Commit
•
8591e44
1
Parent(s):
0c383d9
Model save
Browse files- README.md +150 -0
- adapter_model.safetensors +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Jul22_14-16-26_mlp14/events.out.tfevents.1721650765.mlp14 +2 -2
- runs/Jul22_14-16-26_mlp14/events.out.tfevents.1721712828.mlp14 +3 -0
- tokenizer_config.json +1 -1
- train_results.json +8 -0
- trainer_state.json +1944 -0
README.md
ADDED
|
@@ -0,0 +1,150 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
base_model: microsoft/Phi-3-mini-4k-instruct
|
| 11 |
+
model-index:
|
| 12 |
+
- name: phi-ft-1000000-fp-newsplit
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# phi-ft-1000000-fp-newsplit
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 1.7754
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
### Training hyperparameters
|
| 40 |
+
|
| 41 |
+
The following hyperparameters were used during training:
|
| 42 |
+
- learning_rate: 5e-05
|
| 43 |
+
- train_batch_size: 4
|
| 44 |
+
- eval_batch_size: 4
|
| 45 |
+
- seed: 0
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.2
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 55 |
+
| 3.1002 | 0.0114 | 100 | 3.0505 |
|
| 56 |
+
| 2.1929 | 0.0229 | 200 | 2.0493 |
|
| 57 |
+
| 1.6369 | 0.0343 | 300 | 1.6432 |
|
| 58 |
+
| 1.4618 | 0.0458 | 400 | 1.5580 |
|
| 59 |
+
| 1.317 | 0.0572 | 500 | 1.5410 |
|
| 60 |
+
| 1.1329 | 0.0687 | 600 | 1.6269 |
|
| 61 |
+
| 0.9505 | 0.0801 | 700 | 1.7387 |
|
| 62 |
+
| 0.8334 | 0.0916 | 800 | 1.7443 |
|
| 63 |
+
| 0.7692 | 0.1030 | 900 | 1.7634 |
|
| 64 |
+
| 0.6983 | 0.1145 | 1000 | 1.7546 |
|
| 65 |
+
| 0.6859 | 0.1259 | 1100 | 1.7593 |
|
| 66 |
+
| 0.6671 | 0.1374 | 1200 | 1.7647 |
|
| 67 |
+
| 0.6285 | 0.1488 | 1300 | 1.7951 |
|
| 68 |
+
| 0.6121 | 0.1603 | 1400 | 1.7816 |
|
| 69 |
+
| 0.5923 | 0.1717 | 1500 | 1.8132 |
|
| 70 |
+
| 0.5908 | 0.1832 | 1600 | 1.7664 |
|
| 71 |
+
| 0.5662 | 0.1946 | 1700 | 1.8307 |
|
| 72 |
+
| 0.5637 | 0.2060 | 1800 | 1.7864 |
|
| 73 |
+
| 0.5475 | 0.2175 | 1900 | 1.7988 |
|
| 74 |
+
| 0.5421 | 0.2289 | 2000 | 1.7876 |
|
| 75 |
+
| 0.529 | 0.2404 | 2100 | 1.7661 |
|
| 76 |
+
| 0.5202 | 0.2518 | 2200 | 1.7709 |
|
| 77 |
+
| 0.5287 | 0.2633 | 2300 | 1.7681 |
|
| 78 |
+
| 0.514 | 0.2747 | 2400 | 1.7765 |
|
| 79 |
+
| 0.5026 | 0.2862 | 2500 | 1.7931 |
|
| 80 |
+
| 0.5038 | 0.2976 | 2600 | 1.7808 |
|
| 81 |
+
| 0.5052 | 0.3091 | 2700 | 1.7689 |
|
| 82 |
+
| 0.4918 | 0.3205 | 2800 | 1.7862 |
|
| 83 |
+
| 0.4817 | 0.3320 | 2900 | 1.7916 |
|
| 84 |
+
| 0.4806 | 0.3434 | 3000 | 1.7796 |
|
| 85 |
+
| 0.4849 | 0.3549 | 3100 | 1.7654 |
|
| 86 |
+
| 0.4784 | 0.3663 | 3200 | 1.7576 |
|
| 87 |
+
| 0.4712 | 0.3777 | 3300 | 1.7746 |
|
| 88 |
+
| 0.4715 | 0.3892 | 3400 | 1.7568 |
|
| 89 |
+
| 0.4608 | 0.4006 | 3500 | 1.7424 |
|
| 90 |
+
| 0.4629 | 0.4121 | 3600 | 1.7561 |
|
| 91 |
+
| 0.4591 | 0.4235 | 3700 | 1.7498 |
|
| 92 |
+
| 0.4652 | 0.4350 | 3800 | 1.7366 |
|
| 93 |
+
| 0.461 | 0.4464 | 3900 | 1.7394 |
|
| 94 |
+
| 0.4469 | 0.4579 | 4000 | 1.7397 |
|
| 95 |
+
| 0.4521 | 0.4693 | 4100 | 1.7555 |
|
| 96 |
+
| 0.4498 | 0.4808 | 4200 | 1.7652 |
|
| 97 |
+
| 0.4541 | 0.4922 | 4300 | 1.7583 |
|
| 98 |
+
| 0.4594 | 0.5037 | 4400 | 1.7605 |
|
| 99 |
+
| 0.4514 | 0.5151 | 4500 | 1.7686 |
|
| 100 |
+
| 0.4395 | 0.5266 | 4600 | 1.7714 |
|
| 101 |
+
| 0.4384 | 0.5380 | 4700 | 1.7889 |
|
| 102 |
+
| 0.4392 | 0.5495 | 4800 | 1.7709 |
|
| 103 |
+
| 0.4495 | 0.5609 | 4900 | 1.7554 |
|
| 104 |
+
| 0.4375 | 0.5723 | 5000 | 1.7532 |
|
| 105 |
+
| 0.4441 | 0.5838 | 5100 | 1.7770 |
|
| 106 |
+
| 0.4458 | 0.5952 | 5200 | 1.7528 |
|
| 107 |
+
| 0.4343 | 0.6067 | 5300 | 1.7646 |
|
| 108 |
+
| 0.433 | 0.6181 | 5400 | 1.7689 |
|
| 109 |
+
| 0.4371 | 0.6296 | 5500 | 1.7738 |
|
| 110 |
+
| 0.4376 | 0.6410 | 5600 | 1.7633 |
|
| 111 |
+
| 0.4366 | 0.6525 | 5700 | 1.7810 |
|
| 112 |
+
| 0.43 | 0.6639 | 5800 | 1.7685 |
|
| 113 |
+
| 0.4345 | 0.6754 | 5900 | 1.7761 |
|
| 114 |
+
| 0.4379 | 0.6868 | 6000 | 1.7782 |
|
| 115 |
+
| 0.4294 | 0.6983 | 6100 | 1.7737 |
|
| 116 |
+
| 0.4441 | 0.7097 | 6200 | 1.7646 |
|
| 117 |
+
| 0.4396 | 0.7212 | 6300 | 1.7779 |
|
| 118 |
+
| 0.4307 | 0.7326 | 6400 | 1.7766 |
|
| 119 |
+
| 0.4331 | 0.7440 | 6500 | 1.7733 |
|
| 120 |
+
| 0.4326 | 0.7555 | 6600 | 1.7796 |
|
| 121 |
+
| 0.4286 | 0.7669 | 6700 | 1.7803 |
|
| 122 |
+
| 0.4294 | 0.7784 | 6800 | 1.7787 |
|
| 123 |
+
| 0.4294 | 0.7898 | 6900 | 1.7795 |
|
| 124 |
+
| 0.4364 | 0.8013 | 7000 | 1.7765 |
|
| 125 |
+
| 0.4414 | 0.8127 | 7100 | 1.7783 |
|
| 126 |
+
| 0.4336 | 0.8242 | 7200 | 1.7746 |
|
| 127 |
+
| 0.4324 | 0.8356 | 7300 | 1.7728 |
|
| 128 |
+
| 0.4414 | 0.8471 | 7400 | 1.7765 |
|
| 129 |
+
| 0.4288 | 0.8585 | 7500 | 1.7792 |
|
| 130 |
+
| 0.4359 | 0.8700 | 7600 | 1.7776 |
|
| 131 |
+
| 0.4242 | 0.8814 | 7700 | 1.7762 |
|
| 132 |
+
| 0.4413 | 0.8929 | 7800 | 1.7751 |
|
| 133 |
+
| 0.4402 | 0.9043 | 7900 | 1.7754 |
|
| 134 |
+
| 0.4452 | 0.9158 | 8000 | 1.7750 |
|
| 135 |
+
| 0.4346 | 0.9272 | 8100 | 1.7755 |
|
| 136 |
+
| 0.4396 | 0.9386 | 8200 | 1.7751 |
|
| 137 |
+
| 0.44 | 0.9501 | 8300 | 1.7752 |
|
| 138 |
+
| 0.4333 | 0.9615 | 8400 | 1.7753 |
|
| 139 |
+
| 0.4348 | 0.9730 | 8500 | 1.7754 |
|
| 140 |
+
| 0.4331 | 0.9844 | 8600 | 1.7752 |
|
| 141 |
+
| 0.4326 | 0.9959 | 8700 | 1.7754 |
|
| 142 |
+
|
| 143 |
+
|
| 144 |
+
### Framework versions
|
| 145 |
+
|
| 146 |
+
- PEFT 0.10.0
|
| 147 |
+
- Transformers 4.40.0
|
| 148 |
+
- Pytorch 2.3.0+cu121
|
| 149 |
+
- Datasets 2.16.0
|
| 150 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 38823448
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a31d7efea45f56eead31ecc27b66ff72e0b2f9ae86cf12f75956b0114dd6bb66
|
| 3 |
size 38823448
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.775362253189087,
|
| 4 |
+
"eval_runtime": 124.0975,
|
| 5 |
+
"eval_samples": 10000,
|
| 6 |
+
"eval_samples_per_second": 2.256,
|
| 7 |
+
"eval_steps_per_second": 0.564,
|
| 8 |
+
"total_flos": 1.6067287852253184e+18,
|
| 9 |
+
"train_loss": 0.5878726873860691,
|
| 10 |
+
"train_runtime": 61938.6999,
|
| 11 |
+
"train_samples_per_second": 0.564,
|
| 12 |
+
"train_steps_per_second": 0.141
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.775362253189087,
|
| 4 |
+
"eval_runtime": 124.0975,
|
| 5 |
+
"eval_samples": 10000,
|
| 6 |
+
"eval_samples_per_second": 2.256,
|
| 7 |
+
"eval_steps_per_second": 0.564
|
| 8 |
+
}
|
runs/Jul22_14-16-26_mlp14/events.out.tfevents.1721650765.mlp14
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:04467207782884966a22792299f38eb5760df84f8ff35ee5c6784737745dfa93
|
| 3 |
+
size 65701
|
runs/Jul22_14-16-26_mlp14/events.out.tfevents.1721712828.mlp14
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:39058e8c30961ab8668413a1562168666f86219a1155caae1763efcc946cf5fb
|
| 3 |
+
size 40
|
tokenizer_config.json
CHANGED
|
@@ -122,7 +122,7 @@
|
|
| 122 |
"legacy": false,
|
| 123 |
"model_max_length": 2048,
|
| 124 |
"pad_token": "<unk>",
|
| 125 |
-
"padding_side": "
|
| 126 |
"sp_model_kwargs": {},
|
| 127 |
"tokenizer_class": "LlamaTokenizer",
|
| 128 |
"unk_token": "<unk>",
|
|
|
|
| 122 |
"legacy": false,
|
| 123 |
"model_max_length": 2048,
|
| 124 |
"pad_token": "<unk>",
|
| 125 |
+
"padding_side": "left",
|
| 126 |
"sp_model_kwargs": {},
|
| 127 |
"tokenizer_class": "LlamaTokenizer",
|
| 128 |
"unk_token": "<unk>",
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 1.6067287852253184e+18,
|
| 4 |
+
"train_loss": 0.5878726873860691,
|
| 5 |
+
"train_runtime": 61938.6999,
|
| 6 |
+
"train_samples_per_second": 0.564,
|
| 7 |
+
"train_steps_per_second": 0.141
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1944 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 8736,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005723443223443223,
|
| 13 |
+
"grad_norm": 6.21875,
|
| 14 |
+
"learning_rate": 1.4302059496567508e-06,
|
| 15 |
+
"loss": 3.2916,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.011446886446886446,
|
| 20 |
+
"grad_norm": 3.453125,
|
| 21 |
+
"learning_rate": 2.8604118993135015e-06,
|
| 22 |
+
"loss": 3.1002,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.011446886446886446,
|
| 27 |
+
"eval_loss": 3.050471782684326,
|
| 28 |
+
"eval_runtime": 124.1248,
|
| 29 |
+
"eval_samples_per_second": 2.256,
|
| 30 |
+
"eval_steps_per_second": 0.564,
|
| 31 |
+
"step": 100
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.017170329670329672,
|
| 35 |
+
"grad_norm": 1.8671875,
|
| 36 |
+
"learning_rate": 4.290617848970252e-06,
|
| 37 |
+
"loss": 2.6977,
|
| 38 |
+
"step": 150
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.022893772893772892,
|
| 42 |
+
"grad_norm": 0.91015625,
|
| 43 |
+
"learning_rate": 5.720823798627003e-06,
|
| 44 |
+
"loss": 2.1929,
|
| 45 |
+
"step": 200
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.022893772893772892,
|
| 49 |
+
"eval_loss": 2.049286365509033,
|
| 50 |
+
"eval_runtime": 124.1321,
|
| 51 |
+
"eval_samples_per_second": 2.256,
|
| 52 |
+
"eval_steps_per_second": 0.564,
|
| 53 |
+
"step": 200
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.028617216117216116,
|
| 57 |
+
"grad_norm": 0.625,
|
| 58 |
+
"learning_rate": 7.151029748283754e-06,
|
| 59 |
+
"loss": 1.8298,
|
| 60 |
+
"step": 250
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 0.034340659340659344,
|
| 64 |
+
"grad_norm": 0.486328125,
|
| 65 |
+
"learning_rate": 8.581235697940504e-06,
|
| 66 |
+
"loss": 1.6369,
|
| 67 |
+
"step": 300
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 0.034340659340659344,
|
| 71 |
+
"eval_loss": 1.6432359218597412,
|
| 72 |
+
"eval_runtime": 124.1394,
|
| 73 |
+
"eval_samples_per_second": 2.256,
|
| 74 |
+
"eval_steps_per_second": 0.564,
|
| 75 |
+
"step": 300
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 0.04006410256410257,
|
| 79 |
+
"grad_norm": 0.5078125,
|
| 80 |
+
"learning_rate": 1.0011441647597253e-05,
|
| 81 |
+
"loss": 1.531,
|
| 82 |
+
"step": 350
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.045787545787545784,
|
| 86 |
+
"grad_norm": 0.328125,
|
| 87 |
+
"learning_rate": 1.1441647597254006e-05,
|
| 88 |
+
"loss": 1.4618,
|
| 89 |
+
"step": 400
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.045787545787545784,
|
| 93 |
+
"eval_loss": 1.5580341815948486,
|
| 94 |
+
"eval_runtime": 124.1039,
|
| 95 |
+
"eval_samples_per_second": 2.256,
|
| 96 |
+
"eval_steps_per_second": 0.564,
|
| 97 |
+
"step": 400
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.05151098901098901,
|
| 101 |
+
"grad_norm": 0.28515625,
|
| 102 |
+
"learning_rate": 1.2871853546910755e-05,
|
| 103 |
+
"loss": 1.4061,
|
| 104 |
+
"step": 450
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.05723443223443223,
|
| 108 |
+
"grad_norm": 0.2890625,
|
| 109 |
+
"learning_rate": 1.4302059496567508e-05,
|
| 110 |
+
"loss": 1.317,
|
| 111 |
+
"step": 500
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.05723443223443223,
|
| 115 |
+
"eval_loss": 1.5409735441207886,
|
| 116 |
+
"eval_runtime": 124.0925,
|
| 117 |
+
"eval_samples_per_second": 2.256,
|
| 118 |
+
"eval_steps_per_second": 0.564,
|
| 119 |
+
"step": 500
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.06295787545787546,
|
| 123 |
+
"grad_norm": 0.265625,
|
| 124 |
+
"learning_rate": 1.5732265446224257e-05,
|
| 125 |
+
"loss": 1.2334,
|
| 126 |
+
"step": 550
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.06868131868131869,
|
| 130 |
+
"grad_norm": 0.341796875,
|
| 131 |
+
"learning_rate": 1.716247139588101e-05,
|
| 132 |
+
"loss": 1.1329,
|
| 133 |
+
"step": 600
|
| 134 |
+
},
|
| 135 |
+
{
|
| 136 |
+
"epoch": 0.06868131868131869,
|
| 137 |
+
"eval_loss": 1.6269210577011108,
|
| 138 |
+
"eval_runtime": 124.0849,
|
| 139 |
+
"eval_samples_per_second": 2.257,
|
| 140 |
+
"eval_steps_per_second": 0.564,
|
| 141 |
+
"step": 600
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.0744047619047619,
|
| 145 |
+
"grad_norm": 0.275390625,
|
| 146 |
+
"learning_rate": 1.859267734553776e-05,
|
| 147 |
+
"loss": 1.039,
|
| 148 |
+
"step": 650
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.08012820512820513,
|
| 152 |
+
"grad_norm": 0.33984375,
|
| 153 |
+
"learning_rate": 2.0022883295194507e-05,
|
| 154 |
+
"loss": 0.9505,
|
| 155 |
+
"step": 700
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 0.08012820512820513,
|
| 159 |
+
"eval_loss": 1.738716959953308,
|
| 160 |
+
"eval_runtime": 124.1002,
|
| 161 |
+
"eval_samples_per_second": 2.256,
|
| 162 |
+
"eval_steps_per_second": 0.564,
|
| 163 |
+
"step": 700
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08585164835164835,
|
| 167 |
+
"grad_norm": 0.349609375,
|
| 168 |
+
"learning_rate": 2.145308924485126e-05,
|
| 169 |
+
"loss": 0.8827,
|
| 170 |
+
"step": 750
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.09157509157509157,
|
| 174 |
+
"grad_norm": 0.4296875,
|
| 175 |
+
"learning_rate": 2.2883295194508012e-05,
|
| 176 |
+
"loss": 0.8334,
|
| 177 |
+
"step": 800
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09157509157509157,
|
| 181 |
+
"eval_loss": 1.7442790269851685,
|
| 182 |
+
"eval_runtime": 124.087,
|
| 183 |
+
"eval_samples_per_second": 2.256,
|
| 184 |
+
"eval_steps_per_second": 0.564,
|
| 185 |
+
"step": 800
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.0972985347985348,
|
| 189 |
+
"grad_norm": 0.431640625,
|
| 190 |
+
"learning_rate": 2.431350114416476e-05,
|
| 191 |
+
"loss": 0.8127,
|
| 192 |
+
"step": 850
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.10302197802197802,
|
| 196 |
+
"grad_norm": 0.400390625,
|
| 197 |
+
"learning_rate": 2.574370709382151e-05,
|
| 198 |
+
"loss": 0.7692,
|
| 199 |
+
"step": 900
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.10302197802197802,
|
| 203 |
+
"eval_loss": 1.7634161710739136,
|
| 204 |
+
"eval_runtime": 124.1852,
|
| 205 |
+
"eval_samples_per_second": 2.255,
|
| 206 |
+
"eval_steps_per_second": 0.564,
|
| 207 |
+
"step": 900
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.10874542124542125,
|
| 211 |
+
"grad_norm": 0.455078125,
|
| 212 |
+
"learning_rate": 2.7173913043478262e-05,
|
| 213 |
+
"loss": 0.74,
|
| 214 |
+
"step": 950
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.11446886446886446,
|
| 218 |
+
"grad_norm": 0.48828125,
|
| 219 |
+
"learning_rate": 2.8604118993135016e-05,
|
| 220 |
+
"loss": 0.6983,
|
| 221 |
+
"step": 1000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.11446886446886446,
|
| 225 |
+
"eval_loss": 1.7546391487121582,
|
| 226 |
+
"eval_runtime": 124.0637,
|
| 227 |
+
"eval_samples_per_second": 2.257,
|
| 228 |
+
"eval_steps_per_second": 0.564,
|
| 229 |
+
"step": 1000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.1201923076923077,
|
| 233 |
+
"grad_norm": 0.46484375,
|
| 234 |
+
"learning_rate": 3.0034324942791764e-05,
|
| 235 |
+
"loss": 0.7277,
|
| 236 |
+
"step": 1050
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.1259157509157509,
|
| 240 |
+
"grad_norm": 0.5703125,
|
| 241 |
+
"learning_rate": 3.1464530892448515e-05,
|
| 242 |
+
"loss": 0.6859,
|
| 243 |
+
"step": 1100
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.1259157509157509,
|
| 247 |
+
"eval_loss": 1.7593410015106201,
|
| 248 |
+
"eval_runtime": 124.125,
|
| 249 |
+
"eval_samples_per_second": 2.256,
|
| 250 |
+
"eval_steps_per_second": 0.564,
|
| 251 |
+
"step": 1100
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.13163919413919414,
|
| 255 |
+
"grad_norm": 0.4921875,
|
| 256 |
+
"learning_rate": 3.289473684210527e-05,
|
| 257 |
+
"loss": 0.6844,
|
| 258 |
+
"step": 1150
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.13736263736263737,
|
| 262 |
+
"grad_norm": 0.5546875,
|
| 263 |
+
"learning_rate": 3.432494279176202e-05,
|
| 264 |
+
"loss": 0.6671,
|
| 265 |
+
"step": 1200
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.13736263736263737,
|
| 269 |
+
"eval_loss": 1.7647184133529663,
|
| 270 |
+
"eval_runtime": 124.0962,
|
| 271 |
+
"eval_samples_per_second": 2.256,
|
| 272 |
+
"eval_steps_per_second": 0.564,
|
| 273 |
+
"step": 1200
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 0.14308608058608058,
|
| 277 |
+
"grad_norm": 0.4921875,
|
| 278 |
+
"learning_rate": 3.5755148741418764e-05,
|
| 279 |
+
"loss": 0.6553,
|
| 280 |
+
"step": 1250
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.1488095238095238,
|
| 284 |
+
"grad_norm": 0.5546875,
|
| 285 |
+
"learning_rate": 3.718535469107552e-05,
|
| 286 |
+
"loss": 0.6285,
|
| 287 |
+
"step": 1300
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 0.1488095238095238,
|
| 291 |
+
"eval_loss": 1.7950905561447144,
|
| 292 |
+
"eval_runtime": 124.0984,
|
| 293 |
+
"eval_samples_per_second": 2.256,
|
| 294 |
+
"eval_steps_per_second": 0.564,
|
| 295 |
+
"step": 1300
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 0.15453296703296704,
|
| 299 |
+
"grad_norm": 0.5234375,
|
| 300 |
+
"learning_rate": 3.8615560640732266e-05,
|
| 301 |
+
"loss": 0.6192,
|
| 302 |
+
"step": 1350
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.16025641025641027,
|
| 306 |
+
"grad_norm": 0.5703125,
|
| 307 |
+
"learning_rate": 4.0045766590389014e-05,
|
| 308 |
+
"loss": 0.6121,
|
| 309 |
+
"step": 1400
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 0.16025641025641027,
|
| 313 |
+
"eval_loss": 1.7816270589828491,
|
| 314 |
+
"eval_runtime": 124.0951,
|
| 315 |
+
"eval_samples_per_second": 2.256,
|
| 316 |
+
"eval_steps_per_second": 0.564,
|
| 317 |
+
"step": 1400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16597985347985347,
|
| 321 |
+
"grad_norm": 0.609375,
|
| 322 |
+
"learning_rate": 4.147597254004577e-05,
|
| 323 |
+
"loss": 0.6039,
|
| 324 |
+
"step": 1450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.1717032967032967,
|
| 328 |
+
"grad_norm": 0.5625,
|
| 329 |
+
"learning_rate": 4.290617848970252e-05,
|
| 330 |
+
"loss": 0.5923,
|
| 331 |
+
"step": 1500
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1717032967032967,
|
| 335 |
+
"eval_loss": 1.8131866455078125,
|
| 336 |
+
"eval_runtime": 124.0957,
|
| 337 |
+
"eval_samples_per_second": 2.256,
|
| 338 |
+
"eval_steps_per_second": 0.564,
|
| 339 |
+
"step": 1500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.17742673992673993,
|
| 343 |
+
"grad_norm": 0.61328125,
|
| 344 |
+
"learning_rate": 4.433638443935927e-05,
|
| 345 |
+
"loss": 0.5973,
|
| 346 |
+
"step": 1550
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.18315018315018314,
|
| 350 |
+
"grad_norm": 0.62890625,
|
| 351 |
+
"learning_rate": 4.5766590389016025e-05,
|
| 352 |
+
"loss": 0.5908,
|
| 353 |
+
"step": 1600
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.18315018315018314,
|
| 357 |
+
"eval_loss": 1.7663753032684326,
|
| 358 |
+
"eval_runtime": 124.103,
|
| 359 |
+
"eval_samples_per_second": 2.256,
|
| 360 |
+
"eval_steps_per_second": 0.564,
|
| 361 |
+
"step": 1600
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 0.18887362637362637,
|
| 365 |
+
"grad_norm": 0.63671875,
|
| 366 |
+
"learning_rate": 4.719679633867277e-05,
|
| 367 |
+
"loss": 0.5869,
|
| 368 |
+
"step": 1650
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.1945970695970696,
|
| 372 |
+
"grad_norm": 0.6015625,
|
| 373 |
+
"learning_rate": 4.862700228832952e-05,
|
| 374 |
+
"loss": 0.5662,
|
| 375 |
+
"step": 1700
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 0.1945970695970696,
|
| 379 |
+
"eval_loss": 1.830661416053772,
|
| 380 |
+
"eval_runtime": 124.107,
|
| 381 |
+
"eval_samples_per_second": 2.256,
|
| 382 |
+
"eval_steps_per_second": 0.564,
|
| 383 |
+
"step": 1700
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 0.20032051282051283,
|
| 387 |
+
"grad_norm": 0.65234375,
|
| 388 |
+
"learning_rate": 4.9999989894357565e-05,
|
| 389 |
+
"loss": 0.5641,
|
| 390 |
+
"step": 1750
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 0.20604395604395603,
|
| 394 |
+
"grad_norm": 0.6171875,
|
| 395 |
+
"learning_rate": 4.999316889636665e-05,
|
| 396 |
+
"loss": 0.5637,
|
| 397 |
+
"step": 1800
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 0.20604395604395603,
|
| 401 |
+
"eval_loss": 1.7863534688949585,
|
| 402 |
+
"eval_runtime": 124.0697,
|
| 403 |
+
"eval_samples_per_second": 2.257,
|
| 404 |
+
"eval_steps_per_second": 0.564,
|
| 405 |
+
"step": 1800
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 0.21176739926739926,
|
| 409 |
+
"grad_norm": 0.63671875,
|
| 410 |
+
"learning_rate": 4.9973719827852006e-05,
|
| 411 |
+
"loss": 0.5566,
|
| 412 |
+
"step": 1850
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 0.2174908424908425,
|
| 416 |
+
"grad_norm": 0.578125,
|
| 417 |
+
"learning_rate": 4.994165251566713e-05,
|
| 418 |
+
"loss": 0.5475,
|
| 419 |
+
"step": 1900
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 0.2174908424908425,
|
| 423 |
+
"eval_loss": 1.7987805604934692,
|
| 424 |
+
"eval_runtime": 124.1161,
|
| 425 |
+
"eval_samples_per_second": 2.256,
|
| 426 |
+
"eval_steps_per_second": 0.564,
|
| 427 |
+
"step": 1900
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.22321428571428573,
|
| 431 |
+
"grad_norm": 0.6328125,
|
| 432 |
+
"learning_rate": 4.98969831621704e-05,
|
| 433 |
+
"loss": 0.5544,
|
| 434 |
+
"step": 1950
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.22893772893772893,
|
| 438 |
+
"grad_norm": 0.6953125,
|
| 439 |
+
"learning_rate": 4.98397343370387e-05,
|
| 440 |
+
"loss": 0.5421,
|
| 441 |
+
"step": 2000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.22893772893772893,
|
| 445 |
+
"eval_loss": 1.7876337766647339,
|
| 446 |
+
"eval_runtime": 124.1072,
|
| 447 |
+
"eval_samples_per_second": 2.256,
|
| 448 |
+
"eval_steps_per_second": 0.564,
|
| 449 |
+
"step": 2000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 0.23466117216117216,
|
| 453 |
+
"grad_norm": 0.58984375,
|
| 454 |
+
"learning_rate": 4.976993496586383e-05,
|
| 455 |
+
"loss": 0.5386,
|
| 456 |
+
"step": 2050
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 0.2403846153846154,
|
| 460 |
+
"grad_norm": 0.70703125,
|
| 461 |
+
"learning_rate": 4.968762031553753e-05,
|
| 462 |
+
"loss": 0.529,
|
| 463 |
+
"step": 2100
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 0.2403846153846154,
|
| 467 |
+
"eval_loss": 1.7660707235336304,
|
| 468 |
+
"eval_runtime": 124.1155,
|
| 469 |
+
"eval_samples_per_second": 2.256,
|
| 470 |
+
"eval_steps_per_second": 0.564,
|
| 471 |
+
"step": 2100
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24610805860805862,
|
| 475 |
+
"grad_norm": 0.625,
|
| 476 |
+
"learning_rate": 4.959283197643249e-05,
|
| 477 |
+
"loss": 0.5278,
|
| 478 |
+
"step": 2150
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.2518315018315018,
|
| 482 |
+
"grad_norm": 0.64453125,
|
| 483 |
+
"learning_rate": 4.948561784138841e-05,
|
| 484 |
+
"loss": 0.5202,
|
| 485 |
+
"step": 2200
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.2518315018315018,
|
| 489 |
+
"eval_loss": 1.770936131477356,
|
| 490 |
+
"eval_runtime": 124.1257,
|
| 491 |
+
"eval_samples_per_second": 2.256,
|
| 492 |
+
"eval_steps_per_second": 0.564,
|
| 493 |
+
"step": 2200
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.25755494505494503,
|
| 497 |
+
"grad_norm": 0.59375,
|
| 498 |
+
"learning_rate": 4.936603208151355e-05,
|
| 499 |
+
"loss": 0.5071,
|
| 500 |
+
"step": 2250
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.2632783882783883,
|
| 504 |
+
"grad_norm": 0.6796875,
|
| 505 |
+
"learning_rate": 4.9234135118814246e-05,
|
| 506 |
+
"loss": 0.5287,
|
| 507 |
+
"step": 2300
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.2632783882783883,
|
| 511 |
+
"eval_loss": 1.7680959701538086,
|
| 512 |
+
"eval_runtime": 124.1059,
|
| 513 |
+
"eval_samples_per_second": 2.256,
|
| 514 |
+
"eval_steps_per_second": 0.564,
|
| 515 |
+
"step": 2300
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 0.2690018315018315,
|
| 519 |
+
"grad_norm": 0.6171875,
|
| 520 |
+
"learning_rate": 4.908999359566602e-05,
|
| 521 |
+
"loss": 0.5116,
|
| 522 |
+
"step": 2350
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 0.27472527472527475,
|
| 526 |
+
"grad_norm": 0.61328125,
|
| 527 |
+
"learning_rate": 4.8933680341141775e-05,
|
| 528 |
+
"loss": 0.514,
|
| 529 |
+
"step": 2400
|
| 530 |
+
},
|
| 531 |
+
{
|
| 532 |
+
"epoch": 0.27472527472527475,
|
| 533 |
+
"eval_loss": 1.7765259742736816,
|
| 534 |
+
"eval_runtime": 124.1161,
|
| 535 |
+
"eval_samples_per_second": 2.256,
|
| 536 |
+
"eval_steps_per_second": 0.564,
|
| 537 |
+
"step": 2400
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 0.28044871794871795,
|
| 541 |
+
"grad_norm": 0.6328125,
|
| 542 |
+
"learning_rate": 4.8765274334214116e-05,
|
| 543 |
+
"loss": 0.5099,
|
| 544 |
+
"step": 2450
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 0.28617216117216115,
|
| 548 |
+
"grad_norm": 0.6875,
|
| 549 |
+
"learning_rate": 4.8584860663850404e-05,
|
| 550 |
+
"loss": 0.5026,
|
| 551 |
+
"step": 2500
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 0.28617216117216115,
|
| 555 |
+
"eval_loss": 1.7931022644042969,
|
| 556 |
+
"eval_runtime": 124.0904,
|
| 557 |
+
"eval_samples_per_second": 2.256,
|
| 558 |
+
"eval_steps_per_second": 0.564,
|
| 559 |
+
"step": 2500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 0.2918956043956044,
|
| 563 |
+
"grad_norm": 0.609375,
|
| 564 |
+
"learning_rate": 4.839253048602059e-05,
|
| 565 |
+
"loss": 0.5044,
|
| 566 |
+
"step": 2550
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.2976190476190476,
|
| 570 |
+
"grad_norm": 0.6484375,
|
| 571 |
+
"learning_rate": 4.818838097763967e-05,
|
| 572 |
+
"loss": 0.5038,
|
| 573 |
+
"step": 2600
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 0.2976190476190476,
|
| 577 |
+
"eval_loss": 1.7807551622390747,
|
| 578 |
+
"eval_runtime": 124.0961,
|
| 579 |
+
"eval_samples_per_second": 2.256,
|
| 580 |
+
"eval_steps_per_second": 0.564,
|
| 581 |
+
"step": 2600
|
| 582 |
+
},
|
| 583 |
+
{
|
| 584 |
+
"epoch": 0.3033424908424908,
|
| 585 |
+
"grad_norm": 0.65625,
|
| 586 |
+
"learning_rate": 4.7972515287468e-05,
|
| 587 |
+
"loss": 0.4828,
|
| 588 |
+
"step": 2650
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 0.3090659340659341,
|
| 592 |
+
"grad_norm": 0.64453125,
|
| 593 |
+
"learning_rate": 4.774504248399427e-05,
|
| 594 |
+
"loss": 0.5052,
|
| 595 |
+
"step": 2700
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 0.3090659340659341,
|
| 599 |
+
"eval_loss": 1.7688714265823364,
|
| 600 |
+
"eval_runtime": 124.0947,
|
| 601 |
+
"eval_samples_per_second": 2.256,
|
| 602 |
+
"eval_steps_per_second": 0.564,
|
| 603 |
+
"step": 2700
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 0.3147893772893773,
|
| 607 |
+
"grad_norm": 0.62890625,
|
| 608 |
+
"learning_rate": 4.750607750032748e-05,
|
| 609 |
+
"loss": 0.4938,
|
| 610 |
+
"step": 2750
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 0.32051282051282054,
|
| 614 |
+
"grad_norm": 0.74609375,
|
| 615 |
+
"learning_rate": 4.725574107612567e-05,
|
| 616 |
+
"loss": 0.4918,
|
| 617 |
+
"step": 2800
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 0.32051282051282054,
|
| 621 |
+
"eval_loss": 1.7862409353256226,
|
| 622 |
+
"eval_runtime": 124.0923,
|
| 623 |
+
"eval_samples_per_second": 2.256,
|
| 624 |
+
"eval_steps_per_second": 0.564,
|
| 625 |
+
"step": 2800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.32623626373626374,
|
| 629 |
+
"grad_norm": 0.6796875,
|
| 630 |
+
"learning_rate": 4.699415969659098e-05,
|
| 631 |
+
"loss": 0.4847,
|
| 632 |
+
"step": 2850
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.33195970695970695,
|
| 636 |
+
"grad_norm": 0.69140625,
|
| 637 |
+
"learning_rate": 4.672146552856155e-05,
|
| 638 |
+
"loss": 0.4817,
|
| 639 |
+
"step": 2900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.33195970695970695,
|
| 643 |
+
"eval_loss": 1.7916373014450073,
|
| 644 |
+
"eval_runtime": 124.0857,
|
| 645 |
+
"eval_samples_per_second": 2.257,
|
| 646 |
+
"eval_steps_per_second": 0.564,
|
| 647 |
+
"step": 2900
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.3376831501831502,
|
| 651 |
+
"grad_norm": 0.67578125,
|
| 652 |
+
"learning_rate": 4.6437796353732824e-05,
|
| 653 |
+
"loss": 0.4908,
|
| 654 |
+
"step": 2950
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.3434065934065934,
|
| 658 |
+
"grad_norm": 0.68359375,
|
| 659 |
+
"learning_rate": 4.614329549904187e-05,
|
| 660 |
+
"loss": 0.4806,
|
| 661 |
+
"step": 3000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.3434065934065934,
|
| 665 |
+
"eval_loss": 1.7795602083206177,
|
| 666 |
+
"eval_runtime": 124.1973,
|
| 667 |
+
"eval_samples_per_second": 2.254,
|
| 668 |
+
"eval_steps_per_second": 0.564,
|
| 669 |
+
"step": 3000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 0.3491300366300366,
|
| 673 |
+
"grad_norm": 0.65234375,
|
| 674 |
+
"learning_rate": 4.583811176424984e-05,
|
| 675 |
+
"loss": 0.4831,
|
| 676 |
+
"step": 3050
|
| 677 |
+
},
|
| 678 |
+
{
|
| 679 |
+
"epoch": 0.35485347985347987,
|
| 680 |
+
"grad_norm": 0.6640625,
|
| 681 |
+
"learning_rate": 4.5522399346759304e-05,
|
| 682 |
+
"loss": 0.4849,
|
| 683 |
+
"step": 3100
|
| 684 |
+
},
|
| 685 |
+
{
|
| 686 |
+
"epoch": 0.35485347985347987,
|
| 687 |
+
"eval_loss": 1.76537024974823,
|
| 688 |
+
"eval_runtime": 124.1057,
|
| 689 |
+
"eval_samples_per_second": 2.256,
|
| 690 |
+
"eval_steps_per_second": 0.564,
|
| 691 |
+
"step": 3100
|
| 692 |
+
},
|
| 693 |
+
{
|
| 694 |
+
"epoch": 0.3605769230769231,
|
| 695 |
+
"grad_norm": 0.6171875,
|
| 696 |
+
"learning_rate": 4.51963177637043e-05,
|
| 697 |
+
"loss": 0.4787,
|
| 698 |
+
"step": 3150
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.3663003663003663,
|
| 702 |
+
"grad_norm": 0.640625,
|
| 703 |
+
"learning_rate": 4.4860031771352626e-05,
|
| 704 |
+
"loss": 0.4784,
|
| 705 |
+
"step": 3200
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 0.3663003663003663,
|
| 709 |
+
"eval_loss": 1.7576422691345215,
|
| 710 |
+
"eval_runtime": 124.109,
|
| 711 |
+
"eval_samples_per_second": 2.256,
|
| 712 |
+
"eval_steps_per_second": 0.564,
|
| 713 |
+
"step": 3200
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 0.37202380952380953,
|
| 717 |
+
"grad_norm": 0.70703125,
|
| 718 |
+
"learning_rate": 4.451371128186087e-05,
|
| 719 |
+
"loss": 0.485,
|
| 720 |
+
"step": 3250
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 0.37774725274725274,
|
| 724 |
+
"grad_norm": 0.64453125,
|
| 725 |
+
"learning_rate": 4.4157531277424503e-05,
|
| 726 |
+
"loss": 0.4712,
|
| 727 |
+
"step": 3300
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 0.37774725274725274,
|
| 731 |
+
"eval_loss": 1.7745938301086426,
|
| 732 |
+
"eval_runtime": 124.0704,
|
| 733 |
+
"eval_samples_per_second": 2.257,
|
| 734 |
+
"eval_steps_per_second": 0.564,
|
| 735 |
+
"step": 3300
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 0.383470695970696,
|
| 739 |
+
"grad_norm": 0.6953125,
|
| 740 |
+
"learning_rate": 4.379167172186619e-05,
|
| 741 |
+
"loss": 0.4899,
|
| 742 |
+
"step": 3350
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 0.3891941391941392,
|
| 746 |
+
"grad_norm": 0.671875,
|
| 747 |
+
"learning_rate": 4.3416317469707125e-05,
|
| 748 |
+
"loss": 0.4715,
|
| 749 |
+
"step": 3400
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 0.3891941391941392,
|
| 753 |
+
"eval_loss": 1.7567591667175293,
|
| 754 |
+
"eval_runtime": 124.0669,
|
| 755 |
+
"eval_samples_per_second": 2.257,
|
| 756 |
+
"eval_steps_per_second": 0.564,
|
| 757 |
+
"step": 3400
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 0.3949175824175824,
|
| 761 |
+
"grad_norm": 0.65625,
|
| 762 |
+
"learning_rate": 4.3031658172767266e-05,
|
| 763 |
+
"loss": 0.472,
|
| 764 |
+
"step": 3450
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.40064102564102566,
|
| 768 |
+
"grad_norm": 0.65625,
|
| 769 |
+
"learning_rate": 4.263788818434168e-05,
|
| 770 |
+
"loss": 0.4608,
|
| 771 |
+
"step": 3500
|
| 772 |
+
},
|
| 773 |
+
{
|
| 774 |
+
"epoch": 0.40064102564102566,
|
| 775 |
+
"eval_loss": 1.7424110174179077,
|
| 776 |
+
"eval_runtime": 124.063,
|
| 777 |
+
"eval_samples_per_second": 2.257,
|
| 778 |
+
"eval_steps_per_second": 0.564,
|
| 779 |
+
"step": 3500
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.40636446886446886,
|
| 783 |
+
"grad_norm": 0.63671875,
|
| 784 |
+
"learning_rate": 4.223520646100145e-05,
|
| 785 |
+
"loss": 0.4586,
|
| 786 |
+
"step": 3550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.41208791208791207,
|
| 790 |
+
"grad_norm": 0.68359375,
|
| 791 |
+
"learning_rate": 4.182381646206868e-05,
|
| 792 |
+
"loss": 0.4629,
|
| 793 |
+
"step": 3600
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.41208791208791207,
|
| 797 |
+
"eval_loss": 1.7561120986938477,
|
| 798 |
+
"eval_runtime": 124.1336,
|
| 799 |
+
"eval_samples_per_second": 2.256,
|
| 800 |
+
"eval_steps_per_second": 0.564,
|
| 801 |
+
"step": 3600
|
| 802 |
+
},
|
| 803 |
+
{
|
| 804 |
+
"epoch": 0.4178113553113553,
|
| 805 |
+
"grad_norm": 0.6875,
|
| 806 |
+
"learning_rate": 4.140392604681646e-05,
|
| 807 |
+
"loss": 0.4655,
|
| 808 |
+
"step": 3650
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.42353479853479853,
|
| 812 |
+
"grad_norm": 0.7109375,
|
| 813 |
+
"learning_rate": 4.097574736944575e-05,
|
| 814 |
+
"loss": 0.4591,
|
| 815 |
+
"step": 3700
|
| 816 |
+
},
|
| 817 |
+
{
|
| 818 |
+
"epoch": 0.42353479853479853,
|
| 819 |
+
"eval_loss": 1.7497508525848389,
|
| 820 |
+
"eval_runtime": 124.0856,
|
| 821 |
+
"eval_samples_per_second": 2.257,
|
| 822 |
+
"eval_steps_per_second": 0.564,
|
| 823 |
+
"step": 3700
|
| 824 |
+
},
|
| 825 |
+
{
|
| 826 |
+
"epoch": 0.42925824175824173,
|
| 827 |
+
"grad_norm": 0.7578125,
|
| 828 |
+
"learning_rate": 4.053949677189208e-05,
|
| 829 |
+
"loss": 0.4658,
|
| 830 |
+
"step": 3750
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.434981684981685,
|
| 834 |
+
"grad_norm": 0.671875,
|
| 835 |
+
"learning_rate": 4.0095394674516506e-05,
|
| 836 |
+
"loss": 0.4652,
|
| 837 |
+
"step": 3800
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 0.434981684981685,
|
| 841 |
+
"eval_loss": 1.736577033996582,
|
| 842 |
+
"eval_runtime": 124.0959,
|
| 843 |
+
"eval_samples_per_second": 2.256,
|
| 844 |
+
"eval_steps_per_second": 0.564,
|
| 845 |
+
"step": 3800
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 0.4407051282051282,
|
| 849 |
+
"grad_norm": 0.7109375,
|
| 850 |
+
"learning_rate": 3.96436654647358e-05,
|
| 851 |
+
"loss": 0.4588,
|
| 852 |
+
"step": 3850
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 0.44642857142857145,
|
| 856 |
+
"grad_norm": 0.66796875,
|
| 857 |
+
"learning_rate": 3.918453738364824e-05,
|
| 858 |
+
"loss": 0.461,
|
| 859 |
+
"step": 3900
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 0.44642857142857145,
|
| 863 |
+
"eval_loss": 1.7393635511398315,
|
| 864 |
+
"eval_runtime": 124.0934,
|
| 865 |
+
"eval_samples_per_second": 2.256,
|
| 866 |
+
"eval_steps_per_second": 0.564,
|
| 867 |
+
"step": 3900
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 0.45215201465201466,
|
| 871 |
+
"grad_norm": 0.6875,
|
| 872 |
+
"learning_rate": 3.871824241071236e-05,
|
| 873 |
+
"loss": 0.4493,
|
| 874 |
+
"step": 3950
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 0.45787545787545786,
|
| 878 |
+
"grad_norm": 0.6796875,
|
| 879 |
+
"learning_rate": 3.824501614653676e-05,
|
| 880 |
+
"loss": 0.4469,
|
| 881 |
+
"step": 4000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 0.45787545787545786,
|
| 885 |
+
"eval_loss": 1.7396734952926636,
|
| 886 |
+
"eval_runtime": 124.0876,
|
| 887 |
+
"eval_samples_per_second": 2.256,
|
| 888 |
+
"eval_steps_per_second": 0.564,
|
| 889 |
+
"step": 4000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 0.4635989010989011,
|
| 893 |
+
"grad_norm": 0.73828125,
|
| 894 |
+
"learning_rate": 3.7765097693840385e-05,
|
| 895 |
+
"loss": 0.4496,
|
| 896 |
+
"step": 4050
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.4693223443223443,
|
| 900 |
+
"grad_norm": 0.74609375,
|
| 901 |
+
"learning_rate": 3.727872953664322e-05,
|
| 902 |
+
"loss": 0.4521,
|
| 903 |
+
"step": 4100
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 0.4693223443223443,
|
| 907 |
+
"eval_loss": 1.7555357217788696,
|
| 908 |
+
"eval_runtime": 124.0832,
|
| 909 |
+
"eval_samples_per_second": 2.257,
|
| 910 |
+
"eval_steps_per_second": 0.564,
|
| 911 |
+
"step": 4100
|
| 912 |
+
},
|
| 913 |
+
{
|
| 914 |
+
"epoch": 0.4750457875457875,
|
| 915 |
+
"grad_norm": 0.671875,
|
| 916 |
+
"learning_rate": 3.678615741774861e-05,
|
| 917 |
+
"loss": 0.4598,
|
| 918 |
+
"step": 4150
|
| 919 |
+
},
|
| 920 |
+
{
|
| 921 |
+
"epoch": 0.4807692307692308,
|
| 922 |
+
"grad_norm": 0.77734375,
|
| 923 |
+
"learning_rate": 3.628763021457909e-05,
|
| 924 |
+
"loss": 0.4498,
|
| 925 |
+
"step": 4200
|
| 926 |
+
},
|
| 927 |
+
{
|
| 928 |
+
"epoch": 0.4807692307692308,
|
| 929 |
+
"eval_loss": 1.7651796340942383,
|
| 930 |
+
"eval_runtime": 124.0848,
|
| 931 |
+
"eval_samples_per_second": 2.257,
|
| 932 |
+
"eval_steps_per_second": 0.564,
|
| 933 |
+
"step": 4200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.486492673992674,
|
| 937 |
+
"grad_norm": 0.68359375,
|
| 938 |
+
"learning_rate": 3.57833998134283e-05,
|
| 939 |
+
"loss": 0.4471,
|
| 940 |
+
"step": 4250
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.49221611721611724,
|
| 944 |
+
"grad_norm": 0.6953125,
|
| 945 |
+
"learning_rate": 3.5273720982192716e-05,
|
| 946 |
+
"loss": 0.4541,
|
| 947 |
+
"step": 4300
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.49221611721611724,
|
| 951 |
+
"eval_loss": 1.758300542831421,
|
| 952 |
+
"eval_runtime": 124.0895,
|
| 953 |
+
"eval_samples_per_second": 2.256,
|
| 954 |
+
"eval_steps_per_second": 0.564,
|
| 955 |
+
"step": 4300
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 0.49793956043956045,
|
| 959 |
+
"grad_norm": 0.76953125,
|
| 960 |
+
"learning_rate": 3.475885124164737e-05,
|
| 961 |
+
"loss": 0.4595,
|
| 962 |
+
"step": 4350
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.5036630036630036,
|
| 966 |
+
"grad_norm": 0.65625,
|
| 967 |
+
"learning_rate": 3.4239050735330754e-05,
|
| 968 |
+
"loss": 0.4594,
|
| 969 |
+
"step": 4400
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 0.5036630036630036,
|
| 973 |
+
"eval_loss": 1.7604867219924927,
|
| 974 |
+
"eval_runtime": 124.0898,
|
| 975 |
+
"eval_samples_per_second": 2.256,
|
| 976 |
+
"eval_steps_per_second": 0.564,
|
| 977 |
+
"step": 4400
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 0.5093864468864469,
|
| 981 |
+
"grad_norm": 0.67578125,
|
| 982 |
+
"learning_rate": 3.371458209810437e-05,
|
| 983 |
+
"loss": 0.4584,
|
| 984 |
+
"step": 4450
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 0.5151098901098901,
|
| 988 |
+
"grad_norm": 0.90625,
|
| 989 |
+
"learning_rate": 3.3185710323453684e-05,
|
| 990 |
+
"loss": 0.4514,
|
| 991 |
+
"step": 4500
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 0.5151098901098901,
|
| 995 |
+
"eval_loss": 1.7685655355453491,
|
| 996 |
+
"eval_runtime": 124.0911,
|
| 997 |
+
"eval_samples_per_second": 2.256,
|
| 998 |
+
"eval_steps_per_second": 0.564,
|
| 999 |
+
"step": 4500
|
| 1000 |
+
},
|
| 1001 |
+
{
|
| 1002 |
+
"epoch": 0.5208333333333334,
|
| 1003 |
+
"grad_norm": 0.67578125,
|
| 1004 |
+
"learning_rate": 3.265270262959723e-05,
|
| 1005 |
+
"loss": 0.4523,
|
| 1006 |
+
"step": 4550
|
| 1007 |
+
},
|
| 1008 |
+
{
|
| 1009 |
+
"epoch": 0.5265567765567766,
|
| 1010 |
+
"grad_norm": 0.65625,
|
| 1011 |
+
"learning_rate": 3.211582832447175e-05,
|
| 1012 |
+
"loss": 0.4395,
|
| 1013 |
+
"step": 4600
|
| 1014 |
+
},
|
| 1015 |
+
{
|
| 1016 |
+
"epoch": 0.5265567765567766,
|
| 1017 |
+
"eval_loss": 1.7713632583618164,
|
| 1018 |
+
"eval_runtime": 124.0855,
|
| 1019 |
+
"eval_samples_per_second": 2.257,
|
| 1020 |
+
"eval_steps_per_second": 0.564,
|
| 1021 |
+
"step": 4600
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 0.5322802197802198,
|
| 1025 |
+
"grad_norm": 0.78125,
|
| 1026 |
+
"learning_rate": 3.1575358669661356e-05,
|
| 1027 |
+
"loss": 0.4464,
|
| 1028 |
+
"step": 4650
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 0.538003663003663,
|
| 1032 |
+
"grad_norm": 0.72265625,
|
| 1033 |
+
"learning_rate": 3.103156674333976e-05,
|
| 1034 |
+
"loss": 0.4384,
|
| 1035 |
+
"step": 4700
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 0.538003663003663,
|
| 1039 |
+
"eval_loss": 1.788908839225769,
|
| 1040 |
+
"eval_runtime": 124.0945,
|
| 1041 |
+
"eval_samples_per_second": 2.256,
|
| 1042 |
+
"eval_steps_per_second": 0.564,
|
| 1043 |
+
"step": 4700
|
| 1044 |
+
},
|
| 1045 |
+
{
|
| 1046 |
+
"epoch": 0.5437271062271062,
|
| 1047 |
+
"grad_norm": 0.69921875,
|
| 1048 |
+
"learning_rate": 3.0484727302294475e-05,
|
| 1049 |
+
"loss": 0.4446,
|
| 1050 |
+
"step": 4750
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 0.5494505494505495,
|
| 1054 |
+
"grad_norm": 0.6875,
|
| 1055 |
+
"learning_rate": 2.9935116643102983e-05,
|
| 1056 |
+
"loss": 0.4392,
|
| 1057 |
+
"step": 4800
|
| 1058 |
+
},
|
| 1059 |
+
{
|
| 1060 |
+
"epoch": 0.5494505494505495,
|
| 1061 |
+
"eval_loss": 1.7709113359451294,
|
| 1062 |
+
"eval_runtime": 124.088,
|
| 1063 |
+
"eval_samples_per_second": 2.256,
|
| 1064 |
+
"eval_steps_per_second": 0.564,
|
| 1065 |
+
"step": 4800
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 0.5551739926739927,
|
| 1069 |
+
"grad_norm": 0.7109375,
|
| 1070 |
+
"learning_rate": 2.9383012462530895e-05,
|
| 1071 |
+
"loss": 0.4406,
|
| 1072 |
+
"step": 4850
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 0.5608974358974359,
|
| 1076 |
+
"grad_norm": 0.6953125,
|
| 1077 |
+
"learning_rate": 2.8828693717222625e-05,
|
| 1078 |
+
"loss": 0.4495,
|
| 1079 |
+
"step": 4900
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 0.5608974358974359,
|
| 1083 |
+
"eval_loss": 1.7554136514663696,
|
| 1084 |
+
"eval_runtime": 124.0986,
|
| 1085 |
+
"eval_samples_per_second": 2.256,
|
| 1086 |
+
"eval_steps_per_second": 0.564,
|
| 1087 |
+
"step": 4900
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5666208791208791,
|
| 1091 |
+
"grad_norm": 0.7109375,
|
| 1092 |
+
"learning_rate": 2.8272440482755535e-05,
|
| 1093 |
+
"loss": 0.4433,
|
| 1094 |
+
"step": 4950
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5723443223443223,
|
| 1098 |
+
"grad_norm": 0.7890625,
|
| 1099 |
+
"learning_rate": 2.771453381212865e-05,
|
| 1100 |
+
"loss": 0.4375,
|
| 1101 |
+
"step": 5000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5723443223443223,
|
| 1105 |
+
"eval_loss": 1.7531843185424805,
|
| 1106 |
+
"eval_runtime": 124.0918,
|
| 1107 |
+
"eval_samples_per_second": 2.256,
|
| 1108 |
+
"eval_steps_per_second": 0.564,
|
| 1109 |
+
"step": 5000
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 0.5780677655677655,
|
| 1113 |
+
"grad_norm": 0.7578125,
|
| 1114 |
+
"learning_rate": 2.715525559375764e-05,
|
| 1115 |
+
"loss": 0.4405,
|
| 1116 |
+
"step": 5050
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 0.5837912087912088,
|
| 1120 |
+
"grad_norm": 0.72265625,
|
| 1121 |
+
"learning_rate": 2.6594888409047557e-05,
|
| 1122 |
+
"loss": 0.4441,
|
| 1123 |
+
"step": 5100
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 0.5837912087912088,
|
| 1127 |
+
"eval_loss": 1.7770174741744995,
|
| 1128 |
+
"eval_runtime": 124.0878,
|
| 1129 |
+
"eval_samples_per_second": 2.256,
|
| 1130 |
+
"eval_steps_per_second": 0.564,
|
| 1131 |
+
"step": 5100
|
| 1132 |
+
},
|
| 1133 |
+
{
|
| 1134 |
+
"epoch": 0.589514652014652,
|
| 1135 |
+
"grad_norm": 0.77734375,
|
| 1136 |
+
"learning_rate": 2.6033715389615588e-05,
|
| 1137 |
+
"loss": 0.4403,
|
| 1138 |
+
"step": 5150
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 0.5952380952380952,
|
| 1142 |
+
"grad_norm": 0.7890625,
|
| 1143 |
+
"learning_rate": 2.5472020074235635e-05,
|
| 1144 |
+
"loss": 0.4458,
|
| 1145 |
+
"step": 5200
|
| 1146 |
+
},
|
| 1147 |
+
{
|
| 1148 |
+
"epoch": 0.5952380952380952,
|
| 1149 |
+
"eval_loss": 1.7528095245361328,
|
| 1150 |
+
"eval_runtime": 124.0953,
|
| 1151 |
+
"eval_samples_per_second": 2.256,
|
| 1152 |
+
"eval_steps_per_second": 0.564,
|
| 1153 |
+
"step": 5200
|
| 1154 |
+
},
|
| 1155 |
+
{
|
| 1156 |
+
"epoch": 0.6009615384615384,
|
| 1157 |
+
"grad_norm": 0.62890625,
|
| 1158 |
+
"learning_rate": 2.4910086265577364e-05,
|
| 1159 |
+
"loss": 0.4368,
|
| 1160 |
+
"step": 5250
|
| 1161 |
+
},
|
| 1162 |
+
{
|
| 1163 |
+
"epoch": 0.6066849816849816,
|
| 1164 |
+
"grad_norm": 0.734375,
|
| 1165 |
+
"learning_rate": 2.4348197886811702e-05,
|
| 1166 |
+
"loss": 0.4343,
|
| 1167 |
+
"step": 5300
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 0.6066849816849816,
|
| 1171 |
+
"eval_loss": 1.7645584344863892,
|
| 1172 |
+
"eval_runtime": 124.0831,
|
| 1173 |
+
"eval_samples_per_second": 2.257,
|
| 1174 |
+
"eval_steps_per_second": 0.564,
|
| 1175 |
+
"step": 5300
|
| 1176 |
+
},
|
| 1177 |
+
{
|
| 1178 |
+
"epoch": 0.612408424908425,
|
| 1179 |
+
"grad_norm": 0.73828125,
|
| 1180 |
+
"learning_rate": 2.3786638838155694e-05,
|
| 1181 |
+
"loss": 0.4444,
|
| 1182 |
+
"step": 5350
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 0.6181318681318682,
|
| 1186 |
+
"grad_norm": 0.7578125,
|
| 1187 |
+
"learning_rate": 2.32256928534287e-05,
|
| 1188 |
+
"loss": 0.433,
|
| 1189 |
+
"step": 5400
|
| 1190 |
+
},
|
| 1191 |
+
{
|
| 1192 |
+
"epoch": 0.6181318681318682,
|
| 1193 |
+
"eval_loss": 1.7689203023910522,
|
| 1194 |
+
"eval_runtime": 124.0857,
|
| 1195 |
+
"eval_samples_per_second": 2.257,
|
| 1196 |
+
"eval_steps_per_second": 0.564,
|
| 1197 |
+
"step": 5400
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 0.6238553113553114,
|
| 1201 |
+
"grad_norm": 0.70703125,
|
| 1202 |
+
"learning_rate": 2.2665643356692923e-05,
|
| 1203 |
+
"loss": 0.4344,
|
| 1204 |
+
"step": 5450
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 0.6295787545787546,
|
| 1208 |
+
"grad_norm": 0.65625,
|
| 1209 |
+
"learning_rate": 2.210677331905019e-05,
|
| 1210 |
+
"loss": 0.4371,
|
| 1211 |
+
"step": 5500
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 0.6295787545787546,
|
| 1215 |
+
"eval_loss": 1.7737501859664917,
|
| 1216 |
+
"eval_runtime": 124.0948,
|
| 1217 |
+
"eval_samples_per_second": 2.256,
|
| 1218 |
+
"eval_steps_per_second": 0.564,
|
| 1219 |
+
"step": 5500
|
| 1220 |
+
},
|
| 1221 |
+
{
|
| 1222 |
+
"epoch": 0.6353021978021978,
|
| 1223 |
+
"grad_norm": 0.83984375,
|
| 1224 |
+
"learning_rate": 2.1549365115667853e-05,
|
| 1225 |
+
"loss": 0.4377,
|
| 1226 |
+
"step": 5550
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 0.6410256410256411,
|
| 1230 |
+
"grad_norm": 0.76171875,
|
| 1231 |
+
"learning_rate": 2.099370038310553e-05,
|
| 1232 |
+
"loss": 0.4376,
|
| 1233 |
+
"step": 5600
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 0.6410256410256411,
|
| 1237 |
+
"eval_loss": 1.7633239030838013,
|
| 1238 |
+
"eval_runtime": 124.0669,
|
| 1239 |
+
"eval_samples_per_second": 2.257,
|
| 1240 |
+
"eval_steps_per_second": 0.564,
|
| 1241 |
+
"step": 5600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6467490842490843,
|
| 1245 |
+
"grad_norm": 0.71875,
|
| 1246 |
+
"learning_rate": 2.044005987701531e-05,
|
| 1247 |
+
"loss": 0.4388,
|
| 1248 |
+
"step": 5650
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6524725274725275,
|
| 1252 |
+
"grad_norm": 0.7734375,
|
| 1253 |
+
"learning_rate": 1.9888723330286763e-05,
|
| 1254 |
+
"loss": 0.4366,
|
| 1255 |
+
"step": 5700
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6524725274725275,
|
| 1259 |
+
"eval_loss": 1.7809503078460693,
|
| 1260 |
+
"eval_runtime": 124.1026,
|
| 1261 |
+
"eval_samples_per_second": 2.256,
|
| 1262 |
+
"eval_steps_per_second": 0.564,
|
| 1263 |
+
"step": 5700
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 0.6581959706959707,
|
| 1267 |
+
"grad_norm": 0.78515625,
|
| 1268 |
+
"learning_rate": 1.933996931170898e-05,
|
| 1269 |
+
"loss": 0.4333,
|
| 1270 |
+
"step": 5750
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 0.6639194139194139,
|
| 1274 |
+
"grad_norm": 0.7421875,
|
| 1275 |
+
"learning_rate": 1.879407508522056e-05,
|
| 1276 |
+
"loss": 0.43,
|
| 1277 |
+
"step": 5800
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 0.6639194139194139,
|
| 1281 |
+
"eval_loss": 1.7684820890426636,
|
| 1282 |
+
"eval_runtime": 124.0911,
|
| 1283 |
+
"eval_samples_per_second": 2.256,
|
| 1284 |
+
"eval_steps_per_second": 0.564,
|
| 1285 |
+
"step": 5800
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 0.6696428571428571,
|
| 1289 |
+
"grad_norm": 0.82421875,
|
| 1290 |
+
"learning_rate": 1.8251316469819075e-05,
|
| 1291 |
+
"loss": 0.4318,
|
| 1292 |
+
"step": 5850
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 0.6753663003663004,
|
| 1296 |
+
"grad_norm": 0.765625,
|
| 1297 |
+
"learning_rate": 1.7711967700200435e-05,
|
| 1298 |
+
"loss": 0.4345,
|
| 1299 |
+
"step": 5900
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 0.6753663003663004,
|
| 1303 |
+
"eval_loss": 1.7761142253875732,
|
| 1304 |
+
"eval_runtime": 124.0752,
|
| 1305 |
+
"eval_samples_per_second": 2.257,
|
| 1306 |
+
"eval_steps_per_second": 0.564,
|
| 1307 |
+
"step": 5900
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 0.6810897435897436,
|
| 1311 |
+
"grad_norm": 0.78125,
|
| 1312 |
+
"learning_rate": 1.7176301288198894e-05,
|
| 1313 |
+
"loss": 0.4362,
|
| 1314 |
+
"step": 5950
|
| 1315 |
+
},
|
| 1316 |
+
{
|
| 1317 |
+
"epoch": 0.6868131868131868,
|
| 1318 |
+
"grad_norm": 0.8671875,
|
| 1319 |
+
"learning_rate": 1.6644587885097457e-05,
|
| 1320 |
+
"loss": 0.4379,
|
| 1321 |
+
"step": 6000
|
| 1322 |
+
},
|
| 1323 |
+
{
|
| 1324 |
+
"epoch": 0.6868131868131868,
|
| 1325 |
+
"eval_loss": 1.7782317399978638,
|
| 1326 |
+
"eval_runtime": 124.0838,
|
| 1327 |
+
"eval_samples_per_second": 2.257,
|
| 1328 |
+
"eval_steps_per_second": 0.564,
|
| 1329 |
+
"step": 6000
|
| 1330 |
+
},
|
| 1331 |
+
{
|
| 1332 |
+
"epoch": 0.69253663003663,
|
| 1333 |
+
"grad_norm": 0.80078125,
|
| 1334 |
+
"learning_rate": 1.611709614487835e-05,
|
| 1335 |
+
"loss": 0.4379,
|
| 1336 |
+
"step": 6050
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 0.6982600732600732,
|
| 1340 |
+
"grad_norm": 0.75,
|
| 1341 |
+
"learning_rate": 1.5594092588482718e-05,
|
| 1342 |
+
"loss": 0.4294,
|
| 1343 |
+
"step": 6100
|
| 1344 |
+
},
|
| 1345 |
+
{
|
| 1346 |
+
"epoch": 0.6982600732600732,
|
| 1347 |
+
"eval_loss": 1.7736785411834717,
|
| 1348 |
+
"eval_runtime": 124.0841,
|
| 1349 |
+
"eval_samples_per_second": 2.257,
|
| 1350 |
+
"eval_steps_per_second": 0.564,
|
| 1351 |
+
"step": 6100
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 0.7039835164835165,
|
| 1355 |
+
"grad_norm": 0.75,
|
| 1356 |
+
"learning_rate": 1.5075841469147974e-05,
|
| 1357 |
+
"loss": 0.4356,
|
| 1358 |
+
"step": 6150
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 0.7097069597069597,
|
| 1362 |
+
"grad_norm": 0.84375,
|
| 1363 |
+
"learning_rate": 1.456260463889102e-05,
|
| 1364 |
+
"loss": 0.4441,
|
| 1365 |
+
"step": 6200
|
| 1366 |
+
},
|
| 1367 |
+
{
|
| 1368 |
+
"epoch": 0.7097069597069597,
|
| 1369 |
+
"eval_loss": 1.7646363973617554,
|
| 1370 |
+
"eval_runtime": 124.1036,
|
| 1371 |
+
"eval_samples_per_second": 2.256,
|
| 1372 |
+
"eval_steps_per_second": 0.564,
|
| 1373 |
+
"step": 6200
|
| 1374 |
+
},
|
| 1375 |
+
{
|
| 1376 |
+
"epoch": 0.7154304029304029,
|
| 1377 |
+
"grad_norm": 0.7109375,
|
| 1378 |
+
"learning_rate": 1.4054641416204609e-05,
|
| 1379 |
+
"loss": 0.4347,
|
| 1380 |
+
"step": 6250
|
| 1381 |
+
},
|
| 1382 |
+
{
|
| 1383 |
+
"epoch": 0.7211538461538461,
|
| 1384 |
+
"grad_norm": 0.828125,
|
| 1385 |
+
"learning_rate": 1.3552208455033932e-05,
|
| 1386 |
+
"loss": 0.4396,
|
| 1387 |
+
"step": 6300
|
| 1388 |
+
},
|
| 1389 |
+
{
|
| 1390 |
+
"epoch": 0.7211538461538461,
|
| 1391 |
+
"eval_loss": 1.7779277563095093,
|
| 1392 |
+
"eval_runtime": 124.0892,
|
| 1393 |
+
"eval_samples_per_second": 2.256,
|
| 1394 |
+
"eval_steps_per_second": 0.564,
|
| 1395 |
+
"step": 6300
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.7268772893772893,
|
| 1399 |
+
"grad_norm": 0.75,
|
| 1400 |
+
"learning_rate": 1.3055559615099353e-05,
|
| 1401 |
+
"loss": 0.4399,
|
| 1402 |
+
"step": 6350
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.7326007326007326,
|
| 1406 |
+
"grad_norm": 0.76953125,
|
| 1407 |
+
"learning_rate": 1.256494583363104e-05,
|
| 1408 |
+
"loss": 0.4307,
|
| 1409 |
+
"step": 6400
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.7326007326007326,
|
| 1413 |
+
"eval_loss": 1.7766470909118652,
|
| 1414 |
+
"eval_runtime": 124.1063,
|
| 1415 |
+
"eval_samples_per_second": 2.256,
|
| 1416 |
+
"eval_steps_per_second": 0.564,
|
| 1417 |
+
"step": 6400
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 0.7383241758241759,
|
| 1421 |
+
"grad_norm": 0.79296875,
|
| 1422 |
+
"learning_rate": 1.2080614998580212e-05,
|
| 1423 |
+
"loss": 0.4351,
|
| 1424 |
+
"step": 6450
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 0.7440476190476191,
|
| 1428 |
+
"grad_norm": 0.7265625,
|
| 1429 |
+
"learning_rate": 1.1602811823371069e-05,
|
| 1430 |
+
"loss": 0.4331,
|
| 1431 |
+
"step": 6500
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 0.7440476190476191,
|
| 1435 |
+
"eval_loss": 1.7733304500579834,
|
| 1436 |
+
"eval_runtime": 124.0732,
|
| 1437 |
+
"eval_samples_per_second": 2.257,
|
| 1438 |
+
"eval_steps_per_second": 0.564,
|
| 1439 |
+
"step": 6500
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 0.7497710622710623,
|
| 1443 |
+
"grad_norm": 0.734375,
|
| 1444 |
+
"learning_rate": 1.1131777723256629e-05,
|
| 1445 |
+
"loss": 0.4373,
|
| 1446 |
+
"step": 6550
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 0.7554945054945055,
|
| 1450 |
+
"grad_norm": 0.6484375,
|
| 1451 |
+
"learning_rate": 1.0667750693341072e-05,
|
| 1452 |
+
"loss": 0.4326,
|
| 1453 |
+
"step": 6600
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 0.7554945054945055,
|
| 1457 |
+
"eval_loss": 1.7796032428741455,
|
| 1458 |
+
"eval_runtime": 124.0885,
|
| 1459 |
+
"eval_samples_per_second": 2.256,
|
| 1460 |
+
"eval_steps_per_second": 0.564,
|
| 1461 |
+
"step": 6600
|
| 1462 |
+
},
|
| 1463 |
+
{
|
| 1464 |
+
"epoch": 0.7612179487179487,
|
| 1465 |
+
"grad_norm": 0.71875,
|
| 1466 |
+
"learning_rate": 1.0210965188330119e-05,
|
| 1467 |
+
"loss": 0.4314,
|
| 1468 |
+
"step": 6650
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 0.766941391941392,
|
| 1472 |
+
"grad_norm": 0.76171875,
|
| 1473 |
+
"learning_rate": 9.761652004070241e-06,
|
| 1474 |
+
"loss": 0.4286,
|
| 1475 |
+
"step": 6700
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 0.766941391941392,
|
| 1479 |
+
"eval_loss": 1.7802751064300537,
|
| 1480 |
+
"eval_runtime": 124.0901,
|
| 1481 |
+
"eval_samples_per_second": 2.256,
|
| 1482 |
+
"eval_steps_per_second": 0.564,
|
| 1483 |
+
"step": 6700
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 0.7726648351648352,
|
| 1487 |
+
"grad_norm": 0.84375,
|
| 1488 |
+
"learning_rate": 9.320038160936568e-06,
|
| 1489 |
+
"loss": 0.44,
|
| 1490 |
+
"step": 6750
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 0.7783882783882784,
|
| 1494 |
+
"grad_norm": 0.7890625,
|
| 1495 |
+
"learning_rate": 8.886346789128305e-06,
|
| 1496 |
+
"loss": 0.4294,
|
| 1497 |
+
"step": 6800
|
| 1498 |
+
},
|
| 1499 |
+
{
|
| 1500 |
+
"epoch": 0.7783882783882784,
|
| 1501 |
+
"eval_loss": 1.7787123918533325,
|
| 1502 |
+
"eval_runtime": 124.0869,
|
| 1503 |
+
"eval_samples_per_second": 2.256,
|
| 1504 |
+
"eval_steps_per_second": 0.564,
|
| 1505 |
+
"step": 6800
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 0.7841117216117216,
|
| 1509 |
+
"grad_norm": 0.73828125,
|
| 1510 |
+
"learning_rate": 8.460797015929873e-06,
|
| 1511 |
+
"loss": 0.4401,
|
| 1512 |
+
"step": 6850
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 0.7898351648351648,
|
| 1516 |
+
"grad_norm": 0.71484375,
|
| 1517 |
+
"learning_rate": 8.04360385499437e-06,
|
| 1518 |
+
"loss": 0.4294,
|
| 1519 |
+
"step": 6900
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 0.7898351648351648,
|
| 1523 |
+
"eval_loss": 1.779537558555603,
|
| 1524 |
+
"eval_runtime": 124.0971,
|
| 1525 |
+
"eval_samples_per_second": 2.256,
|
| 1526 |
+
"eval_steps_per_second": 0.564,
|
| 1527 |
+
"step": 6900
|
| 1528 |
+
},
|
| 1529 |
+
{
|
| 1530 |
+
"epoch": 0.795558608058608,
|
| 1531 |
+
"grad_norm": 0.80859375,
|
| 1532 |
+
"learning_rate": 7.63497809770566e-06,
|
| 1533 |
+
"loss": 0.4327,
|
| 1534 |
+
"step": 6950
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 0.8012820512820513,
|
| 1538 |
+
"grad_norm": 0.76953125,
|
| 1539 |
+
"learning_rate": 7.235126206673651e-06,
|
| 1540 |
+
"loss": 0.4364,
|
| 1541 |
+
"step": 7000
|
| 1542 |
+
},
|
| 1543 |
+
{
|
| 1544 |
+
"epoch": 0.8012820512820513,
|
| 1545 |
+
"eval_loss": 1.7765251398086548,
|
| 1546 |
+
"eval_runtime": 124.0761,
|
| 1547 |
+
"eval_samples_per_second": 2.257,
|
| 1548 |
+
"eval_steps_per_second": 0.564,
|
| 1549 |
+
"step": 7000
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.8070054945054945,
|
| 1553 |
+
"grad_norm": 0.734375,
|
| 1554 |
+
"learning_rate": 6.844250211416903e-06,
|
| 1555 |
+
"loss": 0.4392,
|
| 1556 |
+
"step": 7050
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.8127289377289377,
|
| 1560 |
+
"grad_norm": 0.73828125,
|
| 1561 |
+
"learning_rate": 6.462547606284947e-06,
|
| 1562 |
+
"loss": 0.4414,
|
| 1563 |
+
"step": 7100
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.8127289377289377,
|
| 1567 |
+
"eval_loss": 1.7783187627792358,
|
| 1568 |
+
"eval_runtime": 124.1204,
|
| 1569 |
+
"eval_samples_per_second": 2.256,
|
| 1570 |
+
"eval_steps_per_second": 0.564,
|
| 1571 |
+
"step": 7100
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 0.8184523809523809,
|
| 1575 |
+
"grad_norm": 0.7421875,
|
| 1576 |
+
"learning_rate": 6.09021125067217e-06,
|
| 1577 |
+
"loss": 0.4354,
|
| 1578 |
+
"step": 7150
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 0.8241758241758241,
|
| 1582 |
+
"grad_norm": 0.84765625,
|
| 1583 |
+
"learning_rate": 5.7274292715734315e-06,
|
| 1584 |
+
"loss": 0.4336,
|
| 1585 |
+
"step": 7200
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 0.8241758241758241,
|
| 1589 |
+
"eval_loss": 1.7745906114578247,
|
| 1590 |
+
"eval_runtime": 124.1664,
|
| 1591 |
+
"eval_samples_per_second": 2.255,
|
| 1592 |
+
"eval_steps_per_second": 0.564,
|
| 1593 |
+
"step": 7200
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 0.8298992673992674,
|
| 1597 |
+
"grad_norm": 0.7421875,
|
| 1598 |
+
"learning_rate": 5.374384968530952e-06,
|
| 1599 |
+
"loss": 0.4296,
|
| 1600 |
+
"step": 7250
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 0.8356227106227107,
|
| 1604 |
+
"grad_norm": 0.67578125,
|
| 1605 |
+
"learning_rate": 5.031256721020139e-06,
|
| 1606 |
+
"loss": 0.4324,
|
| 1607 |
+
"step": 7300
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 0.8356227106227107,
|
| 1611 |
+
"eval_loss": 1.772754192352295,
|
| 1612 |
+
"eval_runtime": 124.0892,
|
| 1613 |
+
"eval_samples_per_second": 2.256,
|
| 1614 |
+
"eval_steps_per_second": 0.564,
|
| 1615 |
+
"step": 7300
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 0.8413461538461539,
|
| 1619 |
+
"grad_norm": 0.78125,
|
| 1620 |
+
"learning_rate": 4.698217898321483e-06,
|
| 1621 |
+
"loss": 0.4304,
|
| 1622 |
+
"step": 7350
|
| 1623 |
+
},
|
| 1624 |
+
{
|
| 1625 |
+
"epoch": 0.8470695970695971,
|
| 1626 |
+
"grad_norm": 0.71875,
|
| 1627 |
+
"learning_rate": 4.375436771923766e-06,
|
| 1628 |
+
"loss": 0.4414,
|
| 1629 |
+
"step": 7400
|
| 1630 |
+
},
|
| 1631 |
+
{
|
| 1632 |
+
"epoch": 0.8470695970695971,
|
| 1633 |
+
"eval_loss": 1.7765103578567505,
|
| 1634 |
+
"eval_runtime": 124.0978,
|
| 1635 |
+
"eval_samples_per_second": 2.256,
|
| 1636 |
+
"eval_steps_per_second": 0.564,
|
| 1637 |
+
"step": 7400
|
| 1638 |
+
},
|
| 1639 |
+
{
|
| 1640 |
+
"epoch": 0.8527930402930403,
|
| 1641 |
+
"grad_norm": 0.77734375,
|
| 1642 |
+
"learning_rate": 4.063076430503138e-06,
|
| 1643 |
+
"loss": 0.4258,
|
| 1644 |
+
"step": 7450
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 0.8585164835164835,
|
| 1648 |
+
"grad_norm": 0.76953125,
|
| 1649 |
+
"learning_rate": 3.761294697520751e-06,
|
| 1650 |
+
"loss": 0.4288,
|
| 1651 |
+
"step": 7500
|
| 1652 |
+
},
|
| 1653 |
+
{
|
| 1654 |
+
"epoch": 0.8585164835164835,
|
| 1655 |
+
"eval_loss": 1.7792127132415771,
|
| 1656 |
+
"eval_runtime": 124.1222,
|
| 1657 |
+
"eval_samples_per_second": 2.256,
|
| 1658 |
+
"eval_steps_per_second": 0.564,
|
| 1659 |
+
"step": 7500
|
| 1660 |
+
},
|
| 1661 |
+
{
|
| 1662 |
+
"epoch": 0.8642399267399268,
|
| 1663 |
+
"grad_norm": 0.72265625,
|
| 1664 |
+
"learning_rate": 3.4702440514807366e-06,
|
| 1665 |
+
"loss": 0.4365,
|
| 1666 |
+
"step": 7550
|
| 1667 |
+
},
|
| 1668 |
+
{
|
| 1669 |
+
"epoch": 0.86996336996337,
|
| 1670 |
+
"grad_norm": 0.79296875,
|
| 1671 |
+
"learning_rate": 3.1900715488887873e-06,
|
| 1672 |
+
"loss": 0.4359,
|
| 1673 |
+
"step": 7600
|
| 1674 |
+
},
|
| 1675 |
+
{
|
| 1676 |
+
"epoch": 0.86996336996337,
|
| 1677 |
+
"eval_loss": 1.777583122253418,
|
| 1678 |
+
"eval_runtime": 124.1273,
|
| 1679 |
+
"eval_samples_per_second": 2.256,
|
| 1680 |
+
"eval_steps_per_second": 0.564,
|
| 1681 |
+
"step": 7600
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 0.8756868131868132,
|
| 1685 |
+
"grad_norm": 0.7109375,
|
| 1686 |
+
"learning_rate": 2.9209187499502604e-06,
|
| 1687 |
+
"loss": 0.4388,
|
| 1688 |
+
"step": 7650
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 0.8814102564102564,
|
| 1692 |
+
"grad_norm": 0.6953125,
|
| 1693 |
+
"learning_rate": 2.662921647045355e-06,
|
| 1694 |
+
"loss": 0.4242,
|
| 1695 |
+
"step": 7700
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 0.8814102564102564,
|
| 1699 |
+
"eval_loss": 1.7761567831039429,
|
| 1700 |
+
"eval_runtime": 124.1269,
|
| 1701 |
+
"eval_samples_per_second": 2.256,
|
| 1702 |
+
"eval_steps_per_second": 0.564,
|
| 1703 |
+
"step": 7700
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8871336996336996,
|
| 1707 |
+
"grad_norm": 0.73828125,
|
| 1708 |
+
"learning_rate": 2.4162105960174486e-06,
|
| 1709 |
+
"loss": 0.4352,
|
| 1710 |
+
"step": 7750
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8928571428571429,
|
| 1714 |
+
"grad_norm": 0.77734375,
|
| 1715 |
+
"learning_rate": 2.180910250309423e-06,
|
| 1716 |
+
"loss": 0.4413,
|
| 1717 |
+
"step": 7800
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8928571428571429,
|
| 1721 |
+
"eval_loss": 1.7751096487045288,
|
| 1722 |
+
"eval_runtime": 124.1103,
|
| 1723 |
+
"eval_samples_per_second": 2.256,
|
| 1724 |
+
"eval_steps_per_second": 0.564,
|
| 1725 |
+
"step": 7800
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 0.8985805860805861,
|
| 1729 |
+
"grad_norm": 0.734375,
|
| 1730 |
+
"learning_rate": 1.957139497981131e-06,
|
| 1731 |
+
"loss": 0.4293,
|
| 1732 |
+
"step": 7850
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 0.9043040293040293,
|
| 1736 |
+
"grad_norm": 0.8359375,
|
| 1737 |
+
"learning_rate": 1.74501140163994e-06,
|
| 1738 |
+
"loss": 0.4402,
|
| 1739 |
+
"step": 7900
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 0.9043040293040293,
|
| 1743 |
+
"eval_loss": 1.7753708362579346,
|
| 1744 |
+
"eval_runtime": 124.1127,
|
| 1745 |
+
"eval_samples_per_second": 2.256,
|
| 1746 |
+
"eval_steps_per_second": 0.564,
|
| 1747 |
+
"step": 7900
|
| 1748 |
+
},
|
| 1749 |
+
{
|
| 1750 |
+
"epoch": 0.9100274725274725,
|
| 1751 |
+
"grad_norm": 0.75,
|
| 1752 |
+
"learning_rate": 1.5446331413145887e-06,
|
| 1753 |
+
"loss": 0.4436,
|
| 1754 |
+
"step": 7950
|
| 1755 |
+
},
|
| 1756 |
+
{
|
| 1757 |
+
"epoch": 0.9157509157509157,
|
| 1758 |
+
"grad_norm": 0.703125,
|
| 1759 |
+
"learning_rate": 1.3561059603013265e-06,
|
| 1760 |
+
"loss": 0.4452,
|
| 1761 |
+
"step": 8000
|
| 1762 |
+
},
|
| 1763 |
+
{
|
| 1764 |
+
"epoch": 0.9157509157509157,
|
| 1765 |
+
"eval_loss": 1.7749762535095215,
|
| 1766 |
+
"eval_runtime": 124.0977,
|
| 1767 |
+
"eval_samples_per_second": 2.256,
|
| 1768 |
+
"eval_steps_per_second": 0.564,
|
| 1769 |
+
"step": 8000
|
| 1770 |
+
},
|
| 1771 |
+
{
|
| 1772 |
+
"epoch": 0.9214743589743589,
|
| 1773 |
+
"grad_norm": 0.6640625,
|
| 1774 |
+
"learning_rate": 1.1795251140096358e-06,
|
| 1775 |
+
"loss": 0.4458,
|
| 1776 |
+
"step": 8050
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 0.9271978021978022,
|
| 1780 |
+
"grad_norm": 0.74609375,
|
| 1781 |
+
"learning_rate": 1.014979821833395e-06,
|
| 1782 |
+
"loss": 0.4346,
|
| 1783 |
+
"step": 8100
|
| 1784 |
+
},
|
| 1785 |
+
{
|
| 1786 |
+
"epoch": 0.9271978021978022,
|
| 1787 |
+
"eval_loss": 1.775481104850769,
|
| 1788 |
+
"eval_runtime": 124.1005,
|
| 1789 |
+
"eval_samples_per_second": 2.256,
|
| 1790 |
+
"eval_steps_per_second": 0.564,
|
| 1791 |
+
"step": 8100
|
| 1792 |
+
},
|
| 1793 |
+
{
|
| 1794 |
+
"epoch": 0.9329212454212454,
|
| 1795 |
+
"grad_norm": 0.73046875,
|
| 1796 |
+
"learning_rate": 8.625532220718186e-07,
|
| 1797 |
+
"loss": 0.4361,
|
| 1798 |
+
"step": 8150
|
| 1799 |
+
},
|
| 1800 |
+
{
|
| 1801 |
+
"epoch": 0.9386446886446886,
|
| 1802 |
+
"grad_norm": 0.7578125,
|
| 1803 |
+
"learning_rate": 7.223223299229198e-07,
|
| 1804 |
+
"loss": 0.4396,
|
| 1805 |
+
"step": 8200
|
| 1806 |
+
},
|
| 1807 |
+
{
|
| 1808 |
+
"epoch": 0.9386446886446886,
|
| 1809 |
+
"eval_loss": 1.7751343250274658,
|
| 1810 |
+
"eval_runtime": 124.089,
|
| 1811 |
+
"eval_samples_per_second": 2.256,
|
| 1812 |
+
"eval_steps_per_second": 0.564,
|
| 1813 |
+
"step": 8200
|
| 1814 |
+
},
|
| 1815 |
+
{
|
| 1816 |
+
"epoch": 0.9443681318681318,
|
| 1817 |
+
"grad_norm": 0.765625,
|
| 1818 |
+
"learning_rate": 5.943579985707409e-07,
|
| 1819 |
+
"loss": 0.4337,
|
| 1820 |
+
"step": 8250
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 0.950091575091575,
|
| 1824 |
+
"grad_norm": 0.78125,
|
| 1825 |
+
"learning_rate": 4.787248833860114e-07,
|
| 1826 |
+
"loss": 0.44,
|
| 1827 |
+
"step": 8300
|
| 1828 |
+
},
|
| 1829 |
+
{
|
| 1830 |
+
"epoch": 0.950091575091575,
|
| 1831 |
+
"eval_loss": 1.7752093076705933,
|
| 1832 |
+
"eval_runtime": 124.0951,
|
| 1833 |
+
"eval_samples_per_second": 2.256,
|
| 1834 |
+
"eval_steps_per_second": 0.564,
|
| 1835 |
+
"step": 8300
|
| 1836 |
+
},
|
| 1837 |
+
{
|
| 1838 |
+
"epoch": 0.9558150183150184,
|
| 1839 |
+
"grad_norm": 0.7578125,
|
| 1840 |
+
"learning_rate": 3.7548140925833806e-07,
|
| 1841 |
+
"loss": 0.4362,
|
| 1842 |
+
"step": 8350
|
| 1843 |
+
},
|
| 1844 |
+
{
|
| 1845 |
+
"epoch": 0.9615384615384616,
|
| 1846 |
+
"grad_norm": 0.7578125,
|
| 1847 |
+
"learning_rate": 2.8467974107636017e-07,
|
| 1848 |
+
"loss": 0.4333,
|
| 1849 |
+
"step": 8400
|
| 1850 |
+
},
|
| 1851 |
+
{
|
| 1852 |
+
"epoch": 0.9615384615384616,
|
| 1853 |
+
"eval_loss": 1.7753241062164307,
|
| 1854 |
+
"eval_runtime": 124.0863,
|
| 1855 |
+
"eval_samples_per_second": 2.256,
|
| 1856 |
+
"eval_steps_per_second": 0.564,
|
| 1857 |
+
"step": 8400
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.9672619047619048,
|
| 1861 |
+
"grad_norm": 0.83984375,
|
| 1862 |
+
"learning_rate": 2.063657573708966e-07,
|
| 1863 |
+
"loss": 0.4338,
|
| 1864 |
+
"step": 8450
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.972985347985348,
|
| 1868 |
+
"grad_norm": 0.84375,
|
| 1869 |
+
"learning_rate": 1.4057902713431327e-07,
|
| 1870 |
+
"loss": 0.4348,
|
| 1871 |
+
"step": 8500
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.972985347985348,
|
| 1875 |
+
"eval_loss": 1.7753559350967407,
|
| 1876 |
+
"eval_runtime": 124.0677,
|
| 1877 |
+
"eval_samples_per_second": 2.257,
|
| 1878 |
+
"eval_steps_per_second": 0.564,
|
| 1879 |
+
"step": 8500
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 0.9787087912087912,
|
| 1883 |
+
"grad_norm": 0.734375,
|
| 1884 |
+
"learning_rate": 8.735278982785755e-08,
|
| 1885 |
+
"loss": 0.4306,
|
| 1886 |
+
"step": 8550
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 0.9844322344322345,
|
| 1890 |
+
"grad_norm": 0.74609375,
|
| 1891 |
+
"learning_rate": 4.671393858705908e-08,
|
| 1892 |
+
"loss": 0.4331,
|
| 1893 |
+
"step": 8600
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 0.9844322344322345,
|
| 1897 |
+
"eval_loss": 1.77518630027771,
|
| 1898 |
+
"eval_runtime": 124.0873,
|
| 1899 |
+
"eval_samples_per_second": 2.256,
|
| 1900 |
+
"eval_steps_per_second": 0.564,
|
| 1901 |
+
"step": 8600
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 0.9901556776556777,
|
| 1905 |
+
"grad_norm": 0.78125,
|
| 1906 |
+
"learning_rate": 1.868300663367406e-08,
|
| 1907 |
+
"loss": 0.4329,
|
| 1908 |
+
"step": 8650
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 0.9958791208791209,
|
| 1912 |
+
"grad_norm": 0.76171875,
|
| 1913 |
+
"learning_rate": 3.2741569010674712e-09,
|
| 1914 |
+
"loss": 0.4326,
|
| 1915 |
+
"step": 8700
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 0.9958791208791209,
|
| 1919 |
+
"eval_loss": 1.7753793001174927,
|
| 1920 |
+
"eval_runtime": 124.103,
|
| 1921 |
+
"eval_samples_per_second": 2.256,
|
| 1922 |
+
"eval_steps_per_second": 0.564,
|
| 1923 |
+
"step": 8700
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 1.0,
|
| 1927 |
+
"step": 8736,
|
| 1928 |
+
"total_flos": 1.6067287852253184e+18,
|
| 1929 |
+
"train_loss": 0.5878726873860691,
|
| 1930 |
+
"train_runtime": 61938.6999,
|
| 1931 |
+
"train_samples_per_second": 0.564,
|
| 1932 |
+
"train_steps_per_second": 0.141
|
| 1933 |
+
}
|
| 1934 |
+
],
|
| 1935 |
+
"logging_steps": 50,
|
| 1936 |
+
"max_steps": 8736,
|
| 1937 |
+
"num_input_tokens_seen": 0,
|
| 1938 |
+
"num_train_epochs": 1,
|
| 1939 |
+
"save_steps": 250,
|
| 1940 |
+
"total_flos": 1.6067287852253184e+18,
|
| 1941 |
+
"train_batch_size": 4,
|
| 1942 |
+
"trial_name": null,
|
| 1943 |
+
"trial_params": null
|
| 1944 |
+
}
|