Qin Liu
commited on
Model save
Browse files- .gitattributes +1 -0
- README.md +69 -0
- adapter_config.json +31 -0
- adapter_model.safetensors +3 -0
- all_results.json +8 -0
- runs/Apr26_02-29-49_COE-CS-sv003/events.out.tfevents.1714098668.COE-CS-sv003.474751.0 +3 -0
- special_tokens_map.json +28 -0
- tokenizer.json +3 -0
- tokenizer_config.json +70 -0
- train_results.json +8 -0
- trainer_state.json +2812 -0
- training_args.bin +3 -0
.gitattributes
CHANGED
|
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
| 33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
| 34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
| 35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
| 36 |
+
tokenizer.json filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gemma
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: google/gemma-7b
|
| 9 |
+
model-index:
|
| 10 |
+
- name: zephyr-7b-gemma-sft-10p-2048
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
# zephyr-7b-gemma-sft-10p-2048
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the None dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 1.2143
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 4
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- gradient_accumulation_steps: 2
|
| 47 |
+
- total_train_batch_size: 32
|
| 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 |
+
- num_epochs: 3
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 58 |
+
| 0.8945 | 1.0 | 655 | 1.2307 |
|
| 59 |
+
| 0.8664 | 2.0 | 1311 | 1.1823 |
|
| 60 |
+
| 0.8345 | 3.0 | 1965 | 1.2143 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- PEFT 0.7.1
|
| 66 |
+
- Transformers 4.39.0.dev0
|
| 67 |
+
- Pytorch 2.1.2
|
| 68 |
+
- Datasets 2.14.6
|
| 69 |
+
- Tokenizers 0.15.2
|
adapter_config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "google/gemma-7b",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"fan_in_fan_out": false,
|
| 7 |
+
"inference_mode": true,
|
| 8 |
+
"init_lora_weights": true,
|
| 9 |
+
"layers_pattern": null,
|
| 10 |
+
"layers_to_transform": null,
|
| 11 |
+
"loftq_config": {},
|
| 12 |
+
"lora_alpha": 8,
|
| 13 |
+
"lora_dropout": 0.05,
|
| 14 |
+
"megatron_config": null,
|
| 15 |
+
"megatron_core": "megatron.core",
|
| 16 |
+
"modules_to_save": null,
|
| 17 |
+
"peft_type": "LORA",
|
| 18 |
+
"r": 6,
|
| 19 |
+
"rank_pattern": {},
|
| 20 |
+
"revision": null,
|
| 21 |
+
"target_modules": [
|
| 22 |
+
"down_proj",
|
| 23 |
+
"up_proj",
|
| 24 |
+
"o_proj",
|
| 25 |
+
"k_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
+
"gate_proj",
|
| 28 |
+
"v_proj"
|
| 29 |
+
],
|
| 30 |
+
"task_type": "CAUSAL_LM"
|
| 31 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:215a47df1fb327319becff891d72fc3a777fa9837691287409f412ed54560337
|
| 3 |
+
size 37555048
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.1282114190303036,
|
| 4 |
+
"train_runtime": 21563.1342,
|
| 5 |
+
"train_samples": 20971,
|
| 6 |
+
"train_samples_per_second": 2.918,
|
| 7 |
+
"train_steps_per_second": 0.091
|
| 8 |
+
}
|
runs/Apr26_02-29-49_COE-CS-sv003/events.out.tfevents.1714098668.COE-CS-sv003.474751.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bf6da6fd21b37edfa6c29919928cac1cf10ac678539a5e0c5b02e4d820c5c437
|
| 3 |
+
size 89007
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<|im_start|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<|im_end|>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
}
|
| 17 |
+
],
|
| 18 |
+
"bos_token": "<|im_start|>",
|
| 19 |
+
"eos_token": "<|im_end|>",
|
| 20 |
+
"pad_token": "<|im_end|>",
|
| 21 |
+
"unk_token": {
|
| 22 |
+
"content": "<unk>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false
|
| 27 |
+
}
|
| 28 |
+
}
|
tokenizer.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:299f8e59a5c4a3b3941dbe1159a7079d69c7a8d5ca34322ace1be9140ae76cc0
|
| 3 |
+
size 17477572
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"added_tokens_decoder": {
|
| 5 |
+
"0": {
|
| 6 |
+
"content": "<pad>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false,
|
| 11 |
+
"special": true
|
| 12 |
+
},
|
| 13 |
+
"1": {
|
| 14 |
+
"content": "<eos>",
|
| 15 |
+
"lstrip": false,
|
| 16 |
+
"normalized": false,
|
| 17 |
+
"rstrip": false,
|
| 18 |
+
"single_word": false,
|
| 19 |
+
"special": true
|
| 20 |
+
},
|
| 21 |
+
"2": {
|
| 22 |
+
"content": "<bos>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false,
|
| 27 |
+
"special": true
|
| 28 |
+
},
|
| 29 |
+
"3": {
|
| 30 |
+
"content": "<unk>",
|
| 31 |
+
"lstrip": false,
|
| 32 |
+
"normalized": false,
|
| 33 |
+
"rstrip": false,
|
| 34 |
+
"single_word": false,
|
| 35 |
+
"special": true
|
| 36 |
+
},
|
| 37 |
+
"106": {
|
| 38 |
+
"content": "<|im_start|>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": false,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false,
|
| 43 |
+
"special": true
|
| 44 |
+
},
|
| 45 |
+
"107": {
|
| 46 |
+
"content": "<|im_end|>",
|
| 47 |
+
"lstrip": false,
|
| 48 |
+
"normalized": false,
|
| 49 |
+
"rstrip": false,
|
| 50 |
+
"single_word": false,
|
| 51 |
+
"special": true
|
| 52 |
+
}
|
| 53 |
+
},
|
| 54 |
+
"additional_special_tokens": [
|
| 55 |
+
"<|im_start|>",
|
| 56 |
+
"<|im_end|>"
|
| 57 |
+
],
|
| 58 |
+
"bos_token": "<|im_start|>",
|
| 59 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 60 |
+
"clean_up_tokenization_spaces": false,
|
| 61 |
+
"eos_token": "<|im_end|>",
|
| 62 |
+
"legacy": null,
|
| 63 |
+
"model_max_length": 2048,
|
| 64 |
+
"pad_token": "<|im_end|>",
|
| 65 |
+
"sp_model_kwargs": {},
|
| 66 |
+
"spaces_between_special_tokens": false,
|
| 67 |
+
"tokenizer_class": "GemmaTokenizer",
|
| 68 |
+
"unk_token": "<unk>",
|
| 69 |
+
"use_default_system_prompt": false
|
| 70 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"train_loss": 1.1282114190303036,
|
| 4 |
+
"train_runtime": 21563.1342,
|
| 5 |
+
"train_samples": 20971,
|
| 6 |
+
"train_samples_per_second": 2.918,
|
| 7 |
+
"train_steps_per_second": 0.091
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2812 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.997711670480549,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1965,
|
| 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 |
+
"grad_norm": 74.09103341509594,
|
| 14 |
+
"learning_rate": 1.015228426395939e-06,
|
| 15 |
+
"loss": 17.1746,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.01,
|
| 20 |
+
"grad_norm": 72.3290702180604,
|
| 21 |
+
"learning_rate": 5.076142131979695e-06,
|
| 22 |
+
"loss": 16.9572,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.02,
|
| 27 |
+
"grad_norm": 68.83490251260932,
|
| 28 |
+
"learning_rate": 1.015228426395939e-05,
|
| 29 |
+
"loss": 16.4638,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.02,
|
| 34 |
+
"grad_norm": 66.14177674907667,
|
| 35 |
+
"learning_rate": 1.5228426395939088e-05,
|
| 36 |
+
"loss": 15.6072,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.03,
|
| 41 |
+
"grad_norm": 54.81375243549984,
|
| 42 |
+
"learning_rate": 2.030456852791878e-05,
|
| 43 |
+
"loss": 13.4345,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04,
|
| 48 |
+
"grad_norm": 50.488727866909294,
|
| 49 |
+
"learning_rate": 2.5380710659898476e-05,
|
| 50 |
+
"loss": 10.913,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.05,
|
| 55 |
+
"grad_norm": 31.023628043144985,
|
| 56 |
+
"learning_rate": 3.0456852791878175e-05,
|
| 57 |
+
"loss": 7.8116,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.05,
|
| 62 |
+
"grad_norm": 23.598786792296284,
|
| 63 |
+
"learning_rate": 3.553299492385787e-05,
|
| 64 |
+
"loss": 5.6786,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"grad_norm": 14.877927520936698,
|
| 70 |
+
"learning_rate": 4.060913705583756e-05,
|
| 71 |
+
"loss": 4.5198,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.07,
|
| 76 |
+
"grad_norm": 10.064755961584666,
|
| 77 |
+
"learning_rate": 4.568527918781726e-05,
|
| 78 |
+
"loss": 3.6186,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.08,
|
| 83 |
+
"grad_norm": 9.060836838854962,
|
| 84 |
+
"learning_rate": 5.076142131979695e-05,
|
| 85 |
+
"loss": 2.7201,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.08,
|
| 90 |
+
"grad_norm": 7.929927938462856,
|
| 91 |
+
"learning_rate": 5.583756345177665e-05,
|
| 92 |
+
"loss": 2.0934,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.09,
|
| 97 |
+
"grad_norm": 3.6854754029311034,
|
| 98 |
+
"learning_rate": 6.091370558375635e-05,
|
| 99 |
+
"loss": 1.7053,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1,
|
| 104 |
+
"grad_norm": 3.113155800813489,
|
| 105 |
+
"learning_rate": 6.598984771573604e-05,
|
| 106 |
+
"loss": 1.5882,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.11,
|
| 111 |
+
"grad_norm": 2.516100336604323,
|
| 112 |
+
"learning_rate": 7.106598984771574e-05,
|
| 113 |
+
"loss": 1.4144,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.11,
|
| 118 |
+
"grad_norm": 2.301065817041555,
|
| 119 |
+
"learning_rate": 7.614213197969543e-05,
|
| 120 |
+
"loss": 1.2941,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.12,
|
| 125 |
+
"grad_norm": 1.9632078441101082,
|
| 126 |
+
"learning_rate": 8.121827411167512e-05,
|
| 127 |
+
"loss": 1.3128,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.13,
|
| 132 |
+
"grad_norm": 2.0984053055137286,
|
| 133 |
+
"learning_rate": 8.629441624365483e-05,
|
| 134 |
+
"loss": 1.3046,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.14,
|
| 139 |
+
"grad_norm": 1.931380727499494,
|
| 140 |
+
"learning_rate": 9.137055837563452e-05,
|
| 141 |
+
"loss": 1.3041,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.14,
|
| 146 |
+
"grad_norm": 1.6180374677897205,
|
| 147 |
+
"learning_rate": 9.644670050761421e-05,
|
| 148 |
+
"loss": 1.21,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.15,
|
| 153 |
+
"grad_norm": 2.0950245146438564,
|
| 154 |
+
"learning_rate": 0.0001015228426395939,
|
| 155 |
+
"loss": 1.21,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.16,
|
| 160 |
+
"grad_norm": 2.1189031385931925,
|
| 161 |
+
"learning_rate": 0.00010659898477157362,
|
| 162 |
+
"loss": 1.1054,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.17,
|
| 167 |
+
"grad_norm": 2.015635255767197,
|
| 168 |
+
"learning_rate": 0.0001116751269035533,
|
| 169 |
+
"loss": 1.1536,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.18,
|
| 174 |
+
"grad_norm": 1.3137399170764255,
|
| 175 |
+
"learning_rate": 0.000116751269035533,
|
| 176 |
+
"loss": 1.0306,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.18,
|
| 181 |
+
"grad_norm": 1.4459264638000027,
|
| 182 |
+
"learning_rate": 0.0001218274111675127,
|
| 183 |
+
"loss": 1.0956,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.19,
|
| 188 |
+
"grad_norm": 1.8464804044878194,
|
| 189 |
+
"learning_rate": 0.0001269035532994924,
|
| 190 |
+
"loss": 1.1259,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2,
|
| 195 |
+
"grad_norm": 2.1383334768591142,
|
| 196 |
+
"learning_rate": 0.00013197969543147207,
|
| 197 |
+
"loss": 1.0779,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.21,
|
| 202 |
+
"grad_norm": 2.816865870724176,
|
| 203 |
+
"learning_rate": 0.00013705583756345178,
|
| 204 |
+
"loss": 1.0787,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.21,
|
| 209 |
+
"grad_norm": 1.514643987387779,
|
| 210 |
+
"learning_rate": 0.00014213197969543148,
|
| 211 |
+
"loss": 1.1153,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.22,
|
| 216 |
+
"grad_norm": 1.9840936459404523,
|
| 217 |
+
"learning_rate": 0.00014720812182741116,
|
| 218 |
+
"loss": 1.0968,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.23,
|
| 223 |
+
"grad_norm": 2.1892173343482852,
|
| 224 |
+
"learning_rate": 0.00015228426395939087,
|
| 225 |
+
"loss": 1.0693,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.24,
|
| 230 |
+
"grad_norm": 2.0955352886552725,
|
| 231 |
+
"learning_rate": 0.00015736040609137057,
|
| 232 |
+
"loss": 1.0467,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.24,
|
| 237 |
+
"grad_norm": 1.989968599726472,
|
| 238 |
+
"learning_rate": 0.00016243654822335025,
|
| 239 |
+
"loss": 0.9779,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.25,
|
| 244 |
+
"grad_norm": 1.4084167700372459,
|
| 245 |
+
"learning_rate": 0.00016751269035532995,
|
| 246 |
+
"loss": 0.9782,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.26,
|
| 251 |
+
"grad_norm": 1.850084427217725,
|
| 252 |
+
"learning_rate": 0.00017258883248730966,
|
| 253 |
+
"loss": 0.9894,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.27,
|
| 258 |
+
"grad_norm": 2.035510825701115,
|
| 259 |
+
"learning_rate": 0.00017766497461928934,
|
| 260 |
+
"loss": 1.0363,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.27,
|
| 265 |
+
"grad_norm": 1.833391476127078,
|
| 266 |
+
"learning_rate": 0.00018274111675126904,
|
| 267 |
+
"loss": 1.0404,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.28,
|
| 272 |
+
"grad_norm": 2.166606883573874,
|
| 273 |
+
"learning_rate": 0.00018781725888324875,
|
| 274 |
+
"loss": 1.0367,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.29,
|
| 279 |
+
"grad_norm": 1.803084621710003,
|
| 280 |
+
"learning_rate": 0.00019289340101522843,
|
| 281 |
+
"loss": 1.0255,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.3,
|
| 286 |
+
"grad_norm": 1.624927933213703,
|
| 287 |
+
"learning_rate": 0.00019796954314720813,
|
| 288 |
+
"loss": 0.9999,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.31,
|
| 293 |
+
"grad_norm": 1.6910332351905768,
|
| 294 |
+
"learning_rate": 0.00019999857915498025,
|
| 295 |
+
"loss": 1.0179,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.31,
|
| 300 |
+
"grad_norm": 1.7237344176394134,
|
| 301 |
+
"learning_rate": 0.00019998989635940995,
|
| 302 |
+
"loss": 1.073,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.32,
|
| 307 |
+
"grad_norm": 1.5125995515843347,
|
| 308 |
+
"learning_rate": 0.00019997332081116373,
|
| 309 |
+
"loss": 0.9436,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.33,
|
| 314 |
+
"grad_norm": 1.4259560737656336,
|
| 315 |
+
"learning_rate": 0.00019994885381864085,
|
| 316 |
+
"loss": 1.0299,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.34,
|
| 321 |
+
"grad_norm": 1.745342365482267,
|
| 322 |
+
"learning_rate": 0.00019991649731315554,
|
| 323 |
+
"loss": 1.0236,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.34,
|
| 328 |
+
"grad_norm": 1.347141268066664,
|
| 329 |
+
"learning_rate": 0.0001998762538487849,
|
| 330 |
+
"loss": 1.0469,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.35,
|
| 335 |
+
"grad_norm": 1.6280396542343814,
|
| 336 |
+
"learning_rate": 0.00019982812660216702,
|
| 337 |
+
"loss": 1.0283,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.36,
|
| 342 |
+
"grad_norm": 1.8616024137182106,
|
| 343 |
+
"learning_rate": 0.0001997721193722502,
|
| 344 |
+
"loss": 1.018,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.37,
|
| 349 |
+
"grad_norm": 1.6748163706768688,
|
| 350 |
+
"learning_rate": 0.00019970823657999327,
|
| 351 |
+
"loss": 0.9641,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.37,
|
| 356 |
+
"grad_norm": 1.8217157997237183,
|
| 357 |
+
"learning_rate": 0.00019963648326801653,
|
| 358 |
+
"loss": 0.9978,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.38,
|
| 363 |
+
"grad_norm": 1.7192183617890824,
|
| 364 |
+
"learning_rate": 0.00019955686510020372,
|
| 365 |
+
"loss": 1.0003,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.39,
|
| 370 |
+
"grad_norm": 1.6670101592399749,
|
| 371 |
+
"learning_rate": 0.00019946938836125483,
|
| 372 |
+
"loss": 1.0061,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.4,
|
| 377 |
+
"grad_norm": 2.0401527430040987,
|
| 378 |
+
"learning_rate": 0.00019937405995619023,
|
| 379 |
+
"loss": 1.0114,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.4,
|
| 384 |
+
"grad_norm": 1.5731808739124065,
|
| 385 |
+
"learning_rate": 0.0001992708874098054,
|
| 386 |
+
"loss": 0.9585,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.41,
|
| 391 |
+
"grad_norm": 1.2522516219731639,
|
| 392 |
+
"learning_rate": 0.00019915987886607713,
|
| 393 |
+
"loss": 0.9405,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.42,
|
| 398 |
+
"grad_norm": 1.5809359876057816,
|
| 399 |
+
"learning_rate": 0.0001990410430875205,
|
| 400 |
+
"loss": 0.9561,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.43,
|
| 405 |
+
"grad_norm": 1.4808669010565572,
|
| 406 |
+
"learning_rate": 0.00019891438945449745,
|
| 407 |
+
"loss": 0.9935,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.43,
|
| 412 |
+
"grad_norm": 1.4718483710588592,
|
| 413 |
+
"learning_rate": 0.00019877992796447601,
|
| 414 |
+
"loss": 0.951,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.44,
|
| 419 |
+
"grad_norm": 1.2492130220680642,
|
| 420 |
+
"learning_rate": 0.00019863766923124148,
|
| 421 |
+
"loss": 0.9418,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.45,
|
| 426 |
+
"grad_norm": 1.3890861204364473,
|
| 427 |
+
"learning_rate": 0.00019848762448405834,
|
| 428 |
+
"loss": 0.9585,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.46,
|
| 433 |
+
"grad_norm": 2.1030414314285437,
|
| 434 |
+
"learning_rate": 0.00019832980556678406,
|
| 435 |
+
"loss": 0.9363,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.47,
|
| 440 |
+
"grad_norm": 1.3041154321040325,
|
| 441 |
+
"learning_rate": 0.00019816422493693417,
|
| 442 |
+
"loss": 0.9555,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.47,
|
| 447 |
+
"grad_norm": 1.4561950481471824,
|
| 448 |
+
"learning_rate": 0.0001979908956646988,
|
| 449 |
+
"loss": 1.006,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.48,
|
| 454 |
+
"grad_norm": 1.3447506745274114,
|
| 455 |
+
"learning_rate": 0.00019780983143191102,
|
| 456 |
+
"loss": 0.938,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.49,
|
| 461 |
+
"grad_norm": 1.7691372827404117,
|
| 462 |
+
"learning_rate": 0.00019762104653096708,
|
| 463 |
+
"loss": 0.9217,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.5,
|
| 468 |
+
"grad_norm": 1.6326336607001684,
|
| 469 |
+
"learning_rate": 0.00019742455586369784,
|
| 470 |
+
"loss": 0.9642,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.5,
|
| 475 |
+
"grad_norm": 1.6469761536298397,
|
| 476 |
+
"learning_rate": 0.0001972203749401928,
|
| 477 |
+
"loss": 0.9049,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.51,
|
| 482 |
+
"grad_norm": 1.4156285399996218,
|
| 483 |
+
"learning_rate": 0.0001970085198775757,
|
| 484 |
+
"loss": 0.9832,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.52,
|
| 489 |
+
"grad_norm": 1.2369154811780847,
|
| 490 |
+
"learning_rate": 0.00019678900739873226,
|
| 491 |
+
"loss": 0.9586,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.53,
|
| 496 |
+
"grad_norm": 1.184783040387616,
|
| 497 |
+
"learning_rate": 0.00019656185483099024,
|
| 498 |
+
"loss": 0.9512,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.53,
|
| 503 |
+
"grad_norm": 1.5777666184106938,
|
| 504 |
+
"learning_rate": 0.00019632708010475165,
|
| 505 |
+
"loss": 0.9688,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.54,
|
| 510 |
+
"grad_norm": 1.5444096080267231,
|
| 511 |
+
"learning_rate": 0.00019608470175207733,
|
| 512 |
+
"loss": 0.926,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.55,
|
| 517 |
+
"grad_norm": 1.5314192534663842,
|
| 518 |
+
"learning_rate": 0.00019583473890522432,
|
| 519 |
+
"loss": 0.9758,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.56,
|
| 524 |
+
"grad_norm": 1.7369618751197407,
|
| 525 |
+
"learning_rate": 0.0001955772112951354,
|
| 526 |
+
"loss": 0.8986,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.56,
|
| 531 |
+
"grad_norm": 1.9574089323769333,
|
| 532 |
+
"learning_rate": 0.00019531213924988176,
|
| 533 |
+
"loss": 0.9164,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.57,
|
| 538 |
+
"grad_norm": 1.2296445171044836,
|
| 539 |
+
"learning_rate": 0.00019503954369305847,
|
| 540 |
+
"loss": 0.9108,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.58,
|
| 545 |
+
"grad_norm": 1.4908276926290502,
|
| 546 |
+
"learning_rate": 0.00019475944614213262,
|
| 547 |
+
"loss": 0.9318,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.59,
|
| 552 |
+
"grad_norm": 1.3086171099702917,
|
| 553 |
+
"learning_rate": 0.00019447186870674506,
|
| 554 |
+
"loss": 0.9266,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.59,
|
| 559 |
+
"grad_norm": 1.1788811143001057,
|
| 560 |
+
"learning_rate": 0.00019417683408696503,
|
| 561 |
+
"loss": 0.9388,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.6,
|
| 566 |
+
"grad_norm": 1.1833646675888452,
|
| 567 |
+
"learning_rate": 0.00019387436557149841,
|
| 568 |
+
"loss": 0.982,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.61,
|
| 573 |
+
"grad_norm": 1.4649925294006232,
|
| 574 |
+
"learning_rate": 0.00019356448703584933,
|
| 575 |
+
"loss": 0.9697,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.62,
|
| 580 |
+
"grad_norm": 1.3041908495917867,
|
| 581 |
+
"learning_rate": 0.00019324722294043558,
|
| 582 |
+
"loss": 0.9623,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.63,
|
| 587 |
+
"grad_norm": 1.3330225209891455,
|
| 588 |
+
"learning_rate": 0.00019292259832865785,
|
| 589 |
+
"loss": 0.9365,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.63,
|
| 594 |
+
"grad_norm": 1.2450996327905286,
|
| 595 |
+
"learning_rate": 0.00019259063882492292,
|
| 596 |
+
"loss": 1.0171,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.64,
|
| 601 |
+
"grad_norm": 1.3505588278305898,
|
| 602 |
+
"learning_rate": 0.00019225137063262084,
|
| 603 |
+
"loss": 0.9522,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.65,
|
| 608 |
+
"grad_norm": 1.884935596805578,
|
| 609 |
+
"learning_rate": 0.00019190482053205673,
|
| 610 |
+
"loss": 0.922,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.66,
|
| 615 |
+
"grad_norm": 1.549801647217274,
|
| 616 |
+
"learning_rate": 0.00019155101587833677,
|
| 617 |
+
"loss": 0.9534,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.66,
|
| 622 |
+
"grad_norm": 1.0786713746828498,
|
| 623 |
+
"learning_rate": 0.00019118998459920902,
|
| 624 |
+
"loss": 0.9176,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.67,
|
| 629 |
+
"grad_norm": 1.1098145056921047,
|
| 630 |
+
"learning_rate": 0.0001908217551928587,
|
| 631 |
+
"loss": 0.9351,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.68,
|
| 636 |
+
"grad_norm": 1.478688514055702,
|
| 637 |
+
"learning_rate": 0.00019044635672565895,
|
| 638 |
+
"loss": 0.9469,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.69,
|
| 643 |
+
"grad_norm": 1.015346045441744,
|
| 644 |
+
"learning_rate": 0.00019006381882987627,
|
| 645 |
+
"loss": 0.9703,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.69,
|
| 650 |
+
"grad_norm": 1.1820782296841172,
|
| 651 |
+
"learning_rate": 0.0001896741717013315,
|
| 652 |
+
"loss": 0.9601,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.7,
|
| 657 |
+
"grad_norm": 1.2966754894510633,
|
| 658 |
+
"learning_rate": 0.00018927744609701643,
|
| 659 |
+
"loss": 0.9605,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.71,
|
| 664 |
+
"grad_norm": 1.0403285337969717,
|
| 665 |
+
"learning_rate": 0.00018887367333266582,
|
| 666 |
+
"loss": 0.9418,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.72,
|
| 671 |
+
"grad_norm": 1.1722720307755863,
|
| 672 |
+
"learning_rate": 0.00018846288528028555,
|
| 673 |
+
"loss": 0.9741,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.72,
|
| 678 |
+
"grad_norm": 1.2117042399971847,
|
| 679 |
+
"learning_rate": 0.0001880451143656368,
|
| 680 |
+
"loss": 0.9435,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.73,
|
| 685 |
+
"grad_norm": 1.0509023160172017,
|
| 686 |
+
"learning_rate": 0.00018762039356567655,
|
| 687 |
+
"loss": 0.8706,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.74,
|
| 692 |
+
"grad_norm": 1.2265158396763423,
|
| 693 |
+
"learning_rate": 0.00018718875640595434,
|
| 694 |
+
"loss": 0.9051,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.75,
|
| 699 |
+
"grad_norm": 1.2935461295547515,
|
| 700 |
+
"learning_rate": 0.00018675023695796607,
|
| 701 |
+
"loss": 0.9758,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.76,
|
| 706 |
+
"grad_norm": 1.5267873012858113,
|
| 707 |
+
"learning_rate": 0.00018630486983646467,
|
| 708 |
+
"loss": 0.9244,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.76,
|
| 713 |
+
"grad_norm": 1.2742929372558043,
|
| 714 |
+
"learning_rate": 0.00018585269019672743,
|
| 715 |
+
"loss": 0.928,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.77,
|
| 720 |
+
"grad_norm": 1.2358049387637984,
|
| 721 |
+
"learning_rate": 0.00018539373373178123,
|
| 722 |
+
"loss": 0.966,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.78,
|
| 727 |
+
"grad_norm": 1.0460726575553243,
|
| 728 |
+
"learning_rate": 0.00018492803666958515,
|
| 729 |
+
"loss": 0.9259,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.79,
|
| 734 |
+
"grad_norm": 1.361928704446451,
|
| 735 |
+
"learning_rate": 0.00018445563577017062,
|
| 736 |
+
"loss": 0.9742,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.79,
|
| 741 |
+
"grad_norm": 1.139407531386836,
|
| 742 |
+
"learning_rate": 0.0001839765683227398,
|
| 743 |
+
"loss": 0.9318,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.8,
|
| 748 |
+
"grad_norm": 1.242505633372625,
|
| 749 |
+
"learning_rate": 0.00018349087214272224,
|
| 750 |
+
"loss": 0.905,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.81,
|
| 755 |
+
"grad_norm": 1.2336955558098357,
|
| 756 |
+
"learning_rate": 0.00018299858556878972,
|
| 757 |
+
"loss": 0.9233,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.82,
|
| 762 |
+
"grad_norm": 1.1643115008804164,
|
| 763 |
+
"learning_rate": 0.00018249974745983023,
|
| 764 |
+
"loss": 0.8991,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.82,
|
| 769 |
+
"grad_norm": 1.2435271188335837,
|
| 770 |
+
"learning_rate": 0.00018199439719188034,
|
| 771 |
+
"loss": 0.8918,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.83,
|
| 776 |
+
"grad_norm": 1.3025449616558689,
|
| 777 |
+
"learning_rate": 0.00018148257465501716,
|
| 778 |
+
"loss": 0.9272,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.84,
|
| 783 |
+
"grad_norm": 1.639151201490911,
|
| 784 |
+
"learning_rate": 0.0001809643202502097,
|
| 785 |
+
"loss": 0.9188,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.85,
|
| 790 |
+
"grad_norm": 1.1046258635499404,
|
| 791 |
+
"learning_rate": 0.00018043967488612955,
|
| 792 |
+
"loss": 0.846,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.85,
|
| 797 |
+
"grad_norm": 1.1404910374403605,
|
| 798 |
+
"learning_rate": 0.00017990867997592203,
|
| 799 |
+
"loss": 0.9187,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.86,
|
| 804 |
+
"grad_norm": 1.1924053040242124,
|
| 805 |
+
"learning_rate": 0.00017937137743393694,
|
| 806 |
+
"loss": 0.9046,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.87,
|
| 811 |
+
"grad_norm": 1.0135630618037776,
|
| 812 |
+
"learning_rate": 0.0001788278096724203,
|
| 813 |
+
"loss": 0.9146,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.88,
|
| 818 |
+
"grad_norm": 0.9391119950422172,
|
| 819 |
+
"learning_rate": 0.00017827801959816625,
|
| 820 |
+
"loss": 0.9343,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.88,
|
| 825 |
+
"grad_norm": 1.1918282771197495,
|
| 826 |
+
"learning_rate": 0.0001777220506091303,
|
| 827 |
+
"loss": 0.912,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.89,
|
| 832 |
+
"grad_norm": 1.3729834554449354,
|
| 833 |
+
"learning_rate": 0.00017715994659100388,
|
| 834 |
+
"loss": 0.9561,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.9,
|
| 839 |
+
"grad_norm": 1.4315867303078007,
|
| 840 |
+
"learning_rate": 0.00017659175191374983,
|
| 841 |
+
"loss": 0.9503,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.91,
|
| 846 |
+
"grad_norm": 1.1673639446932986,
|
| 847 |
+
"learning_rate": 0.0001760175114281003,
|
| 848 |
+
"loss": 0.8902,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.92,
|
| 853 |
+
"grad_norm": 1.230158811259546,
|
| 854 |
+
"learning_rate": 0.0001754372704620164,
|
| 855 |
+
"loss": 0.8828,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.92,
|
| 860 |
+
"grad_norm": 1.1728800163063369,
|
| 861 |
+
"learning_rate": 0.00017485107481711012,
|
| 862 |
+
"loss": 0.9118,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.93,
|
| 867 |
+
"grad_norm": 1.0131739031610576,
|
| 868 |
+
"learning_rate": 0.00017425897076502902,
|
| 869 |
+
"loss": 0.9403,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.94,
|
| 874 |
+
"grad_norm": 1.351934812093798,
|
| 875 |
+
"learning_rate": 0.00017366100504380378,
|
| 876 |
+
"loss": 0.8876,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.95,
|
| 881 |
+
"grad_norm": 1.0384035475562403,
|
| 882 |
+
"learning_rate": 0.00017305722485415882,
|
| 883 |
+
"loss": 0.9159,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.95,
|
| 888 |
+
"grad_norm": 1.1416340076156324,
|
| 889 |
+
"learning_rate": 0.0001724476778557866,
|
| 890 |
+
"loss": 0.9358,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.96,
|
| 895 |
+
"grad_norm": 1.3419613876196461,
|
| 896 |
+
"learning_rate": 0.00017183241216358553,
|
| 897 |
+
"loss": 0.9729,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.97,
|
| 902 |
+
"grad_norm": 1.6092824682901703,
|
| 903 |
+
"learning_rate": 0.00017121147634386192,
|
| 904 |
+
"loss": 0.8965,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.98,
|
| 909 |
+
"grad_norm": 0.8533351436444618,
|
| 910 |
+
"learning_rate": 0.00017058491941049649,
|
| 911 |
+
"loss": 0.8917,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.98,
|
| 916 |
+
"grad_norm": 1.1738959779265217,
|
| 917 |
+
"learning_rate": 0.0001699527908210754,
|
| 918 |
+
"loss": 0.9366,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.99,
|
| 923 |
+
"grad_norm": 1.0847553199067752,
|
| 924 |
+
"learning_rate": 0.0001693151404729862,
|
| 925 |
+
"loss": 0.9053,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.0,
|
| 930 |
+
"grad_norm": 1.1087892770915242,
|
| 931 |
+
"learning_rate": 0.00016867201869947938,
|
| 932 |
+
"loss": 0.8945,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.0,
|
| 937 |
+
"eval_loss": 1.2306770086288452,
|
| 938 |
+
"eval_runtime": 248.745,
|
| 939 |
+
"eval_samples_per_second": 9.287,
|
| 940 |
+
"eval_steps_per_second": 0.583,
|
| 941 |
+
"step": 655
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 1.01,
|
| 945 |
+
"grad_norm": 0.98970356590023,
|
| 946 |
+
"learning_rate": 0.00016802347626569501,
|
| 947 |
+
"loss": 0.8931,
|
| 948 |
+
"step": 660
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 1.01,
|
| 952 |
+
"grad_norm": 1.1038579119403917,
|
| 953 |
+
"learning_rate": 0.00016736956436465573,
|
| 954 |
+
"loss": 0.934,
|
| 955 |
+
"step": 665
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 1.02,
|
| 959 |
+
"grad_norm": 1.143467118458221,
|
| 960 |
+
"learning_rate": 0.00016671033461322583,
|
| 961 |
+
"loss": 0.8668,
|
| 962 |
+
"step": 670
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 1.03,
|
| 966 |
+
"grad_norm": 0.8924280436486196,
|
| 967 |
+
"learning_rate": 0.00016604583904803672,
|
| 968 |
+
"loss": 0.8548,
|
| 969 |
+
"step": 675
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 1.04,
|
| 973 |
+
"grad_norm": 1.1115437180292647,
|
| 974 |
+
"learning_rate": 0.0001653761301213795,
|
| 975 |
+
"loss": 0.9385,
|
| 976 |
+
"step": 680
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 1.05,
|
| 980 |
+
"grad_norm": 1.6438519669074836,
|
| 981 |
+
"learning_rate": 0.00016470126069706456,
|
| 982 |
+
"loss": 0.9406,
|
| 983 |
+
"step": 685
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 1.05,
|
| 987 |
+
"grad_norm": 1.0490682410760337,
|
| 988 |
+
"learning_rate": 0.00016402128404624882,
|
| 989 |
+
"loss": 0.9381,
|
| 990 |
+
"step": 690
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 1.06,
|
| 994 |
+
"grad_norm": 1.0451537501943204,
|
| 995 |
+
"learning_rate": 0.0001633362538432306,
|
| 996 |
+
"loss": 0.9113,
|
| 997 |
+
"step": 695
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 1.07,
|
| 1001 |
+
"grad_norm": 0.9747283917774375,
|
| 1002 |
+
"learning_rate": 0.0001626462241612129,
|
| 1003 |
+
"loss": 0.9121,
|
| 1004 |
+
"step": 700
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 1.08,
|
| 1008 |
+
"grad_norm": 1.084078539640254,
|
| 1009 |
+
"learning_rate": 0.00016195124946803525,
|
| 1010 |
+
"loss": 0.8831,
|
| 1011 |
+
"step": 705
|
| 1012 |
+
},
|
| 1013 |
+
{
|
| 1014 |
+
"epoch": 1.08,
|
| 1015 |
+
"grad_norm": 0.9719667527581831,
|
| 1016 |
+
"learning_rate": 0.00016125138462187396,
|
| 1017 |
+
"loss": 0.8436,
|
| 1018 |
+
"step": 710
|
| 1019 |
+
},
|
| 1020 |
+
{
|
| 1021 |
+
"epoch": 1.09,
|
| 1022 |
+
"grad_norm": 0.9935638920294466,
|
| 1023 |
+
"learning_rate": 0.00016054668486691203,
|
| 1024 |
+
"loss": 0.9196,
|
| 1025 |
+
"step": 715
|
| 1026 |
+
},
|
| 1027 |
+
{
|
| 1028 |
+
"epoch": 1.1,
|
| 1029 |
+
"grad_norm": 0.9417823234407828,
|
| 1030 |
+
"learning_rate": 0.00015983720582897853,
|
| 1031 |
+
"loss": 0.8577,
|
| 1032 |
+
"step": 720
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 1.11,
|
| 1036 |
+
"grad_norm": 0.8915998902762848,
|
| 1037 |
+
"learning_rate": 0.0001591230035111576,
|
| 1038 |
+
"loss": 0.853,
|
| 1039 |
+
"step": 725
|
| 1040 |
+
},
|
| 1041 |
+
{
|
| 1042 |
+
"epoch": 1.11,
|
| 1043 |
+
"grad_norm": 1.01108160646832,
|
| 1044 |
+
"learning_rate": 0.00015840413428936767,
|
| 1045 |
+
"loss": 0.9148,
|
| 1046 |
+
"step": 730
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 1.12,
|
| 1050 |
+
"grad_norm": 0.9944405742170609,
|
| 1051 |
+
"learning_rate": 0.00015768065490791183,
|
| 1052 |
+
"loss": 0.9151,
|
| 1053 |
+
"step": 735
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 1.13,
|
| 1057 |
+
"grad_norm": 0.9869449635796341,
|
| 1058 |
+
"learning_rate": 0.0001569526224749983,
|
| 1059 |
+
"loss": 0.9144,
|
| 1060 |
+
"step": 740
|
| 1061 |
+
},
|
| 1062 |
+
{
|
| 1063 |
+
"epoch": 1.14,
|
| 1064 |
+
"grad_norm": 1.5578640706977607,
|
| 1065 |
+
"learning_rate": 0.00015622009445823273,
|
| 1066 |
+
"loss": 0.9241,
|
| 1067 |
+
"step": 745
|
| 1068 |
+
},
|
| 1069 |
+
{
|
| 1070 |
+
"epoch": 1.14,
|
| 1071 |
+
"grad_norm": 1.0818607624486927,
|
| 1072 |
+
"learning_rate": 0.00015548312868008196,
|
| 1073 |
+
"loss": 0.9123,
|
| 1074 |
+
"step": 750
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 1.15,
|
| 1078 |
+
"grad_norm": 0.9747826539453109,
|
| 1079 |
+
"learning_rate": 0.0001547417833133097,
|
| 1080 |
+
"loss": 0.8666,
|
| 1081 |
+
"step": 755
|
| 1082 |
+
},
|
| 1083 |
+
{
|
| 1084 |
+
"epoch": 1.16,
|
| 1085 |
+
"grad_norm": 0.9965768127635756,
|
| 1086 |
+
"learning_rate": 0.00015399611687638473,
|
| 1087 |
+
"loss": 0.8808,
|
| 1088 |
+
"step": 760
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 1.17,
|
| 1092 |
+
"grad_norm": 1.1911947572353123,
|
| 1093 |
+
"learning_rate": 0.00015324618822886165,
|
| 1094 |
+
"loss": 0.8824,
|
| 1095 |
+
"step": 765
|
| 1096 |
+
},
|
| 1097 |
+
{
|
| 1098 |
+
"epoch": 1.17,
|
| 1099 |
+
"grad_norm": 1.1537563580786732,
|
| 1100 |
+
"learning_rate": 0.00015249205656673483,
|
| 1101 |
+
"loss": 0.863,
|
| 1102 |
+
"step": 770
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 1.18,
|
| 1106 |
+
"grad_norm": 1.0518456260292857,
|
| 1107 |
+
"learning_rate": 0.00015173378141776568,
|
| 1108 |
+
"loss": 0.8812,
|
| 1109 |
+
"step": 775
|
| 1110 |
+
},
|
| 1111 |
+
{
|
| 1112 |
+
"epoch": 1.19,
|
| 1113 |
+
"grad_norm": 1.387022437551768,
|
| 1114 |
+
"learning_rate": 0.00015097142263678382,
|
| 1115 |
+
"loss": 0.8835,
|
| 1116 |
+
"step": 780
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 1.2,
|
| 1120 |
+
"grad_norm": 1.0408593407980289,
|
| 1121 |
+
"learning_rate": 0.0001502050404009624,
|
| 1122 |
+
"loss": 0.94,
|
| 1123 |
+
"step": 785
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 1.21,
|
| 1127 |
+
"grad_norm": 0.9105554748085787,
|
| 1128 |
+
"learning_rate": 0.00014943469520506807,
|
| 1129 |
+
"loss": 0.9024,
|
| 1130 |
+
"step": 790
|
| 1131 |
+
},
|
| 1132 |
+
{
|
| 1133 |
+
"epoch": 1.21,
|
| 1134 |
+
"grad_norm": 1.0445125806761943,
|
| 1135 |
+
"learning_rate": 0.00014866044785668563,
|
| 1136 |
+
"loss": 0.8973,
|
| 1137 |
+
"step": 795
|
| 1138 |
+
},
|
| 1139 |
+
{
|
| 1140 |
+
"epoch": 1.22,
|
| 1141 |
+
"grad_norm": 1.0453562314924396,
|
| 1142 |
+
"learning_rate": 0.00014788235947141832,
|
| 1143 |
+
"loss": 0.9402,
|
| 1144 |
+
"step": 800
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 1.23,
|
| 1148 |
+
"grad_norm": 0.9691509611927153,
|
| 1149 |
+
"learning_rate": 0.00014710049146806348,
|
| 1150 |
+
"loss": 0.8924,
|
| 1151 |
+
"step": 805
|
| 1152 |
+
},
|
| 1153 |
+
{
|
| 1154 |
+
"epoch": 1.24,
|
| 1155 |
+
"grad_norm": 1.1002288592485436,
|
| 1156 |
+
"learning_rate": 0.00014631490556376458,
|
| 1157 |
+
"loss": 0.8535,
|
| 1158 |
+
"step": 810
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 1.24,
|
| 1162 |
+
"grad_norm": 1.0588060898797662,
|
| 1163 |
+
"learning_rate": 0.00014552566376913943,
|
| 1164 |
+
"loss": 0.9215,
|
| 1165 |
+
"step": 815
|
| 1166 |
+
},
|
| 1167 |
+
{
|
| 1168 |
+
"epoch": 1.25,
|
| 1169 |
+
"grad_norm": 1.1064064596494758,
|
| 1170 |
+
"learning_rate": 0.0001447328283833854,
|
| 1171 |
+
"loss": 0.8822,
|
| 1172 |
+
"step": 820
|
| 1173 |
+
},
|
| 1174 |
+
{
|
| 1175 |
+
"epoch": 1.26,
|
| 1176 |
+
"grad_norm": 0.9443323504920631,
|
| 1177 |
+
"learning_rate": 0.00014393646198936167,
|
| 1178 |
+
"loss": 0.8519,
|
| 1179 |
+
"step": 825
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 1.27,
|
| 1183 |
+
"grad_norm": 1.0288834578773713,
|
| 1184 |
+
"learning_rate": 0.00014313662744864955,
|
| 1185 |
+
"loss": 0.8127,
|
| 1186 |
+
"step": 830
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 1.27,
|
| 1190 |
+
"grad_norm": 0.9217642107444823,
|
| 1191 |
+
"learning_rate": 0.00014233338789659006,
|
| 1192 |
+
"loss": 0.9436,
|
| 1193 |
+
"step": 835
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 1.28,
|
| 1197 |
+
"grad_norm": 1.1636504558701406,
|
| 1198 |
+
"learning_rate": 0.00014152680673730058,
|
| 1199 |
+
"loss": 0.8811,
|
| 1200 |
+
"step": 840
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 1.29,
|
| 1204 |
+
"grad_norm": 1.0596219737156714,
|
| 1205 |
+
"learning_rate": 0.00014071694763866986,
|
| 1206 |
+
"loss": 0.9467,
|
| 1207 |
+
"step": 845
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 1.3,
|
| 1211 |
+
"grad_norm": 1.2066959377309563,
|
| 1212 |
+
"learning_rate": 0.00013990387452733255,
|
| 1213 |
+
"loss": 0.8134,
|
| 1214 |
+
"step": 850
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 1.3,
|
| 1218 |
+
"grad_norm": 1.0002794262680956,
|
| 1219 |
+
"learning_rate": 0.00013908765158362297,
|
| 1220 |
+
"loss": 0.8525,
|
| 1221 |
+
"step": 855
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 1.31,
|
| 1225 |
+
"grad_norm": 0.9370583121034902,
|
| 1226 |
+
"learning_rate": 0.000138268343236509,
|
| 1227 |
+
"loss": 0.9415,
|
| 1228 |
+
"step": 860
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 1.32,
|
| 1232 |
+
"grad_norm": 1.068438209489481,
|
| 1233 |
+
"learning_rate": 0.00013744601415850637,
|
| 1234 |
+
"loss": 0.8345,
|
| 1235 |
+
"step": 865
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 1.33,
|
| 1239 |
+
"grad_norm": 1.0013512513559877,
|
| 1240 |
+
"learning_rate": 0.00013662072926057383,
|
| 1241 |
+
"loss": 0.9228,
|
| 1242 |
+
"step": 870
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 1.33,
|
| 1246 |
+
"grad_norm": 1.1009082230050427,
|
| 1247 |
+
"learning_rate": 0.00013579255368698914,
|
| 1248 |
+
"loss": 0.8734,
|
| 1249 |
+
"step": 875
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 1.34,
|
| 1253 |
+
"grad_norm": 0.9267504277577387,
|
| 1254 |
+
"learning_rate": 0.00013496155281020707,
|
| 1255 |
+
"loss": 0.883,
|
| 1256 |
+
"step": 880
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 1.35,
|
| 1260 |
+
"grad_norm": 0.9055721244854339,
|
| 1261 |
+
"learning_rate": 0.00013412779222569908,
|
| 1262 |
+
"loss": 0.909,
|
| 1263 |
+
"step": 885
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 1.36,
|
| 1267 |
+
"grad_norm": 1.0647099863896714,
|
| 1268 |
+
"learning_rate": 0.00013329133774677552,
|
| 1269 |
+
"loss": 0.8752,
|
| 1270 |
+
"step": 890
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.37,
|
| 1274 |
+
"grad_norm": 1.0280795070472157,
|
| 1275 |
+
"learning_rate": 0.00013245225539939073,
|
| 1276 |
+
"loss": 0.846,
|
| 1277 |
+
"step": 895
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.37,
|
| 1281 |
+
"grad_norm": 0.9768719689574438,
|
| 1282 |
+
"learning_rate": 0.0001316106114169311,
|
| 1283 |
+
"loss": 0.8497,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.38,
|
| 1288 |
+
"grad_norm": 1.1309395586064876,
|
| 1289 |
+
"learning_rate": 0.00013076647223498702,
|
| 1290 |
+
"loss": 0.9077,
|
| 1291 |
+
"step": 905
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.39,
|
| 1295 |
+
"grad_norm": 1.2045776660696499,
|
| 1296 |
+
"learning_rate": 0.0001299199044861086,
|
| 1297 |
+
"loss": 0.8688,
|
| 1298 |
+
"step": 910
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.4,
|
| 1302 |
+
"grad_norm": 0.9841428220188774,
|
| 1303 |
+
"learning_rate": 0.0001290709749945463,
|
| 1304 |
+
"loss": 0.8629,
|
| 1305 |
+
"step": 915
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.4,
|
| 1309 |
+
"grad_norm": 1.325435278640866,
|
| 1310 |
+
"learning_rate": 0.0001282197507709757,
|
| 1311 |
+
"loss": 0.8845,
|
| 1312 |
+
"step": 920
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.41,
|
| 1316 |
+
"grad_norm": 1.001414834995417,
|
| 1317 |
+
"learning_rate": 0.0001273662990072083,
|
| 1318 |
+
"loss": 0.8362,
|
| 1319 |
+
"step": 925
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.42,
|
| 1323 |
+
"grad_norm": 0.9850158774989259,
|
| 1324 |
+
"learning_rate": 0.00012651068707088766,
|
| 1325 |
+
"loss": 0.9111,
|
| 1326 |
+
"step": 930
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.43,
|
| 1330 |
+
"grad_norm": 1.125415537788732,
|
| 1331 |
+
"learning_rate": 0.00012565298250017155,
|
| 1332 |
+
"loss": 0.8847,
|
| 1333 |
+
"step": 935
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.43,
|
| 1337 |
+
"grad_norm": 0.9312415927403098,
|
| 1338 |
+
"learning_rate": 0.00012479325299840095,
|
| 1339 |
+
"loss": 0.8961,
|
| 1340 |
+
"step": 940
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.44,
|
| 1344 |
+
"grad_norm": 1.0186566605726637,
|
| 1345 |
+
"learning_rate": 0.0001239315664287558,
|
| 1346 |
+
"loss": 0.9071,
|
| 1347 |
+
"step": 945
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.45,
|
| 1351 |
+
"grad_norm": 0.9587942024148747,
|
| 1352 |
+
"learning_rate": 0.00012306799080889807,
|
| 1353 |
+
"loss": 0.8553,
|
| 1354 |
+
"step": 950
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.46,
|
| 1358 |
+
"grad_norm": 1.0677123133883124,
|
| 1359 |
+
"learning_rate": 0.000122202594305603,
|
| 1360 |
+
"loss": 0.8459,
|
| 1361 |
+
"step": 955
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.46,
|
| 1365 |
+
"grad_norm": 0.8807263375032743,
|
| 1366 |
+
"learning_rate": 0.00012133544522937805,
|
| 1367 |
+
"loss": 0.8544,
|
| 1368 |
+
"step": 960
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.47,
|
| 1372 |
+
"grad_norm": 1.0893135979400062,
|
| 1373 |
+
"learning_rate": 0.00012046661202907101,
|
| 1374 |
+
"loss": 0.8967,
|
| 1375 |
+
"step": 965
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.48,
|
| 1379 |
+
"grad_norm": 1.2654715562850305,
|
| 1380 |
+
"learning_rate": 0.00011959616328646678,
|
| 1381 |
+
"loss": 0.8554,
|
| 1382 |
+
"step": 970
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.49,
|
| 1386 |
+
"grad_norm": 0.8250130976731411,
|
| 1387 |
+
"learning_rate": 0.00011872416771087395,
|
| 1388 |
+
"loss": 0.9238,
|
| 1389 |
+
"step": 975
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.5,
|
| 1393 |
+
"grad_norm": 0.978527504635651,
|
| 1394 |
+
"learning_rate": 0.00011785069413370111,
|
| 1395 |
+
"loss": 0.8132,
|
| 1396 |
+
"step": 980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.5,
|
| 1400 |
+
"grad_norm": 0.9818769307982783,
|
| 1401 |
+
"learning_rate": 0.00011697581150302362,
|
| 1402 |
+
"loss": 0.8541,
|
| 1403 |
+
"step": 985
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.51,
|
| 1407 |
+
"grad_norm": 0.951052968962308,
|
| 1408 |
+
"learning_rate": 0.00011609958887814129,
|
| 1409 |
+
"loss": 0.8814,
|
| 1410 |
+
"step": 990
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.52,
|
| 1414 |
+
"grad_norm": 0.9722385611491896,
|
| 1415 |
+
"learning_rate": 0.00011522209542412696,
|
| 1416 |
+
"loss": 0.8814,
|
| 1417 |
+
"step": 995
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.53,
|
| 1421 |
+
"grad_norm": 1.1517052356748192,
|
| 1422 |
+
"learning_rate": 0.00011434340040636704,
|
| 1423 |
+
"loss": 0.8906,
|
| 1424 |
+
"step": 1000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.53,
|
| 1428 |
+
"grad_norm": 0.9501728523383044,
|
| 1429 |
+
"learning_rate": 0.00011346357318509395,
|
| 1430 |
+
"loss": 0.8791,
|
| 1431 |
+
"step": 1005
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.54,
|
| 1435 |
+
"grad_norm": 0.9615870877113771,
|
| 1436 |
+
"learning_rate": 0.00011258268320991119,
|
| 1437 |
+
"loss": 0.8547,
|
| 1438 |
+
"step": 1010
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.55,
|
| 1442 |
+
"grad_norm": 1.1528161628232605,
|
| 1443 |
+
"learning_rate": 0.0001117008000143112,
|
| 1444 |
+
"loss": 0.8321,
|
| 1445 |
+
"step": 1015
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.56,
|
| 1449 |
+
"grad_norm": 1.0959125037961595,
|
| 1450 |
+
"learning_rate": 0.00011081799321018695,
|
| 1451 |
+
"loss": 0.8782,
|
| 1452 |
+
"step": 1020
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.56,
|
| 1456 |
+
"grad_norm": 1.1574849427468559,
|
| 1457 |
+
"learning_rate": 0.00010993433248233671,
|
| 1458 |
+
"loss": 0.8611,
|
| 1459 |
+
"step": 1025
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.57,
|
| 1463 |
+
"grad_norm": 0.9005752589116309,
|
| 1464 |
+
"learning_rate": 0.0001090498875829638,
|
| 1465 |
+
"loss": 0.9169,
|
| 1466 |
+
"step": 1030
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.58,
|
| 1470 |
+
"grad_norm": 0.9992289604264046,
|
| 1471 |
+
"learning_rate": 0.00010816472832617037,
|
| 1472 |
+
"loss": 0.8925,
|
| 1473 |
+
"step": 1035
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.59,
|
| 1477 |
+
"grad_norm": 1.1498675103760203,
|
| 1478 |
+
"learning_rate": 0.00010727892458244686,
|
| 1479 |
+
"loss": 0.8942,
|
| 1480 |
+
"step": 1040
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.59,
|
| 1484 |
+
"grad_norm": 0.9081946761117776,
|
| 1485 |
+
"learning_rate": 0.00010639254627315657,
|
| 1486 |
+
"loss": 0.8955,
|
| 1487 |
+
"step": 1045
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.6,
|
| 1491 |
+
"grad_norm": 1.0275540701494208,
|
| 1492 |
+
"learning_rate": 0.00010550566336501637,
|
| 1493 |
+
"loss": 0.8442,
|
| 1494 |
+
"step": 1050
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.61,
|
| 1498 |
+
"grad_norm": 0.9434351529699477,
|
| 1499 |
+
"learning_rate": 0.00010461834586457398,
|
| 1500 |
+
"loss": 0.9419,
|
| 1501 |
+
"step": 1055
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.62,
|
| 1505 |
+
"grad_norm": 0.8829537272599468,
|
| 1506 |
+
"learning_rate": 0.0001037306638126818,
|
| 1507 |
+
"loss": 0.9257,
|
| 1508 |
+
"step": 1060
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.62,
|
| 1512 |
+
"grad_norm": 1.1223033845867914,
|
| 1513 |
+
"learning_rate": 0.00010284268727896833,
|
| 1514 |
+
"loss": 0.9161,
|
| 1515 |
+
"step": 1065
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.63,
|
| 1519 |
+
"grad_norm": 1.2421465538773482,
|
| 1520 |
+
"learning_rate": 0.00010195448635630708,
|
| 1521 |
+
"loss": 0.8701,
|
| 1522 |
+
"step": 1070
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.64,
|
| 1526 |
+
"grad_norm": 1.1218568027877072,
|
| 1527 |
+
"learning_rate": 0.00010106613115528398,
|
| 1528 |
+
"loss": 0.9046,
|
| 1529 |
+
"step": 1075
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.65,
|
| 1533 |
+
"grad_norm": 1.0337432513754021,
|
| 1534 |
+
"learning_rate": 0.00010017769179866279,
|
| 1535 |
+
"loss": 0.8609,
|
| 1536 |
+
"step": 1080
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.66,
|
| 1540 |
+
"grad_norm": 0.8566019611964384,
|
| 1541 |
+
"learning_rate": 9.928923841585024e-05,
|
| 1542 |
+
"loss": 0.8153,
|
| 1543 |
+
"step": 1085
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.66,
|
| 1547 |
+
"grad_norm": 0.9670758845502253,
|
| 1548 |
+
"learning_rate": 9.840084113736024e-05,
|
| 1549 |
+
"loss": 0.8968,
|
| 1550 |
+
"step": 1090
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.67,
|
| 1554 |
+
"grad_norm": 0.8720532251402997,
|
| 1555 |
+
"learning_rate": 9.751257008927803e-05,
|
| 1556 |
+
"loss": 0.9142,
|
| 1557 |
+
"step": 1095
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.68,
|
| 1561 |
+
"grad_norm": 1.005265233752645,
|
| 1562 |
+
"learning_rate": 9.66244953877248e-05,
|
| 1563 |
+
"loss": 0.8797,
|
| 1564 |
+
"step": 1100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.69,
|
| 1568 |
+
"grad_norm": 1.0788499372739098,
|
| 1569 |
+
"learning_rate": 9.573668713332304e-05,
|
| 1570 |
+
"loss": 0.8721,
|
| 1571 |
+
"step": 1105
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.69,
|
| 1575 |
+
"grad_norm": 0.959383830376907,
|
| 1576 |
+
"learning_rate": 9.484921540566307e-05,
|
| 1577 |
+
"loss": 0.9281,
|
| 1578 |
+
"step": 1110
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.7,
|
| 1582 |
+
"grad_norm": 0.9960816420866151,
|
| 1583 |
+
"learning_rate": 9.396215025777139e-05,
|
| 1584 |
+
"loss": 0.8946,
|
| 1585 |
+
"step": 1115
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.71,
|
| 1589 |
+
"grad_norm": 0.9183171097472164,
|
| 1590 |
+
"learning_rate": 9.307556171058085e-05,
|
| 1591 |
+
"loss": 0.8727,
|
| 1592 |
+
"step": 1120
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.72,
|
| 1596 |
+
"grad_norm": 0.9320733649944887,
|
| 1597 |
+
"learning_rate": 9.218951974740354e-05,
|
| 1598 |
+
"loss": 0.9172,
|
| 1599 |
+
"step": 1125
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.72,
|
| 1603 |
+
"grad_norm": 0.8820875642651754,
|
| 1604 |
+
"learning_rate": 9.130409430840678e-05,
|
| 1605 |
+
"loss": 0.903,
|
| 1606 |
+
"step": 1130
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.73,
|
| 1610 |
+
"grad_norm": 0.9886857168118836,
|
| 1611 |
+
"learning_rate": 9.041935528509216e-05,
|
| 1612 |
+
"loss": 0.919,
|
| 1613 |
+
"step": 1135
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.74,
|
| 1617 |
+
"grad_norm": 1.0243023631867323,
|
| 1618 |
+
"learning_rate": 8.953537251477874e-05,
|
| 1619 |
+
"loss": 0.8972,
|
| 1620 |
+
"step": 1140
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.75,
|
| 1624 |
+
"grad_norm": 1.0083372332679066,
|
| 1625 |
+
"learning_rate": 8.865221577509033e-05,
|
| 1626 |
+
"loss": 0.8922,
|
| 1627 |
+
"step": 1145
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.75,
|
| 1631 |
+
"grad_norm": 1.0260814204564148,
|
| 1632 |
+
"learning_rate": 8.776995477844768e-05,
|
| 1633 |
+
"loss": 0.8516,
|
| 1634 |
+
"step": 1150
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.76,
|
| 1638 |
+
"grad_norm": 0.8142375395224646,
|
| 1639 |
+
"learning_rate": 8.688865916656547e-05,
|
| 1640 |
+
"loss": 0.8484,
|
| 1641 |
+
"step": 1155
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.77,
|
| 1645 |
+
"grad_norm": 1.0780009890850535,
|
| 1646 |
+
"learning_rate": 8.600839850495546e-05,
|
| 1647 |
+
"loss": 0.8693,
|
| 1648 |
+
"step": 1160
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.78,
|
| 1652 |
+
"grad_norm": 1.34710148555113,
|
| 1653 |
+
"learning_rate": 8.512924227743481e-05,
|
| 1654 |
+
"loss": 0.8465,
|
| 1655 |
+
"step": 1165
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.78,
|
| 1659 |
+
"grad_norm": 0.8023152238316216,
|
| 1660 |
+
"learning_rate": 8.425125988064197e-05,
|
| 1661 |
+
"loss": 0.8813,
|
| 1662 |
+
"step": 1170
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.79,
|
| 1666 |
+
"grad_norm": 1.0380654418842252,
|
| 1667 |
+
"learning_rate": 8.337452061855814e-05,
|
| 1668 |
+
"loss": 0.9174,
|
| 1669 |
+
"step": 1175
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.8,
|
| 1673 |
+
"grad_norm": 1.0726246573015967,
|
| 1674 |
+
"learning_rate": 8.249909369703728e-05,
|
| 1675 |
+
"loss": 0.8662,
|
| 1676 |
+
"step": 1180
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.81,
|
| 1680 |
+
"grad_norm": 1.0278553172179552,
|
| 1681 |
+
"learning_rate": 8.162504821834295e-05,
|
| 1682 |
+
"loss": 0.8908,
|
| 1683 |
+
"step": 1185
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.82,
|
| 1687 |
+
"grad_norm": 1.3072760509850052,
|
| 1688 |
+
"learning_rate": 8.075245317569395e-05,
|
| 1689 |
+
"loss": 0.8671,
|
| 1690 |
+
"step": 1190
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.82,
|
| 1694 |
+
"grad_norm": 0.9627970898558489,
|
| 1695 |
+
"learning_rate": 7.988137744781805e-05,
|
| 1696 |
+
"loss": 0.8796,
|
| 1697 |
+
"step": 1195
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.83,
|
| 1701 |
+
"grad_norm": 0.9715464599367558,
|
| 1702 |
+
"learning_rate": 7.901188979351526e-05,
|
| 1703 |
+
"loss": 0.8625,
|
| 1704 |
+
"step": 1200
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.84,
|
| 1708 |
+
"grad_norm": 1.0848501852802424,
|
| 1709 |
+
"learning_rate": 7.814405884623011e-05,
|
| 1710 |
+
"loss": 0.855,
|
| 1711 |
+
"step": 1205
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.85,
|
| 1715 |
+
"grad_norm": 0.9626487674718053,
|
| 1716 |
+
"learning_rate": 7.727795310863425e-05,
|
| 1717 |
+
"loss": 0.9057,
|
| 1718 |
+
"step": 1210
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.85,
|
| 1722 |
+
"grad_norm": 0.8586177531466549,
|
| 1723 |
+
"learning_rate": 7.64136409472189e-05,
|
| 1724 |
+
"loss": 0.7599,
|
| 1725 |
+
"step": 1215
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.86,
|
| 1729 |
+
"grad_norm": 0.9382983703101969,
|
| 1730 |
+
"learning_rate": 7.55511905868985e-05,
|
| 1731 |
+
"loss": 0.8729,
|
| 1732 |
+
"step": 1220
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.87,
|
| 1736 |
+
"grad_norm": 1.0198546805846869,
|
| 1737 |
+
"learning_rate": 7.469067010562539e-05,
|
| 1738 |
+
"loss": 0.9047,
|
| 1739 |
+
"step": 1225
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.88,
|
| 1743 |
+
"grad_norm": 1.027987716065299,
|
| 1744 |
+
"learning_rate": 7.383214742901573e-05,
|
| 1745 |
+
"loss": 0.8926,
|
| 1746 |
+
"step": 1230
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.88,
|
| 1750 |
+
"grad_norm": 0.9910955347499838,
|
| 1751 |
+
"learning_rate": 7.297569032498817e-05,
|
| 1752 |
+
"loss": 0.9097,
|
| 1753 |
+
"step": 1235
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.89,
|
| 1757 |
+
"grad_norm": 1.108145730506975,
|
| 1758 |
+
"learning_rate": 7.212136639841418e-05,
|
| 1759 |
+
"loss": 0.8127,
|
| 1760 |
+
"step": 1240
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.9,
|
| 1764 |
+
"grad_norm": 0.78955284168565,
|
| 1765 |
+
"learning_rate": 7.126924308578196e-05,
|
| 1766 |
+
"loss": 0.9052,
|
| 1767 |
+
"step": 1245
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.91,
|
| 1771 |
+
"grad_norm": 0.9172876708503183,
|
| 1772 |
+
"learning_rate": 7.041938764987297e-05,
|
| 1773 |
+
"loss": 0.8864,
|
| 1774 |
+
"step": 1250
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.91,
|
| 1778 |
+
"grad_norm": 0.9468127168909404,
|
| 1779 |
+
"learning_rate": 6.957186717445283e-05,
|
| 1780 |
+
"loss": 0.8877,
|
| 1781 |
+
"step": 1255
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.92,
|
| 1785 |
+
"grad_norm": 0.9264382383778683,
|
| 1786 |
+
"learning_rate": 6.872674855897581e-05,
|
| 1787 |
+
"loss": 0.8407,
|
| 1788 |
+
"step": 1260
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.93,
|
| 1792 |
+
"grad_norm": 0.9055699868252433,
|
| 1793 |
+
"learning_rate": 6.788409851330418e-05,
|
| 1794 |
+
"loss": 0.9067,
|
| 1795 |
+
"step": 1265
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.94,
|
| 1799 |
+
"grad_norm": 0.9776429485158797,
|
| 1800 |
+
"learning_rate": 6.704398355244237e-05,
|
| 1801 |
+
"loss": 0.8893,
|
| 1802 |
+
"step": 1270
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.95,
|
| 1806 |
+
"grad_norm": 1.1033183037079404,
|
| 1807 |
+
"learning_rate": 6.620646999128668e-05,
|
| 1808 |
+
"loss": 0.919,
|
| 1809 |
+
"step": 1275
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.95,
|
| 1813 |
+
"grad_norm": 0.8873640249636788,
|
| 1814 |
+
"learning_rate": 6.537162393939063e-05,
|
| 1815 |
+
"loss": 0.8226,
|
| 1816 |
+
"step": 1280
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.96,
|
| 1820 |
+
"grad_norm": 1.0005367354469088,
|
| 1821 |
+
"learning_rate": 6.453951129574644e-05,
|
| 1822 |
+
"loss": 0.8982,
|
| 1823 |
+
"step": 1285
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.97,
|
| 1827 |
+
"grad_norm": 0.9756199772914784,
|
| 1828 |
+
"learning_rate": 6.371019774358351e-05,
|
| 1829 |
+
"loss": 0.898,
|
| 1830 |
+
"step": 1290
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.98,
|
| 1834 |
+
"grad_norm": 0.8573420891606421,
|
| 1835 |
+
"learning_rate": 6.288374874518348e-05,
|
| 1836 |
+
"loss": 0.8824,
|
| 1837 |
+
"step": 1295
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.98,
|
| 1841 |
+
"grad_norm": 0.8314787002741244,
|
| 1842 |
+
"learning_rate": 6.206022953671303e-05,
|
| 1843 |
+
"loss": 0.8533,
|
| 1844 |
+
"step": 1300
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.99,
|
| 1848 |
+
"grad_norm": 0.999256628643339,
|
| 1849 |
+
"learning_rate": 6.123970512307434e-05,
|
| 1850 |
+
"loss": 0.8766,
|
| 1851 |
+
"step": 1305
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 2.0,
|
| 1855 |
+
"grad_norm": 1.020370445146433,
|
| 1856 |
+
"learning_rate": 6.042224027277403e-05,
|
| 1857 |
+
"loss": 0.8664,
|
| 1858 |
+
"step": 1310
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 2.0,
|
| 1862 |
+
"eval_loss": 1.1823043823242188,
|
| 1863 |
+
"eval_runtime": 248.6275,
|
| 1864 |
+
"eval_samples_per_second": 9.291,
|
| 1865 |
+
"eval_steps_per_second": 0.583,
|
| 1866 |
+
"step": 1311
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 2.01,
|
| 1870 |
+
"grad_norm": 0.9076519760868025,
|
| 1871 |
+
"learning_rate": 5.960789951281052e-05,
|
| 1872 |
+
"loss": 0.8506,
|
| 1873 |
+
"step": 1315
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 2.01,
|
| 1877 |
+
"grad_norm": 0.8435077804141212,
|
| 1878 |
+
"learning_rate": 5.879674712358061e-05,
|
| 1879 |
+
"loss": 0.8293,
|
| 1880 |
+
"step": 1320
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 2.02,
|
| 1884 |
+
"grad_norm": 1.2235293049079738,
|
| 1885 |
+
"learning_rate": 5.798884713380541e-05,
|
| 1886 |
+
"loss": 0.8727,
|
| 1887 |
+
"step": 1325
|
| 1888 |
+
},
|
| 1889 |
+
{
|
| 1890 |
+
"epoch": 2.03,
|
| 1891 |
+
"grad_norm": 0.982943334202588,
|
| 1892 |
+
"learning_rate": 5.718426331547626e-05,
|
| 1893 |
+
"loss": 0.8403,
|
| 1894 |
+
"step": 1330
|
| 1895 |
+
},
|
| 1896 |
+
{
|
| 1897 |
+
"epoch": 2.04,
|
| 1898 |
+
"grad_norm": 1.0225809341397019,
|
| 1899 |
+
"learning_rate": 5.638305917882093e-05,
|
| 1900 |
+
"loss": 0.825,
|
| 1901 |
+
"step": 1335
|
| 1902 |
+
},
|
| 1903 |
+
{
|
| 1904 |
+
"epoch": 2.04,
|
| 1905 |
+
"grad_norm": 1.0068602287969126,
|
| 1906 |
+
"learning_rate": 5.558529796729015e-05,
|
| 1907 |
+
"loss": 0.8253,
|
| 1908 |
+
"step": 1340
|
| 1909 |
+
},
|
| 1910 |
+
{
|
| 1911 |
+
"epoch": 2.05,
|
| 1912 |
+
"grad_norm": 0.8942997135434371,
|
| 1913 |
+
"learning_rate": 5.479104265256583e-05,
|
| 1914 |
+
"loss": 0.8093,
|
| 1915 |
+
"step": 1345
|
| 1916 |
+
},
|
| 1917 |
+
{
|
| 1918 |
+
"epoch": 2.06,
|
| 1919 |
+
"grad_norm": 0.919639915565314,
|
| 1920 |
+
"learning_rate": 5.4000355929589985e-05,
|
| 1921 |
+
"loss": 0.8736,
|
| 1922 |
+
"step": 1350
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 2.07,
|
| 1926 |
+
"grad_norm": 0.9792431007986793,
|
| 1927 |
+
"learning_rate": 5.32133002116161e-05,
|
| 1928 |
+
"loss": 0.8134,
|
| 1929 |
+
"step": 1355
|
| 1930 |
+
},
|
| 1931 |
+
{
|
| 1932 |
+
"epoch": 2.07,
|
| 1933 |
+
"grad_norm": 0.9705003699468054,
|
| 1934 |
+
"learning_rate": 5.242993762528248e-05,
|
| 1935 |
+
"loss": 0.821,
|
| 1936 |
+
"step": 1360
|
| 1937 |
+
},
|
| 1938 |
+
{
|
| 1939 |
+
"epoch": 2.08,
|
| 1940 |
+
"grad_norm": 1.0687461433169771,
|
| 1941 |
+
"learning_rate": 5.165033000570825e-05,
|
| 1942 |
+
"loss": 0.8586,
|
| 1943 |
+
"step": 1365
|
| 1944 |
+
},
|
| 1945 |
+
{
|
| 1946 |
+
"epoch": 2.09,
|
| 1947 |
+
"grad_norm": 0.9864467254623525,
|
| 1948 |
+
"learning_rate": 5.087453889161229e-05,
|
| 1949 |
+
"loss": 0.7872,
|
| 1950 |
+
"step": 1370
|
| 1951 |
+
},
|
| 1952 |
+
{
|
| 1953 |
+
"epoch": 2.1,
|
| 1954 |
+
"grad_norm": 0.9022370559155457,
|
| 1955 |
+
"learning_rate": 5.010262552045573e-05,
|
| 1956 |
+
"loss": 0.8145,
|
| 1957 |
+
"step": 1375
|
| 1958 |
+
},
|
| 1959 |
+
{
|
| 1960 |
+
"epoch": 2.11,
|
| 1961 |
+
"grad_norm": 0.8356213372223533,
|
| 1962 |
+
"learning_rate": 4.933465082360807e-05,
|
| 1963 |
+
"loss": 0.8521,
|
| 1964 |
+
"step": 1380
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 2.11,
|
| 1968 |
+
"grad_norm": 1.0182132748957182,
|
| 1969 |
+
"learning_rate": 4.857067542153768e-05,
|
| 1970 |
+
"loss": 0.8567,
|
| 1971 |
+
"step": 1385
|
| 1972 |
+
},
|
| 1973 |
+
{
|
| 1974 |
+
"epoch": 2.12,
|
| 1975 |
+
"grad_norm": 1.086022764305583,
|
| 1976 |
+
"learning_rate": 4.781075961902656e-05,
|
| 1977 |
+
"loss": 0.804,
|
| 1978 |
+
"step": 1390
|
| 1979 |
+
},
|
| 1980 |
+
{
|
| 1981 |
+
"epoch": 2.13,
|
| 1982 |
+
"grad_norm": 1.0333147014472928,
|
| 1983 |
+
"learning_rate": 4.705496340041009e-05,
|
| 1984 |
+
"loss": 0.8119,
|
| 1985 |
+
"step": 1395
|
| 1986 |
+
},
|
| 1987 |
+
{
|
| 1988 |
+
"epoch": 2.14,
|
| 1989 |
+
"grad_norm": 1.0968659156489957,
|
| 1990 |
+
"learning_rate": 4.630334642484231e-05,
|
| 1991 |
+
"loss": 0.8434,
|
| 1992 |
+
"step": 1400
|
| 1993 |
+
},
|
| 1994 |
+
{
|
| 1995 |
+
"epoch": 2.14,
|
| 1996 |
+
"grad_norm": 1.1297383258429818,
|
| 1997 |
+
"learning_rate": 4.555596802158653e-05,
|
| 1998 |
+
"loss": 0.8388,
|
| 1999 |
+
"step": 1405
|
| 2000 |
+
},
|
| 2001 |
+
{
|
| 2002 |
+
"epoch": 2.15,
|
| 2003 |
+
"grad_norm": 0.9894862065125212,
|
| 2004 |
+
"learning_rate": 4.481288718533238e-05,
|
| 2005 |
+
"loss": 0.8457,
|
| 2006 |
+
"step": 1410
|
| 2007 |
+
},
|
| 2008 |
+
{
|
| 2009 |
+
"epoch": 2.16,
|
| 2010 |
+
"grad_norm": 1.0086593655810214,
|
| 2011 |
+
"learning_rate": 4.407416257153872e-05,
|
| 2012 |
+
"loss": 0.8452,
|
| 2013 |
+
"step": 1415
|
| 2014 |
+
},
|
| 2015 |
+
{
|
| 2016 |
+
"epoch": 2.17,
|
| 2017 |
+
"grad_norm": 1.077271484795884,
|
| 2018 |
+
"learning_rate": 4.33398524918039e-05,
|
| 2019 |
+
"loss": 0.8307,
|
| 2020 |
+
"step": 1420
|
| 2021 |
+
},
|
| 2022 |
+
{
|
| 2023 |
+
"epoch": 2.17,
|
| 2024 |
+
"grad_norm": 1.101473877979803,
|
| 2025 |
+
"learning_rate": 4.261001490926272e-05,
|
| 2026 |
+
"loss": 0.8228,
|
| 2027 |
+
"step": 1425
|
| 2028 |
+
},
|
| 2029 |
+
{
|
| 2030 |
+
"epoch": 2.18,
|
| 2031 |
+
"grad_norm": 1.1024756868438783,
|
| 2032 |
+
"learning_rate": 4.1884707434011295e-05,
|
| 2033 |
+
"loss": 0.8455,
|
| 2034 |
+
"step": 1430
|
| 2035 |
+
},
|
| 2036 |
+
{
|
| 2037 |
+
"epoch": 2.19,
|
| 2038 |
+
"grad_norm": 0.9596705868289979,
|
| 2039 |
+
"learning_rate": 4.116398731855934e-05,
|
| 2040 |
+
"loss": 0.8429,
|
| 2041 |
+
"step": 1435
|
| 2042 |
+
},
|
| 2043 |
+
{
|
| 2044 |
+
"epoch": 2.2,
|
| 2045 |
+
"grad_norm": 1.0578699242751421,
|
| 2046 |
+
"learning_rate": 4.044791145331105e-05,
|
| 2047 |
+
"loss": 0.7775,
|
| 2048 |
+
"step": 1440
|
| 2049 |
+
},
|
| 2050 |
+
{
|
| 2051 |
+
"epoch": 2.2,
|
| 2052 |
+
"grad_norm": 0.8551950143576533,
|
| 2053 |
+
"learning_rate": 3.973653636207437e-05,
|
| 2054 |
+
"loss": 0.8328,
|
| 2055 |
+
"step": 1445
|
| 2056 |
+
},
|
| 2057 |
+
{
|
| 2058 |
+
"epoch": 2.21,
|
| 2059 |
+
"grad_norm": 1.071052415435941,
|
| 2060 |
+
"learning_rate": 3.902991819759937e-05,
|
| 2061 |
+
"loss": 0.8459,
|
| 2062 |
+
"step": 1450
|
| 2063 |
+
},
|
| 2064 |
+
{
|
| 2065 |
+
"epoch": 2.22,
|
| 2066 |
+
"grad_norm": 1.0178868820951739,
|
| 2067 |
+
"learning_rate": 3.832811273714569e-05,
|
| 2068 |
+
"loss": 0.8003,
|
| 2069 |
+
"step": 1455
|
| 2070 |
+
},
|
| 2071 |
+
{
|
| 2072 |
+
"epoch": 2.23,
|
| 2073 |
+
"grad_norm": 1.0388700331104954,
|
| 2074 |
+
"learning_rate": 3.763117537807972e-05,
|
| 2075 |
+
"loss": 0.8579,
|
| 2076 |
+
"step": 1460
|
| 2077 |
+
},
|
| 2078 |
+
{
|
| 2079 |
+
"epoch": 2.23,
|
| 2080 |
+
"grad_norm": 0.9885214128911158,
|
| 2081 |
+
"learning_rate": 3.693916113350182e-05,
|
| 2082 |
+
"loss": 0.8114,
|
| 2083 |
+
"step": 1465
|
| 2084 |
+
},
|
| 2085 |
+
{
|
| 2086 |
+
"epoch": 2.24,
|
| 2087 |
+
"grad_norm": 0.9507588869671092,
|
| 2088 |
+
"learning_rate": 3.625212462790385e-05,
|
| 2089 |
+
"loss": 0.7656,
|
| 2090 |
+
"step": 1470
|
| 2091 |
+
},
|
| 2092 |
+
{
|
| 2093 |
+
"epoch": 2.25,
|
| 2094 |
+
"grad_norm": 1.033384131735441,
|
| 2095 |
+
"learning_rate": 3.55701200928574e-05,
|
| 2096 |
+
"loss": 0.7871,
|
| 2097 |
+
"step": 1475
|
| 2098 |
+
},
|
| 2099 |
+
{
|
| 2100 |
+
"epoch": 2.26,
|
| 2101 |
+
"grad_norm": 0.9158523594001913,
|
| 2102 |
+
"learning_rate": 3.489320136273288e-05,
|
| 2103 |
+
"loss": 0.8385,
|
| 2104 |
+
"step": 1480
|
| 2105 |
+
},
|
| 2106 |
+
{
|
| 2107 |
+
"epoch": 2.27,
|
| 2108 |
+
"grad_norm": 1.0861208797350566,
|
| 2109 |
+
"learning_rate": 3.422142187045011e-05,
|
| 2110 |
+
"loss": 0.8296,
|
| 2111 |
+
"step": 1485
|
| 2112 |
+
},
|
| 2113 |
+
{
|
| 2114 |
+
"epoch": 2.27,
|
| 2115 |
+
"grad_norm": 1.0108364033236483,
|
| 2116 |
+
"learning_rate": 3.3554834643260624e-05,
|
| 2117 |
+
"loss": 0.8405,
|
| 2118 |
+
"step": 1490
|
| 2119 |
+
},
|
| 2120 |
+
{
|
| 2121 |
+
"epoch": 2.28,
|
| 2122 |
+
"grad_norm": 1.017348028808686,
|
| 2123 |
+
"learning_rate": 3.289349229856197e-05,
|
| 2124 |
+
"loss": 0.8041,
|
| 2125 |
+
"step": 1495
|
| 2126 |
+
},
|
| 2127 |
+
{
|
| 2128 |
+
"epoch": 2.29,
|
| 2129 |
+
"grad_norm": 0.9720934709242245,
|
| 2130 |
+
"learning_rate": 3.223744703974413e-05,
|
| 2131 |
+
"loss": 0.8477,
|
| 2132 |
+
"step": 1500
|
| 2133 |
+
},
|
| 2134 |
+
{
|
| 2135 |
+
"epoch": 2.3,
|
| 2136 |
+
"grad_norm": 1.0491113172636481,
|
| 2137 |
+
"learning_rate": 3.1586750652069074e-05,
|
| 2138 |
+
"loss": 0.8455,
|
| 2139 |
+
"step": 1505
|
| 2140 |
+
},
|
| 2141 |
+
{
|
| 2142 |
+
"epoch": 2.3,
|
| 2143 |
+
"grad_norm": 0.969350149657045,
|
| 2144 |
+
"learning_rate": 3.094145449858285e-05,
|
| 2145 |
+
"loss": 0.8373,
|
| 2146 |
+
"step": 1510
|
| 2147 |
+
},
|
| 2148 |
+
{
|
| 2149 |
+
"epoch": 2.31,
|
| 2150 |
+
"grad_norm": 1.1292836391329433,
|
| 2151 |
+
"learning_rate": 3.030160951606138e-05,
|
| 2152 |
+
"loss": 0.8265,
|
| 2153 |
+
"step": 1515
|
| 2154 |
+
},
|
| 2155 |
+
{
|
| 2156 |
+
"epoch": 2.32,
|
| 2157 |
+
"grad_norm": 1.1377550753616885,
|
| 2158 |
+
"learning_rate": 2.966726621098963e-05,
|
| 2159 |
+
"loss": 0.8724,
|
| 2160 |
+
"step": 1520
|
| 2161 |
+
},
|
| 2162 |
+
{
|
| 2163 |
+
"epoch": 2.33,
|
| 2164 |
+
"grad_norm": 0.9248650902738645,
|
| 2165 |
+
"learning_rate": 2.9038474655574897e-05,
|
| 2166 |
+
"loss": 0.8247,
|
| 2167 |
+
"step": 1525
|
| 2168 |
+
},
|
| 2169 |
+
{
|
| 2170 |
+
"epoch": 2.33,
|
| 2171 |
+
"grad_norm": 1.24903061509406,
|
| 2172 |
+
"learning_rate": 2.84152844837943e-05,
|
| 2173 |
+
"loss": 0.8407,
|
| 2174 |
+
"step": 1530
|
| 2175 |
+
},
|
| 2176 |
+
{
|
| 2177 |
+
"epoch": 2.34,
|
| 2178 |
+
"grad_norm": 0.9300456009720597,
|
| 2179 |
+
"learning_rate": 2.779774488747704e-05,
|
| 2180 |
+
"loss": 0.8017,
|
| 2181 |
+
"step": 1535
|
| 2182 |
+
},
|
| 2183 |
+
{
|
| 2184 |
+
"epoch": 2.35,
|
| 2185 |
+
"grad_norm": 0.9247283493036916,
|
| 2186 |
+
"learning_rate": 2.7185904612421176e-05,
|
| 2187 |
+
"loss": 0.7873,
|
| 2188 |
+
"step": 1540
|
| 2189 |
+
},
|
| 2190 |
+
{
|
| 2191 |
+
"epoch": 2.36,
|
| 2192 |
+
"grad_norm": 0.984266641348006,
|
| 2193 |
+
"learning_rate": 2.6579811954546052e-05,
|
| 2194 |
+
"loss": 0.8231,
|
| 2195 |
+
"step": 1545
|
| 2196 |
+
},
|
| 2197 |
+
{
|
| 2198 |
+
"epoch": 2.36,
|
| 2199 |
+
"grad_norm": 1.0283812513796944,
|
| 2200 |
+
"learning_rate": 2.5979514756079914e-05,
|
| 2201 |
+
"loss": 0.8149,
|
| 2202 |
+
"step": 1550
|
| 2203 |
+
},
|
| 2204 |
+
{
|
| 2205 |
+
"epoch": 2.37,
|
| 2206 |
+
"grad_norm": 1.1106242321896724,
|
| 2207 |
+
"learning_rate": 2.5385060401783566e-05,
|
| 2208 |
+
"loss": 0.8276,
|
| 2209 |
+
"step": 1555
|
| 2210 |
+
},
|
| 2211 |
+
{
|
| 2212 |
+
"epoch": 2.38,
|
| 2213 |
+
"grad_norm": 0.9567893029629212,
|
| 2214 |
+
"learning_rate": 2.4796495815209996e-05,
|
| 2215 |
+
"loss": 0.816,
|
| 2216 |
+
"step": 1560
|
| 2217 |
+
},
|
| 2218 |
+
{
|
| 2219 |
+
"epoch": 2.39,
|
| 2220 |
+
"grad_norm": 1.0867782224252835,
|
| 2221 |
+
"learning_rate": 2.4213867455000338e-05,
|
| 2222 |
+
"loss": 0.892,
|
| 2223 |
+
"step": 1565
|
| 2224 |
+
},
|
| 2225 |
+
{
|
| 2226 |
+
"epoch": 2.4,
|
| 2227 |
+
"grad_norm": 1.1389935740624955,
|
| 2228 |
+
"learning_rate": 2.363722131121675e-05,
|
| 2229 |
+
"loss": 0.8478,
|
| 2230 |
+
"step": 1570
|
| 2231 |
+
},
|
| 2232 |
+
{
|
| 2233 |
+
"epoch": 2.4,
|
| 2234 |
+
"grad_norm": 0.9718824376096208,
|
| 2235 |
+
"learning_rate": 2.3066602901712108e-05,
|
| 2236 |
+
"loss": 0.8496,
|
| 2237 |
+
"step": 1575
|
| 2238 |
+
},
|
| 2239 |
+
{
|
| 2240 |
+
"epoch": 2.41,
|
| 2241 |
+
"grad_norm": 1.1771964742364636,
|
| 2242 |
+
"learning_rate": 2.2502057268537137e-05,
|
| 2243 |
+
"loss": 0.8449,
|
| 2244 |
+
"step": 1580
|
| 2245 |
+
},
|
| 2246 |
+
{
|
| 2247 |
+
"epoch": 2.42,
|
| 2248 |
+
"grad_norm": 1.4160024538664562,
|
| 2249 |
+
"learning_rate": 2.1943628974384857e-05,
|
| 2250 |
+
"loss": 0.8491,
|
| 2251 |
+
"step": 1585
|
| 2252 |
+
},
|
| 2253 |
+
{
|
| 2254 |
+
"epoch": 2.43,
|
| 2255 |
+
"grad_norm": 1.0415796126412569,
|
| 2256 |
+
"learning_rate": 2.1391362099073032e-05,
|
| 2257 |
+
"loss": 0.802,
|
| 2258 |
+
"step": 1590
|
| 2259 |
+
},
|
| 2260 |
+
{
|
| 2261 |
+
"epoch": 2.43,
|
| 2262 |
+
"grad_norm": 0.8715990916453478,
|
| 2263 |
+
"learning_rate": 2.0845300236064757e-05,
|
| 2264 |
+
"loss": 0.8696,
|
| 2265 |
+
"step": 1595
|
| 2266 |
+
},
|
| 2267 |
+
{
|
| 2268 |
+
"epoch": 2.44,
|
| 2269 |
+
"grad_norm": 1.049043541881097,
|
| 2270 |
+
"learning_rate": 2.0305486489027424e-05,
|
| 2271 |
+
"loss": 0.8302,
|
| 2272 |
+
"step": 1600
|
| 2273 |
+
},
|
| 2274 |
+
{
|
| 2275 |
+
"epoch": 2.45,
|
| 2276 |
+
"grad_norm": 0.9586867830799501,
|
| 2277 |
+
"learning_rate": 1.9771963468430187e-05,
|
| 2278 |
+
"loss": 0.8173,
|
| 2279 |
+
"step": 1605
|
| 2280 |
+
},
|
| 2281 |
+
{
|
| 2282 |
+
"epoch": 2.46,
|
| 2283 |
+
"grad_norm": 1.0529652964302254,
|
| 2284 |
+
"learning_rate": 1.924477328818054e-05,
|
| 2285 |
+
"loss": 0.8617,
|
| 2286 |
+
"step": 1610
|
| 2287 |
+
},
|
| 2288 |
+
{
|
| 2289 |
+
"epoch": 2.46,
|
| 2290 |
+
"grad_norm": 1.0449868664649247,
|
| 2291 |
+
"learning_rate": 1.8723957562300133e-05,
|
| 2292 |
+
"loss": 0.8291,
|
| 2293 |
+
"step": 1615
|
| 2294 |
+
},
|
| 2295 |
+
{
|
| 2296 |
+
"epoch": 2.47,
|
| 2297 |
+
"grad_norm": 1.0564336584456713,
|
| 2298 |
+
"learning_rate": 1.8209557401639732e-05,
|
| 2299 |
+
"loss": 0.8175,
|
| 2300 |
+
"step": 1620
|
| 2301 |
+
},
|
| 2302 |
+
{
|
| 2303 |
+
"epoch": 2.48,
|
| 2304 |
+
"grad_norm": 0.9643616639994842,
|
| 2305 |
+
"learning_rate": 1.7701613410634365e-05,
|
| 2306 |
+
"loss": 0.7645,
|
| 2307 |
+
"step": 1625
|
| 2308 |
+
},
|
| 2309 |
+
{
|
| 2310 |
+
"epoch": 2.49,
|
| 2311 |
+
"grad_norm": 1.1870712950124547,
|
| 2312 |
+
"learning_rate": 1.720016568409799e-05,
|
| 2313 |
+
"loss": 0.8509,
|
| 2314 |
+
"step": 1630
|
| 2315 |
+
},
|
| 2316 |
+
{
|
| 2317 |
+
"epoch": 2.49,
|
| 2318 |
+
"grad_norm": 1.06154190477567,
|
| 2319 |
+
"learning_rate": 1.6705253804058675e-05,
|
| 2320 |
+
"loss": 0.8725,
|
| 2321 |
+
"step": 1635
|
| 2322 |
+
},
|
| 2323 |
+
{
|
| 2324 |
+
"epoch": 2.5,
|
| 2325 |
+
"grad_norm": 1.0388956977852164,
|
| 2326 |
+
"learning_rate": 1.621691683663418e-05,
|
| 2327 |
+
"loss": 0.8139,
|
| 2328 |
+
"step": 1640
|
| 2329 |
+
},
|
| 2330 |
+
{
|
| 2331 |
+
"epoch": 2.51,
|
| 2332 |
+
"grad_norm": 1.0678593249410824,
|
| 2333 |
+
"learning_rate": 1.573519332894824e-05,
|
| 2334 |
+
"loss": 0.7882,
|
| 2335 |
+
"step": 1645
|
| 2336 |
+
},
|
| 2337 |
+
{
|
| 2338 |
+
"epoch": 2.52,
|
| 2339 |
+
"grad_norm": 1.0504305711608912,
|
| 2340 |
+
"learning_rate": 1.5260121306087828e-05,
|
| 2341 |
+
"loss": 0.8289,
|
| 2342 |
+
"step": 1650
|
| 2343 |
+
},
|
| 2344 |
+
{
|
| 2345 |
+
"epoch": 2.52,
|
| 2346 |
+
"grad_norm": 1.039744935641794,
|
| 2347 |
+
"learning_rate": 1.4791738268101584e-05,
|
| 2348 |
+
"loss": 0.7841,
|
| 2349 |
+
"step": 1655
|
| 2350 |
+
},
|
| 2351 |
+
{
|
| 2352 |
+
"epoch": 2.53,
|
| 2353 |
+
"grad_norm": 1.0672936964776525,
|
| 2354 |
+
"learning_rate": 1.433008118703979e-05,
|
| 2355 |
+
"loss": 0.7895,
|
| 2356 |
+
"step": 1660
|
| 2357 |
+
},
|
| 2358 |
+
{
|
| 2359 |
+
"epoch": 2.54,
|
| 2360 |
+
"grad_norm": 1.095135634651846,
|
| 2361 |
+
"learning_rate": 1.3875186504035965e-05,
|
| 2362 |
+
"loss": 0.8143,
|
| 2363 |
+
"step": 1665
|
| 2364 |
+
},
|
| 2365 |
+
{
|
| 2366 |
+
"epoch": 2.55,
|
| 2367 |
+
"grad_norm": 1.0672595953625994,
|
| 2368 |
+
"learning_rate": 1.3427090126430342e-05,
|
| 2369 |
+
"loss": 0.8258,
|
| 2370 |
+
"step": 1670
|
| 2371 |
+
},
|
| 2372 |
+
{
|
| 2373 |
+
"epoch": 2.56,
|
| 2374 |
+
"grad_norm": 1.0487848694270845,
|
| 2375 |
+
"learning_rate": 1.298582742493546e-05,
|
| 2376 |
+
"loss": 0.8433,
|
| 2377 |
+
"step": 1675
|
| 2378 |
+
},
|
| 2379 |
+
{
|
| 2380 |
+
"epoch": 2.56,
|
| 2381 |
+
"grad_norm": 0.8678608016330454,
|
| 2382 |
+
"learning_rate": 1.2551433230844224e-05,
|
| 2383 |
+
"loss": 0.7898,
|
| 2384 |
+
"step": 1680
|
| 2385 |
+
},
|
| 2386 |
+
{
|
| 2387 |
+
"epoch": 2.57,
|
| 2388 |
+
"grad_norm": 0.929677257792853,
|
| 2389 |
+
"learning_rate": 1.2123941833280473e-05,
|
| 2390 |
+
"loss": 0.7707,
|
| 2391 |
+
"step": 1685
|
| 2392 |
+
},
|
| 2393 |
+
{
|
| 2394 |
+
"epoch": 2.58,
|
| 2395 |
+
"grad_norm": 1.0288546508719252,
|
| 2396 |
+
"learning_rate": 1.1703386976492337e-05,
|
| 2397 |
+
"loss": 0.8115,
|
| 2398 |
+
"step": 1690
|
| 2399 |
+
},
|
| 2400 |
+
{
|
| 2401 |
+
"epoch": 2.59,
|
| 2402 |
+
"grad_norm": 0.9439251135887912,
|
| 2403 |
+
"learning_rate": 1.1289801857188587e-05,
|
| 2404 |
+
"loss": 0.8584,
|
| 2405 |
+
"step": 1695
|
| 2406 |
+
},
|
| 2407 |
+
{
|
| 2408 |
+
"epoch": 2.59,
|
| 2409 |
+
"grad_norm": 0.9292081088950934,
|
| 2410 |
+
"learning_rate": 1.0883219121918265e-05,
|
| 2411 |
+
"loss": 0.8164,
|
| 2412 |
+
"step": 1700
|
| 2413 |
+
},
|
| 2414 |
+
{
|
| 2415 |
+
"epoch": 2.6,
|
| 2416 |
+
"grad_norm": 0.9421757462427118,
|
| 2417 |
+
"learning_rate": 1.0483670864493778e-05,
|
| 2418 |
+
"loss": 0.795,
|
| 2419 |
+
"step": 1705
|
| 2420 |
+
},
|
| 2421 |
+
{
|
| 2422 |
+
"epoch": 2.61,
|
| 2423 |
+
"grad_norm": 0.999919346225195,
|
| 2424 |
+
"learning_rate": 1.0091188623457415e-05,
|
| 2425 |
+
"loss": 0.8184,
|
| 2426 |
+
"step": 1710
|
| 2427 |
+
},
|
| 2428 |
+
{
|
| 2429 |
+
"epoch": 2.62,
|
| 2430 |
+
"grad_norm": 0.9742832828955831,
|
| 2431 |
+
"learning_rate": 9.70580337959196e-06,
|
| 2432 |
+
"loss": 0.8299,
|
| 2433 |
+
"step": 1715
|
| 2434 |
+
},
|
| 2435 |
+
{
|
| 2436 |
+
"epoch": 2.62,
|
| 2437 |
+
"grad_norm": 0.9089507431330445,
|
| 2438 |
+
"learning_rate": 9.327545553475114e-06,
|
| 2439 |
+
"loss": 0.7723,
|
| 2440 |
+
"step": 1720
|
| 2441 |
+
},
|
| 2442 |
+
{
|
| 2443 |
+
"epoch": 2.63,
|
| 2444 |
+
"grad_norm": 0.9732825474788489,
|
| 2445 |
+
"learning_rate": 8.95644500307835e-06,
|
| 2446 |
+
"loss": 0.8501,
|
| 2447 |
+
"step": 1725
|
| 2448 |
+
},
|
| 2449 |
+
{
|
| 2450 |
+
"epoch": 2.64,
|
| 2451 |
+
"grad_norm": 0.9960143529194506,
|
| 2452 |
+
"learning_rate": 8.592531021409945e-06,
|
| 2453 |
+
"loss": 0.9096,
|
| 2454 |
+
"step": 1730
|
| 2455 |
+
},
|
| 2456 |
+
{
|
| 2457 |
+
"epoch": 2.65,
|
| 2458 |
+
"grad_norm": 0.9339895800344439,
|
| 2459 |
+
"learning_rate": 8.235832334202776e-06,
|
| 2460 |
+
"loss": 0.8828,
|
| 2461 |
+
"step": 1735
|
| 2462 |
+
},
|
| 2463 |
+
{
|
| 2464 |
+
"epoch": 2.65,
|
| 2465 |
+
"grad_norm": 1.056692915239154,
|
| 2466 |
+
"learning_rate": 7.886377097646792e-06,
|
| 2467 |
+
"loss": 0.8749,
|
| 2468 |
+
"step": 1740
|
| 2469 |
+
},
|
| 2470 |
+
{
|
| 2471 |
+
"epoch": 2.66,
|
| 2472 |
+
"grad_norm": 0.9711524416206577,
|
| 2473 |
+
"learning_rate": 7.544192896166568e-06,
|
| 2474 |
+
"loss": 0.815,
|
| 2475 |
+
"step": 1745
|
| 2476 |
+
},
|
| 2477 |
+
{
|
| 2478 |
+
"epoch": 2.67,
|
| 2479 |
+
"grad_norm": 1.0985213550852182,
|
| 2480 |
+
"learning_rate": 7.209306740243893e-06,
|
| 2481 |
+
"loss": 0.8256,
|
| 2482 |
+
"step": 1750
|
| 2483 |
+
},
|
| 2484 |
+
{
|
| 2485 |
+
"epoch": 2.68,
|
| 2486 |
+
"grad_norm": 1.0441483688013498,
|
| 2487 |
+
"learning_rate": 6.881745064285617e-06,
|
| 2488 |
+
"loss": 0.8083,
|
| 2489 |
+
"step": 1755
|
| 2490 |
+
},
|
| 2491 |
+
{
|
| 2492 |
+
"epoch": 2.68,
|
| 2493 |
+
"grad_norm": 0.9998637235645016,
|
| 2494 |
+
"learning_rate": 6.561533724537095e-06,
|
| 2495 |
+
"loss": 0.8116,
|
| 2496 |
+
"step": 1760
|
| 2497 |
+
},
|
| 2498 |
+
{
|
| 2499 |
+
"epoch": 2.69,
|
| 2500 |
+
"grad_norm": 1.1263808784059275,
|
| 2501 |
+
"learning_rate": 6.248697997041219e-06,
|
| 2502 |
+
"loss": 0.811,
|
| 2503 |
+
"step": 1765
|
| 2504 |
+
},
|
| 2505 |
+
{
|
| 2506 |
+
"epoch": 2.7,
|
| 2507 |
+
"grad_norm": 0.9529944445283575,
|
| 2508 |
+
"learning_rate": 5.943262575643238e-06,
|
| 2509 |
+
"loss": 0.8225,
|
| 2510 |
+
"step": 1770
|
| 2511 |
+
},
|
| 2512 |
+
{
|
| 2513 |
+
"epoch": 2.71,
|
| 2514 |
+
"grad_norm": 0.9459961315540705,
|
| 2515 |
+
"learning_rate": 5.645251570041488e-06,
|
| 2516 |
+
"loss": 0.7937,
|
| 2517 |
+
"step": 1775
|
| 2518 |
+
},
|
| 2519 |
+
{
|
| 2520 |
+
"epoch": 2.72,
|
| 2521 |
+
"grad_norm": 0.9556679665444172,
|
| 2522 |
+
"learning_rate": 5.354688503884342e-06,
|
| 2523 |
+
"loss": 0.8109,
|
| 2524 |
+
"step": 1780
|
| 2525 |
+
},
|
| 2526 |
+
{
|
| 2527 |
+
"epoch": 2.72,
|
| 2528 |
+
"grad_norm": 1.162754539840243,
|
| 2529 |
+
"learning_rate": 5.071596312913329e-06,
|
| 2530 |
+
"loss": 0.8337,
|
| 2531 |
+
"step": 1785
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 2.73,
|
| 2535 |
+
"grad_norm": 0.987118403127622,
|
| 2536 |
+
"learning_rate": 4.79599734315267e-06,
|
| 2537 |
+
"loss": 0.8208,
|
| 2538 |
+
"step": 1790
|
| 2539 |
+
},
|
| 2540 |
+
{
|
| 2541 |
+
"epoch": 2.74,
|
| 2542 |
+
"grad_norm": 1.0185381683595645,
|
| 2543 |
+
"learning_rate": 4.527913349145441e-06,
|
| 2544 |
+
"loss": 0.7854,
|
| 2545 |
+
"step": 1795
|
| 2546 |
+
},
|
| 2547 |
+
{
|
| 2548 |
+
"epoch": 2.75,
|
| 2549 |
+
"grad_norm": 1.036855323975471,
|
| 2550 |
+
"learning_rate": 4.267365492236319e-06,
|
| 2551 |
+
"loss": 0.8138,
|
| 2552 |
+
"step": 1800
|
| 2553 |
+
},
|
| 2554 |
+
{
|
| 2555 |
+
"epoch": 2.75,
|
| 2556 |
+
"grad_norm": 0.906565660746568,
|
| 2557 |
+
"learning_rate": 4.014374338901206e-06,
|
| 2558 |
+
"loss": 0.782,
|
| 2559 |
+
"step": 1805
|
| 2560 |
+
},
|
| 2561 |
+
{
|
| 2562 |
+
"epoch": 2.76,
|
| 2563 |
+
"grad_norm": 0.9770509541103689,
|
| 2564 |
+
"learning_rate": 3.768959859123822e-06,
|
| 2565 |
+
"loss": 0.8191,
|
| 2566 |
+
"step": 1810
|
| 2567 |
+
},
|
| 2568 |
+
{
|
| 2569 |
+
"epoch": 2.77,
|
| 2570 |
+
"grad_norm": 1.0486391035448164,
|
| 2571 |
+
"learning_rate": 3.531141424819373e-06,
|
| 2572 |
+
"loss": 0.8603,
|
| 2573 |
+
"step": 1815
|
| 2574 |
+
},
|
| 2575 |
+
{
|
| 2576 |
+
"epoch": 2.78,
|
| 2577 |
+
"grad_norm": 1.0773318638652334,
|
| 2578 |
+
"learning_rate": 3.300937808305371e-06,
|
| 2579 |
+
"loss": 0.8521,
|
| 2580 |
+
"step": 1820
|
| 2581 |
+
},
|
| 2582 |
+
{
|
| 2583 |
+
"epoch": 2.78,
|
| 2584 |
+
"grad_norm": 0.985804357345318,
|
| 2585 |
+
"learning_rate": 3.0783671808198635e-06,
|
| 2586 |
+
"loss": 0.8141,
|
| 2587 |
+
"step": 1825
|
| 2588 |
+
},
|
| 2589 |
+
{
|
| 2590 |
+
"epoch": 2.79,
|
| 2591 |
+
"grad_norm": 1.1529169013999445,
|
| 2592 |
+
"learning_rate": 2.8634471110871032e-06,
|
| 2593 |
+
"loss": 0.7832,
|
| 2594 |
+
"step": 1830
|
| 2595 |
+
},
|
| 2596 |
+
{
|
| 2597 |
+
"epoch": 2.8,
|
| 2598 |
+
"grad_norm": 0.9031631771021017,
|
| 2599 |
+
"learning_rate": 2.656194563930714e-06,
|
| 2600 |
+
"loss": 0.8585,
|
| 2601 |
+
"step": 1835
|
| 2602 |
+
},
|
| 2603 |
+
{
|
| 2604 |
+
"epoch": 2.81,
|
| 2605 |
+
"grad_norm": 0.8637751952649739,
|
| 2606 |
+
"learning_rate": 2.4566258989345726e-06,
|
| 2607 |
+
"loss": 0.7996,
|
| 2608 |
+
"step": 1840
|
| 2609 |
+
},
|
| 2610 |
+
{
|
| 2611 |
+
"epoch": 2.81,
|
| 2612 |
+
"grad_norm": 1.0164681340381803,
|
| 2613 |
+
"learning_rate": 2.264756869151441e-06,
|
| 2614 |
+
"loss": 0.8044,
|
| 2615 |
+
"step": 1845
|
| 2616 |
+
},
|
| 2617 |
+
{
|
| 2618 |
+
"epoch": 2.82,
|
| 2619 |
+
"grad_norm": 1.0556120127779176,
|
| 2620 |
+
"learning_rate": 2.0806026198595173e-06,
|
| 2621 |
+
"loss": 0.7943,
|
| 2622 |
+
"step": 1850
|
| 2623 |
+
},
|
| 2624 |
+
{
|
| 2625 |
+
"epoch": 2.83,
|
| 2626 |
+
"grad_norm": 1.1048150934514633,
|
| 2627 |
+
"learning_rate": 1.9041776873669704e-06,
|
| 2628 |
+
"loss": 0.8225,
|
| 2629 |
+
"step": 1855
|
| 2630 |
+
},
|
| 2631 |
+
{
|
| 2632 |
+
"epoch": 2.84,
|
| 2633 |
+
"grad_norm": 0.9507298133478834,
|
| 2634 |
+
"learning_rate": 1.7354959978644225e-06,
|
| 2635 |
+
"loss": 0.8107,
|
| 2636 |
+
"step": 1860
|
| 2637 |
+
},
|
| 2638 |
+
{
|
| 2639 |
+
"epoch": 2.85,
|
| 2640 |
+
"grad_norm": 1.1226388926103825,
|
| 2641 |
+
"learning_rate": 1.5745708663257197e-06,
|
| 2642 |
+
"loss": 0.8282,
|
| 2643 |
+
"step": 1865
|
| 2644 |
+
},
|
| 2645 |
+
{
|
| 2646 |
+
"epoch": 2.85,
|
| 2647 |
+
"grad_norm": 0.929914623001807,
|
| 2648 |
+
"learning_rate": 1.4214149954569268e-06,
|
| 2649 |
+
"loss": 0.7743,
|
| 2650 |
+
"step": 1870
|
| 2651 |
+
},
|
| 2652 |
+
{
|
| 2653 |
+
"epoch": 2.86,
|
| 2654 |
+
"grad_norm": 0.8451615970437887,
|
| 2655 |
+
"learning_rate": 1.2760404746936183e-06,
|
| 2656 |
+
"loss": 0.799,
|
| 2657 |
+
"step": 1875
|
| 2658 |
+
},
|
| 2659 |
+
{
|
| 2660 |
+
"epoch": 2.87,
|
| 2661 |
+
"grad_norm": 0.9722281685095321,
|
| 2662 |
+
"learning_rate": 1.1384587792465872e-06,
|
| 2663 |
+
"loss": 0.8559,
|
| 2664 |
+
"step": 1880
|
| 2665 |
+
},
|
| 2666 |
+
{
|
| 2667 |
+
"epoch": 2.88,
|
| 2668 |
+
"grad_norm": 0.9340387136482731,
|
| 2669 |
+
"learning_rate": 1.0086807691960244e-06,
|
| 2670 |
+
"loss": 0.822,
|
| 2671 |
+
"step": 1885
|
| 2672 |
+
},
|
| 2673 |
+
{
|
| 2674 |
+
"epoch": 2.88,
|
| 2675 |
+
"grad_norm": 0.9046388002266489,
|
| 2676 |
+
"learning_rate": 8.867166886343592e-07,
|
| 2677 |
+
"loss": 0.8384,
|
| 2678 |
+
"step": 1890
|
| 2679 |
+
},
|
| 2680 |
+
{
|
| 2681 |
+
"epoch": 2.89,
|
| 2682 |
+
"grad_norm": 0.9601062532338479,
|
| 2683 |
+
"learning_rate": 7.725761648574969e-07,
|
| 2684 |
+
"loss": 0.8342,
|
| 2685 |
+
"step": 1895
|
| 2686 |
+
},
|
| 2687 |
+
{
|
| 2688 |
+
"epoch": 2.9,
|
| 2689 |
+
"grad_norm": 0.99460636173352,
|
| 2690 |
+
"learning_rate": 6.662682076050031e-07,
|
| 2691 |
+
"loss": 0.8721,
|
| 2692 |
+
"step": 1900
|
| 2693 |
+
},
|
| 2694 |
+
{
|
| 2695 |
+
"epoch": 2.91,
|
| 2696 |
+
"grad_norm": 0.9721186202775147,
|
| 2697 |
+
"learning_rate": 5.678012083488727e-07,
|
| 2698 |
+
"loss": 0.8559,
|
| 2699 |
+
"step": 1905
|
| 2700 |
+
},
|
| 2701 |
+
{
|
| 2702 |
+
"epoch": 2.91,
|
| 2703 |
+
"grad_norm": 1.2579065537174423,
|
| 2704 |
+
"learning_rate": 4.771829396311267e-07,
|
| 2705 |
+
"loss": 0.7897,
|
| 2706 |
+
"step": 1910
|
| 2707 |
+
},
|
| 2708 |
+
{
|
| 2709 |
+
"epoch": 2.92,
|
| 2710 |
+
"grad_norm": 1.011827783498998,
|
| 2711 |
+
"learning_rate": 3.9442055445029214e-07,
|
| 2712 |
+
"loss": 0.7987,
|
| 2713 |
+
"step": 1915
|
| 2714 |
+
},
|
| 2715 |
+
{
|
| 2716 |
+
"epoch": 2.93,
|
| 2717 |
+
"grad_norm": 1.133610710351972,
|
| 2718 |
+
"learning_rate": 3.1952058569681975e-07,
|
| 2719 |
+
"loss": 0.8083,
|
| 2720 |
+
"step": 1920
|
| 2721 |
+
},
|
| 2722 |
+
{
|
| 2723 |
+
"epoch": 2.94,
|
| 2724 |
+
"grad_norm": 0.9463613366312343,
|
| 2725 |
+
"learning_rate": 2.5248894563735247e-07,
|
| 2726 |
+
"loss": 0.848,
|
| 2727 |
+
"step": 1925
|
| 2728 |
+
},
|
| 2729 |
+
{
|
| 2730 |
+
"epoch": 2.94,
|
| 2731 |
+
"grad_norm": 1.0867623661570065,
|
| 2732 |
+
"learning_rate": 1.933309254480542e-07,
|
| 2733 |
+
"loss": 0.83,
|
| 2734 |
+
"step": 1930
|
| 2735 |
+
},
|
| 2736 |
+
{
|
| 2737 |
+
"epoch": 2.95,
|
| 2738 |
+
"grad_norm": 0.8980549292174614,
|
| 2739 |
+
"learning_rate": 1.4205119479693274e-07,
|
| 2740 |
+
"loss": 0.8238,
|
| 2741 |
+
"step": 1935
|
| 2742 |
+
},
|
| 2743 |
+
{
|
| 2744 |
+
"epoch": 2.96,
|
| 2745 |
+
"grad_norm": 0.8870015504798887,
|
| 2746 |
+
"learning_rate": 9.865380147530134e-08,
|
| 2747 |
+
"loss": 0.8323,
|
| 2748 |
+
"step": 1940
|
| 2749 |
+
},
|
| 2750 |
+
{
|
| 2751 |
+
"epoch": 2.97,
|
| 2752 |
+
"grad_norm": 1.1236359515597645,
|
| 2753 |
+
"learning_rate": 6.314217107817877e-08,
|
| 2754 |
+
"loss": 0.8563,
|
| 2755 |
+
"step": 1945
|
| 2756 |
+
},
|
| 2757 |
+
{
|
| 2758 |
+
"epoch": 2.97,
|
| 2759 |
+
"grad_norm": 1.0317942271237952,
|
| 2760 |
+
"learning_rate": 3.551910673394998e-08,
|
| 2761 |
+
"loss": 0.851,
|
| 2762 |
+
"step": 1950
|
| 2763 |
+
},
|
| 2764 |
+
{
|
| 2765 |
+
"epoch": 2.98,
|
| 2766 |
+
"grad_norm": 1.0177460190731211,
|
| 2767 |
+
"learning_rate": 1.5786788883065394e-08,
|
| 2768 |
+
"loss": 0.8421,
|
| 2769 |
+
"step": 1955
|
| 2770 |
+
},
|
| 2771 |
+
{
|
| 2772 |
+
"epoch": 2.99,
|
| 2773 |
+
"grad_norm": 1.0956132894984734,
|
| 2774 |
+
"learning_rate": 3.9467751059341085e-09,
|
| 2775 |
+
"loss": 0.8274,
|
| 2776 |
+
"step": 1960
|
| 2777 |
+
},
|
| 2778 |
+
{
|
| 2779 |
+
"epoch": 3.0,
|
| 2780 |
+
"grad_norm": 1.2600799171859316,
|
| 2781 |
+
"learning_rate": 0.0,
|
| 2782 |
+
"loss": 0.8345,
|
| 2783 |
+
"step": 1965
|
| 2784 |
+
},
|
| 2785 |
+
{
|
| 2786 |
+
"epoch": 3.0,
|
| 2787 |
+
"eval_loss": 1.2143404483795166,
|
| 2788 |
+
"eval_runtime": 248.6803,
|
| 2789 |
+
"eval_samples_per_second": 9.289,
|
| 2790 |
+
"eval_steps_per_second": 0.583,
|
| 2791 |
+
"step": 1965
|
| 2792 |
+
},
|
| 2793 |
+
{
|
| 2794 |
+
"epoch": 3.0,
|
| 2795 |
+
"step": 1965,
|
| 2796 |
+
"total_flos": 4085393396662272.0,
|
| 2797 |
+
"train_loss": 1.1282114190303036,
|
| 2798 |
+
"train_runtime": 21563.1342,
|
| 2799 |
+
"train_samples_per_second": 2.918,
|
| 2800 |
+
"train_steps_per_second": 0.091
|
| 2801 |
+
}
|
| 2802 |
+
],
|
| 2803 |
+
"logging_steps": 5,
|
| 2804 |
+
"max_steps": 1965,
|
| 2805 |
+
"num_input_tokens_seen": 0,
|
| 2806 |
+
"num_train_epochs": 3,
|
| 2807 |
+
"save_steps": 500,
|
| 2808 |
+
"total_flos": 4085393396662272.0,
|
| 2809 |
+
"train_batch_size": 4,
|
| 2810 |
+
"trial_name": null,
|
| 2811 |
+
"trial_params": null
|
| 2812 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a1fe485c713c909e975a4de69235986f1675bb57181954ad65b31cd36ee566a3
|
| 3 |
+
size 6200
|