Model save
Browse files
README.md
ADDED
|
@@ -0,0 +1,67 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama2
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
base_model: meta-llama/Llama-2-7b-hf
|
| 9 |
+
model-index:
|
| 10 |
+
- name: llama-2-qlora-vicuna-processed-indicator-0.6
|
| 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 |
+
# llama-2-qlora-vicuna-processed-indicator-0.6
|
| 18 |
+
|
| 19 |
+
This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
|
| 20 |
+
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.7582
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.0002
|
| 41 |
+
- train_batch_size: 2
|
| 42 |
+
- eval_batch_size: 4
|
| 43 |
+
- seed: 42
|
| 44 |
+
- distributed_type: multi-GPU
|
| 45 |
+
- num_devices: 4
|
| 46 |
+
- gradient_accumulation_steps: 8
|
| 47 |
+
- total_train_batch_size: 64
|
| 48 |
+
- total_eval_batch_size: 16
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: cosine
|
| 51 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 52 |
+
- num_epochs: 1
|
| 53 |
+
|
| 54 |
+
### Training results
|
| 55 |
+
|
| 56 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 57 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 58 |
+
| 0.7995 | 0.9998 | 1463 | 0.7582 |
|
| 59 |
+
|
| 60 |
+
|
| 61 |
+
### Framework versions
|
| 62 |
+
|
| 63 |
+
- PEFT 0.11.1
|
| 64 |
+
- Transformers 4.40.1
|
| 65 |
+
- Pytorch 2.3.0+cu121
|
| 66 |
+
- Datasets 2.19.0
|
| 67 |
+
- Tokenizers 0.19.1
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 604336032
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1ac24f0b335e497e29061b264a9f226e0b723ef174c9da5b53de6920808c6eca
|
| 3 |
size 604336032
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9998291474457544,
|
| 3 |
+
"total_flos": 2.720070830863155e+16,
|
| 4 |
+
"train_loss": 0.8341796821873235,
|
| 5 |
+
"train_runtime": 76059.1967,
|
| 6 |
+
"train_samples": 93645,
|
| 7 |
+
"train_samples_per_second": 1.231,
|
| 8 |
+
"train_steps_per_second": 0.019
|
| 9 |
+
}
|
runs/Jul30_14-25-22_illidan.cs.ucla.edu/events.out.tfevents.1722375545.illidan.cs.ucla.edu.4110553.0
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 sha256:9f7861d6ed79ec1d462f274d70011b04a310d0bec8be149b723fd98274265be0
|
| 3 |
+
size 67339
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.9998291474457544,
|
| 3 |
+
"total_flos": 2.720070830863155e+16,
|
| 4 |
+
"train_loss": 0.8341796821873235,
|
| 5 |
+
"train_runtime": 76059.1967,
|
| 6 |
+
"train_samples": 93645,
|
| 7 |
+
"train_samples_per_second": 1.231,
|
| 8 |
+
"train_steps_per_second": 0.019
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2089 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9998291474457544,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1463,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0006834102169827439,
|
| 13 |
+
"grad_norm": 1.2020306596069663,
|
| 14 |
+
"learning_rate": 1.360544217687075e-06,
|
| 15 |
+
"loss": 2.3228,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0034170510849137197,
|
| 20 |
+
"grad_norm": 1.3355474092979494,
|
| 21 |
+
"learning_rate": 6.802721088435375e-06,
|
| 22 |
+
"loss": 2.0724,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006834102169827439,
|
| 27 |
+
"grad_norm": 0.9273189715373761,
|
| 28 |
+
"learning_rate": 1.360544217687075e-05,
|
| 29 |
+
"loss": 1.9699,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.010251153254741158,
|
| 34 |
+
"grad_norm": 1.3295103616049666,
|
| 35 |
+
"learning_rate": 2.0408163265306123e-05,
|
| 36 |
+
"loss": 1.8654,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.013668204339654879,
|
| 41 |
+
"grad_norm": 1.2009437077024674,
|
| 42 |
+
"learning_rate": 2.72108843537415e-05,
|
| 43 |
+
"loss": 1.9025,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.017085255424568596,
|
| 48 |
+
"grad_norm": 1.3726855590601976,
|
| 49 |
+
"learning_rate": 3.401360544217687e-05,
|
| 50 |
+
"loss": 1.7878,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.020502306509482315,
|
| 55 |
+
"grad_norm": 2.657045849796256,
|
| 56 |
+
"learning_rate": 4.0816326530612245e-05,
|
| 57 |
+
"loss": 1.6113,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.023919357594396035,
|
| 62 |
+
"grad_norm": 0.9478658789778001,
|
| 63 |
+
"learning_rate": 4.761904761904762e-05,
|
| 64 |
+
"loss": 1.4247,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.027336408679309757,
|
| 69 |
+
"grad_norm": 0.6332212749744585,
|
| 70 |
+
"learning_rate": 5.4421768707483e-05,
|
| 71 |
+
"loss": 1.2716,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.030753459764223477,
|
| 76 |
+
"grad_norm": 0.7179118855118389,
|
| 77 |
+
"learning_rate": 6.122448979591838e-05,
|
| 78 |
+
"loss": 1.1143,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.03417051084913719,
|
| 83 |
+
"grad_norm": 0.2701049312098903,
|
| 84 |
+
"learning_rate": 6.802721088435374e-05,
|
| 85 |
+
"loss": 1.0379,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.03758756193405091,
|
| 90 |
+
"grad_norm": 0.3332688283691693,
|
| 91 |
+
"learning_rate": 7.482993197278913e-05,
|
| 92 |
+
"loss": 0.9675,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04100461301896463,
|
| 97 |
+
"grad_norm": 0.18106807415672466,
|
| 98 |
+
"learning_rate": 8.163265306122449e-05,
|
| 99 |
+
"loss": 0.9228,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.04442166410387835,
|
| 104 |
+
"grad_norm": 0.10994265965320522,
|
| 105 |
+
"learning_rate": 8.843537414965987e-05,
|
| 106 |
+
"loss": 0.9031,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.04783871518879207,
|
| 111 |
+
"grad_norm": 0.091102677734405,
|
| 112 |
+
"learning_rate": 9.523809523809524e-05,
|
| 113 |
+
"loss": 0.8894,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.051255766273705795,
|
| 118 |
+
"grad_norm": 0.08026180351235926,
|
| 119 |
+
"learning_rate": 0.00010204081632653062,
|
| 120 |
+
"loss": 0.9169,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.054672817358619515,
|
| 125 |
+
"grad_norm": 0.11831157299648261,
|
| 126 |
+
"learning_rate": 0.000108843537414966,
|
| 127 |
+
"loss": 0.8814,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.058089868443533234,
|
| 132 |
+
"grad_norm": 0.06914745377313133,
|
| 133 |
+
"learning_rate": 0.00011564625850340137,
|
| 134 |
+
"loss": 0.8715,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.06150691952844695,
|
| 139 |
+
"grad_norm": 0.13920764271687064,
|
| 140 |
+
"learning_rate": 0.00012244897959183676,
|
| 141 |
+
"loss": 0.8216,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.06492397061336067,
|
| 146 |
+
"grad_norm": 0.11535706946797097,
|
| 147 |
+
"learning_rate": 0.00012925170068027212,
|
| 148 |
+
"loss": 0.8672,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.06834102169827438,
|
| 153 |
+
"grad_norm": 0.06263173224021892,
|
| 154 |
+
"learning_rate": 0.00013605442176870748,
|
| 155 |
+
"loss": 0.8932,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.0717580727831881,
|
| 160 |
+
"grad_norm": 0.0635597195613304,
|
| 161 |
+
"learning_rate": 0.00014285714285714287,
|
| 162 |
+
"loss": 0.8785,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.07517512386810182,
|
| 167 |
+
"grad_norm": 0.0668356688357063,
|
| 168 |
+
"learning_rate": 0.00014965986394557826,
|
| 169 |
+
"loss": 0.8699,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.07859217495301554,
|
| 174 |
+
"grad_norm": 0.21135761738053174,
|
| 175 |
+
"learning_rate": 0.00015646258503401362,
|
| 176 |
+
"loss": 0.854,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.08200922603792926,
|
| 181 |
+
"grad_norm": 0.06434087208368008,
|
| 182 |
+
"learning_rate": 0.00016326530612244898,
|
| 183 |
+
"loss": 0.8176,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.08542627712284298,
|
| 188 |
+
"grad_norm": 0.07781059944854175,
|
| 189 |
+
"learning_rate": 0.00017006802721088434,
|
| 190 |
+
"loss": 0.8484,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.0888433282077567,
|
| 195 |
+
"grad_norm": 0.061240585932445024,
|
| 196 |
+
"learning_rate": 0.00017687074829931973,
|
| 197 |
+
"loss": 0.8268,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.09226037929267042,
|
| 202 |
+
"grad_norm": 0.06103221480342776,
|
| 203 |
+
"learning_rate": 0.00018367346938775512,
|
| 204 |
+
"loss": 0.862,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.09567743037758414,
|
| 209 |
+
"grad_norm": 0.06475817648590411,
|
| 210 |
+
"learning_rate": 0.00019047619047619048,
|
| 211 |
+
"loss": 0.8348,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.09909448146249786,
|
| 216 |
+
"grad_norm": 0.05121427902237011,
|
| 217 |
+
"learning_rate": 0.00019727891156462587,
|
| 218 |
+
"loss": 0.8489,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.10251153254741159,
|
| 223 |
+
"grad_norm": 0.04622680230570971,
|
| 224 |
+
"learning_rate": 0.00019999743552408648,
|
| 225 |
+
"loss": 0.8033,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.10592858363232531,
|
| 230 |
+
"grad_norm": 0.04756612785734163,
|
| 231 |
+
"learning_rate": 0.00019998176420316002,
|
| 232 |
+
"loss": 0.8156,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.10934563471723903,
|
| 237 |
+
"grad_norm": 0.053273317057135,
|
| 238 |
+
"learning_rate": 0.00019995184850013148,
|
| 239 |
+
"loss": 0.8312,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.11276268580215275,
|
| 244 |
+
"grad_norm": 0.05329549631085044,
|
| 245 |
+
"learning_rate": 0.00019990769267708516,
|
| 246 |
+
"loss": 0.8369,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.11617973688706647,
|
| 251 |
+
"grad_norm": 0.04820617549260813,
|
| 252 |
+
"learning_rate": 0.0001998493030248924,
|
| 253 |
+
"loss": 0.7983,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.11959678797198019,
|
| 258 |
+
"grad_norm": 0.057369974131407014,
|
| 259 |
+
"learning_rate": 0.00019977668786231534,
|
| 260 |
+
"loss": 0.8437,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1230138390568939,
|
| 265 |
+
"grad_norm": 0.04809057102621255,
|
| 266 |
+
"learning_rate": 0.0001996898575348218,
|
| 267 |
+
"loss": 0.7988,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.1264308901418076,
|
| 272 |
+
"grad_norm": 0.05206514813436257,
|
| 273 |
+
"learning_rate": 0.00019958882441311126,
|
| 274 |
+
"loss": 0.8653,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.12984794122672133,
|
| 279 |
+
"grad_norm": 0.05340020667776524,
|
| 280 |
+
"learning_rate": 0.0001994736028913526,
|
| 281 |
+
"loss": 0.8169,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.13326499231163505,
|
| 286 |
+
"grad_norm": 0.055601924920779144,
|
| 287 |
+
"learning_rate": 0.00019934420938513313,
|
| 288 |
+
"loss": 0.8532,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.13668204339654877,
|
| 293 |
+
"grad_norm": 0.049961377636725916,
|
| 294 |
+
"learning_rate": 0.00019920066232911992,
|
| 295 |
+
"loss": 0.8551,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.1400990944814625,
|
| 300 |
+
"grad_norm": 0.06440486804586033,
|
| 301 |
+
"learning_rate": 0.00019904298217443366,
|
| 302 |
+
"loss": 0.7731,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.1435161455663762,
|
| 307 |
+
"grad_norm": 0.04520767268171852,
|
| 308 |
+
"learning_rate": 0.0001988711913857346,
|
| 309 |
+
"loss": 0.8517,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.14693319665128993,
|
| 314 |
+
"grad_norm": 0.0593580261290197,
|
| 315 |
+
"learning_rate": 0.0001986853144380224,
|
| 316 |
+
"loss": 0.8656,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.15035024773620365,
|
| 321 |
+
"grad_norm": 0.042745224707827656,
|
| 322 |
+
"learning_rate": 0.00019848537781314883,
|
| 323 |
+
"loss": 0.8116,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.15376729882111737,
|
| 328 |
+
"grad_norm": 0.0460299850700426,
|
| 329 |
+
"learning_rate": 0.0001982714099960452,
|
| 330 |
+
"loss": 0.853,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.15718434990603108,
|
| 335 |
+
"grad_norm": 0.052905877151890414,
|
| 336 |
+
"learning_rate": 0.00019804344147066397,
|
| 337 |
+
"loss": 0.8467,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.1606014009909448,
|
| 342 |
+
"grad_norm": 0.05229076173383141,
|
| 343 |
+
"learning_rate": 0.00019780150471563558,
|
| 344 |
+
"loss": 0.8689,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.16401845207585852,
|
| 349 |
+
"grad_norm": 0.05176820659932698,
|
| 350 |
+
"learning_rate": 0.00019754563419964165,
|
| 351 |
+
"loss": 0.8173,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.16743550316077224,
|
| 356 |
+
"grad_norm": 0.04367228208575328,
|
| 357 |
+
"learning_rate": 0.00019727586637650373,
|
| 358 |
+
"loss": 0.8102,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.17085255424568596,
|
| 363 |
+
"grad_norm": 0.05321184874347303,
|
| 364 |
+
"learning_rate": 0.00019699223967999005,
|
| 365 |
+
"loss": 0.8349,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.17426960533059968,
|
| 370 |
+
"grad_norm": 0.04834762129215642,
|
| 371 |
+
"learning_rate": 0.00019669479451833976,
|
| 372 |
+
"loss": 0.7996,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.1776866564155134,
|
| 377 |
+
"grad_norm": 0.05613807628331564,
|
| 378 |
+
"learning_rate": 0.00019638357326850602,
|
| 379 |
+
"loss": 0.8214,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.18110370750042712,
|
| 384 |
+
"grad_norm": 0.04883314227497837,
|
| 385 |
+
"learning_rate": 0.00019605862027011856,
|
| 386 |
+
"loss": 0.8212,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.18452075858534084,
|
| 391 |
+
"grad_norm": 0.050484613376210824,
|
| 392 |
+
"learning_rate": 0.0001957199818191665,
|
| 393 |
+
"loss": 0.8183,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.18793780967025456,
|
| 398 |
+
"grad_norm": 0.05357806995093009,
|
| 399 |
+
"learning_rate": 0.00019536770616140276,
|
| 400 |
+
"loss": 0.7989,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.19135486075516828,
|
| 405 |
+
"grad_norm": 0.04129143018621023,
|
| 406 |
+
"learning_rate": 0.00019500184348547042,
|
| 407 |
+
"loss": 0.7925,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.194771911840082,
|
| 412 |
+
"grad_norm": 0.04085892932038951,
|
| 413 |
+
"learning_rate": 0.00019462244591575222,
|
| 414 |
+
"loss": 0.8475,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.19818896292499572,
|
| 419 |
+
"grad_norm": 0.06675683827394115,
|
| 420 |
+
"learning_rate": 0.00019422956750494473,
|
| 421 |
+
"loss": 0.7807,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.20160601400990943,
|
| 426 |
+
"grad_norm": 0.05536342161943553,
|
| 427 |
+
"learning_rate": 0.00019382326422635705,
|
| 428 |
+
"loss": 0.8386,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.20502306509482318,
|
| 433 |
+
"grad_norm": 0.04899666779274121,
|
| 434 |
+
"learning_rate": 0.00019340359396593672,
|
| 435 |
+
"loss": 0.804,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.2084401161797369,
|
| 440 |
+
"grad_norm": 0.05443653155294473,
|
| 441 |
+
"learning_rate": 0.00019297061651402236,
|
| 442 |
+
"loss": 0.8228,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.21185716726465062,
|
| 447 |
+
"grad_norm": 0.04672292134290588,
|
| 448 |
+
"learning_rate": 0.00019252439355682573,
|
| 449 |
+
"loss": 0.7986,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.21527421834956434,
|
| 454 |
+
"grad_norm": 0.05171464516573939,
|
| 455 |
+
"learning_rate": 0.00019206498866764288,
|
| 456 |
+
"loss": 0.8158,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.21869126943447806,
|
| 461 |
+
"grad_norm": 0.04153138083699566,
|
| 462 |
+
"learning_rate": 0.00019159246729779725,
|
| 463 |
+
"loss": 0.7744,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.22210832051939178,
|
| 468 |
+
"grad_norm": 0.0555673656216969,
|
| 469 |
+
"learning_rate": 0.00019110689676731454,
|
| 470 |
+
"loss": 0.8087,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.2255253716043055,
|
| 475 |
+
"grad_norm": 0.044931623696595056,
|
| 476 |
+
"learning_rate": 0.00019060834625533196,
|
| 477 |
+
"loss": 0.856,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.22894242268921922,
|
| 482 |
+
"grad_norm": 0.049318573438951276,
|
| 483 |
+
"learning_rate": 0.0001900968867902419,
|
| 484 |
+
"loss": 0.8035,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.23235947377413294,
|
| 489 |
+
"grad_norm": 0.061664060206440885,
|
| 490 |
+
"learning_rate": 0.00018957259123957295,
|
| 491 |
+
"loss": 0.8527,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.23577652485904665,
|
| 496 |
+
"grad_norm": 0.05596265390941409,
|
| 497 |
+
"learning_rate": 0.00018903553429960802,
|
| 498 |
+
"loss": 0.8479,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.23919357594396037,
|
| 503 |
+
"grad_norm": 0.04165765621122209,
|
| 504 |
+
"learning_rate": 0.00018848579248474288,
|
| 505 |
+
"loss": 0.8174,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.2426106270288741,
|
| 510 |
+
"grad_norm": 0.044612085840488855,
|
| 511 |
+
"learning_rate": 0.00018792344411658468,
|
| 512 |
+
"loss": 0.7853,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.2460276781137878,
|
| 517 |
+
"grad_norm": 0.05075354589883105,
|
| 518 |
+
"learning_rate": 0.0001873485693127939,
|
| 519 |
+
"loss": 0.8029,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.24944472919870153,
|
| 524 |
+
"grad_norm": 0.04667928330485948,
|
| 525 |
+
"learning_rate": 0.0001867612499756697,
|
| 526 |
+
"loss": 0.8143,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.2528617802836152,
|
| 531 |
+
"grad_norm": 0.047551720416092784,
|
| 532 |
+
"learning_rate": 0.00018616156978048146,
|
| 533 |
+
"loss": 0.8186,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.25627883136852897,
|
| 538 |
+
"grad_norm": 0.05128991713009436,
|
| 539 |
+
"learning_rate": 0.0001855496141635476,
|
| 540 |
+
"loss": 0.8193,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.25969588245344266,
|
| 545 |
+
"grad_norm": 0.05752820484101754,
|
| 546 |
+
"learning_rate": 0.00018492547031006336,
|
| 547 |
+
"loss": 0.863,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.2631129335383564,
|
| 552 |
+
"grad_norm": 0.04668740520192892,
|
| 553 |
+
"learning_rate": 0.0001842892271416797,
|
| 554 |
+
"loss": 0.7979,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.2665299846232701,
|
| 559 |
+
"grad_norm": 0.050727440707260214,
|
| 560 |
+
"learning_rate": 0.00018364097530383457,
|
| 561 |
+
"loss": 0.8447,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.26994703570818385,
|
| 566 |
+
"grad_norm": 0.048252979993068,
|
| 567 |
+
"learning_rate": 0.00018298080715283858,
|
| 568 |
+
"loss": 0.7944,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.27336408679309754,
|
| 573 |
+
"grad_norm": 0.0476226694515863,
|
| 574 |
+
"learning_rate": 0.00018230881674271722,
|
| 575 |
+
"loss": 0.7988,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.2767811378780113,
|
| 580 |
+
"grad_norm": 0.04795689299493362,
|
| 581 |
+
"learning_rate": 0.00018162509981181084,
|
| 582 |
+
"loss": 0.8044,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.280198188962925,
|
| 587 |
+
"grad_norm": 0.05352148295316883,
|
| 588 |
+
"learning_rate": 0.0001809297537691349,
|
| 589 |
+
"loss": 0.8272,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.2836152400478387,
|
| 594 |
+
"grad_norm": 0.045370780069930124,
|
| 595 |
+
"learning_rate": 0.00018022287768050202,
|
| 596 |
+
"loss": 0.8111,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.2870322911327524,
|
| 601 |
+
"grad_norm": 0.0444718008442191,
|
| 602 |
+
"learning_rate": 0.0001795045722544083,
|
| 603 |
+
"loss": 0.8097,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.29044934221766616,
|
| 608 |
+
"grad_norm": 0.04429513957599892,
|
| 609 |
+
"learning_rate": 0.00017877493982768527,
|
| 610 |
+
"loss": 0.7924,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.29386639330257985,
|
| 615 |
+
"grad_norm": 0.047432068029914336,
|
| 616 |
+
"learning_rate": 0.0001780340843509199,
|
| 617 |
+
"loss": 0.7866,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.2972834443874936,
|
| 622 |
+
"grad_norm": 0.049740183816650754,
|
| 623 |
+
"learning_rate": 0.00017728211137364489,
|
| 624 |
+
"loss": 0.7852,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.3007004954724073,
|
| 629 |
+
"grad_norm": 0.04924473396769746,
|
| 630 |
+
"learning_rate": 0.00017651912802930112,
|
| 631 |
+
"loss": 0.8293,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.30411754655732104,
|
| 636 |
+
"grad_norm": 0.057228556129132115,
|
| 637 |
+
"learning_rate": 0.00017574524301997423,
|
| 638 |
+
"loss": 0.8132,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.30753459764223473,
|
| 643 |
+
"grad_norm": 0.053526590918154204,
|
| 644 |
+
"learning_rate": 0.0001749605666009079,
|
| 645 |
+
"loss": 0.7762,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.3109516487271485,
|
| 650 |
+
"grad_norm": 0.050268696392289904,
|
| 651 |
+
"learning_rate": 0.00017416521056479577,
|
| 652 |
+
"loss": 0.8249,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.31436869981206217,
|
| 657 |
+
"grad_norm": 0.04766777047847375,
|
| 658 |
+
"learning_rate": 0.00017335928822585447,
|
| 659 |
+
"loss": 0.799,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.3177857508969759,
|
| 664 |
+
"grad_norm": 0.044354607610623285,
|
| 665 |
+
"learning_rate": 0.00017254291440367968,
|
| 666 |
+
"loss": 0.838,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.3212028019818896,
|
| 671 |
+
"grad_norm": 0.046906502149708616,
|
| 672 |
+
"learning_rate": 0.00017171620540688782,
|
| 673 |
+
"loss": 0.8276,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.32461985306680335,
|
| 678 |
+
"grad_norm": 0.05146499212308855,
|
| 679 |
+
"learning_rate": 0.00017087927901654557,
|
| 680 |
+
"loss": 0.8168,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.32803690415171705,
|
| 685 |
+
"grad_norm": 0.047105565138273846,
|
| 686 |
+
"learning_rate": 0.00017003225446938965,
|
| 687 |
+
"loss": 0.815,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.3314539552366308,
|
| 692 |
+
"grad_norm": 0.05499606447911948,
|
| 693 |
+
"learning_rate": 0.00016917525244083918,
|
| 694 |
+
"loss": 0.79,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.3348710063215445,
|
| 699 |
+
"grad_norm": 0.0419744220807428,
|
| 700 |
+
"learning_rate": 0.0001683083950278031,
|
| 701 |
+
"loss": 0.8005,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.33828805740645823,
|
| 706 |
+
"grad_norm": 0.049938428892942886,
|
| 707 |
+
"learning_rate": 0.00016743180573128495,
|
| 708 |
+
"loss": 0.7954,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.3417051084913719,
|
| 713 |
+
"grad_norm": 0.046951024104632456,
|
| 714 |
+
"learning_rate": 0.00016654560943878783,
|
| 715 |
+
"loss": 0.8447,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.34512215957628567,
|
| 720 |
+
"grad_norm": 0.05271620074783484,
|
| 721 |
+
"learning_rate": 0.0001656499324065217,
|
| 722 |
+
"loss": 0.8019,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.34853921066119936,
|
| 727 |
+
"grad_norm": 0.0471240782044328,
|
| 728 |
+
"learning_rate": 0.0001647449022414155,
|
| 729 |
+
"loss": 0.8244,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.3519562617461131,
|
| 734 |
+
"grad_norm": 0.05807042014433393,
|
| 735 |
+
"learning_rate": 0.0001638306478829373,
|
| 736 |
+
"loss": 0.8177,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.3553733128310268,
|
| 741 |
+
"grad_norm": 0.05008851116233917,
|
| 742 |
+
"learning_rate": 0.000162907299584724,
|
| 743 |
+
"loss": 0.8101,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.35879036391594055,
|
| 748 |
+
"grad_norm": 0.05138344221543279,
|
| 749 |
+
"learning_rate": 0.00016197498889602448,
|
| 750 |
+
"loss": 0.7888,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.36220741500085424,
|
| 755 |
+
"grad_norm": 0.06306896659470204,
|
| 756 |
+
"learning_rate": 0.0001610338486429575,
|
| 757 |
+
"loss": 0.8229,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.365624466085768,
|
| 762 |
+
"grad_norm": 0.04978065788218456,
|
| 763 |
+
"learning_rate": 0.00016008401290958807,
|
| 764 |
+
"loss": 0.766,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.3690415171706817,
|
| 769 |
+
"grad_norm": 0.053225124642096215,
|
| 770 |
+
"learning_rate": 0.00015912561701882463,
|
| 771 |
+
"loss": 0.8059,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.3724585682555954,
|
| 776 |
+
"grad_norm": 0.06456086867109719,
|
| 777 |
+
"learning_rate": 0.00015815879751313955,
|
| 778 |
+
"loss": 0.7995,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.3758756193405091,
|
| 783 |
+
"grad_norm": 0.05173010381007458,
|
| 784 |
+
"learning_rate": 0.00015718369213511585,
|
| 785 |
+
"loss": 0.8098,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.37929267042542286,
|
| 790 |
+
"grad_norm": 0.04825429265824465,
|
| 791 |
+
"learning_rate": 0.00015620043980782327,
|
| 792 |
+
"loss": 0.8022,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.38270972151033655,
|
| 797 |
+
"grad_norm": 0.04794004820515472,
|
| 798 |
+
"learning_rate": 0.00015520918061502569,
|
| 799 |
+
"loss": 0.8245,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.3861267725952503,
|
| 804 |
+
"grad_norm": 0.045386997453962415,
|
| 805 |
+
"learning_rate": 0.00015421005578122356,
|
| 806 |
+
"loss": 0.7673,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.389543823680164,
|
| 811 |
+
"grad_norm": 0.04809541737862576,
|
| 812 |
+
"learning_rate": 0.00015320320765153367,
|
| 813 |
+
"loss": 0.8501,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.39296087476507774,
|
| 818 |
+
"grad_norm": 0.05241051596332237,
|
| 819 |
+
"learning_rate": 0.0001521887796714092,
|
| 820 |
+
"loss": 0.7942,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.39637792584999143,
|
| 825 |
+
"grad_norm": 0.0435869638715258,
|
| 826 |
+
"learning_rate": 0.00015116691636620318,
|
| 827 |
+
"loss": 0.8019,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.3997949769349052,
|
| 832 |
+
"grad_norm": 0.04978017776276555,
|
| 833 |
+
"learning_rate": 0.00015013776332057786,
|
| 834 |
+
"loss": 0.8177,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.40321202801981887,
|
| 839 |
+
"grad_norm": 0.046469710298847465,
|
| 840 |
+
"learning_rate": 0.00014910146715776338,
|
| 841 |
+
"loss": 0.8312,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.4066290791047326,
|
| 846 |
+
"grad_norm": 0.05396734895489411,
|
| 847 |
+
"learning_rate": 0.00014805817551866838,
|
| 848 |
+
"loss": 0.8062,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.41004613018964636,
|
| 853 |
+
"grad_norm": 0.04459842219250921,
|
| 854 |
+
"learning_rate": 0.00014700803704084564,
|
| 855 |
+
"loss": 0.7739,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.41346318127456005,
|
| 860 |
+
"grad_norm": 0.044911492362826454,
|
| 861 |
+
"learning_rate": 0.00014595120133731565,
|
| 862 |
+
"loss": 0.8243,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.4168802323594738,
|
| 867 |
+
"grad_norm": 0.062259732471486105,
|
| 868 |
+
"learning_rate": 0.00014488781897525131,
|
| 869 |
+
"loss": 0.8035,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.4202972834443875,
|
| 874 |
+
"grad_norm": 0.049059941472946,
|
| 875 |
+
"learning_rate": 0.0001438180414545267,
|
| 876 |
+
"loss": 0.7758,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.42371433452930124,
|
| 881 |
+
"grad_norm": 0.05098968718552912,
|
| 882 |
+
"learning_rate": 0.00014274202118613294,
|
| 883 |
+
"loss": 0.8207,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.42713138561421493,
|
| 888 |
+
"grad_norm": 0.0446401173743016,
|
| 889 |
+
"learning_rate": 0.00014165991147046403,
|
| 890 |
+
"loss": 0.7974,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.4305484366991287,
|
| 895 |
+
"grad_norm": 0.05004490347786539,
|
| 896 |
+
"learning_rate": 0.0001405718664754764,
|
| 897 |
+
"loss": 0.7936,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.43396548778404237,
|
| 902 |
+
"grad_norm": 0.04852978946452643,
|
| 903 |
+
"learning_rate": 0.0001394780412147245,
|
| 904 |
+
"loss": 0.801,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.4373825388689561,
|
| 909 |
+
"grad_norm": 0.04806463749182831,
|
| 910 |
+
"learning_rate": 0.00013837859152527623,
|
| 911 |
+
"loss": 0.7788,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.4407995899538698,
|
| 916 |
+
"grad_norm": 0.04862497331803918,
|
| 917 |
+
"learning_rate": 0.00013727367404551055,
|
| 918 |
+
"loss": 0.7825,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.44421664103878356,
|
| 923 |
+
"grad_norm": 0.049862937300937236,
|
| 924 |
+
"learning_rate": 0.00013616344619280156,
|
| 925 |
+
"loss": 0.7849,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.44763369212369725,
|
| 930 |
+
"grad_norm": 0.045510232942769194,
|
| 931 |
+
"learning_rate": 0.00013504806614109098,
|
| 932 |
+
"loss": 0.7961,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.451050743208611,
|
| 937 |
+
"grad_norm": 0.05369028256273588,
|
| 938 |
+
"learning_rate": 0.00013392769279835354,
|
| 939 |
+
"loss": 0.7573,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.4544677942935247,
|
| 944 |
+
"grad_norm": 0.05026656335694924,
|
| 945 |
+
"learning_rate": 0.0001328024857839569,
|
| 946 |
+
"loss": 0.7828,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.45788484537843843,
|
| 951 |
+
"grad_norm": 0.04857535606631572,
|
| 952 |
+
"learning_rate": 0.00013167260540592114,
|
| 953 |
+
"loss": 0.7593,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.4613018964633521,
|
| 958 |
+
"grad_norm": 0.048520728276942086,
|
| 959 |
+
"learning_rate": 0.00013053821263807946,
|
| 960 |
+
"loss": 0.7675,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.46471894754826587,
|
| 965 |
+
"grad_norm": 0.04667501054011369,
|
| 966 |
+
"learning_rate": 0.00012939946909714433,
|
| 967 |
+
"loss": 0.8009,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.46813599863317956,
|
| 972 |
+
"grad_norm": 0.057240522592767185,
|
| 973 |
+
"learning_rate": 0.000128256537019682,
|
| 974 |
+
"loss": 0.8031,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.4715530497180933,
|
| 979 |
+
"grad_norm": 0.052631048423238694,
|
| 980 |
+
"learning_rate": 0.0001271095792389987,
|
| 981 |
+
"loss": 0.7996,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.474970100803007,
|
| 986 |
+
"grad_norm": 0.04746921614132779,
|
| 987 |
+
"learning_rate": 0.00012595875916194188,
|
| 988 |
+
"loss": 0.7814,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.47838715188792075,
|
| 993 |
+
"grad_norm": 0.05533339114072001,
|
| 994 |
+
"learning_rate": 0.00012480424074561933,
|
| 995 |
+
"loss": 0.7658,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.48180420297283444,
|
| 1000 |
+
"grad_norm": 0.04827072377712223,
|
| 1001 |
+
"learning_rate": 0.0001236461884740409,
|
| 1002 |
+
"loss": 0.7643,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.4852212540577482,
|
| 1007 |
+
"grad_norm": 0.04645041541707932,
|
| 1008 |
+
"learning_rate": 0.00012248476733468368,
|
| 1009 |
+
"loss": 0.8182,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.4886383051426619,
|
| 1014 |
+
"grad_norm": 0.05525909772402828,
|
| 1015 |
+
"learning_rate": 0.00012132014279498703,
|
| 1016 |
+
"loss": 0.8095,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.4920553562275756,
|
| 1021 |
+
"grad_norm": 0.055659491763221325,
|
| 1022 |
+
"learning_rate": 0.0001201524807787779,
|
| 1023 |
+
"loss": 0.786,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.4954724073124893,
|
| 1028 |
+
"grad_norm": 0.05673553372313532,
|
| 1029 |
+
"learning_rate": 0.00011898194764263197,
|
| 1030 |
+
"loss": 0.8332,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.49888945839740306,
|
| 1035 |
+
"grad_norm": 0.05515832759906232,
|
| 1036 |
+
"learning_rate": 0.0001178087101521729,
|
| 1037 |
+
"loss": 0.7668,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.5023065094823168,
|
| 1042 |
+
"grad_norm": 0.05901869759826189,
|
| 1043 |
+
"learning_rate": 0.00011663293545831302,
|
| 1044 |
+
"loss": 0.7922,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.5057235605672304,
|
| 1049 |
+
"grad_norm": 0.057677078146364885,
|
| 1050 |
+
"learning_rate": 0.00011545479107343963,
|
| 1051 |
+
"loss": 0.7849,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.5091406116521442,
|
| 1056 |
+
"grad_norm": 0.05124349837241442,
|
| 1057 |
+
"learning_rate": 0.0001142744448475494,
|
| 1058 |
+
"loss": 0.8001,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.5125576627370579,
|
| 1063 |
+
"grad_norm": 0.05373639047755876,
|
| 1064 |
+
"learning_rate": 0.00011309206494433488,
|
| 1065 |
+
"loss": 0.8011,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.5159747138219717,
|
| 1070 |
+
"grad_norm": 0.052106174258665414,
|
| 1071 |
+
"learning_rate": 0.00011190781981722623,
|
| 1072 |
+
"loss": 0.8019,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.5193917649068853,
|
| 1077 |
+
"grad_norm": 0.05720659004653996,
|
| 1078 |
+
"learning_rate": 0.00011072187818539159,
|
| 1079 |
+
"loss": 0.7834,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.5228088159917991,
|
| 1084 |
+
"grad_norm": 0.04968260034464468,
|
| 1085 |
+
"learning_rate": 0.00010953440900969994,
|
| 1086 |
+
"loss": 0.763,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.5262258670767128,
|
| 1091 |
+
"grad_norm": 0.051174734080506955,
|
| 1092 |
+
"learning_rate": 0.000108345581468649,
|
| 1093 |
+
"loss": 0.7542,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.5296429181616266,
|
| 1098 |
+
"grad_norm": 0.05202154641165268,
|
| 1099 |
+
"learning_rate": 0.00010715556493426262,
|
| 1100 |
+
"loss": 0.8402,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.5330599692465402,
|
| 1105 |
+
"grad_norm": 0.05276394171114056,
|
| 1106 |
+
"learning_rate": 0.00010596452894796028,
|
| 1107 |
+
"loss": 0.7978,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.536477020331454,
|
| 1112 |
+
"grad_norm": 0.05377226234936245,
|
| 1113 |
+
"learning_rate": 0.00010477264319640252,
|
| 1114 |
+
"loss": 0.7923,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.5398940714163677,
|
| 1119 |
+
"grad_norm": 0.05764668909046145,
|
| 1120 |
+
"learning_rate": 0.00010358007748731582,
|
| 1121 |
+
"loss": 0.7842,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.5433111225012814,
|
| 1126 |
+
"grad_norm": 0.05020673174055731,
|
| 1127 |
+
"learning_rate": 0.00010238700172530009,
|
| 1128 |
+
"loss": 0.8392,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.5467281735861951,
|
| 1133 |
+
"grad_norm": 0.062063862040811144,
|
| 1134 |
+
"learning_rate": 0.00010119358588762232,
|
| 1135 |
+
"loss": 0.8354,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.5501452246711088,
|
| 1140 |
+
"grad_norm": 0.052012383832107364,
|
| 1141 |
+
"learning_rate": 0.0001,
|
| 1142 |
+
"loss": 0.7635,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.5535622757560226,
|
| 1147 |
+
"grad_norm": 0.053730950289275915,
|
| 1148 |
+
"learning_rate": 9.880641411237772e-05,
|
| 1149 |
+
"loss": 0.7928,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.5569793268409363,
|
| 1154 |
+
"grad_norm": 0.052091960918167694,
|
| 1155 |
+
"learning_rate": 9.761299827469992e-05,
|
| 1156 |
+
"loss": 0.8637,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.56039637792585,
|
| 1161 |
+
"grad_norm": 0.06417204337115318,
|
| 1162 |
+
"learning_rate": 9.641992251268419e-05,
|
| 1163 |
+
"loss": 0.8059,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.5638134290107637,
|
| 1168 |
+
"grad_norm": 0.05147939412647928,
|
| 1169 |
+
"learning_rate": 9.52273568035975e-05,
|
| 1170 |
+
"loss": 0.7883,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.5672304800956774,
|
| 1175 |
+
"grad_norm": 0.0496321920196591,
|
| 1176 |
+
"learning_rate": 9.403547105203974e-05,
|
| 1177 |
+
"loss": 0.814,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.5706475311805912,
|
| 1182 |
+
"grad_norm": 0.05031872933033937,
|
| 1183 |
+
"learning_rate": 9.28444350657374e-05,
|
| 1184 |
+
"loss": 0.7816,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.5740645822655048,
|
| 1189 |
+
"grad_norm": 0.07016242369513874,
|
| 1190 |
+
"learning_rate": 9.165441853135104e-05,
|
| 1191 |
+
"loss": 0.8329,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.5774816333504186,
|
| 1196 |
+
"grad_norm": 0.04971424475300946,
|
| 1197 |
+
"learning_rate": 9.046559099030012e-05,
|
| 1198 |
+
"loss": 0.7799,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.5808986844353323,
|
| 1203 |
+
"grad_norm": 0.0522755462077452,
|
| 1204 |
+
"learning_rate": 8.927812181460843e-05,
|
| 1205 |
+
"loss": 0.7938,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.5843157355202461,
|
| 1210 |
+
"grad_norm": 0.0512047310154528,
|
| 1211 |
+
"learning_rate": 8.809218018277378e-05,
|
| 1212 |
+
"loss": 0.7986,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.5877327866051597,
|
| 1217 |
+
"grad_norm": 0.06380254072567644,
|
| 1218 |
+
"learning_rate": 8.690793505566511e-05,
|
| 1219 |
+
"loss": 0.7792,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.5911498376900735,
|
| 1224 |
+
"grad_norm": 0.051049981907943395,
|
| 1225 |
+
"learning_rate": 8.57255551524506e-05,
|
| 1226 |
+
"loss": 0.8093,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.5945668887749872,
|
| 1231 |
+
"grad_norm": 0.05340993003962982,
|
| 1232 |
+
"learning_rate": 8.454520892656038e-05,
|
| 1233 |
+
"loss": 0.7856,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.597983939859901,
|
| 1238 |
+
"grad_norm": 0.058549604896924366,
|
| 1239 |
+
"learning_rate": 8.336706454168701e-05,
|
| 1240 |
+
"loss": 0.7921,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.6014009909448146,
|
| 1245 |
+
"grad_norm": 0.05465665880133671,
|
| 1246 |
+
"learning_rate": 8.219128984782712e-05,
|
| 1247 |
+
"loss": 0.775,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.6048180420297283,
|
| 1252 |
+
"grad_norm": 0.055556269694792465,
|
| 1253 |
+
"learning_rate": 8.101805235736804e-05,
|
| 1254 |
+
"loss": 0.8248,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.6082350931146421,
|
| 1259 |
+
"grad_norm": 0.05456060311041398,
|
| 1260 |
+
"learning_rate": 7.984751922122214e-05,
|
| 1261 |
+
"loss": 0.7868,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.6116521441995558,
|
| 1266 |
+
"grad_norm": 0.045746939739603906,
|
| 1267 |
+
"learning_rate": 7.867985720501301e-05,
|
| 1268 |
+
"loss": 0.7543,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.6150691952844695,
|
| 1273 |
+
"grad_norm": 0.07588505104354382,
|
| 1274 |
+
"learning_rate": 7.751523266531634e-05,
|
| 1275 |
+
"loss": 0.8306,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.6184862463693832,
|
| 1280 |
+
"grad_norm": 0.07330950373801204,
|
| 1281 |
+
"learning_rate": 7.635381152595915e-05,
|
| 1282 |
+
"loss": 0.8365,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.621903297454297,
|
| 1287 |
+
"grad_norm": 0.050877734105124285,
|
| 1288 |
+
"learning_rate": 7.519575925438067e-05,
|
| 1289 |
+
"loss": 0.7798,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.6253203485392107,
|
| 1294 |
+
"grad_norm": 0.060652606963752305,
|
| 1295 |
+
"learning_rate": 7.404124083805819e-05,
|
| 1296 |
+
"loss": 0.7894,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.6287373996241243,
|
| 1301 |
+
"grad_norm": 0.051776470909645156,
|
| 1302 |
+
"learning_rate": 7.289042076100132e-05,
|
| 1303 |
+
"loss": 0.7573,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.6321544507090381,
|
| 1308 |
+
"grad_norm": 0.05310980558006848,
|
| 1309 |
+
"learning_rate": 7.174346298031804e-05,
|
| 1310 |
+
"loss": 0.7859,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.6355715017939518,
|
| 1315 |
+
"grad_norm": 0.05227069362568692,
|
| 1316 |
+
"learning_rate": 7.060053090285572e-05,
|
| 1317 |
+
"loss": 0.8129,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.6389885528788656,
|
| 1322 |
+
"grad_norm": 0.05951427084679483,
|
| 1323 |
+
"learning_rate": 6.946178736192053e-05,
|
| 1324 |
+
"loss": 0.7709,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.6424056039637792,
|
| 1329 |
+
"grad_norm": 0.051385803262068645,
|
| 1330 |
+
"learning_rate": 6.832739459407885e-05,
|
| 1331 |
+
"loss": 0.7958,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.645822655048693,
|
| 1336 |
+
"grad_norm": 0.0699197425530053,
|
| 1337 |
+
"learning_rate": 6.719751421604309e-05,
|
| 1338 |
+
"loss": 0.7725,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.6492397061336067,
|
| 1343 |
+
"grad_norm": 0.05971638114237985,
|
| 1344 |
+
"learning_rate": 6.607230720164647e-05,
|
| 1345 |
+
"loss": 0.8066,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.6526567572185205,
|
| 1350 |
+
"grad_norm": 0.061174773260503496,
|
| 1351 |
+
"learning_rate": 6.495193385890901e-05,
|
| 1352 |
+
"loss": 0.7676,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.6560738083034341,
|
| 1357 |
+
"grad_norm": 0.05624437053551885,
|
| 1358 |
+
"learning_rate": 6.383655380719848e-05,
|
| 1359 |
+
"loss": 0.7995,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.6594908593883478,
|
| 1364 |
+
"grad_norm": 0.0506865636845252,
|
| 1365 |
+
"learning_rate": 6.272632595448947e-05,
|
| 1366 |
+
"loss": 0.8109,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.6629079104732616,
|
| 1371 |
+
"grad_norm": 0.052397675907368615,
|
| 1372 |
+
"learning_rate": 6.162140847472381e-05,
|
| 1373 |
+
"loss": 0.805,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.6663249615581753,
|
| 1378 |
+
"grad_norm": 0.05888057649474354,
|
| 1379 |
+
"learning_rate": 6.05219587852755e-05,
|
| 1380 |
+
"loss": 0.7887,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.669742012643089,
|
| 1385 |
+
"grad_norm": 0.05311115757434341,
|
| 1386 |
+
"learning_rate": 5.9428133524523646e-05,
|
| 1387 |
+
"loss": 0.7787,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.6731590637280027,
|
| 1392 |
+
"grad_norm": 0.052096208789878605,
|
| 1393 |
+
"learning_rate": 5.834008852953603e-05,
|
| 1394 |
+
"loss": 0.794,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.6765761148129165,
|
| 1399 |
+
"grad_norm": 0.06038508504122752,
|
| 1400 |
+
"learning_rate": 5.7257978813867094e-05,
|
| 1401 |
+
"loss": 0.7965,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.6799931658978302,
|
| 1406 |
+
"grad_norm": 0.06296714469265219,
|
| 1407 |
+
"learning_rate": 5.6181958545473325e-05,
|
| 1408 |
+
"loss": 0.7931,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.6834102169827438,
|
| 1413 |
+
"grad_norm": 0.04915194607567942,
|
| 1414 |
+
"learning_rate": 5.511218102474872e-05,
|
| 1415 |
+
"loss": 0.8123,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.6868272680676576,
|
| 1420 |
+
"grad_norm": 0.06403272365242713,
|
| 1421 |
+
"learning_rate": 5.4048798662684376e-05,
|
| 1422 |
+
"loss": 0.7897,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.6902443191525713,
|
| 1427 |
+
"grad_norm": 0.05295988577722599,
|
| 1428 |
+
"learning_rate": 5.299196295915441e-05,
|
| 1429 |
+
"loss": 0.8112,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.6936613702374851,
|
| 1434 |
+
"grad_norm": 0.052501801842771235,
|
| 1435 |
+
"learning_rate": 5.1941824481331626e-05,
|
| 1436 |
+
"loss": 0.7997,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.6970784213223987,
|
| 1441 |
+
"grad_norm": 0.05826194574882407,
|
| 1442 |
+
"learning_rate": 5.089853284223667e-05,
|
| 1443 |
+
"loss": 0.8301,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.7004954724073125,
|
| 1448 |
+
"grad_norm": 0.06250337842792936,
|
| 1449 |
+
"learning_rate": 4.986223667942214e-05,
|
| 1450 |
+
"loss": 0.8135,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.7039125234922262,
|
| 1455 |
+
"grad_norm": 0.05760873048674179,
|
| 1456 |
+
"learning_rate": 4.8833083633796826e-05,
|
| 1457 |
+
"loss": 0.8062,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.70732957457714,
|
| 1462 |
+
"grad_norm": 0.05324285237345129,
|
| 1463 |
+
"learning_rate": 4.781122032859079e-05,
|
| 1464 |
+
"loss": 0.7931,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.7107466256620536,
|
| 1469 |
+
"grad_norm": 0.0599011246936521,
|
| 1470 |
+
"learning_rate": 4.6796792348466356e-05,
|
| 1471 |
+
"loss": 0.78,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.7141636767469673,
|
| 1476 |
+
"grad_norm": 0.04893354818829853,
|
| 1477 |
+
"learning_rate": 4.578994421877645e-05,
|
| 1478 |
+
"loss": 0.8154,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.7175807278318811,
|
| 1483 |
+
"grad_norm": 0.05182450287923114,
|
| 1484 |
+
"learning_rate": 4.479081938497435e-05,
|
| 1485 |
+
"loss": 0.7852,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.7209977789167948,
|
| 1490 |
+
"grad_norm": 0.05550201225603795,
|
| 1491 |
+
"learning_rate": 4.379956019217675e-05,
|
| 1492 |
+
"loss": 0.7835,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.7244148300017085,
|
| 1497 |
+
"grad_norm": 0.05173446489569767,
|
| 1498 |
+
"learning_rate": 4.281630786488418e-05,
|
| 1499 |
+
"loss": 0.8116,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.7278318810866222,
|
| 1504 |
+
"grad_norm": 0.056948504215535725,
|
| 1505 |
+
"learning_rate": 4.184120248686048e-05,
|
| 1506 |
+
"loss": 0.8377,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.731248932171536,
|
| 1511 |
+
"grad_norm": 0.061134339997307,
|
| 1512 |
+
"learning_rate": 4.087438298117536e-05,
|
| 1513 |
+
"loss": 0.8125,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.7346659832564497,
|
| 1518 |
+
"grad_norm": 0.04956030468932823,
|
| 1519 |
+
"learning_rate": 3.991598709041196e-05,
|
| 1520 |
+
"loss": 0.7923,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.7380830343413634,
|
| 1525 |
+
"grad_norm": 0.04927896659942923,
|
| 1526 |
+
"learning_rate": 3.896615135704251e-05,
|
| 1527 |
+
"loss": 0.8063,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.7415000854262771,
|
| 1532 |
+
"grad_norm": 0.049440200527473416,
|
| 1533 |
+
"learning_rate": 3.802501110397553e-05,
|
| 1534 |
+
"loss": 0.8469,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.7449171365111908,
|
| 1539 |
+
"grad_norm": 0.04986683012842273,
|
| 1540 |
+
"learning_rate": 3.709270041527599e-05,
|
| 1541 |
+
"loss": 0.7906,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.7483341875961046,
|
| 1546 |
+
"grad_norm": 0.05602482087134134,
|
| 1547 |
+
"learning_rate": 3.616935211706275e-05,
|
| 1548 |
+
"loss": 0.7721,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.7517512386810182,
|
| 1553 |
+
"grad_norm": 0.04687315536980261,
|
| 1554 |
+
"learning_rate": 3.525509775858451e-05,
|
| 1555 |
+
"loss": 0.7597,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.755168289765932,
|
| 1560 |
+
"grad_norm": 0.056244634522858154,
|
| 1561 |
+
"learning_rate": 3.4350067593478356e-05,
|
| 1562 |
+
"loss": 0.8166,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.7585853408508457,
|
| 1567 |
+
"grad_norm": 0.06127414968737818,
|
| 1568 |
+
"learning_rate": 3.345439056121216e-05,
|
| 1569 |
+
"loss": 0.7837,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.7620023919357595,
|
| 1574 |
+
"grad_norm": 0.05399366743123272,
|
| 1575 |
+
"learning_rate": 3.256819426871507e-05,
|
| 1576 |
+
"loss": 0.778,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.7654194430206731,
|
| 1581 |
+
"grad_norm": 0.06458330411214924,
|
| 1582 |
+
"learning_rate": 3.169160497219692e-05,
|
| 1583 |
+
"loss": 0.8348,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.7688364941055869,
|
| 1588 |
+
"grad_norm": 0.05215467522818499,
|
| 1589 |
+
"learning_rate": 3.0824747559160836e-05,
|
| 1590 |
+
"loss": 0.7972,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.7722535451905006,
|
| 1595 |
+
"grad_norm": 0.05781545861178535,
|
| 1596 |
+
"learning_rate": 2.9967745530610357e-05,
|
| 1597 |
+
"loss": 0.8017,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.7756705962754143,
|
| 1602 |
+
"grad_norm": 0.05543397909633525,
|
| 1603 |
+
"learning_rate": 2.9120720983454463e-05,
|
| 1604 |
+
"loss": 0.803,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.779087647360328,
|
| 1609 |
+
"grad_norm": 0.04870099285910927,
|
| 1610 |
+
"learning_rate": 2.828379459311219e-05,
|
| 1611 |
+
"loss": 0.7606,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.7825046984452417,
|
| 1616 |
+
"grad_norm": 0.0667187917753979,
|
| 1617 |
+
"learning_rate": 2.745708559632032e-05,
|
| 1618 |
+
"loss": 0.8097,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.7859217495301555,
|
| 1623 |
+
"grad_norm": 0.05222086951191216,
|
| 1624 |
+
"learning_rate": 2.6640711774145543e-05,
|
| 1625 |
+
"loss": 0.7795,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.7893388006150692,
|
| 1630 |
+
"grad_norm": 0.04711247830692807,
|
| 1631 |
+
"learning_rate": 2.5834789435204243e-05,
|
| 1632 |
+
"loss": 0.7978,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.7927558516999829,
|
| 1637 |
+
"grad_norm": 0.04903142437310661,
|
| 1638 |
+
"learning_rate": 2.503943339909214e-05,
|
| 1639 |
+
"loss": 0.8113,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.7961729027848966,
|
| 1644 |
+
"grad_norm": 0.05187782322058754,
|
| 1645 |
+
"learning_rate": 2.4254756980025773e-05,
|
| 1646 |
+
"loss": 0.8056,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.7995899538698104,
|
| 1651 |
+
"grad_norm": 0.07430744833255934,
|
| 1652 |
+
"learning_rate": 2.348087197069889e-05,
|
| 1653 |
+
"loss": 0.7875,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.8030070049547241,
|
| 1658 |
+
"grad_norm": 0.05224096237020666,
|
| 1659 |
+
"learning_rate": 2.2717888626355134e-05,
|
| 1660 |
+
"loss": 0.7539,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.8064240560396377,
|
| 1665 |
+
"grad_norm": 0.06939065704978066,
|
| 1666 |
+
"learning_rate": 2.196591564908016e-05,
|
| 1667 |
+
"loss": 0.805,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.8098411071245515,
|
| 1672 |
+
"grad_norm": 0.05628523497923318,
|
| 1673 |
+
"learning_rate": 2.122506017231477e-05,
|
| 1674 |
+
"loss": 0.7774,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.8132581582094652,
|
| 1679 |
+
"grad_norm": 0.05874291639598103,
|
| 1680 |
+
"learning_rate": 2.04954277455917e-05,
|
| 1681 |
+
"loss": 0.8078,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.816675209294379,
|
| 1686 |
+
"grad_norm": 0.060226053962080625,
|
| 1687 |
+
"learning_rate": 1.9777122319497986e-05,
|
| 1688 |
+
"loss": 0.7899,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.8200922603792927,
|
| 1693 |
+
"grad_norm": 0.05099402535993439,
|
| 1694 |
+
"learning_rate": 1.907024623086515e-05,
|
| 1695 |
+
"loss": 0.7816,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.8235093114642064,
|
| 1700 |
+
"grad_norm": 0.057975740535861416,
|
| 1701 |
+
"learning_rate": 1.837490018818917e-05,
|
| 1702 |
+
"loss": 0.7726,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.8269263625491201,
|
| 1707 |
+
"grad_norm": 0.059484775752965784,
|
| 1708 |
+
"learning_rate": 1.7691183257282772e-05,
|
| 1709 |
+
"loss": 0.7563,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.8303434136340339,
|
| 1714 |
+
"grad_norm": 0.05370824040165683,
|
| 1715 |
+
"learning_rate": 1.7019192847161425e-05,
|
| 1716 |
+
"loss": 0.7812,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.8337604647189476,
|
| 1721 |
+
"grad_norm": 0.06107368550618316,
|
| 1722 |
+
"learning_rate": 1.635902469616544e-05,
|
| 1723 |
+
"loss": 0.8212,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.8371775158038612,
|
| 1728 |
+
"grad_norm": 0.06800386851846588,
|
| 1729 |
+
"learning_rate": 1.57107728583203e-05,
|
| 1730 |
+
"loss": 0.8013,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.840594566888775,
|
| 1735 |
+
"grad_norm": 0.07161023273562961,
|
| 1736 |
+
"learning_rate": 1.5074529689936645e-05,
|
| 1737 |
+
"loss": 0.7903,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.8440116179736887,
|
| 1742 |
+
"grad_norm": 0.06460741141183791,
|
| 1743 |
+
"learning_rate": 1.4450385836452429e-05,
|
| 1744 |
+
"loss": 0.8216,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.8474286690586025,
|
| 1749 |
+
"grad_norm": 0.0610751932799301,
|
| 1750 |
+
"learning_rate": 1.383843021951855e-05,
|
| 1751 |
+
"loss": 0.814,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.8508457201435161,
|
| 1756 |
+
"grad_norm": 0.05080799197596389,
|
| 1757 |
+
"learning_rate": 1.3238750024330338e-05,
|
| 1758 |
+
"loss": 0.808,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.8542627712284299,
|
| 1763 |
+
"grad_norm": 0.051114360900508575,
|
| 1764 |
+
"learning_rate": 1.2651430687206112e-05,
|
| 1765 |
+
"loss": 0.7891,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.8576798223133436,
|
| 1770 |
+
"grad_norm": 0.06232632387186641,
|
| 1771 |
+
"learning_rate": 1.207655588341534e-05,
|
| 1772 |
+
"loss": 0.8154,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.8610968733982574,
|
| 1777 |
+
"grad_norm": 0.050991722404101165,
|
| 1778 |
+
"learning_rate": 1.1514207515257147e-05,
|
| 1779 |
+
"loss": 0.8033,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.864513924483171,
|
| 1784 |
+
"grad_norm": 0.05489592548763728,
|
| 1785 |
+
"learning_rate": 1.096446570039198e-05,
|
| 1786 |
+
"loss": 0.7717,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.8679309755680847,
|
| 1791 |
+
"grad_norm": 0.055894258580256755,
|
| 1792 |
+
"learning_rate": 1.0427408760427093e-05,
|
| 1793 |
+
"loss": 0.7856,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.8713480266529985,
|
| 1798 |
+
"grad_norm": 0.05522200907166403,
|
| 1799 |
+
"learning_rate": 9.903113209758096e-06,
|
| 1800 |
+
"loss": 0.7621,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.8747650777379122,
|
| 1805 |
+
"grad_norm": 0.05047477471682088,
|
| 1806 |
+
"learning_rate": 9.391653744668072e-06,
|
| 1807 |
+
"loss": 0.786,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.8781821288228259,
|
| 1812 |
+
"grad_norm": 0.05005408324574757,
|
| 1813 |
+
"learning_rate": 8.89310323268544e-06,
|
| 1814 |
+
"loss": 0.7687,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.8815991799077396,
|
| 1819 |
+
"grad_norm": 0.061644507314616744,
|
| 1820 |
+
"learning_rate": 8.40753270220277e-06,
|
| 1821 |
+
"loss": 0.8259,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.8850162309926534,
|
| 1826 |
+
"grad_norm": 0.05765541923821856,
|
| 1827 |
+
"learning_rate": 7.935011332357112e-06,
|
| 1828 |
+
"loss": 0.7769,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.8884332820775671,
|
| 1833 |
+
"grad_norm": 0.05364219973356691,
|
| 1834 |
+
"learning_rate": 7.475606443174288e-06,
|
| 1835 |
+
"loss": 0.7701,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.8918503331624807,
|
| 1840 |
+
"grad_norm": 0.07263092996256254,
|
| 1841 |
+
"learning_rate": 7.029383485977625e-06,
|
| 1842 |
+
"loss": 0.7792,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.8952673842473945,
|
| 1847 |
+
"grad_norm": 0.060017401840350565,
|
| 1848 |
+
"learning_rate": 6.596406034063318e-06,
|
| 1849 |
+
"loss": 0.7845,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.8986844353323082,
|
| 1854 |
+
"grad_norm": 0.05492038881398355,
|
| 1855 |
+
"learning_rate": 6.176735773642961e-06,
|
| 1856 |
+
"loss": 0.7764,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.902101486417222,
|
| 1861 |
+
"grad_norm": 0.06526851328526875,
|
| 1862 |
+
"learning_rate": 5.770432495055311e-06,
|
| 1863 |
+
"loss": 0.8039,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.9055185375021356,
|
| 1868 |
+
"grad_norm": 0.050050687943016064,
|
| 1869 |
+
"learning_rate": 5.377554084247771e-06,
|
| 1870 |
+
"loss": 0.7889,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.9089355885870494,
|
| 1875 |
+
"grad_norm": 0.05623811926805879,
|
| 1876 |
+
"learning_rate": 4.998156514529595e-06,
|
| 1877 |
+
"loss": 0.7867,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.9123526396719631,
|
| 1882 |
+
"grad_norm": 0.051017981131755645,
|
| 1883 |
+
"learning_rate": 4.632293838597246e-06,
|
| 1884 |
+
"loss": 0.8321,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.9157696907568769,
|
| 1889 |
+
"grad_norm": 0.05412601343623542,
|
| 1890 |
+
"learning_rate": 4.280018180833501e-06,
|
| 1891 |
+
"loss": 0.804,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.9191867418417905,
|
| 1896 |
+
"grad_norm": 0.0595456214327722,
|
| 1897 |
+
"learning_rate": 3.941379729881456e-06,
|
| 1898 |
+
"loss": 0.7845,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.9226037929267042,
|
| 1903 |
+
"grad_norm": 0.05195812070434534,
|
| 1904 |
+
"learning_rate": 3.6164267314939713e-06,
|
| 1905 |
+
"loss": 0.8123,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.926020844011618,
|
| 1910 |
+
"grad_norm": 0.07421106031821466,
|
| 1911 |
+
"learning_rate": 3.3052054816602452e-06,
|
| 1912 |
+
"loss": 0.7904,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 0.9294378950965317,
|
| 1917 |
+
"grad_norm": 0.05963573412011266,
|
| 1918 |
+
"learning_rate": 3.007760320009967e-06,
|
| 1919 |
+
"loss": 0.7794,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 0.9328549461814454,
|
| 1924 |
+
"grad_norm": 0.059269509717240765,
|
| 1925 |
+
"learning_rate": 2.7241336234962944e-06,
|
| 1926 |
+
"loss": 0.8089,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 0.9362719972663591,
|
| 1931 |
+
"grad_norm": 0.059259649761895836,
|
| 1932 |
+
"learning_rate": 2.4543658003583604e-06,
|
| 1933 |
+
"loss": 0.7626,
|
| 1934 |
+
"step": 1370
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 0.9396890483512729,
|
| 1938 |
+
"grad_norm": 0.06762123294868294,
|
| 1939 |
+
"learning_rate": 2.19849528436441e-06,
|
| 1940 |
+
"loss": 0.802,
|
| 1941 |
+
"step": 1375
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 0.9431060994361866,
|
| 1945 |
+
"grad_norm": 0.060831926872792076,
|
| 1946 |
+
"learning_rate": 1.956558529336061e-06,
|
| 1947 |
+
"loss": 0.7518,
|
| 1948 |
+
"step": 1380
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 0.9465231505211003,
|
| 1952 |
+
"grad_norm": 0.051796739233864277,
|
| 1953 |
+
"learning_rate": 1.7285900039547998e-06,
|
| 1954 |
+
"loss": 0.7966,
|
| 1955 |
+
"step": 1385
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 0.949940201606014,
|
| 1959 |
+
"grad_norm": 0.055065582393637,
|
| 1960 |
+
"learning_rate": 1.5146221868511668e-06,
|
| 1961 |
+
"loss": 0.799,
|
| 1962 |
+
"step": 1390
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 0.9533572526909277,
|
| 1966 |
+
"grad_norm": 0.05414380180341436,
|
| 1967 |
+
"learning_rate": 1.3146855619776134e-06,
|
| 1968 |
+
"loss": 0.8069,
|
| 1969 |
+
"step": 1395
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 0.9567743037758415,
|
| 1973 |
+
"grad_norm": 0.07040840761575913,
|
| 1974 |
+
"learning_rate": 1.1288086142653864e-06,
|
| 1975 |
+
"loss": 0.8167,
|
| 1976 |
+
"step": 1400
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 0.9601913548607551,
|
| 1980 |
+
"grad_norm": 0.05744269684792452,
|
| 1981 |
+
"learning_rate": 9.570178255663532e-07,
|
| 1982 |
+
"loss": 0.7983,
|
| 1983 |
+
"step": 1405
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 0.9636084059456689,
|
| 1987 |
+
"grad_norm": 0.06440333782704491,
|
| 1988 |
+
"learning_rate": 7.993376708800848e-07,
|
| 1989 |
+
"loss": 0.7664,
|
| 1990 |
+
"step": 1410
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 0.9670254570305826,
|
| 1994 |
+
"grad_norm": 0.061502951799294485,
|
| 1995 |
+
"learning_rate": 6.557906148669024e-07,
|
| 1996 |
+
"loss": 0.8054,
|
| 1997 |
+
"step": 1415
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 0.9704425081154964,
|
| 2001 |
+
"grad_norm": 0.05555024265947659,
|
| 2002 |
+
"learning_rate": 5.26397108647414e-07,
|
| 2003 |
+
"loss": 0.8238,
|
| 2004 |
+
"step": 1420
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 0.97385955920041,
|
| 2008 |
+
"grad_norm": 0.053575566054838014,
|
| 2009 |
+
"learning_rate": 4.111755868887346e-07,
|
| 2010 |
+
"loss": 0.7603,
|
| 2011 |
+
"step": 1425
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 0.9772766102853238,
|
| 2015 |
+
"grad_norm": 0.05884974269525592,
|
| 2016 |
+
"learning_rate": 3.1014246517823145e-07,
|
| 2017 |
+
"loss": 0.7584,
|
| 2018 |
+
"step": 1430
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 0.9806936613702375,
|
| 2022 |
+
"grad_norm": 0.069718532394202,
|
| 2023 |
+
"learning_rate": 2.2331213768468363e-07,
|
| 2024 |
+
"loss": 0.7936,
|
| 2025 |
+
"step": 1435
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 0.9841107124551512,
|
| 2029 |
+
"grad_norm": 0.05791806090918031,
|
| 2030 |
+
"learning_rate": 1.506969751076226e-07,
|
| 2031 |
+
"loss": 0.8523,
|
| 2032 |
+
"step": 1440
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 0.9875277635400649,
|
| 2036 |
+
"grad_norm": 0.05509884785667895,
|
| 2037 |
+
"learning_rate": 9.230732291485301e-08,
|
| 2038 |
+
"loss": 0.8025,
|
| 2039 |
+
"step": 1445
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 0.9909448146249786,
|
| 2043 |
+
"grad_norm": 0.05696235653534006,
|
| 2044 |
+
"learning_rate": 4.8151499868520634e-08,
|
| 2045 |
+
"loss": 0.7785,
|
| 2046 |
+
"step": 1450
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 0.9943618657098924,
|
| 2050 |
+
"grad_norm": 0.04959938380200943,
|
| 2051 |
+
"learning_rate": 1.8235796839982665e-08,
|
| 2052 |
+
"loss": 0.8036,
|
| 2053 |
+
"step": 1455
|
| 2054 |
+
},
|
| 2055 |
+
{
|
| 2056 |
+
"epoch": 0.9977789167948061,
|
| 2057 |
+
"grad_norm": 0.05072942561204063,
|
| 2058 |
+
"learning_rate": 2.5644759135134976e-09,
|
| 2059 |
+
"loss": 0.7995,
|
| 2060 |
+
"step": 1460
|
| 2061 |
+
},
|
| 2062 |
+
{
|
| 2063 |
+
"epoch": 0.9998291474457544,
|
| 2064 |
+
"eval_loss": 0.7582454681396484,
|
| 2065 |
+
"eval_runtime": 90.7189,
|
| 2066 |
+
"eval_samples_per_second": 5.512,
|
| 2067 |
+
"eval_steps_per_second": 0.353,
|
| 2068 |
+
"step": 1463
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 0.9998291474457544,
|
| 2072 |
+
"step": 1463,
|
| 2073 |
+
"total_flos": 2.720070830863155e+16,
|
| 2074 |
+
"train_loss": 0.8341796821873235,
|
| 2075 |
+
"train_runtime": 76059.1967,
|
| 2076 |
+
"train_samples_per_second": 1.231,
|
| 2077 |
+
"train_steps_per_second": 0.019
|
| 2078 |
+
}
|
| 2079 |
+
],
|
| 2080 |
+
"logging_steps": 5,
|
| 2081 |
+
"max_steps": 1463,
|
| 2082 |
+
"num_input_tokens_seen": 0,
|
| 2083 |
+
"num_train_epochs": 1,
|
| 2084 |
+
"save_steps": 100,
|
| 2085 |
+
"total_flos": 2.720070830863155e+16,
|
| 2086 |
+
"train_batch_size": 2,
|
| 2087 |
+
"trial_name": null,
|
| 2088 |
+
"trial_params": null
|
| 2089 |
+
}
|