MHGanainy/gpt2-xl-lora-multi-512-k5-1
Browse files- README.md +2 -0
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Oct21_20-48-15_057200f7f556/events.out.tfevents.1729550228.057200f7f556.780.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +1372 -0
README.md
CHANGED
|
@@ -15,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
# gpt2-xl-lora-multi-512-k5-1
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
|
|
|
|
|
|
| 18 |
|
| 19 |
## Model description
|
| 20 |
|
|
|
|
| 15 |
# gpt2-xl-lora-multi-512-k5-1
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [openai-community/gpt2-xl](https://huggingface.co/openai-community/gpt2-xl) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.8533
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.8533389568328857,
|
| 4 |
+
"eval_runtime": 429.9452,
|
| 5 |
+
"eval_samples_per_second": 141.255,
|
| 6 |
+
"eval_steps_per_second": 8.829,
|
| 7 |
+
"perplexity": 6.3810901765889225,
|
| 8 |
+
"total_flos": 2.790413903660581e+18,
|
| 9 |
+
"train_loss": 2.1026804686029656,
|
| 10 |
+
"train_runtime": 6094.1614,
|
| 11 |
+
"train_samples_per_second": 50.133,
|
| 12 |
+
"train_steps_per_second": 3.133
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"eval_loss": 1.8533389568328857,
|
| 4 |
+
"eval_runtime": 429.9452,
|
| 5 |
+
"eval_samples_per_second": 141.255,
|
| 6 |
+
"eval_steps_per_second": 8.829,
|
| 7 |
+
"perplexity": 6.3810901765889225
|
| 8 |
+
}
|
runs/Oct21_20-48-15_057200f7f556/events.out.tfevents.1729550228.057200f7f556.780.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d83c47ea07f740cb0a0200b5b41c7feb25e24bbcd6f8e78a935f21ced3e733a
|
| 3 |
+
size 316
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"total_flos": 2.790413903660581e+18,
|
| 4 |
+
"train_loss": 2.1026804686029656,
|
| 5 |
+
"train_runtime": 6094.1614,
|
| 6 |
+
"train_samples_per_second": 50.133,
|
| 7 |
+
"train_steps_per_second": 3.133
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1372 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 19095,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.005236973029588898,
|
| 13 |
+
"grad_norm": 0.07153668254613876,
|
| 14 |
+
"learning_rate": 1.3093289689034372e-07,
|
| 15 |
+
"loss": 2.2441,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.010473946059177796,
|
| 20 |
+
"grad_norm": 0.06819326430559158,
|
| 21 |
+
"learning_rate": 2.6186579378068744e-07,
|
| 22 |
+
"loss": 2.2612,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.015710919088766692,
|
| 27 |
+
"grad_norm": 0.0710565522313118,
|
| 28 |
+
"learning_rate": 3.9279869067103113e-07,
|
| 29 |
+
"loss": 2.241,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.020947892118355592,
|
| 34 |
+
"grad_norm": 0.07100865989923477,
|
| 35 |
+
"learning_rate": 5.237315875613749e-07,
|
| 36 |
+
"loss": 2.2467,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.026184865147944488,
|
| 41 |
+
"grad_norm": 0.10397951304912567,
|
| 42 |
+
"learning_rate": 6.546644844517186e-07,
|
| 43 |
+
"loss": 2.2484,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.031421838177533384,
|
| 48 |
+
"grad_norm": 0.08388083428144455,
|
| 49 |
+
"learning_rate": 7.855973813420623e-07,
|
| 50 |
+
"loss": 2.2341,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.036658811207122284,
|
| 55 |
+
"grad_norm": 0.0907754898071289,
|
| 56 |
+
"learning_rate": 9.165302782324059e-07,
|
| 57 |
+
"loss": 2.2355,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.041895784236711184,
|
| 62 |
+
"grad_norm": 0.11433129012584686,
|
| 63 |
+
"learning_rate": 1.0474631751227498e-06,
|
| 64 |
+
"loss": 2.2507,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.04713275726630008,
|
| 69 |
+
"grad_norm": 0.11222276836633682,
|
| 70 |
+
"learning_rate": 1.1783960720130934e-06,
|
| 71 |
+
"loss": 2.2331,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.052369730295888976,
|
| 76 |
+
"grad_norm": 0.12000411748886108,
|
| 77 |
+
"learning_rate": 1.3093289689034372e-06,
|
| 78 |
+
"loss": 2.2416,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.057606703325477876,
|
| 83 |
+
"grad_norm": 0.12595705687999725,
|
| 84 |
+
"learning_rate": 1.4402618657937809e-06,
|
| 85 |
+
"loss": 2.2436,
|
| 86 |
+
"step": 1100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.06284367635506677,
|
| 90 |
+
"grad_norm": 0.1335853487253189,
|
| 91 |
+
"learning_rate": 1.5711947626841245e-06,
|
| 92 |
+
"loss": 2.2279,
|
| 93 |
+
"step": 1200
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06808064938465568,
|
| 97 |
+
"grad_norm": 0.15639664232730865,
|
| 98 |
+
"learning_rate": 1.7021276595744682e-06,
|
| 99 |
+
"loss": 2.2123,
|
| 100 |
+
"step": 1300
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.07331762241424457,
|
| 104 |
+
"grad_norm": 0.15709149837493896,
|
| 105 |
+
"learning_rate": 1.8330605564648118e-06,
|
| 106 |
+
"loss": 2.2212,
|
| 107 |
+
"step": 1400
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07855459544383346,
|
| 111 |
+
"grad_norm": 0.16985486447811127,
|
| 112 |
+
"learning_rate": 1.9639934533551554e-06,
|
| 113 |
+
"loss": 2.2262,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08379156847342237,
|
| 118 |
+
"grad_norm": 0.1816212385892868,
|
| 119 |
+
"learning_rate": 2.0949263502454995e-06,
|
| 120 |
+
"loss": 2.1965,
|
| 121 |
+
"step": 1600
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08902854150301126,
|
| 125 |
+
"grad_norm": 0.1750434935092926,
|
| 126 |
+
"learning_rate": 2.225859247135843e-06,
|
| 127 |
+
"loss": 2.2329,
|
| 128 |
+
"step": 1700
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.09426551453260015,
|
| 132 |
+
"grad_norm": 0.18720516562461853,
|
| 133 |
+
"learning_rate": 2.3567921440261868e-06,
|
| 134 |
+
"loss": 2.22,
|
| 135 |
+
"step": 1800
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.09950248756218906,
|
| 139 |
+
"grad_norm": 0.1913694590330124,
|
| 140 |
+
"learning_rate": 2.486415711947627e-06,
|
| 141 |
+
"loss": 2.2004,
|
| 142 |
+
"step": 1900
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.10473946059177795,
|
| 146 |
+
"grad_norm": 0.21759122610092163,
|
| 147 |
+
"learning_rate": 2.6173486088379706e-06,
|
| 148 |
+
"loss": 2.2146,
|
| 149 |
+
"step": 2000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.10997643362136685,
|
| 153 |
+
"grad_norm": 0.2021426558494568,
|
| 154 |
+
"learning_rate": 2.7482815057283147e-06,
|
| 155 |
+
"loss": 2.2058,
|
| 156 |
+
"step": 2100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.11521340665095575,
|
| 160 |
+
"grad_norm": 0.24332384765148163,
|
| 161 |
+
"learning_rate": 2.879214402618658e-06,
|
| 162 |
+
"loss": 2.2055,
|
| 163 |
+
"step": 2200
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.12045037968054464,
|
| 167 |
+
"grad_norm": 0.2356160432100296,
|
| 168 |
+
"learning_rate": 3.010147299509002e-06,
|
| 169 |
+
"loss": 2.1812,
|
| 170 |
+
"step": 2300
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12568735271013354,
|
| 174 |
+
"grad_norm": 0.23526005446910858,
|
| 175 |
+
"learning_rate": 3.141080196399345e-06,
|
| 176 |
+
"loss": 2.1911,
|
| 177 |
+
"step": 2400
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.13092432573972243,
|
| 181 |
+
"grad_norm": 0.25922101736068726,
|
| 182 |
+
"learning_rate": 3.2720130932896892e-06,
|
| 183 |
+
"loss": 2.1958,
|
| 184 |
+
"step": 2500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13616129876931135,
|
| 188 |
+
"grad_norm": 0.2532810866832733,
|
| 189 |
+
"learning_rate": 3.4029459901800333e-06,
|
| 190 |
+
"loss": 2.1964,
|
| 191 |
+
"step": 2600
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.14139827179890024,
|
| 195 |
+
"grad_norm": 0.2663579285144806,
|
| 196 |
+
"learning_rate": 3.5338788870703765e-06,
|
| 197 |
+
"loss": 2.2024,
|
| 198 |
+
"step": 2700
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.14663524482848914,
|
| 202 |
+
"grad_norm": 0.25965484976768494,
|
| 203 |
+
"learning_rate": 3.6648117839607206e-06,
|
| 204 |
+
"loss": 2.1886,
|
| 205 |
+
"step": 2800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.15187221785807803,
|
| 209 |
+
"grad_norm": 0.27700600028038025,
|
| 210 |
+
"learning_rate": 3.7957446808510638e-06,
|
| 211 |
+
"loss": 2.1968,
|
| 212 |
+
"step": 2900
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.15710919088766692,
|
| 216 |
+
"grad_norm": 0.4916733205318451,
|
| 217 |
+
"learning_rate": 3.926677577741408e-06,
|
| 218 |
+
"loss": 2.1866,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.16234616391725581,
|
| 223 |
+
"grad_norm": 0.29351043701171875,
|
| 224 |
+
"learning_rate": 4.0576104746317515e-06,
|
| 225 |
+
"loss": 2.1858,
|
| 226 |
+
"step": 3100
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.16758313694684474,
|
| 230 |
+
"grad_norm": 0.2852916121482849,
|
| 231 |
+
"learning_rate": 4.1885433715220955e-06,
|
| 232 |
+
"loss": 2.1768,
|
| 233 |
+
"step": 3200
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.17282010997643363,
|
| 237 |
+
"grad_norm": 0.3035738170146942,
|
| 238 |
+
"learning_rate": 4.319476268412439e-06,
|
| 239 |
+
"loss": 2.1566,
|
| 240 |
+
"step": 3300
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.17805708300602252,
|
| 244 |
+
"grad_norm": 0.31292301416397095,
|
| 245 |
+
"learning_rate": 4.450409165302783e-06,
|
| 246 |
+
"loss": 2.1877,
|
| 247 |
+
"step": 3400
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1832940560356114,
|
| 251 |
+
"grad_norm": 0.33194518089294434,
|
| 252 |
+
"learning_rate": 4.581342062193127e-06,
|
| 253 |
+
"loss": 2.1823,
|
| 254 |
+
"step": 3500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1885310290652003,
|
| 258 |
+
"grad_norm": 0.3144392967224121,
|
| 259 |
+
"learning_rate": 4.71227495908347e-06,
|
| 260 |
+
"loss": 2.1914,
|
| 261 |
+
"step": 3600
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.1937680020947892,
|
| 265 |
+
"grad_norm": 0.3243958652019501,
|
| 266 |
+
"learning_rate": 4.843207855973814e-06,
|
| 267 |
+
"loss": 2.1756,
|
| 268 |
+
"step": 3700
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.19900497512437812,
|
| 272 |
+
"grad_norm": 0.33841511607170105,
|
| 273 |
+
"learning_rate": 4.974140752864157e-06,
|
| 274 |
+
"loss": 2.1779,
|
| 275 |
+
"step": 3800
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.204241948153967,
|
| 279 |
+
"grad_norm": 0.35222944617271423,
|
| 280 |
+
"learning_rate": 5.1050736497545014e-06,
|
| 281 |
+
"loss": 2.1518,
|
| 282 |
+
"step": 3900
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.2094789211835559,
|
| 286 |
+
"grad_norm": 0.341221421957016,
|
| 287 |
+
"learning_rate": 5.2360065466448455e-06,
|
| 288 |
+
"loss": 2.1765,
|
| 289 |
+
"step": 4000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.2147158942131448,
|
| 293 |
+
"grad_norm": 0.3837369680404663,
|
| 294 |
+
"learning_rate": 5.366939443535189e-06,
|
| 295 |
+
"loss": 2.1485,
|
| 296 |
+
"step": 4100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.2199528672427337,
|
| 300 |
+
"grad_norm": 0.3704512119293213,
|
| 301 |
+
"learning_rate": 5.497872340425532e-06,
|
| 302 |
+
"loss": 2.1733,
|
| 303 |
+
"step": 4200
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.2251898402723226,
|
| 307 |
+
"grad_norm": 0.38055258989334106,
|
| 308 |
+
"learning_rate": 5.628805237315876e-06,
|
| 309 |
+
"loss": 2.1536,
|
| 310 |
+
"step": 4300
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.2304268133019115,
|
| 314 |
+
"grad_norm": 0.388332724571228,
|
| 315 |
+
"learning_rate": 5.75973813420622e-06,
|
| 316 |
+
"loss": 2.1491,
|
| 317 |
+
"step": 4400
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.2356637863315004,
|
| 321 |
+
"grad_norm": 0.38129305839538574,
|
| 322 |
+
"learning_rate": 5.890671031096563e-06,
|
| 323 |
+
"loss": 2.1434,
|
| 324 |
+
"step": 4500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2409007593610893,
|
| 328 |
+
"grad_norm": 0.3961154818534851,
|
| 329 |
+
"learning_rate": 6.021603927986907e-06,
|
| 330 |
+
"loss": 2.1546,
|
| 331 |
+
"step": 4600
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.24613773239067818,
|
| 335 |
+
"grad_norm": 0.39760035276412964,
|
| 336 |
+
"learning_rate": 6.152536824877251e-06,
|
| 337 |
+
"loss": 2.137,
|
| 338 |
+
"step": 4700
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2513747054202671,
|
| 342 |
+
"grad_norm": 0.3933262228965759,
|
| 343 |
+
"learning_rate": 6.283469721767595e-06,
|
| 344 |
+
"loss": 2.1547,
|
| 345 |
+
"step": 4800
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.256611678449856,
|
| 349 |
+
"grad_norm": 0.3820164203643799,
|
| 350 |
+
"learning_rate": 6.414402618657938e-06,
|
| 351 |
+
"loss": 2.1526,
|
| 352 |
+
"step": 4900
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.26184865147944486,
|
| 356 |
+
"grad_norm": 0.4237716794013977,
|
| 357 |
+
"learning_rate": 6.545335515548282e-06,
|
| 358 |
+
"loss": 2.1509,
|
| 359 |
+
"step": 5000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.2670856245090338,
|
| 363 |
+
"grad_norm": 0.38590240478515625,
|
| 364 |
+
"learning_rate": 6.676268412438626e-06,
|
| 365 |
+
"loss": 2.1492,
|
| 366 |
+
"step": 5100
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2723225975386227,
|
| 370 |
+
"grad_norm": 0.4500404894351959,
|
| 371 |
+
"learning_rate": 6.807201309328969e-06,
|
| 372 |
+
"loss": 2.142,
|
| 373 |
+
"step": 5200
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.27755957056821157,
|
| 377 |
+
"grad_norm": 0.43991413712501526,
|
| 378 |
+
"learning_rate": 6.938134206219313e-06,
|
| 379 |
+
"loss": 2.151,
|
| 380 |
+
"step": 5300
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2827965435978005,
|
| 384 |
+
"grad_norm": 0.4152943193912506,
|
| 385 |
+
"learning_rate": 7.069067103109657e-06,
|
| 386 |
+
"loss": 2.1458,
|
| 387 |
+
"step": 5400
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.28803351662738935,
|
| 391 |
+
"grad_norm": 0.43392306566238403,
|
| 392 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 393 |
+
"loss": 2.1464,
|
| 394 |
+
"step": 5500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.2932704896569783,
|
| 398 |
+
"grad_norm": 0.47078412771224976,
|
| 399 |
+
"learning_rate": 7.330932896890345e-06,
|
| 400 |
+
"loss": 2.1323,
|
| 401 |
+
"step": 5600
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.29850746268656714,
|
| 405 |
+
"grad_norm": 0.5018095970153809,
|
| 406 |
+
"learning_rate": 7.461865793780688e-06,
|
| 407 |
+
"loss": 2.142,
|
| 408 |
+
"step": 5700
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.30374443571615606,
|
| 412 |
+
"grad_norm": 0.4851648211479187,
|
| 413 |
+
"learning_rate": 7.592798690671032e-06,
|
| 414 |
+
"loss": 2.1202,
|
| 415 |
+
"step": 5800
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.308981408745745,
|
| 419 |
+
"grad_norm": 0.46947425603866577,
|
| 420 |
+
"learning_rate": 7.723731587561375e-06,
|
| 421 |
+
"loss": 2.1279,
|
| 422 |
+
"step": 5900
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.31421838177533384,
|
| 426 |
+
"grad_norm": 0.4687449336051941,
|
| 427 |
+
"learning_rate": 7.85466448445172e-06,
|
| 428 |
+
"loss": 2.1206,
|
| 429 |
+
"step": 6000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.31945535480492276,
|
| 433 |
+
"grad_norm": 0.5196475982666016,
|
| 434 |
+
"learning_rate": 7.98428805237316e-06,
|
| 435 |
+
"loss": 2.1426,
|
| 436 |
+
"step": 6100
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.32469232783451163,
|
| 440 |
+
"grad_norm": 0.4802483916282654,
|
| 441 |
+
"learning_rate": 8.115220949263503e-06,
|
| 442 |
+
"loss": 2.1241,
|
| 443 |
+
"step": 6200
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.32992930086410055,
|
| 447 |
+
"grad_norm": 0.4807850122451782,
|
| 448 |
+
"learning_rate": 8.246153846153848e-06,
|
| 449 |
+
"loss": 2.1223,
|
| 450 |
+
"step": 6300
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.33516627389368947,
|
| 454 |
+
"grad_norm": 0.4742112457752228,
|
| 455 |
+
"learning_rate": 8.377086743044191e-06,
|
| 456 |
+
"loss": 2.1268,
|
| 457 |
+
"step": 6400
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.34040324692327834,
|
| 461 |
+
"grad_norm": 0.47567641735076904,
|
| 462 |
+
"learning_rate": 8.508019639934534e-06,
|
| 463 |
+
"loss": 2.1103,
|
| 464 |
+
"step": 6500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.34564021995286726,
|
| 468 |
+
"grad_norm": 0.5237625241279602,
|
| 469 |
+
"learning_rate": 8.638952536824878e-06,
|
| 470 |
+
"loss": 2.1279,
|
| 471 |
+
"step": 6600
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.3508771929824561,
|
| 475 |
+
"grad_norm": 0.5335693359375,
|
| 476 |
+
"learning_rate": 8.769885433715222e-06,
|
| 477 |
+
"loss": 2.1268,
|
| 478 |
+
"step": 6700
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.35611416601204504,
|
| 482 |
+
"grad_norm": 0.518767237663269,
|
| 483 |
+
"learning_rate": 8.900818330605566e-06,
|
| 484 |
+
"loss": 2.1271,
|
| 485 |
+
"step": 6800
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.36135113904163396,
|
| 489 |
+
"grad_norm": 0.539432168006897,
|
| 490 |
+
"learning_rate": 9.031751227495909e-06,
|
| 491 |
+
"loss": 2.1138,
|
| 492 |
+
"step": 6900
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.3665881120712228,
|
| 496 |
+
"grad_norm": 0.535347580909729,
|
| 497 |
+
"learning_rate": 9.162684124386254e-06,
|
| 498 |
+
"loss": 2.1331,
|
| 499 |
+
"step": 7000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.37182508510081175,
|
| 503 |
+
"grad_norm": 0.5684192776679993,
|
| 504 |
+
"learning_rate": 9.293617021276597e-06,
|
| 505 |
+
"loss": 2.1228,
|
| 506 |
+
"step": 7100
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3770620581304006,
|
| 510 |
+
"grad_norm": 0.5739738345146179,
|
| 511 |
+
"learning_rate": 9.42454991816694e-06,
|
| 512 |
+
"loss": 2.1108,
|
| 513 |
+
"step": 7200
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.38229903115998953,
|
| 517 |
+
"grad_norm": 0.5791832804679871,
|
| 518 |
+
"learning_rate": 9.555482815057283e-06,
|
| 519 |
+
"loss": 2.1157,
|
| 520 |
+
"step": 7300
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.3875360041895784,
|
| 524 |
+
"grad_norm": 0.5687488317489624,
|
| 525 |
+
"learning_rate": 9.686415711947628e-06,
|
| 526 |
+
"loss": 2.1253,
|
| 527 |
+
"step": 7400
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.3927729772191673,
|
| 531 |
+
"grad_norm": 0.5615626573562622,
|
| 532 |
+
"learning_rate": 9.817348608837972e-06,
|
| 533 |
+
"loss": 2.1048,
|
| 534 |
+
"step": 7500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.39800995024875624,
|
| 538 |
+
"grad_norm": 0.5687493085861206,
|
| 539 |
+
"learning_rate": 9.948281505728315e-06,
|
| 540 |
+
"loss": 2.0961,
|
| 541 |
+
"step": 7600
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4032469232783451,
|
| 545 |
+
"grad_norm": 0.5530877709388733,
|
| 546 |
+
"learning_rate": 1.0079214402618658e-05,
|
| 547 |
+
"loss": 2.0956,
|
| 548 |
+
"step": 7700
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.408483896307934,
|
| 552 |
+
"grad_norm": 0.5689095258712769,
|
| 553 |
+
"learning_rate": 1.0210147299509003e-05,
|
| 554 |
+
"loss": 2.1009,
|
| 555 |
+
"step": 7800
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.4137208693375229,
|
| 559 |
+
"grad_norm": 0.5516002178192139,
|
| 560 |
+
"learning_rate": 1.0341080196399346e-05,
|
| 561 |
+
"loss": 2.1012,
|
| 562 |
+
"step": 7900
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.4189578423671118,
|
| 566 |
+
"grad_norm": 0.5762948989868164,
|
| 567 |
+
"learning_rate": 1.0472013093289691e-05,
|
| 568 |
+
"loss": 2.1028,
|
| 569 |
+
"step": 8000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.42419481539670073,
|
| 573 |
+
"grad_norm": 0.5746406316757202,
|
| 574 |
+
"learning_rate": 1.0602945990180034e-05,
|
| 575 |
+
"loss": 2.1113,
|
| 576 |
+
"step": 8100
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.4294317884262896,
|
| 580 |
+
"grad_norm": 0.6148165464401245,
|
| 581 |
+
"learning_rate": 1.0733878887070377e-05,
|
| 582 |
+
"loss": 2.1046,
|
| 583 |
+
"step": 8200
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.4346687614558785,
|
| 587 |
+
"grad_norm": 0.5808936953544617,
|
| 588 |
+
"learning_rate": 1.086481178396072e-05,
|
| 589 |
+
"loss": 2.1005,
|
| 590 |
+
"step": 8300
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.4399057344854674,
|
| 594 |
+
"grad_norm": 0.6299387812614441,
|
| 595 |
+
"learning_rate": 1.0995744680851064e-05,
|
| 596 |
+
"loss": 2.0907,
|
| 597 |
+
"step": 8400
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.4451427075150563,
|
| 601 |
+
"grad_norm": 0.6052182912826538,
|
| 602 |
+
"learning_rate": 1.1126677577741409e-05,
|
| 603 |
+
"loss": 2.1108,
|
| 604 |
+
"step": 8500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4503796805446452,
|
| 608 |
+
"grad_norm": 0.6362836360931396,
|
| 609 |
+
"learning_rate": 1.1257610474631752e-05,
|
| 610 |
+
"loss": 2.0662,
|
| 611 |
+
"step": 8600
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.4556166535742341,
|
| 615 |
+
"grad_norm": 0.6669042110443115,
|
| 616 |
+
"learning_rate": 1.1387234042553192e-05,
|
| 617 |
+
"loss": 2.0811,
|
| 618 |
+
"step": 8700
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.460853626603823,
|
| 622 |
+
"grad_norm": 0.6090083718299866,
|
| 623 |
+
"learning_rate": 1.1518166939443535e-05,
|
| 624 |
+
"loss": 2.0872,
|
| 625 |
+
"step": 8800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.4660905996334119,
|
| 629 |
+
"grad_norm": 0.6355557441711426,
|
| 630 |
+
"learning_rate": 1.164909983633388e-05,
|
| 631 |
+
"loss": 2.0939,
|
| 632 |
+
"step": 8900
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4713275726630008,
|
| 636 |
+
"grad_norm": 0.7169618606567383,
|
| 637 |
+
"learning_rate": 1.1780032733224223e-05,
|
| 638 |
+
"loss": 2.0963,
|
| 639 |
+
"step": 9000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.47656454569258966,
|
| 643 |
+
"grad_norm": 0.5874924659729004,
|
| 644 |
+
"learning_rate": 1.1910965630114568e-05,
|
| 645 |
+
"loss": 2.1036,
|
| 646 |
+
"step": 9100
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.4818015187221786,
|
| 650 |
+
"grad_norm": 0.6330558657646179,
|
| 651 |
+
"learning_rate": 1.2041898527004911e-05,
|
| 652 |
+
"loss": 2.0823,
|
| 653 |
+
"step": 9200
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.4870384917517675,
|
| 657 |
+
"grad_norm": 0.6011582016944885,
|
| 658 |
+
"learning_rate": 1.2172831423895254e-05,
|
| 659 |
+
"loss": 2.0705,
|
| 660 |
+
"step": 9300
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.49227546478135636,
|
| 664 |
+
"grad_norm": 0.6566969156265259,
|
| 665 |
+
"learning_rate": 1.2303764320785598e-05,
|
| 666 |
+
"loss": 2.0875,
|
| 667 |
+
"step": 9400
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.4975124378109453,
|
| 671 |
+
"grad_norm": 0.7280591726303101,
|
| 672 |
+
"learning_rate": 1.2434697217675943e-05,
|
| 673 |
+
"loss": 2.0799,
|
| 674 |
+
"step": 9500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5027494108405341,
|
| 678 |
+
"grad_norm": 0.6386376023292542,
|
| 679 |
+
"learning_rate": 1.2565630114566286e-05,
|
| 680 |
+
"loss": 2.1116,
|
| 681 |
+
"step": 9600
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5079863838701231,
|
| 685 |
+
"grad_norm": 0.5829023122787476,
|
| 686 |
+
"learning_rate": 1.2696563011456629e-05,
|
| 687 |
+
"loss": 2.0771,
|
| 688 |
+
"step": 9700
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.513223356899712,
|
| 692 |
+
"grad_norm": 0.586355984210968,
|
| 693 |
+
"learning_rate": 1.2827495908346974e-05,
|
| 694 |
+
"loss": 2.0789,
|
| 695 |
+
"step": 9800
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5184603299293009,
|
| 699 |
+
"grad_norm": 0.6114677786827087,
|
| 700 |
+
"learning_rate": 1.2958428805237317e-05,
|
| 701 |
+
"loss": 2.0834,
|
| 702 |
+
"step": 9900
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5236973029588897,
|
| 706 |
+
"grad_norm": 0.5855059623718262,
|
| 707 |
+
"learning_rate": 1.308936170212766e-05,
|
| 708 |
+
"loss": 2.0817,
|
| 709 |
+
"step": 10000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5289342759884786,
|
| 713 |
+
"grad_norm": 0.6457023024559021,
|
| 714 |
+
"learning_rate": 1.3220294599018004e-05,
|
| 715 |
+
"loss": 2.0918,
|
| 716 |
+
"step": 10100
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5341712490180676,
|
| 720 |
+
"grad_norm": 0.693600058555603,
|
| 721 |
+
"learning_rate": 1.3351227495908348e-05,
|
| 722 |
+
"loss": 2.0968,
|
| 723 |
+
"step": 10200
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.5394082220476565,
|
| 727 |
+
"grad_norm": 0.6309637427330017,
|
| 728 |
+
"learning_rate": 1.3482160392798692e-05,
|
| 729 |
+
"loss": 2.0908,
|
| 730 |
+
"step": 10300
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5446451950772454,
|
| 734 |
+
"grad_norm": 0.6856634616851807,
|
| 735 |
+
"learning_rate": 1.3613093289689035e-05,
|
| 736 |
+
"loss": 2.0928,
|
| 737 |
+
"step": 10400
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5498821681068342,
|
| 741 |
+
"grad_norm": 0.6567341685295105,
|
| 742 |
+
"learning_rate": 1.374402618657938e-05,
|
| 743 |
+
"loss": 2.0847,
|
| 744 |
+
"step": 10500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5551191411364231,
|
| 748 |
+
"grad_norm": 0.5995367765426636,
|
| 749 |
+
"learning_rate": 1.3874959083469723e-05,
|
| 750 |
+
"loss": 2.0768,
|
| 751 |
+
"step": 10600
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.560356114166012,
|
| 755 |
+
"grad_norm": 0.6619229316711426,
|
| 756 |
+
"learning_rate": 1.4005891980360066e-05,
|
| 757 |
+
"loss": 2.0902,
|
| 758 |
+
"step": 10700
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.565593087195601,
|
| 762 |
+
"grad_norm": 0.658575713634491,
|
| 763 |
+
"learning_rate": 1.413682487725041e-05,
|
| 764 |
+
"loss": 2.0809,
|
| 765 |
+
"step": 10800
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.5708300602251899,
|
| 769 |
+
"grad_norm": 0.7023152112960815,
|
| 770 |
+
"learning_rate": 1.4267757774140754e-05,
|
| 771 |
+
"loss": 2.0527,
|
| 772 |
+
"step": 10900
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.5760670332547787,
|
| 776 |
+
"grad_norm": 0.6093494892120361,
|
| 777 |
+
"learning_rate": 1.4398690671031098e-05,
|
| 778 |
+
"loss": 2.0826,
|
| 779 |
+
"step": 11000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.5813040062843676,
|
| 783 |
+
"grad_norm": 0.6743735671043396,
|
| 784 |
+
"learning_rate": 1.452962356792144e-05,
|
| 785 |
+
"loss": 2.0829,
|
| 786 |
+
"step": 11100
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.5865409793139565,
|
| 790 |
+
"grad_norm": 0.616309404373169,
|
| 791 |
+
"learning_rate": 1.4660556464811786e-05,
|
| 792 |
+
"loss": 2.0632,
|
| 793 |
+
"step": 11200
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.5917779523435455,
|
| 797 |
+
"grad_norm": 0.6099916696548462,
|
| 798 |
+
"learning_rate": 1.4791489361702129e-05,
|
| 799 |
+
"loss": 2.0566,
|
| 800 |
+
"step": 11300
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.5970149253731343,
|
| 804 |
+
"grad_norm": 0.6355324387550354,
|
| 805 |
+
"learning_rate": 1.4922422258592472e-05,
|
| 806 |
+
"loss": 2.0677,
|
| 807 |
+
"step": 11400
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6022518984027232,
|
| 811 |
+
"grad_norm": 0.6354811191558838,
|
| 812 |
+
"learning_rate": 1.5053355155482815e-05,
|
| 813 |
+
"loss": 2.0634,
|
| 814 |
+
"step": 11500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6074888714323121,
|
| 818 |
+
"grad_norm": 0.6543029546737671,
|
| 819 |
+
"learning_rate": 1.518428805237316e-05,
|
| 820 |
+
"loss": 2.0708,
|
| 821 |
+
"step": 11600
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.612725844461901,
|
| 825 |
+
"grad_norm": 0.6910520195960999,
|
| 826 |
+
"learning_rate": 1.5315220949263502e-05,
|
| 827 |
+
"loss": 2.0637,
|
| 828 |
+
"step": 11700
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.61796281749149,
|
| 832 |
+
"grad_norm": 0.9914392828941345,
|
| 833 |
+
"learning_rate": 1.544615384615385e-05,
|
| 834 |
+
"loss": 2.0662,
|
| 835 |
+
"step": 11800
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6231997905210788,
|
| 839 |
+
"grad_norm": 0.6324429512023926,
|
| 840 |
+
"learning_rate": 1.557708674304419e-05,
|
| 841 |
+
"loss": 2.0744,
|
| 842 |
+
"step": 11900
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6284367635506677,
|
| 846 |
+
"grad_norm": 0.6607252955436707,
|
| 847 |
+
"learning_rate": 1.5708019639934535e-05,
|
| 848 |
+
"loss": 2.084,
|
| 849 |
+
"step": 12000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6336737365802566,
|
| 853 |
+
"grad_norm": 0.6728180646896362,
|
| 854 |
+
"learning_rate": 1.5838952536824878e-05,
|
| 855 |
+
"loss": 2.0656,
|
| 856 |
+
"step": 12100
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.6389107096098455,
|
| 860 |
+
"grad_norm": 0.7279180884361267,
|
| 861 |
+
"learning_rate": 1.596988543371522e-05,
|
| 862 |
+
"loss": 2.0644,
|
| 863 |
+
"step": 12200
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6441476826394344,
|
| 867 |
+
"grad_norm": 0.6170722842216492,
|
| 868 |
+
"learning_rate": 1.6100818330605564e-05,
|
| 869 |
+
"loss": 2.0591,
|
| 870 |
+
"step": 12300
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6493846556690233,
|
| 874 |
+
"grad_norm": 0.6168251633644104,
|
| 875 |
+
"learning_rate": 1.6231751227495908e-05,
|
| 876 |
+
"loss": 2.0691,
|
| 877 |
+
"step": 12400
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6546216286986122,
|
| 881 |
+
"grad_norm": 0.6229096055030823,
|
| 882 |
+
"learning_rate": 1.6362684124386254e-05,
|
| 883 |
+
"loss": 2.069,
|
| 884 |
+
"step": 12500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6598586017282011,
|
| 888 |
+
"grad_norm": 0.6832563281059265,
|
| 889 |
+
"learning_rate": 1.6493617021276598e-05,
|
| 890 |
+
"loss": 2.0429,
|
| 891 |
+
"step": 12600
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.66509557475779,
|
| 895 |
+
"grad_norm": 0.696476936340332,
|
| 896 |
+
"learning_rate": 1.662454991816694e-05,
|
| 897 |
+
"loss": 2.0573,
|
| 898 |
+
"step": 12700
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.6703325477873789,
|
| 902 |
+
"grad_norm": 0.7355523705482483,
|
| 903 |
+
"learning_rate": 1.6755482815057284e-05,
|
| 904 |
+
"loss": 2.0324,
|
| 905 |
+
"step": 12800
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.6755695208169678,
|
| 909 |
+
"grad_norm": 0.6899946928024292,
|
| 910 |
+
"learning_rate": 1.6886415711947627e-05,
|
| 911 |
+
"loss": 2.0568,
|
| 912 |
+
"step": 12900
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.6808064938465567,
|
| 916 |
+
"grad_norm": 0.7369588017463684,
|
| 917 |
+
"learning_rate": 1.701734860883797e-05,
|
| 918 |
+
"loss": 2.056,
|
| 919 |
+
"step": 13000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.6860434668761456,
|
| 923 |
+
"grad_norm": 0.645595371723175,
|
| 924 |
+
"learning_rate": 1.7148281505728314e-05,
|
| 925 |
+
"loss": 2.0609,
|
| 926 |
+
"step": 13100
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.6912804399057345,
|
| 930 |
+
"grad_norm": 0.6963063478469849,
|
| 931 |
+
"learning_rate": 1.727921440261866e-05,
|
| 932 |
+
"loss": 2.0409,
|
| 933 |
+
"step": 13200
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.6965174129353234,
|
| 937 |
+
"grad_norm": 0.614264726638794,
|
| 938 |
+
"learning_rate": 1.7410147299509003e-05,
|
| 939 |
+
"loss": 2.0473,
|
| 940 |
+
"step": 13300
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.7017543859649122,
|
| 944 |
+
"grad_norm": 0.6803929209709167,
|
| 945 |
+
"learning_rate": 1.7541080196399347e-05,
|
| 946 |
+
"loss": 2.0481,
|
| 947 |
+
"step": 13400
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7069913589945012,
|
| 951 |
+
"grad_norm": 0.6785172820091248,
|
| 952 |
+
"learning_rate": 1.767201309328969e-05,
|
| 953 |
+
"loss": 2.0663,
|
| 954 |
+
"step": 13500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7122283320240901,
|
| 958 |
+
"grad_norm": 0.6221017241477966,
|
| 959 |
+
"learning_rate": 1.7802945990180033e-05,
|
| 960 |
+
"loss": 2.046,
|
| 961 |
+
"step": 13600
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.717465305053679,
|
| 965 |
+
"grad_norm": 0.7000072002410889,
|
| 966 |
+
"learning_rate": 1.7933878887070376e-05,
|
| 967 |
+
"loss": 2.0513,
|
| 968 |
+
"step": 13700
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7227022780832679,
|
| 972 |
+
"grad_norm": 0.6962223052978516,
|
| 973 |
+
"learning_rate": 1.806481178396072e-05,
|
| 974 |
+
"loss": 2.0539,
|
| 975 |
+
"step": 13800
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.7279392511128567,
|
| 979 |
+
"grad_norm": 0.6127362847328186,
|
| 980 |
+
"learning_rate": 1.8195744680851066e-05,
|
| 981 |
+
"loss": 2.05,
|
| 982 |
+
"step": 13900
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7331762241424457,
|
| 986 |
+
"grad_norm": 0.7580384612083435,
|
| 987 |
+
"learning_rate": 1.832667757774141e-05,
|
| 988 |
+
"loss": 2.0488,
|
| 989 |
+
"step": 14000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7384131971720346,
|
| 993 |
+
"grad_norm": 0.6130460500717163,
|
| 994 |
+
"learning_rate": 1.8457610474631753e-05,
|
| 995 |
+
"loss": 2.0407,
|
| 996 |
+
"step": 14100
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.7436501702016235,
|
| 1000 |
+
"grad_norm": 0.782464861869812,
|
| 1001 |
+
"learning_rate": 1.8588543371522096e-05,
|
| 1002 |
+
"loss": 2.0328,
|
| 1003 |
+
"step": 14200
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7488871432312123,
|
| 1007 |
+
"grad_norm": 0.6607732772827148,
|
| 1008 |
+
"learning_rate": 1.871947626841244e-05,
|
| 1009 |
+
"loss": 2.0504,
|
| 1010 |
+
"step": 14300
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.7541241162608012,
|
| 1014 |
+
"grad_norm": 0.6285542249679565,
|
| 1015 |
+
"learning_rate": 1.8850409165302782e-05,
|
| 1016 |
+
"loss": 2.0513,
|
| 1017 |
+
"step": 14400
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.7593610892903901,
|
| 1021 |
+
"grad_norm": 0.723513662815094,
|
| 1022 |
+
"learning_rate": 1.8981342062193125e-05,
|
| 1023 |
+
"loss": 2.038,
|
| 1024 |
+
"step": 14500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.7645980623199791,
|
| 1028 |
+
"grad_norm": 0.6517907977104187,
|
| 1029 |
+
"learning_rate": 1.9112274959083472e-05,
|
| 1030 |
+
"loss": 2.0375,
|
| 1031 |
+
"step": 14600
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.769835035349568,
|
| 1035 |
+
"grad_norm": 0.6433988213539124,
|
| 1036 |
+
"learning_rate": 1.9243207855973815e-05,
|
| 1037 |
+
"loss": 2.0662,
|
| 1038 |
+
"step": 14700
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.7750720083791568,
|
| 1042 |
+
"grad_norm": 0.8419864177703857,
|
| 1043 |
+
"learning_rate": 1.937414075286416e-05,
|
| 1044 |
+
"loss": 2.0273,
|
| 1045 |
+
"step": 14800
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.7803089814087457,
|
| 1049 |
+
"grad_norm": 0.6121337413787842,
|
| 1050 |
+
"learning_rate": 1.95050736497545e-05,
|
| 1051 |
+
"loss": 2.051,
|
| 1052 |
+
"step": 14900
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.7855459544383346,
|
| 1056 |
+
"grad_norm": 0.6917948126792908,
|
| 1057 |
+
"learning_rate": 1.9634697217675943e-05,
|
| 1058 |
+
"loss": 2.0316,
|
| 1059 |
+
"step": 15000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.7907829274679236,
|
| 1063 |
+
"grad_norm": 0.5825837850570679,
|
| 1064 |
+
"learning_rate": 1.9765630114566286e-05,
|
| 1065 |
+
"loss": 2.0212,
|
| 1066 |
+
"step": 15100
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.7960199004975125,
|
| 1070 |
+
"grad_norm": 0.6557635068893433,
|
| 1071 |
+
"learning_rate": 1.989656301145663e-05,
|
| 1072 |
+
"loss": 2.0417,
|
| 1073 |
+
"step": 15200
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8012568735271013,
|
| 1077 |
+
"grad_norm": 0.7703428268432617,
|
| 1078 |
+
"learning_rate": 1.999850868008123e-05,
|
| 1079 |
+
"loss": 2.0376,
|
| 1080 |
+
"step": 15300
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.8064938465566902,
|
| 1084 |
+
"grad_norm": 0.6174185872077942,
|
| 1085 |
+
"learning_rate": 1.995052846880561e-05,
|
| 1086 |
+
"loss": 2.0401,
|
| 1087 |
+
"step": 15400
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8117308195862791,
|
| 1091 |
+
"grad_norm": 0.7646263837814331,
|
| 1092 |
+
"learning_rate": 1.9835285545781057e-05,
|
| 1093 |
+
"loss": 2.0355,
|
| 1094 |
+
"step": 15500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.816967792615868,
|
| 1098 |
+
"grad_norm": 0.6554729342460632,
|
| 1099 |
+
"learning_rate": 1.9653558920035776e-05,
|
| 1100 |
+
"loss": 2.06,
|
| 1101 |
+
"step": 15600
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.822204765645457,
|
| 1105 |
+
"grad_norm": 0.6709089279174805,
|
| 1106 |
+
"learning_rate": 1.940657701131479e-05,
|
| 1107 |
+
"loss": 2.0339,
|
| 1108 |
+
"step": 15700
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8274417386750458,
|
| 1112 |
+
"grad_norm": 0.6536616086959839,
|
| 1113 |
+
"learning_rate": 1.9096009346315668e-05,
|
| 1114 |
+
"loss": 2.0411,
|
| 1115 |
+
"step": 15800
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8326787117046347,
|
| 1119 |
+
"grad_norm": 0.6588115096092224,
|
| 1120 |
+
"learning_rate": 1.8723955273167993e-05,
|
| 1121 |
+
"loss": 2.0341,
|
| 1122 |
+
"step": 15900
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8379156847342236,
|
| 1126 |
+
"grad_norm": 0.6266176104545593,
|
| 1127 |
+
"learning_rate": 1.8292929770443562e-05,
|
| 1128 |
+
"loss": 2.0349,
|
| 1129 |
+
"step": 16000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8431526577638125,
|
| 1133 |
+
"grad_norm": 0.6606239676475525,
|
| 1134 |
+
"learning_rate": 1.780584644662429e-05,
|
| 1135 |
+
"loss": 2.041,
|
| 1136 |
+
"step": 16100
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.8483896307934015,
|
| 1140 |
+
"grad_norm": 0.6853938102722168,
|
| 1141 |
+
"learning_rate": 1.726599784494649e-05,
|
| 1142 |
+
"loss": 2.0409,
|
| 1143 |
+
"step": 16200
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.8536266038229903,
|
| 1147 |
+
"grad_norm": 0.6542646288871765,
|
| 1148 |
+
"learning_rate": 1.66770331867552e-05,
|
| 1149 |
+
"loss": 2.0401,
|
| 1150 |
+
"step": 16300
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.8588635768525792,
|
| 1154 |
+
"grad_norm": 0.6488157510757446,
|
| 1155 |
+
"learning_rate": 1.6042933703816995e-05,
|
| 1156 |
+
"loss": 2.0329,
|
| 1157 |
+
"step": 16400
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.8641005498821681,
|
| 1161 |
+
"grad_norm": 0.6016569137573242,
|
| 1162 |
+
"learning_rate": 1.5367985726337818e-05,
|
| 1163 |
+
"loss": 2.0276,
|
| 1164 |
+
"step": 16500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.869337522911757,
|
| 1168 |
+
"grad_norm": 0.6777680516242981,
|
| 1169 |
+
"learning_rate": 1.4656751708603102e-05,
|
| 1170 |
+
"loss": 2.0357,
|
| 1171 |
+
"step": 16600
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.874574495941346,
|
| 1175 |
+
"grad_norm": 0.5954397916793823,
|
| 1176 |
+
"learning_rate": 1.3914039388098432e-05,
|
| 1177 |
+
"loss": 2.0235,
|
| 1178 |
+
"step": 16700
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.8798114689709348,
|
| 1182 |
+
"grad_norm": 0.6781564950942993,
|
| 1183 |
+
"learning_rate": 1.3144869286586354e-05,
|
| 1184 |
+
"loss": 2.0114,
|
| 1185 |
+
"step": 16800
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.8850484420005237,
|
| 1189 |
+
"grad_norm": 0.6111444234848022,
|
| 1190 |
+
"learning_rate": 1.2354440772822623e-05,
|
| 1191 |
+
"loss": 2.0232,
|
| 1192 |
+
"step": 16900
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.8902854150301126,
|
| 1196 |
+
"grad_norm": 0.6590002179145813,
|
| 1197 |
+
"learning_rate": 1.1548096916318175e-05,
|
| 1198 |
+
"loss": 2.0327,
|
| 1199 |
+
"step": 17000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.8955223880597015,
|
| 1203 |
+
"grad_norm": 0.6230506300926208,
|
| 1204 |
+
"learning_rate": 1.0731288369725337e-05,
|
| 1205 |
+
"loss": 2.0153,
|
| 1206 |
+
"step": 17100
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.9007593610892904,
|
| 1210 |
+
"grad_norm": 0.7188674211502075,
|
| 1211 |
+
"learning_rate": 9.909536523992925e-06,
|
| 1212 |
+
"loss": 2.0266,
|
| 1213 |
+
"step": 17200
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.9059963341188793,
|
| 1217 |
+
"grad_norm": 0.6554442048072815,
|
| 1218 |
+
"learning_rate": 9.08839618535077e-06,
|
| 1219 |
+
"loss": 2.0305,
|
| 1220 |
+
"step": 17300
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9112333071484682,
|
| 1224 |
+
"grad_norm": 0.6578665971755981,
|
| 1225 |
+
"learning_rate": 8.273418026416578e-06,
|
| 1226 |
+
"loss": 2.0425,
|
| 1227 |
+
"step": 17400
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.9164702801780571,
|
| 1231 |
+
"grad_norm": 0.7747833132743835,
|
| 1232 |
+
"learning_rate": 7.470111065244805e-06,
|
| 1233 |
+
"loss": 2.0261,
|
| 1234 |
+
"step": 17500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.921707253207646,
|
| 1238 |
+
"grad_norm": 0.6754171252250671,
|
| 1239 |
+
"learning_rate": 6.6916652667519855e-06,
|
| 1240 |
+
"loss": 2.022,
|
| 1241 |
+
"step": 17600
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.9269442262372348,
|
| 1245 |
+
"grad_norm": 0.7135418057441711,
|
| 1246 |
+
"learning_rate": 5.927625476285426e-06,
|
| 1247 |
+
"loss": 2.0263,
|
| 1248 |
+
"step": 17700
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9321811992668237,
|
| 1252 |
+
"grad_norm": 0.607435405254364,
|
| 1253 |
+
"learning_rate": 5.191113766822905e-06,
|
| 1254 |
+
"loss": 2.0193,
|
| 1255 |
+
"step": 17800
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.9374181722964127,
|
| 1259 |
+
"grad_norm": 0.6597747802734375,
|
| 1260 |
+
"learning_rate": 4.487108745778958e-06,
|
| 1261 |
+
"loss": 2.0126,
|
| 1262 |
+
"step": 17900
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9426551453260016,
|
| 1266 |
+
"grad_norm": 0.6818113923072815,
|
| 1267 |
+
"learning_rate": 3.820369284699823e-06,
|
| 1268 |
+
"loss": 2.0266,
|
| 1269 |
+
"step": 18000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 0.9478921183555905,
|
| 1273 |
+
"grad_norm": 0.6001537442207336,
|
| 1274 |
+
"learning_rate": 3.195402350659945e-06,
|
| 1275 |
+
"loss": 2.0342,
|
| 1276 |
+
"step": 18100
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 0.9531290913851793,
|
| 1280 |
+
"grad_norm": 0.5956875085830688,
|
| 1281 |
+
"learning_rate": 2.616432540460255e-06,
|
| 1282 |
+
"loss": 2.0364,
|
| 1283 |
+
"step": 18200
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 0.9583660644147682,
|
| 1287 |
+
"grad_norm": 0.663137674331665,
|
| 1288 |
+
"learning_rate": 2.0873735235683535e-06,
|
| 1289 |
+
"loss": 2.0167,
|
| 1290 |
+
"step": 18300
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 0.9636030374443572,
|
| 1294 |
+
"grad_norm": 0.6755833625793457,
|
| 1295 |
+
"learning_rate": 1.6118015868380387e-06,
|
| 1296 |
+
"loss": 2.0252,
|
| 1297 |
+
"step": 18400
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 0.9688400104739461,
|
| 1301 |
+
"grad_norm": 0.638041615486145,
|
| 1302 |
+
"learning_rate": 1.1929314598383423e-06,
|
| 1303 |
+
"loss": 2.0212,
|
| 1304 |
+
"step": 18500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 0.974076983503535,
|
| 1308 |
+
"grad_norm": 0.6840391755104065,
|
| 1309 |
+
"learning_rate": 8.335945842058524e-07,
|
| 1310 |
+
"loss": 2.0244,
|
| 1311 |
+
"step": 18600
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 0.9793139565331238,
|
| 1315 |
+
"grad_norm": 0.733739972114563,
|
| 1316 |
+
"learning_rate": 5.362199739132656e-07,
|
| 1317 |
+
"loss": 2.0167,
|
| 1318 |
+
"step": 18700
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 0.9845509295627127,
|
| 1322 |
+
"grad_norm": 0.5909525752067566,
|
| 1323 |
+
"learning_rate": 3.028177958332512e-07,
|
| 1324 |
+
"loss": 2.0278,
|
| 1325 |
+
"step": 18800
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 0.9897879025923016,
|
| 1329 |
+
"grad_norm": 0.7036847472190857,
|
| 1330 |
+
"learning_rate": 1.349657815883032e-07,
|
| 1331 |
+
"loss": 2.0359,
|
| 1332 |
+
"step": 18900
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 0.9950248756218906,
|
| 1336 |
+
"grad_norm": 0.6682682037353516,
|
| 1337 |
+
"learning_rate": 3.379856253855951e-08,
|
| 1338 |
+
"loss": 2.0143,
|
| 1339 |
+
"step": 19000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 1.0,
|
| 1343 |
+
"step": 19095,
|
| 1344 |
+
"total_flos": 2.790413903660581e+18,
|
| 1345 |
+
"train_loss": 2.1026804686029656,
|
| 1346 |
+
"train_runtime": 6094.1614,
|
| 1347 |
+
"train_samples_per_second": 50.133,
|
| 1348 |
+
"train_steps_per_second": 3.133
|
| 1349 |
+
}
|
| 1350 |
+
],
|
| 1351 |
+
"logging_steps": 100,
|
| 1352 |
+
"max_steps": 19095,
|
| 1353 |
+
"num_input_tokens_seen": 0,
|
| 1354 |
+
"num_train_epochs": 1,
|
| 1355 |
+
"save_steps": 500,
|
| 1356 |
+
"stateful_callbacks": {
|
| 1357 |
+
"TrainerControl": {
|
| 1358 |
+
"args": {
|
| 1359 |
+
"should_epoch_stop": false,
|
| 1360 |
+
"should_evaluate": false,
|
| 1361 |
+
"should_log": false,
|
| 1362 |
+
"should_save": true,
|
| 1363 |
+
"should_training_stop": true
|
| 1364 |
+
},
|
| 1365 |
+
"attributes": {}
|
| 1366 |
+
}
|
| 1367 |
+
},
|
| 1368 |
+
"total_flos": 2.790413903660581e+18,
|
| 1369 |
+
"train_batch_size": 2,
|
| 1370 |
+
"trial_name": null,
|
| 1371 |
+
"trial_params": null
|
| 1372 |
+
}
|