End of training
Browse files- README.md +2 -1
- all_results.json +12 -0
- eval_results.json +7 -0
- train_results.json +8 -0
- trainer_state.json +675 -0
- training_eval_loss.png +0 -0
- training_loss.png +0 -0
README.md
CHANGED
|
@@ -4,6 +4,7 @@ license: llama3.1
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
|
|
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: oh_teknium_scaling_down_random_0.9
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# oh_teknium_scaling_down_random_0.9
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.6034
|
| 21 |
|
|
|
|
| 4 |
base_model: meta-llama/Meta-Llama-3.1-8B
|
| 5 |
tags:
|
| 6 |
- llama-factory
|
| 7 |
+
- full
|
| 8 |
- generated_from_trainer
|
| 9 |
model-index:
|
| 10 |
- name: oh_teknium_scaling_down_random_0.9
|
|
|
|
| 16 |
|
| 17 |
# oh_teknium_scaling_down_random_0.9
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/oh_teknium_scaling_down_random_0.9 dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6034
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9961522017956392,
|
| 3 |
+
"eval_loss": 0.6033630967140198,
|
| 4 |
+
"eval_runtime": 199.0551,
|
| 5 |
+
"eval_samples_per_second": 39.567,
|
| 6 |
+
"eval_steps_per_second": 0.623,
|
| 7 |
+
"total_flos": 1467123247349760.0,
|
| 8 |
+
"train_loss": 0.5650538080903493,
|
| 9 |
+
"train_runtime": 28810.5739,
|
| 10 |
+
"train_samples_per_second": 15.582,
|
| 11 |
+
"train_steps_per_second": 0.03
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9961522017956392,
|
| 3 |
+
"eval_loss": 0.6033630967140198,
|
| 4 |
+
"eval_runtime": 199.0551,
|
| 5 |
+
"eval_samples_per_second": 39.567,
|
| 6 |
+
"eval_steps_per_second": 0.623
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 2.9961522017956392,
|
| 3 |
+
"total_flos": 1467123247349760.0,
|
| 4 |
+
"train_loss": 0.5650538080903493,
|
| 5 |
+
"train_runtime": 28810.5739,
|
| 6 |
+
"train_samples_per_second": 15.582,
|
| 7 |
+
"train_steps_per_second": 0.03
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,675 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9961522017956392,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 876,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.034202650705429674,
|
| 13 |
+
"grad_norm": 1.6804444186183018,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.788,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06840530141085935,
|
| 20 |
+
"grad_norm": 1.2870091906780483,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.7236,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10260795211628901,
|
| 27 |
+
"grad_norm": 2.3640555348664254,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.6985,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.1368106028217187,
|
| 34 |
+
"grad_norm": 2.7430590909120953,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.6844,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.17101325352714836,
|
| 41 |
+
"grad_norm": 0.9746905467359548,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.6715,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.20521590423257802,
|
| 48 |
+
"grad_norm": 0.8507207569814836,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.6674,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.2394185549380077,
|
| 55 |
+
"grad_norm": 1.0995524010187212,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.6524,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2736212056434374,
|
| 62 |
+
"grad_norm": 0.6280153768789657,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.6362,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.30782385634886705,
|
| 69 |
+
"grad_norm": 0.5688437842480452,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.6299,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3420265070542967,
|
| 76 |
+
"grad_norm": 1.431931070075041,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.629,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3762291577597264,
|
| 83 |
+
"grad_norm": 1.6920363679606414,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.6307,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.41043180846515603,
|
| 90 |
+
"grad_norm": 0.580804701324865,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.6138,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.4446344591705857,
|
| 97 |
+
"grad_norm": 0.705027663006494,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.624,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4788371098760154,
|
| 104 |
+
"grad_norm": 0.7502862161170595,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6205,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.5130397605814451,
|
| 111 |
+
"grad_norm": 0.6708553413300853,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.616,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5472424112868748,
|
| 118 |
+
"grad_norm": 0.8494244770944251,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.627,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.5814450619923044,
|
| 125 |
+
"grad_norm": 0.53626174143875,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.6147,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.6156477126977341,
|
| 132 |
+
"grad_norm": 0.6364357464747067,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6195,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6498503634031637,
|
| 139 |
+
"grad_norm": 0.6082222049720536,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.6069,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.6840530141085934,
|
| 146 |
+
"grad_norm": 0.566328820213013,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6072,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.718255664814023,
|
| 153 |
+
"grad_norm": 0.49888078998310204,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.602,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7524583155194527,
|
| 160 |
+
"grad_norm": 0.4747529862310286,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6168,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7866609662248825,
|
| 167 |
+
"grad_norm": 0.5260744936940637,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.6046,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.8208636169303121,
|
| 174 |
+
"grad_norm": 0.5136960498148959,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6091,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.8550662676357418,
|
| 181 |
+
"grad_norm": 0.5616383990535931,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.6057,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.8892689183411714,
|
| 188 |
+
"grad_norm": 0.6884011630886375,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.6137,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.9234715690466011,
|
| 195 |
+
"grad_norm": 0.6448568689058315,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.6094,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.9576742197520308,
|
| 202 |
+
"grad_norm": 0.7693749759884462,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.6101,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9918768704574604,
|
| 209 |
+
"grad_norm": 0.5239302029558732,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.6053,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.9987174005985464,
|
| 216 |
+
"eval_loss": 0.607873260974884,
|
| 217 |
+
"eval_runtime": 205.6424,
|
| 218 |
+
"eval_samples_per_second": 38.299,
|
| 219 |
+
"eval_steps_per_second": 0.603,
|
| 220 |
+
"step": 292
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.0260795211628901,
|
| 224 |
+
"grad_norm": 0.5873382254143713,
|
| 225 |
+
"learning_rate": 5e-06,
|
| 226 |
+
"loss": 0.6057,
|
| 227 |
+
"step": 300
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.0602821718683197,
|
| 231 |
+
"grad_norm": 0.5561542049663418,
|
| 232 |
+
"learning_rate": 5e-06,
|
| 233 |
+
"loss": 0.5509,
|
| 234 |
+
"step": 310
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.0944848225737496,
|
| 238 |
+
"grad_norm": 0.580839172660231,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.5532,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.1286874732791792,
|
| 245 |
+
"grad_norm": 1.1795363810290622,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.5608,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.1628901239846088,
|
| 252 |
+
"grad_norm": 0.5245548599902146,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.551,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.1970927746900384,
|
| 259 |
+
"grad_norm": 0.7325061869705158,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.5542,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.2312954253954682,
|
| 266 |
+
"grad_norm": 0.5758834321007024,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.5569,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.2654980761008978,
|
| 273 |
+
"grad_norm": 0.5517090543506249,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.56,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.2997007268063274,
|
| 280 |
+
"grad_norm": 0.54026871791332,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.5567,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.3339033775117572,
|
| 287 |
+
"grad_norm": 0.6625845470909241,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.5539,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.3681060282171869,
|
| 294 |
+
"grad_norm": 0.6679250166336651,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.5546,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.4023086789226165,
|
| 301 |
+
"grad_norm": 0.43568230991463597,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.5548,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.436511329628046,
|
| 308 |
+
"grad_norm": 0.7673132417511955,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.5539,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.470713980333476,
|
| 315 |
+
"grad_norm": 0.8267199515857039,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.5609,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.5049166310389055,
|
| 322 |
+
"grad_norm": 0.7420049089951272,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.5526,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.5391192817443353,
|
| 329 |
+
"grad_norm": 0.5711771430617684,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.5459,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.573321932449765,
|
| 336 |
+
"grad_norm": 0.5059415250675989,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.5559,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.6075245831551945,
|
| 343 |
+
"grad_norm": 0.5634757605615536,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.5527,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.6417272338606241,
|
| 350 |
+
"grad_norm": 0.47121211159709975,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.5586,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6759298845660537,
|
| 357 |
+
"grad_norm": 0.5677050425607371,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.5598,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.7101325352714836,
|
| 364 |
+
"grad_norm": 0.610525785684426,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.5509,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.7443351859769132,
|
| 371 |
+
"grad_norm": 0.6011597954395561,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.5551,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.778537836682343,
|
| 378 |
+
"grad_norm": 0.4613495220552048,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.5492,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.8127404873877726,
|
| 385 |
+
"grad_norm": 0.5256335189092972,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.5528,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.8469431380932022,
|
| 392 |
+
"grad_norm": 0.4659992912811747,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.5438,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.8811457887986318,
|
| 399 |
+
"grad_norm": 0.49388622607101107,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.5516,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.9153484395040614,
|
| 406 |
+
"grad_norm": 0.49235853413145225,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.5499,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.9495510902094912,
|
| 413 |
+
"grad_norm": 0.48530934948386684,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.5491,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.983753740914921,
|
| 420 |
+
"grad_norm": 0.681088284788218,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.5603,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.9974348011970928,
|
| 427 |
+
"eval_loss": 0.5989111065864563,
|
| 428 |
+
"eval_runtime": 198.5327,
|
| 429 |
+
"eval_samples_per_second": 39.671,
|
| 430 |
+
"eval_steps_per_second": 0.625,
|
| 431 |
+
"step": 584
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.0179563916203507,
|
| 435 |
+
"grad_norm": 0.6929323229832084,
|
| 436 |
+
"learning_rate": 5e-06,
|
| 437 |
+
"loss": 0.5622,
|
| 438 |
+
"step": 590
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.0521590423257803,
|
| 442 |
+
"grad_norm": 0.5875304861764742,
|
| 443 |
+
"learning_rate": 5e-06,
|
| 444 |
+
"loss": 0.4938,
|
| 445 |
+
"step": 600
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.08636169303121,
|
| 449 |
+
"grad_norm": 0.6162810457588677,
|
| 450 |
+
"learning_rate": 5e-06,
|
| 451 |
+
"loss": 0.5075,
|
| 452 |
+
"step": 610
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.1205643437366395,
|
| 456 |
+
"grad_norm": 0.5609910238130167,
|
| 457 |
+
"learning_rate": 5e-06,
|
| 458 |
+
"loss": 0.4983,
|
| 459 |
+
"step": 620
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.154766994442069,
|
| 463 |
+
"grad_norm": 0.5851708963453293,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.5001,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.188969645147499,
|
| 470 |
+
"grad_norm": 0.5635902260827742,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.5018,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.2231722958529287,
|
| 477 |
+
"grad_norm": 0.7189631820244675,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.4981,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.2573749465583584,
|
| 484 |
+
"grad_norm": 0.5675999743836657,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.4955,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.291577597263788,
|
| 491 |
+
"grad_norm": 0.5230970038040867,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.5011,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.3257802479692176,
|
| 498 |
+
"grad_norm": 0.5536244794373915,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.5016,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.359982898674647,
|
| 505 |
+
"grad_norm": 0.4987813092132446,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.5107,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.3941855493800768,
|
| 512 |
+
"grad_norm": 0.5979352098881308,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.5026,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.428388200085507,
|
| 519 |
+
"grad_norm": 0.508657041803482,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.5053,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.4625908507909364,
|
| 526 |
+
"grad_norm": 0.5418322815471002,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.4999,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.496793501496366,
|
| 533 |
+
"grad_norm": 2.189328511686068,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.5039,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.5309961522017956,
|
| 540 |
+
"grad_norm": 0.5032483059171474,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.5086,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.5651988029072252,
|
| 547 |
+
"grad_norm": 0.6897627250322256,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.4967,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.599401453612655,
|
| 554 |
+
"grad_norm": 0.6125595774876328,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.5082,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.633604104318085,
|
| 561 |
+
"grad_norm": 0.5478411600953903,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.5008,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.6678067550235145,
|
| 568 |
+
"grad_norm": 0.4805016073434078,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.5015,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.702009405728944,
|
| 575 |
+
"grad_norm": 0.5561978743904621,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.5045,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.7362120564343737,
|
| 582 |
+
"grad_norm": 0.4899190841497652,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.5024,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.7704147071398033,
|
| 589 |
+
"grad_norm": 0.513594637059708,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.4987,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.804617357845233,
|
| 596 |
+
"grad_norm": 0.5314442863240219,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.5077,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.8388200085506625,
|
| 603 |
+
"grad_norm": 0.5397227572021858,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.5084,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.873022659256092,
|
| 610 |
+
"grad_norm": 0.5577858554366566,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.5085,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.907225309961522,
|
| 617 |
+
"grad_norm": 0.5436188950593666,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.51,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.941427960666952,
|
| 624 |
+
"grad_norm": 0.5548892577477849,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.5111,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.9756306113723814,
|
| 631 |
+
"grad_norm": 0.5037692940815492,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.5083,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.9961522017956392,
|
| 638 |
+
"eval_loss": 0.6033630967140198,
|
| 639 |
+
"eval_runtime": 196.2227,
|
| 640 |
+
"eval_samples_per_second": 40.138,
|
| 641 |
+
"eval_steps_per_second": 0.632,
|
| 642 |
+
"step": 876
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.9961522017956392,
|
| 646 |
+
"step": 876,
|
| 647 |
+
"total_flos": 1467123247349760.0,
|
| 648 |
+
"train_loss": 0.5650538080903493,
|
| 649 |
+
"train_runtime": 28810.5739,
|
| 650 |
+
"train_samples_per_second": 15.582,
|
| 651 |
+
"train_steps_per_second": 0.03
|
| 652 |
+
}
|
| 653 |
+
],
|
| 654 |
+
"logging_steps": 10,
|
| 655 |
+
"max_steps": 876,
|
| 656 |
+
"num_input_tokens_seen": 0,
|
| 657 |
+
"num_train_epochs": 3,
|
| 658 |
+
"save_steps": 500,
|
| 659 |
+
"stateful_callbacks": {
|
| 660 |
+
"TrainerControl": {
|
| 661 |
+
"args": {
|
| 662 |
+
"should_epoch_stop": false,
|
| 663 |
+
"should_evaluate": false,
|
| 664 |
+
"should_log": false,
|
| 665 |
+
"should_save": true,
|
| 666 |
+
"should_training_stop": true
|
| 667 |
+
},
|
| 668 |
+
"attributes": {}
|
| 669 |
+
}
|
| 670 |
+
},
|
| 671 |
+
"total_flos": 1467123247349760.0,
|
| 672 |
+
"train_batch_size": 8,
|
| 673 |
+
"trial_name": null,
|
| 674 |
+
"trial_params": null
|
| 675 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|