Commit
•
61e5666
1
Parent(s):
fd53d71
initialize model for Dark Souls Item Description in Chinese
Browse files- config.json +33 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- result.xlsx +0 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1071 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "bigscience/bloomz-560m",
|
| 3 |
+
"apply_residual_connection_post_layernorm": false,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"BloomForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"attention_softmax_in_fp32": true,
|
| 9 |
+
"bias_dropout_fusion": true,
|
| 10 |
+
"bos_token_id": 1,
|
| 11 |
+
"eos_token_id": 2,
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 1024,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"layer_norm_epsilon": 1e-05,
|
| 16 |
+
"masked_softmax_fusion": true,
|
| 17 |
+
"model_type": "bloom",
|
| 18 |
+
"n_head": 16,
|
| 19 |
+
"n_inner": null,
|
| 20 |
+
"n_layer": 24,
|
| 21 |
+
"offset_alibi": 100,
|
| 22 |
+
"pad_token_id": 3,
|
| 23 |
+
"pretraining_tp": 1,
|
| 24 |
+
"seq_length": 2048,
|
| 25 |
+
"skip_bias_add": true,
|
| 26 |
+
"skip_bias_add_qkv": false,
|
| 27 |
+
"slow_but_exact": false,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.39.2",
|
| 30 |
+
"unk_token_id": 0,
|
| 31 |
+
"use_cache": true,
|
| 32 |
+
"vocab_size": 250880
|
| 33 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 3,
|
| 6 |
+
"transformers_version": "4.39.2"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b98f507cdc3268b53bcd171eb915af083f9352b5f24888923f74adbd8d35766
|
| 3 |
+
size 2236892304
|
result.xlsx
ADDED
|
Binary file (11.5 kB). View file
|
|
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1784c9e20ffdc46b706882695c2108245d7626a328b6d70a37d079ad1fbbc989
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4c015a10e0c470ec1b84d8950d7dc21746e37ddfe7a03134fabd9c91c2997c6a
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1071 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.4140012070006036,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.02,
|
| 13 |
+
"grad_norm": 53.90464782714844,
|
| 14 |
+
"learning_rate": 9.946323134728933e-06,
|
| 15 |
+
"loss": 5.2914,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03,
|
| 20 |
+
"grad_norm": 53.05311965942383,
|
| 21 |
+
"learning_rate": 9.892646269457864e-06,
|
| 22 |
+
"loss": 4.9539,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05,
|
| 27 |
+
"grad_norm": 64.5956802368164,
|
| 28 |
+
"learning_rate": 9.838969404186796e-06,
|
| 29 |
+
"loss": 4.7374,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06,
|
| 34 |
+
"grad_norm": 54.43781280517578,
|
| 35 |
+
"learning_rate": 9.785292538915728e-06,
|
| 36 |
+
"loss": 4.5988,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.08,
|
| 41 |
+
"grad_norm": 48.784854888916016,
|
| 42 |
+
"learning_rate": 9.73161567364466e-06,
|
| 43 |
+
"loss": 4.4899,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.1,
|
| 48 |
+
"grad_norm": 56.800575256347656,
|
| 49 |
+
"learning_rate": 9.677938808373591e-06,
|
| 50 |
+
"loss": 4.4391,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.11,
|
| 55 |
+
"grad_norm": 51.58082962036133,
|
| 56 |
+
"learning_rate": 9.624261943102525e-06,
|
| 57 |
+
"loss": 4.3347,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.13,
|
| 62 |
+
"grad_norm": 50.625308990478516,
|
| 63 |
+
"learning_rate": 9.570585077831455e-06,
|
| 64 |
+
"loss": 4.3145,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.14,
|
| 69 |
+
"grad_norm": 46.56022262573242,
|
| 70 |
+
"learning_rate": 9.516908212560388e-06,
|
| 71 |
+
"loss": 4.1742,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.16,
|
| 76 |
+
"grad_norm": 51.267852783203125,
|
| 77 |
+
"learning_rate": 9.463231347289318e-06,
|
| 78 |
+
"loss": 4.1308,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.18,
|
| 83 |
+
"grad_norm": 52.627098083496094,
|
| 84 |
+
"learning_rate": 9.40955448201825e-06,
|
| 85 |
+
"loss": 4.1251,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.19,
|
| 90 |
+
"grad_norm": 51.49211120605469,
|
| 91 |
+
"learning_rate": 9.355877616747183e-06,
|
| 92 |
+
"loss": 4.073,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.21,
|
| 97 |
+
"grad_norm": 50.88691329956055,
|
| 98 |
+
"learning_rate": 9.302200751476115e-06,
|
| 99 |
+
"loss": 3.9998,
|
| 100 |
+
"step": 130
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.23,
|
| 104 |
+
"grad_norm": 45.862796783447266,
|
| 105 |
+
"learning_rate": 9.248523886205046e-06,
|
| 106 |
+
"loss": 3.8666,
|
| 107 |
+
"step": 140
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.24,
|
| 111 |
+
"grad_norm": 51.27406692504883,
|
| 112 |
+
"learning_rate": 9.194847020933978e-06,
|
| 113 |
+
"loss": 3.9477,
|
| 114 |
+
"step": 150
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.26,
|
| 118 |
+
"grad_norm": 47.50687026977539,
|
| 119 |
+
"learning_rate": 9.14117015566291e-06,
|
| 120 |
+
"loss": 3.7899,
|
| 121 |
+
"step": 160
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.27,
|
| 125 |
+
"grad_norm": 48.58837127685547,
|
| 126 |
+
"learning_rate": 9.087493290391842e-06,
|
| 127 |
+
"loss": 3.7295,
|
| 128 |
+
"step": 170
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.29,
|
| 132 |
+
"grad_norm": 48.3990592956543,
|
| 133 |
+
"learning_rate": 9.033816425120775e-06,
|
| 134 |
+
"loss": 3.7023,
|
| 135 |
+
"step": 180
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.31,
|
| 139 |
+
"grad_norm": 46.30027770996094,
|
| 140 |
+
"learning_rate": 8.980139559849705e-06,
|
| 141 |
+
"loss": 3.6479,
|
| 142 |
+
"step": 190
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.32,
|
| 146 |
+
"grad_norm": 47.15484619140625,
|
| 147 |
+
"learning_rate": 8.926462694578637e-06,
|
| 148 |
+
"loss": 3.6161,
|
| 149 |
+
"step": 200
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.34,
|
| 153 |
+
"grad_norm": 49.22386932373047,
|
| 154 |
+
"learning_rate": 8.87278582930757e-06,
|
| 155 |
+
"loss": 3.5503,
|
| 156 |
+
"step": 210
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.35,
|
| 160 |
+
"grad_norm": 45.485557556152344,
|
| 161 |
+
"learning_rate": 8.819108964036502e-06,
|
| 162 |
+
"loss": 3.4459,
|
| 163 |
+
"step": 220
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.37,
|
| 167 |
+
"grad_norm": 50.84394454956055,
|
| 168 |
+
"learning_rate": 8.765432098765432e-06,
|
| 169 |
+
"loss": 3.4918,
|
| 170 |
+
"step": 230
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.39,
|
| 174 |
+
"grad_norm": 43.17815399169922,
|
| 175 |
+
"learning_rate": 8.711755233494365e-06,
|
| 176 |
+
"loss": 3.3527,
|
| 177 |
+
"step": 240
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.4,
|
| 181 |
+
"grad_norm": 41.90092849731445,
|
| 182 |
+
"learning_rate": 8.658078368223295e-06,
|
| 183 |
+
"loss": 3.347,
|
| 184 |
+
"step": 250
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.42,
|
| 188 |
+
"grad_norm": 43.23625564575195,
|
| 189 |
+
"learning_rate": 8.60440150295223e-06,
|
| 190 |
+
"loss": 3.3433,
|
| 191 |
+
"step": 260
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.43,
|
| 195 |
+
"grad_norm": 44.05680847167969,
|
| 196 |
+
"learning_rate": 8.55072463768116e-06,
|
| 197 |
+
"loss": 3.2831,
|
| 198 |
+
"step": 270
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.45,
|
| 202 |
+
"grad_norm": 46.9146842956543,
|
| 203 |
+
"learning_rate": 8.497047772410092e-06,
|
| 204 |
+
"loss": 3.2932,
|
| 205 |
+
"step": 280
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.47,
|
| 209 |
+
"grad_norm": 48.542701721191406,
|
| 210 |
+
"learning_rate": 8.443370907139024e-06,
|
| 211 |
+
"loss": 3.2108,
|
| 212 |
+
"step": 290
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.48,
|
| 216 |
+
"grad_norm": 39.1197509765625,
|
| 217 |
+
"learning_rate": 8.389694041867955e-06,
|
| 218 |
+
"loss": 3.1704,
|
| 219 |
+
"step": 300
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.5,
|
| 223 |
+
"grad_norm": 45.300697326660156,
|
| 224 |
+
"learning_rate": 8.336017176596887e-06,
|
| 225 |
+
"loss": 3.2239,
|
| 226 |
+
"step": 310
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.51,
|
| 230 |
+
"grad_norm": 47.439823150634766,
|
| 231 |
+
"learning_rate": 8.28234031132582e-06,
|
| 232 |
+
"loss": 3.1709,
|
| 233 |
+
"step": 320
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.53,
|
| 237 |
+
"grad_norm": 42.67869567871094,
|
| 238 |
+
"learning_rate": 8.228663446054752e-06,
|
| 239 |
+
"loss": 3.0708,
|
| 240 |
+
"step": 330
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.55,
|
| 244 |
+
"grad_norm": 46.0591926574707,
|
| 245 |
+
"learning_rate": 8.174986580783682e-06,
|
| 246 |
+
"loss": 3.0959,
|
| 247 |
+
"step": 340
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.56,
|
| 251 |
+
"grad_norm": 42.061805725097656,
|
| 252 |
+
"learning_rate": 8.121309715512614e-06,
|
| 253 |
+
"loss": 3.0442,
|
| 254 |
+
"step": 350
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.58,
|
| 258 |
+
"grad_norm": 44.487159729003906,
|
| 259 |
+
"learning_rate": 8.067632850241547e-06,
|
| 260 |
+
"loss": 2.8881,
|
| 261 |
+
"step": 360
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.6,
|
| 265 |
+
"grad_norm": 44.4419059753418,
|
| 266 |
+
"learning_rate": 8.013955984970479e-06,
|
| 267 |
+
"loss": 2.9994,
|
| 268 |
+
"step": 370
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.61,
|
| 272 |
+
"grad_norm": 40.00767135620117,
|
| 273 |
+
"learning_rate": 7.96027911969941e-06,
|
| 274 |
+
"loss": 2.8931,
|
| 275 |
+
"step": 380
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.63,
|
| 279 |
+
"grad_norm": 44.14014434814453,
|
| 280 |
+
"learning_rate": 7.906602254428342e-06,
|
| 281 |
+
"loss": 2.9288,
|
| 282 |
+
"step": 390
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.64,
|
| 286 |
+
"grad_norm": 42.418888092041016,
|
| 287 |
+
"learning_rate": 7.852925389157274e-06,
|
| 288 |
+
"loss": 2.7776,
|
| 289 |
+
"step": 400
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.66,
|
| 293 |
+
"grad_norm": 49.98628234863281,
|
| 294 |
+
"learning_rate": 7.799248523886206e-06,
|
| 295 |
+
"loss": 2.8408,
|
| 296 |
+
"step": 410
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.68,
|
| 300 |
+
"grad_norm": 39.240352630615234,
|
| 301 |
+
"learning_rate": 7.745571658615137e-06,
|
| 302 |
+
"loss": 2.7995,
|
| 303 |
+
"step": 420
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.69,
|
| 307 |
+
"grad_norm": 43.185569763183594,
|
| 308 |
+
"learning_rate": 7.691894793344069e-06,
|
| 309 |
+
"loss": 2.7486,
|
| 310 |
+
"step": 430
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.71,
|
| 314 |
+
"grad_norm": 37.41790008544922,
|
| 315 |
+
"learning_rate": 7.638217928073001e-06,
|
| 316 |
+
"loss": 2.7459,
|
| 317 |
+
"step": 440
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.72,
|
| 321 |
+
"grad_norm": 38.520973205566406,
|
| 322 |
+
"learning_rate": 7.584541062801934e-06,
|
| 323 |
+
"loss": 2.6785,
|
| 324 |
+
"step": 450
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.74,
|
| 328 |
+
"grad_norm": 42.01523971557617,
|
| 329 |
+
"learning_rate": 7.530864197530865e-06,
|
| 330 |
+
"loss": 2.7745,
|
| 331 |
+
"step": 460
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.76,
|
| 335 |
+
"grad_norm": 40.55296325683594,
|
| 336 |
+
"learning_rate": 7.477187332259796e-06,
|
| 337 |
+
"loss": 2.7046,
|
| 338 |
+
"step": 470
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.77,
|
| 342 |
+
"grad_norm": 38.31270217895508,
|
| 343 |
+
"learning_rate": 7.423510466988728e-06,
|
| 344 |
+
"loss": 2.6057,
|
| 345 |
+
"step": 480
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.79,
|
| 349 |
+
"grad_norm": 37.31733322143555,
|
| 350 |
+
"learning_rate": 7.369833601717661e-06,
|
| 351 |
+
"loss": 2.5229,
|
| 352 |
+
"step": 490
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.8,
|
| 356 |
+
"grad_norm": 37.55893325805664,
|
| 357 |
+
"learning_rate": 7.316156736446592e-06,
|
| 358 |
+
"loss": 2.5846,
|
| 359 |
+
"step": 500
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.82,
|
| 363 |
+
"grad_norm": 40.64212417602539,
|
| 364 |
+
"learning_rate": 7.262479871175524e-06,
|
| 365 |
+
"loss": 2.4992,
|
| 366 |
+
"step": 510
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.84,
|
| 370 |
+
"grad_norm": 41.34495544433594,
|
| 371 |
+
"learning_rate": 7.208803005904456e-06,
|
| 372 |
+
"loss": 2.5575,
|
| 373 |
+
"step": 520
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.85,
|
| 377 |
+
"grad_norm": 38.53882598876953,
|
| 378 |
+
"learning_rate": 7.155126140633387e-06,
|
| 379 |
+
"loss": 2.5468,
|
| 380 |
+
"step": 530
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.87,
|
| 384 |
+
"grad_norm": 39.23023986816406,
|
| 385 |
+
"learning_rate": 7.10144927536232e-06,
|
| 386 |
+
"loss": 2.5009,
|
| 387 |
+
"step": 540
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.89,
|
| 391 |
+
"grad_norm": 35.68672180175781,
|
| 392 |
+
"learning_rate": 7.047772410091251e-06,
|
| 393 |
+
"loss": 2.2929,
|
| 394 |
+
"step": 550
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.9,
|
| 398 |
+
"grad_norm": 38.0786247253418,
|
| 399 |
+
"learning_rate": 6.994095544820183e-06,
|
| 400 |
+
"loss": 2.3178,
|
| 401 |
+
"step": 560
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.92,
|
| 405 |
+
"grad_norm": 42.21394348144531,
|
| 406 |
+
"learning_rate": 6.940418679549115e-06,
|
| 407 |
+
"loss": 2.4645,
|
| 408 |
+
"step": 570
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.93,
|
| 412 |
+
"grad_norm": 37.197696685791016,
|
| 413 |
+
"learning_rate": 6.886741814278046e-06,
|
| 414 |
+
"loss": 2.4797,
|
| 415 |
+
"step": 580
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.95,
|
| 419 |
+
"grad_norm": 40.341514587402344,
|
| 420 |
+
"learning_rate": 6.833064949006979e-06,
|
| 421 |
+
"loss": 2.3048,
|
| 422 |
+
"step": 590
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.97,
|
| 426 |
+
"grad_norm": 36.51962661743164,
|
| 427 |
+
"learning_rate": 6.779388083735911e-06,
|
| 428 |
+
"loss": 2.4652,
|
| 429 |
+
"step": 600
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.98,
|
| 433 |
+
"grad_norm": 35.32337951660156,
|
| 434 |
+
"learning_rate": 6.725711218464842e-06,
|
| 435 |
+
"loss": 2.2979,
|
| 436 |
+
"step": 610
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.0,
|
| 440 |
+
"grad_norm": 40.75404739379883,
|
| 441 |
+
"learning_rate": 6.672034353193773e-06,
|
| 442 |
+
"loss": 2.354,
|
| 443 |
+
"step": 620
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.01,
|
| 447 |
+
"grad_norm": 30.498510360717773,
|
| 448 |
+
"learning_rate": 6.6183574879227065e-06,
|
| 449 |
+
"loss": 1.8592,
|
| 450 |
+
"step": 630
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.03,
|
| 454 |
+
"grad_norm": 31.13385772705078,
|
| 455 |
+
"learning_rate": 6.564680622651638e-06,
|
| 456 |
+
"loss": 1.7776,
|
| 457 |
+
"step": 640
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.05,
|
| 461 |
+
"grad_norm": 34.9401969909668,
|
| 462 |
+
"learning_rate": 6.511003757380569e-06,
|
| 463 |
+
"loss": 1.9329,
|
| 464 |
+
"step": 650
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.06,
|
| 468 |
+
"grad_norm": 32.90480041503906,
|
| 469 |
+
"learning_rate": 6.457326892109501e-06,
|
| 470 |
+
"loss": 1.8312,
|
| 471 |
+
"step": 660
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.08,
|
| 475 |
+
"grad_norm": 32.94902420043945,
|
| 476 |
+
"learning_rate": 6.403650026838433e-06,
|
| 477 |
+
"loss": 1.8556,
|
| 478 |
+
"step": 670
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.09,
|
| 482 |
+
"grad_norm": 32.02881622314453,
|
| 483 |
+
"learning_rate": 6.349973161567365e-06,
|
| 484 |
+
"loss": 1.8142,
|
| 485 |
+
"step": 680
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.11,
|
| 489 |
+
"grad_norm": 32.752323150634766,
|
| 490 |
+
"learning_rate": 6.296296296296297e-06,
|
| 491 |
+
"loss": 1.7429,
|
| 492 |
+
"step": 690
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.13,
|
| 496 |
+
"grad_norm": 31.938289642333984,
|
| 497 |
+
"learning_rate": 6.242619431025229e-06,
|
| 498 |
+
"loss": 1.821,
|
| 499 |
+
"step": 700
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.14,
|
| 503 |
+
"grad_norm": 32.64255142211914,
|
| 504 |
+
"learning_rate": 6.18894256575416e-06,
|
| 505 |
+
"loss": 1.7492,
|
| 506 |
+
"step": 710
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.16,
|
| 510 |
+
"grad_norm": 30.172483444213867,
|
| 511 |
+
"learning_rate": 6.135265700483092e-06,
|
| 512 |
+
"loss": 1.7661,
|
| 513 |
+
"step": 720
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.17,
|
| 517 |
+
"grad_norm": 32.1895637512207,
|
| 518 |
+
"learning_rate": 6.081588835212025e-06,
|
| 519 |
+
"loss": 1.6979,
|
| 520 |
+
"step": 730
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.19,
|
| 524 |
+
"grad_norm": 32.555870056152344,
|
| 525 |
+
"learning_rate": 6.027911969940956e-06,
|
| 526 |
+
"loss": 1.778,
|
| 527 |
+
"step": 740
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.21,
|
| 531 |
+
"grad_norm": 31.702539443969727,
|
| 532 |
+
"learning_rate": 5.974235104669888e-06,
|
| 533 |
+
"loss": 1.7263,
|
| 534 |
+
"step": 750
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.22,
|
| 538 |
+
"grad_norm": 32.07310104370117,
|
| 539 |
+
"learning_rate": 5.920558239398819e-06,
|
| 540 |
+
"loss": 1.5491,
|
| 541 |
+
"step": 760
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.24,
|
| 545 |
+
"grad_norm": 31.130224227905273,
|
| 546 |
+
"learning_rate": 5.866881374127752e-06,
|
| 547 |
+
"loss": 1.6249,
|
| 548 |
+
"step": 770
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.26,
|
| 552 |
+
"grad_norm": 39.838436126708984,
|
| 553 |
+
"learning_rate": 5.8132045088566835e-06,
|
| 554 |
+
"loss": 1.6721,
|
| 555 |
+
"step": 780
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.27,
|
| 559 |
+
"grad_norm": 33.75567626953125,
|
| 560 |
+
"learning_rate": 5.759527643585615e-06,
|
| 561 |
+
"loss": 1.6779,
|
| 562 |
+
"step": 790
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.29,
|
| 566 |
+
"grad_norm": 31.251935958862305,
|
| 567 |
+
"learning_rate": 5.705850778314546e-06,
|
| 568 |
+
"loss": 1.6401,
|
| 569 |
+
"step": 800
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.3,
|
| 573 |
+
"grad_norm": 31.644649505615234,
|
| 574 |
+
"learning_rate": 5.652173913043479e-06,
|
| 575 |
+
"loss": 1.6647,
|
| 576 |
+
"step": 810
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.32,
|
| 580 |
+
"grad_norm": 30.14424705505371,
|
| 581 |
+
"learning_rate": 5.598497047772411e-06,
|
| 582 |
+
"loss": 1.6887,
|
| 583 |
+
"step": 820
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.34,
|
| 587 |
+
"grad_norm": 29.70695686340332,
|
| 588 |
+
"learning_rate": 5.544820182501342e-06,
|
| 589 |
+
"loss": 1.5387,
|
| 590 |
+
"step": 830
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.35,
|
| 594 |
+
"grad_norm": 31.330068588256836,
|
| 595 |
+
"learning_rate": 5.4911433172302745e-06,
|
| 596 |
+
"loss": 1.6414,
|
| 597 |
+
"step": 840
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.37,
|
| 601 |
+
"grad_norm": 32.08658981323242,
|
| 602 |
+
"learning_rate": 5.437466451959206e-06,
|
| 603 |
+
"loss": 1.6158,
|
| 604 |
+
"step": 850
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.38,
|
| 608 |
+
"grad_norm": 33.42084503173828,
|
| 609 |
+
"learning_rate": 5.383789586688137e-06,
|
| 610 |
+
"loss": 1.6728,
|
| 611 |
+
"step": 860
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.4,
|
| 615 |
+
"grad_norm": 32.10792922973633,
|
| 616 |
+
"learning_rate": 5.3301127214170704e-06,
|
| 617 |
+
"loss": 1.5365,
|
| 618 |
+
"step": 870
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.42,
|
| 622 |
+
"grad_norm": 34.231239318847656,
|
| 623 |
+
"learning_rate": 5.276435856146002e-06,
|
| 624 |
+
"loss": 1.584,
|
| 625 |
+
"step": 880
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.43,
|
| 629 |
+
"grad_norm": 32.19587326049805,
|
| 630 |
+
"learning_rate": 5.222758990874933e-06,
|
| 631 |
+
"loss": 1.6233,
|
| 632 |
+
"step": 890
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.45,
|
| 636 |
+
"grad_norm": 30.36279296875,
|
| 637 |
+
"learning_rate": 5.169082125603865e-06,
|
| 638 |
+
"loss": 1.5246,
|
| 639 |
+
"step": 900
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.46,
|
| 643 |
+
"grad_norm": 33.34714889526367,
|
| 644 |
+
"learning_rate": 5.115405260332798e-06,
|
| 645 |
+
"loss": 1.5514,
|
| 646 |
+
"step": 910
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.48,
|
| 650 |
+
"grad_norm": 32.581424713134766,
|
| 651 |
+
"learning_rate": 5.061728395061729e-06,
|
| 652 |
+
"loss": 1.495,
|
| 653 |
+
"step": 920
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.5,
|
| 657 |
+
"grad_norm": 33.158203125,
|
| 658 |
+
"learning_rate": 5.0080515297906606e-06,
|
| 659 |
+
"loss": 1.5546,
|
| 660 |
+
"step": 930
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.51,
|
| 664 |
+
"grad_norm": 29.796606063842773,
|
| 665 |
+
"learning_rate": 4.954374664519592e-06,
|
| 666 |
+
"loss": 1.5141,
|
| 667 |
+
"step": 940
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.53,
|
| 671 |
+
"grad_norm": 31.936180114746094,
|
| 672 |
+
"learning_rate": 4.900697799248524e-06,
|
| 673 |
+
"loss": 1.5156,
|
| 674 |
+
"step": 950
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.54,
|
| 678 |
+
"grad_norm": 30.770095825195312,
|
| 679 |
+
"learning_rate": 4.847020933977456e-06,
|
| 680 |
+
"loss": 1.5262,
|
| 681 |
+
"step": 960
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.56,
|
| 685 |
+
"grad_norm": 32.497520446777344,
|
| 686 |
+
"learning_rate": 4.793344068706388e-06,
|
| 687 |
+
"loss": 1.5303,
|
| 688 |
+
"step": 970
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.58,
|
| 692 |
+
"grad_norm": 31.067218780517578,
|
| 693 |
+
"learning_rate": 4.739667203435319e-06,
|
| 694 |
+
"loss": 1.4994,
|
| 695 |
+
"step": 980
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.59,
|
| 699 |
+
"grad_norm": 27.720073699951172,
|
| 700 |
+
"learning_rate": 4.6859903381642516e-06,
|
| 701 |
+
"loss": 1.4268,
|
| 702 |
+
"step": 990
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.61,
|
| 706 |
+
"grad_norm": 30.310941696166992,
|
| 707 |
+
"learning_rate": 4.632313472893184e-06,
|
| 708 |
+
"loss": 1.4636,
|
| 709 |
+
"step": 1000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.63,
|
| 713 |
+
"grad_norm": 33.62602996826172,
|
| 714 |
+
"learning_rate": 4.578636607622115e-06,
|
| 715 |
+
"loss": 1.4783,
|
| 716 |
+
"step": 1010
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.64,
|
| 720 |
+
"grad_norm": 28.9564266204834,
|
| 721 |
+
"learning_rate": 4.5249597423510475e-06,
|
| 722 |
+
"loss": 1.359,
|
| 723 |
+
"step": 1020
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.66,
|
| 727 |
+
"grad_norm": 29.886262893676758,
|
| 728 |
+
"learning_rate": 4.471282877079979e-06,
|
| 729 |
+
"loss": 1.4405,
|
| 730 |
+
"step": 1030
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.67,
|
| 734 |
+
"grad_norm": 26.291038513183594,
|
| 735 |
+
"learning_rate": 4.417606011808911e-06,
|
| 736 |
+
"loss": 1.3914,
|
| 737 |
+
"step": 1040
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.69,
|
| 741 |
+
"grad_norm": 30.628904342651367,
|
| 742 |
+
"learning_rate": 4.3639291465378425e-06,
|
| 743 |
+
"loss": 1.4335,
|
| 744 |
+
"step": 1050
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.71,
|
| 748 |
+
"grad_norm": 27.96939468383789,
|
| 749 |
+
"learning_rate": 4.310252281266775e-06,
|
| 750 |
+
"loss": 1.3577,
|
| 751 |
+
"step": 1060
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.72,
|
| 755 |
+
"grad_norm": 29.119224548339844,
|
| 756 |
+
"learning_rate": 4.256575415995706e-06,
|
| 757 |
+
"loss": 1.3808,
|
| 758 |
+
"step": 1070
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.74,
|
| 762 |
+
"grad_norm": 30.36097526550293,
|
| 763 |
+
"learning_rate": 4.202898550724638e-06,
|
| 764 |
+
"loss": 1.3545,
|
| 765 |
+
"step": 1080
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.75,
|
| 769 |
+
"grad_norm": 30.843242645263672,
|
| 770 |
+
"learning_rate": 4.14922168545357e-06,
|
| 771 |
+
"loss": 1.3751,
|
| 772 |
+
"step": 1090
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.77,
|
| 776 |
+
"grad_norm": 29.29217529296875,
|
| 777 |
+
"learning_rate": 4.095544820182501e-06,
|
| 778 |
+
"loss": 1.3649,
|
| 779 |
+
"step": 1100
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.79,
|
| 783 |
+
"grad_norm": 30.685625076293945,
|
| 784 |
+
"learning_rate": 4.0418679549114335e-06,
|
| 785 |
+
"loss": 1.4354,
|
| 786 |
+
"step": 1110
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.8,
|
| 790 |
+
"grad_norm": 26.101669311523438,
|
| 791 |
+
"learning_rate": 3.988191089640365e-06,
|
| 792 |
+
"loss": 1.3355,
|
| 793 |
+
"step": 1120
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.82,
|
| 797 |
+
"grad_norm": 29.12729835510254,
|
| 798 |
+
"learning_rate": 3.934514224369297e-06,
|
| 799 |
+
"loss": 1.3568,
|
| 800 |
+
"step": 1130
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.83,
|
| 804 |
+
"grad_norm": 27.82271957397461,
|
| 805 |
+
"learning_rate": 3.880837359098229e-06,
|
| 806 |
+
"loss": 1.3702,
|
| 807 |
+
"step": 1140
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.85,
|
| 811 |
+
"grad_norm": 26.432327270507812,
|
| 812 |
+
"learning_rate": 3.827160493827161e-06,
|
| 813 |
+
"loss": 1.3231,
|
| 814 |
+
"step": 1150
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.87,
|
| 818 |
+
"grad_norm": 30.632972717285156,
|
| 819 |
+
"learning_rate": 3.7734836285560927e-06,
|
| 820 |
+
"loss": 1.3283,
|
| 821 |
+
"step": 1160
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.88,
|
| 825 |
+
"grad_norm": 27.142309188842773,
|
| 826 |
+
"learning_rate": 3.7198067632850245e-06,
|
| 827 |
+
"loss": 1.3159,
|
| 828 |
+
"step": 1170
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.9,
|
| 832 |
+
"grad_norm": 27.63045310974121,
|
| 833 |
+
"learning_rate": 3.6661298980139563e-06,
|
| 834 |
+
"loss": 1.3777,
|
| 835 |
+
"step": 1180
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.92,
|
| 839 |
+
"grad_norm": 30.256242752075195,
|
| 840 |
+
"learning_rate": 3.612453032742888e-06,
|
| 841 |
+
"loss": 1.2845,
|
| 842 |
+
"step": 1190
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.93,
|
| 846 |
+
"grad_norm": 28.592174530029297,
|
| 847 |
+
"learning_rate": 3.5587761674718204e-06,
|
| 848 |
+
"loss": 1.3163,
|
| 849 |
+
"step": 1200
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.95,
|
| 853 |
+
"grad_norm": 29.088247299194336,
|
| 854 |
+
"learning_rate": 3.505099302200752e-06,
|
| 855 |
+
"loss": 1.3145,
|
| 856 |
+
"step": 1210
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.96,
|
| 860 |
+
"grad_norm": 27.801074981689453,
|
| 861 |
+
"learning_rate": 3.4514224369296832e-06,
|
| 862 |
+
"loss": 1.3675,
|
| 863 |
+
"step": 1220
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.98,
|
| 867 |
+
"grad_norm": 28.81484603881836,
|
| 868 |
+
"learning_rate": 3.3977455716586155e-06,
|
| 869 |
+
"loss": 1.2854,
|
| 870 |
+
"step": 1230
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.0,
|
| 874 |
+
"grad_norm": 28.966217041015625,
|
| 875 |
+
"learning_rate": 3.3440687063875473e-06,
|
| 876 |
+
"loss": 1.3431,
|
| 877 |
+
"step": 1240
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.01,
|
| 881 |
+
"grad_norm": 23.021453857421875,
|
| 882 |
+
"learning_rate": 3.290391841116479e-06,
|
| 883 |
+
"loss": 1.0516,
|
| 884 |
+
"step": 1250
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.03,
|
| 888 |
+
"grad_norm": 25.622419357299805,
|
| 889 |
+
"learning_rate": 3.236714975845411e-06,
|
| 890 |
+
"loss": 0.9667,
|
| 891 |
+
"step": 1260
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.04,
|
| 895 |
+
"grad_norm": 26.45795249938965,
|
| 896 |
+
"learning_rate": 3.1830381105743428e-06,
|
| 897 |
+
"loss": 0.9341,
|
| 898 |
+
"step": 1270
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.06,
|
| 902 |
+
"grad_norm": 26.28618812561035,
|
| 903 |
+
"learning_rate": 3.1293612453032746e-06,
|
| 904 |
+
"loss": 0.9024,
|
| 905 |
+
"step": 1280
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.08,
|
| 909 |
+
"grad_norm": 24.80799102783203,
|
| 910 |
+
"learning_rate": 3.075684380032206e-06,
|
| 911 |
+
"loss": 0.8784,
|
| 912 |
+
"step": 1290
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 2.09,
|
| 916 |
+
"grad_norm": 25.70199966430664,
|
| 917 |
+
"learning_rate": 3.0220075147611383e-06,
|
| 918 |
+
"loss": 0.9307,
|
| 919 |
+
"step": 1300
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 2.11,
|
| 923 |
+
"grad_norm": 24.88735580444336,
|
| 924 |
+
"learning_rate": 2.9683306494900697e-06,
|
| 925 |
+
"loss": 0.9037,
|
| 926 |
+
"step": 1310
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 2.12,
|
| 930 |
+
"grad_norm": 26.51141929626465,
|
| 931 |
+
"learning_rate": 2.914653784219002e-06,
|
| 932 |
+
"loss": 0.9884,
|
| 933 |
+
"step": 1320
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 2.14,
|
| 937 |
+
"grad_norm": 25.662946701049805,
|
| 938 |
+
"learning_rate": 2.8609769189479338e-06,
|
| 939 |
+
"loss": 0.8631,
|
| 940 |
+
"step": 1330
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 2.16,
|
| 944 |
+
"grad_norm": 22.733741760253906,
|
| 945 |
+
"learning_rate": 2.8073000536768656e-06,
|
| 946 |
+
"loss": 0.9273,
|
| 947 |
+
"step": 1340
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 2.17,
|
| 951 |
+
"grad_norm": 24.159793853759766,
|
| 952 |
+
"learning_rate": 2.7536231884057974e-06,
|
| 953 |
+
"loss": 0.9262,
|
| 954 |
+
"step": 1350
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.19,
|
| 958 |
+
"grad_norm": 23.92421531677246,
|
| 959 |
+
"learning_rate": 2.699946323134729e-06,
|
| 960 |
+
"loss": 0.9048,
|
| 961 |
+
"step": 1360
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.2,
|
| 965 |
+
"grad_norm": 28.564496994018555,
|
| 966 |
+
"learning_rate": 2.646269457863661e-06,
|
| 967 |
+
"loss": 0.9879,
|
| 968 |
+
"step": 1370
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.22,
|
| 972 |
+
"grad_norm": 25.430883407592773,
|
| 973 |
+
"learning_rate": 2.5925925925925925e-06,
|
| 974 |
+
"loss": 0.892,
|
| 975 |
+
"step": 1380
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.24,
|
| 979 |
+
"grad_norm": 23.307687759399414,
|
| 980 |
+
"learning_rate": 2.5389157273215247e-06,
|
| 981 |
+
"loss": 0.894,
|
| 982 |
+
"step": 1390
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.25,
|
| 986 |
+
"grad_norm": 25.83247184753418,
|
| 987 |
+
"learning_rate": 2.4852388620504566e-06,
|
| 988 |
+
"loss": 0.8817,
|
| 989 |
+
"step": 1400
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.27,
|
| 993 |
+
"grad_norm": 25.72507095336914,
|
| 994 |
+
"learning_rate": 2.4315619967793884e-06,
|
| 995 |
+
"loss": 0.9155,
|
| 996 |
+
"step": 1410
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.29,
|
| 1000 |
+
"grad_norm": 26.67945098876953,
|
| 1001 |
+
"learning_rate": 2.3778851315083202e-06,
|
| 1002 |
+
"loss": 0.9325,
|
| 1003 |
+
"step": 1420
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.3,
|
| 1007 |
+
"grad_norm": 25.82522964477539,
|
| 1008 |
+
"learning_rate": 2.324208266237252e-06,
|
| 1009 |
+
"loss": 0.9055,
|
| 1010 |
+
"step": 1430
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.32,
|
| 1014 |
+
"grad_norm": 22.66315269470215,
|
| 1015 |
+
"learning_rate": 2.270531400966184e-06,
|
| 1016 |
+
"loss": 0.8631,
|
| 1017 |
+
"step": 1440
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.33,
|
| 1021 |
+
"grad_norm": 25.832313537597656,
|
| 1022 |
+
"learning_rate": 2.2168545356951157e-06,
|
| 1023 |
+
"loss": 0.9539,
|
| 1024 |
+
"step": 1450
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.35,
|
| 1028 |
+
"grad_norm": 23.865262985229492,
|
| 1029 |
+
"learning_rate": 2.163177670424047e-06,
|
| 1030 |
+
"loss": 0.8461,
|
| 1031 |
+
"step": 1460
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.37,
|
| 1035 |
+
"grad_norm": 23.32217025756836,
|
| 1036 |
+
"learning_rate": 2.109500805152979e-06,
|
| 1037 |
+
"loss": 0.8595,
|
| 1038 |
+
"step": 1470
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.38,
|
| 1042 |
+
"grad_norm": 24.299062728881836,
|
| 1043 |
+
"learning_rate": 2.0558239398819112e-06,
|
| 1044 |
+
"loss": 0.899,
|
| 1045 |
+
"step": 1480
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.4,
|
| 1049 |
+
"grad_norm": 25.582359313964844,
|
| 1050 |
+
"learning_rate": 2.002147074610843e-06,
|
| 1051 |
+
"loss": 0.898,
|
| 1052 |
+
"step": 1490
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.41,
|
| 1056 |
+
"grad_norm": 25.416006088256836,
|
| 1057 |
+
"learning_rate": 1.948470209339775e-06,
|
| 1058 |
+
"loss": 0.898,
|
| 1059 |
+
"step": 1500
|
| 1060 |
+
}
|
| 1061 |
+
],
|
| 1062 |
+
"logging_steps": 10,
|
| 1063 |
+
"max_steps": 1863,
|
| 1064 |
+
"num_input_tokens_seen": 0,
|
| 1065 |
+
"num_train_epochs": 3,
|
| 1066 |
+
"save_steps": 500,
|
| 1067 |
+
"total_flos": 5730542923874304.0,
|
| 1068 |
+
"train_batch_size": 4,
|
| 1069 |
+
"trial_name": null,
|
| 1070 |
+
"trial_params": null
|
| 1071 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4508efb6e0eb484b44f1b3f4335989489474ba1b0a09e069d078b9435fe807e
|
| 3 |
+
size 4475
|