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 +717 -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: llama3-1_8b_oh_v3.1_wo_glaive_code_assistant
|
|
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# llama3-1_8b_oh_v3.1_wo_glaive_code_assistant
|
| 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: 0.6763
|
| 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: llama3-1_8b_oh_v3.1_wo_glaive_code_assistant
|
|
|
|
| 16 |
|
| 17 |
# llama3-1_8b_oh_v3.1_wo_glaive_code_assistant
|
| 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_v3.1_wo_glaive_code_assistant dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
- Loss: 0.6763
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.6763173341751099,
|
| 4 |
+
"eval_runtime": 29.5942,
|
| 5 |
+
"eval_samples_per_second": 284.313,
|
| 6 |
+
"eval_steps_per_second": 1.115,
|
| 7 |
+
"total_flos": 1572859973468160.0,
|
| 8 |
+
"train_loss": 0.648996304907103,
|
| 9 |
+
"train_runtime": 5978.5297,
|
| 10 |
+
"train_samples_per_second": 80.212,
|
| 11 |
+
"train_steps_per_second": 0.157
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"eval_loss": 0.6763173341751099,
|
| 4 |
+
"eval_runtime": 29.5942,
|
| 5 |
+
"eval_samples_per_second": 284.313,
|
| 6 |
+
"eval_steps_per_second": 1.115
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 1572859973468160.0,
|
| 4 |
+
"train_loss": 0.648996304907103,
|
| 5 |
+
"train_runtime": 5978.5297,
|
| 6 |
+
"train_samples_per_second": 80.212,
|
| 7 |
+
"train_steps_per_second": 0.157
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,717 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 939,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03194888178913738,
|
| 13 |
+
"grad_norm": 4.9730574359401585,
|
| 14 |
+
"learning_rate": 5e-06,
|
| 15 |
+
"loss": 0.9178,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06389776357827476,
|
| 20 |
+
"grad_norm": 2.414364422517365,
|
| 21 |
+
"learning_rate": 5e-06,
|
| 22 |
+
"loss": 0.8185,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.09584664536741214,
|
| 27 |
+
"grad_norm": 2.355627570386528,
|
| 28 |
+
"learning_rate": 5e-06,
|
| 29 |
+
"loss": 0.7955,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.12779552715654952,
|
| 34 |
+
"grad_norm": 0.9368274495065052,
|
| 35 |
+
"learning_rate": 5e-06,
|
| 36 |
+
"loss": 0.777,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.1597444089456869,
|
| 41 |
+
"grad_norm": 1.0303864547714687,
|
| 42 |
+
"learning_rate": 5e-06,
|
| 43 |
+
"loss": 0.7544,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.19169329073482427,
|
| 48 |
+
"grad_norm": 0.8799746481470822,
|
| 49 |
+
"learning_rate": 5e-06,
|
| 50 |
+
"loss": 0.7414,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.22364217252396165,
|
| 55 |
+
"grad_norm": 2.571145704302205,
|
| 56 |
+
"learning_rate": 5e-06,
|
| 57 |
+
"loss": 0.7335,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.25559105431309903,
|
| 62 |
+
"grad_norm": 1.4093624036341388,
|
| 63 |
+
"learning_rate": 5e-06,
|
| 64 |
+
"loss": 0.7174,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.28753993610223644,
|
| 69 |
+
"grad_norm": 0.8615478628829397,
|
| 70 |
+
"learning_rate": 5e-06,
|
| 71 |
+
"loss": 0.7318,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.3194888178913738,
|
| 76 |
+
"grad_norm": 0.5831162653054314,
|
| 77 |
+
"learning_rate": 5e-06,
|
| 78 |
+
"loss": 0.72,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.3514376996805112,
|
| 83 |
+
"grad_norm": 0.6173922126615862,
|
| 84 |
+
"learning_rate": 5e-06,
|
| 85 |
+
"loss": 0.716,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.38338658146964855,
|
| 90 |
+
"grad_norm": 0.8888214325930446,
|
| 91 |
+
"learning_rate": 5e-06,
|
| 92 |
+
"loss": 0.7134,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.41533546325878595,
|
| 97 |
+
"grad_norm": 0.6979078907766634,
|
| 98 |
+
"learning_rate": 5e-06,
|
| 99 |
+
"loss": 0.7021,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.4472843450479233,
|
| 104 |
+
"grad_norm": 0.6491577783946951,
|
| 105 |
+
"learning_rate": 5e-06,
|
| 106 |
+
"loss": 0.6991,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.4792332268370607,
|
| 111 |
+
"grad_norm": 0.5395727376776203,
|
| 112 |
+
"learning_rate": 5e-06,
|
| 113 |
+
"loss": 0.7068,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.5111821086261981,
|
| 118 |
+
"grad_norm": 0.689753166346745,
|
| 119 |
+
"learning_rate": 5e-06,
|
| 120 |
+
"loss": 0.6935,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.5431309904153354,
|
| 125 |
+
"grad_norm": 0.5521859868964065,
|
| 126 |
+
"learning_rate": 5e-06,
|
| 127 |
+
"loss": 0.705,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.5750798722044729,
|
| 132 |
+
"grad_norm": 0.5708494730680989,
|
| 133 |
+
"learning_rate": 5e-06,
|
| 134 |
+
"loss": 0.6961,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6070287539936102,
|
| 139 |
+
"grad_norm": 0.7742403437422439,
|
| 140 |
+
"learning_rate": 5e-06,
|
| 141 |
+
"loss": 0.7045,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.6389776357827476,
|
| 146 |
+
"grad_norm": 0.6022796802365044,
|
| 147 |
+
"learning_rate": 5e-06,
|
| 148 |
+
"loss": 0.6911,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.670926517571885,
|
| 153 |
+
"grad_norm": 0.5364796054217524,
|
| 154 |
+
"learning_rate": 5e-06,
|
| 155 |
+
"loss": 0.695,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.7028753993610224,
|
| 160 |
+
"grad_norm": 0.47930189735330775,
|
| 161 |
+
"learning_rate": 5e-06,
|
| 162 |
+
"loss": 0.6961,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.7348242811501597,
|
| 167 |
+
"grad_norm": 0.5376571771457028,
|
| 168 |
+
"learning_rate": 5e-06,
|
| 169 |
+
"loss": 0.7035,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.7667731629392971,
|
| 174 |
+
"grad_norm": 0.7570921090383397,
|
| 175 |
+
"learning_rate": 5e-06,
|
| 176 |
+
"loss": 0.6801,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.7987220447284346,
|
| 181 |
+
"grad_norm": 0.5375423682771385,
|
| 182 |
+
"learning_rate": 5e-06,
|
| 183 |
+
"loss": 0.6809,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.8306709265175719,
|
| 188 |
+
"grad_norm": 0.5836884654889642,
|
| 189 |
+
"learning_rate": 5e-06,
|
| 190 |
+
"loss": 0.6864,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.8626198083067093,
|
| 195 |
+
"grad_norm": 0.598608895789362,
|
| 196 |
+
"learning_rate": 5e-06,
|
| 197 |
+
"loss": 0.6902,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.8945686900958466,
|
| 202 |
+
"grad_norm": 0.5331459973786111,
|
| 203 |
+
"learning_rate": 5e-06,
|
| 204 |
+
"loss": 0.6879,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9265175718849841,
|
| 209 |
+
"grad_norm": 0.5067341091409512,
|
| 210 |
+
"learning_rate": 5e-06,
|
| 211 |
+
"loss": 0.6881,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.9584664536741214,
|
| 216 |
+
"grad_norm": 0.5948122722251207,
|
| 217 |
+
"learning_rate": 5e-06,
|
| 218 |
+
"loss": 0.6868,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.9904153354632588,
|
| 223 |
+
"grad_norm": 0.6883144495229085,
|
| 224 |
+
"learning_rate": 5e-06,
|
| 225 |
+
"loss": 0.6818,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.0,
|
| 230 |
+
"eval_loss": 0.680176854133606,
|
| 231 |
+
"eval_runtime": 30.2696,
|
| 232 |
+
"eval_samples_per_second": 277.968,
|
| 233 |
+
"eval_steps_per_second": 1.09,
|
| 234 |
+
"step": 313
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.0223642172523961,
|
| 238 |
+
"grad_norm": 0.7468894293006159,
|
| 239 |
+
"learning_rate": 5e-06,
|
| 240 |
+
"loss": 0.656,
|
| 241 |
+
"step": 320
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.0543130990415335,
|
| 245 |
+
"grad_norm": 0.6942190077761362,
|
| 246 |
+
"learning_rate": 5e-06,
|
| 247 |
+
"loss": 0.6387,
|
| 248 |
+
"step": 330
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.0862619808306708,
|
| 252 |
+
"grad_norm": 0.6242193449068775,
|
| 253 |
+
"learning_rate": 5e-06,
|
| 254 |
+
"loss": 0.6407,
|
| 255 |
+
"step": 340
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.1182108626198084,
|
| 259 |
+
"grad_norm": 0.6329861406612104,
|
| 260 |
+
"learning_rate": 5e-06,
|
| 261 |
+
"loss": 0.6428,
|
| 262 |
+
"step": 350
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.1501597444089458,
|
| 266 |
+
"grad_norm": 0.6142460964851904,
|
| 267 |
+
"learning_rate": 5e-06,
|
| 268 |
+
"loss": 0.6356,
|
| 269 |
+
"step": 360
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.182108626198083,
|
| 273 |
+
"grad_norm": 0.6850789950403501,
|
| 274 |
+
"learning_rate": 5e-06,
|
| 275 |
+
"loss": 0.636,
|
| 276 |
+
"step": 370
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.2140575079872205,
|
| 280 |
+
"grad_norm": 0.5355828508870474,
|
| 281 |
+
"learning_rate": 5e-06,
|
| 282 |
+
"loss": 0.6382,
|
| 283 |
+
"step": 380
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.2460063897763578,
|
| 287 |
+
"grad_norm": 0.5629126693045496,
|
| 288 |
+
"learning_rate": 5e-06,
|
| 289 |
+
"loss": 0.6344,
|
| 290 |
+
"step": 390
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.2779552715654952,
|
| 294 |
+
"grad_norm": 0.5204488937445614,
|
| 295 |
+
"learning_rate": 5e-06,
|
| 296 |
+
"loss": 0.6401,
|
| 297 |
+
"step": 400
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.3099041533546325,
|
| 301 |
+
"grad_norm": 0.5396176523426488,
|
| 302 |
+
"learning_rate": 5e-06,
|
| 303 |
+
"loss": 0.6382,
|
| 304 |
+
"step": 410
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.34185303514377,
|
| 308 |
+
"grad_norm": 0.6023672235090743,
|
| 309 |
+
"learning_rate": 5e-06,
|
| 310 |
+
"loss": 0.6417,
|
| 311 |
+
"step": 420
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.3738019169329074,
|
| 315 |
+
"grad_norm": 0.5459880912136352,
|
| 316 |
+
"learning_rate": 5e-06,
|
| 317 |
+
"loss": 0.6434,
|
| 318 |
+
"step": 430
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 1.4057507987220448,
|
| 322 |
+
"grad_norm": 0.5092923506095731,
|
| 323 |
+
"learning_rate": 5e-06,
|
| 324 |
+
"loss": 0.6366,
|
| 325 |
+
"step": 440
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.4376996805111821,
|
| 329 |
+
"grad_norm": 0.6382479698360073,
|
| 330 |
+
"learning_rate": 5e-06,
|
| 331 |
+
"loss": 0.6353,
|
| 332 |
+
"step": 450
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 1.4696485623003195,
|
| 336 |
+
"grad_norm": 0.7533985947305522,
|
| 337 |
+
"learning_rate": 5e-06,
|
| 338 |
+
"loss": 0.6392,
|
| 339 |
+
"step": 460
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 1.5015974440894568,
|
| 343 |
+
"grad_norm": 0.7229864015275371,
|
| 344 |
+
"learning_rate": 5e-06,
|
| 345 |
+
"loss": 0.6381,
|
| 346 |
+
"step": 470
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.5335463258785942,
|
| 350 |
+
"grad_norm": 0.5758239404648058,
|
| 351 |
+
"learning_rate": 5e-06,
|
| 352 |
+
"loss": 0.636,
|
| 353 |
+
"step": 480
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.5654952076677318,
|
| 357 |
+
"grad_norm": 0.5422584017870065,
|
| 358 |
+
"learning_rate": 5e-06,
|
| 359 |
+
"loss": 0.6393,
|
| 360 |
+
"step": 490
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.5974440894568689,
|
| 364 |
+
"grad_norm": 0.500259651199132,
|
| 365 |
+
"learning_rate": 5e-06,
|
| 366 |
+
"loss": 0.6441,
|
| 367 |
+
"step": 500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.6293929712460065,
|
| 371 |
+
"grad_norm": 0.8813431746686585,
|
| 372 |
+
"learning_rate": 5e-06,
|
| 373 |
+
"loss": 0.6308,
|
| 374 |
+
"step": 510
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.6613418530351438,
|
| 378 |
+
"grad_norm": 0.49923586083798,
|
| 379 |
+
"learning_rate": 5e-06,
|
| 380 |
+
"loss": 0.6365,
|
| 381 |
+
"step": 520
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.6932907348242812,
|
| 385 |
+
"grad_norm": 0.46329712814558677,
|
| 386 |
+
"learning_rate": 5e-06,
|
| 387 |
+
"loss": 0.6279,
|
| 388 |
+
"step": 530
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.7252396166134185,
|
| 392 |
+
"grad_norm": 0.7606638693711257,
|
| 393 |
+
"learning_rate": 5e-06,
|
| 394 |
+
"loss": 0.6341,
|
| 395 |
+
"step": 540
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.7571884984025559,
|
| 399 |
+
"grad_norm": 0.5490152416615772,
|
| 400 |
+
"learning_rate": 5e-06,
|
| 401 |
+
"loss": 0.6378,
|
| 402 |
+
"step": 550
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.7891373801916934,
|
| 406 |
+
"grad_norm": 0.5845892096604395,
|
| 407 |
+
"learning_rate": 5e-06,
|
| 408 |
+
"loss": 0.6395,
|
| 409 |
+
"step": 560
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 1.8210862619808306,
|
| 413 |
+
"grad_norm": 0.5147171055711276,
|
| 414 |
+
"learning_rate": 5e-06,
|
| 415 |
+
"loss": 0.6297,
|
| 416 |
+
"step": 570
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.8530351437699681,
|
| 420 |
+
"grad_norm": 0.5052763678219258,
|
| 421 |
+
"learning_rate": 5e-06,
|
| 422 |
+
"loss": 0.6342,
|
| 423 |
+
"step": 580
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 1.8849840255591053,
|
| 427 |
+
"grad_norm": 0.5390734384968956,
|
| 428 |
+
"learning_rate": 5e-06,
|
| 429 |
+
"loss": 0.6369,
|
| 430 |
+
"step": 590
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 1.9169329073482428,
|
| 434 |
+
"grad_norm": 0.6413516519865096,
|
| 435 |
+
"learning_rate": 5e-06,
|
| 436 |
+
"loss": 0.6418,
|
| 437 |
+
"step": 600
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 1.9488817891373802,
|
| 441 |
+
"grad_norm": 0.5239174298332788,
|
| 442 |
+
"learning_rate": 5e-06,
|
| 443 |
+
"loss": 0.6274,
|
| 444 |
+
"step": 610
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 1.9808306709265175,
|
| 448 |
+
"grad_norm": 0.7039020492664947,
|
| 449 |
+
"learning_rate": 5e-06,
|
| 450 |
+
"loss": 0.6385,
|
| 451 |
+
"step": 620
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.0,
|
| 455 |
+
"eval_loss": 0.6707362532615662,
|
| 456 |
+
"eval_runtime": 29.7735,
|
| 457 |
+
"eval_samples_per_second": 282.6,
|
| 458 |
+
"eval_steps_per_second": 1.108,
|
| 459 |
+
"step": 626
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.012779552715655,
|
| 463 |
+
"grad_norm": 0.9555964669975626,
|
| 464 |
+
"learning_rate": 5e-06,
|
| 465 |
+
"loss": 0.6152,
|
| 466 |
+
"step": 630
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.0447284345047922,
|
| 470 |
+
"grad_norm": 0.6929923210576585,
|
| 471 |
+
"learning_rate": 5e-06,
|
| 472 |
+
"loss": 0.5881,
|
| 473 |
+
"step": 640
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.07667731629393,
|
| 477 |
+
"grad_norm": 0.5186537877634542,
|
| 478 |
+
"learning_rate": 5e-06,
|
| 479 |
+
"loss": 0.5871,
|
| 480 |
+
"step": 650
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.108626198083067,
|
| 484 |
+
"grad_norm": 0.5532963474572669,
|
| 485 |
+
"learning_rate": 5e-06,
|
| 486 |
+
"loss": 0.5896,
|
| 487 |
+
"step": 660
|
| 488 |
+
},
|
| 489 |
+
{
|
| 490 |
+
"epoch": 2.1405750798722045,
|
| 491 |
+
"grad_norm": 0.5667781850304423,
|
| 492 |
+
"learning_rate": 5e-06,
|
| 493 |
+
"loss": 0.589,
|
| 494 |
+
"step": 670
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.1725239616613417,
|
| 498 |
+
"grad_norm": 0.5285098402420302,
|
| 499 |
+
"learning_rate": 5e-06,
|
| 500 |
+
"loss": 0.5901,
|
| 501 |
+
"step": 680
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 2.2044728434504792,
|
| 505 |
+
"grad_norm": 0.536646702155888,
|
| 506 |
+
"learning_rate": 5e-06,
|
| 507 |
+
"loss": 0.5863,
|
| 508 |
+
"step": 690
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 2.236421725239617,
|
| 512 |
+
"grad_norm": 0.5015237529663761,
|
| 513 |
+
"learning_rate": 5e-06,
|
| 514 |
+
"loss": 0.5846,
|
| 515 |
+
"step": 700
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.268370607028754,
|
| 519 |
+
"grad_norm": 0.6152245702781334,
|
| 520 |
+
"learning_rate": 5e-06,
|
| 521 |
+
"loss": 0.5885,
|
| 522 |
+
"step": 710
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.3003194888178915,
|
| 526 |
+
"grad_norm": 0.6897591535184826,
|
| 527 |
+
"learning_rate": 5e-06,
|
| 528 |
+
"loss": 0.584,
|
| 529 |
+
"step": 720
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 2.3322683706070286,
|
| 533 |
+
"grad_norm": 0.48574756268499075,
|
| 534 |
+
"learning_rate": 5e-06,
|
| 535 |
+
"loss": 0.5888,
|
| 536 |
+
"step": 730
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 2.364217252396166,
|
| 540 |
+
"grad_norm": 0.7776088479205505,
|
| 541 |
+
"learning_rate": 5e-06,
|
| 542 |
+
"loss": 0.591,
|
| 543 |
+
"step": 740
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 2.3961661341853033,
|
| 547 |
+
"grad_norm": 0.5643419759040877,
|
| 548 |
+
"learning_rate": 5e-06,
|
| 549 |
+
"loss": 0.5975,
|
| 550 |
+
"step": 750
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 2.428115015974441,
|
| 554 |
+
"grad_norm": 0.7618894474745995,
|
| 555 |
+
"learning_rate": 5e-06,
|
| 556 |
+
"loss": 0.5907,
|
| 557 |
+
"step": 760
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 2.460063897763578,
|
| 561 |
+
"grad_norm": 0.566521654030305,
|
| 562 |
+
"learning_rate": 5e-06,
|
| 563 |
+
"loss": 0.5902,
|
| 564 |
+
"step": 770
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.4920127795527156,
|
| 568 |
+
"grad_norm": 0.539754953993636,
|
| 569 |
+
"learning_rate": 5e-06,
|
| 570 |
+
"loss": 0.5884,
|
| 571 |
+
"step": 780
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.523961661341853,
|
| 575 |
+
"grad_norm": 0.5625428853361434,
|
| 576 |
+
"learning_rate": 5e-06,
|
| 577 |
+
"loss": 0.5818,
|
| 578 |
+
"step": 790
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.5559105431309903,
|
| 582 |
+
"grad_norm": 0.6279591943173818,
|
| 583 |
+
"learning_rate": 5e-06,
|
| 584 |
+
"loss": 0.588,
|
| 585 |
+
"step": 800
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.587859424920128,
|
| 589 |
+
"grad_norm": 0.6224175939518917,
|
| 590 |
+
"learning_rate": 5e-06,
|
| 591 |
+
"loss": 0.5934,
|
| 592 |
+
"step": 810
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 2.619808306709265,
|
| 596 |
+
"grad_norm": 0.5642613523825399,
|
| 597 |
+
"learning_rate": 5e-06,
|
| 598 |
+
"loss": 0.5865,
|
| 599 |
+
"step": 820
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 2.6517571884984026,
|
| 603 |
+
"grad_norm": 0.6865432310936253,
|
| 604 |
+
"learning_rate": 5e-06,
|
| 605 |
+
"loss": 0.5926,
|
| 606 |
+
"step": 830
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.68370607028754,
|
| 610 |
+
"grad_norm": 0.5373522424367686,
|
| 611 |
+
"learning_rate": 5e-06,
|
| 612 |
+
"loss": 0.5921,
|
| 613 |
+
"step": 840
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 2.7156549520766773,
|
| 617 |
+
"grad_norm": 0.567457215981538,
|
| 618 |
+
"learning_rate": 5e-06,
|
| 619 |
+
"loss": 0.5954,
|
| 620 |
+
"step": 850
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.747603833865815,
|
| 624 |
+
"grad_norm": 0.6426061724785515,
|
| 625 |
+
"learning_rate": 5e-06,
|
| 626 |
+
"loss": 0.5988,
|
| 627 |
+
"step": 860
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 2.779552715654952,
|
| 631 |
+
"grad_norm": 0.59442339830324,
|
| 632 |
+
"learning_rate": 5e-06,
|
| 633 |
+
"loss": 0.6008,
|
| 634 |
+
"step": 870
|
| 635 |
+
},
|
| 636 |
+
{
|
| 637 |
+
"epoch": 2.8115015974440896,
|
| 638 |
+
"grad_norm": 0.5592579452365836,
|
| 639 |
+
"learning_rate": 5e-06,
|
| 640 |
+
"loss": 0.6022,
|
| 641 |
+
"step": 880
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.8434504792332267,
|
| 645 |
+
"grad_norm": 0.5094979594863669,
|
| 646 |
+
"learning_rate": 5e-06,
|
| 647 |
+
"loss": 0.592,
|
| 648 |
+
"step": 890
|
| 649 |
+
},
|
| 650 |
+
{
|
| 651 |
+
"epoch": 2.8753993610223643,
|
| 652 |
+
"grad_norm": 0.5091387458726695,
|
| 653 |
+
"learning_rate": 5e-06,
|
| 654 |
+
"loss": 0.6,
|
| 655 |
+
"step": 900
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.9073482428115014,
|
| 659 |
+
"grad_norm": 0.5094298110498509,
|
| 660 |
+
"learning_rate": 5e-06,
|
| 661 |
+
"loss": 0.5938,
|
| 662 |
+
"step": 910
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.939297124600639,
|
| 666 |
+
"grad_norm": 0.5130578982623665,
|
| 667 |
+
"learning_rate": 5e-06,
|
| 668 |
+
"loss": 0.586,
|
| 669 |
+
"step": 920
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 2.9712460063897765,
|
| 673 |
+
"grad_norm": 0.4928009047818368,
|
| 674 |
+
"learning_rate": 5e-06,
|
| 675 |
+
"loss": 0.5911,
|
| 676 |
+
"step": 930
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 3.0,
|
| 680 |
+
"eval_loss": 0.6763173341751099,
|
| 681 |
+
"eval_runtime": 29.5519,
|
| 682 |
+
"eval_samples_per_second": 284.719,
|
| 683 |
+
"eval_steps_per_second": 1.117,
|
| 684 |
+
"step": 939
|
| 685 |
+
},
|
| 686 |
+
{
|
| 687 |
+
"epoch": 3.0,
|
| 688 |
+
"step": 939,
|
| 689 |
+
"total_flos": 1572859973468160.0,
|
| 690 |
+
"train_loss": 0.648996304907103,
|
| 691 |
+
"train_runtime": 5978.5297,
|
| 692 |
+
"train_samples_per_second": 80.212,
|
| 693 |
+
"train_steps_per_second": 0.157
|
| 694 |
+
}
|
| 695 |
+
],
|
| 696 |
+
"logging_steps": 10,
|
| 697 |
+
"max_steps": 939,
|
| 698 |
+
"num_input_tokens_seen": 0,
|
| 699 |
+
"num_train_epochs": 3,
|
| 700 |
+
"save_steps": 500,
|
| 701 |
+
"stateful_callbacks": {
|
| 702 |
+
"TrainerControl": {
|
| 703 |
+
"args": {
|
| 704 |
+
"should_epoch_stop": false,
|
| 705 |
+
"should_evaluate": false,
|
| 706 |
+
"should_log": false,
|
| 707 |
+
"should_save": true,
|
| 708 |
+
"should_training_stop": true
|
| 709 |
+
},
|
| 710 |
+
"attributes": {}
|
| 711 |
+
}
|
| 712 |
+
},
|
| 713 |
+
"total_flos": 1572859973468160.0,
|
| 714 |
+
"train_batch_size": 16,
|
| 715 |
+
"trial_name": null,
|
| 716 |
+
"trial_params": null
|
| 717 |
+
}
|
training_eval_loss.png
ADDED
|
training_loss.png
ADDED
|