718163957849,
|
| 256 |
+
"eval_f1_weighted": 0.6618855058014609,
|
| 257 |
+
"eval_loss": 0.8832775950431824,
|
| 258 |
+
"eval_runtime": 1344.9811,
|
| 259 |
+
"eval_samples_per_second": 8.326,
|
| 260 |
+
"eval_steps_per_second": 0.26,
|
| 261 |
+
"step": 350
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.23076923076923078,
|
| 265 |
+
"grad_norm": 63.84526824951172,
|
| 266 |
+
"learning_rate": 6.25e-06,
|
| 267 |
+
"loss": 0.9499,
|
| 268 |
+
"step": 375
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.23076923076923078,
|
| 272 |
+
"eval_f1_macro": 0.646307301425064,
|
| 273 |
+
"eval_f1_micro": 0.6585104482943382,
|
| 274 |
+
"eval_f1_weighted": 0.659331813949001,
|
| 275 |
+
"eval_loss": 0.8836826682090759,
|
| 276 |
+
"eval_runtime": 1287.5574,
|
| 277 |
+
"eval_samples_per_second": 8.697,
|
| 278 |
+
"eval_steps_per_second": 0.272,
|
| 279 |
+
"step": 375
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.24615384615384617,
|
| 283 |
+
"grad_norm": 21.614879608154297,
|
| 284 |
+
"learning_rate": 0.0,
|
| 285 |
+
"loss": 0.9721,
|
| 286 |
+
"step": 400
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.24615384615384617,
|
| 290 |
+
"eval_f1_macro": 0.6475943444675745,
|
| 291 |
+
"eval_f1_micro": 0.6614574031076978,
|
| 292 |
+
"eval_f1_weighted": 0.6614511612983146,
|
| 293 |
+
"eval_loss": 0.8807787895202637,
|
| 294 |
+
"eval_runtime": 1313.0891,
|
| 295 |
+
"eval_samples_per_second": 8.528,
|
| 296 |
+
"eval_steps_per_second": 0.267,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.24615384615384617,
|
| 301 |
+
"step": 400,
|
| 302 |
+
"total_flos": 1.690284412871639e+17,
|
| 303 |
+
"train_loss": 1.104334650039673,
|
| 304 |
+
"train_runtime": 22485.3021,
|
| 305 |
+
"train_samples_per_second": 0.569,
|
| 306 |
+
"train_steps_per_second": 0.018
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.24615384615384617,
|
| 310 |
+
"eval_f1_macro": 0.6475943444675745,
|
| 311 |
+
"eval_f1_micro": 0.6614574031076978,
|
| 312 |
+
"eval_f1_weighted": 0.6614511612983146,
|
| 313 |
+
"eval_loss": 0.8807787895202637,
|
| 314 |
+
"eval_runtime": 1299.8097,
|
| 315 |
+
"eval_samples_per_second": 8.615,
|
| 316 |
+
"eval_steps_per_second": 0.269,
|
| 317 |
+
"step": 400
|
| 318 |
+
}
|
| 319 |
+
],
|
| 320 |
+
"logging_steps": 25,
|
| 321 |
+
"max_steps": 400,
|
| 322 |
+
"num_input_tokens_seen": 0,
|
| 323 |
+
"num_train_epochs": 1,
|
| 324 |
+
"save_steps": 25,
|
| 325 |
+
"total_flos": 1.690284412871639e+17,
|
| 326 |
+
"train_batch_size": 32,
|
| 327 |
+
"trial_name": null,
|
| 328 |
+
"trial_params": null
|
| 329 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4920
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3745867cf9811ba4f8713c7ef3ed9214e2a6decf9a602ecbc1d1be5e8770d17c
|
| 3 |
size 4920
|