Model save
Browse files- README.md +63 -0
- all_results.json +8 -0
- generation_config.json +6 -0
- model.safetensors +1 -1
- runs/Apr10_04-50-43_compsci-cluster-fitz-05/events.out.tfevents.1712739053.compsci-cluster-fitz-05.898746.0 +2 -2
- train_results.json +8 -0
- trainer_state.json +1956 -0
README.md
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
base_model: openai-community/gpt2-medium
|
| 4 |
+
tags:
|
| 5 |
+
- trl
|
| 6 |
+
- sft
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- generator
|
| 10 |
+
model-index:
|
| 11 |
+
- name: gpt2-sft
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# gpt2-sft
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [openai-community/gpt2-medium](https://huggingface.co/openai-community/gpt2-medium) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 2.1977
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 2e-05
|
| 42 |
+
- train_batch_size: 8
|
| 43 |
+
- eval_batch_size: 8
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
+
- lr_scheduler_type: cosine
|
| 48 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 1
|
| 50 |
+
|
| 51 |
+
### Training results
|
| 52 |
+
|
| 53 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 55 |
+
| 2.1966 | 1.0 | 1366 | 2.1977 |
|
| 56 |
+
|
| 57 |
+
|
| 58 |
+
### Framework versions
|
| 59 |
+
|
| 60 |
+
- Transformers 4.39.0.dev0
|
| 61 |
+
- Pytorch 2.2.1+cu121
|
| 62 |
+
- Datasets 2.14.6
|
| 63 |
+
- Tokenizers 0.15.2
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.6064129045034048,
|
| 4 |
+
"train_runtime": 690.5421,
|
| 5 |
+
"train_samples": 64320,
|
| 6 |
+
"train_samples_per_second": 15.821,
|
| 7 |
+
"train_steps_per_second": 1.978
|
| 8 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.39.0.dev0"
|
| 6 |
+
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 709676664
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:92c934097e50f1d44b571afffe0e8f25204fb20758bf1b6f5af521ed6b5cbada
|
| 3 |
size 709676664
|
runs/Apr10_04-50-43_compsci-cluster-fitz-05/events.out.tfevents.1712739053.compsci-cluster-fitz-05.898746.0
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 sha256:08fd83ce11565ecafad95a6234860d0ebb9524675204c9fb2d03f32c7fe3313e
|
| 3 |
+
size 21011
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 0.6064129045034048,
|
| 4 |
+
"train_runtime": 690.5421,
|
| 5 |
+
"train_samples": 64320,
|
| 6 |
+
"train_samples_per_second": 15.821,
|
| 7 |
+
"train_steps_per_second": 1.978
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1956 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1366,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0,
|
| 13 |
+
"grad_norm": 15.1875,
|
| 14 |
+
"learning_rate": 1.4598540145985402e-07,
|
| 15 |
+
"loss": 3.5409,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0,
|
| 20 |
+
"grad_norm": 13.625,
|
| 21 |
+
"learning_rate": 7.299270072992701e-07,
|
| 22 |
+
"loss": 3.4713,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.01,
|
| 27 |
+
"grad_norm": 13.0,
|
| 28 |
+
"learning_rate": 1.4598540145985402e-06,
|
| 29 |
+
"loss": 3.4904,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.01,
|
| 34 |
+
"grad_norm": 13.8125,
|
| 35 |
+
"learning_rate": 2.1897810218978103e-06,
|
| 36 |
+
"loss": 3.4687,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.01,
|
| 41 |
+
"grad_norm": 15.9375,
|
| 42 |
+
"learning_rate": 2.9197080291970804e-06,
|
| 43 |
+
"loss": 3.4864,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.02,
|
| 48 |
+
"grad_norm": 13.4375,
|
| 49 |
+
"learning_rate": 3.6496350364963505e-06,
|
| 50 |
+
"loss": 3.4514,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.02,
|
| 55 |
+
"grad_norm": 14.0625,
|
| 56 |
+
"learning_rate": 4.379562043795621e-06,
|
| 57 |
+
"loss": 3.4462,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.03,
|
| 62 |
+
"grad_norm": 13.0,
|
| 63 |
+
"learning_rate": 5.1094890510948916e-06,
|
| 64 |
+
"loss": 3.407,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.03,
|
| 69 |
+
"grad_norm": 15.3125,
|
| 70 |
+
"learning_rate": 5.839416058394161e-06,
|
| 71 |
+
"loss": 3.4481,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.03,
|
| 76 |
+
"grad_norm": 15.25,
|
| 77 |
+
"learning_rate": 6.569343065693431e-06,
|
| 78 |
+
"loss": 3.4538,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.04,
|
| 83 |
+
"grad_norm": 12.75,
|
| 84 |
+
"learning_rate": 7.299270072992701e-06,
|
| 85 |
+
"loss": 3.3903,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.04,
|
| 90 |
+
"grad_norm": 12.25,
|
| 91 |
+
"learning_rate": 8.029197080291972e-06,
|
| 92 |
+
"loss": 3.349,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.04,
|
| 97 |
+
"grad_norm": 13.875,
|
| 98 |
+
"learning_rate": 8.759124087591241e-06,
|
| 99 |
+
"loss": 3.357,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.05,
|
| 104 |
+
"grad_norm": 11.8125,
|
| 105 |
+
"learning_rate": 9.48905109489051e-06,
|
| 106 |
+
"loss": 3.2995,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.05,
|
| 111 |
+
"grad_norm": 11.75,
|
| 112 |
+
"learning_rate": 1.0218978102189783e-05,
|
| 113 |
+
"loss": 3.2453,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.05,
|
| 118 |
+
"grad_norm": 11.875,
|
| 119 |
+
"learning_rate": 1.0948905109489052e-05,
|
| 120 |
+
"loss": 3.2581,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.06,
|
| 125 |
+
"grad_norm": 12.75,
|
| 126 |
+
"learning_rate": 1.1678832116788322e-05,
|
| 127 |
+
"loss": 3.2102,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.06,
|
| 132 |
+
"grad_norm": 10.25,
|
| 133 |
+
"learning_rate": 1.2408759124087593e-05,
|
| 134 |
+
"loss": 3.1574,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.07,
|
| 139 |
+
"grad_norm": 9.875,
|
| 140 |
+
"learning_rate": 1.3138686131386862e-05,
|
| 141 |
+
"loss": 3.0819,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.07,
|
| 146 |
+
"grad_norm": 8.8125,
|
| 147 |
+
"learning_rate": 1.3868613138686133e-05,
|
| 148 |
+
"loss": 3.0446,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.07,
|
| 153 |
+
"grad_norm": 8.4375,
|
| 154 |
+
"learning_rate": 1.4598540145985402e-05,
|
| 155 |
+
"loss": 2.961,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.08,
|
| 160 |
+
"grad_norm": 7.78125,
|
| 161 |
+
"learning_rate": 1.5328467153284673e-05,
|
| 162 |
+
"loss": 2.9056,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.08,
|
| 167 |
+
"grad_norm": 6.5625,
|
| 168 |
+
"learning_rate": 1.6058394160583944e-05,
|
| 169 |
+
"loss": 2.8261,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.08,
|
| 174 |
+
"grad_norm": 5.75,
|
| 175 |
+
"learning_rate": 1.678832116788321e-05,
|
| 176 |
+
"loss": 2.7407,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.09,
|
| 181 |
+
"grad_norm": 5.34375,
|
| 182 |
+
"learning_rate": 1.7518248175182482e-05,
|
| 183 |
+
"loss": 2.6689,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.09,
|
| 188 |
+
"grad_norm": 4.46875,
|
| 189 |
+
"learning_rate": 1.8248175182481753e-05,
|
| 190 |
+
"loss": 2.6545,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.1,
|
| 195 |
+
"grad_norm": 4.59375,
|
| 196 |
+
"learning_rate": 1.897810218978102e-05,
|
| 197 |
+
"loss": 2.5854,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.1,
|
| 202 |
+
"grad_norm": 3.75,
|
| 203 |
+
"learning_rate": 1.9708029197080295e-05,
|
| 204 |
+
"loss": 2.5589,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.1,
|
| 209 |
+
"grad_norm": 3.40625,
|
| 210 |
+
"learning_rate": 1.99997059600332e-05,
|
| 211 |
+
"loss": 2.5376,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.11,
|
| 216 |
+
"grad_norm": 3.390625,
|
| 217 |
+
"learning_rate": 1.9997909111745443e-05,
|
| 218 |
+
"loss": 2.5087,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.11,
|
| 223 |
+
"grad_norm": 3.375,
|
| 224 |
+
"learning_rate": 1.9994479063873808e-05,
|
| 225 |
+
"loss": 2.4925,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.11,
|
| 230 |
+
"grad_norm": 2.984375,
|
| 231 |
+
"learning_rate": 1.998941637673072e-05,
|
| 232 |
+
"loss": 2.4023,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.12,
|
| 237 |
+
"grad_norm": 3.015625,
|
| 238 |
+
"learning_rate": 1.9982721877326954e-05,
|
| 239 |
+
"loss": 2.4525,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.12,
|
| 244 |
+
"grad_norm": 2.890625,
|
| 245 |
+
"learning_rate": 1.9974396659236522e-05,
|
| 246 |
+
"loss": 2.4526,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.12,
|
| 251 |
+
"grad_norm": 3.203125,
|
| 252 |
+
"learning_rate": 1.9964442082418057e-05,
|
| 253 |
+
"loss": 2.416,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.13,
|
| 258 |
+
"grad_norm": 2.953125,
|
| 259 |
+
"learning_rate": 1.9952859772992626e-05,
|
| 260 |
+
"loss": 2.4271,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.13,
|
| 265 |
+
"grad_norm": 3.0,
|
| 266 |
+
"learning_rate": 1.9939651622978127e-05,
|
| 267 |
+
"loss": 2.4125,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.14,
|
| 272 |
+
"grad_norm": 2.71875,
|
| 273 |
+
"learning_rate": 1.9924819789980204e-05,
|
| 274 |
+
"loss": 2.421,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.14,
|
| 279 |
+
"grad_norm": 2.765625,
|
| 280 |
+
"learning_rate": 1.990836669683979e-05,
|
| 281 |
+
"loss": 2.3559,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.14,
|
| 286 |
+
"grad_norm": 2.609375,
|
| 287 |
+
"learning_rate": 1.9890295031237336e-05,
|
| 288 |
+
"loss": 2.3899,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.15,
|
| 293 |
+
"grad_norm": 2.609375,
|
| 294 |
+
"learning_rate": 1.9870607745253765e-05,
|
| 295 |
+
"loss": 2.3836,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.15,
|
| 300 |
+
"grad_norm": 2.859375,
|
| 301 |
+
"learning_rate": 1.9849308054888235e-05,
|
| 302 |
+
"loss": 2.3803,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.15,
|
| 307 |
+
"grad_norm": 2.734375,
|
| 308 |
+
"learning_rate": 1.9826399439532787e-05,
|
| 309 |
+
"loss": 2.3862,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.16,
|
| 314 |
+
"grad_norm": 2.6875,
|
| 315 |
+
"learning_rate": 1.980188564140399e-05,
|
| 316 |
+
"loss": 2.332,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.16,
|
| 321 |
+
"grad_norm": 2.84375,
|
| 322 |
+
"learning_rate": 1.9775770664931613e-05,
|
| 323 |
+
"loss": 2.3934,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.16,
|
| 328 |
+
"grad_norm": 3.03125,
|
| 329 |
+
"learning_rate": 1.97480587761045e-05,
|
| 330 |
+
"loss": 2.3672,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.17,
|
| 335 |
+
"grad_norm": 2.828125,
|
| 336 |
+
"learning_rate": 1.9718754501773688e-05,
|
| 337 |
+
"loss": 2.3645,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.17,
|
| 342 |
+
"grad_norm": 2.703125,
|
| 343 |
+
"learning_rate": 1.9687862628912952e-05,
|
| 344 |
+
"loss": 2.3382,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.18,
|
| 349 |
+
"grad_norm": 2.625,
|
| 350 |
+
"learning_rate": 1.9655388203836804e-05,
|
| 351 |
+
"loss": 2.3214,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.18,
|
| 356 |
+
"grad_norm": 2.6875,
|
| 357 |
+
"learning_rate": 1.962133653137618e-05,
|
| 358 |
+
"loss": 2.344,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.18,
|
| 363 |
+
"grad_norm": 2.5,
|
| 364 |
+
"learning_rate": 1.958571317401186e-05,
|
| 365 |
+
"loss": 2.3284,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.19,
|
| 370 |
+
"grad_norm": 2.421875,
|
| 371 |
+
"learning_rate": 1.954852395096582e-05,
|
| 372 |
+
"loss": 2.3248,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.19,
|
| 377 |
+
"grad_norm": 2.421875,
|
| 378 |
+
"learning_rate": 1.9509774937250646e-05,
|
| 379 |
+
"loss": 2.3123,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.19,
|
| 384 |
+
"grad_norm": 2.546875,
|
| 385 |
+
"learning_rate": 1.9469472462677128e-05,
|
| 386 |
+
"loss": 2.2931,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.2,
|
| 391 |
+
"grad_norm": 2.578125,
|
| 392 |
+
"learning_rate": 1.9427623110820295e-05,
|
| 393 |
+
"loss": 2.3049,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2,
|
| 398 |
+
"grad_norm": 2.484375,
|
| 399 |
+
"learning_rate": 1.9384233717943934e-05,
|
| 400 |
+
"loss": 2.3491,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.2,
|
| 405 |
+
"grad_norm": 2.78125,
|
| 406 |
+
"learning_rate": 1.933931137188387e-05,
|
| 407 |
+
"loss": 2.3268,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.21,
|
| 412 |
+
"grad_norm": 2.671875,
|
| 413 |
+
"learning_rate": 1.9292863410890138e-05,
|
| 414 |
+
"loss": 2.3244,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.21,
|
| 419 |
+
"grad_norm": 2.34375,
|
| 420 |
+
"learning_rate": 1.9244897422428254e-05,
|
| 421 |
+
"loss": 2.2832,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.22,
|
| 426 |
+
"grad_norm": 2.5625,
|
| 427 |
+
"learning_rate": 1.919542124193976e-05,
|
| 428 |
+
"loss": 2.309,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.22,
|
| 433 |
+
"grad_norm": 2.734375,
|
| 434 |
+
"learning_rate": 1.9144442951562275e-05,
|
| 435 |
+
"loss": 2.313,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.22,
|
| 440 |
+
"grad_norm": 2.609375,
|
| 441 |
+
"learning_rate": 1.909197087880925e-05,
|
| 442 |
+
"loss": 2.3164,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.23,
|
| 447 |
+
"grad_norm": 2.5625,
|
| 448 |
+
"learning_rate": 1.9038013595209647e-05,
|
| 449 |
+
"loss": 2.3098,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.23,
|
| 454 |
+
"grad_norm": 2.53125,
|
| 455 |
+
"learning_rate": 1.8982579914907706e-05,
|
| 456 |
+
"loss": 2.3103,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.23,
|
| 461 |
+
"grad_norm": 2.4375,
|
| 462 |
+
"learning_rate": 1.8925678893223155e-05,
|
| 463 |
+
"loss": 2.293,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.24,
|
| 468 |
+
"grad_norm": 2.375,
|
| 469 |
+
"learning_rate": 1.8867319825171972e-05,
|
| 470 |
+
"loss": 2.3174,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.24,
|
| 475 |
+
"grad_norm": 2.609375,
|
| 476 |
+
"learning_rate": 1.880751224394801e-05,
|
| 477 |
+
"loss": 2.3213,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.25,
|
| 482 |
+
"grad_norm": 2.375,
|
| 483 |
+
"learning_rate": 1.8746265919365707e-05,
|
| 484 |
+
"loss": 2.288,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.25,
|
| 489 |
+
"grad_norm": 2.5,
|
| 490 |
+
"learning_rate": 1.8683590856264137e-05,
|
| 491 |
+
"loss": 2.2918,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.25,
|
| 496 |
+
"grad_norm": 2.5625,
|
| 497 |
+
"learning_rate": 1.8619497292872702e-05,
|
| 498 |
+
"loss": 2.2748,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.26,
|
| 503 |
+
"grad_norm": 2.28125,
|
| 504 |
+
"learning_rate": 1.855399569913866e-05,
|
| 505 |
+
"loss": 2.2933,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.26,
|
| 510 |
+
"grad_norm": 2.390625,
|
| 511 |
+
"learning_rate": 1.8487096775016807e-05,
|
| 512 |
+
"loss": 2.3154,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.26,
|
| 517 |
+
"grad_norm": 2.515625,
|
| 518 |
+
"learning_rate": 1.841881144872162e-05,
|
| 519 |
+
"loss": 2.2824,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.27,
|
| 524 |
+
"grad_norm": 2.1875,
|
| 525 |
+
"learning_rate": 1.8349150874942067e-05,
|
| 526 |
+
"loss": 2.3012,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.27,
|
| 531 |
+
"grad_norm": 2.25,
|
| 532 |
+
"learning_rate": 1.8278126433019453e-05,
|
| 533 |
+
"loss": 2.3538,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.27,
|
| 538 |
+
"grad_norm": 2.546875,
|
| 539 |
+
"learning_rate": 1.8205749725088556e-05,
|
| 540 |
+
"loss": 2.2788,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.28,
|
| 545 |
+
"grad_norm": 2.453125,
|
| 546 |
+
"learning_rate": 1.8132032574182367e-05,
|
| 547 |
+
"loss": 2.2883,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.28,
|
| 552 |
+
"grad_norm": 2.25,
|
| 553 |
+
"learning_rate": 1.8056987022300746e-05,
|
| 554 |
+
"loss": 2.3311,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.29,
|
| 559 |
+
"grad_norm": 2.171875,
|
| 560 |
+
"learning_rate": 1.798062532844333e-05,
|
| 561 |
+
"loss": 2.3094,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.29,
|
| 566 |
+
"grad_norm": 2.203125,
|
| 567 |
+
"learning_rate": 1.7902959966606954e-05,
|
| 568 |
+
"loss": 2.3103,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.29,
|
| 573 |
+
"grad_norm": 2.71875,
|
| 574 |
+
"learning_rate": 1.7824003623747984e-05,
|
| 575 |
+
"loss": 2.2717,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.3,
|
| 580 |
+
"grad_norm": 2.46875,
|
| 581 |
+
"learning_rate": 1.7743769197709836e-05,
|
| 582 |
+
"loss": 2.313,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.3,
|
| 587 |
+
"grad_norm": 2.234375,
|
| 588 |
+
"learning_rate": 1.7662269795116093e-05,
|
| 589 |
+
"loss": 2.2838,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.3,
|
| 594 |
+
"grad_norm": 2.28125,
|
| 595 |
+
"learning_rate": 1.757951872922944e-05,
|
| 596 |
+
"loss": 2.2888,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.31,
|
| 601 |
+
"grad_norm": 2.3125,
|
| 602 |
+
"learning_rate": 1.749552951777693e-05,
|
| 603 |
+
"loss": 2.2648,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.31,
|
| 608 |
+
"grad_norm": 2.921875,
|
| 609 |
+
"learning_rate": 1.7410315880741765e-05,
|
| 610 |
+
"loss": 2.3002,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.31,
|
| 615 |
+
"grad_norm": 2.1875,
|
| 616 |
+
"learning_rate": 1.7323891738122126e-05,
|
| 617 |
+
"loss": 2.2758,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.32,
|
| 622 |
+
"grad_norm": 2.5,
|
| 623 |
+
"learning_rate": 1.7236271207657246e-05,
|
| 624 |
+
"loss": 2.3409,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.32,
|
| 629 |
+
"grad_norm": 2.390625,
|
| 630 |
+
"learning_rate": 1.714746860252124e-05,
|
| 631 |
+
"loss": 2.3065,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.33,
|
| 636 |
+
"grad_norm": 2.171875,
|
| 637 |
+
"learning_rate": 1.7057498428984988e-05,
|
| 638 |
+
"loss": 2.2724,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.33,
|
| 643 |
+
"grad_norm": 2.40625,
|
| 644 |
+
"learning_rate": 1.696637538404646e-05,
|
| 645 |
+
"loss": 2.2881,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.33,
|
| 650 |
+
"grad_norm": 2.328125,
|
| 651 |
+
"learning_rate": 1.6874114353029915e-05,
|
| 652 |
+
"loss": 2.3293,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.34,
|
| 657 |
+
"grad_norm": 2.84375,
|
| 658 |
+
"learning_rate": 1.678073040715433e-05,
|
| 659 |
+
"loss": 2.2792,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.34,
|
| 664 |
+
"grad_norm": 2.21875,
|
| 665 |
+
"learning_rate": 1.6686238801071435e-05,
|
| 666 |
+
"loss": 2.2359,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.34,
|
| 671 |
+
"grad_norm": 2.265625,
|
| 672 |
+
"learning_rate": 1.659065497037381e-05,
|
| 673 |
+
"loss": 2.2881,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.35,
|
| 678 |
+
"grad_norm": 2.09375,
|
| 679 |
+
"learning_rate": 1.649399452907343e-05,
|
| 680 |
+
"loss": 2.2918,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.35,
|
| 685 |
+
"grad_norm": 2.4375,
|
| 686 |
+
"learning_rate": 1.639627326705103e-05,
|
| 687 |
+
"loss": 2.2801,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.36,
|
| 692 |
+
"grad_norm": 2.265625,
|
| 693 |
+
"learning_rate": 1.629750714747677e-05,
|
| 694 |
+
"loss": 2.2948,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.36,
|
| 699 |
+
"grad_norm": 2.546875,
|
| 700 |
+
"learning_rate": 1.619771230420261e-05,
|
| 701 |
+
"loss": 2.2611,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.36,
|
| 706 |
+
"grad_norm": 2.359375,
|
| 707 |
+
"learning_rate": 1.609690503912674e-05,
|
| 708 |
+
"loss": 2.3027,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.37,
|
| 713 |
+
"grad_norm": 2.25,
|
| 714 |
+
"learning_rate": 1.599510181953062e-05,
|
| 715 |
+
"loss": 2.2495,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.37,
|
| 720 |
+
"grad_norm": 2.40625,
|
| 721 |
+
"learning_rate": 1.589231927538898e-05,
|
| 722 |
+
"loss": 2.2587,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.37,
|
| 727 |
+
"grad_norm": 2.265625,
|
| 728 |
+
"learning_rate": 1.578857419665325e-05,
|
| 729 |
+
"loss": 2.2614,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.38,
|
| 734 |
+
"grad_norm": 2.34375,
|
| 735 |
+
"learning_rate": 1.568388353050883e-05,
|
| 736 |
+
"loss": 2.2618,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.38,
|
| 741 |
+
"grad_norm": 2.21875,
|
| 742 |
+
"learning_rate": 1.5578264378606727e-05,
|
| 743 |
+
"loss": 2.3015,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.38,
|
| 748 |
+
"grad_norm": 2.296875,
|
| 749 |
+
"learning_rate": 1.5471733994269916e-05,
|
| 750 |
+
"loss": 2.2542,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.39,
|
| 755 |
+
"grad_norm": 2.34375,
|
| 756 |
+
"learning_rate": 1.5364309779674925e-05,
|
| 757 |
+
"loss": 2.2573,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.39,
|
| 762 |
+
"grad_norm": 2.390625,
|
| 763 |
+
"learning_rate": 1.5256009283009148e-05,
|
| 764 |
+
"loss": 2.291,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.4,
|
| 769 |
+
"grad_norm": 2.640625,
|
| 770 |
+
"learning_rate": 1.5146850195604255e-05,
|
| 771 |
+
"loss": 2.2856,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.4,
|
| 776 |
+
"grad_norm": 2.21875,
|
| 777 |
+
"learning_rate": 1.5036850349046236e-05,
|
| 778 |
+
"loss": 2.2403,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.4,
|
| 783 |
+
"grad_norm": 2.265625,
|
| 784 |
+
"learning_rate": 1.4926027712262565e-05,
|
| 785 |
+
"loss": 2.324,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.41,
|
| 790 |
+
"grad_norm": 2.21875,
|
| 791 |
+
"learning_rate": 1.4814400388586871e-05,
|
| 792 |
+
"loss": 2.2583,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.41,
|
| 797 |
+
"grad_norm": 2.21875,
|
| 798 |
+
"learning_rate": 1.4701986612801699e-05,
|
| 799 |
+
"loss": 2.2787,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.41,
|
| 804 |
+
"grad_norm": 2.28125,
|
| 805 |
+
"learning_rate": 1.4588804748159783e-05,
|
| 806 |
+
"loss": 2.2246,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.42,
|
| 811 |
+
"grad_norm": 2.234375,
|
| 812 |
+
"learning_rate": 1.447487328338434e-05,
|
| 813 |
+
"loss": 2.3084,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.42,
|
| 818 |
+
"grad_norm": 2.15625,
|
| 819 |
+
"learning_rate": 1.4360210829648834e-05,
|
| 820 |
+
"loss": 2.2681,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.42,
|
| 825 |
+
"grad_norm": 2.40625,
|
| 826 |
+
"learning_rate": 1.4244836117536797e-05,
|
| 827 |
+
"loss": 2.2876,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.43,
|
| 832 |
+
"grad_norm": 2.390625,
|
| 833 |
+
"learning_rate": 1.4128767993982097e-05,
|
| 834 |
+
"loss": 2.2373,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.43,
|
| 839 |
+
"grad_norm": 2.3125,
|
| 840 |
+
"learning_rate": 1.401202541919021e-05,
|
| 841 |
+
"loss": 2.2318,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.44,
|
| 846 |
+
"grad_norm": 2.28125,
|
| 847 |
+
"learning_rate": 1.389462746354099e-05,
|
| 848 |
+
"loss": 2.2876,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.44,
|
| 853 |
+
"grad_norm": 2.28125,
|
| 854 |
+
"learning_rate": 1.3776593304473464e-05,
|
| 855 |
+
"loss": 2.2743,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.44,
|
| 860 |
+
"grad_norm": 2.25,
|
| 861 |
+
"learning_rate": 1.3657942223353093e-05,
|
| 862 |
+
"loss": 2.3012,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.45,
|
| 867 |
+
"grad_norm": 2.28125,
|
| 868 |
+
"learning_rate": 1.3538693602322102e-05,
|
| 869 |
+
"loss": 2.2909,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.45,
|
| 874 |
+
"grad_norm": 2.3125,
|
| 875 |
+
"learning_rate": 1.3418866921133309e-05,
|
| 876 |
+
"loss": 2.2919,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.45,
|
| 881 |
+
"grad_norm": 2.09375,
|
| 882 |
+
"learning_rate": 1.3298481753968044e-05,
|
| 883 |
+
"loss": 2.2759,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.46,
|
| 888 |
+
"grad_norm": 2.421875,
|
| 889 |
+
"learning_rate": 1.317755776623862e-05,
|
| 890 |
+
"loss": 2.2374,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.46,
|
| 895 |
+
"grad_norm": 2.328125,
|
| 896 |
+
"learning_rate": 1.3056114711375898e-05,
|
| 897 |
+
"loss": 2.2749,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.46,
|
| 902 |
+
"grad_norm": 2.171875,
|
| 903 |
+
"learning_rate": 1.2934172427602499e-05,
|
| 904 |
+
"loss": 2.2678,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.47,
|
| 909 |
+
"grad_norm": 2.390625,
|
| 910 |
+
"learning_rate": 1.2811750834692134e-05,
|
| 911 |
+
"loss": 2.2902,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.47,
|
| 916 |
+
"grad_norm": 2.25,
|
| 917 |
+
"learning_rate": 1.268886993071564e-05,
|
| 918 |
+
"loss": 2.2685,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.48,
|
| 923 |
+
"grad_norm": 2.1875,
|
| 924 |
+
"learning_rate": 1.2565549788774198e-05,
|
| 925 |
+
"loss": 2.2695,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.48,
|
| 930 |
+
"grad_norm": 2.28125,
|
| 931 |
+
"learning_rate": 1.2441810553720324e-05,
|
| 932 |
+
"loss": 2.2628,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.48,
|
| 937 |
+
"grad_norm": 2.59375,
|
| 938 |
+
"learning_rate": 1.2317672438867114e-05,
|
| 939 |
+
"loss": 2.2308,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.49,
|
| 944 |
+
"grad_norm": 2.125,
|
| 945 |
+
"learning_rate": 1.2193155722686326e-05,
|
| 946 |
+
"loss": 2.2787,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.49,
|
| 951 |
+
"grad_norm": 2.359375,
|
| 952 |
+
"learning_rate": 1.2068280745495797e-05,
|
| 953 |
+
"loss": 2.2848,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.49,
|
| 958 |
+
"grad_norm": 2.328125,
|
| 959 |
+
"learning_rate": 1.1943067906136772e-05,
|
| 960 |
+
"loss": 2.2716,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.5,
|
| 965 |
+
"grad_norm": 2.328125,
|
| 966 |
+
"learning_rate": 1.1817537658641677e-05,
|
| 967 |
+
"loss": 2.2833,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.5,
|
| 972 |
+
"grad_norm": 2.28125,
|
| 973 |
+
"learning_rate": 1.1691710508892859e-05,
|
| 974 |
+
"loss": 2.2403,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.51,
|
| 979 |
+
"grad_norm": 2.265625,
|
| 980 |
+
"learning_rate": 1.1565607011272874e-05,
|
| 981 |
+
"loss": 2.2913,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.51,
|
| 986 |
+
"grad_norm": 2.28125,
|
| 987 |
+
"learning_rate": 1.1439247765306835e-05,
|
| 988 |
+
"loss": 2.2717,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.51,
|
| 993 |
+
"grad_norm": 2.109375,
|
| 994 |
+
"learning_rate": 1.13126534122974e-05,
|
| 995 |
+
"loss": 2.255,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.52,
|
| 1000 |
+
"grad_norm": 2.359375,
|
| 1001 |
+
"learning_rate": 1.1185844631952926e-05,
|
| 1002 |
+
"loss": 2.2996,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.52,
|
| 1007 |
+
"grad_norm": 2.25,
|
| 1008 |
+
"learning_rate": 1.1058842139009369e-05,
|
| 1009 |
+
"loss": 2.2673,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.52,
|
| 1014 |
+
"grad_norm": 2.234375,
|
| 1015 |
+
"learning_rate": 1.0931666679846427e-05,
|
| 1016 |
+
"loss": 2.2508,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.53,
|
| 1021 |
+
"grad_norm": 2.296875,
|
| 1022 |
+
"learning_rate": 1.0804339029098542e-05,
|
| 1023 |
+
"loss": 2.2279,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.53,
|
| 1028 |
+
"grad_norm": 2.296875,
|
| 1029 |
+
"learning_rate": 1.0676879986261274e-05,
|
| 1030 |
+
"loss": 2.2841,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.53,
|
| 1035 |
+
"grad_norm": 2.09375,
|
| 1036 |
+
"learning_rate": 1.0549310372293631e-05,
|
| 1037 |
+
"loss": 2.2626,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.54,
|
| 1042 |
+
"grad_norm": 2.15625,
|
| 1043 |
+
"learning_rate": 1.0421651026216859e-05,
|
| 1044 |
+
"loss": 2.2737,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.54,
|
| 1049 |
+
"grad_norm": 2.28125,
|
| 1050 |
+
"learning_rate": 1.0293922801710322e-05,
|
| 1051 |
+
"loss": 2.2713,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.55,
|
| 1056 |
+
"grad_norm": 2.40625,
|
| 1057 |
+
"learning_rate": 1.0166146563704953e-05,
|
| 1058 |
+
"loss": 2.2836,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.55,
|
| 1063 |
+
"grad_norm": 2.125,
|
| 1064 |
+
"learning_rate": 1.003834318497489e-05,
|
| 1065 |
+
"loss": 2.2993,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.55,
|
| 1070 |
+
"grad_norm": 2.734375,
|
| 1071 |
+
"learning_rate": 9.910533542727826e-06,
|
| 1072 |
+
"loss": 2.2349,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.56,
|
| 1077 |
+
"grad_norm": 2.125,
|
| 1078 |
+
"learning_rate": 9.782738515194611e-06,
|
| 1079 |
+
"loss": 2.2954,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.56,
|
| 1084 |
+
"grad_norm": 2.09375,
|
| 1085 |
+
"learning_rate": 9.654978978218735e-06,
|
| 1086 |
+
"loss": 2.2879,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.56,
|
| 1091 |
+
"grad_norm": 2.4375,
|
| 1092 |
+
"learning_rate": 9.527275801846148e-06,
|
| 1093 |
+
"loss": 2.2637,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.57,
|
| 1098 |
+
"grad_norm": 2.15625,
|
| 1099 |
+
"learning_rate": 9.399649846916075e-06,
|
| 1100 |
+
"loss": 2.2558,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.57,
|
| 1105 |
+
"grad_norm": 2.265625,
|
| 1106 |
+
"learning_rate": 9.272121961653293e-06,
|
| 1107 |
+
"loss": 2.2513,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.57,
|
| 1112 |
+
"grad_norm": 2.421875,
|
| 1113 |
+
"learning_rate": 9.1447129782625e-06,
|
| 1114 |
+
"loss": 2.2602,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.58,
|
| 1119 |
+
"grad_norm": 2.515625,
|
| 1120 |
+
"learning_rate": 9.017443709525278e-06,
|
| 1121 |
+
"loss": 2.2518,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.58,
|
| 1126 |
+
"grad_norm": 2.96875,
|
| 1127 |
+
"learning_rate": 8.890334945400256e-06,
|
| 1128 |
+
"loss": 2.2511,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.59,
|
| 1133 |
+
"grad_norm": 2.1875,
|
| 1134 |
+
"learning_rate": 8.763407449626956e-06,
|
| 1135 |
+
"loss": 2.2424,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.59,
|
| 1140 |
+
"grad_norm": 2.1875,
|
| 1141 |
+
"learning_rate": 8.636681956333992e-06,
|
| 1142 |
+
"loss": 2.2414,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.59,
|
| 1147 |
+
"grad_norm": 2.28125,
|
| 1148 |
+
"learning_rate": 8.510179166652023e-06,
|
| 1149 |
+
"loss": 2.2786,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.6,
|
| 1154 |
+
"grad_norm": 2.3125,
|
| 1155 |
+
"learning_rate": 8.383919745332171e-06,
|
| 1156 |
+
"loss": 2.2856,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.6,
|
| 1161 |
+
"grad_norm": 1.984375,
|
| 1162 |
+
"learning_rate": 8.257924317370323e-06,
|
| 1163 |
+
"loss": 2.283,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.6,
|
| 1168 |
+
"grad_norm": 2.46875,
|
| 1169 |
+
"learning_rate": 8.132213464637972e-06,
|
| 1170 |
+
"loss": 2.2562,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.61,
|
| 1175 |
+
"grad_norm": 2.0625,
|
| 1176 |
+
"learning_rate": 8.006807722520073e-06,
|
| 1177 |
+
"loss": 2.2338,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.61,
|
| 1182 |
+
"grad_norm": 2.484375,
|
| 1183 |
+
"learning_rate": 7.881727576560513e-06,
|
| 1184 |
+
"loss": 2.2892,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.61,
|
| 1189 |
+
"grad_norm": 2.203125,
|
| 1190 |
+
"learning_rate": 7.756993459115696e-06,
|
| 1191 |
+
"loss": 2.2578,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.62,
|
| 1196 |
+
"grad_norm": 2.546875,
|
| 1197 |
+
"learning_rate": 7.632625746016859e-06,
|
| 1198 |
+
"loss": 2.2318,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.62,
|
| 1203 |
+
"grad_norm": 2.125,
|
| 1204 |
+
"learning_rate": 7.508644753241568e-06,
|
| 1205 |
+
"loss": 2.2991,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.63,
|
| 1210 |
+
"grad_norm": 2.234375,
|
| 1211 |
+
"learning_rate": 7.385070733595044e-06,
|
| 1212 |
+
"loss": 2.2888,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.63,
|
| 1217 |
+
"grad_norm": 2.21875,
|
| 1218 |
+
"learning_rate": 7.261923873401762e-06,
|
| 1219 |
+
"loss": 2.3097,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.63,
|
| 1224 |
+
"grad_norm": 2.375,
|
| 1225 |
+
"learning_rate": 7.139224289207958e-06,
|
| 1226 |
+
"loss": 2.2517,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.64,
|
| 1231 |
+
"grad_norm": 2.203125,
|
| 1232 |
+
"learning_rate": 7.016992024495475e-06,
|
| 1233 |
+
"loss": 2.2512,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.64,
|
| 1238 |
+
"grad_norm": 3.015625,
|
| 1239 |
+
"learning_rate": 6.895247046407616e-06,
|
| 1240 |
+
"loss": 2.2245,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.64,
|
| 1245 |
+
"grad_norm": 2.171875,
|
| 1246 |
+
"learning_rate": 6.774009242487393e-06,
|
| 1247 |
+
"loss": 2.2485,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.65,
|
| 1252 |
+
"grad_norm": 2.765625,
|
| 1253 |
+
"learning_rate": 6.653298417428841e-06,
|
| 1254 |
+
"loss": 2.2257,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.65,
|
| 1259 |
+
"grad_norm": 2.109375,
|
| 1260 |
+
"learning_rate": 6.533134289841822e-06,
|
| 1261 |
+
"loss": 2.2616,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.66,
|
| 1266 |
+
"grad_norm": 2.25,
|
| 1267 |
+
"learning_rate": 6.4135364890309245e-06,
|
| 1268 |
+
"loss": 2.2599,
|
| 1269 |
+
"step": 895
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.66,
|
| 1273 |
+
"grad_norm": 2.15625,
|
| 1274 |
+
"learning_rate": 6.294524551788928e-06,
|
| 1275 |
+
"loss": 2.2262,
|
| 1276 |
+
"step": 900
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.66,
|
| 1280 |
+
"grad_norm": 2.078125,
|
| 1281 |
+
"learning_rate": 6.176117919205378e-06,
|
| 1282 |
+
"loss": 2.2752,
|
| 1283 |
+
"step": 905
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.67,
|
| 1287 |
+
"grad_norm": 2.09375,
|
| 1288 |
+
"learning_rate": 6.0583359334908275e-06,
|
| 1289 |
+
"loss": 2.2541,
|
| 1290 |
+
"step": 910
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.67,
|
| 1294 |
+
"grad_norm": 2.375,
|
| 1295 |
+
"learning_rate": 5.941197834817183e-06,
|
| 1296 |
+
"loss": 2.2642,
|
| 1297 |
+
"step": 915
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.67,
|
| 1301 |
+
"grad_norm": 2.421875,
|
| 1302 |
+
"learning_rate": 5.824722758174763e-06,
|
| 1303 |
+
"loss": 2.2879,
|
| 1304 |
+
"step": 920
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.68,
|
| 1308 |
+
"grad_norm": 2.34375,
|
| 1309 |
+
"learning_rate": 5.708929730246501e-06,
|
| 1310 |
+
"loss": 2.2794,
|
| 1311 |
+
"step": 925
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.68,
|
| 1315 |
+
"grad_norm": 2.21875,
|
| 1316 |
+
"learning_rate": 5.593837666299871e-06,
|
| 1317 |
+
"loss": 2.2787,
|
| 1318 |
+
"step": 930
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.68,
|
| 1322 |
+
"grad_norm": 2.109375,
|
| 1323 |
+
"learning_rate": 5.479465367096999e-06,
|
| 1324 |
+
"loss": 2.2598,
|
| 1325 |
+
"step": 935
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.69,
|
| 1329 |
+
"grad_norm": 2.25,
|
| 1330 |
+
"learning_rate": 5.365831515823478e-06,
|
| 1331 |
+
"loss": 2.2633,
|
| 1332 |
+
"step": 940
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.69,
|
| 1336 |
+
"grad_norm": 2.3125,
|
| 1337 |
+
"learning_rate": 5.252954675036384e-06,
|
| 1338 |
+
"loss": 2.2602,
|
| 1339 |
+
"step": 945
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 0.7,
|
| 1343 |
+
"grad_norm": 2.28125,
|
| 1344 |
+
"learning_rate": 5.140853283632039e-06,
|
| 1345 |
+
"loss": 2.2585,
|
| 1346 |
+
"step": 950
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 0.7,
|
| 1350 |
+
"grad_norm": 2.3125,
|
| 1351 |
+
"learning_rate": 5.029545653833887e-06,
|
| 1352 |
+
"loss": 2.279,
|
| 1353 |
+
"step": 955
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 0.7,
|
| 1357 |
+
"grad_norm": 2.125,
|
| 1358 |
+
"learning_rate": 4.919049968201182e-06,
|
| 1359 |
+
"loss": 2.264,
|
| 1360 |
+
"step": 960
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 0.71,
|
| 1364 |
+
"grad_norm": 2.15625,
|
| 1365 |
+
"learning_rate": 4.809384276658728e-06,
|
| 1366 |
+
"loss": 2.2765,
|
| 1367 |
+
"step": 965
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 0.71,
|
| 1371 |
+
"grad_norm": 2.578125,
|
| 1372 |
+
"learning_rate": 4.7005664935483995e-06,
|
| 1373 |
+
"loss": 2.273,
|
| 1374 |
+
"step": 970
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 0.71,
|
| 1378 |
+
"grad_norm": 2.25,
|
| 1379 |
+
"learning_rate": 4.592614394702731e-06,
|
| 1380 |
+
"loss": 2.2695,
|
| 1381 |
+
"step": 975
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 0.72,
|
| 1385 |
+
"grad_norm": 2.234375,
|
| 1386 |
+
"learning_rate": 4.4855456145411754e-06,
|
| 1387 |
+
"loss": 2.2415,
|
| 1388 |
+
"step": 980
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 0.72,
|
| 1392 |
+
"grad_norm": 2.328125,
|
| 1393 |
+
"learning_rate": 4.379377643189444e-06,
|
| 1394 |
+
"loss": 2.2602,
|
| 1395 |
+
"step": 985
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 0.72,
|
| 1399 |
+
"grad_norm": 2.03125,
|
| 1400 |
+
"learning_rate": 4.27412782362242e-06,
|
| 1401 |
+
"loss": 2.2589,
|
| 1402 |
+
"step": 990
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 0.73,
|
| 1406 |
+
"grad_norm": 2.109375,
|
| 1407 |
+
"learning_rate": 4.169813348831121e-06,
|
| 1408 |
+
"loss": 2.2712,
|
| 1409 |
+
"step": 995
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 0.73,
|
| 1413 |
+
"grad_norm": 2.5,
|
| 1414 |
+
"learning_rate": 4.0664512590141415e-06,
|
| 1415 |
+
"loss": 2.27,
|
| 1416 |
+
"step": 1000
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 0.74,
|
| 1420 |
+
"grad_norm": 2.1875,
|
| 1421 |
+
"learning_rate": 3.9640584387940765e-06,
|
| 1422 |
+
"loss": 2.2953,
|
| 1423 |
+
"step": 1005
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 0.74,
|
| 1427 |
+
"grad_norm": 2.203125,
|
| 1428 |
+
"learning_rate": 3.862651614459346e-06,
|
| 1429 |
+
"loss": 2.2796,
|
| 1430 |
+
"step": 1010
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 0.74,
|
| 1434 |
+
"grad_norm": 2.53125,
|
| 1435 |
+
"learning_rate": 3.7622473512318745e-06,
|
| 1436 |
+
"loss": 2.286,
|
| 1437 |
+
"step": 1015
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 0.75,
|
| 1441 |
+
"grad_norm": 2.234375,
|
| 1442 |
+
"learning_rate": 3.662862050561128e-06,
|
| 1443 |
+
"loss": 2.274,
|
| 1444 |
+
"step": 1020
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 0.75,
|
| 1448 |
+
"grad_norm": 2.140625,
|
| 1449 |
+
"learning_rate": 3.564511947444821e-06,
|
| 1450 |
+
"loss": 2.2617,
|
| 1451 |
+
"step": 1025
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 0.75,
|
| 1455 |
+
"grad_norm": 2.1875,
|
| 1456 |
+
"learning_rate": 3.4672131077769056e-06,
|
| 1457 |
+
"loss": 2.2636,
|
| 1458 |
+
"step": 1030
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 0.76,
|
| 1462 |
+
"grad_norm": 2.3125,
|
| 1463 |
+
"learning_rate": 3.37098142572312e-06,
|
| 1464 |
+
"loss": 2.2809,
|
| 1465 |
+
"step": 1035
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 0.76,
|
| 1469 |
+
"grad_norm": 2.328125,
|
| 1470 |
+
"learning_rate": 3.275832621124615e-06,
|
| 1471 |
+
"loss": 2.2541,
|
| 1472 |
+
"step": 1040
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 0.77,
|
| 1476 |
+
"grad_norm": 2.46875,
|
| 1477 |
+
"learning_rate": 3.18178223693006e-06,
|
| 1478 |
+
"loss": 2.297,
|
| 1479 |
+
"step": 1045
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 0.77,
|
| 1483 |
+
"grad_norm": 2.5,
|
| 1484 |
+
"learning_rate": 3.0888456366566165e-06,
|
| 1485 |
+
"loss": 2.2861,
|
| 1486 |
+
"step": 1050
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 0.77,
|
| 1490 |
+
"grad_norm": 2.15625,
|
| 1491 |
+
"learning_rate": 2.997038001880287e-06,
|
| 1492 |
+
"loss": 2.2931,
|
| 1493 |
+
"step": 1055
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 0.78,
|
| 1497 |
+
"grad_norm": 2.125,
|
| 1498 |
+
"learning_rate": 2.9063743297558843e-06,
|
| 1499 |
+
"loss": 2.2441,
|
| 1500 |
+
"step": 1060
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 0.78,
|
| 1504 |
+
"grad_norm": 2.34375,
|
| 1505 |
+
"learning_rate": 2.8168694305672473e-06,
|
| 1506 |
+
"loss": 2.2687,
|
| 1507 |
+
"step": 1065
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 0.78,
|
| 1511 |
+
"grad_norm": 2.15625,
|
| 1512 |
+
"learning_rate": 2.7285379253078626e-06,
|
| 1513 |
+
"loss": 2.2732,
|
| 1514 |
+
"step": 1070
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 0.79,
|
| 1518 |
+
"grad_norm": 2.1875,
|
| 1519 |
+
"learning_rate": 2.6413942432925065e-06,
|
| 1520 |
+
"loss": 2.2441,
|
| 1521 |
+
"step": 1075
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 0.79,
|
| 1525 |
+
"grad_norm": 2.140625,
|
| 1526 |
+
"learning_rate": 2.555452619800135e-06,
|
| 1527 |
+
"loss": 2.3106,
|
| 1528 |
+
"step": 1080
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 0.79,
|
| 1532 |
+
"grad_norm": 2.046875,
|
| 1533 |
+
"learning_rate": 2.4707270937485038e-06,
|
| 1534 |
+
"loss": 2.2372,
|
| 1535 |
+
"step": 1085
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 0.8,
|
| 1539 |
+
"grad_norm": 1.96875,
|
| 1540 |
+
"learning_rate": 2.3872315054008457e-06,
|
| 1541 |
+
"loss": 2.2571,
|
| 1542 |
+
"step": 1090
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 0.8,
|
| 1546 |
+
"grad_norm": 2.140625,
|
| 1547 |
+
"learning_rate": 2.304979494105011e-06,
|
| 1548 |
+
"loss": 2.2538,
|
| 1549 |
+
"step": 1095
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 0.81,
|
| 1553 |
+
"grad_norm": 2.234375,
|
| 1554 |
+
"learning_rate": 2.22398449606542e-06,
|
| 1555 |
+
"loss": 2.2863,
|
| 1556 |
+
"step": 1100
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 0.81,
|
| 1560 |
+
"grad_norm": 2.203125,
|
| 1561 |
+
"learning_rate": 2.1442597421482124e-06,
|
| 1562 |
+
"loss": 2.2811,
|
| 1563 |
+
"step": 1105
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 0.81,
|
| 1567 |
+
"grad_norm": 2.15625,
|
| 1568 |
+
"learning_rate": 2.0658182557199137e-06,
|
| 1569 |
+
"loss": 2.2884,
|
| 1570 |
+
"step": 1110
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 0.82,
|
| 1574 |
+
"grad_norm": 2.1875,
|
| 1575 |
+
"learning_rate": 1.9886728505200436e-06,
|
| 1576 |
+
"loss": 2.2409,
|
| 1577 |
+
"step": 1115
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 0.82,
|
| 1581 |
+
"grad_norm": 2.296875,
|
| 1582 |
+
"learning_rate": 1.9128361285678986e-06,
|
| 1583 |
+
"loss": 2.2267,
|
| 1584 |
+
"step": 1120
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 0.82,
|
| 1588 |
+
"grad_norm": 2.171875,
|
| 1589 |
+
"learning_rate": 1.8383204781040065e-06,
|
| 1590 |
+
"loss": 2.2654,
|
| 1591 |
+
"step": 1125
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 0.83,
|
| 1595 |
+
"grad_norm": 2.0625,
|
| 1596 |
+
"learning_rate": 1.7651380715664124e-06,
|
| 1597 |
+
"loss": 2.3065,
|
| 1598 |
+
"step": 1130
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 0.83,
|
| 1602 |
+
"grad_norm": 2.109375,
|
| 1603 |
+
"learning_rate": 1.693300863602294e-06,
|
| 1604 |
+
"loss": 2.25,
|
| 1605 |
+
"step": 1135
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 0.83,
|
| 1609 |
+
"grad_norm": 2.484375,
|
| 1610 |
+
"learning_rate": 1.6228205891151027e-06,
|
| 1611 |
+
"loss": 2.2566,
|
| 1612 |
+
"step": 1140
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 0.84,
|
| 1616 |
+
"grad_norm": 2.28125,
|
| 1617 |
+
"learning_rate": 1.5537087613476255e-06,
|
| 1618 |
+
"loss": 2.2508,
|
| 1619 |
+
"step": 1145
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 0.84,
|
| 1623 |
+
"grad_norm": 2.328125,
|
| 1624 |
+
"learning_rate": 1.4859766700012478e-06,
|
| 1625 |
+
"loss": 2.2346,
|
| 1626 |
+
"step": 1150
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 0.85,
|
| 1630 |
+
"grad_norm": 2.78125,
|
| 1631 |
+
"learning_rate": 1.419635379391736e-06,
|
| 1632 |
+
"loss": 2.2482,
|
| 1633 |
+
"step": 1155
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 0.85,
|
| 1637 |
+
"grad_norm": 2.1875,
|
| 1638 |
+
"learning_rate": 1.3546957266418315e-06,
|
| 1639 |
+
"loss": 2.2644,
|
| 1640 |
+
"step": 1160
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 0.85,
|
| 1644 |
+
"grad_norm": 2.203125,
|
| 1645 |
+
"learning_rate": 1.291168319910967e-06,
|
| 1646 |
+
"loss": 2.2521,
|
| 1647 |
+
"step": 1165
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 0.86,
|
| 1651 |
+
"grad_norm": 2.09375,
|
| 1652 |
+
"learning_rate": 1.2290635366623805e-06,
|
| 1653 |
+
"loss": 2.2304,
|
| 1654 |
+
"step": 1170
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 0.86,
|
| 1658 |
+
"grad_norm": 2.15625,
|
| 1659 |
+
"learning_rate": 1.1683915219679054e-06,
|
| 1660 |
+
"loss": 2.2861,
|
| 1661 |
+
"step": 1175
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 0.86,
|
| 1665 |
+
"grad_norm": 1.96875,
|
| 1666 |
+
"learning_rate": 1.1091621868507418e-06,
|
| 1667 |
+
"loss": 2.287,
|
| 1668 |
+
"step": 1180
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 0.87,
|
| 1672 |
+
"grad_norm": 2.234375,
|
| 1673 |
+
"learning_rate": 1.0513852066664454e-06,
|
| 1674 |
+
"loss": 2.2172,
|
| 1675 |
+
"step": 1185
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 0.87,
|
| 1679 |
+
"grad_norm": 2.5,
|
| 1680 |
+
"learning_rate": 9.950700195224194e-07,
|
| 1681 |
+
"loss": 2.2312,
|
| 1682 |
+
"step": 1190
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 0.87,
|
| 1686 |
+
"grad_norm": 2.578125,
|
| 1687 |
+
"learning_rate": 9.402258247361584e-07,
|
| 1688 |
+
"loss": 2.2529,
|
| 1689 |
+
"step": 1195
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 0.88,
|
| 1693 |
+
"grad_norm": 2.390625,
|
| 1694 |
+
"learning_rate": 8.868615813325054e-07,
|
| 1695 |
+
"loss": 2.268,
|
| 1696 |
+
"step": 1200
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 0.88,
|
| 1700 |
+
"grad_norm": 2.109375,
|
| 1701 |
+
"learning_rate": 8.349860065801563e-07,
|
| 1702 |
+
"loss": 2.2757,
|
| 1703 |
+
"step": 1205
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 0.89,
|
| 1707 |
+
"grad_norm": 2.375,
|
| 1708 |
+
"learning_rate": 7.846075745676584e-07,
|
| 1709 |
+
"loss": 2.2693,
|
| 1710 |
+
"step": 1210
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 0.89,
|
| 1714 |
+
"grad_norm": 2.1875,
|
| 1715 |
+
"learning_rate": 7.357345148191297e-07,
|
| 1716 |
+
"loss": 2.2866,
|
| 1717 |
+
"step": 1215
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 0.89,
|
| 1721 |
+
"grad_norm": 2.21875,
|
| 1722 |
+
"learning_rate": 6.883748109499455e-07,
|
| 1723 |
+
"loss": 2.2496,
|
| 1724 |
+
"step": 1220
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 0.9,
|
| 1728 |
+
"grad_norm": 2.28125,
|
| 1729 |
+
"learning_rate": 6.425361993625589e-07,
|
| 1730 |
+
"loss": 2.2746,
|
| 1731 |
+
"step": 1225
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 0.9,
|
| 1735 |
+
"grad_norm": 2.1875,
|
| 1736 |
+
"learning_rate": 5.982261679827561e-07,
|
| 1737 |
+
"loss": 2.2776,
|
| 1738 |
+
"step": 1230
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 0.9,
|
| 1742 |
+
"grad_norm": 2.015625,
|
| 1743 |
+
"learning_rate": 5.554519550364456e-07,
|
| 1744 |
+
"loss": 2.2243,
|
| 1745 |
+
"step": 1235
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 0.91,
|
| 1749 |
+
"grad_norm": 1.9765625,
|
| 1750 |
+
"learning_rate": 5.142205478672824e-07,
|
| 1751 |
+
"loss": 2.2735,
|
| 1752 |
+
"step": 1240
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 0.91,
|
| 1756 |
+
"grad_norm": 2.203125,
|
| 1757 |
+
"learning_rate": 4.745386817952502e-07,
|
| 1758 |
+
"loss": 2.2766,
|
| 1759 |
+
"step": 1245
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 0.92,
|
| 1763 |
+
"grad_norm": 2.28125,
|
| 1764 |
+
"learning_rate": 4.3641283901641794e-07,
|
| 1765 |
+
"loss": 2.2487,
|
| 1766 |
+
"step": 1250
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 0.92,
|
| 1770 |
+
"grad_norm": 2.515625,
|
| 1771 |
+
"learning_rate": 3.9984924754405296e-07,
|
| 1772 |
+
"loss": 2.2951,
|
| 1773 |
+
"step": 1255
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 0.92,
|
| 1777 |
+
"grad_norm": 2.265625,
|
| 1778 |
+
"learning_rate": 3.6485388019123826e-07,
|
| 1779 |
+
"loss": 2.2378,
|
| 1780 |
+
"step": 1260
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 0.93,
|
| 1784 |
+
"grad_norm": 2.40625,
|
| 1785 |
+
"learning_rate": 3.314324535952007e-07,
|
| 1786 |
+
"loss": 2.281,
|
| 1787 |
+
"step": 1265
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 0.93,
|
| 1791 |
+
"grad_norm": 2.234375,
|
| 1792 |
+
"learning_rate": 2.995904272834671e-07,
|
| 1793 |
+
"loss": 2.26,
|
| 1794 |
+
"step": 1270
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 0.93,
|
| 1798 |
+
"grad_norm": 2.4375,
|
| 1799 |
+
"learning_rate": 2.6933300278202844e-07,
|
| 1800 |
+
"loss": 2.2277,
|
| 1801 |
+
"step": 1275
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 0.94,
|
| 1805 |
+
"grad_norm": 2.078125,
|
| 1806 |
+
"learning_rate": 2.406651227656576e-07,
|
| 1807 |
+
"loss": 2.266,
|
| 1808 |
+
"step": 1280
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 0.94,
|
| 1812 |
+
"grad_norm": 2.515625,
|
| 1813 |
+
"learning_rate": 2.135914702504871e-07,
|
| 1814 |
+
"loss": 2.2392,
|
| 1815 |
+
"step": 1285
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 0.94,
|
| 1819 |
+
"grad_norm": 2.15625,
|
| 1820 |
+
"learning_rate": 1.8811646782903814e-07,
|
| 1821 |
+
"loss": 2.2394,
|
| 1822 |
+
"step": 1290
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 0.95,
|
| 1826 |
+
"grad_norm": 2.0625,
|
| 1827 |
+
"learning_rate": 1.6424427694775702e-07,
|
| 1828 |
+
"loss": 2.2977,
|
| 1829 |
+
"step": 1295
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 0.95,
|
| 1833 |
+
"grad_norm": 2.015625,
|
| 1834 |
+
"learning_rate": 1.4197879722723018e-07,
|
| 1835 |
+
"loss": 2.2837,
|
| 1836 |
+
"step": 1300
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 0.96,
|
| 1840 |
+
"grad_norm": 2.046875,
|
| 1841 |
+
"learning_rate": 1.213236658251704e-07,
|
| 1842 |
+
"loss": 2.2568,
|
| 1843 |
+
"step": 1305
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 0.96,
|
| 1847 |
+
"grad_norm": 2.375,
|
| 1848 |
+
"learning_rate": 1.0228225684226434e-07,
|
| 1849 |
+
"loss": 2.2856,
|
| 1850 |
+
"step": 1310
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 0.96,
|
| 1854 |
+
"grad_norm": 1.953125,
|
| 1855 |
+
"learning_rate": 8.485768077100443e-08,
|
| 1856 |
+
"loss": 2.2729,
|
| 1857 |
+
"step": 1315
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 0.97,
|
| 1861 |
+
"grad_norm": 2.234375,
|
| 1862 |
+
"learning_rate": 6.905278398757432e-08,
|
| 1863 |
+
"loss": 2.2387,
|
| 1864 |
+
"step": 1320
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 0.97,
|
| 1868 |
+
"grad_norm": 2.34375,
|
| 1869 |
+
"learning_rate": 5.48701482868852e-08,
|
| 1870 |
+
"loss": 2.2797,
|
| 1871 |
+
"step": 1325
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 0.97,
|
| 1875 |
+
"grad_norm": 2.21875,
|
| 1876 |
+
"learning_rate": 4.231209046082207e-08,
|
| 1877 |
+
"loss": 2.2219,
|
| 1878 |
+
"step": 1330
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 0.98,
|
| 1882 |
+
"grad_norm": 2.40625,
|
| 1883 |
+
"learning_rate": 3.1380661919796495e-08,
|
| 1884 |
+
"loss": 2.2422,
|
| 1885 |
+
"step": 1335
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 0.98,
|
| 1889 |
+
"grad_norm": 2.3125,
|
| 1890 |
+
"learning_rate": 2.2077648357629044e-08,
|
| 1891 |
+
"loss": 2.2803,
|
| 1892 |
+
"step": 1340
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 0.98,
|
| 1896 |
+
"grad_norm": 2.09375,
|
| 1897 |
+
"learning_rate": 1.4404569459858242e-08,
|
| 1898 |
+
"loss": 2.2881,
|
| 1899 |
+
"step": 1345
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 0.99,
|
| 1903 |
+
"grad_norm": 2.25,
|
| 1904 |
+
"learning_rate": 8.36267865548912e-09,
|
| 1905 |
+
"loss": 2.2251,
|
| 1906 |
+
"step": 1350
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 0.99,
|
| 1910 |
+
"grad_norm": 2.265625,
|
| 1911 |
+
"learning_rate": 3.952962912238123e-09,
|
| 1912 |
+
"loss": 2.2907,
|
| 1913 |
+
"step": 1355
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 1.0,
|
| 1917 |
+
"grad_norm": 2.515625,
|
| 1918 |
+
"learning_rate": 1.1761425753142875e-09,
|
| 1919 |
+
"loss": 2.2715,
|
| 1920 |
+
"step": 1360
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 1.0,
|
| 1924 |
+
"grad_norm": 2.375,
|
| 1925 |
+
"learning_rate": 3.2671249743376766e-11,
|
| 1926 |
+
"loss": 2.1966,
|
| 1927 |
+
"step": 1365
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 1.0,
|
| 1931 |
+
"eval_loss": 2.1976709365844727,
|
| 1932 |
+
"eval_runtime": 128.8685,
|
| 1933 |
+
"eval_samples_per_second": 21.534,
|
| 1934 |
+
"eval_steps_per_second": 2.693,
|
| 1935 |
+
"step": 1366
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.0,
|
| 1939 |
+
"step": 1366,
|
| 1940 |
+
"total_flos": 2.02921101754368e+16,
|
| 1941 |
+
"train_loss": 0.6064129045034048,
|
| 1942 |
+
"train_runtime": 690.5421,
|
| 1943 |
+
"train_samples_per_second": 15.821,
|
| 1944 |
+
"train_steps_per_second": 1.978
|
| 1945 |
+
}
|
| 1946 |
+
],
|
| 1947 |
+
"logging_steps": 5,
|
| 1948 |
+
"max_steps": 1366,
|
| 1949 |
+
"num_input_tokens_seen": 0,
|
| 1950 |
+
"num_train_epochs": 1,
|
| 1951 |
+
"save_steps": 100,
|
| 1952 |
+
"total_flos": 2.02921101754368e+16,
|
| 1953 |
+
"train_batch_size": 8,
|
| 1954 |
+
"trial_name": null,
|
| 1955 |
+
"trial_params": null
|
| 1956 |
+
}
|