Uploading /ephemeral/hossein/output/prog-y-tiny-llama-CDL-21
Browse files- README.md +61 -0
- all_results.json +8 -0
- config.json +1 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- runs/progressive-yoco-Tinyllama-Layer21/events.out.tfevents.1732513661.creative-turing-2.1196220.0 +3 -0
- special_tokens_map.json +24 -0
- tokenizer.json +0 -0
- tokenizer.model +3 -0
- tokenizer_config.json +44 -0
- train_results.json +8 -0
- trainer_log.jsonl +151 -0
- trainer_state.json +1092 -0
- training_args.bin +3 -0
- training_loss.png +0 -0
README.md
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: transformers
|
| 3 |
+
license: other
|
| 4 |
+
base_model: TinyLlama/TinyLlama_v1.1
|
| 5 |
+
tags:
|
| 6 |
+
- llama-factory
|
| 7 |
+
- full
|
| 8 |
+
- generated_from_trainer
|
| 9 |
+
model-index:
|
| 10 |
+
- name: progressive-yoco-tiny-llama-CDL-21
|
| 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 |
+
# progressive-yoco-tiny-llama-CDL-21
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [/ephemeral/hossein/hf_models/y-tinyllama/](https://huggingface.co//ephemeral/hossein/hf_models/y-tinyllama/) on the reformatted_ultrachat_200k dataset.
|
| 20 |
+
|
| 21 |
+
## Model description
|
| 22 |
+
|
| 23 |
+
More information needed
|
| 24 |
+
|
| 25 |
+
## Intended uses & limitations
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Training and evaluation data
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training procedure
|
| 34 |
+
|
| 35 |
+
### Training hyperparameters
|
| 36 |
+
|
| 37 |
+
The following hyperparameters were used during training:
|
| 38 |
+
- learning_rate: 2e-05
|
| 39 |
+
- train_batch_size: 58
|
| 40 |
+
- eval_batch_size: 1
|
| 41 |
+
- seed: 42
|
| 42 |
+
- distributed_type: multi-GPU
|
| 43 |
+
- num_devices: 8
|
| 44 |
+
- gradient_accumulation_steps: 3
|
| 45 |
+
- total_train_batch_size: 1392
|
| 46 |
+
- total_eval_batch_size: 8
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: cosine
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.005
|
| 50 |
+
- training_steps: 150
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
|
| 55 |
+
|
| 56 |
+
### Framework versions
|
| 57 |
+
|
| 58 |
+
- Transformers 4.45.2
|
| 59 |
+
- Pytorch 2.5.1+cu124
|
| 60 |
+
- Datasets 3.1.0
|
| 61 |
+
- Tokenizers 0.20.3
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.6423357664233578,
|
| 3 |
+
"total_flos": 241175567007744.0,
|
| 4 |
+
"train_loss": 1.6121594174702962,
|
| 5 |
+
"train_runtime": 3593.9445,
|
| 6 |
+
"train_samples_per_second": 58.098,
|
| 7 |
+
"train_steps_per_second": 0.042
|
| 8 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"_name_or_path": "TinyLlama/TinyLlama_v1.1", "architectures": ["ProgressiveYocoLlamaForCausalLM"], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "crossDecoder_start_idx": 5, "eos_token_id": 2, "hidden_act": "silu", "hidden_size": 2048, "initializer_range": 0.02, "intermediate_size": 5632, "max_position_embeddings": 2048, "mlp_bias": false, "model_type": "progressive_yoco_llama", "num_attention_heads": 32, "num_hidden_layers": 22, "num_key_value_heads": 4, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 10000.0, "tie_word_embeddings": false, "torch_dtype": "bfloat16", "transformers_version": "4.45.2", "use_cache": false, "vocab_size": 32000}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 1,
|
| 3 |
+
"eos_token_id": 2,
|
| 4 |
+
"max_length": 2048,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.45.2"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8c0edbc339ac8bd1ca9e86050f3b91d6f4dde6a49bd0f2097065dc0ef3df3a3a
|
| 3 |
+
size 2191734544
|
runs/progressive-yoco-Tinyllama-Layer21/events.out.tfevents.1732513661.creative-turing-2.1196220.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7daf275e9e495f56e425529e12253f0e58897c3d99785e2f75819112dd601e42
|
| 3 |
+
size 36891
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "</s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": "</s>",
|
| 17 |
+
"unk_token": {
|
| 18 |
+
"content": "<unk>",
|
| 19 |
+
"lstrip": false,
|
| 20 |
+
"normalized": false,
|
| 21 |
+
"rstrip": false,
|
| 22 |
+
"single_word": false
|
| 23 |
+
}
|
| 24 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
|
| 3 |
+
size 499723
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": true,
|
| 3 |
+
"add_eos_token": false,
|
| 4 |
+
"add_prefix_space": null,
|
| 5 |
+
"added_tokens_decoder": {
|
| 6 |
+
"0": {
|
| 7 |
+
"content": "<unk>",
|
| 8 |
+
"lstrip": false,
|
| 9 |
+
"normalized": false,
|
| 10 |
+
"rstrip": false,
|
| 11 |
+
"single_word": false,
|
| 12 |
+
"special": true
|
| 13 |
+
},
|
| 14 |
+
"1": {
|
| 15 |
+
"content": "<s>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": false,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false,
|
| 20 |
+
"special": true
|
| 21 |
+
},
|
| 22 |
+
"2": {
|
| 23 |
+
"content": "</s>",
|
| 24 |
+
"lstrip": false,
|
| 25 |
+
"normalized": false,
|
| 26 |
+
"rstrip": false,
|
| 27 |
+
"single_word": false,
|
| 28 |
+
"special": true
|
| 29 |
+
}
|
| 30 |
+
},
|
| 31 |
+
"bos_token": "<s>",
|
| 32 |
+
"chat_template": "{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}",
|
| 33 |
+
"clean_up_tokenization_spaces": false,
|
| 34 |
+
"eos_token": "</s>",
|
| 35 |
+
"legacy": false,
|
| 36 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 37 |
+
"pad_token": "</s>",
|
| 38 |
+
"padding_side": "right",
|
| 39 |
+
"sp_model_kwargs": {},
|
| 40 |
+
"split_special_tokens": false,
|
| 41 |
+
"tokenizer_class": "LlamaTokenizer",
|
| 42 |
+
"unk_token": "<unk>",
|
| 43 |
+
"use_default_system_prompt": false
|
| 44 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.6423357664233578,
|
| 3 |
+
"total_flos": 241175567007744.0,
|
| 4 |
+
"train_loss": 1.6121594174702962,
|
| 5 |
+
"train_runtime": 3593.9445,
|
| 6 |
+
"train_samples_per_second": 58.098,
|
| 7 |
+
"train_steps_per_second": 0.042
|
| 8 |
+
}
|
trainer_log.jsonl
ADDED
|
@@ -0,0 +1,151 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"current_steps": 1, "total_steps": 150, "loss": 5.8607, "lr": 2e-05, "epoch": 0.010948905109489052, "percentage": 0.67, "elapsed_time": "0:00:27", "remaining_time": "1:07:55"}
|
| 2 |
+
{"current_steps": 2, "total_steps": 150, "loss": 5.8686, "lr": 1.999777729859618e-05, "epoch": 0.021897810218978103, "percentage": 1.33, "elapsed_time": "0:00:50", "remaining_time": "1:02:33"}
|
| 3 |
+
{"current_steps": 3, "total_steps": 150, "loss": 5.8515, "lr": 1.9991110182465032e-05, "epoch": 0.032846715328467155, "percentage": 2.0, "elapsed_time": "0:01:14", "remaining_time": "1:00:37"}
|
| 4 |
+
{"current_steps": 4, "total_steps": 150, "loss": 4.5812, "lr": 1.9980001615408228e-05, "epoch": 0.043795620437956206, "percentage": 2.67, "elapsed_time": "0:01:37", "remaining_time": "0:59:31"}
|
| 5 |
+
{"current_steps": 5, "total_steps": 150, "loss": 3.9891, "lr": 1.9964456535631287e-05, "epoch": 0.05474452554744526, "percentage": 3.33, "elapsed_time": "0:02:01", "remaining_time": "0:58:43"}
|
| 6 |
+
{"current_steps": 6, "total_steps": 150, "loss": 3.3771, "lr": 1.9944481853548335e-05, "epoch": 0.06569343065693431, "percentage": 4.0, "elapsed_time": "0:02:25", "remaining_time": "0:58:05"}
|
| 7 |
+
{"current_steps": 7, "total_steps": 150, "loss": 2.935, "lr": 1.9920086448710162e-05, "epoch": 0.07664233576642336, "percentage": 4.67, "elapsed_time": "0:02:48", "remaining_time": "0:57:30"}
|
| 8 |
+
{"current_steps": 8, "total_steps": 150, "loss": 2.4909, "lr": 1.9891281165856876e-05, "epoch": 0.08759124087591241, "percentage": 5.33, "elapsed_time": "0:03:12", "remaining_time": "0:56:59"}
|
| 9 |
+
{"current_steps": 9, "total_steps": 150, "loss": 2.3074, "lr": 1.9858078810097004e-05, "epoch": 0.09854014598540146, "percentage": 6.0, "elapsed_time": "0:03:36", "remaining_time": "0:56:29"}
|
| 10 |
+
{"current_steps": 10, "total_steps": 150, "loss": 2.168, "lr": 1.98204941412151e-05, "epoch": 0.10948905109489052, "percentage": 6.67, "elapsed_time": "0:04:00", "remaining_time": "0:56:02"}
|
| 11 |
+
{"current_steps": 11, "total_steps": 150, "loss": 2.0632, "lr": 1.9778543867110428e-05, "epoch": 0.12043795620437957, "percentage": 7.33, "elapsed_time": "0:04:23", "remaining_time": "0:55:35"}
|
| 12 |
+
{"current_steps": 12, "total_steps": 150, "loss": 1.9709, "lr": 1.9732246636369605e-05, "epoch": 0.13138686131386862, "percentage": 8.0, "elapsed_time": "0:04:47", "remaining_time": "0:55:08"}
|
| 13 |
+
{"current_steps": 13, "total_steps": 150, "loss": 1.9209, "lr": 1.968162302997659e-05, "epoch": 0.14233576642335766, "percentage": 8.67, "elapsed_time": "0:05:11", "remaining_time": "0:54:42"}
|
| 14 |
+
{"current_steps": 14, "total_steps": 150, "loss": 1.8743, "lr": 1.962669555216358e-05, "epoch": 0.15328467153284672, "percentage": 9.33, "elapsed_time": "0:05:35", "remaining_time": "0:54:16"}
|
| 15 |
+
{"current_steps": 15, "total_steps": 150, "loss": 1.8264, "lr": 1.9567488620406984e-05, "epoch": 0.16423357664233576, "percentage": 10.0, "elapsed_time": "0:05:59", "remaining_time": "0:53:51"}
|
| 16 |
+
{"current_steps": 16, "total_steps": 150, "loss": 1.7879, "lr": 1.9504028554572865e-05, "epoch": 0.17518248175182483, "percentage": 10.67, "elapsed_time": "0:06:22", "remaining_time": "0:53:25"}
|
| 17 |
+
{"current_steps": 17, "total_steps": 150, "loss": 1.7542, "lr": 1.943634356521671e-05, "epoch": 0.18613138686131386, "percentage": 11.33, "elapsed_time": "0:06:46", "remaining_time": "0:53:00"}
|
| 18 |
+
{"current_steps": 18, "total_steps": 150, "loss": 1.7429, "lr": 1.9364463741042694e-05, "epoch": 0.19708029197080293, "percentage": 12.0, "elapsed_time": "0:07:10", "remaining_time": "0:52:35"}
|
| 19 |
+
{"current_steps": 19, "total_steps": 150, "loss": 1.707, "lr": 1.928842103552803e-05, "epoch": 0.20802919708029197, "percentage": 12.67, "elapsed_time": "0:07:34", "remaining_time": "0:52:10"}
|
| 20 |
+
{"current_steps": 20, "total_steps": 150, "loss": 1.6844, "lr": 1.920824925271838e-05, "epoch": 0.21897810218978103, "percentage": 13.33, "elapsed_time": "0:07:57", "remaining_time": "0:51:45"}
|
| 21 |
+
{"current_steps": 21, "total_steps": 150, "loss": 1.6516, "lr": 1.9123984032200586e-05, "epoch": 0.22992700729927007, "percentage": 14.0, "elapsed_time": "0:08:21", "remaining_time": "0:51:21"}
|
| 22 |
+
{"current_steps": 22, "total_steps": 150, "loss": 1.6474, "lr": 1.9035662833259433e-05, "epoch": 0.24087591240875914, "percentage": 14.67, "elapsed_time": "0:08:45", "remaining_time": "0:50:56"}
|
| 23 |
+
{"current_steps": 23, "total_steps": 150, "loss": 1.6311, "lr": 1.8943324918225495e-05, "epoch": 0.2518248175182482, "percentage": 15.33, "elapsed_time": "0:09:09", "remaining_time": "0:50:32"}
|
| 24 |
+
{"current_steps": 24, "total_steps": 150, "loss": 1.6035, "lr": 1.8847011335021447e-05, "epoch": 0.26277372262773724, "percentage": 16.0, "elapsed_time": "0:09:32", "remaining_time": "0:50:07"}
|
| 25 |
+
{"current_steps": 25, "total_steps": 150, "loss": 1.6014, "lr": 1.874676489891461e-05, "epoch": 0.2737226277372263, "percentage": 16.67, "elapsed_time": "0:09:56", "remaining_time": "0:49:43"}
|
| 26 |
+
{"current_steps": 26, "total_steps": 150, "loss": 1.5806, "lr": 1.8642630173483832e-05, "epoch": 0.2846715328467153, "percentage": 17.33, "elapsed_time": "0:10:20", "remaining_time": "0:49:19"}
|
| 27 |
+
{"current_steps": 27, "total_steps": 150, "loss": 1.5736, "lr": 1.85346534508092e-05, "epoch": 0.2956204379562044, "percentage": 18.0, "elapsed_time": "0:10:44", "remaining_time": "0:48:54"}
|
| 28 |
+
{"current_steps": 28, "total_steps": 150, "loss": 1.5642, "lr": 1.8422882730893323e-05, "epoch": 0.30656934306569344, "percentage": 18.67, "elapsed_time": "0:11:07", "remaining_time": "0:48:30"}
|
| 29 |
+
{"current_steps": 29, "total_steps": 150, "loss": 1.5504, "lr": 1.8307367700323412e-05, "epoch": 0.3175182481751825, "percentage": 19.33, "elapsed_time": "0:11:31", "remaining_time": "0:48:06"}
|
| 30 |
+
{"current_steps": 30, "total_steps": 150, "loss": 1.5498, "lr": 1.8188159710183595e-05, "epoch": 0.3284671532846715, "percentage": 20.0, "elapsed_time": "0:11:55", "remaining_time": "0:47:41"}
|
| 31 |
+
{"current_steps": 31, "total_steps": 150, "loss": 1.5283, "lr": 1.8065311753227272e-05, "epoch": 0.33941605839416056, "percentage": 20.67, "elapsed_time": "0:12:19", "remaining_time": "0:47:17"}
|
| 32 |
+
{"current_steps": 32, "total_steps": 150, "loss": 1.5455, "lr": 1.7938878440319722e-05, "epoch": 0.35036496350364965, "percentage": 21.33, "elapsed_time": "0:12:42", "remaining_time": "0:46:53"}
|
| 33 |
+
{"current_steps": 33, "total_steps": 150, "loss": 1.4982, "lr": 1.7808915976161364e-05, "epoch": 0.3613138686131387, "percentage": 22.0, "elapsed_time": "0:13:06", "remaining_time": "0:46:29"}
|
| 34 |
+
{"current_steps": 34, "total_steps": 150, "loss": 1.5187, "lr": 1.7675482134302503e-05, "epoch": 0.3722627737226277, "percentage": 22.67, "elapsed_time": "0:13:30", "remaining_time": "0:46:05"}
|
| 35 |
+
{"current_steps": 35, "total_steps": 150, "loss": 1.5082, "lr": 1.753863623146066e-05, "epoch": 0.38321167883211676, "percentage": 23.33, "elapsed_time": "0:13:54", "remaining_time": "0:45:41"}
|
| 36 |
+
{"current_steps": 36, "total_steps": 150, "loss": 1.5055, "lr": 1.7398439101151908e-05, "epoch": 0.39416058394160586, "percentage": 24.0, "elapsed_time": "0:14:18", "remaining_time": "0:45:17"}
|
| 37 |
+
{"current_steps": 37, "total_steps": 150, "loss": 1.5039, "lr": 1.7254953066647915e-05, "epoch": 0.4051094890510949, "percentage": 24.67, "elapsed_time": "0:14:41", "remaining_time": "0:44:53"}
|
| 38 |
+
{"current_steps": 38, "total_steps": 150, "loss": 1.4877, "lr": 1.710824191327075e-05, "epoch": 0.41605839416058393, "percentage": 25.33, "elapsed_time": "0:15:05", "remaining_time": "0:44:29"}
|
| 39 |
+
{"current_steps": 39, "total_steps": 150, "loss": 1.4882, "lr": 1.695837086003772e-05, "epoch": 0.42700729927007297, "percentage": 26.0, "elapsed_time": "0:15:29", "remaining_time": "0:44:04"}
|
| 40 |
+
{"current_steps": 40, "total_steps": 150, "loss": 1.4765, "lr": 1.680540653066891e-05, "epoch": 0.43795620437956206, "percentage": 26.67, "elapsed_time": "0:15:53", "remaining_time": "0:43:40"}
|
| 41 |
+
{"current_steps": 41, "total_steps": 150, "loss": 1.4868, "lr": 1.6649416923970248e-05, "epoch": 0.4489051094890511, "percentage": 27.33, "elapsed_time": "0:16:16", "remaining_time": "0:43:16"}
|
| 42 |
+
{"current_steps": 42, "total_steps": 150, "loss": 1.4797, "lr": 1.649047138360529e-05, "epoch": 0.45985401459854014, "percentage": 28.0, "elapsed_time": "0:16:40", "remaining_time": "0:42:52"}
|
| 43 |
+
{"current_steps": 43, "total_steps": 150, "loss": 1.476, "lr": 1.632864056726917e-05, "epoch": 0.4708029197080292, "percentage": 28.67, "elapsed_time": "0:17:04", "remaining_time": "0:42:28"}
|
| 44 |
+
{"current_steps": 44, "total_steps": 150, "loss": 1.4471, "lr": 1.6163996415278423e-05, "epoch": 0.48175182481751827, "percentage": 29.33, "elapsed_time": "0:17:28", "remaining_time": "0:42:04"}
|
| 45 |
+
{"current_steps": 45, "total_steps": 150, "loss": 1.4603, "lr": 1.5996612118590604e-05, "epoch": 0.4927007299270073, "percentage": 30.0, "elapsed_time": "0:17:51", "remaining_time": "0:41:41"}
|
| 46 |
+
{"current_steps": 46, "total_steps": 150, "loss": 1.475, "lr": 1.5826562086267956e-05, "epoch": 0.5036496350364964, "percentage": 30.67, "elapsed_time": "0:18:15", "remaining_time": "0:41:17"}
|
| 47 |
+
{"current_steps": 47, "total_steps": 150, "loss": 1.4634, "lr": 1.565392191239959e-05, "epoch": 0.5145985401459854, "percentage": 31.33, "elapsed_time": "0:18:39", "remaining_time": "0:40:53"}
|
| 48 |
+
{"current_steps": 48, "total_steps": 150, "loss": 1.4551, "lr": 1.5478768342496872e-05, "epoch": 0.5255474452554745, "percentage": 32.0, "elapsed_time": "0:19:03", "remaining_time": "0:40:29"}
|
| 49 |
+
{"current_steps": 49, "total_steps": 150, "loss": 1.4407, "lr": 1.5301179239376936e-05, "epoch": 0.5364963503649635, "percentage": 32.67, "elapsed_time": "0:19:26", "remaining_time": "0:40:05"}
|
| 50 |
+
{"current_steps": 50, "total_steps": 150, "loss": 1.4532, "lr": 1.512123354854955e-05, "epoch": 0.5474452554744526, "percentage": 33.33, "elapsed_time": "0:19:50", "remaining_time": "0:39:41"}
|
| 51 |
+
{"current_steps": 51, "total_steps": 150, "loss": 1.4549, "lr": 1.4939011263122635e-05, "epoch": 0.5583941605839416, "percentage": 34.0, "elapsed_time": "0:20:21", "remaining_time": "0:39:30"}
|
| 52 |
+
{"current_steps": 52, "total_steps": 150, "loss": 1.4429, "lr": 1.4754593388242117e-05, "epoch": 0.5693430656934306, "percentage": 34.67, "elapsed_time": "0:20:45", "remaining_time": "0:39:06"}
|
| 53 |
+
{"current_steps": 53, "total_steps": 150, "loss": 1.4331, "lr": 1.4568061905081874e-05, "epoch": 0.5802919708029197, "percentage": 35.33, "elapsed_time": "0:21:08", "remaining_time": "0:38:42"}
|
| 54 |
+
{"current_steps": 54, "total_steps": 150, "loss": 1.4411, "lr": 1.4379499734399797e-05, "epoch": 0.5912408759124088, "percentage": 36.0, "elapsed_time": "0:21:32", "remaining_time": "0:38:17"}
|
| 55 |
+
{"current_steps": 55, "total_steps": 150, "loss": 1.4412, "lr": 1.4188990699676186e-05, "epoch": 0.6021897810218978, "percentage": 36.67, "elapsed_time": "0:21:56", "remaining_time": "0:37:53"}
|
| 56 |
+
{"current_steps": 56, "total_steps": 150, "loss": 1.4259, "lr": 1.3996619489850822e-05, "epoch": 0.6131386861313869, "percentage": 37.33, "elapsed_time": "0:22:19", "remaining_time": "0:37:29"}
|
| 57 |
+
{"current_steps": 57, "total_steps": 150, "loss": 1.4256, "lr": 1.3802471621675337e-05, "epoch": 0.6240875912408759, "percentage": 38.0, "elapsed_time": "0:22:43", "remaining_time": "0:37:05"}
|
| 58 |
+
{"current_steps": 58, "total_steps": 150, "loss": 1.4301, "lr": 1.3606633401697557e-05, "epoch": 0.635036496350365, "percentage": 38.67, "elapsed_time": "0:23:07", "remaining_time": "0:36:40"}
|
| 59 |
+
{"current_steps": 59, "total_steps": 150, "loss": 1.4315, "lr": 1.340919188789477e-05, "epoch": 0.6459854014598541, "percentage": 39.33, "elapsed_time": "0:23:31", "remaining_time": "0:36:16"}
|
| 60 |
+
{"current_steps": 60, "total_steps": 150, "loss": 1.4214, "lr": 1.3210234850972966e-05, "epoch": 0.656934306569343, "percentage": 40.0, "elapsed_time": "0:23:55", "remaining_time": "0:35:52"}
|
| 61 |
+
{"current_steps": 61, "total_steps": 150, "loss": 1.4078, "lr": 1.300985073534919e-05, "epoch": 0.6678832116788321, "percentage": 40.67, "elapsed_time": "0:24:18", "remaining_time": "0:35:28"}
|
| 62 |
+
{"current_steps": 62, "total_steps": 150, "loss": 1.4176, "lr": 1.280812861983446e-05, "epoch": 0.6788321167883211, "percentage": 41.33, "elapsed_time": "0:24:42", "remaining_time": "0:35:04"}
|
| 63 |
+
{"current_steps": 63, "total_steps": 150, "loss": 1.4128, "lr": 1.2605158178034656e-05, "epoch": 0.6897810218978102, "percentage": 42.0, "elapsed_time": "0:25:06", "remaining_time": "0:34:40"}
|
| 64 |
+
{"current_steps": 64, "total_steps": 150, "loss": 1.3984, "lr": 1.2401029638486952e-05, "epoch": 0.7007299270072993, "percentage": 42.67, "elapsed_time": "0:25:30", "remaining_time": "0:34:16"}
|
| 65 |
+
{"current_steps": 65, "total_steps": 150, "loss": 1.4189, "lr": 1.219583374454963e-05, "epoch": 0.7116788321167883, "percentage": 43.33, "elapsed_time": "0:25:53", "remaining_time": "0:33:51"}
|
| 66 |
+
{"current_steps": 66, "total_steps": 150, "loss": 1.4168, "lr": 1.1989661714063e-05, "epoch": 0.7226277372262774, "percentage": 44.0, "elapsed_time": "0:26:17", "remaining_time": "0:33:27"}
|
| 67 |
+
{"current_steps": 67, "total_steps": 150, "loss": 1.4154, "lr": 1.1782605198799371e-05, "epoch": 0.7335766423357665, "percentage": 44.67, "elapsed_time": "0:26:41", "remaining_time": "0:33:03"}
|
| 68 |
+
{"current_steps": 68, "total_steps": 150, "loss": 1.4091, "lr": 1.157475624372018e-05, "epoch": 0.7445255474452555, "percentage": 45.33, "elapsed_time": "0:27:05", "remaining_time": "0:32:39"}
|
| 69 |
+
{"current_steps": 69, "total_steps": 150, "loss": 1.413, "lr": 1.1366207246058269e-05, "epoch": 0.7554744525547445, "percentage": 46.0, "elapsed_time": "0:27:28", "remaining_time": "0:32:15"}
|
| 70 |
+
{"current_steps": 70, "total_steps": 150, "loss": 1.4095, "lr": 1.1157050914243614e-05, "epoch": 0.7664233576642335, "percentage": 46.67, "elapsed_time": "0:27:52", "remaining_time": "0:31:51"}
|
| 71 |
+
{"current_steps": 71, "total_steps": 150, "loss": 1.4192, "lr": 1.0947380226690686e-05, "epoch": 0.7773722627737226, "percentage": 47.33, "elapsed_time": "0:28:16", "remaining_time": "0:31:27"}
|
| 72 |
+
{"current_steps": 72, "total_steps": 150, "loss": 1.4148, "lr": 1.0737288390465792e-05, "epoch": 0.7883211678832117, "percentage": 48.0, "elapsed_time": "0:28:40", "remaining_time": "0:31:03"}
|
| 73 |
+
{"current_steps": 73, "total_steps": 150, "loss": 1.4103, "lr": 1.0526868799852797e-05, "epoch": 0.7992700729927007, "percentage": 48.67, "elapsed_time": "0:29:03", "remaining_time": "0:30:39"}
|
| 74 |
+
{"current_steps": 74, "total_steps": 150, "loss": 1.4039, "lr": 1.031621499483559e-05, "epoch": 0.8102189781021898, "percentage": 49.33, "elapsed_time": "0:29:27", "remaining_time": "0:30:15"}
|
| 75 |
+
{"current_steps": 75, "total_steps": 150, "loss": 1.4038, "lr": 1.0105420619515798e-05, "epoch": 0.8211678832116789, "percentage": 50.0, "elapsed_time": "0:29:51", "remaining_time": "0:29:51"}
|
| 76 |
+
{"current_steps": 76, "total_steps": 150, "loss": 1.3998, "lr": 9.894579380484206e-06, "epoch": 0.8321167883211679, "percentage": 50.67, "elapsed_time": "0:30:15", "remaining_time": "0:29:27"}
|
| 77 |
+
{"current_steps": 77, "total_steps": 150, "loss": 1.3963, "lr": 9.683785005164412e-06, "epoch": 0.843065693430657, "percentage": 51.33, "elapsed_time": "0:30:39", "remaining_time": "0:29:03"}
|
| 78 |
+
{"current_steps": 78, "total_steps": 150, "loss": 1.4056, "lr": 9.473131200147205e-06, "epoch": 0.8540145985401459, "percentage": 52.0, "elapsed_time": "0:31:02", "remaining_time": "0:28:39"}
|
| 79 |
+
{"current_steps": 79, "total_steps": 150, "loss": 1.3924, "lr": 9.262711609534211e-06, "epoch": 0.864963503649635, "percentage": 52.67, "elapsed_time": "0:31:26", "remaining_time": "0:28:15"}
|
| 80 |
+
{"current_steps": 80, "total_steps": 150, "loss": 1.3882, "lr": 9.052619773309318e-06, "epoch": 0.8759124087591241, "percentage": 53.33, "elapsed_time": "0:31:50", "remaining_time": "0:27:51"}
|
| 81 |
+
{"current_steps": 81, "total_steps": 150, "loss": 1.4019, "lr": 8.842949085756389e-06, "epoch": 0.8868613138686131, "percentage": 54.0, "elapsed_time": "0:32:14", "remaining_time": "0:27:27"}
|
| 82 |
+
{"current_steps": 82, "total_steps": 150, "loss": 1.395, "lr": 8.633792753941733e-06, "epoch": 0.8978102189781022, "percentage": 54.67, "elapsed_time": "0:32:37", "remaining_time": "0:27:03"}
|
| 83 |
+
{"current_steps": 83, "total_steps": 150, "loss": 1.3888, "lr": 8.425243756279824e-06, "epoch": 0.9087591240875912, "percentage": 55.33, "elapsed_time": "0:33:01", "remaining_time": "0:26:39"}
|
| 84 |
+
{"current_steps": 84, "total_steps": 150, "loss": 1.3986, "lr": 8.217394801200632e-06, "epoch": 0.9197080291970803, "percentage": 56.0, "elapsed_time": "0:33:25", "remaining_time": "0:26:15"}
|
| 85 |
+
{"current_steps": 85, "total_steps": 150, "loss": 1.3744, "lr": 8.010338285937006e-06, "epoch": 0.9306569343065694, "percentage": 56.67, "elapsed_time": "0:33:49", "remaining_time": "0:25:51"}
|
| 86 |
+
{"current_steps": 86, "total_steps": 150, "loss": 1.3805, "lr": 7.804166255450372e-06, "epoch": 0.9416058394160584, "percentage": 57.33, "elapsed_time": "0:34:12", "remaining_time": "0:25:27"}
|
| 87 |
+
{"current_steps": 87, "total_steps": 150, "loss": 1.3812, "lr": 7.598970361513052e-06, "epoch": 0.9525547445255474, "percentage": 58.0, "elapsed_time": "0:34:36", "remaining_time": "0:25:03"}
|
| 88 |
+
{"current_steps": 88, "total_steps": 150, "loss": 1.3908, "lr": 7.394841821965345e-06, "epoch": 0.9635036496350365, "percentage": 58.67, "elapsed_time": "0:35:00", "remaining_time": "0:24:39"}
|
| 89 |
+
{"current_steps": 89, "total_steps": 150, "loss": 1.3952, "lr": 7.191871380165538e-06, "epoch": 0.9744525547445255, "percentage": 59.33, "elapsed_time": "0:35:24", "remaining_time": "0:24:15"}
|
| 90 |
+
{"current_steps": 90, "total_steps": 150, "loss": 1.3853, "lr": 6.990149264650814e-06, "epoch": 0.9854014598540146, "percentage": 60.0, "elapsed_time": "0:35:48", "remaining_time": "0:23:52"}
|
| 91 |
+
{"current_steps": 91, "total_steps": 150, "loss": 1.3802, "lr": 6.789765149027039e-06, "epoch": 0.9963503649635036, "percentage": 60.67, "elapsed_time": "0:36:11", "remaining_time": "0:23:28"}
|
| 92 |
+
{"current_steps": 92, "total_steps": 150, "loss": 1.3781, "lr": 6.590808112105232e-06, "epoch": 1.0072992700729928, "percentage": 61.33, "elapsed_time": "0:36:36", "remaining_time": "0:23:04"}
|
| 93 |
+
{"current_steps": 93, "total_steps": 150, "loss": 1.3737, "lr": 6.3933665983024465e-06, "epoch": 1.0182481751824817, "percentage": 62.0, "elapsed_time": "0:37:00", "remaining_time": "0:22:40"}
|
| 94 |
+
{"current_steps": 94, "total_steps": 150, "loss": 1.3829, "lr": 6.197528378324664e-06, "epoch": 1.0291970802919708, "percentage": 62.67, "elapsed_time": "0:37:23", "remaining_time": "0:22:16"}
|
| 95 |
+
{"current_steps": 95, "total_steps": 150, "loss": 1.3637, "lr": 6.003380510149179e-06, "epoch": 1.0401459854014599, "percentage": 63.33, "elapsed_time": "0:37:47", "remaining_time": "0:21:52"}
|
| 96 |
+
{"current_steps": 96, "total_steps": 150, "loss": 1.3671, "lr": 5.8110093003238175e-06, "epoch": 1.051094890510949, "percentage": 64.0, "elapsed_time": "0:38:11", "remaining_time": "0:21:28"}
|
| 97 |
+
{"current_steps": 97, "total_steps": 150, "loss": 1.3736, "lr": 5.620500265600206e-06, "epoch": 1.062043795620438, "percentage": 64.67, "elapsed_time": "0:38:35", "remaining_time": "0:21:04"}
|
| 98 |
+
{"current_steps": 98, "total_steps": 150, "loss": 1.3671, "lr": 5.431938094918132e-06, "epoch": 1.072992700729927, "percentage": 65.33, "elapsed_time": "0:38:58", "remaining_time": "0:20:41"}
|
| 99 |
+
{"current_steps": 99, "total_steps": 150, "loss": 1.3565, "lr": 5.245406611757882e-06, "epoch": 1.083941605839416, "percentage": 66.0, "elapsed_time": "0:39:22", "remaining_time": "0:20:17"}
|
| 100 |
+
{"current_steps": 100, "total_steps": 150, "loss": 1.3724, "lr": 5.060988736877366e-06, "epoch": 1.094890510948905, "percentage": 66.67, "elapsed_time": "0:39:46", "remaining_time": "0:19:53"}
|
| 101 |
+
{"current_steps": 101, "total_steps": 150, "loss": 1.3812, "lr": 4.878766451450451e-06, "epoch": 1.1058394160583942, "percentage": 67.33, "elapsed_time": "0:40:20", "remaining_time": "0:19:34"}
|
| 102 |
+
{"current_steps": 102, "total_steps": 150, "loss": 1.3809, "lr": 4.698820760623064e-06, "epoch": 1.1167883211678833, "percentage": 68.0, "elapsed_time": "0:40:44", "remaining_time": "0:19:10"}
|
| 103 |
+
{"current_steps": 103, "total_steps": 150, "loss": 1.3671, "lr": 4.5212316575031325e-06, "epoch": 1.1277372262773722, "percentage": 68.67, "elapsed_time": "0:41:07", "remaining_time": "0:18:46"}
|
| 104 |
+
{"current_steps": 104, "total_steps": 150, "loss": 1.3706, "lr": 4.346078087600411e-06, "epoch": 1.1386861313868613, "percentage": 69.33, "elapsed_time": "0:41:31", "remaining_time": "0:18:22"}
|
| 105 |
+
{"current_steps": 105, "total_steps": 150, "loss": 1.3729, "lr": 4.173437913732048e-06, "epoch": 1.1496350364963503, "percentage": 70.0, "elapsed_time": "0:41:55", "remaining_time": "0:17:58"}
|
| 106 |
+
{"current_steps": 106, "total_steps": 150, "loss": 1.3658, "lr": 4.003387881409397e-06, "epoch": 1.1605839416058394, "percentage": 70.67, "elapsed_time": "0:42:19", "remaining_time": "0:17:34"}
|
| 107 |
+
{"current_steps": 107, "total_steps": 150, "loss": 1.379, "lr": 3.836003584721577e-06, "epoch": 1.1715328467153285, "percentage": 71.33, "elapsed_time": "0:42:42", "remaining_time": "0:17:09"}
|
| 108 |
+
{"current_steps": 108, "total_steps": 150, "loss": 1.3731, "lr": 3.6713594327308343e-06, "epoch": 1.1824817518248176, "percentage": 72.0, "elapsed_time": "0:43:06", "remaining_time": "0:16:45"}
|
| 109 |
+
{"current_steps": 109, "total_steps": 150, "loss": 1.3543, "lr": 3.509528616394716e-06, "epoch": 1.1934306569343065, "percentage": 72.67, "elapsed_time": "0:43:30", "remaining_time": "0:16:21"}
|
| 110 |
+
{"current_steps": 110, "total_steps": 150, "loss": 1.3717, "lr": 3.3505830760297543e-06, "epoch": 1.2043795620437956, "percentage": 73.33, "elapsed_time": "0:43:54", "remaining_time": "0:15:57"}
|
| 111 |
+
{"current_steps": 111, "total_steps": 150, "loss": 1.367, "lr": 3.1945934693310897e-06, "epoch": 1.2153284671532847, "percentage": 74.0, "elapsed_time": "0:44:18", "remaining_time": "0:15:33"}
|
| 112 |
+
{"current_steps": 112, "total_steps": 150, "loss": 1.3491, "lr": 3.0416291399622834e-06, "epoch": 1.2262773722627738, "percentage": 74.67, "elapsed_time": "0:44:41", "remaining_time": "0:15:09"}
|
| 113 |
+
{"current_steps": 113, "total_steps": 150, "loss": 1.3851, "lr": 2.891758086729253e-06, "epoch": 1.2372262773722627, "percentage": 75.33, "elapsed_time": "0:45:05", "remaining_time": "0:14:45"}
|
| 114 |
+
{"current_steps": 114, "total_steps": 150, "loss": 1.3635, "lr": 2.7450469333520856e-06, "epoch": 1.2481751824817517, "percentage": 76.0, "elapsed_time": "0:45:29", "remaining_time": "0:14:21"}
|
| 115 |
+
{"current_steps": 115, "total_steps": 150, "loss": 1.3653, "lr": 2.6015608988480956e-06, "epoch": 1.2591240875912408, "percentage": 76.67, "elapsed_time": "0:45:53", "remaining_time": "0:13:57"}
|
| 116 |
+
{"current_steps": 116, "total_steps": 150, "loss": 1.3577, "lr": 2.4613637685393433e-06, "epoch": 1.27007299270073, "percentage": 77.33, "elapsed_time": "0:46:16", "remaining_time": "0:13:33"}
|
| 117 |
+
{"current_steps": 117, "total_steps": 150, "loss": 1.3593, "lr": 2.324517865697501e-06, "epoch": 1.281021897810219, "percentage": 78.0, "elapsed_time": "0:46:40", "remaining_time": "0:13:09"}
|
| 118 |
+
{"current_steps": 118, "total_steps": 150, "loss": 1.3629, "lr": 2.19108402383864e-06, "epoch": 1.2919708029197081, "percentage": 78.67, "elapsed_time": "0:47:04", "remaining_time": "0:12:45"}
|
| 119 |
+
{"current_steps": 119, "total_steps": 150, "loss": 1.3708, "lr": 2.06112155968028e-06, "epoch": 1.3029197080291972, "percentage": 79.33, "elapsed_time": "0:47:28", "remaining_time": "0:12:21"}
|
| 120 |
+
{"current_steps": 120, "total_steps": 150, "loss": 1.3667, "lr": 1.9346882467727323e-06, "epoch": 1.313868613138686, "percentage": 80.0, "elapsed_time": "0:47:51", "remaining_time": "0:11:57"}
|
| 121 |
+
{"current_steps": 121, "total_steps": 150, "loss": 1.3668, "lr": 1.811840289816409e-06, "epoch": 1.3248175182481752, "percentage": 80.67, "elapsed_time": "0:48:15", "remaining_time": "0:11:33"}
|
| 122 |
+
{"current_steps": 122, "total_steps": 150, "loss": 1.3587, "lr": 1.6926322996765899e-06, "epoch": 1.3357664233576643, "percentage": 81.33, "elapsed_time": "0:48:39", "remaining_time": "0:11:10"}
|
| 123 |
+
{"current_steps": 123, "total_steps": 150, "loss": 1.3681, "lr": 1.5771172691066793e-06, "epoch": 1.3467153284671534, "percentage": 82.0, "elapsed_time": "0:49:03", "remaining_time": "0:10:46"}
|
| 124 |
+
{"current_steps": 124, "total_steps": 150, "loss": 1.3797, "lr": 1.4653465491908003e-06, "epoch": 1.3576642335766422, "percentage": 82.67, "elapsed_time": "0:49:26", "remaining_time": "0:10:22"}
|
| 125 |
+
{"current_steps": 125, "total_steps": 150, "loss": 1.3734, "lr": 1.3573698265161683e-06, "epoch": 1.3686131386861313, "percentage": 83.33, "elapsed_time": "0:49:50", "remaining_time": "0:09:58"}
|
| 126 |
+
{"current_steps": 126, "total_steps": 150, "loss": 1.3712, "lr": 1.2532351010853916e-06, "epoch": 1.3795620437956204, "percentage": 84.0, "elapsed_time": "0:50:14", "remaining_time": "0:09:34"}
|
| 127 |
+
{"current_steps": 127, "total_steps": 150, "loss": 1.3698, "lr": 1.152988664978556e-06, "epoch": 1.3905109489051095, "percentage": 84.67, "elapsed_time": "0:50:38", "remaining_time": "0:09:10"}
|
| 128 |
+
{"current_steps": 128, "total_steps": 150, "loss": 1.3621, "lr": 1.0566750817745076e-06, "epoch": 1.4014598540145986, "percentage": 85.33, "elapsed_time": "0:51:01", "remaining_time": "0:08:46"}
|
| 129 |
+
{"current_steps": 129, "total_steps": 150, "loss": 1.3715, "lr": 9.6433716674057e-07, "epoch": 1.4124087591240877, "percentage": 86.0, "elapsed_time": "0:51:25", "remaining_time": "0:08:22"}
|
| 130 |
+
{"current_steps": 130, "total_steps": 150, "loss": 1.368, "lr": 8.760159677994174e-07, "epoch": 1.4233576642335766, "percentage": 86.67, "elapsed_time": "0:51:49", "remaining_time": "0:07:58"}
|
| 131 |
+
{"current_steps": 131, "total_steps": 150, "loss": 1.3485, "lr": 7.91750747281621e-07, "epoch": 1.4343065693430657, "percentage": 87.33, "elapsed_time": "0:52:13", "remaining_time": "0:07:34"}
|
| 132 |
+
{"current_steps": 132, "total_steps": 150, "loss": 1.3564, "lr": 7.115789644719728e-07, "epoch": 1.4452554744525548, "percentage": 88.0, "elapsed_time": "0:52:36", "remaining_time": "0:07:10"}
|
| 133 |
+
{"current_steps": 133, "total_steps": 150, "loss": 1.3532, "lr": 6.355362589573078e-07, "epoch": 1.4562043795620438, "percentage": 88.67, "elapsed_time": "0:53:00", "remaining_time": "0:06:46"}
|
| 134 |
+
{"current_steps": 134, "total_steps": 150, "loss": 1.3701, "lr": 5.636564347832907e-07, "epoch": 1.4671532846715327, "percentage": 89.33, "elapsed_time": "0:53:24", "remaining_time": "0:06:22"}
|
| 135 |
+
{"current_steps": 135, "total_steps": 150, "loss": 1.3676, "lr": 4.95971445427137e-07, "epoch": 1.4781021897810218, "percentage": 90.0, "elapsed_time": "0:53:48", "remaining_time": "0:05:58"}
|
| 136 |
+
{"current_steps": 136, "total_steps": 150, "loss": 1.3556, "lr": 4.3251137959302023e-07, "epoch": 1.489051094890511, "percentage": 90.67, "elapsed_time": "0:54:11", "remaining_time": "0:05:34"}
|
| 137 |
+
{"current_steps": 137, "total_steps": 150, "loss": 1.3695, "lr": 3.733044478364234e-07, "epoch": 1.5, "percentage": 91.33, "elapsed_time": "0:54:35", "remaining_time": "0:05:10"}
|
| 138 |
+
{"current_steps": 138, "total_steps": 150, "loss": 1.3615, "lr": 3.1837697002341293e-07, "epoch": 1.510948905109489, "percentage": 92.0, "elapsed_time": "0:54:59", "remaining_time": "0:04:46"}
|
| 139 |
+
{"current_steps": 139, "total_steps": 150, "loss": 1.363, "lr": 2.677533636303964e-07, "epoch": 1.5218978102189782, "percentage": 92.67, "elapsed_time": "0:55:23", "remaining_time": "0:04:22"}
|
| 140 |
+
{"current_steps": 140, "total_steps": 150, "loss": 1.3669, "lr": 2.214561328895748e-07, "epoch": 1.5328467153284673, "percentage": 93.33, "elapsed_time": "0:55:46", "remaining_time": "0:03:59"}
|
| 141 |
+
{"current_steps": 141, "total_steps": 150, "loss": 1.355, "lr": 1.7950585878489856e-07, "epoch": 1.5437956204379562, "percentage": 94.0, "elapsed_time": "0:56:10", "remaining_time": "0:03:35"}
|
| 142 |
+
{"current_steps": 142, "total_steps": 150, "loss": 1.3725, "lr": 1.419211899029971e-07, "epoch": 1.5547445255474452, "percentage": 94.67, "elapsed_time": "0:56:34", "remaining_time": "0:03:11"}
|
| 143 |
+
{"current_steps": 143, "total_steps": 150, "loss": 1.3773, "lr": 1.0871883414312778e-07, "epoch": 1.5656934306569343, "percentage": 95.33, "elapsed_time": "0:56:58", "remaining_time": "0:02:47"}
|
| 144 |
+
{"current_steps": 144, "total_steps": 150, "loss": 1.351, "lr": 7.99135512898408e-08, "epoch": 1.5766423357664232, "percentage": 96.0, "elapsed_time": "0:57:21", "remaining_time": "0:02:23"}
|
| 145 |
+
{"current_steps": 145, "total_steps": 150, "loss": 1.3626, "lr": 5.55181464516652e-08, "epoch": 1.5875912408759123, "percentage": 96.67, "elapsed_time": "0:57:45", "remaining_time": "0:01:59"}
|
| 146 |
+
{"current_steps": 146, "total_steps": 150, "loss": 1.375, "lr": 3.554346436871581e-08, "epoch": 1.5985401459854014, "percentage": 97.33, "elapsed_time": "0:58:09", "remaining_time": "0:01:35"}
|
| 147 |
+
{"current_steps": 147, "total_steps": 150, "loss": 1.3599, "lr": 1.9998384591773945e-08, "epoch": 1.6094890510948905, "percentage": 98.0, "elapsed_time": "0:58:33", "remaining_time": "0:01:11"}
|
| 148 |
+
{"current_steps": 148, "total_steps": 150, "loss": 1.3476, "lr": 8.889817534969425e-09, "epoch": 1.6204379562043796, "percentage": 98.67, "elapsed_time": "0:58:56", "remaining_time": "0:00:47"}
|
| 149 |
+
{"current_steps": 149, "total_steps": 150, "loss": 1.3655, "lr": 2.222701403818972e-09, "epoch": 1.6313868613138687, "percentage": 99.33, "elapsed_time": "0:59:20", "remaining_time": "0:00:23"}
|
| 150 |
+
{"current_steps": 150, "total_steps": 150, "loss": 1.3705, "lr": 0.0, "epoch": 1.6423357664233578, "percentage": 100.0, "elapsed_time": "0:59:44", "remaining_time": "0:00:00"}
|
| 151 |
+
{"current_steps": 150, "total_steps": 150, "epoch": 1.6423357664233578, "percentage": 100.0, "elapsed_time": "0:59:53", "remaining_time": "0:00:00"}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1092 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.6423357664233578,
|
| 5 |
+
"eval_steps": 50,
|
| 6 |
+
"global_step": 150,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010948905109489052,
|
| 13 |
+
"grad_norm": 1307753144.573482,
|
| 14 |
+
"learning_rate": 2e-05,
|
| 15 |
+
"loss": 5.8607,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.021897810218978103,
|
| 20 |
+
"grad_norm": 834902425.6972803,
|
| 21 |
+
"learning_rate": 1.999777729859618e-05,
|
| 22 |
+
"loss": 5.8686,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.032846715328467155,
|
| 27 |
+
"grad_norm": 808.9016350206678,
|
| 28 |
+
"learning_rate": 1.9991110182465032e-05,
|
| 29 |
+
"loss": 5.8515,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.043795620437956206,
|
| 34 |
+
"grad_norm": 56.34317531908879,
|
| 35 |
+
"learning_rate": 1.9980001615408228e-05,
|
| 36 |
+
"loss": 4.5812,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.05474452554744526,
|
| 41 |
+
"grad_norm": 26.93030207315231,
|
| 42 |
+
"learning_rate": 1.9964456535631287e-05,
|
| 43 |
+
"loss": 3.9891,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06569343065693431,
|
| 48 |
+
"grad_norm": 20.344543886007664,
|
| 49 |
+
"learning_rate": 1.9944481853548335e-05,
|
| 50 |
+
"loss": 3.3771,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07664233576642336,
|
| 55 |
+
"grad_norm": 12.6575820414914,
|
| 56 |
+
"learning_rate": 1.9920086448710162e-05,
|
| 57 |
+
"loss": 2.935,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.08759124087591241,
|
| 62 |
+
"grad_norm": 7.990285688703706,
|
| 63 |
+
"learning_rate": 1.9891281165856876e-05,
|
| 64 |
+
"loss": 2.4909,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09854014598540146,
|
| 69 |
+
"grad_norm": 4.526434637254955,
|
| 70 |
+
"learning_rate": 1.9858078810097004e-05,
|
| 71 |
+
"loss": 2.3074,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10948905109489052,
|
| 76 |
+
"grad_norm": 3.432415062463592,
|
| 77 |
+
"learning_rate": 1.98204941412151e-05,
|
| 78 |
+
"loss": 2.168,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.12043795620437957,
|
| 83 |
+
"grad_norm": 2.8983437684817286,
|
| 84 |
+
"learning_rate": 1.9778543867110428e-05,
|
| 85 |
+
"loss": 2.0632,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.13138686131386862,
|
| 90 |
+
"grad_norm": 2.1547654165559886,
|
| 91 |
+
"learning_rate": 1.9732246636369605e-05,
|
| 92 |
+
"loss": 1.9709,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.14233576642335766,
|
| 97 |
+
"grad_norm": 1.5632105837184251,
|
| 98 |
+
"learning_rate": 1.968162302997659e-05,
|
| 99 |
+
"loss": 1.9209,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.15328467153284672,
|
| 104 |
+
"grad_norm": 1.433284775693193,
|
| 105 |
+
"learning_rate": 1.962669555216358e-05,
|
| 106 |
+
"loss": 1.8743,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.16423357664233576,
|
| 111 |
+
"grad_norm": 1.3058467215685399,
|
| 112 |
+
"learning_rate": 1.9567488620406984e-05,
|
| 113 |
+
"loss": 1.8264,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.17518248175182483,
|
| 118 |
+
"grad_norm": 1.0822327666073748,
|
| 119 |
+
"learning_rate": 1.9504028554572865e-05,
|
| 120 |
+
"loss": 1.7879,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.18613138686131386,
|
| 125 |
+
"grad_norm": 1.3231585817643108,
|
| 126 |
+
"learning_rate": 1.943634356521671e-05,
|
| 127 |
+
"loss": 1.7542,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.19708029197080293,
|
| 132 |
+
"grad_norm": 1.1242088253564448,
|
| 133 |
+
"learning_rate": 1.9364463741042694e-05,
|
| 134 |
+
"loss": 1.7429,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.20802919708029197,
|
| 139 |
+
"grad_norm": 1.0033003244965946,
|
| 140 |
+
"learning_rate": 1.928842103552803e-05,
|
| 141 |
+
"loss": 1.707,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.21897810218978103,
|
| 146 |
+
"grad_norm": 0.7481922790889302,
|
| 147 |
+
"learning_rate": 1.920824925271838e-05,
|
| 148 |
+
"loss": 1.6844,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.22992700729927007,
|
| 153 |
+
"grad_norm": 0.6188106546280582,
|
| 154 |
+
"learning_rate": 1.9123984032200586e-05,
|
| 155 |
+
"loss": 1.6516,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.24087591240875914,
|
| 160 |
+
"grad_norm": 0.963664679847666,
|
| 161 |
+
"learning_rate": 1.9035662833259433e-05,
|
| 162 |
+
"loss": 1.6474,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2518248175182482,
|
| 167 |
+
"grad_norm": 0.8398357524796255,
|
| 168 |
+
"learning_rate": 1.8943324918225495e-05,
|
| 169 |
+
"loss": 1.6311,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.26277372262773724,
|
| 174 |
+
"grad_norm": 0.5601109858987311,
|
| 175 |
+
"learning_rate": 1.8847011335021447e-05,
|
| 176 |
+
"loss": 1.6035,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.2737226277372263,
|
| 181 |
+
"grad_norm": 0.5532013562786996,
|
| 182 |
+
"learning_rate": 1.874676489891461e-05,
|
| 183 |
+
"loss": 1.6014,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.2846715328467153,
|
| 188 |
+
"grad_norm": 0.4738300388302612,
|
| 189 |
+
"learning_rate": 1.8642630173483832e-05,
|
| 190 |
+
"loss": 1.5806,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.2956204379562044,
|
| 195 |
+
"grad_norm": 0.5790284650469533,
|
| 196 |
+
"learning_rate": 1.85346534508092e-05,
|
| 197 |
+
"loss": 1.5736,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.30656934306569344,
|
| 202 |
+
"grad_norm": 0.48957818083767896,
|
| 203 |
+
"learning_rate": 1.8422882730893323e-05,
|
| 204 |
+
"loss": 1.5642,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3175182481751825,
|
| 209 |
+
"grad_norm": 0.4481599979449311,
|
| 210 |
+
"learning_rate": 1.8307367700323412e-05,
|
| 211 |
+
"loss": 1.5504,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3284671532846715,
|
| 216 |
+
"grad_norm": 0.47622632947338317,
|
| 217 |
+
"learning_rate": 1.8188159710183595e-05,
|
| 218 |
+
"loss": 1.5498,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.33941605839416056,
|
| 223 |
+
"grad_norm": 0.4579285162469448,
|
| 224 |
+
"learning_rate": 1.8065311753227272e-05,
|
| 225 |
+
"loss": 1.5283,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.35036496350364965,
|
| 230 |
+
"grad_norm": 0.5718221087613807,
|
| 231 |
+
"learning_rate": 1.7938878440319722e-05,
|
| 232 |
+
"loss": 1.5455,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.3613138686131387,
|
| 237 |
+
"grad_norm": 0.6248596987677889,
|
| 238 |
+
"learning_rate": 1.7808915976161364e-05,
|
| 239 |
+
"loss": 1.4982,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.3722627737226277,
|
| 244 |
+
"grad_norm": 0.6344703083428059,
|
| 245 |
+
"learning_rate": 1.7675482134302503e-05,
|
| 246 |
+
"loss": 1.5187,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.38321167883211676,
|
| 251 |
+
"grad_norm": 0.4293094757287632,
|
| 252 |
+
"learning_rate": 1.753863623146066e-05,
|
| 253 |
+
"loss": 1.5082,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.39416058394160586,
|
| 258 |
+
"grad_norm": 0.48618342412405535,
|
| 259 |
+
"learning_rate": 1.7398439101151908e-05,
|
| 260 |
+
"loss": 1.5055,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4051094890510949,
|
| 265 |
+
"grad_norm": 0.6352381109587321,
|
| 266 |
+
"learning_rate": 1.7254953066647915e-05,
|
| 267 |
+
"loss": 1.5039,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.41605839416058393,
|
| 272 |
+
"grad_norm": 0.4611200601887447,
|
| 273 |
+
"learning_rate": 1.710824191327075e-05,
|
| 274 |
+
"loss": 1.4877,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.42700729927007297,
|
| 279 |
+
"grad_norm": 0.4504118609016994,
|
| 280 |
+
"learning_rate": 1.695837086003772e-05,
|
| 281 |
+
"loss": 1.4882,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.43795620437956206,
|
| 286 |
+
"grad_norm": 0.5887114679045612,
|
| 287 |
+
"learning_rate": 1.680540653066891e-05,
|
| 288 |
+
"loss": 1.4765,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.4489051094890511,
|
| 293 |
+
"grad_norm": 0.5350891546396978,
|
| 294 |
+
"learning_rate": 1.6649416923970248e-05,
|
| 295 |
+
"loss": 1.4868,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.45985401459854014,
|
| 300 |
+
"grad_norm": 0.5294860342681897,
|
| 301 |
+
"learning_rate": 1.649047138360529e-05,
|
| 302 |
+
"loss": 1.4797,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.4708029197080292,
|
| 307 |
+
"grad_norm": 0.5902917267127649,
|
| 308 |
+
"learning_rate": 1.632864056726917e-05,
|
| 309 |
+
"loss": 1.476,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.48175182481751827,
|
| 314 |
+
"grad_norm": 0.484973856854526,
|
| 315 |
+
"learning_rate": 1.6163996415278423e-05,
|
| 316 |
+
"loss": 1.4471,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.4927007299270073,
|
| 321 |
+
"grad_norm": 0.5629759129619559,
|
| 322 |
+
"learning_rate": 1.5996612118590604e-05,
|
| 323 |
+
"loss": 1.4603,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5036496350364964,
|
| 328 |
+
"grad_norm": 0.6761133066832582,
|
| 329 |
+
"learning_rate": 1.5826562086267956e-05,
|
| 330 |
+
"loss": 1.475,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.5145985401459854,
|
| 335 |
+
"grad_norm": 0.9155358687424147,
|
| 336 |
+
"learning_rate": 1.565392191239959e-05,
|
| 337 |
+
"loss": 1.4634,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.5255474452554745,
|
| 342 |
+
"grad_norm": 0.8771334841836861,
|
| 343 |
+
"learning_rate": 1.5478768342496872e-05,
|
| 344 |
+
"loss": 1.4551,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.5364963503649635,
|
| 349 |
+
"grad_norm": 0.5454958307781915,
|
| 350 |
+
"learning_rate": 1.5301179239376936e-05,
|
| 351 |
+
"loss": 1.4407,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.5474452554744526,
|
| 356 |
+
"grad_norm": 0.7936525781349211,
|
| 357 |
+
"learning_rate": 1.512123354854955e-05,
|
| 358 |
+
"loss": 1.4532,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.5583941605839416,
|
| 363 |
+
"grad_norm": 0.554879116460015,
|
| 364 |
+
"learning_rate": 1.4939011263122635e-05,
|
| 365 |
+
"loss": 1.4549,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.5693430656934306,
|
| 370 |
+
"grad_norm": 0.5747684461689259,
|
| 371 |
+
"learning_rate": 1.4754593388242117e-05,
|
| 372 |
+
"loss": 1.4429,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.5802919708029197,
|
| 377 |
+
"grad_norm": 0.48704829694816915,
|
| 378 |
+
"learning_rate": 1.4568061905081874e-05,
|
| 379 |
+
"loss": 1.4331,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.5912408759124088,
|
| 384 |
+
"grad_norm": 0.5327887513190984,
|
| 385 |
+
"learning_rate": 1.4379499734399797e-05,
|
| 386 |
+
"loss": 1.4411,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.6021897810218978,
|
| 391 |
+
"grad_norm": 0.4187125331546464,
|
| 392 |
+
"learning_rate": 1.4188990699676186e-05,
|
| 393 |
+
"loss": 1.4412,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.6131386861313869,
|
| 398 |
+
"grad_norm": 0.4714230775749121,
|
| 399 |
+
"learning_rate": 1.3996619489850822e-05,
|
| 400 |
+
"loss": 1.4259,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.6240875912408759,
|
| 405 |
+
"grad_norm": 0.4381564152922412,
|
| 406 |
+
"learning_rate": 1.3802471621675337e-05,
|
| 407 |
+
"loss": 1.4256,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.635036496350365,
|
| 412 |
+
"grad_norm": 0.4231730743091842,
|
| 413 |
+
"learning_rate": 1.3606633401697557e-05,
|
| 414 |
+
"loss": 1.4301,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.6459854014598541,
|
| 419 |
+
"grad_norm": 0.4418906525966298,
|
| 420 |
+
"learning_rate": 1.340919188789477e-05,
|
| 421 |
+
"loss": 1.4315,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.656934306569343,
|
| 426 |
+
"grad_norm": 0.4185857819687774,
|
| 427 |
+
"learning_rate": 1.3210234850972966e-05,
|
| 428 |
+
"loss": 1.4214,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.6678832116788321,
|
| 433 |
+
"grad_norm": 0.4532947265015192,
|
| 434 |
+
"learning_rate": 1.300985073534919e-05,
|
| 435 |
+
"loss": 1.4078,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.6788321167883211,
|
| 440 |
+
"grad_norm": 0.40915663629636617,
|
| 441 |
+
"learning_rate": 1.280812861983446e-05,
|
| 442 |
+
"loss": 1.4176,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.6897810218978102,
|
| 447 |
+
"grad_norm": 0.439973666230761,
|
| 448 |
+
"learning_rate": 1.2605158178034656e-05,
|
| 449 |
+
"loss": 1.4128,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.7007299270072993,
|
| 454 |
+
"grad_norm": 0.388332392545993,
|
| 455 |
+
"learning_rate": 1.2401029638486952e-05,
|
| 456 |
+
"loss": 1.3984,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.7116788321167883,
|
| 461 |
+
"grad_norm": 0.4186724653879167,
|
| 462 |
+
"learning_rate": 1.219583374454963e-05,
|
| 463 |
+
"loss": 1.4189,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.7226277372262774,
|
| 468 |
+
"grad_norm": 0.3939338896067437,
|
| 469 |
+
"learning_rate": 1.1989661714063e-05,
|
| 470 |
+
"loss": 1.4168,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.7335766423357665,
|
| 475 |
+
"grad_norm": 0.40861994135430135,
|
| 476 |
+
"learning_rate": 1.1782605198799371e-05,
|
| 477 |
+
"loss": 1.4154,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.7445255474452555,
|
| 482 |
+
"grad_norm": 0.43804898800624004,
|
| 483 |
+
"learning_rate": 1.157475624372018e-05,
|
| 484 |
+
"loss": 1.4091,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.7554744525547445,
|
| 489 |
+
"grad_norm": 0.3713868282180065,
|
| 490 |
+
"learning_rate": 1.1366207246058269e-05,
|
| 491 |
+
"loss": 1.413,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.7664233576642335,
|
| 496 |
+
"grad_norm": 0.40050112634034546,
|
| 497 |
+
"learning_rate": 1.1157050914243614e-05,
|
| 498 |
+
"loss": 1.4095,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.7773722627737226,
|
| 503 |
+
"grad_norm": 0.5545190963222433,
|
| 504 |
+
"learning_rate": 1.0947380226690686e-05,
|
| 505 |
+
"loss": 1.4192,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.7883211678832117,
|
| 510 |
+
"grad_norm": 0.6373132292613183,
|
| 511 |
+
"learning_rate": 1.0737288390465792e-05,
|
| 512 |
+
"loss": 1.4148,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.7992700729927007,
|
| 517 |
+
"grad_norm": 0.9691724096220773,
|
| 518 |
+
"learning_rate": 1.0526868799852797e-05,
|
| 519 |
+
"loss": 1.4103,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.8102189781021898,
|
| 524 |
+
"grad_norm": 1.523100464337697,
|
| 525 |
+
"learning_rate": 1.031621499483559e-05,
|
| 526 |
+
"loss": 1.4039,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.8211678832116789,
|
| 531 |
+
"grad_norm": 0.4509061687013781,
|
| 532 |
+
"learning_rate": 1.0105420619515798e-05,
|
| 533 |
+
"loss": 1.4038,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.8321167883211679,
|
| 538 |
+
"grad_norm": 1.256231131490188,
|
| 539 |
+
"learning_rate": 9.894579380484206e-06,
|
| 540 |
+
"loss": 1.3998,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.843065693430657,
|
| 545 |
+
"grad_norm": 1.2122171916647917,
|
| 546 |
+
"learning_rate": 9.683785005164412e-06,
|
| 547 |
+
"loss": 1.3963,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.8540145985401459,
|
| 552 |
+
"grad_norm": 0.5090380253354413,
|
| 553 |
+
"learning_rate": 9.473131200147205e-06,
|
| 554 |
+
"loss": 1.4056,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.864963503649635,
|
| 559 |
+
"grad_norm": 1.1907187887866326,
|
| 560 |
+
"learning_rate": 9.262711609534211e-06,
|
| 561 |
+
"loss": 1.3924,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.8759124087591241,
|
| 566 |
+
"grad_norm": 0.8697170065867255,
|
| 567 |
+
"learning_rate": 9.052619773309318e-06,
|
| 568 |
+
"loss": 1.3882,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.8868613138686131,
|
| 573 |
+
"grad_norm": 2.8000591331483067,
|
| 574 |
+
"learning_rate": 8.842949085756389e-06,
|
| 575 |
+
"loss": 1.4019,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8978102189781022,
|
| 580 |
+
"grad_norm": 0.8565040683588356,
|
| 581 |
+
"learning_rate": 8.633792753941733e-06,
|
| 582 |
+
"loss": 1.395,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.9087591240875912,
|
| 587 |
+
"grad_norm": 0.6637373975286237,
|
| 588 |
+
"learning_rate": 8.425243756279824e-06,
|
| 589 |
+
"loss": 1.3888,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.9197080291970803,
|
| 594 |
+
"grad_norm": 0.5820949034093401,
|
| 595 |
+
"learning_rate": 8.217394801200632e-06,
|
| 596 |
+
"loss": 1.3986,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.9306569343065694,
|
| 601 |
+
"grad_norm": 0.6183959176435854,
|
| 602 |
+
"learning_rate": 8.010338285937006e-06,
|
| 603 |
+
"loss": 1.3744,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.9416058394160584,
|
| 608 |
+
"grad_norm": 0.4885651952245995,
|
| 609 |
+
"learning_rate": 7.804166255450372e-06,
|
| 610 |
+
"loss": 1.3805,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.9525547445255474,
|
| 615 |
+
"grad_norm": 0.6428503364685796,
|
| 616 |
+
"learning_rate": 7.598970361513052e-06,
|
| 617 |
+
"loss": 1.3812,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.9635036496350365,
|
| 622 |
+
"grad_norm": 0.49052562802655075,
|
| 623 |
+
"learning_rate": 7.394841821965345e-06,
|
| 624 |
+
"loss": 1.3908,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.9744525547445255,
|
| 629 |
+
"grad_norm": 0.5101699571131805,
|
| 630 |
+
"learning_rate": 7.191871380165538e-06,
|
| 631 |
+
"loss": 1.3952,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.9854014598540146,
|
| 636 |
+
"grad_norm": 0.4636600468548116,
|
| 637 |
+
"learning_rate": 6.990149264650814e-06,
|
| 638 |
+
"loss": 1.3853,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.9963503649635036,
|
| 643 |
+
"grad_norm": 0.4452445674294053,
|
| 644 |
+
"learning_rate": 6.789765149027039e-06,
|
| 645 |
+
"loss": 1.3802,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.0072992700729928,
|
| 650 |
+
"grad_norm": 0.43937585741530943,
|
| 651 |
+
"learning_rate": 6.590808112105232e-06,
|
| 652 |
+
"loss": 1.3781,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.0182481751824817,
|
| 657 |
+
"grad_norm": 0.40693017546372506,
|
| 658 |
+
"learning_rate": 6.3933665983024465e-06,
|
| 659 |
+
"loss": 1.3737,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.0291970802919708,
|
| 664 |
+
"grad_norm": 0.38839616070894256,
|
| 665 |
+
"learning_rate": 6.197528378324664e-06,
|
| 666 |
+
"loss": 1.3829,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.0401459854014599,
|
| 671 |
+
"grad_norm": 0.38097535630426005,
|
| 672 |
+
"learning_rate": 6.003380510149179e-06,
|
| 673 |
+
"loss": 1.3637,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.051094890510949,
|
| 678 |
+
"grad_norm": 0.39702149034001294,
|
| 679 |
+
"learning_rate": 5.8110093003238175e-06,
|
| 680 |
+
"loss": 1.3671,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.062043795620438,
|
| 685 |
+
"grad_norm": 0.3605457629404061,
|
| 686 |
+
"learning_rate": 5.620500265600206e-06,
|
| 687 |
+
"loss": 1.3736,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.072992700729927,
|
| 692 |
+
"grad_norm": 0.36989167656043737,
|
| 693 |
+
"learning_rate": 5.431938094918132e-06,
|
| 694 |
+
"loss": 1.3671,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.083941605839416,
|
| 699 |
+
"grad_norm": 0.3618853171615048,
|
| 700 |
+
"learning_rate": 5.245406611757882e-06,
|
| 701 |
+
"loss": 1.3565,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.094890510948905,
|
| 706 |
+
"grad_norm": 0.3674144215955233,
|
| 707 |
+
"learning_rate": 5.060988736877366e-06,
|
| 708 |
+
"loss": 1.3724,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.1058394160583942,
|
| 713 |
+
"grad_norm": 0.3639943086109723,
|
| 714 |
+
"learning_rate": 4.878766451450451e-06,
|
| 715 |
+
"loss": 1.3812,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.1167883211678833,
|
| 720 |
+
"grad_norm": 0.3498082169635515,
|
| 721 |
+
"learning_rate": 4.698820760623064e-06,
|
| 722 |
+
"loss": 1.3809,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.1277372262773722,
|
| 727 |
+
"grad_norm": 0.3598091848786093,
|
| 728 |
+
"learning_rate": 4.5212316575031325e-06,
|
| 729 |
+
"loss": 1.3671,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.1386861313868613,
|
| 734 |
+
"grad_norm": 0.3400003839686111,
|
| 735 |
+
"learning_rate": 4.346078087600411e-06,
|
| 736 |
+
"loss": 1.3706,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.1496350364963503,
|
| 741 |
+
"grad_norm": 0.3432926361881467,
|
| 742 |
+
"learning_rate": 4.173437913732048e-06,
|
| 743 |
+
"loss": 1.3729,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.1605839416058394,
|
| 748 |
+
"grad_norm": 0.31910315994505895,
|
| 749 |
+
"learning_rate": 4.003387881409397e-06,
|
| 750 |
+
"loss": 1.3658,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.1715328467153285,
|
| 755 |
+
"grad_norm": 0.2921252235377192,
|
| 756 |
+
"learning_rate": 3.836003584721577e-06,
|
| 757 |
+
"loss": 1.379,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.1824817518248176,
|
| 762 |
+
"grad_norm": 0.3254128955866072,
|
| 763 |
+
"learning_rate": 3.6713594327308343e-06,
|
| 764 |
+
"loss": 1.3731,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.1934306569343065,
|
| 769 |
+
"grad_norm": 0.28899540616764935,
|
| 770 |
+
"learning_rate": 3.509528616394716e-06,
|
| 771 |
+
"loss": 1.3543,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.2043795620437956,
|
| 776 |
+
"grad_norm": 0.29199974979520216,
|
| 777 |
+
"learning_rate": 3.3505830760297543e-06,
|
| 778 |
+
"loss": 1.3717,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.2153284671532847,
|
| 783 |
+
"grad_norm": 0.29125633952293084,
|
| 784 |
+
"learning_rate": 3.1945934693310897e-06,
|
| 785 |
+
"loss": 1.367,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.2262773722627738,
|
| 790 |
+
"grad_norm": 0.2684849755154101,
|
| 791 |
+
"learning_rate": 3.0416291399622834e-06,
|
| 792 |
+
"loss": 1.3491,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.2372262773722627,
|
| 797 |
+
"grad_norm": 0.2788745976030149,
|
| 798 |
+
"learning_rate": 2.891758086729253e-06,
|
| 799 |
+
"loss": 1.3851,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.2481751824817517,
|
| 804 |
+
"grad_norm": 0.2629488608174756,
|
| 805 |
+
"learning_rate": 2.7450469333520856e-06,
|
| 806 |
+
"loss": 1.3635,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.2591240875912408,
|
| 811 |
+
"grad_norm": 0.282739899934587,
|
| 812 |
+
"learning_rate": 2.6015608988480956e-06,
|
| 813 |
+
"loss": 1.3653,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.27007299270073,
|
| 818 |
+
"grad_norm": 0.255405519285792,
|
| 819 |
+
"learning_rate": 2.4613637685393433e-06,
|
| 820 |
+
"loss": 1.3577,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.281021897810219,
|
| 825 |
+
"grad_norm": 0.2684645750068251,
|
| 826 |
+
"learning_rate": 2.324517865697501e-06,
|
| 827 |
+
"loss": 1.3593,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.2919708029197081,
|
| 832 |
+
"grad_norm": 0.2780965942652864,
|
| 833 |
+
"learning_rate": 2.19108402383864e-06,
|
| 834 |
+
"loss": 1.3629,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.3029197080291972,
|
| 839 |
+
"grad_norm": 0.24973583733367183,
|
| 840 |
+
"learning_rate": 2.06112155968028e-06,
|
| 841 |
+
"loss": 1.3708,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.313868613138686,
|
| 846 |
+
"grad_norm": 0.2734070791097322,
|
| 847 |
+
"learning_rate": 1.9346882467727323e-06,
|
| 848 |
+
"loss": 1.3667,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.3248175182481752,
|
| 853 |
+
"grad_norm": 0.25336631629673606,
|
| 854 |
+
"learning_rate": 1.811840289816409e-06,
|
| 855 |
+
"loss": 1.3668,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.3357664233576643,
|
| 860 |
+
"grad_norm": 0.23850115305021719,
|
| 861 |
+
"learning_rate": 1.6926322996765899e-06,
|
| 862 |
+
"loss": 1.3587,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.3467153284671534,
|
| 867 |
+
"grad_norm": 0.2385807140151342,
|
| 868 |
+
"learning_rate": 1.5771172691066793e-06,
|
| 869 |
+
"loss": 1.3681,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.3576642335766422,
|
| 874 |
+
"grad_norm": 0.22733939752218302,
|
| 875 |
+
"learning_rate": 1.4653465491908003e-06,
|
| 876 |
+
"loss": 1.3797,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.3686131386861313,
|
| 881 |
+
"grad_norm": 0.23834207160803472,
|
| 882 |
+
"learning_rate": 1.3573698265161683e-06,
|
| 883 |
+
"loss": 1.3734,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.3795620437956204,
|
| 888 |
+
"grad_norm": 0.2504353241857666,
|
| 889 |
+
"learning_rate": 1.2532351010853916e-06,
|
| 890 |
+
"loss": 1.3712,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.3905109489051095,
|
| 895 |
+
"grad_norm": 0.22583139736183075,
|
| 896 |
+
"learning_rate": 1.152988664978556e-06,
|
| 897 |
+
"loss": 1.3698,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.4014598540145986,
|
| 902 |
+
"grad_norm": 0.21952373195540156,
|
| 903 |
+
"learning_rate": 1.0566750817745076e-06,
|
| 904 |
+
"loss": 1.3621,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.4124087591240877,
|
| 909 |
+
"grad_norm": 0.22729848379470638,
|
| 910 |
+
"learning_rate": 9.6433716674057e-07,
|
| 911 |
+
"loss": 1.3715,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.4233576642335766,
|
| 916 |
+
"grad_norm": 0.22809639230959283,
|
| 917 |
+
"learning_rate": 8.760159677994174e-07,
|
| 918 |
+
"loss": 1.368,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.4343065693430657,
|
| 923 |
+
"grad_norm": 0.2308357892799694,
|
| 924 |
+
"learning_rate": 7.91750747281621e-07,
|
| 925 |
+
"loss": 1.3485,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.4452554744525548,
|
| 930 |
+
"grad_norm": 0.20774665085058974,
|
| 931 |
+
"learning_rate": 7.115789644719728e-07,
|
| 932 |
+
"loss": 1.3564,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.4562043795620438,
|
| 937 |
+
"grad_norm": 0.21597224231188505,
|
| 938 |
+
"learning_rate": 6.355362589573078e-07,
|
| 939 |
+
"loss": 1.3532,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.4671532846715327,
|
| 944 |
+
"grad_norm": 0.20753949627306614,
|
| 945 |
+
"learning_rate": 5.636564347832907e-07,
|
| 946 |
+
"loss": 1.3701,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.4781021897810218,
|
| 951 |
+
"grad_norm": 0.20846728274059312,
|
| 952 |
+
"learning_rate": 4.95971445427137e-07,
|
| 953 |
+
"loss": 1.3676,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.489051094890511,
|
| 958 |
+
"grad_norm": 0.21154433030560368,
|
| 959 |
+
"learning_rate": 4.3251137959302023e-07,
|
| 960 |
+
"loss": 1.3556,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.5,
|
| 965 |
+
"grad_norm": 0.20814486343879773,
|
| 966 |
+
"learning_rate": 3.733044478364234e-07,
|
| 967 |
+
"loss": 1.3695,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.510948905109489,
|
| 972 |
+
"grad_norm": 0.20892044294616594,
|
| 973 |
+
"learning_rate": 3.1837697002341293e-07,
|
| 974 |
+
"loss": 1.3615,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.5218978102189782,
|
| 979 |
+
"grad_norm": 0.21441434395596953,
|
| 980 |
+
"learning_rate": 2.677533636303964e-07,
|
| 981 |
+
"loss": 1.363,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.5328467153284673,
|
| 986 |
+
"grad_norm": 0.20670867098808735,
|
| 987 |
+
"learning_rate": 2.214561328895748e-07,
|
| 988 |
+
"loss": 1.3669,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.5437956204379562,
|
| 993 |
+
"grad_norm": 0.21061722568448937,
|
| 994 |
+
"learning_rate": 1.7950585878489856e-07,
|
| 995 |
+
"loss": 1.355,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.5547445255474452,
|
| 1000 |
+
"grad_norm": 0.20534097539240212,
|
| 1001 |
+
"learning_rate": 1.419211899029971e-07,
|
| 1002 |
+
"loss": 1.3725,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.5656934306569343,
|
| 1007 |
+
"grad_norm": 0.20552602600669798,
|
| 1008 |
+
"learning_rate": 1.0871883414312778e-07,
|
| 1009 |
+
"loss": 1.3773,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.5766423357664232,
|
| 1014 |
+
"grad_norm": 0.20856223955306896,
|
| 1015 |
+
"learning_rate": 7.99135512898408e-08,
|
| 1016 |
+
"loss": 1.351,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.5875912408759123,
|
| 1021 |
+
"grad_norm": 0.19725835281917636,
|
| 1022 |
+
"learning_rate": 5.55181464516652e-08,
|
| 1023 |
+
"loss": 1.3626,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.5985401459854014,
|
| 1028 |
+
"grad_norm": 0.1966955279845087,
|
| 1029 |
+
"learning_rate": 3.554346436871581e-08,
|
| 1030 |
+
"loss": 1.375,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.6094890510948905,
|
| 1035 |
+
"grad_norm": 0.1929701456222177,
|
| 1036 |
+
"learning_rate": 1.9998384591773945e-08,
|
| 1037 |
+
"loss": 1.3599,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.6204379562043796,
|
| 1042 |
+
"grad_norm": 0.19208684973155887,
|
| 1043 |
+
"learning_rate": 8.889817534969425e-09,
|
| 1044 |
+
"loss": 1.3476,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.6313868613138687,
|
| 1049 |
+
"grad_norm": 0.19723243941498336,
|
| 1050 |
+
"learning_rate": 2.222701403818972e-09,
|
| 1051 |
+
"loss": 1.3655,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.6423357664233578,
|
| 1056 |
+
"grad_norm": 0.2044024566794594,
|
| 1057 |
+
"learning_rate": 0.0,
|
| 1058 |
+
"loss": 1.3705,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.6423357664233578,
|
| 1063 |
+
"step": 150,
|
| 1064 |
+
"total_flos": 241175567007744.0,
|
| 1065 |
+
"train_loss": 1.6121594174702962,
|
| 1066 |
+
"train_runtime": 3593.9445,
|
| 1067 |
+
"train_samples_per_second": 58.098,
|
| 1068 |
+
"train_steps_per_second": 0.042
|
| 1069 |
+
}
|
| 1070 |
+
],
|
| 1071 |
+
"logging_steps": 1,
|
| 1072 |
+
"max_steps": 150,
|
| 1073 |
+
"num_input_tokens_seen": 0,
|
| 1074 |
+
"num_train_epochs": 2,
|
| 1075 |
+
"save_steps": 50,
|
| 1076 |
+
"stateful_callbacks": {
|
| 1077 |
+
"TrainerControl": {
|
| 1078 |
+
"args": {
|
| 1079 |
+
"should_epoch_stop": false,
|
| 1080 |
+
"should_evaluate": false,
|
| 1081 |
+
"should_log": false,
|
| 1082 |
+
"should_save": true,
|
| 1083 |
+
"should_training_stop": true
|
| 1084 |
+
},
|
| 1085 |
+
"attributes": {}
|
| 1086 |
+
}
|
| 1087 |
+
},
|
| 1088 |
+
"total_flos": 241175567007744.0,
|
| 1089 |
+
"train_batch_size": 58,
|
| 1090 |
+
"trial_name": null,
|
| 1091 |
+
"trial_params": null
|
| 1092 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1190d6bb8e69c94c8a7924202931a7c966a91cf1010899d48a1db20fac760225
|
| 3 |
+
size 7224
|
training_loss.png
ADDED
|