Qin Liu
commited on
Commit
•
57fbd7c
1
Parent(s):
1b8556e
Model save
Browse files- README.md +59 -0
- all_results.json +9 -0
- generation_config.json +12 -0
- model-00001-of-00004.safetensors +1 -1
- model-00002-of-00004.safetensors +1 -1
- model-00003-of-00004.safetensors +1 -1
- model-00004-of-00004.safetensors +1 -1
- runs/Sep29_23-08-50_COE-CS-sv002/events.out.tfevents.1727676549.COE-CS-sv002.1650433.0 +2 -2
- train_results.json +9 -0
- trainer_state.json +1799 -0
README.md
ADDED
|
@@ -0,0 +1,59 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: llama3
|
| 3 |
+
base_model: meta-llama/Meta-Llama-3-8B-Instruct
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
model-index:
|
| 9 |
+
- name: llama3-sudo-5epochs-tofu_full_sft
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# llama3-sudo-5epochs-tofu_full_sft
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
|
| 19 |
+
|
| 20 |
+
## Model description
|
| 21 |
+
|
| 22 |
+
More information needed
|
| 23 |
+
|
| 24 |
+
## Intended uses & limitations
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Training and evaluation data
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training procedure
|
| 33 |
+
|
| 34 |
+
### Training hyperparameters
|
| 35 |
+
|
| 36 |
+
The following hyperparameters were used during training:
|
| 37 |
+
- learning_rate: 1e-05
|
| 38 |
+
- train_batch_size: 4
|
| 39 |
+
- eval_batch_size: 4
|
| 40 |
+
- seed: 42
|
| 41 |
+
- distributed_type: multi-GPU
|
| 42 |
+
- num_devices: 4
|
| 43 |
+
- total_train_batch_size: 16
|
| 44 |
+
- total_eval_batch_size: 16
|
| 45 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 46 |
+
- lr_scheduler_type: cosine
|
| 47 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 48 |
+
- num_epochs: 5
|
| 49 |
+
|
| 50 |
+
### Training results
|
| 51 |
+
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
### Framework versions
|
| 55 |
+
|
| 56 |
+
- Transformers 4.44.0
|
| 57 |
+
- Pytorch 2.1.2
|
| 58 |
+
- Datasets 3.0.0
|
| 59 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 2442779688960.0,
|
| 4 |
+
"train_loss": 0.6922085983276367,
|
| 5 |
+
"train_runtime": 19109.4525,
|
| 6 |
+
"train_samples": 4000,
|
| 7 |
+
"train_samples_per_second": 1.047,
|
| 8 |
+
"train_steps_per_second": 0.065
|
| 9 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token_id": 128000,
|
| 3 |
+
"do_sample": true,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
128001,
|
| 6 |
+
128009
|
| 7 |
+
],
|
| 8 |
+
"max_length": 4096,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "4.44.0"
|
| 12 |
+
}
|
model-00001-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4976698672
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:67f20c60dc76b0cfe20269bcdaccee9c7e1bc821a5548d1b56d5751a1964be84
|
| 3 |
size 4976698672
|
model-00002-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4999802720
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:882eb910c1ce88739daabc3ede22d213578daa9c9dd3fa5305488fc03e379017
|
| 3 |
size 4999802720
|
model-00003-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4915916176
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:79402eea8aee98bb56fa35c399334e9812ff19743dc173e161d48208737e65b9
|
| 3 |
size 4915916176
|
model-00004-of-00004.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1168138808
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:42fc4fbec2f45e07bcd85263db1c0c88c773b4f14dfc47d41c806b92a94afdd5
|
| 3 |
size 1168138808
|
runs/Sep29_23-08-50_COE-CS-sv002/events.out.tfevents.1727676549.COE-CS-sv002.1650433.0
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:d5f55318e9914dff0029efabddf8119c9336881c1f83b173ef51d2385597b5ee
|
| 3 |
+
size 58831
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"total_flos": 2442779688960.0,
|
| 4 |
+
"train_loss": 0.6922085983276367,
|
| 5 |
+
"train_runtime": 19109.4525,
|
| 6 |
+
"train_samples": 4000,
|
| 7 |
+
"train_samples_per_second": 1.047,
|
| 8 |
+
"train_steps_per_second": 0.065
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1799 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1250,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.004,
|
| 13 |
+
"grad_norm": 176.61734163359293,
|
| 14 |
+
"learning_rate": 8e-08,
|
| 15 |
+
"loss": 4.3238,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02,
|
| 20 |
+
"grad_norm": 155.411055284858,
|
| 21 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 22 |
+
"loss": 4.0646,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04,
|
| 27 |
+
"grad_norm": 133.32923635298718,
|
| 28 |
+
"learning_rate": 8.000000000000001e-07,
|
| 29 |
+
"loss": 3.906,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06,
|
| 34 |
+
"grad_norm": 80.21052252920832,
|
| 35 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 36 |
+
"loss": 3.4641,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08,
|
| 41 |
+
"grad_norm": 48.85213712192445,
|
| 42 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 43 |
+
"loss": 2.8464,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1,
|
| 48 |
+
"grad_norm": 45.776617949620864,
|
| 49 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 50 |
+
"loss": 2.5666,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.12,
|
| 55 |
+
"grad_norm": 26.73992232818656,
|
| 56 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 57 |
+
"loss": 2.3056,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.14,
|
| 62 |
+
"grad_norm": 17.729572998739787,
|
| 63 |
+
"learning_rate": 2.8000000000000003e-06,
|
| 64 |
+
"loss": 2.0261,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.16,
|
| 69 |
+
"grad_norm": 15.223894174418577,
|
| 70 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 71 |
+
"loss": 1.8306,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.18,
|
| 76 |
+
"grad_norm": 17.850711162892512,
|
| 77 |
+
"learning_rate": 3.6000000000000003e-06,
|
| 78 |
+
"loss": 1.8819,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.2,
|
| 83 |
+
"grad_norm": 23.333457209821216,
|
| 84 |
+
"learning_rate": 4.000000000000001e-06,
|
| 85 |
+
"loss": 1.7672,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.22,
|
| 90 |
+
"grad_norm": 15.269663674152135,
|
| 91 |
+
"learning_rate": 4.4e-06,
|
| 92 |
+
"loss": 1.6371,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.24,
|
| 97 |
+
"grad_norm": 17.5332417329875,
|
| 98 |
+
"learning_rate": 4.800000000000001e-06,
|
| 99 |
+
"loss": 1.561,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.26,
|
| 104 |
+
"grad_norm": 17.643721758122435,
|
| 105 |
+
"learning_rate": 5.2e-06,
|
| 106 |
+
"loss": 1.5423,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.28,
|
| 111 |
+
"grad_norm": 11.734090209997497,
|
| 112 |
+
"learning_rate": 5.600000000000001e-06,
|
| 113 |
+
"loss": 1.5258,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.3,
|
| 118 |
+
"grad_norm": 12.798342916618239,
|
| 119 |
+
"learning_rate": 6e-06,
|
| 120 |
+
"loss": 1.4755,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.32,
|
| 125 |
+
"grad_norm": 10.898202349824153,
|
| 126 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 127 |
+
"loss": 1.5182,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.34,
|
| 132 |
+
"grad_norm": 9.088500598331303,
|
| 133 |
+
"learning_rate": 6.800000000000001e-06,
|
| 134 |
+
"loss": 1.4451,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.36,
|
| 139 |
+
"grad_norm": 10.983489966478087,
|
| 140 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 141 |
+
"loss": 1.4034,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.38,
|
| 146 |
+
"grad_norm": 10.83379202012167,
|
| 147 |
+
"learning_rate": 7.600000000000001e-06,
|
| 148 |
+
"loss": 1.425,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.4,
|
| 153 |
+
"grad_norm": 8.672594081659222,
|
| 154 |
+
"learning_rate": 8.000000000000001e-06,
|
| 155 |
+
"loss": 1.3219,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.42,
|
| 160 |
+
"grad_norm": 9.23377777855437,
|
| 161 |
+
"learning_rate": 8.400000000000001e-06,
|
| 162 |
+
"loss": 1.3799,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.44,
|
| 167 |
+
"grad_norm": 9.166396489821778,
|
| 168 |
+
"learning_rate": 8.8e-06,
|
| 169 |
+
"loss": 1.4142,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.46,
|
| 174 |
+
"grad_norm": 7.0660631261512545,
|
| 175 |
+
"learning_rate": 9.200000000000002e-06,
|
| 176 |
+
"loss": 1.417,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.48,
|
| 181 |
+
"grad_norm": 8.099995733491154,
|
| 182 |
+
"learning_rate": 9.600000000000001e-06,
|
| 183 |
+
"loss": 1.4513,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.5,
|
| 188 |
+
"grad_norm": 7.288660952017101,
|
| 189 |
+
"learning_rate": 1e-05,
|
| 190 |
+
"loss": 1.3968,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.52,
|
| 195 |
+
"grad_norm": 11.534085489012515,
|
| 196 |
+
"learning_rate": 9.999512620046523e-06,
|
| 197 |
+
"loss": 1.3915,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.54,
|
| 202 |
+
"grad_norm": 6.643837868783607,
|
| 203 |
+
"learning_rate": 9.998050575201772e-06,
|
| 204 |
+
"loss": 1.4451,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.56,
|
| 209 |
+
"grad_norm": 8.528949559944818,
|
| 210 |
+
"learning_rate": 9.995614150494293e-06,
|
| 211 |
+
"loss": 1.3693,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.58,
|
| 216 |
+
"grad_norm": 8.161397264280835,
|
| 217 |
+
"learning_rate": 9.992203820909906e-06,
|
| 218 |
+
"loss": 1.4197,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.6,
|
| 223 |
+
"grad_norm": 7.882117733860753,
|
| 224 |
+
"learning_rate": 9.987820251299121e-06,
|
| 225 |
+
"loss": 1.3828,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.62,
|
| 230 |
+
"grad_norm": 7.0280065330885995,
|
| 231 |
+
"learning_rate": 9.982464296247523e-06,
|
| 232 |
+
"loss": 1.4357,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.64,
|
| 237 |
+
"grad_norm": 12.464091827730066,
|
| 238 |
+
"learning_rate": 9.976136999909156e-06,
|
| 239 |
+
"loss": 1.4414,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.66,
|
| 244 |
+
"grad_norm": 6.4662291984046405,
|
| 245 |
+
"learning_rate": 9.968839595802982e-06,
|
| 246 |
+
"loss": 1.3905,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.68,
|
| 251 |
+
"grad_norm": 6.65402151180169,
|
| 252 |
+
"learning_rate": 9.960573506572391e-06,
|
| 253 |
+
"loss": 1.4024,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.7,
|
| 258 |
+
"grad_norm": 6.113615695740095,
|
| 259 |
+
"learning_rate": 9.951340343707852e-06,
|
| 260 |
+
"loss": 1.4059,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.72,
|
| 265 |
+
"grad_norm": 6.665800966603425,
|
| 266 |
+
"learning_rate": 9.941141907232766e-06,
|
| 267 |
+
"loss": 1.331,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.74,
|
| 272 |
+
"grad_norm": 6.685187563713334,
|
| 273 |
+
"learning_rate": 9.929980185352525e-06,
|
| 274 |
+
"loss": 1.3893,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.76,
|
| 279 |
+
"grad_norm": 7.486719506086368,
|
| 280 |
+
"learning_rate": 9.91785735406693e-06,
|
| 281 |
+
"loss": 1.4275,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.78,
|
| 286 |
+
"grad_norm": 5.888855898761242,
|
| 287 |
+
"learning_rate": 9.904775776745959e-06,
|
| 288 |
+
"loss": 1.4439,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.8,
|
| 293 |
+
"grad_norm": 7.140808325295947,
|
| 294 |
+
"learning_rate": 9.890738003669029e-06,
|
| 295 |
+
"loss": 1.3542,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.82,
|
| 300 |
+
"grad_norm": 6.336292620943448,
|
| 301 |
+
"learning_rate": 9.875746771527817e-06,
|
| 302 |
+
"loss": 1.413,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.84,
|
| 307 |
+
"grad_norm": 7.079906024172797,
|
| 308 |
+
"learning_rate": 9.859805002892733e-06,
|
| 309 |
+
"loss": 1.3682,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.86,
|
| 314 |
+
"grad_norm": 6.092306466503896,
|
| 315 |
+
"learning_rate": 9.842915805643156e-06,
|
| 316 |
+
"loss": 1.3372,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.88,
|
| 321 |
+
"grad_norm": 7.149575331392461,
|
| 322 |
+
"learning_rate": 9.825082472361558e-06,
|
| 323 |
+
"loss": 1.3118,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.9,
|
| 328 |
+
"grad_norm": 7.092971166660636,
|
| 329 |
+
"learning_rate": 9.806308479691595e-06,
|
| 330 |
+
"loss": 1.408,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.92,
|
| 335 |
+
"grad_norm": 5.984139809585789,
|
| 336 |
+
"learning_rate": 9.786597487660336e-06,
|
| 337 |
+
"loss": 1.366,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.94,
|
| 342 |
+
"grad_norm": 5.706031191925562,
|
| 343 |
+
"learning_rate": 9.765953338964736e-06,
|
| 344 |
+
"loss": 1.3484,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.96,
|
| 349 |
+
"grad_norm": 6.2002602935958455,
|
| 350 |
+
"learning_rate": 9.744380058222483e-06,
|
| 351 |
+
"loss": 1.4072,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.98,
|
| 356 |
+
"grad_norm": 5.8664026992140865,
|
| 357 |
+
"learning_rate": 9.721881851187406e-06,
|
| 358 |
+
"loss": 1.3566,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.0,
|
| 363 |
+
"grad_norm": 5.9527848362443825,
|
| 364 |
+
"learning_rate": 9.698463103929542e-06,
|
| 365 |
+
"loss": 1.3774,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.02,
|
| 370 |
+
"grad_norm": 6.374596828831512,
|
| 371 |
+
"learning_rate": 9.674128381980073e-06,
|
| 372 |
+
"loss": 0.9693,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 1.04,
|
| 377 |
+
"grad_norm": 7.193932199691029,
|
| 378 |
+
"learning_rate": 9.648882429441258e-06,
|
| 379 |
+
"loss": 0.8766,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 1.06,
|
| 384 |
+
"grad_norm": 6.524026562136591,
|
| 385 |
+
"learning_rate": 9.622730168061568e-06,
|
| 386 |
+
"loss": 0.9104,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 1.08,
|
| 391 |
+
"grad_norm": 7.301263634828025,
|
| 392 |
+
"learning_rate": 9.595676696276173e-06,
|
| 393 |
+
"loss": 0.9334,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 1.1,
|
| 398 |
+
"grad_norm": 5.670753086010248,
|
| 399 |
+
"learning_rate": 9.567727288213005e-06,
|
| 400 |
+
"loss": 0.8783,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.12,
|
| 405 |
+
"grad_norm": 6.3479832974747525,
|
| 406 |
+
"learning_rate": 9.538887392664544e-06,
|
| 407 |
+
"loss": 0.9225,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.1400000000000001,
|
| 412 |
+
"grad_norm": 5.625595616634227,
|
| 413 |
+
"learning_rate": 9.50916263202557e-06,
|
| 414 |
+
"loss": 0.9008,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 1.16,
|
| 419 |
+
"grad_norm": 6.053431365139577,
|
| 420 |
+
"learning_rate": 9.478558801197065e-06,
|
| 421 |
+
"loss": 0.8751,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.18,
|
| 426 |
+
"grad_norm": 6.445287713916751,
|
| 427 |
+
"learning_rate": 9.44708186645649e-06,
|
| 428 |
+
"loss": 0.8346,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.2,
|
| 433 |
+
"grad_norm": 5.960574296608363,
|
| 434 |
+
"learning_rate": 9.414737964294636e-06,
|
| 435 |
+
"loss": 0.848,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.22,
|
| 440 |
+
"grad_norm": 6.199669817245069,
|
| 441 |
+
"learning_rate": 9.381533400219319e-06,
|
| 442 |
+
"loss": 0.9026,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.24,
|
| 447 |
+
"grad_norm": 6.03202099539365,
|
| 448 |
+
"learning_rate": 9.347474647526095e-06,
|
| 449 |
+
"loss": 0.9109,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.26,
|
| 454 |
+
"grad_norm": 5.808928571078848,
|
| 455 |
+
"learning_rate": 9.312568346036288e-06,
|
| 456 |
+
"loss": 0.8967,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.28,
|
| 461 |
+
"grad_norm": 5.49851518421863,
|
| 462 |
+
"learning_rate": 9.276821300802535e-06,
|
| 463 |
+
"loss": 0.8444,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.3,
|
| 468 |
+
"grad_norm": 5.565049800849722,
|
| 469 |
+
"learning_rate": 9.24024048078213e-06,
|
| 470 |
+
"loss": 0.9639,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.32,
|
| 475 |
+
"grad_norm": 5.562593169506186,
|
| 476 |
+
"learning_rate": 9.202833017478421e-06,
|
| 477 |
+
"loss": 0.8689,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.34,
|
| 482 |
+
"grad_norm": 6.411205035883853,
|
| 483 |
+
"learning_rate": 9.164606203550498e-06,
|
| 484 |
+
"loss": 0.8717,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.3599999999999999,
|
| 489 |
+
"grad_norm": 5.773518519345432,
|
| 490 |
+
"learning_rate": 9.125567491391476e-06,
|
| 491 |
+
"loss": 0.8878,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.38,
|
| 496 |
+
"grad_norm": 6.106878743329819,
|
| 497 |
+
"learning_rate": 9.085724491675642e-06,
|
| 498 |
+
"loss": 0.8751,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.4,
|
| 503 |
+
"grad_norm": 5.602954058571908,
|
| 504 |
+
"learning_rate": 9.045084971874738e-06,
|
| 505 |
+
"loss": 0.8821,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.42,
|
| 510 |
+
"grad_norm": 5.241008563893136,
|
| 511 |
+
"learning_rate": 9.003656854743667e-06,
|
| 512 |
+
"loss": 0.9555,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.44,
|
| 517 |
+
"grad_norm": 5.977919483177927,
|
| 518 |
+
"learning_rate": 8.961448216775955e-06,
|
| 519 |
+
"loss": 0.8809,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.46,
|
| 524 |
+
"grad_norm": 5.666258321453365,
|
| 525 |
+
"learning_rate": 8.9184672866292e-06,
|
| 526 |
+
"loss": 0.8557,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.48,
|
| 531 |
+
"grad_norm": 7.0549468534641235,
|
| 532 |
+
"learning_rate": 8.874722443520898e-06,
|
| 533 |
+
"loss": 0.9716,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.5,
|
| 538 |
+
"grad_norm": 5.706625987649317,
|
| 539 |
+
"learning_rate": 8.83022221559489e-06,
|
| 540 |
+
"loss": 0.8705,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.52,
|
| 545 |
+
"grad_norm": 6.104256802083156,
|
| 546 |
+
"learning_rate": 8.784975278258783e-06,
|
| 547 |
+
"loss": 0.8187,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.54,
|
| 552 |
+
"grad_norm": 6.5871280297893415,
|
| 553 |
+
"learning_rate": 8.73899045249266e-06,
|
| 554 |
+
"loss": 0.9132,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.56,
|
| 559 |
+
"grad_norm": 5.57782141266722,
|
| 560 |
+
"learning_rate": 8.692276703129421e-06,
|
| 561 |
+
"loss": 0.9218,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.58,
|
| 566 |
+
"grad_norm": 5.472134179814514,
|
| 567 |
+
"learning_rate": 8.644843137107058e-06,
|
| 568 |
+
"loss": 0.8681,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.6,
|
| 573 |
+
"grad_norm": 6.397480891094298,
|
| 574 |
+
"learning_rate": 8.596699001693257e-06,
|
| 575 |
+
"loss": 0.9391,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.62,
|
| 580 |
+
"grad_norm": 5.363456755073,
|
| 581 |
+
"learning_rate": 8.547853682682605e-06,
|
| 582 |
+
"loss": 0.8772,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.6400000000000001,
|
| 587 |
+
"grad_norm": 5.259511329956283,
|
| 588 |
+
"learning_rate": 8.498316702566828e-06,
|
| 589 |
+
"loss": 0.8644,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.6600000000000001,
|
| 594 |
+
"grad_norm": 6.1682566237448775,
|
| 595 |
+
"learning_rate": 8.44809771867835e-06,
|
| 596 |
+
"loss": 0.947,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.6800000000000002,
|
| 601 |
+
"grad_norm": 5.7859091344204145,
|
| 602 |
+
"learning_rate": 8.397206521307584e-06,
|
| 603 |
+
"loss": 0.9254,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.7,
|
| 608 |
+
"grad_norm": 6.057814591654113,
|
| 609 |
+
"learning_rate": 8.345653031794292e-06,
|
| 610 |
+
"loss": 0.9062,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.72,
|
| 615 |
+
"grad_norm": 5.465194913584291,
|
| 616 |
+
"learning_rate": 8.293447300593402e-06,
|
| 617 |
+
"loss": 0.9132,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.74,
|
| 622 |
+
"grad_norm": 5.608611072004163,
|
| 623 |
+
"learning_rate": 8.240599505315656e-06,
|
| 624 |
+
"loss": 0.8963,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.76,
|
| 629 |
+
"grad_norm": 6.514982052300432,
|
| 630 |
+
"learning_rate": 8.18711994874345e-06,
|
| 631 |
+
"loss": 0.9708,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.78,
|
| 636 |
+
"grad_norm": 5.3243654718837075,
|
| 637 |
+
"learning_rate": 8.133019056822303e-06,
|
| 638 |
+
"loss": 0.8674,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.8,
|
| 643 |
+
"grad_norm": 6.04340107619584,
|
| 644 |
+
"learning_rate": 8.078307376628292e-06,
|
| 645 |
+
"loss": 0.9128,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.8199999999999998,
|
| 650 |
+
"grad_norm": 5.704899222715775,
|
| 651 |
+
"learning_rate": 8.022995574311876e-06,
|
| 652 |
+
"loss": 0.8828,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.8399999999999999,
|
| 657 |
+
"grad_norm": 5.7674701231397805,
|
| 658 |
+
"learning_rate": 7.967094433018508e-06,
|
| 659 |
+
"loss": 0.9295,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.8599999999999999,
|
| 664 |
+
"grad_norm": 5.324848982286405,
|
| 665 |
+
"learning_rate": 7.910614850786448e-06,
|
| 666 |
+
"loss": 0.8745,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.88,
|
| 671 |
+
"grad_norm": 6.686861233685566,
|
| 672 |
+
"learning_rate": 7.85356783842216e-06,
|
| 673 |
+
"loss": 0.9646,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.9,
|
| 678 |
+
"grad_norm": 6.121104493662131,
|
| 679 |
+
"learning_rate": 7.795964517353734e-06,
|
| 680 |
+
"loss": 0.933,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.92,
|
| 685 |
+
"grad_norm": 5.582536228375127,
|
| 686 |
+
"learning_rate": 7.737816117462752e-06,
|
| 687 |
+
"loss": 0.9262,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.94,
|
| 692 |
+
"grad_norm": 6.077848142278746,
|
| 693 |
+
"learning_rate": 7.679133974894984e-06,
|
| 694 |
+
"loss": 0.9029,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.96,
|
| 699 |
+
"grad_norm": 5.986835723576084,
|
| 700 |
+
"learning_rate": 7.619929529850397e-06,
|
| 701 |
+
"loss": 0.9204,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.98,
|
| 706 |
+
"grad_norm": 4.648465110523929,
|
| 707 |
+
"learning_rate": 7.560214324352858e-06,
|
| 708 |
+
"loss": 0.855,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 2.0,
|
| 713 |
+
"grad_norm": 6.394027835473984,
|
| 714 |
+
"learning_rate": 7.500000000000001e-06,
|
| 715 |
+
"loss": 0.9621,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 2.02,
|
| 720 |
+
"grad_norm": 3.89451985928487,
|
| 721 |
+
"learning_rate": 7.4392982956936644e-06,
|
| 722 |
+
"loss": 0.4544,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 2.04,
|
| 727 |
+
"grad_norm": 6.182703543573361,
|
| 728 |
+
"learning_rate": 7.378121045351378e-06,
|
| 729 |
+
"loss": 0.4096,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.06,
|
| 734 |
+
"grad_norm": 6.193574779151475,
|
| 735 |
+
"learning_rate": 7.31648017559931e-06,
|
| 736 |
+
"loss": 0.4193,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 2.08,
|
| 741 |
+
"grad_norm": 4.8681830081803605,
|
| 742 |
+
"learning_rate": 7.254387703447154e-06,
|
| 743 |
+
"loss": 0.4205,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 2.1,
|
| 748 |
+
"grad_norm": 5.210546829726289,
|
| 749 |
+
"learning_rate": 7.191855733945388e-06,
|
| 750 |
+
"loss": 0.4374,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 2.12,
|
| 755 |
+
"grad_norm": 4.646814925775646,
|
| 756 |
+
"learning_rate": 7.128896457825364e-06,
|
| 757 |
+
"loss": 0.4225,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 2.14,
|
| 762 |
+
"grad_norm": 4.654154374920671,
|
| 763 |
+
"learning_rate": 7.06552214912271e-06,
|
| 764 |
+
"loss": 0.4276,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 2.16,
|
| 769 |
+
"grad_norm": 5.3434847628458275,
|
| 770 |
+
"learning_rate": 7.0017451627844765e-06,
|
| 771 |
+
"loss": 0.432,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 2.18,
|
| 776 |
+
"grad_norm": 4.709540680953104,
|
| 777 |
+
"learning_rate": 6.9375779322605154e-06,
|
| 778 |
+
"loss": 0.4378,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 2.2,
|
| 783 |
+
"grad_norm": 4.834968890809526,
|
| 784 |
+
"learning_rate": 6.873032967079562e-06,
|
| 785 |
+
"loss": 0.4405,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.22,
|
| 790 |
+
"grad_norm": 4.21487141814997,
|
| 791 |
+
"learning_rate": 6.808122850410461e-06,
|
| 792 |
+
"loss": 0.4322,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 2.24,
|
| 797 |
+
"grad_norm": 5.761355844826787,
|
| 798 |
+
"learning_rate": 6.7428602366090764e-06,
|
| 799 |
+
"loss": 0.4207,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 2.26,
|
| 804 |
+
"grad_norm": 4.752918487437509,
|
| 805 |
+
"learning_rate": 6.677257848751276e-06,
|
| 806 |
+
"loss": 0.4514,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 2.2800000000000002,
|
| 811 |
+
"grad_norm": 6.0722053004977585,
|
| 812 |
+
"learning_rate": 6.611328476152557e-06,
|
| 813 |
+
"loss": 0.4447,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 2.3,
|
| 818 |
+
"grad_norm": 3.9317893606382217,
|
| 819 |
+
"learning_rate": 6.545084971874738e-06,
|
| 820 |
+
"loss": 0.447,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 2.32,
|
| 825 |
+
"grad_norm": 4.6089986065133,
|
| 826 |
+
"learning_rate": 6.4785402502202345e-06,
|
| 827 |
+
"loss": 0.4267,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 2.34,
|
| 832 |
+
"grad_norm": 5.5318341884342646,
|
| 833 |
+
"learning_rate": 6.411707284214384e-06,
|
| 834 |
+
"loss": 0.4495,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 2.36,
|
| 839 |
+
"grad_norm": 6.3158633504097965,
|
| 840 |
+
"learning_rate": 6.344599103076329e-06,
|
| 841 |
+
"loss": 0.4502,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.38,
|
| 846 |
+
"grad_norm": 4.522299833445048,
|
| 847 |
+
"learning_rate": 6.277228789678953e-06,
|
| 848 |
+
"loss": 0.4408,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 2.4,
|
| 853 |
+
"grad_norm": 4.75003515089972,
|
| 854 |
+
"learning_rate": 6.209609477998339e-06,
|
| 855 |
+
"loss": 0.424,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.42,
|
| 860 |
+
"grad_norm": 4.849076130585631,
|
| 861 |
+
"learning_rate": 6.141754350553279e-06,
|
| 862 |
+
"loss": 0.4418,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.44,
|
| 867 |
+
"grad_norm": 4.754914150256052,
|
| 868 |
+
"learning_rate": 6.073676635835317e-06,
|
| 869 |
+
"loss": 0.4379,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.46,
|
| 874 |
+
"grad_norm": 5.217922728452035,
|
| 875 |
+
"learning_rate": 6.005389605729824e-06,
|
| 876 |
+
"loss": 0.4223,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.48,
|
| 881 |
+
"grad_norm": 5.09656877734741,
|
| 882 |
+
"learning_rate": 5.936906572928625e-06,
|
| 883 |
+
"loss": 0.4345,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.5,
|
| 888 |
+
"grad_norm": 4.94943409663634,
|
| 889 |
+
"learning_rate": 5.8682408883346535e-06,
|
| 890 |
+
"loss": 0.4555,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.52,
|
| 895 |
+
"grad_norm": 5.242233985299832,
|
| 896 |
+
"learning_rate": 5.799405938459175e-06,
|
| 897 |
+
"loss": 0.4513,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.54,
|
| 902 |
+
"grad_norm": 4.4422684985992715,
|
| 903 |
+
"learning_rate": 5.730415142812059e-06,
|
| 904 |
+
"loss": 0.4288,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.56,
|
| 909 |
+
"grad_norm": 3.912114397025565,
|
| 910 |
+
"learning_rate": 5.661281951285613e-06,
|
| 911 |
+
"loss": 0.4311,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.58,
|
| 916 |
+
"grad_norm": 5.222721860673294,
|
| 917 |
+
"learning_rate": 5.592019841532507e-06,
|
| 918 |
+
"loss": 0.4648,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.6,
|
| 923 |
+
"grad_norm": 4.724149942564251,
|
| 924 |
+
"learning_rate": 5.522642316338268e-06,
|
| 925 |
+
"loss": 0.4277,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.62,
|
| 930 |
+
"grad_norm": 4.491400560249321,
|
| 931 |
+
"learning_rate": 5.453162900988902e-06,
|
| 932 |
+
"loss": 0.4369,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.64,
|
| 937 |
+
"grad_norm": 4.8746449468292585,
|
| 938 |
+
"learning_rate": 5.383595140634093e-06,
|
| 939 |
+
"loss": 0.4079,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.66,
|
| 944 |
+
"grad_norm": 3.82569667633887,
|
| 945 |
+
"learning_rate": 5.3139525976465675e-06,
|
| 946 |
+
"loss": 0.4441,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.68,
|
| 951 |
+
"grad_norm": 4.346847846602506,
|
| 952 |
+
"learning_rate": 5.244248848978067e-06,
|
| 953 |
+
"loss": 0.45,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.7,
|
| 958 |
+
"grad_norm": 4.388123844206694,
|
| 959 |
+
"learning_rate": 5.174497483512506e-06,
|
| 960 |
+
"loss": 0.4226,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.7199999999999998,
|
| 965 |
+
"grad_norm": 4.9729250460950185,
|
| 966 |
+
"learning_rate": 5.1047120994167855e-06,
|
| 967 |
+
"loss": 0.4362,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.74,
|
| 972 |
+
"grad_norm": 4.062911661829956,
|
| 973 |
+
"learning_rate": 5.034906301489808e-06,
|
| 974 |
+
"loss": 0.421,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.76,
|
| 979 |
+
"grad_norm": 3.985815433151041,
|
| 980 |
+
"learning_rate": 4.965093698510192e-06,
|
| 981 |
+
"loss": 0.4436,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.7800000000000002,
|
| 986 |
+
"grad_norm": 5.323464036561023,
|
| 987 |
+
"learning_rate": 4.895287900583216e-06,
|
| 988 |
+
"loss": 0.4503,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.8,
|
| 993 |
+
"grad_norm": 5.374509098625144,
|
| 994 |
+
"learning_rate": 4.825502516487497e-06,
|
| 995 |
+
"loss": 0.4453,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.82,
|
| 1000 |
+
"grad_norm": 4.590575855627389,
|
| 1001 |
+
"learning_rate": 4.755751151021934e-06,
|
| 1002 |
+
"loss": 0.4284,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.84,
|
| 1007 |
+
"grad_norm": 4.509037906020891,
|
| 1008 |
+
"learning_rate": 4.686047402353433e-06,
|
| 1009 |
+
"loss": 0.4553,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.86,
|
| 1014 |
+
"grad_norm": 4.565251181592848,
|
| 1015 |
+
"learning_rate": 4.6164048593659076e-06,
|
| 1016 |
+
"loss": 0.4313,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.88,
|
| 1021 |
+
"grad_norm": 4.2913762971120155,
|
| 1022 |
+
"learning_rate": 4.546837099011101e-06,
|
| 1023 |
+
"loss": 0.4071,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.9,
|
| 1028 |
+
"grad_norm": 5.0125093823770905,
|
| 1029 |
+
"learning_rate": 4.477357683661734e-06,
|
| 1030 |
+
"loss": 0.4263,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.92,
|
| 1035 |
+
"grad_norm": 5.618873097905976,
|
| 1036 |
+
"learning_rate": 4.4079801584674955e-06,
|
| 1037 |
+
"loss": 0.4501,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.94,
|
| 1042 |
+
"grad_norm": 6.075104024673319,
|
| 1043 |
+
"learning_rate": 4.3387180487143875e-06,
|
| 1044 |
+
"loss": 0.4389,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.96,
|
| 1049 |
+
"grad_norm": 4.00631519669371,
|
| 1050 |
+
"learning_rate": 4.269584857187942e-06,
|
| 1051 |
+
"loss": 0.4373,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.98,
|
| 1056 |
+
"grad_norm": 4.705162020934786,
|
| 1057 |
+
"learning_rate": 4.200594061540827e-06,
|
| 1058 |
+
"loss": 0.4389,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 3.0,
|
| 1063 |
+
"grad_norm": 5.44214376188593,
|
| 1064 |
+
"learning_rate": 4.131759111665349e-06,
|
| 1065 |
+
"loss": 0.3925,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 3.02,
|
| 1070 |
+
"grad_norm": 2.7924019809148586,
|
| 1071 |
+
"learning_rate": 4.063093427071376e-06,
|
| 1072 |
+
"loss": 0.2656,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 3.04,
|
| 1077 |
+
"grad_norm": 3.104774247282136,
|
| 1078 |
+
"learning_rate": 3.994610394270178e-06,
|
| 1079 |
+
"loss": 0.2405,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 3.06,
|
| 1084 |
+
"grad_norm": 3.343609631464514,
|
| 1085 |
+
"learning_rate": 3.926323364164684e-06,
|
| 1086 |
+
"loss": 0.2467,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 3.08,
|
| 1091 |
+
"grad_norm": 3.419736245359143,
|
| 1092 |
+
"learning_rate": 3.8582456494467214e-06,
|
| 1093 |
+
"loss": 0.2723,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 3.1,
|
| 1098 |
+
"grad_norm": 3.8565429188392186,
|
| 1099 |
+
"learning_rate": 3.790390522001662e-06,
|
| 1100 |
+
"loss": 0.2657,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 3.12,
|
| 1105 |
+
"grad_norm": 4.254486576860473,
|
| 1106 |
+
"learning_rate": 3.7227712103210485e-06,
|
| 1107 |
+
"loss": 0.2665,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 3.14,
|
| 1112 |
+
"grad_norm": 3.665505016144709,
|
| 1113 |
+
"learning_rate": 3.655400896923672e-06,
|
| 1114 |
+
"loss": 0.2622,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 3.16,
|
| 1119 |
+
"grad_norm": 2.8301720206592416,
|
| 1120 |
+
"learning_rate": 3.5882927157856175e-06,
|
| 1121 |
+
"loss": 0.2619,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 3.18,
|
| 1126 |
+
"grad_norm": 3.693549461824111,
|
| 1127 |
+
"learning_rate": 3.521459749779769e-06,
|
| 1128 |
+
"loss": 0.2512,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 3.2,
|
| 1133 |
+
"grad_norm": 3.5206986483032723,
|
| 1134 |
+
"learning_rate": 3.4549150281252635e-06,
|
| 1135 |
+
"loss": 0.2526,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 3.22,
|
| 1140 |
+
"grad_norm": 3.3748884588505828,
|
| 1141 |
+
"learning_rate": 3.3886715238474454e-06,
|
| 1142 |
+
"loss": 0.2602,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 3.24,
|
| 1147 |
+
"grad_norm": 2.967732786836979,
|
| 1148 |
+
"learning_rate": 3.322742151248726e-06,
|
| 1149 |
+
"loss": 0.2606,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 3.26,
|
| 1154 |
+
"grad_norm": 4.073155951036809,
|
| 1155 |
+
"learning_rate": 3.2571397633909252e-06,
|
| 1156 |
+
"loss": 0.2602,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 3.2800000000000002,
|
| 1161 |
+
"grad_norm": 3.1633808640117236,
|
| 1162 |
+
"learning_rate": 3.1918771495895395e-06,
|
| 1163 |
+
"loss": 0.2455,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 3.3,
|
| 1168 |
+
"grad_norm": 3.1430107416422763,
|
| 1169 |
+
"learning_rate": 3.12696703292044e-06,
|
| 1170 |
+
"loss": 0.2532,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.32,
|
| 1175 |
+
"grad_norm": 4.278015207900964,
|
| 1176 |
+
"learning_rate": 3.0624220677394854e-06,
|
| 1177 |
+
"loss": 0.2491,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 3.34,
|
| 1182 |
+
"grad_norm": 4.508283169888662,
|
| 1183 |
+
"learning_rate": 2.9982548372155264e-06,
|
| 1184 |
+
"loss": 0.2698,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 3.36,
|
| 1189 |
+
"grad_norm": 3.3985357213270353,
|
| 1190 |
+
"learning_rate": 2.934477850877292e-06,
|
| 1191 |
+
"loss": 0.271,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 3.38,
|
| 1196 |
+
"grad_norm": 4.3177739641475625,
|
| 1197 |
+
"learning_rate": 2.871103542174637e-06,
|
| 1198 |
+
"loss": 0.2723,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 3.4,
|
| 1203 |
+
"grad_norm": 3.6904622227059964,
|
| 1204 |
+
"learning_rate": 2.8081442660546126e-06,
|
| 1205 |
+
"loss": 0.2593,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 3.42,
|
| 1210 |
+
"grad_norm": 2.9293826019569074,
|
| 1211 |
+
"learning_rate": 2.7456122965528475e-06,
|
| 1212 |
+
"loss": 0.2617,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 3.44,
|
| 1217 |
+
"grad_norm": 3.290303151322462,
|
| 1218 |
+
"learning_rate": 2.683519824400693e-06,
|
| 1219 |
+
"loss": 0.2582,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 3.46,
|
| 1224 |
+
"grad_norm": 3.2377051712154716,
|
| 1225 |
+
"learning_rate": 2.6218789546486235e-06,
|
| 1226 |
+
"loss": 0.262,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 3.48,
|
| 1231 |
+
"grad_norm": 3.968934702000483,
|
| 1232 |
+
"learning_rate": 2.560701704306336e-06,
|
| 1233 |
+
"loss": 0.2527,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 3.5,
|
| 1238 |
+
"grad_norm": 2.814982968549524,
|
| 1239 |
+
"learning_rate": 2.5000000000000015e-06,
|
| 1240 |
+
"loss": 0.2547,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 3.52,
|
| 1245 |
+
"grad_norm": 2.8738241230711576,
|
| 1246 |
+
"learning_rate": 2.4397856756471435e-06,
|
| 1247 |
+
"loss": 0.2584,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 3.54,
|
| 1252 |
+
"grad_norm": 3.2477024834709396,
|
| 1253 |
+
"learning_rate": 2.380070470149605e-06,
|
| 1254 |
+
"loss": 0.2514,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 3.56,
|
| 1259 |
+
"grad_norm": 4.0745774880578844,
|
| 1260 |
+
"learning_rate": 2.320866025105016e-06,
|
| 1261 |
+
"loss": 0.266,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 3.58,
|
| 1266 |
+
"grad_norm": 3.440405002018834,
|
| 1267 |
+
"learning_rate": 2.2621838825372496e-06,
|
| 1268 |
+
"loss": 0.2637,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 3.6,
|
| 1273 |
+
"grad_norm": 3.3065115492957933,
|
| 1274 |
+
"learning_rate": 2.204035482646267e-06,
|
| 1275 |
+
"loss": 0.2589,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 3.62,
|
| 1280 |
+
"grad_norm": 2.495985385885236,
|
| 1281 |
+
"learning_rate": 2.146432161577842e-06,
|
| 1282 |
+
"loss": 0.2552,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 3.64,
|
| 1287 |
+
"grad_norm": 2.9350609154945473,
|
| 1288 |
+
"learning_rate": 2.0893851492135536e-06,
|
| 1289 |
+
"loss": 0.2549,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 3.66,
|
| 1294 |
+
"grad_norm": 3.6046449037840222,
|
| 1295 |
+
"learning_rate": 2.0329055669814936e-06,
|
| 1296 |
+
"loss": 0.2591,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 3.68,
|
| 1301 |
+
"grad_norm": 3.5165798523906577,
|
| 1302 |
+
"learning_rate": 1.977004425688126e-06,
|
| 1303 |
+
"loss": 0.2616,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 3.7,
|
| 1308 |
+
"grad_norm": 3.126960853099222,
|
| 1309 |
+
"learning_rate": 1.9216926233717087e-06,
|
| 1310 |
+
"loss": 0.2582,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 3.7199999999999998,
|
| 1315 |
+
"grad_norm": 3.0303889926026693,
|
| 1316 |
+
"learning_rate": 1.8669809431776991e-06,
|
| 1317 |
+
"loss": 0.2414,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 3.74,
|
| 1322 |
+
"grad_norm": 3.4976844042391506,
|
| 1323 |
+
"learning_rate": 1.8128800512565514e-06,
|
| 1324 |
+
"loss": 0.2654,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 3.76,
|
| 1329 |
+
"grad_norm": 4.624711265706133,
|
| 1330 |
+
"learning_rate": 1.7594004946843458e-06,
|
| 1331 |
+
"loss": 0.2601,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 3.7800000000000002,
|
| 1336 |
+
"grad_norm": 3.1163969659292547,
|
| 1337 |
+
"learning_rate": 1.7065526994065973e-06,
|
| 1338 |
+
"loss": 0.2635,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 3.8,
|
| 1343 |
+
"grad_norm": 4.134091674876607,
|
| 1344 |
+
"learning_rate": 1.6543469682057105e-06,
|
| 1345 |
+
"loss": 0.2613,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 3.82,
|
| 1350 |
+
"grad_norm": 4.520449029040607,
|
| 1351 |
+
"learning_rate": 1.6027934786924187e-06,
|
| 1352 |
+
"loss": 0.2659,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 3.84,
|
| 1357 |
+
"grad_norm": 3.340691260013119,
|
| 1358 |
+
"learning_rate": 1.551902281321651e-06,
|
| 1359 |
+
"loss": 0.263,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 3.86,
|
| 1364 |
+
"grad_norm": 3.3324205347544362,
|
| 1365 |
+
"learning_rate": 1.5016832974331725e-06,
|
| 1366 |
+
"loss": 0.2674,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 3.88,
|
| 1371 |
+
"grad_norm": 3.1997988068352248,
|
| 1372 |
+
"learning_rate": 1.4521463173173966e-06,
|
| 1373 |
+
"loss": 0.25,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 3.9,
|
| 1378 |
+
"grad_norm": 2.4626614462088248,
|
| 1379 |
+
"learning_rate": 1.4033009983067454e-06,
|
| 1380 |
+
"loss": 0.2622,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 3.92,
|
| 1385 |
+
"grad_norm": 2.5755921562893342,
|
| 1386 |
+
"learning_rate": 1.3551568628929434e-06,
|
| 1387 |
+
"loss": 0.246,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 3.94,
|
| 1392 |
+
"grad_norm": 2.9337304572329717,
|
| 1393 |
+
"learning_rate": 1.3077232968705805e-06,
|
| 1394 |
+
"loss": 0.2403,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 3.96,
|
| 1399 |
+
"grad_norm": 3.6309216538671354,
|
| 1400 |
+
"learning_rate": 1.2610095475073415e-06,
|
| 1401 |
+
"loss": 0.2708,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 3.98,
|
| 1406 |
+
"grad_norm": 2.6536626242412056,
|
| 1407 |
+
"learning_rate": 1.2150247217412186e-06,
|
| 1408 |
+
"loss": 0.2495,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 4.0,
|
| 1413 |
+
"grad_norm": 3.8919822758192404,
|
| 1414 |
+
"learning_rate": 1.1697777844051105e-06,
|
| 1415 |
+
"loss": 0.2763,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 4.02,
|
| 1420 |
+
"grad_norm": 2.6296022264068797,
|
| 1421 |
+
"learning_rate": 1.1252775564791023e-06,
|
| 1422 |
+
"loss": 0.2009,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 4.04,
|
| 1427 |
+
"grad_norm": 2.0687616816587715,
|
| 1428 |
+
"learning_rate": 1.0815327133708015e-06,
|
| 1429 |
+
"loss": 0.1978,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 4.06,
|
| 1434 |
+
"grad_norm": 2.036715506169882,
|
| 1435 |
+
"learning_rate": 1.0385517832240472e-06,
|
| 1436 |
+
"loss": 0.195,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 4.08,
|
| 1441 |
+
"grad_norm": 2.1444679630444647,
|
| 1442 |
+
"learning_rate": 9.963431452563331e-07,
|
| 1443 |
+
"loss": 0.195,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 4.1,
|
| 1448 |
+
"grad_norm": 2.3208387437528306,
|
| 1449 |
+
"learning_rate": 9.549150281252633e-07,
|
| 1450 |
+
"loss": 0.2003,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 4.12,
|
| 1455 |
+
"grad_norm": 2.6475787949798635,
|
| 1456 |
+
"learning_rate": 9.142755083243577e-07,
|
| 1457 |
+
"loss": 0.2012,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 4.14,
|
| 1462 |
+
"grad_norm": 2.0747944566043417,
|
| 1463 |
+
"learning_rate": 8.744325086085248e-07,
|
| 1464 |
+
"loss": 0.1839,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 4.16,
|
| 1469 |
+
"grad_norm": 2.5790204025275174,
|
| 1470 |
+
"learning_rate": 8.353937964495029e-07,
|
| 1471 |
+
"loss": 0.2078,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 4.18,
|
| 1476 |
+
"grad_norm": 2.279012511901917,
|
| 1477 |
+
"learning_rate": 7.971669825215789e-07,
|
| 1478 |
+
"loss": 0.2033,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 4.2,
|
| 1483 |
+
"grad_norm": 2.263082794933402,
|
| 1484 |
+
"learning_rate": 7.597595192178702e-07,
|
| 1485 |
+
"loss": 0.2013,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 4.22,
|
| 1490 |
+
"grad_norm": 1.746174251156355,
|
| 1491 |
+
"learning_rate": 7.23178699197467e-07,
|
| 1492 |
+
"loss": 0.1942,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 4.24,
|
| 1497 |
+
"grad_norm": 1.8889622200380844,
|
| 1498 |
+
"learning_rate": 6.874316539637127e-07,
|
| 1499 |
+
"loss": 0.192,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 4.26,
|
| 1504 |
+
"grad_norm": 1.9542919003413253,
|
| 1505 |
+
"learning_rate": 6.52525352473905e-07,
|
| 1506 |
+
"loss": 0.2073,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 4.28,
|
| 1511 |
+
"grad_norm": 2.110295600169004,
|
| 1512 |
+
"learning_rate": 6.184665997806832e-07,
|
| 1513 |
+
"loss": 0.1926,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 4.3,
|
| 1518 |
+
"grad_norm": 2.2234924989126563,
|
| 1519 |
+
"learning_rate": 5.852620357053651e-07,
|
| 1520 |
+
"loss": 0.2033,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 4.32,
|
| 1525 |
+
"grad_norm": 2.6413829746557504,
|
| 1526 |
+
"learning_rate": 5.529181335435124e-07,
|
| 1527 |
+
"loss": 0.198,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 4.34,
|
| 1532 |
+
"grad_norm": 2.097744663828809,
|
| 1533 |
+
"learning_rate": 5.214411988029355e-07,
|
| 1534 |
+
"loss": 0.2017,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 4.36,
|
| 1539 |
+
"grad_norm": 2.3107650980894263,
|
| 1540 |
+
"learning_rate": 4.908373679744316e-07,
|
| 1541 |
+
"loss": 0.2004,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 4.38,
|
| 1546 |
+
"grad_norm": 2.0392932247609665,
|
| 1547 |
+
"learning_rate": 4.6111260733545714e-07,
|
| 1548 |
+
"loss": 0.1978,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 4.4,
|
| 1553 |
+
"grad_norm": 1.9108697879670782,
|
| 1554 |
+
"learning_rate": 4.322727117869951e-07,
|
| 1555 |
+
"loss": 0.2064,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 4.42,
|
| 1560 |
+
"grad_norm": 2.0940301187428645,
|
| 1561 |
+
"learning_rate": 4.043233037238281e-07,
|
| 1562 |
+
"loss": 0.1992,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 4.44,
|
| 1567 |
+
"grad_norm": 2.1573212002556192,
|
| 1568 |
+
"learning_rate": 3.772698319384349e-07,
|
| 1569 |
+
"loss": 0.1978,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 4.46,
|
| 1574 |
+
"grad_norm": 1.7681339799095577,
|
| 1575 |
+
"learning_rate": 3.511175705587433e-07,
|
| 1576 |
+
"loss": 0.1949,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 4.48,
|
| 1581 |
+
"grad_norm": 2.6507223455216335,
|
| 1582 |
+
"learning_rate": 3.258716180199278e-07,
|
| 1583 |
+
"loss": 0.1974,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 4.5,
|
| 1588 |
+
"grad_norm": 4.899195247798816,
|
| 1589 |
+
"learning_rate": 3.015368960704584e-07,
|
| 1590 |
+
"loss": 0.1951,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 4.52,
|
| 1595 |
+
"grad_norm": 2.362208067582404,
|
| 1596 |
+
"learning_rate": 2.7811814881259503e-07,
|
| 1597 |
+
"loss": 0.1898,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 4.54,
|
| 1602 |
+
"grad_norm": 1.9828163543826154,
|
| 1603 |
+
"learning_rate": 2.556199417775174e-07,
|
| 1604 |
+
"loss": 0.2081,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 4.5600000000000005,
|
| 1609 |
+
"grad_norm": 1.9187249803005326,
|
| 1610 |
+
"learning_rate": 2.3404666103526542e-07,
|
| 1611 |
+
"loss": 0.2027,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 4.58,
|
| 1616 |
+
"grad_norm": 2.050797728763808,
|
| 1617 |
+
"learning_rate": 2.134025123396638e-07,
|
| 1618 |
+
"loss": 0.195,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 4.6,
|
| 1623 |
+
"grad_norm": 2.434262719508876,
|
| 1624 |
+
"learning_rate": 1.9369152030840553e-07,
|
| 1625 |
+
"loss": 0.1935,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 4.62,
|
| 1630 |
+
"grad_norm": 2.7114986108898496,
|
| 1631 |
+
"learning_rate": 1.7491752763844294e-07,
|
| 1632 |
+
"loss": 0.201,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 4.64,
|
| 1637 |
+
"grad_norm": 2.342848968749728,
|
| 1638 |
+
"learning_rate": 1.5708419435684463e-07,
|
| 1639 |
+
"loss": 0.2045,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 4.66,
|
| 1644 |
+
"grad_norm": 2.50611015184642,
|
| 1645 |
+
"learning_rate": 1.4019499710726913e-07,
|
| 1646 |
+
"loss": 0.2037,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 4.68,
|
| 1651 |
+
"grad_norm": 1.9845962796916226,
|
| 1652 |
+
"learning_rate": 1.2425322847218368e-07,
|
| 1653 |
+
"loss": 0.1902,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 4.7,
|
| 1658 |
+
"grad_norm": 2.066929322984826,
|
| 1659 |
+
"learning_rate": 1.0926199633097156e-07,
|
| 1660 |
+
"loss": 0.1956,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 4.72,
|
| 1665 |
+
"grad_norm": 2.4138648473696853,
|
| 1666 |
+
"learning_rate": 9.522422325404234e-08,
|
| 1667 |
+
"loss": 0.1985,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 4.74,
|
| 1672 |
+
"grad_norm": 2.911095557310394,
|
| 1673 |
+
"learning_rate": 8.214264593307097e-08,
|
| 1674 |
+
"loss": 0.1965,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 4.76,
|
| 1679 |
+
"grad_norm": 2.030674439873152,
|
| 1680 |
+
"learning_rate": 7.001981464747565e-08,
|
| 1681 |
+
"loss": 0.1905,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 4.78,
|
| 1686 |
+
"grad_norm": 2.5046761639631696,
|
| 1687 |
+
"learning_rate": 5.8858092767236084e-08,
|
| 1688 |
+
"loss": 0.2016,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 4.8,
|
| 1693 |
+
"grad_norm": 2.6259164045048715,
|
| 1694 |
+
"learning_rate": 4.865965629214819e-08,
|
| 1695 |
+
"loss": 0.1991,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 4.82,
|
| 1700 |
+
"grad_norm": 2.5096024708693725,
|
| 1701 |
+
"learning_rate": 3.9426493427611177e-08,
|
| 1702 |
+
"loss": 0.1987,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 4.84,
|
| 1707 |
+
"grad_norm": 2.1146883842387783,
|
| 1708 |
+
"learning_rate": 3.1160404197018155e-08,
|
| 1709 |
+
"loss": 0.1915,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 4.86,
|
| 1714 |
+
"grad_norm": 2.1798112190278136,
|
| 1715 |
+
"learning_rate": 2.386300009084408e-08,
|
| 1716 |
+
"loss": 0.2055,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 4.88,
|
| 1721 |
+
"grad_norm": 2.1283754821360152,
|
| 1722 |
+
"learning_rate": 1.753570375247815e-08,
|
| 1723 |
+
"loss": 0.1941,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 4.9,
|
| 1728 |
+
"grad_norm": 2.945625263690615,
|
| 1729 |
+
"learning_rate": 1.2179748700879013e-08,
|
| 1730 |
+
"loss": 0.2051,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 4.92,
|
| 1735 |
+
"grad_norm": 2.3318033551646793,
|
| 1736 |
+
"learning_rate": 7.796179090094891e-09,
|
| 1737 |
+
"loss": 0.1929,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 4.9399999999999995,
|
| 1742 |
+
"grad_norm": 2.7773560139669944,
|
| 1743 |
+
"learning_rate": 4.385849505708084e-09,
|
| 1744 |
+
"loss": 0.1958,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 4.96,
|
| 1749 |
+
"grad_norm": 2.2518024677328685,
|
| 1750 |
+
"learning_rate": 1.9494247982282386e-09,
|
| 1751 |
+
"loss": 0.2032,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 4.98,
|
| 1756 |
+
"grad_norm": 2.0203646274765132,
|
| 1757 |
+
"learning_rate": 4.87379953478806e-10,
|
| 1758 |
+
"loss": 0.2075,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 5.0,
|
| 1763 |
+
"grad_norm": 2.9166287851342645,
|
| 1764 |
+
"learning_rate": 0.0,
|
| 1765 |
+
"loss": 0.1981,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 5.0,
|
| 1770 |
+
"step": 1250,
|
| 1771 |
+
"total_flos": 2442779688960.0,
|
| 1772 |
+
"train_loss": 0.6922085983276367,
|
| 1773 |
+
"train_runtime": 19109.4525,
|
| 1774 |
+
"train_samples_per_second": 1.047,
|
| 1775 |
+
"train_steps_per_second": 0.065
|
| 1776 |
+
}
|
| 1777 |
+
],
|
| 1778 |
+
"logging_steps": 5,
|
| 1779 |
+
"max_steps": 1250,
|
| 1780 |
+
"num_input_tokens_seen": 0,
|
| 1781 |
+
"num_train_epochs": 5,
|
| 1782 |
+
"save_steps": 100,
|
| 1783 |
+
"stateful_callbacks": {
|
| 1784 |
+
"TrainerControl": {
|
| 1785 |
+
"args": {
|
| 1786 |
+
"should_epoch_stop": false,
|
| 1787 |
+
"should_evaluate": false,
|
| 1788 |
+
"should_log": false,
|
| 1789 |
+
"should_save": true,
|
| 1790 |
+
"should_training_stop": true
|
| 1791 |
+
},
|
| 1792 |
+
"attributes": {}
|
| 1793 |
+
}
|
| 1794 |
+
},
|
| 1795 |
+
"total_flos": 2442779688960.0,
|
| 1796 |
+
"train_batch_size": 4,
|
| 1797 |
+
"trial_name": null,
|
| 1798 |
+
"trial_params": null
|
| 1799 |
+
}
|