Model save
Browse files- README.md +7 -4
- all_results.json +5 -5
- lora_4/adapter_config.json +4 -4
- lora_4/adapter_model.safetensors +1 -1
- train_results.json +5 -5
- trainer_state.json +1136 -56
README.md
CHANGED
|
@@ -2,12 +2,12 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
| 5 |
-
- alignment-handbook
|
| 6 |
- trl
|
| 7 |
- sft
|
|
|
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
-
-
|
| 11 |
base_model: mistralai/Mistral-7B-v0.1
|
| 12 |
model-index:
|
| 13 |
- name: zephyr-7b-sft-qlora
|
|
@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 19 |
|
| 20 |
# zephyr-7b-sft-qlora
|
| 21 |
|
| 22 |
-
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
@@ -43,10 +43,13 @@ The following hyperparameters were used during training:
|
|
| 43 |
- eval_batch_size: 4
|
| 44 |
- seed: 42
|
| 45 |
- distributed_type: multi-GPU
|
|
|
|
|
|
|
|
|
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: cosine
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
-
- training_steps:
|
| 50 |
- mixed_precision_training: Native AMP
|
| 51 |
|
| 52 |
### Training results
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
library_name: peft
|
| 4 |
tags:
|
|
|
|
| 5 |
- trl
|
| 6 |
- sft
|
| 7 |
+
- alignment-handbook
|
| 8 |
- generated_from_trainer
|
| 9 |
datasets:
|
| 10 |
+
- generator
|
| 11 |
base_model: mistralai/Mistral-7B-v0.1
|
| 12 |
model-index:
|
| 13 |
- name: zephyr-7b-sft-qlora
|
|
|
|
| 19 |
|
| 20 |
# zephyr-7b-sft-qlora
|
| 21 |
|
| 22 |
+
This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 43 |
- eval_batch_size: 4
|
| 44 |
- seed: 42
|
| 45 |
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 4
|
| 47 |
+
- total_train_batch_size: 8
|
| 48 |
+
- total_eval_batch_size: 16
|
| 49 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
- lr_scheduler_type: cosine
|
| 51 |
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- training_steps: 1000
|
| 53 |
- mixed_precision_training: Native AMP
|
| 54 |
|
| 55 |
### Training results
|
all_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 207865,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.06,
|
| 3 |
+
"train_loss": -130335.20808300782,
|
| 4 |
+
"train_runtime": 5587.0346,
|
| 5 |
"train_samples": 207865,
|
| 6 |
+
"train_samples_per_second": 1.432,
|
| 7 |
+
"train_steps_per_second": 0.179
|
| 8 |
}
|
lora_4/adapter_config.json
CHANGED
|
@@ -23,12 +23,12 @@
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
| 26 |
-
"o_proj",
|
| 27 |
-
"up_proj",
|
| 28 |
-
"q_proj",
|
| 29 |
"down_proj",
|
| 30 |
-
"gate_proj",
|
| 31 |
"k_proj",
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
"v_proj"
|
| 33 |
],
|
| 34 |
"task_type": null,
|
|
|
|
| 23 |
"rank_pattern": {},
|
| 24 |
"revision": null,
|
| 25 |
"target_modules": [
|
|
|
|
|
|
|
|
|
|
| 26 |
"down_proj",
|
|
|
|
| 27 |
"k_proj",
|
| 28 |
+
"gate_proj",
|
| 29 |
+
"q_proj",
|
| 30 |
+
"o_proj",
|
| 31 |
+
"up_proj",
|
| 32 |
"v_proj"
|
| 33 |
],
|
| 34 |
"task_type": null,
|
lora_4/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 167832240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:69fd14c7e249a3d390d5f29ab54efb1b8ef865e9f2b0e88fce88a3d087c69f71
|
| 3 |
size 167832240
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch": 0.
|
| 3 |
-
"train_loss":
|
| 4 |
-
"train_runtime":
|
| 5 |
"train_samples": 207865,
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 0.06,
|
| 3 |
+
"train_loss": -130335.20808300782,
|
| 4 |
+
"train_runtime": 5587.0346,
|
| 5 |
"train_samples": 207865,
|
| 6 |
+
"train_samples_per_second": 1.432,
|
| 7 |
+
"train_steps_per_second": 0.179
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,155 +1,1235 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.0,
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
"step": 5
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.0,
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss":
|
| 27 |
"step": 10
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.0,
|
| 31 |
-
"learning_rate":
|
| 32 |
-
"loss":
|
| 33 |
"step": 15
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.0,
|
| 37 |
-
"learning_rate":
|
| 38 |
-
"loss":
|
| 39 |
"step": 20
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.0,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss":
|
| 45 |
"step": 25
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.0,
|
| 49 |
-
"learning_rate":
|
| 50 |
-
"loss":
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.0,
|
| 55 |
-
"learning_rate":
|
| 56 |
-
"loss":
|
| 57 |
"step": 35
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.0,
|
| 61 |
-
"learning_rate":
|
| 62 |
-
"loss":
|
| 63 |
"step": 40
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.0,
|
| 67 |
-
"learning_rate":
|
| 68 |
-
"loss":
|
| 69 |
"step": 45
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.0,
|
| 73 |
-
"learning_rate": 0.
|
| 74 |
-
"loss":
|
| 75 |
"step": 50
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.0,
|
| 79 |
-
"learning_rate": 0.
|
| 80 |
-
"loss":
|
| 81 |
"step": 55
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.0,
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss":
|
| 87 |
"step": 60
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.0,
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss":
|
| 93 |
"step": 65
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.0,
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss":
|
| 99 |
"step": 70
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.0,
|
| 103 |
-
"learning_rate":
|
| 104 |
-
"loss":
|
| 105 |
"step": 75
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.0,
|
| 109 |
-
"learning_rate":
|
| 110 |
-
"loss":
|
| 111 |
"step": 80
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.0,
|
| 115 |
-
"learning_rate":
|
| 116 |
-
"loss":
|
| 117 |
"step": 85
|
| 118 |
},
|
| 119 |
{
|
| 120 |
-
"epoch": 0.
|
| 121 |
-
"learning_rate":
|
| 122 |
-
"loss":
|
| 123 |
"step": 90
|
| 124 |
},
|
| 125 |
{
|
| 126 |
-
"epoch": 0.
|
| 127 |
-
"learning_rate":
|
| 128 |
-
"loss":
|
| 129 |
"step": 95
|
| 130 |
},
|
| 131 |
{
|
| 132 |
-
"epoch": 0.
|
| 133 |
-
"learning_rate": 0.
|
| 134 |
-
"loss":
|
| 135 |
"step": 100
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"epoch": 0.
|
| 139 |
-
"
|
| 140 |
-
"
|
| 141 |
-
"
|
| 142 |
-
|
| 143 |
-
|
| 144 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 145 |
}
|
| 146 |
],
|
| 147 |
"logging_steps": 5,
|
| 148 |
-
"max_steps":
|
| 149 |
"num_input_tokens_seen": 0,
|
| 150 |
"num_train_epochs": 1,
|
| 151 |
"save_steps": 100,
|
| 152 |
-
"total_flos":
|
| 153 |
"train_batch_size": 2,
|
| 154 |
"trial_name": null,
|
| 155 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.05737563830397613,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1000,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0,
|
| 13 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 14 |
+
"loss": -7547.9619,
|
| 15 |
"step": 1
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.0,
|
| 19 |
+
"learning_rate": 1e-05,
|
| 20 |
+
"loss": -9106.0537,
|
| 21 |
"step": 5
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.0,
|
| 25 |
+
"learning_rate": 2e-05,
|
| 26 |
+
"loss": -10521.2672,
|
| 27 |
"step": 10
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.0,
|
| 31 |
+
"learning_rate": 3e-05,
|
| 32 |
+
"loss": -13425.4438,
|
| 33 |
"step": 15
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.0,
|
| 37 |
+
"learning_rate": 4e-05,
|
| 38 |
+
"loss": -18578.7203,
|
| 39 |
"step": 20
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 0.0,
|
| 43 |
+
"learning_rate": 5e-05,
|
| 44 |
+
"loss": -29817.3,
|
| 45 |
"step": 25
|
| 46 |
},
|
| 47 |
{
|
| 48 |
"epoch": 0.0,
|
| 49 |
+
"learning_rate": 6e-05,
|
| 50 |
+
"loss": -61245.15,
|
| 51 |
"step": 30
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.0,
|
| 55 |
+
"learning_rate": 7e-05,
|
| 56 |
+
"loss": -120348.9625,
|
| 57 |
"step": 35
|
| 58 |
},
|
| 59 |
{
|
| 60 |
"epoch": 0.0,
|
| 61 |
+
"learning_rate": 8e-05,
|
| 62 |
+
"loss": -156901.3375,
|
| 63 |
"step": 40
|
| 64 |
},
|
| 65 |
{
|
| 66 |
"epoch": 0.0,
|
| 67 |
+
"learning_rate": 9e-05,
|
| 68 |
+
"loss": -177861.1,
|
| 69 |
"step": 45
|
| 70 |
},
|
| 71 |
{
|
| 72 |
"epoch": 0.0,
|
| 73 |
+
"learning_rate": 0.0001,
|
| 74 |
+
"loss": -189531.375,
|
| 75 |
"step": 50
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 0.0,
|
| 79 |
+
"learning_rate": 0.00011000000000000002,
|
| 80 |
+
"loss": -170585.1375,
|
| 81 |
"step": 55
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 0.0,
|
| 85 |
+
"learning_rate": 0.00012,
|
| 86 |
+
"loss": -174312.9,
|
| 87 |
"step": 60
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 0.0,
|
| 91 |
+
"learning_rate": 0.00013000000000000002,
|
| 92 |
+
"loss": -179821.775,
|
| 93 |
"step": 65
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 0.0,
|
| 97 |
+
"learning_rate": 0.00014,
|
| 98 |
+
"loss": -182256.625,
|
| 99 |
"step": 70
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 0.0,
|
| 103 |
+
"learning_rate": 0.00015000000000000001,
|
| 104 |
+
"loss": -191381.75,
|
| 105 |
"step": 75
|
| 106 |
},
|
| 107 |
{
|
| 108 |
"epoch": 0.0,
|
| 109 |
+
"learning_rate": 0.00016,
|
| 110 |
+
"loss": -197267.975,
|
| 111 |
"step": 80
|
| 112 |
},
|
| 113 |
{
|
| 114 |
"epoch": 0.0,
|
| 115 |
+
"learning_rate": 0.00017,
|
| 116 |
+
"loss": -203530.25,
|
| 117 |
"step": 85
|
| 118 |
},
|
| 119 |
{
|
| 120 |
+
"epoch": 0.01,
|
| 121 |
+
"learning_rate": 0.00018,
|
| 122 |
+
"loss": -207515.1875,
|
| 123 |
"step": 90
|
| 124 |
},
|
| 125 |
{
|
| 126 |
+
"epoch": 0.01,
|
| 127 |
+
"learning_rate": 0.00019,
|
| 128 |
+
"loss": -209935.85,
|
| 129 |
"step": 95
|
| 130 |
},
|
| 131 |
{
|
| 132 |
+
"epoch": 0.01,
|
| 133 |
+
"learning_rate": 0.0002,
|
| 134 |
+
"loss": -213990.225,
|
| 135 |
"step": 100
|
| 136 |
},
|
| 137 |
{
|
| 138 |
+
"epoch": 0.01,
|
| 139 |
+
"learning_rate": 0.00019998476951563915,
|
| 140 |
+
"loss": -216354.45,
|
| 141 |
+
"step": 105
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.01,
|
| 145 |
+
"learning_rate": 0.0001999390827019096,
|
| 146 |
+
"loss": -186972.275,
|
| 147 |
+
"step": 110
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.01,
|
| 151 |
+
"learning_rate": 0.0001998629534754574,
|
| 152 |
+
"loss": -121758.175,
|
| 153 |
+
"step": 115
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.01,
|
| 157 |
+
"learning_rate": 0.00019975640502598244,
|
| 158 |
+
"loss": -121319.3,
|
| 159 |
+
"step": 120
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.01,
|
| 163 |
+
"learning_rate": 0.00019961946980917456,
|
| 164 |
+
"loss": -120151.6,
|
| 165 |
+
"step": 125
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.01,
|
| 169 |
+
"learning_rate": 0.00019945218953682734,
|
| 170 |
+
"loss": -122962.075,
|
| 171 |
+
"step": 130
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.01,
|
| 175 |
+
"learning_rate": 0.00019925461516413223,
|
| 176 |
+
"loss": -121680.5875,
|
| 177 |
+
"step": 135
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.01,
|
| 181 |
+
"learning_rate": 0.00019902680687415705,
|
| 182 |
+
"loss": -124972.4125,
|
| 183 |
+
"step": 140
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 0.01,
|
| 187 |
+
"learning_rate": 0.00019876883405951377,
|
| 188 |
+
"loss": -123293.25,
|
| 189 |
+
"step": 145
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 0.01,
|
| 193 |
+
"learning_rate": 0.00019848077530122083,
|
| 194 |
+
"loss": -123971.2875,
|
| 195 |
+
"step": 150
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.01,
|
| 199 |
+
"learning_rate": 0.00019816271834476642,
|
| 200 |
+
"loss": -121120.7125,
|
| 201 |
+
"step": 155
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.01,
|
| 205 |
+
"learning_rate": 0.00019781476007338058,
|
| 206 |
+
"loss": -125770.65,
|
| 207 |
+
"step": 160
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.01,
|
| 211 |
+
"learning_rate": 0.00019743700647852354,
|
| 212 |
+
"loss": -122050.65,
|
| 213 |
+
"step": 165
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.01,
|
| 217 |
+
"learning_rate": 0.00019702957262759965,
|
| 218 |
+
"loss": -124877.2,
|
| 219 |
+
"step": 170
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.01,
|
| 223 |
+
"learning_rate": 0.00019659258262890683,
|
| 224 |
+
"loss": -124584.5,
|
| 225 |
+
"step": 175
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 0.01,
|
| 229 |
+
"learning_rate": 0.0001961261695938319,
|
| 230 |
+
"loss": -123650.8,
|
| 231 |
+
"step": 180
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 0.01,
|
| 235 |
+
"learning_rate": 0.00019563047559630357,
|
| 236 |
+
"loss": -122581.825,
|
| 237 |
+
"step": 185
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 0.01,
|
| 241 |
+
"learning_rate": 0.00019510565162951537,
|
| 242 |
+
"loss": -126474.3125,
|
| 243 |
+
"step": 190
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.01,
|
| 247 |
+
"learning_rate": 0.0001945518575599317,
|
| 248 |
+
"loss": -125734.775,
|
| 249 |
+
"step": 195
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 0.01,
|
| 253 |
+
"learning_rate": 0.00019396926207859084,
|
| 254 |
+
"loss": -124667.3875,
|
| 255 |
+
"step": 200
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.01,
|
| 259 |
+
"learning_rate": 0.00019335804264972018,
|
| 260 |
+
"loss": -125573.05,
|
| 261 |
+
"step": 205
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.01,
|
| 265 |
+
"learning_rate": 0.00019271838545667876,
|
| 266 |
+
"loss": -123766.075,
|
| 267 |
+
"step": 210
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 0.01,
|
| 271 |
+
"learning_rate": 0.00019205048534524406,
|
| 272 |
+
"loss": -127037.1,
|
| 273 |
+
"step": 215
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.01,
|
| 277 |
+
"learning_rate": 0.0001913545457642601,
|
| 278 |
+
"loss": -126327.7375,
|
| 279 |
+
"step": 220
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.01,
|
| 283 |
+
"learning_rate": 0.000190630778703665,
|
| 284 |
+
"loss": -124661.9375,
|
| 285 |
+
"step": 225
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.01,
|
| 289 |
+
"learning_rate": 0.0001898794046299167,
|
| 290 |
+
"loss": -126650.575,
|
| 291 |
+
"step": 230
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 0.01,
|
| 295 |
+
"learning_rate": 0.0001891006524188368,
|
| 296 |
+
"loss": -125326.3375,
|
| 297 |
+
"step": 235
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.01,
|
| 301 |
+
"learning_rate": 0.00018829475928589271,
|
| 302 |
+
"loss": -126330.2125,
|
| 303 |
+
"step": 240
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.01,
|
| 307 |
+
"learning_rate": 0.00018746197071393958,
|
| 308 |
+
"loss": -129089.6,
|
| 309 |
+
"step": 245
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.01,
|
| 313 |
+
"learning_rate": 0.00018660254037844388,
|
| 314 |
+
"loss": -127315.075,
|
| 315 |
+
"step": 250
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 0.01,
|
| 319 |
+
"learning_rate": 0.00018571673007021123,
|
| 320 |
+
"loss": -128970.2875,
|
| 321 |
+
"step": 255
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.01,
|
| 325 |
+
"learning_rate": 0.0001848048096156426,
|
| 326 |
+
"loss": -127177.8625,
|
| 327 |
+
"step": 260
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 0.02,
|
| 331 |
+
"learning_rate": 0.00018386705679454242,
|
| 332 |
+
"loss": -126255.525,
|
| 333 |
+
"step": 265
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 0.02,
|
| 337 |
+
"learning_rate": 0.00018290375725550417,
|
| 338 |
+
"loss": -126107.25,
|
| 339 |
+
"step": 270
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.02,
|
| 343 |
+
"learning_rate": 0.0001819152044288992,
|
| 344 |
+
"loss": -128510.775,
|
| 345 |
+
"step": 275
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.02,
|
| 349 |
+
"learning_rate": 0.00018090169943749476,
|
| 350 |
+
"loss": -125616.225,
|
| 351 |
+
"step": 280
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 0.02,
|
| 355 |
+
"learning_rate": 0.00017986355100472928,
|
| 356 |
+
"loss": -129116.6125,
|
| 357 |
+
"step": 285
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 0.02,
|
| 361 |
+
"learning_rate": 0.00017880107536067218,
|
| 362 |
+
"loss": -129702.9375,
|
| 363 |
+
"step": 290
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.02,
|
| 367 |
+
"learning_rate": 0.0001777145961456971,
|
| 368 |
+
"loss": -129070.425,
|
| 369 |
+
"step": 295
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 0.02,
|
| 373 |
+
"learning_rate": 0.0001766044443118978,
|
| 374 |
+
"loss": -128198.4875,
|
| 375 |
+
"step": 300
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.02,
|
| 379 |
+
"learning_rate": 0.00017547095802227723,
|
| 380 |
+
"loss": -127598.8,
|
| 381 |
+
"step": 305
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.02,
|
| 385 |
+
"learning_rate": 0.00017431448254773944,
|
| 386 |
+
"loss": -128338.525,
|
| 387 |
+
"step": 310
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.02,
|
| 391 |
+
"learning_rate": 0.00017313537016191706,
|
| 392 |
+
"loss": -127601.575,
|
| 393 |
+
"step": 315
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 0.02,
|
| 397 |
+
"learning_rate": 0.0001719339800338651,
|
| 398 |
+
"loss": -126842.4125,
|
| 399 |
+
"step": 320
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.02,
|
| 403 |
+
"learning_rate": 0.00017071067811865476,
|
| 404 |
+
"loss": -131300.9625,
|
| 405 |
+
"step": 325
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.02,
|
| 409 |
+
"learning_rate": 0.00016946583704589973,
|
| 410 |
+
"loss": -126271.675,
|
| 411 |
+
"step": 330
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 0.02,
|
| 415 |
+
"learning_rate": 0.00016819983600624986,
|
| 416 |
+
"loss": -130244.475,
|
| 417 |
+
"step": 335
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 0.02,
|
| 421 |
+
"learning_rate": 0.00016691306063588583,
|
| 422 |
+
"loss": -127375.9625,
|
| 423 |
+
"step": 340
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.02,
|
| 427 |
+
"learning_rate": 0.00016560590289905073,
|
| 428 |
+
"loss": -126334.775,
|
| 429 |
+
"step": 345
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.02,
|
| 433 |
+
"learning_rate": 0.00016427876096865394,
|
| 434 |
+
"loss": -127501.675,
|
| 435 |
+
"step": 350
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 0.02,
|
| 439 |
+
"learning_rate": 0.00016293203910498376,
|
| 440 |
+
"loss": -128556.2,
|
| 441 |
+
"step": 355
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.02,
|
| 445 |
+
"learning_rate": 0.0001615661475325658,
|
| 446 |
+
"loss": -129236.6,
|
| 447 |
+
"step": 360
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 0.02,
|
| 451 |
+
"learning_rate": 0.00016018150231520486,
|
| 452 |
+
"loss": -128883.8375,
|
| 453 |
+
"step": 365
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 0.02,
|
| 457 |
+
"learning_rate": 0.00015877852522924732,
|
| 458 |
+
"loss": -128258.1625,
|
| 459 |
+
"step": 370
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 0.02,
|
| 463 |
+
"learning_rate": 0.0001573576436351046,
|
| 464 |
+
"loss": -127891.6125,
|
| 465 |
+
"step": 375
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.02,
|
| 469 |
+
"learning_rate": 0.0001559192903470747,
|
| 470 |
+
"loss": -127299.3375,
|
| 471 |
+
"step": 380
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.02,
|
| 475 |
+
"learning_rate": 0.00015446390350150273,
|
| 476 |
+
"loss": -134468.3625,
|
| 477 |
+
"step": 385
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.02,
|
| 481 |
+
"learning_rate": 0.0001529919264233205,
|
| 482 |
+
"loss": -129607.4875,
|
| 483 |
+
"step": 390
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 0.02,
|
| 487 |
+
"learning_rate": 0.00015150380749100545,
|
| 488 |
+
"loss": -128010.7,
|
| 489 |
+
"step": 395
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 0.02,
|
| 493 |
+
"learning_rate": 0.00015000000000000001,
|
| 494 |
+
"loss": -129447.0625,
|
| 495 |
+
"step": 400
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 0.02,
|
| 499 |
+
"learning_rate": 0.00014848096202463372,
|
| 500 |
+
"loss": -127959.8875,
|
| 501 |
+
"step": 405
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 0.02,
|
| 505 |
+
"learning_rate": 0.00014694715627858908,
|
| 506 |
+
"loss": -129912.275,
|
| 507 |
+
"step": 410
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.02,
|
| 511 |
+
"learning_rate": 0.00014539904997395468,
|
| 512 |
+
"loss": -125926.1,
|
| 513 |
+
"step": 415
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.02,
|
| 517 |
+
"learning_rate": 0.00014383711467890774,
|
| 518 |
+
"loss": -127398.7875,
|
| 519 |
+
"step": 420
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.02,
|
| 523 |
+
"learning_rate": 0.00014226182617406996,
|
| 524 |
+
"loss": -129303.4,
|
| 525 |
+
"step": 425
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 0.02,
|
| 529 |
+
"learning_rate": 0.00014067366430758004,
|
| 530 |
+
"loss": -127134.8875,
|
| 531 |
+
"step": 430
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 0.02,
|
| 535 |
+
"learning_rate": 0.00013907311284892736,
|
| 536 |
+
"loss": -127413.35,
|
| 537 |
+
"step": 435
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.03,
|
| 541 |
+
"learning_rate": 0.00013746065934159123,
|
| 542 |
+
"loss": -128643.9125,
|
| 543 |
+
"step": 440
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 0.03,
|
| 547 |
+
"learning_rate": 0.00013583679495453,
|
| 548 |
+
"loss": -127599.2,
|
| 549 |
+
"step": 445
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.03,
|
| 553 |
+
"learning_rate": 0.00013420201433256689,
|
| 554 |
+
"loss": -130022.475,
|
| 555 |
+
"step": 450
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.03,
|
| 559 |
+
"learning_rate": 0.00013255681544571568,
|
| 560 |
+
"loss": -128485.2625,
|
| 561 |
+
"step": 455
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 0.03,
|
| 565 |
+
"learning_rate": 0.00013090169943749476,
|
| 566 |
+
"loss": -129286.0125,
|
| 567 |
+
"step": 460
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 0.03,
|
| 571 |
+
"learning_rate": 0.00012923717047227368,
|
| 572 |
+
"loss": -128943.0125,
|
| 573 |
+
"step": 465
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 0.03,
|
| 577 |
+
"learning_rate": 0.0001275637355816999,
|
| 578 |
+
"loss": -130092.2,
|
| 579 |
+
"step": 470
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 0.03,
|
| 583 |
+
"learning_rate": 0.00012588190451025207,
|
| 584 |
+
"loss": -128025.6375,
|
| 585 |
+
"step": 475
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 0.03,
|
| 589 |
+
"learning_rate": 0.00012419218955996676,
|
| 590 |
+
"loss": -127411.2875,
|
| 591 |
+
"step": 480
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.03,
|
| 595 |
+
"learning_rate": 0.0001224951054343865,
|
| 596 |
+
"loss": -128624.175,
|
| 597 |
+
"step": 485
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.03,
|
| 601 |
+
"learning_rate": 0.00012079116908177593,
|
| 602 |
+
"loss": -129188.0875,
|
| 603 |
+
"step": 490
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.03,
|
| 607 |
+
"learning_rate": 0.00011908089953765449,
|
| 608 |
+
"loss": -130780.65,
|
| 609 |
+
"step": 495
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 0.03,
|
| 613 |
+
"learning_rate": 0.00011736481776669306,
|
| 614 |
+
"loss": -128979.9875,
|
| 615 |
+
"step": 500
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 0.03,
|
| 619 |
+
"learning_rate": 0.0001156434465040231,
|
| 620 |
+
"loss": -128289.55,
|
| 621 |
+
"step": 505
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 0.03,
|
| 625 |
+
"learning_rate": 0.00011391731009600654,
|
| 626 |
+
"loss": -129264.7,
|
| 627 |
+
"step": 510
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 0.03,
|
| 631 |
+
"learning_rate": 0.00011218693434051475,
|
| 632 |
+
"loss": -129166.7,
|
| 633 |
+
"step": 515
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.03,
|
| 637 |
+
"learning_rate": 0.00011045284632676536,
|
| 638 |
+
"loss": -130517.05,
|
| 639 |
+
"step": 520
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.03,
|
| 643 |
+
"learning_rate": 0.00010871557427476583,
|
| 644 |
+
"loss": -129060.7125,
|
| 645 |
+
"step": 525
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 0.03,
|
| 649 |
+
"learning_rate": 0.00010697564737441252,
|
| 650 |
+
"loss": -129417.6,
|
| 651 |
+
"step": 530
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 0.03,
|
| 655 |
+
"learning_rate": 0.0001052335956242944,
|
| 656 |
+
"loss": -128037.825,
|
| 657 |
+
"step": 535
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 0.03,
|
| 661 |
+
"learning_rate": 0.00010348994967025012,
|
| 662 |
+
"loss": -128152.5375,
|
| 663 |
+
"step": 540
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 0.03,
|
| 667 |
+
"learning_rate": 0.00010174524064372837,
|
| 668 |
+
"loss": -129852.7125,
|
| 669 |
+
"step": 545
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.03,
|
| 673 |
+
"learning_rate": 0.0001,
|
| 674 |
+
"loss": -130428.5375,
|
| 675 |
+
"step": 550
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.03,
|
| 679 |
+
"learning_rate": 9.825475935627165e-05,
|
| 680 |
+
"loss": -128036.5,
|
| 681 |
+
"step": 555
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.03,
|
| 685 |
+
"learning_rate": 9.651005032974994e-05,
|
| 686 |
+
"loss": -130735.8875,
|
| 687 |
+
"step": 560
|
| 688 |
+
},
|
| 689 |
+
{
|
| 690 |
+
"epoch": 0.03,
|
| 691 |
+
"learning_rate": 9.476640437570562e-05,
|
| 692 |
+
"loss": -128337.95,
|
| 693 |
+
"step": 565
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 0.03,
|
| 697 |
+
"learning_rate": 9.302435262558747e-05,
|
| 698 |
+
"loss": -131162.225,
|
| 699 |
+
"step": 570
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 0.03,
|
| 703 |
+
"learning_rate": 9.128442572523417e-05,
|
| 704 |
+
"loss": -127932.4,
|
| 705 |
+
"step": 575
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.03,
|
| 709 |
+
"learning_rate": 8.954715367323468e-05,
|
| 710 |
+
"loss": -129994.8125,
|
| 711 |
+
"step": 580
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.03,
|
| 715 |
+
"learning_rate": 8.781306565948528e-05,
|
| 716 |
+
"loss": -132943.675,
|
| 717 |
+
"step": 585
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.03,
|
| 721 |
+
"learning_rate": 8.608268990399349e-05,
|
| 722 |
+
"loss": -129275.15,
|
| 723 |
+
"step": 590
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.03,
|
| 727 |
+
"learning_rate": 8.435655349597689e-05,
|
| 728 |
+
"loss": -134494.475,
|
| 729 |
+
"step": 595
|
| 730 |
+
},
|
| 731 |
+
{
|
| 732 |
+
"epoch": 0.03,
|
| 733 |
+
"learning_rate": 8.263518223330697e-05,
|
| 734 |
+
"loss": -129055.075,
|
| 735 |
+
"step": 600
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 0.03,
|
| 739 |
+
"learning_rate": 8.091910046234552e-05,
|
| 740 |
+
"loss": -130328.3625,
|
| 741 |
+
"step": 605
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 0.03,
|
| 745 |
+
"learning_rate": 7.920883091822408e-05,
|
| 746 |
+
"loss": -130358.1,
|
| 747 |
+
"step": 610
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 0.04,
|
| 751 |
+
"learning_rate": 7.750489456561352e-05,
|
| 752 |
+
"loss": -131378.525,
|
| 753 |
+
"step": 615
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 0.04,
|
| 757 |
+
"learning_rate": 7.580781044003324e-05,
|
| 758 |
+
"loss": -133240.4375,
|
| 759 |
+
"step": 620
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.04,
|
| 763 |
+
"learning_rate": 7.411809548974792e-05,
|
| 764 |
+
"loss": -129869.4,
|
| 765 |
+
"step": 625
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.04,
|
| 769 |
+
"learning_rate": 7.243626441830009e-05,
|
| 770 |
+
"loss": -129762.225,
|
| 771 |
+
"step": 630
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.04,
|
| 775 |
+
"learning_rate": 7.076282952772633e-05,
|
| 776 |
+
"loss": -129086.725,
|
| 777 |
+
"step": 635
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 0.04,
|
| 781 |
+
"learning_rate": 6.909830056250527e-05,
|
| 782 |
+
"loss": -132211.9,
|
| 783 |
+
"step": 640
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 0.04,
|
| 787 |
+
"learning_rate": 6.744318455428436e-05,
|
| 788 |
+
"loss": -130901.375,
|
| 789 |
+
"step": 645
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 0.04,
|
| 793 |
+
"learning_rate": 6.579798566743314e-05,
|
| 794 |
+
"loss": -128798.4,
|
| 795 |
+
"step": 650
|
| 796 |
+
},
|
| 797 |
+
{
|
| 798 |
+
"epoch": 0.04,
|
| 799 |
+
"learning_rate": 6.416320504546997e-05,
|
| 800 |
+
"loss": -128713.4375,
|
| 801 |
+
"step": 655
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.04,
|
| 805 |
+
"learning_rate": 6.25393406584088e-05,
|
| 806 |
+
"loss": -129254.1,
|
| 807 |
+
"step": 660
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.04,
|
| 811 |
+
"learning_rate": 6.092688715107264e-05,
|
| 812 |
+
"loss": -129648.825,
|
| 813 |
+
"step": 665
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 0.04,
|
| 817 |
+
"learning_rate": 5.9326335692419995e-05,
|
| 818 |
+
"loss": -129720.0375,
|
| 819 |
+
"step": 670
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 0.04,
|
| 823 |
+
"learning_rate": 5.773817382593008e-05,
|
| 824 |
+
"loss": -130483.5,
|
| 825 |
+
"step": 675
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 0.04,
|
| 829 |
+
"learning_rate": 5.616288532109225e-05,
|
| 830 |
+
"loss": -132085.2625,
|
| 831 |
+
"step": 680
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 0.04,
|
| 835 |
+
"learning_rate": 5.4600950026045326e-05,
|
| 836 |
+
"loss": -129280.825,
|
| 837 |
+
"step": 685
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.04,
|
| 841 |
+
"learning_rate": 5.305284372141095e-05,
|
| 842 |
+
"loss": -128499.675,
|
| 843 |
+
"step": 690
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 0.04,
|
| 847 |
+
"learning_rate": 5.15190379753663e-05,
|
| 848 |
+
"loss": -131128.7125,
|
| 849 |
+
"step": 695
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.04,
|
| 853 |
+
"learning_rate": 5.000000000000002e-05,
|
| 854 |
+
"loss": -128292.3125,
|
| 855 |
+
"step": 700
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 0.04,
|
| 859 |
+
"learning_rate": 4.8496192508994576e-05,
|
| 860 |
+
"loss": -128954.8375,
|
| 861 |
+
"step": 705
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 0.04,
|
| 865 |
+
"learning_rate": 4.700807357667952e-05,
|
| 866 |
+
"loss": -130455.6375,
|
| 867 |
+
"step": 710
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.04,
|
| 871 |
+
"learning_rate": 4.5536096498497295e-05,
|
| 872 |
+
"loss": -130186.5625,
|
| 873 |
+
"step": 715
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 0.04,
|
| 877 |
+
"learning_rate": 4.4080709652925336e-05,
|
| 878 |
+
"loss": -128929.5125,
|
| 879 |
+
"step": 720
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 0.04,
|
| 883 |
+
"learning_rate": 4.264235636489542e-05,
|
| 884 |
+
"loss": -129662.825,
|
| 885 |
+
"step": 725
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 0.04,
|
| 889 |
+
"learning_rate": 4.12214747707527e-05,
|
| 890 |
+
"loss": -130658.1875,
|
| 891 |
+
"step": 730
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.04,
|
| 895 |
+
"learning_rate": 3.981849768479517e-05,
|
| 896 |
+
"loss": -129584.05,
|
| 897 |
+
"step": 735
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 0.04,
|
| 901 |
+
"learning_rate": 3.843385246743417e-05,
|
| 902 |
+
"loss": -129051.1875,
|
| 903 |
+
"step": 740
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 0.04,
|
| 907 |
+
"learning_rate": 3.7067960895016275e-05,
|
| 908 |
+
"loss": -129906.5,
|
| 909 |
+
"step": 745
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 0.04,
|
| 913 |
+
"learning_rate": 3.5721239031346066e-05,
|
| 914 |
+
"loss": -131726.125,
|
| 915 |
+
"step": 750
|
| 916 |
+
},
|
| 917 |
+
{
|
| 918 |
+
"epoch": 0.04,
|
| 919 |
+
"learning_rate": 3.439409710094929e-05,
|
| 920 |
+
"loss": -130614.0125,
|
| 921 |
+
"step": 755
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 0.04,
|
| 925 |
+
"learning_rate": 3.308693936411421e-05,
|
| 926 |
+
"loss": -129400.2,
|
| 927 |
+
"step": 760
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 0.04,
|
| 931 |
+
"learning_rate": 3.1800163993750166e-05,
|
| 932 |
+
"loss": -132052.375,
|
| 933 |
+
"step": 765
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.04,
|
| 937 |
+
"learning_rate": 3.053416295410026e-05,
|
| 938 |
+
"loss": -136370.1,
|
| 939 |
+
"step": 770
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 0.04,
|
| 943 |
+
"learning_rate": 2.9289321881345254e-05,
|
| 944 |
+
"loss": -130621.2,
|
| 945 |
+
"step": 775
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 0.04,
|
| 949 |
+
"learning_rate": 2.8066019966134904e-05,
|
| 950 |
+
"loss": -131630.3,
|
| 951 |
+
"step": 780
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 0.05,
|
| 955 |
+
"learning_rate": 2.6864629838082956e-05,
|
| 956 |
+
"loss": -131564.9125,
|
| 957 |
+
"step": 785
|
| 958 |
+
},
|
| 959 |
+
{
|
| 960 |
+
"epoch": 0.05,
|
| 961 |
+
"learning_rate": 2.5685517452260567e-05,
|
| 962 |
+
"loss": -132246.5375,
|
| 963 |
+
"step": 790
|
| 964 |
+
},
|
| 965 |
+
{
|
| 966 |
+
"epoch": 0.05,
|
| 967 |
+
"learning_rate": 2.45290419777228e-05,
|
| 968 |
+
"loss": -129737.325,
|
| 969 |
+
"step": 795
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.05,
|
| 973 |
+
"learning_rate": 2.339555568810221e-05,
|
| 974 |
+
"loss": -129589.5625,
|
| 975 |
+
"step": 800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.05,
|
| 979 |
+
"learning_rate": 2.2285403854302912e-05,
|
| 980 |
+
"loss": -129878.975,
|
| 981 |
+
"step": 805
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 0.05,
|
| 985 |
+
"learning_rate": 2.119892463932781e-05,
|
| 986 |
+
"loss": -131492.4375,
|
| 987 |
+
"step": 810
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 0.05,
|
| 991 |
+
"learning_rate": 2.013644899527074e-05,
|
| 992 |
+
"loss": -128856.9,
|
| 993 |
+
"step": 815
|
| 994 |
+
},
|
| 995 |
+
{
|
| 996 |
+
"epoch": 0.05,
|
| 997 |
+
"learning_rate": 1.9098300562505266e-05,
|
| 998 |
+
"loss": -131127.25,
|
| 999 |
+
"step": 820
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.05,
|
| 1003 |
+
"learning_rate": 1.808479557110081e-05,
|
| 1004 |
+
"loss": -129349.7375,
|
| 1005 |
+
"step": 825
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 0.05,
|
| 1009 |
+
"learning_rate": 1.7096242744495837e-05,
|
| 1010 |
+
"loss": -133686.3375,
|
| 1011 |
+
"step": 830
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 0.05,
|
| 1015 |
+
"learning_rate": 1.6132943205457606e-05,
|
| 1016 |
+
"loss": -130341.85,
|
| 1017 |
+
"step": 835
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.05,
|
| 1021 |
+
"learning_rate": 1.5195190384357404e-05,
|
| 1022 |
+
"loss": -129862.95,
|
| 1023 |
+
"step": 840
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 0.05,
|
| 1027 |
+
"learning_rate": 1.4283269929788779e-05,
|
| 1028 |
+
"loss": -129905.8875,
|
| 1029 |
+
"step": 845
|
| 1030 |
+
},
|
| 1031 |
+
{
|
| 1032 |
+
"epoch": 0.05,
|
| 1033 |
+
"learning_rate": 1.339745962155613e-05,
|
| 1034 |
+
"loss": -133136.0,
|
| 1035 |
+
"step": 850
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 0.05,
|
| 1039 |
+
"learning_rate": 1.2538029286060426e-05,
|
| 1040 |
+
"loss": -129610.375,
|
| 1041 |
+
"step": 855
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 0.05,
|
| 1045 |
+
"learning_rate": 1.1705240714107302e-05,
|
| 1046 |
+
"loss": -128013.6375,
|
| 1047 |
+
"step": 860
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 0.05,
|
| 1051 |
+
"learning_rate": 1.0899347581163221e-05,
|
| 1052 |
+
"loss": -128977.225,
|
| 1053 |
+
"step": 865
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 0.05,
|
| 1057 |
+
"learning_rate": 1.0120595370083318e-05,
|
| 1058 |
+
"loss": -127813.3125,
|
| 1059 |
+
"step": 870
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.05,
|
| 1063 |
+
"learning_rate": 9.369221296335006e-06,
|
| 1064 |
+
"loss": -129546.625,
|
| 1065 |
+
"step": 875
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.05,
|
| 1069 |
+
"learning_rate": 8.645454235739903e-06,
|
| 1070 |
+
"loss": -131817.925,
|
| 1071 |
+
"step": 880
|
| 1072 |
+
},
|
| 1073 |
+
{
|
| 1074 |
+
"epoch": 0.05,
|
| 1075 |
+
"learning_rate": 7.949514654755962e-06,
|
| 1076 |
+
"loss": -130486.9625,
|
| 1077 |
+
"step": 885
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 0.05,
|
| 1081 |
+
"learning_rate": 7.281614543321269e-06,
|
| 1082 |
+
"loss": -130403.5625,
|
| 1083 |
+
"step": 890
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 0.05,
|
| 1087 |
+
"learning_rate": 6.6419573502798374e-06,
|
| 1088 |
+
"loss": -127931.8625,
|
| 1089 |
+
"step": 895
|
| 1090 |
+
},
|
| 1091 |
+
{
|
| 1092 |
+
"epoch": 0.05,
|
| 1093 |
+
"learning_rate": 6.030737921409169e-06,
|
| 1094 |
+
"loss": -129240.275,
|
| 1095 |
+
"step": 900
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 0.05,
|
| 1099 |
+
"learning_rate": 5.448142440068316e-06,
|
| 1100 |
+
"loss": -130990.2875,
|
| 1101 |
+
"step": 905
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.05,
|
| 1105 |
+
"learning_rate": 4.8943483704846475e-06,
|
| 1106 |
+
"loss": -132186.325,
|
| 1107 |
+
"step": 910
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 0.05,
|
| 1111 |
+
"learning_rate": 4.369524403696457e-06,
|
| 1112 |
+
"loss": -131533.55,
|
| 1113 |
+
"step": 915
|
| 1114 |
+
},
|
| 1115 |
+
{
|
| 1116 |
+
"epoch": 0.05,
|
| 1117 |
+
"learning_rate": 3.873830406168111e-06,
|
| 1118 |
+
"loss": -133344.5,
|
| 1119 |
+
"step": 920
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 0.05,
|
| 1123 |
+
"learning_rate": 3.40741737109318e-06,
|
| 1124 |
+
"loss": -130234.625,
|
| 1125 |
+
"step": 925
|
| 1126 |
+
},
|
| 1127 |
+
{
|
| 1128 |
+
"epoch": 0.05,
|
| 1129 |
+
"learning_rate": 2.970427372400353e-06,
|
| 1130 |
+
"loss": -130692.8375,
|
| 1131 |
+
"step": 930
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.05,
|
| 1135 |
+
"learning_rate": 2.5629935214764865e-06,
|
| 1136 |
+
"loss": -132096.6625,
|
| 1137 |
+
"step": 935
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 0.05,
|
| 1141 |
+
"learning_rate": 2.1852399266194314e-06,
|
| 1142 |
+
"loss": -132693.075,
|
| 1143 |
+
"step": 940
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.05,
|
| 1147 |
+
"learning_rate": 1.8372816552336026e-06,
|
| 1148 |
+
"loss": -133697.4875,
|
| 1149 |
+
"step": 945
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 0.05,
|
| 1153 |
+
"learning_rate": 1.5192246987791981e-06,
|
| 1154 |
+
"loss": -129152.375,
|
| 1155 |
+
"step": 950
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 0.05,
|
| 1159 |
+
"learning_rate": 1.231165940486234e-06,
|
| 1160 |
+
"loss": -130557.1375,
|
| 1161 |
+
"step": 955
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 0.06,
|
| 1165 |
+
"learning_rate": 9.731931258429638e-07,
|
| 1166 |
+
"loss": -129561.8625,
|
| 1167 |
+
"step": 960
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 0.06,
|
| 1171 |
+
"learning_rate": 7.453848358678017e-07,
|
| 1172 |
+
"loss": -128434.575,
|
| 1173 |
+
"step": 965
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 0.06,
|
| 1177 |
+
"learning_rate": 5.478104631726711e-07,
|
| 1178 |
+
"loss": -131701.65,
|
| 1179 |
+
"step": 970
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 0.06,
|
| 1183 |
+
"learning_rate": 3.805301908254455e-07,
|
| 1184 |
+
"loss": -130711.15,
|
| 1185 |
+
"step": 975
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.06,
|
| 1189 |
+
"learning_rate": 2.4359497401758024e-07,
|
| 1190 |
+
"loss": -132491.45,
|
| 1191 |
+
"step": 980
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 0.06,
|
| 1195 |
+
"learning_rate": 1.3704652454261668e-07,
|
| 1196 |
+
"loss": -135885.35,
|
| 1197 |
+
"step": 985
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 0.06,
|
| 1201 |
+
"learning_rate": 6.09172980904238e-08,
|
| 1202 |
+
"loss": -130926.525,
|
| 1203 |
+
"step": 990
|
| 1204 |
+
},
|
| 1205 |
+
{
|
| 1206 |
+
"epoch": 0.06,
|
| 1207 |
+
"learning_rate": 1.5230484360873044e-08,
|
| 1208 |
+
"loss": -130808.1,
|
| 1209 |
+
"step": 995
|
| 1210 |
+
},
|
| 1211 |
+
{
|
| 1212 |
+
"epoch": 0.06,
|
| 1213 |
+
"learning_rate": 0.0,
|
| 1214 |
+
"loss": -132145.975,
|
| 1215 |
+
"step": 1000
|
| 1216 |
+
},
|
| 1217 |
+
{
|
| 1218 |
+
"epoch": 0.06,
|
| 1219 |
+
"step": 1000,
|
| 1220 |
+
"total_flos": 7.072526939948319e+17,
|
| 1221 |
+
"train_loss": -130335.20808300782,
|
| 1222 |
+
"train_runtime": 5587.0346,
|
| 1223 |
+
"train_samples_per_second": 1.432,
|
| 1224 |
+
"train_steps_per_second": 0.179
|
| 1225 |
}
|
| 1226 |
],
|
| 1227 |
"logging_steps": 5,
|
| 1228 |
+
"max_steps": 1000,
|
| 1229 |
"num_input_tokens_seen": 0,
|
| 1230 |
"num_train_epochs": 1,
|
| 1231 |
"save_steps": 100,
|
| 1232 |
+
"total_flos": 7.072526939948319e+17,
|
| 1233 |
"train_batch_size": 2,
|
| 1234 |
"trial_name": null,
|
| 1235 |
"trial_params": null
|