Model save
Browse files- README.md +85 -0
- all_results.json +9 -0
- runs/Sep24_06-36-44_2e80285b3fe9/events.out.tfevents.1727159848.2e80285b3fe9.283796.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +2248 -0
README.md
ADDED
|
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: google/gemma-7b
|
| 3 |
+
datasets:
|
| 4 |
+
- generator
|
| 5 |
+
library_name: peft
|
| 6 |
+
license: gemma
|
| 7 |
+
tags:
|
| 8 |
+
- trl
|
| 9 |
+
- sft
|
| 10 |
+
- generated_from_trainer
|
| 11 |
+
model-index:
|
| 12 |
+
- name: gemma7b-gpt4o_1k_summarize-kasalora-auxloss
|
| 13 |
+
results: []
|
| 14 |
+
---
|
| 15 |
+
|
| 16 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 17 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 18 |
+
|
| 19 |
+
# gemma7b-gpt4o_1k_summarize-kasalora-auxloss
|
| 20 |
+
|
| 21 |
+
This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
|
| 22 |
+
It achieves the following results on the evaluation set:
|
| 23 |
+
- Loss: 34.3196
|
| 24 |
+
|
| 25 |
+
## Model description
|
| 26 |
+
|
| 27 |
+
More information needed
|
| 28 |
+
|
| 29 |
+
## Intended uses & limitations
|
| 30 |
+
|
| 31 |
+
More information needed
|
| 32 |
+
|
| 33 |
+
## Training and evaluation data
|
| 34 |
+
|
| 35 |
+
More information needed
|
| 36 |
+
|
| 37 |
+
## Training procedure
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
The following `bitsandbytes` quantization config was used during training:
|
| 41 |
+
- quant_method: QuantizationMethod.BITS_AND_BYTES
|
| 42 |
+
- _load_in_8bit: True
|
| 43 |
+
- _load_in_4bit: False
|
| 44 |
+
- llm_int8_threshold: 6.0
|
| 45 |
+
- llm_int8_skip_modules: None
|
| 46 |
+
- llm_int8_enable_fp32_cpu_offload: False
|
| 47 |
+
- llm_int8_has_fp16_weight: False
|
| 48 |
+
- bnb_4bit_quant_type: fp4
|
| 49 |
+
- bnb_4bit_use_double_quant: False
|
| 50 |
+
- bnb_4bit_compute_dtype: float32
|
| 51 |
+
- bnb_4bit_quant_storage: uint8
|
| 52 |
+
- load_in_4bit: False
|
| 53 |
+
- load_in_8bit: True
|
| 54 |
+
|
| 55 |
+
### Training hyperparameters
|
| 56 |
+
|
| 57 |
+
The following hyperparameters were used during training:
|
| 58 |
+
- learning_rate: 0.0003
|
| 59 |
+
- train_batch_size: 1
|
| 60 |
+
- eval_batch_size: 2
|
| 61 |
+
- seed: 42
|
| 62 |
+
- distributed_type: multi-GPU
|
| 63 |
+
- num_devices: 4
|
| 64 |
+
- gradient_accumulation_steps: 2
|
| 65 |
+
- total_train_batch_size: 8
|
| 66 |
+
- total_eval_batch_size: 8
|
| 67 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 68 |
+
- lr_scheduler_type: cosine
|
| 69 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 70 |
+
- num_epochs: 1
|
| 71 |
+
|
| 72 |
+
### Training results
|
| 73 |
+
|
| 74 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 75 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 76 |
+
| 32.8196 | 0.9997 | 1567 | 34.3196 |
|
| 77 |
+
|
| 78 |
+
|
| 79 |
+
### Framework versions
|
| 80 |
+
|
| 81 |
+
- PEFT 0.6.3.dev0
|
| 82 |
+
- Transformers 4.44.2
|
| 83 |
+
- Pytorch 2.4.1+cu121
|
| 84 |
+
- Datasets 3.0.0
|
| 85 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996810207336523,
|
| 3 |
+
"total_flos": 6.008655759456338e+17,
|
| 4 |
+
"train_loss": 33.50232638594598,
|
| 5 |
+
"train_runtime": 9663.2811,
|
| 6 |
+
"train_samples": 115376,
|
| 7 |
+
"train_samples_per_second": 1.297,
|
| 8 |
+
"train_steps_per_second": 0.162
|
| 9 |
+
}
|
runs/Sep24_06-36-44_2e80285b3fe9/events.out.tfevents.1727159848.2e80285b3fe9.283796.0
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2516ff7ec859b0a12a806a4f09a9dd373a71e11eaed71bdff7f493dc7b654c9c
|
| 3 |
+
size 73463
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9996810207336523,
|
| 3 |
+
"total_flos": 6.008655759456338e+17,
|
| 4 |
+
"train_loss": 33.50232638594598,
|
| 5 |
+
"train_runtime": 9663.2811,
|
| 6 |
+
"train_samples": 115376,
|
| 7 |
+
"train_samples_per_second": 1.297,
|
| 8 |
+
"train_steps_per_second": 0.162
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2248 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9996810207336523,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1567,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006379585326953748,
|
| 13 |
+
"grad_norm": 682.9805297851562,
|
| 14 |
+
"learning_rate": 1.910828025477707e-06,
|
| 15 |
+
"loss": 77.2852,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.003189792663476874,
|
| 20 |
+
"grad_norm": 520.5723266601562,
|
| 21 |
+
"learning_rate": 9.554140127388534e-06,
|
| 22 |
+
"loss": 77.5951,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006379585326953748,
|
| 27 |
+
"grad_norm": 189.88075256347656,
|
| 28 |
+
"learning_rate": 1.9108280254777068e-05,
|
| 29 |
+
"loss": 64.9391,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.009569377990430622,
|
| 34 |
+
"grad_norm": 71.3049087524414,
|
| 35 |
+
"learning_rate": 2.8662420382165606e-05,
|
| 36 |
+
"loss": 55.1326,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.012759170653907496,
|
| 41 |
+
"grad_norm": 19.6275577545166,
|
| 42 |
+
"learning_rate": 3.8216560509554137e-05,
|
| 43 |
+
"loss": 52.7711,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.01594896331738437,
|
| 48 |
+
"grad_norm": 10.905985832214355,
|
| 49 |
+
"learning_rate": 4.777070063694267e-05,
|
| 50 |
+
"loss": 51.6146,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.019138755980861243,
|
| 55 |
+
"grad_norm": 13.994418144226074,
|
| 56 |
+
"learning_rate": 5.732484076433121e-05,
|
| 57 |
+
"loss": 50.0751,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.022328548644338118,
|
| 62 |
+
"grad_norm": 35.459774017333984,
|
| 63 |
+
"learning_rate": 6.687898089171974e-05,
|
| 64 |
+
"loss": 47.8509,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.025518341307814992,
|
| 69 |
+
"grad_norm": 112.83365631103516,
|
| 70 |
+
"learning_rate": 7.643312101910827e-05,
|
| 71 |
+
"loss": 43.2866,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.028708133971291867,
|
| 76 |
+
"grad_norm": 11.995097160339355,
|
| 77 |
+
"learning_rate": 8.59872611464968e-05,
|
| 78 |
+
"loss": 35.0235,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03189792663476874,
|
| 83 |
+
"grad_norm": 5.324495792388916,
|
| 84 |
+
"learning_rate": 9.554140127388533e-05,
|
| 85 |
+
"loss": 34.1333,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03508771929824561,
|
| 90 |
+
"grad_norm": 3.124239444732666,
|
| 91 |
+
"learning_rate": 0.00010509554140127387,
|
| 92 |
+
"loss": 33.767,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.03827751196172249,
|
| 97 |
+
"grad_norm": 2.8586628437042236,
|
| 98 |
+
"learning_rate": 0.00011464968152866242,
|
| 99 |
+
"loss": 33.6247,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04146730462519936,
|
| 104 |
+
"grad_norm": 5.365510940551758,
|
| 105 |
+
"learning_rate": 0.00012420382165605095,
|
| 106 |
+
"loss": 33.4812,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.044657097288676235,
|
| 111 |
+
"grad_norm": 3.8166890144348145,
|
| 112 |
+
"learning_rate": 0.00013375796178343948,
|
| 113 |
+
"loss": 33.3342,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.04784688995215311,
|
| 118 |
+
"grad_norm": 4.826046466827393,
|
| 119 |
+
"learning_rate": 0.00014331210191082802,
|
| 120 |
+
"loss": 33.3135,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.051036682615629984,
|
| 125 |
+
"grad_norm": 2.350775957107544,
|
| 126 |
+
"learning_rate": 0.00015286624203821655,
|
| 127 |
+
"loss": 33.2625,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.05422647527910686,
|
| 132 |
+
"grad_norm": 4.356655120849609,
|
| 133 |
+
"learning_rate": 0.00016242038216560508,
|
| 134 |
+
"loss": 33.2039,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.05741626794258373,
|
| 139 |
+
"grad_norm": 5.798741817474365,
|
| 140 |
+
"learning_rate": 0.0001719745222929936,
|
| 141 |
+
"loss": 33.1815,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06060606060606061,
|
| 146 |
+
"grad_norm": 7.0147705078125,
|
| 147 |
+
"learning_rate": 0.00018152866242038214,
|
| 148 |
+
"loss": 33.1506,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06379585326953748,
|
| 153 |
+
"grad_norm": 5.96809720993042,
|
| 154 |
+
"learning_rate": 0.00019108280254777067,
|
| 155 |
+
"loss": 33.121,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.06698564593301436,
|
| 160 |
+
"grad_norm": 7.044692516326904,
|
| 161 |
+
"learning_rate": 0.0002006369426751592,
|
| 162 |
+
"loss": 33.0931,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.07017543859649122,
|
| 167 |
+
"grad_norm": 3.396991729736328,
|
| 168 |
+
"learning_rate": 0.00021019108280254773,
|
| 169 |
+
"loss": 33.0729,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.0733652312599681,
|
| 174 |
+
"grad_norm": 1.8412262201309204,
|
| 175 |
+
"learning_rate": 0.00021974522292993626,
|
| 176 |
+
"loss": 33.0843,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.07655502392344497,
|
| 181 |
+
"grad_norm": 2.1479506492614746,
|
| 182 |
+
"learning_rate": 0.00022929936305732485,
|
| 183 |
+
"loss": 33.069,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.07974481658692185,
|
| 188 |
+
"grad_norm": 4.789393901824951,
|
| 189 |
+
"learning_rate": 0.00023885350318471338,
|
| 190 |
+
"loss": 33.0806,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.08293460925039872,
|
| 195 |
+
"grad_norm": 11.621363639831543,
|
| 196 |
+
"learning_rate": 0.0002484076433121019,
|
| 197 |
+
"loss": 33.1734,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.0861244019138756,
|
| 202 |
+
"grad_norm": 73.14874267578125,
|
| 203 |
+
"learning_rate": 0.00025796178343949044,
|
| 204 |
+
"loss": 33.1719,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.08931419457735247,
|
| 209 |
+
"grad_norm": 1.356505274772644,
|
| 210 |
+
"learning_rate": 0.00026751592356687897,
|
| 211 |
+
"loss": 33.0876,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09250398724082935,
|
| 216 |
+
"grad_norm": 1.3292990922927856,
|
| 217 |
+
"learning_rate": 0.0002770700636942675,
|
| 218 |
+
"loss": 33.0986,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.09569377990430622,
|
| 223 |
+
"grad_norm": 27.297182083129883,
|
| 224 |
+
"learning_rate": 0.00028662420382165603,
|
| 225 |
+
"loss": 33.0662,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.09888357256778309,
|
| 230 |
+
"grad_norm": 3.962141275405884,
|
| 231 |
+
"learning_rate": 0.00029617834394904456,
|
| 232 |
+
"loss": 33.103,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10207336523125997,
|
| 237 |
+
"grad_norm": 2.6187081336975098,
|
| 238 |
+
"learning_rate": 0.0002999966490829603,
|
| 239 |
+
"loss": 33.0548,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.10526315789473684,
|
| 244 |
+
"grad_norm": 2.732147693634033,
|
| 245 |
+
"learning_rate": 0.00029997617179878324,
|
| 246 |
+
"loss": 33.0522,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.10845295055821372,
|
| 251 |
+
"grad_norm": 1.41147780418396,
|
| 252 |
+
"learning_rate": 0.0002999370813893047,
|
| 253 |
+
"loss": 33.0373,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11164274322169059,
|
| 258 |
+
"grad_norm": 6.294727802276611,
|
| 259 |
+
"learning_rate": 0.00029987938270592676,
|
| 260 |
+
"loss": 33.0836,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.11483253588516747,
|
| 265 |
+
"grad_norm": 5.494381904602051,
|
| 266 |
+
"learning_rate": 0.0002998030829094724,
|
| 267 |
+
"loss": 33.0821,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.11802232854864433,
|
| 272 |
+
"grad_norm": 2.7319934368133545,
|
| 273 |
+
"learning_rate": 0.00029970819146929694,
|
| 274 |
+
"loss": 33.0523,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12121212121212122,
|
| 279 |
+
"grad_norm": 1.471151351928711,
|
| 280 |
+
"learning_rate": 0.0002995947201621131,
|
| 281 |
+
"loss": 33.0452,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.12440191387559808,
|
| 286 |
+
"grad_norm": 2.9658069610595703,
|
| 287 |
+
"learning_rate": 0.00029946268307052886,
|
| 288 |
+
"loss": 33.0214,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.12759170653907495,
|
| 293 |
+
"grad_norm": 6.95133113861084,
|
| 294 |
+
"learning_rate": 0.0002993120965813003,
|
| 295 |
+
"loss": 33.011,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.13078149920255183,
|
| 300 |
+
"grad_norm": 2.7009997367858887,
|
| 301 |
+
"learning_rate": 0.0002991429793832975,
|
| 302 |
+
"loss": 33.0114,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1339712918660287,
|
| 307 |
+
"grad_norm": 2.7211108207702637,
|
| 308 |
+
"learning_rate": 0.00029895535246518524,
|
| 309 |
+
"loss": 33.0345,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.1371610845295056,
|
| 314 |
+
"grad_norm": 1.866679072380066,
|
| 315 |
+
"learning_rate": 0.0002987492391128182,
|
| 316 |
+
"loss": 33.01,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.14035087719298245,
|
| 321 |
+
"grad_norm": 2.352771759033203,
|
| 322 |
+
"learning_rate": 0.0002985246649063509,
|
| 323 |
+
"loss": 32.9968,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.14354066985645933,
|
| 328 |
+
"grad_norm": 1.8509665727615356,
|
| 329 |
+
"learning_rate": 0.0002982816577170631,
|
| 330 |
+
"loss": 33.0003,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.1467304625199362,
|
| 335 |
+
"grad_norm": 2.1403417587280273,
|
| 336 |
+
"learning_rate": 0.00029802024770390087,
|
| 337 |
+
"loss": 33.0033,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.14992025518341306,
|
| 342 |
+
"grad_norm": 1.2801026105880737,
|
| 343 |
+
"learning_rate": 0.00029774046730973334,
|
| 344 |
+
"loss": 32.9999,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.15311004784688995,
|
| 349 |
+
"grad_norm": 1.6241658926010132,
|
| 350 |
+
"learning_rate": 0.00029744235125732664,
|
| 351 |
+
"loss": 32.9961,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.15629984051036683,
|
| 356 |
+
"grad_norm": 10.04849910736084,
|
| 357 |
+
"learning_rate": 0.0002971259365450344,
|
| 358 |
+
"loss": 33.0072,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.1594896331738437,
|
| 363 |
+
"grad_norm": 6.750959873199463,
|
| 364 |
+
"learning_rate": 0.00029679126244220596,
|
| 365 |
+
"loss": 32.986,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.16267942583732056,
|
| 370 |
+
"grad_norm": 1.4271435737609863,
|
| 371 |
+
"learning_rate": 0.00029643837048431293,
|
| 372 |
+
"loss": 33.0021,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.16586921850079744,
|
| 377 |
+
"grad_norm": 12.441051483154297,
|
| 378 |
+
"learning_rate": 0.0002960673044677939,
|
| 379 |
+
"loss": 33.0079,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.16905901116427433,
|
| 384 |
+
"grad_norm": 1.6060891151428223,
|
| 385 |
+
"learning_rate": 0.00029567811044461977,
|
| 386 |
+
"loss": 32.9779,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.1722488038277512,
|
| 391 |
+
"grad_norm": 2.5357677936553955,
|
| 392 |
+
"learning_rate": 0.00029527083671657746,
|
| 393 |
+
"loss": 33.0001,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.17543859649122806,
|
| 398 |
+
"grad_norm": 1.4731814861297607,
|
| 399 |
+
"learning_rate": 0.00029484553382927594,
|
| 400 |
+
"loss": 32.9766,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.17862838915470494,
|
| 405 |
+
"grad_norm": 1.6727865934371948,
|
| 406 |
+
"learning_rate": 0.000294402254565873,
|
| 407 |
+
"loss": 32.9865,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.18181818181818182,
|
| 412 |
+
"grad_norm": 2.122405529022217,
|
| 413 |
+
"learning_rate": 0.00029394105394052434,
|
| 414 |
+
"loss": 32.9938,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.1850079744816587,
|
| 419 |
+
"grad_norm": 1.9852848052978516,
|
| 420 |
+
"learning_rate": 0.00029346198919155616,
|
| 421 |
+
"loss": 33.0059,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.18819776714513556,
|
| 426 |
+
"grad_norm": 6.7702178955078125,
|
| 427 |
+
"learning_rate": 0.00029296511977436107,
|
| 428 |
+
"loss": 32.9958,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.19138755980861244,
|
| 433 |
+
"grad_norm": 2.152496576309204,
|
| 434 |
+
"learning_rate": 0.0002924505073540198,
|
| 435 |
+
"loss": 32.9785,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.19457735247208932,
|
| 440 |
+
"grad_norm": 1.5221149921417236,
|
| 441 |
+
"learning_rate": 0.0002919182157976476,
|
| 442 |
+
"loss": 32.9521,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.19776714513556617,
|
| 447 |
+
"grad_norm": 1.7718403339385986,
|
| 448 |
+
"learning_rate": 0.00029136831116646815,
|
| 449 |
+
"loss": 32.9534,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.20095693779904306,
|
| 454 |
+
"grad_norm": 1.301917552947998,
|
| 455 |
+
"learning_rate": 0.000290800861707615,
|
| 456 |
+
"loss": 32.9369,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.20414673046251994,
|
| 461 |
+
"grad_norm": 1.795551061630249,
|
| 462 |
+
"learning_rate": 0.00029021593784566113,
|
| 463 |
+
"loss": 32.9692,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.20733652312599682,
|
| 468 |
+
"grad_norm": 1.36078679561615,
|
| 469 |
+
"learning_rate": 0.0002896136121738793,
|
| 470 |
+
"loss": 32.9731,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.21052631578947367,
|
| 475 |
+
"grad_norm": 1.4519957304000854,
|
| 476 |
+
"learning_rate": 0.0002889939594452323,
|
| 477 |
+
"loss": 32.9647,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.21371610845295055,
|
| 482 |
+
"grad_norm": 1.334033489227295,
|
| 483 |
+
"learning_rate": 0.00028835705656309583,
|
| 484 |
+
"loss": 32.9777,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.21690590111642744,
|
| 489 |
+
"grad_norm": 1.4112995862960815,
|
| 490 |
+
"learning_rate": 0.0002877029825717142,
|
| 491 |
+
"loss": 32.9767,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.22009569377990432,
|
| 496 |
+
"grad_norm": 3.528616428375244,
|
| 497 |
+
"learning_rate": 0.0002870318186463901,
|
| 498 |
+
"loss": 32.9354,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.22328548644338117,
|
| 503 |
+
"grad_norm": 3.2746047973632812,
|
| 504 |
+
"learning_rate": 0.0002863436480834105,
|
| 505 |
+
"loss": 32.9897,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.22647527910685805,
|
| 510 |
+
"grad_norm": 1.641404151916504,
|
| 511 |
+
"learning_rate": 0.00028563855628970886,
|
| 512 |
+
"loss": 32.9667,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.22966507177033493,
|
| 517 |
+
"grad_norm": 1.5198549032211304,
|
| 518 |
+
"learning_rate": 0.0002849166307722653,
|
| 519 |
+
"loss": 32.9846,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.23285486443381181,
|
| 524 |
+
"grad_norm": 2.942479372024536,
|
| 525 |
+
"learning_rate": 0.00028417796112724684,
|
| 526 |
+
"loss": 32.9962,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.23604465709728867,
|
| 531 |
+
"grad_norm": 1.860962986946106,
|
| 532 |
+
"learning_rate": 0.0002834226390288873,
|
| 533 |
+
"loss": 32.9502,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.23923444976076555,
|
| 538 |
+
"grad_norm": 1.3557507991790771,
|
| 539 |
+
"learning_rate": 0.0002826507582181103,
|
| 540 |
+
"loss": 32.9439,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.24242424242424243,
|
| 545 |
+
"grad_norm": 2.8216800689697266,
|
| 546 |
+
"learning_rate": 0.00028186241449089524,
|
| 547 |
+
"loss": 32.9648,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.24561403508771928,
|
| 552 |
+
"grad_norm": 1.4041414260864258,
|
| 553 |
+
"learning_rate": 0.000281057705686388,
|
| 554 |
+
"loss": 32.9507,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.24880382775119617,
|
| 559 |
+
"grad_norm": 1.0593173503875732,
|
| 560 |
+
"learning_rate": 0.0002802367316747589,
|
| 561 |
+
"loss": 32.9521,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.25199362041467305,
|
| 566 |
+
"grad_norm": 1.5495184659957886,
|
| 567 |
+
"learning_rate": 0.0002793995943448078,
|
| 568 |
+
"loss": 32.9331,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.2551834130781499,
|
| 573 |
+
"grad_norm": 1.3299857378005981,
|
| 574 |
+
"learning_rate": 0.00027854639759131893,
|
| 575 |
+
"loss": 32.9373,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2583732057416268,
|
| 580 |
+
"grad_norm": 4.341277122497559,
|
| 581 |
+
"learning_rate": 0.00027767724730216696,
|
| 582 |
+
"loss": 32.9751,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.26156299840510366,
|
| 587 |
+
"grad_norm": 2.890934944152832,
|
| 588 |
+
"learning_rate": 0.0002767922513451754,
|
| 589 |
+
"loss": 32.9733,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2647527910685805,
|
| 594 |
+
"grad_norm": 1.1594318151474,
|
| 595 |
+
"learning_rate": 0.00027589151955472965,
|
| 596 |
+
"loss": 32.96,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2679425837320574,
|
| 601 |
+
"grad_norm": 1.5049182176589966,
|
| 602 |
+
"learning_rate": 0.00027497516371814543,
|
| 603 |
+
"loss": 32.9408,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.2711323763955343,
|
| 608 |
+
"grad_norm": 1.2559763193130493,
|
| 609 |
+
"learning_rate": 0.00027404329756179537,
|
| 610 |
+
"loss": 32.949,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.2743221690590112,
|
| 615 |
+
"grad_norm": 3.811816453933716,
|
| 616 |
+
"learning_rate": 0.0002730960367369949,
|
| 617 |
+
"loss": 32.9946,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.27751196172248804,
|
| 622 |
+
"grad_norm": 1.7879546880722046,
|
| 623 |
+
"learning_rate": 0.00027213349880564873,
|
| 624 |
+
"loss": 32.9684,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.2807017543859649,
|
| 629 |
+
"grad_norm": 1.7998263835906982,
|
| 630 |
+
"learning_rate": 0.0002711558032256607,
|
| 631 |
+
"loss": 32.9474,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.2838915470494418,
|
| 636 |
+
"grad_norm": 1.705267071723938,
|
| 637 |
+
"learning_rate": 0.0002701630713361085,
|
| 638 |
+
"loss": 32.9497,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.28708133971291866,
|
| 643 |
+
"grad_norm": 1.5598078966140747,
|
| 644 |
+
"learning_rate": 0.00026915542634218403,
|
| 645 |
+
"loss": 32.9503,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.2902711323763955,
|
| 650 |
+
"grad_norm": 1.2488480806350708,
|
| 651 |
+
"learning_rate": 0.00026813299329990335,
|
| 652 |
+
"loss": 32.9554,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.2934609250398724,
|
| 657 |
+
"grad_norm": 1.4403730630874634,
|
| 658 |
+
"learning_rate": 0.0002670958991005859,
|
| 659 |
+
"loss": 32.9399,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.2966507177033493,
|
| 664 |
+
"grad_norm": 1.6609704494476318,
|
| 665 |
+
"learning_rate": 0.0002660442724551065,
|
| 666 |
+
"loss": 32.9369,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.29984051036682613,
|
| 671 |
+
"grad_norm": 1.792934536933899,
|
| 672 |
+
"learning_rate": 0.00026497824387792146,
|
| 673 |
+
"loss": 32.9469,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.30303030303030304,
|
| 678 |
+
"grad_norm": 1.1857388019561768,
|
| 679 |
+
"learning_rate": 0.00026389794567087085,
|
| 680 |
+
"loss": 32.947,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.3062200956937799,
|
| 685 |
+
"grad_norm": 3.0106189250946045,
|
| 686 |
+
"learning_rate": 0.0002628035119067586,
|
| 687 |
+
"loss": 32.9288,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3094098883572568,
|
| 692 |
+
"grad_norm": 1.5340945720672607,
|
| 693 |
+
"learning_rate": 0.0002616950784127135,
|
| 694 |
+
"loss": 32.9379,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.31259968102073366,
|
| 699 |
+
"grad_norm": 14.717156410217285,
|
| 700 |
+
"learning_rate": 0.00026057278275333165,
|
| 701 |
+
"loss": 32.9656,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.3157894736842105,
|
| 706 |
+
"grad_norm": 1.2622380256652832,
|
| 707 |
+
"learning_rate": 0.00025943676421360395,
|
| 708 |
+
"loss": 32.9441,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3189792663476874,
|
| 713 |
+
"grad_norm": 2.068324089050293,
|
| 714 |
+
"learning_rate": 0.00025828716378163,
|
| 715 |
+
"loss": 32.9572,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.32216905901116427,
|
| 720 |
+
"grad_norm": 1.5519779920578003,
|
| 721 |
+
"learning_rate": 0.00025712412413112006,
|
| 722 |
+
"loss": 32.9444,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.3253588516746411,
|
| 727 |
+
"grad_norm": 5.29410982131958,
|
| 728 |
+
"learning_rate": 0.00025594778960368844,
|
| 729 |
+
"loss": 32.9654,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.32854864433811803,
|
| 734 |
+
"grad_norm": 4.592552185058594,
|
| 735 |
+
"learning_rate": 0.0002547583061909396,
|
| 736 |
+
"loss": 32.9438,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3317384370015949,
|
| 741 |
+
"grad_norm": 2.020326614379883,
|
| 742 |
+
"learning_rate": 0.00025355582151634956,
|
| 743 |
+
"loss": 32.9603,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.3349282296650718,
|
| 748 |
+
"grad_norm": 10.48231029510498,
|
| 749 |
+
"learning_rate": 0.00025234048481694477,
|
| 750 |
+
"loss": 32.9279,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.33811802232854865,
|
| 755 |
+
"grad_norm": 2.3022897243499756,
|
| 756 |
+
"learning_rate": 0.0002511124469247809,
|
| 757 |
+
"loss": 32.9371,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.3413078149920255,
|
| 762 |
+
"grad_norm": 1.5293172597885132,
|
| 763 |
+
"learning_rate": 0.00024987186024822295,
|
| 764 |
+
"loss": 32.917,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3444976076555024,
|
| 769 |
+
"grad_norm": 1.6426869630813599,
|
| 770 |
+
"learning_rate": 0.0002486188787530309,
|
| 771 |
+
"loss": 32.9352,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.34768740031897927,
|
| 776 |
+
"grad_norm": 1.266628623008728,
|
| 777 |
+
"learning_rate": 0.00024735365794325117,
|
| 778 |
+
"loss": 32.9531,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3508771929824561,
|
| 783 |
+
"grad_norm": 1.558414340019226,
|
| 784 |
+
"learning_rate": 0.0002460763548419172,
|
| 785 |
+
"loss": 32.9282,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.35406698564593303,
|
| 790 |
+
"grad_norm": 1.2409437894821167,
|
| 791 |
+
"learning_rate": 0.0002447871279715624,
|
| 792 |
+
"loss": 32.9508,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.3572567783094099,
|
| 797 |
+
"grad_norm": 1.6733072996139526,
|
| 798 |
+
"learning_rate": 0.00024348613733454565,
|
| 799 |
+
"loss": 32.9415,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.36044657097288674,
|
| 804 |
+
"grad_norm": 4.34638786315918,
|
| 805 |
+
"learning_rate": 0.00024217354439319427,
|
| 806 |
+
"loss": 32.9175,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.36363636363636365,
|
| 811 |
+
"grad_norm": 1.4273444414138794,
|
| 812 |
+
"learning_rate": 0.00024084951204976528,
|
| 813 |
+
"loss": 32.9106,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.3668261562998405,
|
| 818 |
+
"grad_norm": 2.605161190032959,
|
| 819 |
+
"learning_rate": 0.0002395142046262281,
|
| 820 |
+
"loss": 32.9428,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.3700159489633174,
|
| 825 |
+
"grad_norm": 4.460203647613525,
|
| 826 |
+
"learning_rate": 0.00023816778784387094,
|
| 827 |
+
"loss": 32.923,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.37320574162679426,
|
| 832 |
+
"grad_norm": 1.232483983039856,
|
| 833 |
+
"learning_rate": 0.0002368104288027336,
|
| 834 |
+
"loss": 32.9239,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.3763955342902711,
|
| 839 |
+
"grad_norm": 10.035587310791016,
|
| 840 |
+
"learning_rate": 0.0002354422959608692,
|
| 841 |
+
"loss": 32.9288,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.379585326953748,
|
| 846 |
+
"grad_norm": 1.621656894683838,
|
| 847 |
+
"learning_rate": 0.00023406355911343717,
|
| 848 |
+
"loss": 32.9499,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.3827751196172249,
|
| 853 |
+
"grad_norm": 1.5809745788574219,
|
| 854 |
+
"learning_rate": 0.00023267438937163077,
|
| 855 |
+
"loss": 32.9158,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.38596491228070173,
|
| 860 |
+
"grad_norm": 2.634787082672119,
|
| 861 |
+
"learning_rate": 0.00023127495914144051,
|
| 862 |
+
"loss": 32.936,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.38915470494417864,
|
| 867 |
+
"grad_norm": 1.3653333187103271,
|
| 868 |
+
"learning_rate": 0.00022986544210225774,
|
| 869 |
+
"loss": 32.9167,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.3923444976076555,
|
| 874 |
+
"grad_norm": 1.6077592372894287,
|
| 875 |
+
"learning_rate": 0.00022844601318531955,
|
| 876 |
+
"loss": 32.9071,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.39553429027113235,
|
| 881 |
+
"grad_norm": 2.3299481868743896,
|
| 882 |
+
"learning_rate": 0.00022701684855199857,
|
| 883 |
+
"loss": 32.9561,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.39872408293460926,
|
| 888 |
+
"grad_norm": 1.0616035461425781,
|
| 889 |
+
"learning_rate": 0.00022557812557194,
|
| 890 |
+
"loss": 32.904,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4019138755980861,
|
| 895 |
+
"grad_norm": 1.510393500328064,
|
| 896 |
+
"learning_rate": 0.00022413002280104915,
|
| 897 |
+
"loss": 32.8996,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.405103668261563,
|
| 902 |
+
"grad_norm": 1.354466438293457,
|
| 903 |
+
"learning_rate": 0.00022267271995933074,
|
| 904 |
+
"loss": 32.9283,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4082934609250399,
|
| 909 |
+
"grad_norm": 1.6015855073928833,
|
| 910 |
+
"learning_rate": 0.00022120639790858482,
|
| 911 |
+
"loss": 32.9179,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.41148325358851673,
|
| 916 |
+
"grad_norm": 1.3500322103500366,
|
| 917 |
+
"learning_rate": 0.00021973123862996044,
|
| 918 |
+
"loss": 32.9267,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.41467304625199364,
|
| 923 |
+
"grad_norm": 1.5598021745681763,
|
| 924 |
+
"learning_rate": 0.00021824742520137026,
|
| 925 |
+
"loss": 32.9303,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.4178628389154705,
|
| 930 |
+
"grad_norm": 1.3947721719741821,
|
| 931 |
+
"learning_rate": 0.00021675514177476945,
|
| 932 |
+
"loss": 32.9193,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.42105263157894735,
|
| 937 |
+
"grad_norm": 1.5799477100372314,
|
| 938 |
+
"learning_rate": 0.0002152545735533012,
|
| 939 |
+
"loss": 32.9358,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.42424242424242425,
|
| 944 |
+
"grad_norm": 2.971487045288086,
|
| 945 |
+
"learning_rate": 0.00021374590676831136,
|
| 946 |
+
"loss": 32.9176,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.4274322169059011,
|
| 951 |
+
"grad_norm": 1.6111208200454712,
|
| 952 |
+
"learning_rate": 0.00021222932865623605,
|
| 953 |
+
"loss": 32.9286,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.430622009569378,
|
| 958 |
+
"grad_norm": 1.521829605102539,
|
| 959 |
+
"learning_rate": 0.00021070502743536414,
|
| 960 |
+
"loss": 32.8895,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.43381180223285487,
|
| 965 |
+
"grad_norm": 2.311312675476074,
|
| 966 |
+
"learning_rate": 0.00020917319228247805,
|
| 967 |
+
"loss": 32.8955,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.4370015948963317,
|
| 972 |
+
"grad_norm": 1.3634883165359497,
|
| 973 |
+
"learning_rate": 0.00020763401330937555,
|
| 974 |
+
"loss": 32.9083,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.44019138755980863,
|
| 979 |
+
"grad_norm": 1.6512216329574585,
|
| 980 |
+
"learning_rate": 0.00020608768153927546,
|
| 981 |
+
"loss": 32.9185,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.4433811802232855,
|
| 986 |
+
"grad_norm": 1.24175226688385,
|
| 987 |
+
"learning_rate": 0.00020453438888311042,
|
| 988 |
+
"loss": 32.903,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.44657097288676234,
|
| 993 |
+
"grad_norm": 1.2090448141098022,
|
| 994 |
+
"learning_rate": 0.00020297432811570916,
|
| 995 |
+
"loss": 32.9181,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.44976076555023925,
|
| 1000 |
+
"grad_norm": 1.4872608184814453,
|
| 1001 |
+
"learning_rate": 0.00020140769285187187,
|
| 1002 |
+
"loss": 32.8974,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4529505582137161,
|
| 1007 |
+
"grad_norm": 1.8509513139724731,
|
| 1008 |
+
"learning_rate": 0.00019983467752234132,
|
| 1009 |
+
"loss": 32.8918,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.45614035087719296,
|
| 1014 |
+
"grad_norm": 1.2917256355285645,
|
| 1015 |
+
"learning_rate": 0.0001982554773496723,
|
| 1016 |
+
"loss": 32.9323,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.45933014354066987,
|
| 1021 |
+
"grad_norm": 1.2343769073486328,
|
| 1022 |
+
"learning_rate": 0.00019667028832400345,
|
| 1023 |
+
"loss": 32.8814,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4625199362041467,
|
| 1028 |
+
"grad_norm": 1.5799022912979126,
|
| 1029 |
+
"learning_rate": 0.00019507930717873313,
|
| 1030 |
+
"loss": 32.9088,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.46570972886762363,
|
| 1035 |
+
"grad_norm": 2.4687390327453613,
|
| 1036 |
+
"learning_rate": 0.00019348273136610364,
|
| 1037 |
+
"loss": 32.9116,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.4688995215311005,
|
| 1042 |
+
"grad_norm": 3.06180477142334,
|
| 1043 |
+
"learning_rate": 0.00019188075903269587,
|
| 1044 |
+
"loss": 32.9153,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.47208931419457734,
|
| 1049 |
+
"grad_norm": 1.394356608390808,
|
| 1050 |
+
"learning_rate": 0.00019027358899483776,
|
| 1051 |
+
"loss": 32.9041,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.47527910685805425,
|
| 1056 |
+
"grad_norm": 2.034818410873413,
|
| 1057 |
+
"learning_rate": 0.00018866142071393013,
|
| 1058 |
+
"loss": 32.9037,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.4784688995215311,
|
| 1063 |
+
"grad_norm": 1.3607292175292969,
|
| 1064 |
+
"learning_rate": 0.00018704445427169156,
|
| 1065 |
+
"loss": 32.9138,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.48165869218500795,
|
| 1070 |
+
"grad_norm": 1.2347332239151,
|
| 1071 |
+
"learning_rate": 0.00018542289034532733,
|
| 1072 |
+
"loss": 32.9023,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.48484848484848486,
|
| 1077 |
+
"grad_norm": 1.3349498510360718,
|
| 1078 |
+
"learning_rate": 0.00018379693018262349,
|
| 1079 |
+
"loss": 32.9105,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.4880382775119617,
|
| 1084 |
+
"grad_norm": 1.068160891532898,
|
| 1085 |
+
"learning_rate": 0.00018216677557697083,
|
| 1086 |
+
"loss": 32.9054,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.49122807017543857,
|
| 1091 |
+
"grad_norm": 1.386084794998169,
|
| 1092 |
+
"learning_rate": 0.00018053262884232078,
|
| 1093 |
+
"loss": 32.8915,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.4944178628389155,
|
| 1098 |
+
"grad_norm": 1.655228614807129,
|
| 1099 |
+
"learning_rate": 0.0001788946927880768,
|
| 1100 |
+
"loss": 32.9004,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.49760765550239233,
|
| 1105 |
+
"grad_norm": 1.4305061101913452,
|
| 1106 |
+
"learning_rate": 0.00017725317069392418,
|
| 1107 |
+
"loss": 32.8904,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.5007974481658692,
|
| 1112 |
+
"grad_norm": 1.2984495162963867,
|
| 1113 |
+
"learning_rate": 0.00017560826628460182,
|
| 1114 |
+
"loss": 32.8991,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5039872408293461,
|
| 1119 |
+
"grad_norm": 1.272292971611023,
|
| 1120 |
+
"learning_rate": 0.00017396018370461808,
|
| 1121 |
+
"loss": 32.9138,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.507177033492823,
|
| 1126 |
+
"grad_norm": 2.4082605838775635,
|
| 1127 |
+
"learning_rate": 0.00017230912749291547,
|
| 1128 |
+
"loss": 32.9214,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5103668261562998,
|
| 1133 |
+
"grad_norm": 1.3513133525848389,
|
| 1134 |
+
"learning_rate": 0.00017065530255748557,
|
| 1135 |
+
"loss": 32.9055,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5135566188197768,
|
| 1140 |
+
"grad_norm": 1.1499751806259155,
|
| 1141 |
+
"learning_rate": 0.00016899891414993854,
|
| 1142 |
+
"loss": 32.9026,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5167464114832536,
|
| 1147 |
+
"grad_norm": 1.3720273971557617,
|
| 1148 |
+
"learning_rate": 0.00016734016784002994,
|
| 1149 |
+
"loss": 32.9024,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5199362041467305,
|
| 1154 |
+
"grad_norm": 1.423717737197876,
|
| 1155 |
+
"learning_rate": 0.000165679269490148,
|
| 1156 |
+
"loss": 32.8942,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.5231259968102073,
|
| 1161 |
+
"grad_norm": 1.2919470071792603,
|
| 1162 |
+
"learning_rate": 0.0001640164252297648,
|
| 1163 |
+
"loss": 32.9024,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5263157894736842,
|
| 1168 |
+
"grad_norm": 1.6845824718475342,
|
| 1169 |
+
"learning_rate": 0.00016235184142985367,
|
| 1170 |
+
"loss": 32.8939,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.529505582137161,
|
| 1175 |
+
"grad_norm": 1.803841471672058,
|
| 1176 |
+
"learning_rate": 0.00016068572467727762,
|
| 1177 |
+
"loss": 32.9116,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.532695374800638,
|
| 1182 |
+
"grad_norm": 1.202531099319458,
|
| 1183 |
+
"learning_rate": 0.00015901828174915005,
|
| 1184 |
+
"loss": 32.9034,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5358851674641149,
|
| 1189 |
+
"grad_norm": 1.2204720973968506,
|
| 1190 |
+
"learning_rate": 0.00015734971958717228,
|
| 1191 |
+
"loss": 32.9029,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5390749601275917,
|
| 1196 |
+
"grad_norm": 1.5371347665786743,
|
| 1197 |
+
"learning_rate": 0.00015568024527195067,
|
| 1198 |
+
"loss": 32.9148,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5422647527910686,
|
| 1203 |
+
"grad_norm": 1.621806263923645,
|
| 1204 |
+
"learning_rate": 0.0001540100659972963,
|
| 1205 |
+
"loss": 32.8704,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5454545454545454,
|
| 1210 |
+
"grad_norm": 1.5353915691375732,
|
| 1211 |
+
"learning_rate": 0.000152339389044511,
|
| 1212 |
+
"loss": 32.9093,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5486443381180224,
|
| 1217 |
+
"grad_norm": 1.5789800882339478,
|
| 1218 |
+
"learning_rate": 0.00015066842175666186,
|
| 1219 |
+
"loss": 32.8946,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5518341307814992,
|
| 1224 |
+
"grad_norm": 1.1287790536880493,
|
| 1225 |
+
"learning_rate": 0.0001489973715128487,
|
| 1226 |
+
"loss": 32.8768,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5550239234449761,
|
| 1231 |
+
"grad_norm": 2.7509007453918457,
|
| 1232 |
+
"learning_rate": 0.00014732644570246675,
|
| 1233 |
+
"loss": 32.8786,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.5582137161084529,
|
| 1238 |
+
"grad_norm": 1.461751937866211,
|
| 1239 |
+
"learning_rate": 0.00014565585169946824,
|
| 1240 |
+
"loss": 32.8935,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.5614035087719298,
|
| 1245 |
+
"grad_norm": 1.2226332426071167,
|
| 1246 |
+
"learning_rate": 0.0001439857968366256,
|
| 1247 |
+
"loss": 32.8947,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.5645933014354066,
|
| 1252 |
+
"grad_norm": 1.1934312582015991,
|
| 1253 |
+
"learning_rate": 0.00014231648837980022,
|
| 1254 |
+
"loss": 32.8689,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.5677830940988836,
|
| 1259 |
+
"grad_norm": 1.0734748840332031,
|
| 1260 |
+
"learning_rate": 0.00014064813350221894,
|
| 1261 |
+
"loss": 32.9223,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.5709728867623605,
|
| 1266 |
+
"grad_norm": 1.6119298934936523,
|
| 1267 |
+
"learning_rate": 0.00013898093925876267,
|
| 1268 |
+
"loss": 32.8659,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.5741626794258373,
|
| 1273 |
+
"grad_norm": 1.1905502080917358,
|
| 1274 |
+
"learning_rate": 0.00013731511256026913,
|
| 1275 |
+
"loss": 32.876,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.5773524720893142,
|
| 1280 |
+
"grad_norm": 1.1675747632980347,
|
| 1281 |
+
"learning_rate": 0.00013565086014785406,
|
| 1282 |
+
"loss": 32.8991,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.580542264752791,
|
| 1287 |
+
"grad_norm": 1.0815011262893677,
|
| 1288 |
+
"learning_rate": 0.00013398838856725257,
|
| 1289 |
+
"loss": 32.8836,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.583732057416268,
|
| 1294 |
+
"grad_norm": 1.3470786809921265,
|
| 1295 |
+
"learning_rate": 0.00013232790414318608,
|
| 1296 |
+
"loss": 32.8723,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.5869218500797448,
|
| 1301 |
+
"grad_norm": 1.1291898488998413,
|
| 1302 |
+
"learning_rate": 0.0001306696129537553,
|
| 1303 |
+
"loss": 32.8919,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.5901116427432217,
|
| 1308 |
+
"grad_norm": 1.2471877336502075,
|
| 1309 |
+
"learning_rate": 0.00012901372080486472,
|
| 1310 |
+
"loss": 32.8716,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.5933014354066986,
|
| 1315 |
+
"grad_norm": 1.2810490131378174,
|
| 1316 |
+
"learning_rate": 0.00012736043320468073,
|
| 1317 |
+
"loss": 32.8679,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.5964912280701754,
|
| 1322 |
+
"grad_norm": 1.1581730842590332,
|
| 1323 |
+
"learning_rate": 0.0001257099553381262,
|
| 1324 |
+
"loss": 32.8792,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.5996810207336523,
|
| 1329 |
+
"grad_norm": 1.0867726802825928,
|
| 1330 |
+
"learning_rate": 0.00012406249204141603,
|
| 1331 |
+
"loss": 32.8761,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.6028708133971292,
|
| 1336 |
+
"grad_norm": 1.279363989830017,
|
| 1337 |
+
"learning_rate": 0.000122418247776635,
|
| 1338 |
+
"loss": 32.881,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6060606060606061,
|
| 1343 |
+
"grad_norm": 1.2689155340194702,
|
| 1344 |
+
"learning_rate": 0.00012077742660636299,
|
| 1345 |
+
"loss": 32.8814,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6092503987240829,
|
| 1350 |
+
"grad_norm": 0.9977864027023315,
|
| 1351 |
+
"learning_rate": 0.00011914023216834904,
|
| 1352 |
+
"loss": 32.8547,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6124401913875598,
|
| 1357 |
+
"grad_norm": 1.4236512184143066,
|
| 1358 |
+
"learning_rate": 0.0001175068676502386,
|
| 1359 |
+
"loss": 32.8553,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6156299840510366,
|
| 1364 |
+
"grad_norm": 1.0569161176681519,
|
| 1365 |
+
"learning_rate": 0.00011587753576435634,
|
| 1366 |
+
"loss": 32.8551,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6188197767145136,
|
| 1371 |
+
"grad_norm": 1.3193695545196533,
|
| 1372 |
+
"learning_rate": 0.00011425243872254835,
|
| 1373 |
+
"loss": 32.8631,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6220095693779905,
|
| 1378 |
+
"grad_norm": 1.0458968877792358,
|
| 1379 |
+
"learning_rate": 0.00011263177821108573,
|
| 1380 |
+
"loss": 32.8778,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.6251993620414673,
|
| 1385 |
+
"grad_norm": 1.097130298614502,
|
| 1386 |
+
"learning_rate": 0.00011101575536563433,
|
| 1387 |
+
"loss": 32.8635,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6283891547049442,
|
| 1392 |
+
"grad_norm": 1.4956417083740234,
|
| 1393 |
+
"learning_rate": 0.00010940457074629217,
|
| 1394 |
+
"loss": 32.8633,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.631578947368421,
|
| 1399 |
+
"grad_norm": 1.9950311183929443,
|
| 1400 |
+
"learning_rate": 0.00010779842431269843,
|
| 1401 |
+
"loss": 32.8721,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6347687400318979,
|
| 1406 |
+
"grad_norm": 1.3927932977676392,
|
| 1407 |
+
"learning_rate": 0.0001061975153992172,
|
| 1408 |
+
"loss": 32.8595,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6379585326953748,
|
| 1413 |
+
"grad_norm": 1.089880347251892,
|
| 1414 |
+
"learning_rate": 0.00010460204269019829,
|
| 1415 |
+
"loss": 32.8519,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6411483253588517,
|
| 1420 |
+
"grad_norm": 1.4656952619552612,
|
| 1421 |
+
"learning_rate": 0.0001030122041953196,
|
| 1422 |
+
"loss": 32.8695,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6443381180223285,
|
| 1427 |
+
"grad_norm": 1.343774437904358,
|
| 1428 |
+
"learning_rate": 0.0001014281972250121,
|
| 1429 |
+
"loss": 32.8688,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6475279106858054,
|
| 1434 |
+
"grad_norm": 1.0298471450805664,
|
| 1435 |
+
"learning_rate": 9.985021836597273e-05,
|
| 1436 |
+
"loss": 32.8466,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6507177033492823,
|
| 1441 |
+
"grad_norm": 1.0400863885879517,
|
| 1442 |
+
"learning_rate": 9.827846345676614e-05,
|
| 1443 |
+
"loss": 32.8529,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.6539074960127592,
|
| 1448 |
+
"grad_norm": 1.187381625175476,
|
| 1449 |
+
"learning_rate": 9.671312756351998e-05,
|
| 1450 |
+
"loss": 32.8502,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.6570972886762361,
|
| 1455 |
+
"grad_norm": 1.0974117517471313,
|
| 1456 |
+
"learning_rate": 9.515440495571569e-05,
|
| 1457 |
+
"loss": 32.8592,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.6602870813397129,
|
| 1462 |
+
"grad_norm": 1.0766081809997559,
|
| 1463 |
+
"learning_rate": 9.360248908207813e-05,
|
| 1464 |
+
"loss": 32.8525,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.6634768740031898,
|
| 1469 |
+
"grad_norm": 1.3774298429489136,
|
| 1470 |
+
"learning_rate": 9.205757254656755e-05,
|
| 1471 |
+
"loss": 32.836,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.6666666666666666,
|
| 1476 |
+
"grad_norm": 1.1813623905181885,
|
| 1477 |
+
"learning_rate": 9.05198470844756e-05,
|
| 1478 |
+
"loss": 32.8414,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.6698564593301436,
|
| 1483 |
+
"grad_norm": 1.556418538093567,
|
| 1484 |
+
"learning_rate": 8.898950353862998e-05,
|
| 1485 |
+
"loss": 32.8629,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.6730462519936204,
|
| 1490 |
+
"grad_norm": 1.267234444618225,
|
| 1491 |
+
"learning_rate": 8.746673183570923e-05,
|
| 1492 |
+
"loss": 32.8663,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.6762360446570973,
|
| 1497 |
+
"grad_norm": 1.3755508661270142,
|
| 1498 |
+
"learning_rate": 8.595172096267157e-05,
|
| 1499 |
+
"loss": 32.8899,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.6794258373205742,
|
| 1504 |
+
"grad_norm": 1.3215252161026,
|
| 1505 |
+
"learning_rate": 8.444465894330024e-05,
|
| 1506 |
+
"loss": 32.838,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.682615629984051,
|
| 1511 |
+
"grad_norm": 1.0346448421478271,
|
| 1512 |
+
"learning_rate": 8.294573281486828e-05,
|
| 1513 |
+
"loss": 32.849,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.6858054226475279,
|
| 1518 |
+
"grad_norm": 1.00798761844635,
|
| 1519 |
+
"learning_rate": 8.145512860492596e-05,
|
| 1520 |
+
"loss": 32.8436,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.6889952153110048,
|
| 1525 |
+
"grad_norm": 1.5634852647781372,
|
| 1526 |
+
"learning_rate": 7.997303130821362e-05,
|
| 1527 |
+
"loss": 32.8626,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.6921850079744817,
|
| 1532 |
+
"grad_norm": 1.0482699871063232,
|
| 1533 |
+
"learning_rate": 7.849962486370206e-05,
|
| 1534 |
+
"loss": 32.8506,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.6953748006379585,
|
| 1539 |
+
"grad_norm": 1.077114224433899,
|
| 1540 |
+
"learning_rate": 7.703509213176451e-05,
|
| 1541 |
+
"loss": 32.8685,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.6985645933014354,
|
| 1546 |
+
"grad_norm": 1.0572760105133057,
|
| 1547 |
+
"learning_rate": 7.557961487148272e-05,
|
| 1548 |
+
"loss": 32.8712,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7017543859649122,
|
| 1553 |
+
"grad_norm": 0.9180437922477722,
|
| 1554 |
+
"learning_rate": 7.413337371808884e-05,
|
| 1555 |
+
"loss": 32.8458,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.7049441786283892,
|
| 1560 |
+
"grad_norm": 1.1923459768295288,
|
| 1561 |
+
"learning_rate": 7.269654816054756e-05,
|
| 1562 |
+
"loss": 32.8429,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7081339712918661,
|
| 1567 |
+
"grad_norm": 1.5775012969970703,
|
| 1568 |
+
"learning_rate": 7.126931651928012e-05,
|
| 1569 |
+
"loss": 32.8366,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7113237639553429,
|
| 1574 |
+
"grad_norm": 1.0124591588974,
|
| 1575 |
+
"learning_rate": 6.985185592403367e-05,
|
| 1576 |
+
"loss": 32.8589,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7145135566188198,
|
| 1581 |
+
"grad_norm": 1.6170281171798706,
|
| 1582 |
+
"learning_rate": 6.844434229189787e-05,
|
| 1583 |
+
"loss": 32.8525,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7177033492822966,
|
| 1588 |
+
"grad_norm": 1.2746340036392212,
|
| 1589 |
+
"learning_rate": 6.704695030547252e-05,
|
| 1590 |
+
"loss": 32.8356,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7208931419457735,
|
| 1595 |
+
"grad_norm": 1.6307950019836426,
|
| 1596 |
+
"learning_rate": 6.56598533911881e-05,
|
| 1597 |
+
"loss": 32.8478,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7240829346092504,
|
| 1602 |
+
"grad_norm": 1.007103443145752,
|
| 1603 |
+
"learning_rate": 6.428322369778254e-05,
|
| 1604 |
+
"loss": 32.8573,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.7272727272727273,
|
| 1609 |
+
"grad_norm": 1.051518201828003,
|
| 1610 |
+
"learning_rate": 6.291723207493577e-05,
|
| 1611 |
+
"loss": 32.8216,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7304625199362041,
|
| 1616 |
+
"grad_norm": 1.5083210468292236,
|
| 1617 |
+
"learning_rate": 6.15620480520666e-05,
|
| 1618 |
+
"loss": 32.8387,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.733652312599681,
|
| 1623 |
+
"grad_norm": 1.4043292999267578,
|
| 1624 |
+
"learning_rate": 6.0217839817292765e-05,
|
| 1625 |
+
"loss": 32.8262,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7368421052631579,
|
| 1630 |
+
"grad_norm": 0.9952952861785889,
|
| 1631 |
+
"learning_rate": 5.888477419655733e-05,
|
| 1632 |
+
"loss": 32.854,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7400318979266348,
|
| 1637 |
+
"grad_norm": 1.1356273889541626,
|
| 1638 |
+
"learning_rate": 5.7563016632924555e-05,
|
| 1639 |
+
"loss": 32.8296,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7432216905901117,
|
| 1644 |
+
"grad_norm": 1.1537383794784546,
|
| 1645 |
+
"learning_rate": 5.625273116604728e-05,
|
| 1646 |
+
"loss": 32.8622,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7464114832535885,
|
| 1651 |
+
"grad_norm": 1.2923429012298584,
|
| 1652 |
+
"learning_rate": 5.495408041180829e-05,
|
| 1653 |
+
"loss": 32.866,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.7496012759170654,
|
| 1658 |
+
"grad_norm": 1.1965166330337524,
|
| 1659 |
+
"learning_rate": 5.3667225542138507e-05,
|
| 1660 |
+
"loss": 32.8139,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.7527910685805422,
|
| 1665 |
+
"grad_norm": 0.9976691007614136,
|
| 1666 |
+
"learning_rate": 5.239232626501464e-05,
|
| 1667 |
+
"loss": 32.8625,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.7559808612440191,
|
| 1672 |
+
"grad_norm": 1.6022858619689941,
|
| 1673 |
+
"learning_rate": 5.1129540804637906e-05,
|
| 1674 |
+
"loss": 32.8465,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.759170653907496,
|
| 1679 |
+
"grad_norm": 1.2044793367385864,
|
| 1680 |
+
"learning_rate": 4.9879025881797466e-05,
|
| 1681 |
+
"loss": 32.8336,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.7623604465709729,
|
| 1686 |
+
"grad_norm": 1.1089948415756226,
|
| 1687 |
+
"learning_rate": 4.8640936694420245e-05,
|
| 1688 |
+
"loss": 32.8279,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.7655502392344498,
|
| 1693 |
+
"grad_norm": 1.0508229732513428,
|
| 1694 |
+
"learning_rate": 4.7415426898309704e-05,
|
| 1695 |
+
"loss": 32.8212,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.7687400318979266,
|
| 1700 |
+
"grad_norm": 1.102999210357666,
|
| 1701 |
+
"learning_rate": 4.62026485880761e-05,
|
| 1702 |
+
"loss": 32.8288,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.7719298245614035,
|
| 1707 |
+
"grad_norm": 1.1736469268798828,
|
| 1708 |
+
"learning_rate": 4.50027522782603e-05,
|
| 1709 |
+
"loss": 32.8339,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.7751196172248804,
|
| 1714 |
+
"grad_norm": 1.3593286275863647,
|
| 1715 |
+
"learning_rate": 4.3815886884654136e-05,
|
| 1716 |
+
"loss": 32.8248,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.7783094098883573,
|
| 1721 |
+
"grad_norm": 1.3843350410461426,
|
| 1722 |
+
"learning_rate": 4.264219970581854e-05,
|
| 1723 |
+
"loss": 32.8508,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.7814992025518341,
|
| 1728 |
+
"grad_norm": 1.1827675104141235,
|
| 1729 |
+
"learning_rate": 4.148183640480293e-05,
|
| 1730 |
+
"loss": 32.8585,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.784688995215311,
|
| 1735 |
+
"grad_norm": 1.0259432792663574,
|
| 1736 |
+
"learning_rate": 4.0334940991067276e-05,
|
| 1737 |
+
"loss": 32.846,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.7878787878787878,
|
| 1742 |
+
"grad_norm": 0.9389122724533081,
|
| 1743 |
+
"learning_rate": 3.920165580260973e-05,
|
| 1744 |
+
"loss": 32.8456,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.7910685805422647,
|
| 1749 |
+
"grad_norm": 0.9574523568153381,
|
| 1750 |
+
"learning_rate": 3.808212148830095e-05,
|
| 1751 |
+
"loss": 32.8244,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.7942583732057417,
|
| 1756 |
+
"grad_norm": 1.0494085550308228,
|
| 1757 |
+
"learning_rate": 3.697647699042918e-05,
|
| 1758 |
+
"loss": 32.8273,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.7974481658692185,
|
| 1763 |
+
"grad_norm": 1.1580866575241089,
|
| 1764 |
+
"learning_rate": 3.5884859527455995e-05,
|
| 1765 |
+
"loss": 32.8352,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8006379585326954,
|
| 1770 |
+
"grad_norm": 1.2154430150985718,
|
| 1771 |
+
"learning_rate": 3.48074045769868e-05,
|
| 1772 |
+
"loss": 32.8394,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8038277511961722,
|
| 1777 |
+
"grad_norm": 1.0774935483932495,
|
| 1778 |
+
"learning_rate": 3.3744245858956967e-05,
|
| 1779 |
+
"loss": 32.8679,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.8070175438596491,
|
| 1784 |
+
"grad_norm": 1.0384732484817505,
|
| 1785 |
+
"learning_rate": 3.2695515319036186e-05,
|
| 1786 |
+
"loss": 32.8412,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.810207336523126,
|
| 1791 |
+
"grad_norm": 0.9459766149520874,
|
| 1792 |
+
"learning_rate": 3.1661343112253304e-05,
|
| 1793 |
+
"loss": 32.8298,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8133971291866029,
|
| 1798 |
+
"grad_norm": 1.0627589225769043,
|
| 1799 |
+
"learning_rate": 3.064185758684265e-05,
|
| 1800 |
+
"loss": 32.813,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8165869218500797,
|
| 1805 |
+
"grad_norm": 0.955592691898346,
|
| 1806 |
+
"learning_rate": 2.96371852683157e-05,
|
| 1807 |
+
"loss": 32.8367,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8197767145135566,
|
| 1812 |
+
"grad_norm": 1.0569384098052979,
|
| 1813 |
+
"learning_rate": 2.8647450843757897e-05,
|
| 1814 |
+
"loss": 32.8481,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8229665071770335,
|
| 1819 |
+
"grad_norm": 1.1720012426376343,
|
| 1820 |
+
"learning_rate": 2.7672777146354246e-05,
|
| 1821 |
+
"loss": 32.8415,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8261562998405104,
|
| 1826 |
+
"grad_norm": 1.108202576637268,
|
| 1827 |
+
"learning_rate": 2.6713285140144802e-05,
|
| 1828 |
+
"loss": 32.8471,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8293460925039873,
|
| 1833 |
+
"grad_norm": 1.3925212621688843,
|
| 1834 |
+
"learning_rate": 2.5769093905012333e-05,
|
| 1835 |
+
"loss": 32.8317,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8325358851674641,
|
| 1840 |
+
"grad_norm": 1.017764687538147,
|
| 1841 |
+
"learning_rate": 2.4840320621903253e-05,
|
| 1842 |
+
"loss": 32.8244,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.835725677830941,
|
| 1847 |
+
"grad_norm": 1.1128332614898682,
|
| 1848 |
+
"learning_rate": 2.392708055828495e-05,
|
| 1849 |
+
"loss": 32.8317,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8389154704944178,
|
| 1854 |
+
"grad_norm": 1.0427892208099365,
|
| 1855 |
+
"learning_rate": 2.3029487053840295e-05,
|
| 1856 |
+
"loss": 32.8307,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.8421052631578947,
|
| 1861 |
+
"grad_norm": 0.95958411693573,
|
| 1862 |
+
"learning_rate": 2.214765150640108e-05,
|
| 1863 |
+
"loss": 32.8408,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.8452950558213717,
|
| 1868 |
+
"grad_norm": 1.0608030557632446,
|
| 1869 |
+
"learning_rate": 2.1281683358122996e-05,
|
| 1870 |
+
"loss": 32.8259,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.8484848484848485,
|
| 1875 |
+
"grad_norm": 0.8998379707336426,
|
| 1876 |
+
"learning_rate": 2.043169008190289e-05,
|
| 1877 |
+
"loss": 32.8468,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.8516746411483254,
|
| 1882 |
+
"grad_norm": 0.9696708917617798,
|
| 1883 |
+
"learning_rate": 1.9597777168040872e-05,
|
| 1884 |
+
"loss": 32.8465,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.8548644338118022,
|
| 1889 |
+
"grad_norm": 0.9957761764526367,
|
| 1890 |
+
"learning_rate": 1.8780048111147776e-05,
|
| 1891 |
+
"loss": 32.8459,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.8580542264752791,
|
| 1896 |
+
"grad_norm": 1.0414236783981323,
|
| 1897 |
+
"learning_rate": 1.797860439730126e-05,
|
| 1898 |
+
"loss": 32.7988,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.861244019138756,
|
| 1903 |
+
"grad_norm": 1.0067895650863647,
|
| 1904 |
+
"learning_rate": 1.7193545491450183e-05,
|
| 1905 |
+
"loss": 32.8391,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.8644338118022329,
|
| 1910 |
+
"grad_norm": 0.9359643459320068,
|
| 1911 |
+
"learning_rate": 1.6424968825070567e-05,
|
| 1912 |
+
"loss": 32.8153,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.8676236044657097,
|
| 1917 |
+
"grad_norm": 1.0279393196105957,
|
| 1918 |
+
"learning_rate": 1.567296978407353e-05,
|
| 1919 |
+
"loss": 32.8272,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.8708133971291866,
|
| 1924 |
+
"grad_norm": 0.9850777387619019,
|
| 1925 |
+
"learning_rate": 1.4937641696967245e-05,
|
| 1926 |
+
"loss": 32.8449,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.8740031897926634,
|
| 1931 |
+
"grad_norm": 1.01199209690094,
|
| 1932 |
+
"learning_rate": 1.4219075823274251e-05,
|
| 1933 |
+
"loss": 32.8309,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.8771929824561403,
|
| 1938 |
+
"grad_norm": 0.9314611554145813,
|
| 1939 |
+
"learning_rate": 1.3517361342205295e-05,
|
| 1940 |
+
"loss": 32.8226,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.8803827751196173,
|
| 1945 |
+
"grad_norm": 0.9059179425239563,
|
| 1946 |
+
"learning_rate": 1.2832585341591844e-05,
|
| 1947 |
+
"loss": 32.8015,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.8835725677830941,
|
| 1952 |
+
"grad_norm": 1.2151134014129639,
|
| 1953 |
+
"learning_rate": 1.2164832807077585e-05,
|
| 1954 |
+
"loss": 32.833,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.886762360446571,
|
| 1959 |
+
"grad_norm": 1.0453729629516602,
|
| 1960 |
+
"learning_rate": 1.151418661157122e-05,
|
| 1961 |
+
"loss": 32.824,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.8899521531100478,
|
| 1966 |
+
"grad_norm": 0.9662382006645203,
|
| 1967 |
+
"learning_rate": 1.0880727504961339e-05,
|
| 1968 |
+
"loss": 32.8157,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.8931419457735247,
|
| 1973 |
+
"grad_norm": 1.0310821533203125,
|
| 1974 |
+
"learning_rate": 1.0264534104094812e-05,
|
| 1975 |
+
"loss": 32.836,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.8963317384370016,
|
| 1980 |
+
"grad_norm": 0.9117385745048523,
|
| 1981 |
+
"learning_rate": 9.665682883019732e-06,
|
| 1982 |
+
"loss": 32.8227,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.8995215311004785,
|
| 1987 |
+
"grad_norm": 0.8850731253623962,
|
| 1988 |
+
"learning_rate": 9.084248163494511e-06,
|
| 1989 |
+
"loss": 32.8276,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9027113237639554,
|
| 1994 |
+
"grad_norm": 1.0270276069641113,
|
| 1995 |
+
"learning_rate": 8.520302105764148e-06,
|
| 1996 |
+
"loss": 32.8239,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9059011164274322,
|
| 2001 |
+
"grad_norm": 0.9645489454269409,
|
| 2002 |
+
"learning_rate": 7.973914699604367e-06,
|
| 2003 |
+
"loss": 32.8235,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.9090909090909091,
|
| 2008 |
+
"grad_norm": 1.0180425643920898,
|
| 2009 |
+
"learning_rate": 7.445153755635569e-06,
|
| 2010 |
+
"loss": 32.834,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9122807017543859,
|
| 2015 |
+
"grad_norm": 0.9252223372459412,
|
| 2016 |
+
"learning_rate": 6.934084896906983e-06,
|
| 2017 |
+
"loss": 32.8312,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9154704944178629,
|
| 2022 |
+
"grad_norm": 1.1004729270935059,
|
| 2023 |
+
"learning_rate": 6.440771550752377e-06,
|
| 2024 |
+
"loss": 32.8397,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9186602870813397,
|
| 2029 |
+
"grad_norm": 0.9907602071762085,
|
| 2030 |
+
"learning_rate": 5.965274940918274e-06,
|
| 2031 |
+
"loss": 32.8319,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9218500797448166,
|
| 2036 |
+
"grad_norm": 0.9156322479248047,
|
| 2037 |
+
"learning_rate": 5.507654079965612e-06,
|
| 2038 |
+
"loss": 32.8098,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9250398724082934,
|
| 2043 |
+
"grad_norm": 1.07799232006073,
|
| 2044 |
+
"learning_rate": 5.067965761945869e-06,
|
| 2045 |
+
"loss": 32.8379,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9282296650717703,
|
| 2050 |
+
"grad_norm": 0.8984599113464355,
|
| 2051 |
+
"learning_rate": 4.646264555352586e-06,
|
| 2052 |
+
"loss": 32.8193,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9314194577352473,
|
| 2057 |
+
"grad_norm": 0.9720312356948853,
|
| 2058 |
+
"learning_rate": 4.242602796348915e-06,
|
| 2059 |
+
"loss": 32.8124,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9346092503987241,
|
| 2064 |
+
"grad_norm": 1.0742676258087158,
|
| 2065 |
+
"learning_rate": 3.857030582272369e-06,
|
| 2066 |
+
"loss": 32.8109,
|
| 2067 |
+
"step": 1465
|
| 2068 |
+
},
|
| 2069 |
+
{
|
| 2070 |
+
"epoch": 0.937799043062201,
|
| 2071 |
+
"grad_norm": 1.0529791116714478,
|
| 2072 |
+
"learning_rate": 3.489595765417441e-06,
|
| 2073 |
+
"loss": 32.7938,
|
| 2074 |
+
"step": 1470
|
| 2075 |
+
},
|
| 2076 |
+
{
|
| 2077 |
+
"epoch": 0.9409888357256778,
|
| 2078 |
+
"grad_norm": 0.9238316416740417,
|
| 2079 |
+
"learning_rate": 3.140343947096624e-06,
|
| 2080 |
+
"loss": 32.7966,
|
| 2081 |
+
"step": 1475
|
| 2082 |
+
},
|
| 2083 |
+
{
|
| 2084 |
+
"epoch": 0.9441786283891547,
|
| 2085 |
+
"grad_norm": 0.9737393260002136,
|
| 2086 |
+
"learning_rate": 2.80931847198117e-06,
|
| 2087 |
+
"loss": 32.8523,
|
| 2088 |
+
"step": 1480
|
| 2089 |
+
},
|
| 2090 |
+
{
|
| 2091 |
+
"epoch": 0.9473684210526315,
|
| 2092 |
+
"grad_norm": 1.0758775472640991,
|
| 2093 |
+
"learning_rate": 2.4965604227215774e-06,
|
| 2094 |
+
"loss": 32.8347,
|
| 2095 |
+
"step": 1485
|
| 2096 |
+
},
|
| 2097 |
+
{
|
| 2098 |
+
"epoch": 0.9505582137161085,
|
| 2099 |
+
"grad_norm": 0.958068311214447,
|
| 2100 |
+
"learning_rate": 2.202108614848885e-06,
|
| 2101 |
+
"loss": 32.8136,
|
| 2102 |
+
"step": 1490
|
| 2103 |
+
},
|
| 2104 |
+
{
|
| 2105 |
+
"epoch": 0.9537480063795853,
|
| 2106 |
+
"grad_norm": 0.9180477857589722,
|
| 2107 |
+
"learning_rate": 1.925999591957561e-06,
|
| 2108 |
+
"loss": 32.813,
|
| 2109 |
+
"step": 1495
|
| 2110 |
+
},
|
| 2111 |
+
{
|
| 2112 |
+
"epoch": 0.9569377990430622,
|
| 2113 |
+
"grad_norm": 0.9640750288963318,
|
| 2114 |
+
"learning_rate": 1.6682676211700107e-06,
|
| 2115 |
+
"loss": 32.8268,
|
| 2116 |
+
"step": 1500
|
| 2117 |
+
},
|
| 2118 |
+
{
|
| 2119 |
+
"epoch": 0.960127591706539,
|
| 2120 |
+
"grad_norm": 0.9315332174301147,
|
| 2121 |
+
"learning_rate": 1.4289446888838652e-06,
|
| 2122 |
+
"loss": 32.8336,
|
| 2123 |
+
"step": 1505
|
| 2124 |
+
},
|
| 2125 |
+
{
|
| 2126 |
+
"epoch": 0.9633173843700159,
|
| 2127 |
+
"grad_norm": 1.051619529724121,
|
| 2128 |
+
"learning_rate": 1.2080604968022378e-06,
|
| 2129 |
+
"loss": 32.8047,
|
| 2130 |
+
"step": 1510
|
| 2131 |
+
},
|
| 2132 |
+
{
|
| 2133 |
+
"epoch": 0.9665071770334929,
|
| 2134 |
+
"grad_norm": 0.9639779925346375,
|
| 2135 |
+
"learning_rate": 1.0056424582474575e-06,
|
| 2136 |
+
"loss": 32.8118,
|
| 2137 |
+
"step": 1515
|
| 2138 |
+
},
|
| 2139 |
+
{
|
| 2140 |
+
"epoch": 0.9696969696969697,
|
| 2141 |
+
"grad_norm": 0.9129908680915833,
|
| 2142 |
+
"learning_rate": 8.217156947590064e-07,
|
| 2143 |
+
"loss": 32.8089,
|
| 2144 |
+
"step": 1520
|
| 2145 |
+
},
|
| 2146 |
+
{
|
| 2147 |
+
"epoch": 0.9728867623604466,
|
| 2148 |
+
"grad_norm": 1.0344650745391846,
|
| 2149 |
+
"learning_rate": 6.563030329755969e-07,
|
| 2150 |
+
"loss": 32.8023,
|
| 2151 |
+
"step": 1525
|
| 2152 |
+
},
|
| 2153 |
+
{
|
| 2154 |
+
"epoch": 0.9760765550239234,
|
| 2155 |
+
"grad_norm": 0.9805324077606201,
|
| 2156 |
+
"learning_rate": 5.094250018023715e-07,
|
| 2157 |
+
"loss": 32.8376,
|
| 2158 |
+
"step": 1530
|
| 2159 |
+
},
|
| 2160 |
+
{
|
| 2161 |
+
"epoch": 0.9792663476874003,
|
| 2162 |
+
"grad_norm": 0.9154828190803528,
|
| 2163 |
+
"learning_rate": 3.8109982986300747e-07,
|
| 2164 |
+
"loss": 32.8375,
|
| 2165 |
+
"step": 1535
|
| 2166 |
+
},
|
| 2167 |
+
{
|
| 2168 |
+
"epoch": 0.9824561403508771,
|
| 2169 |
+
"grad_norm": 0.9236858487129211,
|
| 2170 |
+
"learning_rate": 2.7134344323747616e-07,
|
| 2171 |
+
"loss": 32.808,
|
| 2172 |
+
"step": 1540
|
| 2173 |
+
},
|
| 2174 |
+
{
|
| 2175 |
+
"epoch": 0.9856459330143541,
|
| 2176 |
+
"grad_norm": 0.9518983364105225,
|
| 2177 |
+
"learning_rate": 1.801694634854578e-07,
|
| 2178 |
+
"loss": 32.8399,
|
| 2179 |
+
"step": 1545
|
| 2180 |
+
},
|
| 2181 |
+
{
|
| 2182 |
+
"epoch": 0.988835725677831,
|
| 2183 |
+
"grad_norm": 0.9915338158607483,
|
| 2184 |
+
"learning_rate": 1.075892059558603e-07,
|
| 2185 |
+
"loss": 32.8379,
|
| 2186 |
+
"step": 1550
|
| 2187 |
+
},
|
| 2188 |
+
{
|
| 2189 |
+
"epoch": 0.9920255183413078,
|
| 2190 |
+
"grad_norm": 0.9686278700828552,
|
| 2191 |
+
"learning_rate": 5.3611678382442516e-08,
|
| 2192 |
+
"loss": 32.8208,
|
| 2193 |
+
"step": 1555
|
| 2194 |
+
},
|
| 2195 |
+
{
|
| 2196 |
+
"epoch": 0.9952153110047847,
|
| 2197 |
+
"grad_norm": 0.9356927871704102,
|
| 2198 |
+
"learning_rate": 1.824357976594193e-08,
|
| 2199 |
+
"loss": 32.7948,
|
| 2200 |
+
"step": 1560
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 0.9984051036682615,
|
| 2204 |
+
"grad_norm": 0.9721894264221191,
|
| 2205 |
+
"learning_rate": 1.4892995426396548e-09,
|
| 2206 |
+
"loss": 32.8196,
|
| 2207 |
+
"step": 1565
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 0.9996810207336523,
|
| 2211 |
+
"eval_loss": 34.31958770751953,
|
| 2212 |
+
"eval_runtime": 2.9694,
|
| 2213 |
+
"eval_samples_per_second": 3.368,
|
| 2214 |
+
"eval_steps_per_second": 0.674,
|
| 2215 |
+
"step": 1567
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.9996810207336523,
|
| 2219 |
+
"step": 1567,
|
| 2220 |
+
"total_flos": 6.008655759456338e+17,
|
| 2221 |
+
"train_loss": 33.50232638594598,
|
| 2222 |
+
"train_runtime": 9663.2811,
|
| 2223 |
+
"train_samples_per_second": 1.297,
|
| 2224 |
+
"train_steps_per_second": 0.162
|
| 2225 |
+
}
|
| 2226 |
+
],
|
| 2227 |
+
"logging_steps": 5,
|
| 2228 |
+
"max_steps": 1567,
|
| 2229 |
+
"num_input_tokens_seen": 0,
|
| 2230 |
+
"num_train_epochs": 1,
|
| 2231 |
+
"save_steps": 100,
|
| 2232 |
+
"stateful_callbacks": {
|
| 2233 |
+
"TrainerControl": {
|
| 2234 |
+
"args": {
|
| 2235 |
+
"should_epoch_stop": false,
|
| 2236 |
+
"should_evaluate": false,
|
| 2237 |
+
"should_log": false,
|
| 2238 |
+
"should_save": true,
|
| 2239 |
+
"should_training_stop": true
|
| 2240 |
+
},
|
| 2241 |
+
"attributes": {}
|
| 2242 |
+
}
|
| 2243 |
+
},
|
| 2244 |
+
"total_flos": 6.008655759456338e+17,
|
| 2245 |
+
"train_batch_size": 1,
|
| 2246 |
+
"trial_name": null,
|
| 2247 |
+
"trial_params": null
|
| 2248 |
+
}
|