Commit
·
d083258
1
Parent(s):
0162a38
Upload 9 files
Browse files- config.json +34 -0
- generation_config.json +7 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- spiece.model +3 -0
- spiece.vocab +0 -0
- trainer_state.json +1501 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "./mengzi-t5-base",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"T5ForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"classifier_dropout": 0.0,
|
| 7 |
+
"d_ff": 2048,
|
| 8 |
+
"d_kv": 64,
|
| 9 |
+
"d_model": 768,
|
| 10 |
+
"decoder_start_token_id": 0,
|
| 11 |
+
"dense_act_fn": "gelu_new",
|
| 12 |
+
"dropout_rate": 0.1,
|
| 13 |
+
"eos_token_id": 1,
|
| 14 |
+
"feed_forward_proj": "gated-gelu",
|
| 15 |
+
"gradient_checkpointing": false,
|
| 16 |
+
"initializer_factor": 1.0,
|
| 17 |
+
"is_encoder_decoder": true,
|
| 18 |
+
"is_gated_act": true,
|
| 19 |
+
"layer_norm_epsilon": 1e-06,
|
| 20 |
+
"max_length": 128,
|
| 21 |
+
"model_type": "t5",
|
| 22 |
+
"num_decoder_layers": 12,
|
| 23 |
+
"num_heads": 12,
|
| 24 |
+
"num_layers": 12,
|
| 25 |
+
"output_past": true,
|
| 26 |
+
"pad_token_id": 0,
|
| 27 |
+
"relative_attention_max_distance": 128,
|
| 28 |
+
"relative_attention_num_buckets": 32,
|
| 29 |
+
"tie_word_embeddings": false,
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.33.1",
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 32128
|
| 34 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"decoder_start_token_id": 0,
|
| 4 |
+
"eos_token_id": 1,
|
| 5 |
+
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "4.33.1"
|
| 7 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a21db8566d9dce481fab20d80e89ebea7991abc6c5e2c42ffc2d06c7fc91cb4
|
| 3 |
+
size 990408885
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:afc0d01bfbe6459e279c07eec2b4bf976ab12dd38eecbe5c6abb8b311789f9f9
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fe5c2ea4b8b03f48659e18306cbd2dd06e7e0d96f2f750c0a875b04dc0ac1412
|
| 3 |
+
size 627
|
spiece.model
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0e9093e83e8bc5acd23fade7f1369dd78494753d72d1bd1f4148fd318220cff7
|
| 3 |
+
size 659277
|
spiece.vocab
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1501 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.524030177821643,
|
| 5 |
+
"eval_steps": 2500,
|
| 6 |
+
"global_step": 97500,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03,
|
| 13 |
+
"learning_rate": 0.0005,
|
| 14 |
+
"loss": 0.5631,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.07,
|
| 19 |
+
"learning_rate": 0.0004974874371859296,
|
| 20 |
+
"loss": 0.14,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.1,
|
| 25 |
+
"learning_rate": 0.0004949748743718593,
|
| 26 |
+
"loss": 0.0804,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.13,
|
| 31 |
+
"learning_rate": 0.0004924623115577889,
|
| 32 |
+
"loss": 0.0597,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.17,
|
| 37 |
+
"learning_rate": 0.0004899497487437187,
|
| 38 |
+
"loss": 0.0501,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.17,
|
| 43 |
+
"eval_loss": 0.033905673772096634,
|
| 44 |
+
"eval_runtime": 8.0076,
|
| 45 |
+
"eval_samples_per_second": 124.881,
|
| 46 |
+
"eval_steps_per_second": 1.249,
|
| 47 |
+
"step": 2500
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 0.2,
|
| 51 |
+
"learning_rate": 0.00048743718592964827,
|
| 52 |
+
"loss": 0.0436,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.23,
|
| 57 |
+
"learning_rate": 0.0004849246231155779,
|
| 58 |
+
"loss": 0.0398,
|
| 59 |
+
"step": 3500
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.27,
|
| 63 |
+
"learning_rate": 0.00048241206030150754,
|
| 64 |
+
"loss": 0.0374,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.3,
|
| 69 |
+
"learning_rate": 0.0004798994974874372,
|
| 70 |
+
"loss": 0.035,
|
| 71 |
+
"step": 4500
|
| 72 |
+
},
|
| 73 |
+
{
|
| 74 |
+
"epoch": 0.33,
|
| 75 |
+
"learning_rate": 0.00047738693467336686,
|
| 76 |
+
"loss": 0.0331,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.33,
|
| 81 |
+
"eval_loss": 0.023899264633655548,
|
| 82 |
+
"eval_runtime": 7.9042,
|
| 83 |
+
"eval_samples_per_second": 126.515,
|
| 84 |
+
"eval_steps_per_second": 1.265,
|
| 85 |
+
"step": 5000
|
| 86 |
+
},
|
| 87 |
+
{
|
| 88 |
+
"epoch": 0.37,
|
| 89 |
+
"learning_rate": 0.00047487437185929647,
|
| 90 |
+
"loss": 0.0313,
|
| 91 |
+
"step": 5500
|
| 92 |
+
},
|
| 93 |
+
{
|
| 94 |
+
"epoch": 0.4,
|
| 95 |
+
"learning_rate": 0.00047236180904522613,
|
| 96 |
+
"loss": 0.03,
|
| 97 |
+
"step": 6000
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.43,
|
| 101 |
+
"learning_rate": 0.0004698492462311558,
|
| 102 |
+
"loss": 0.0287,
|
| 103 |
+
"step": 6500
|
| 104 |
+
},
|
| 105 |
+
{
|
| 106 |
+
"epoch": 0.47,
|
| 107 |
+
"learning_rate": 0.00046733668341708545,
|
| 108 |
+
"loss": 0.0277,
|
| 109 |
+
"step": 7000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5,
|
| 113 |
+
"learning_rate": 0.0004648241206030151,
|
| 114 |
+
"loss": 0.0271,
|
| 115 |
+
"step": 7500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.5,
|
| 119 |
+
"eval_loss": 0.01935673877596855,
|
| 120 |
+
"eval_runtime": 8.4605,
|
| 121 |
+
"eval_samples_per_second": 118.197,
|
| 122 |
+
"eval_steps_per_second": 1.182,
|
| 123 |
+
"step": 7500
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.54,
|
| 127 |
+
"learning_rate": 0.0004623115577889447,
|
| 128 |
+
"loss": 0.0262,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.57,
|
| 133 |
+
"learning_rate": 0.0004597989949748744,
|
| 134 |
+
"loss": 0.0253,
|
| 135 |
+
"step": 8500
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.6,
|
| 139 |
+
"learning_rate": 0.000457286432160804,
|
| 140 |
+
"loss": 0.0244,
|
| 141 |
+
"step": 9000
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 0.64,
|
| 145 |
+
"learning_rate": 0.0004547738693467337,
|
| 146 |
+
"loss": 0.0239,
|
| 147 |
+
"step": 9500
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 0.67,
|
| 151 |
+
"learning_rate": 0.0004522613065326633,
|
| 152 |
+
"loss": 0.0238,
|
| 153 |
+
"step": 10000
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.67,
|
| 157 |
+
"eval_loss": 0.017778538167476654,
|
| 158 |
+
"eval_runtime": 7.6198,
|
| 159 |
+
"eval_samples_per_second": 131.237,
|
| 160 |
+
"eval_steps_per_second": 1.312,
|
| 161 |
+
"step": 10000
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 0.7,
|
| 165 |
+
"learning_rate": 0.000449748743718593,
|
| 166 |
+
"loss": 0.0229,
|
| 167 |
+
"step": 10500
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 0.74,
|
| 171 |
+
"learning_rate": 0.0004472361809045226,
|
| 172 |
+
"loss": 0.0223,
|
| 173 |
+
"step": 11000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.77,
|
| 177 |
+
"learning_rate": 0.00044472361809045225,
|
| 178 |
+
"loss": 0.0219,
|
| 179 |
+
"step": 11500
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.8,
|
| 183 |
+
"learning_rate": 0.00044221105527638197,
|
| 184 |
+
"loss": 0.0216,
|
| 185 |
+
"step": 12000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.84,
|
| 189 |
+
"learning_rate": 0.0004396984924623116,
|
| 190 |
+
"loss": 0.0212,
|
| 191 |
+
"step": 12500
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.84,
|
| 195 |
+
"eval_loss": 0.014742564409971237,
|
| 196 |
+
"eval_runtime": 7.6159,
|
| 197 |
+
"eval_samples_per_second": 131.305,
|
| 198 |
+
"eval_steps_per_second": 1.313,
|
| 199 |
+
"step": 12500
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.87,
|
| 203 |
+
"learning_rate": 0.00043718592964824124,
|
| 204 |
+
"loss": 0.0206,
|
| 205 |
+
"step": 13000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.9,
|
| 209 |
+
"learning_rate": 0.00043467336683417085,
|
| 210 |
+
"loss": 0.0205,
|
| 211 |
+
"step": 13500
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 0.94,
|
| 215 |
+
"learning_rate": 0.0004321608040201005,
|
| 216 |
+
"loss": 0.0198,
|
| 217 |
+
"step": 14000
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 0.97,
|
| 221 |
+
"learning_rate": 0.0004296482412060301,
|
| 222 |
+
"loss": 0.0198,
|
| 223 |
+
"step": 14500
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 1.0,
|
| 227 |
+
"learning_rate": 0.00042713567839195983,
|
| 228 |
+
"loss": 0.0193,
|
| 229 |
+
"step": 15000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 1.0,
|
| 233 |
+
"eval_loss": 0.01463437918573618,
|
| 234 |
+
"eval_runtime": 7.6369,
|
| 235 |
+
"eval_samples_per_second": 130.944,
|
| 236 |
+
"eval_steps_per_second": 1.309,
|
| 237 |
+
"step": 15000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 1.04,
|
| 241 |
+
"learning_rate": 0.00042462311557788944,
|
| 242 |
+
"loss": 0.0178,
|
| 243 |
+
"step": 15500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 1.07,
|
| 247 |
+
"learning_rate": 0.0004221105527638191,
|
| 248 |
+
"loss": 0.0177,
|
| 249 |
+
"step": 16000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 1.1,
|
| 253 |
+
"learning_rate": 0.00041959798994974876,
|
| 254 |
+
"loss": 0.0172,
|
| 255 |
+
"step": 16500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.14,
|
| 259 |
+
"learning_rate": 0.00041708542713567837,
|
| 260 |
+
"loss": 0.017,
|
| 261 |
+
"step": 17000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 1.17,
|
| 265 |
+
"learning_rate": 0.0004145728643216081,
|
| 266 |
+
"loss": 0.0172,
|
| 267 |
+
"step": 17500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.17,
|
| 271 |
+
"eval_loss": 0.0131097212433815,
|
| 272 |
+
"eval_runtime": 7.63,
|
| 273 |
+
"eval_samples_per_second": 131.062,
|
| 274 |
+
"eval_steps_per_second": 1.311,
|
| 275 |
+
"step": 17500
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 1.2,
|
| 279 |
+
"learning_rate": 0.0004120603015075377,
|
| 280 |
+
"loss": 0.0168,
|
| 281 |
+
"step": 18000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.24,
|
| 285 |
+
"learning_rate": 0.00040954773869346736,
|
| 286 |
+
"loss": 0.017,
|
| 287 |
+
"step": 18500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 1.27,
|
| 291 |
+
"learning_rate": 0.00040703517587939697,
|
| 292 |
+
"loss": 0.0165,
|
| 293 |
+
"step": 19000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.3,
|
| 297 |
+
"learning_rate": 0.00040452261306532663,
|
| 298 |
+
"loss": 0.0163,
|
| 299 |
+
"step": 19500
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.34,
|
| 303 |
+
"learning_rate": 0.0004020100502512563,
|
| 304 |
+
"loss": 0.0163,
|
| 305 |
+
"step": 20000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 1.34,
|
| 309 |
+
"eval_loss": 0.012624073773622513,
|
| 310 |
+
"eval_runtime": 7.6613,
|
| 311 |
+
"eval_samples_per_second": 130.527,
|
| 312 |
+
"eval_steps_per_second": 1.305,
|
| 313 |
+
"step": 20000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 1.37,
|
| 317 |
+
"learning_rate": 0.00039949748743718595,
|
| 318 |
+
"loss": 0.0161,
|
| 319 |
+
"step": 20500
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 1.41,
|
| 323 |
+
"learning_rate": 0.00039698492462311556,
|
| 324 |
+
"loss": 0.0162,
|
| 325 |
+
"step": 21000
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 1.44,
|
| 329 |
+
"learning_rate": 0.0003944723618090452,
|
| 330 |
+
"loss": 0.0162,
|
| 331 |
+
"step": 21500
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.47,
|
| 335 |
+
"learning_rate": 0.0003919597989949749,
|
| 336 |
+
"loss": 0.0158,
|
| 337 |
+
"step": 22000
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 1.51,
|
| 341 |
+
"learning_rate": 0.00038944723618090455,
|
| 342 |
+
"loss": 0.0157,
|
| 343 |
+
"step": 22500
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.51,
|
| 347 |
+
"eval_loss": 0.0122187789529562,
|
| 348 |
+
"eval_runtime": 7.6253,
|
| 349 |
+
"eval_samples_per_second": 131.143,
|
| 350 |
+
"eval_steps_per_second": 1.311,
|
| 351 |
+
"step": 22500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 1.54,
|
| 355 |
+
"learning_rate": 0.0003869346733668342,
|
| 356 |
+
"loss": 0.0289,
|
| 357 |
+
"step": 23000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.57,
|
| 361 |
+
"learning_rate": 0.0003844221105527638,
|
| 362 |
+
"loss": 0.0157,
|
| 363 |
+
"step": 23500
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 1.61,
|
| 367 |
+
"learning_rate": 0.0003819095477386935,
|
| 368 |
+
"loss": 0.0148,
|
| 369 |
+
"step": 24000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 1.64,
|
| 373 |
+
"learning_rate": 0.0003793969849246231,
|
| 374 |
+
"loss": 0.0148,
|
| 375 |
+
"step": 24500
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 1.67,
|
| 379 |
+
"learning_rate": 0.0003768844221105528,
|
| 380 |
+
"loss": 0.0147,
|
| 381 |
+
"step": 25000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.67,
|
| 385 |
+
"eval_loss": 0.011391976848244667,
|
| 386 |
+
"eval_runtime": 7.5953,
|
| 387 |
+
"eval_samples_per_second": 131.66,
|
| 388 |
+
"eval_steps_per_second": 1.317,
|
| 389 |
+
"step": 25000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.71,
|
| 393 |
+
"learning_rate": 0.0003743718592964824,
|
| 394 |
+
"loss": 0.0148,
|
| 395 |
+
"step": 25500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 1.74,
|
| 399 |
+
"learning_rate": 0.00037185929648241207,
|
| 400 |
+
"loss": 0.0146,
|
| 401 |
+
"step": 26000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 1.77,
|
| 405 |
+
"learning_rate": 0.0003693467336683417,
|
| 406 |
+
"loss": 0.0145,
|
| 407 |
+
"step": 26500
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.81,
|
| 411 |
+
"learning_rate": 0.00036683417085427134,
|
| 412 |
+
"loss": 0.0143,
|
| 413 |
+
"step": 27000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.84,
|
| 417 |
+
"learning_rate": 0.00036432160804020106,
|
| 418 |
+
"loss": 0.0144,
|
| 419 |
+
"step": 27500
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 1.84,
|
| 423 |
+
"eval_loss": 0.011031309142708778,
|
| 424 |
+
"eval_runtime": 7.6444,
|
| 425 |
+
"eval_samples_per_second": 130.814,
|
| 426 |
+
"eval_steps_per_second": 1.308,
|
| 427 |
+
"step": 27500
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.87,
|
| 431 |
+
"learning_rate": 0.00036180904522613067,
|
| 432 |
+
"loss": 0.0142,
|
| 433 |
+
"step": 28000
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 1.91,
|
| 437 |
+
"learning_rate": 0.00035929648241206033,
|
| 438 |
+
"loss": 0.014,
|
| 439 |
+
"step": 28500
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 1.94,
|
| 443 |
+
"learning_rate": 0.00035678391959798994,
|
| 444 |
+
"loss": 0.0143,
|
| 445 |
+
"step": 29000
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 1.97,
|
| 449 |
+
"learning_rate": 0.0003542713567839196,
|
| 450 |
+
"loss": 0.0144,
|
| 451 |
+
"step": 29500
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 2.01,
|
| 455 |
+
"learning_rate": 0.00035175879396984926,
|
| 456 |
+
"loss": 0.0136,
|
| 457 |
+
"step": 30000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 2.01,
|
| 461 |
+
"eval_loss": 0.010939952917397022,
|
| 462 |
+
"eval_runtime": 7.6117,
|
| 463 |
+
"eval_samples_per_second": 131.376,
|
| 464 |
+
"eval_steps_per_second": 1.314,
|
| 465 |
+
"step": 30000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 2.04,
|
| 469 |
+
"learning_rate": 0.0003492462311557789,
|
| 470 |
+
"loss": 0.0121,
|
| 471 |
+
"step": 30500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 2.07,
|
| 475 |
+
"learning_rate": 0.00034673366834170853,
|
| 476 |
+
"loss": 0.0121,
|
| 477 |
+
"step": 31000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 2.11,
|
| 481 |
+
"learning_rate": 0.0003442211055276382,
|
| 482 |
+
"loss": 0.0122,
|
| 483 |
+
"step": 31500
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 2.14,
|
| 487 |
+
"learning_rate": 0.0003417085427135678,
|
| 488 |
+
"loss": 0.012,
|
| 489 |
+
"step": 32000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.17,
|
| 493 |
+
"learning_rate": 0.0003391959798994975,
|
| 494 |
+
"loss": 0.0121,
|
| 495 |
+
"step": 32500
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 2.17,
|
| 499 |
+
"eval_loss": 0.010623764246702194,
|
| 500 |
+
"eval_runtime": 7.623,
|
| 501 |
+
"eval_samples_per_second": 131.182,
|
| 502 |
+
"eval_steps_per_second": 1.312,
|
| 503 |
+
"step": 32500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.21,
|
| 507 |
+
"learning_rate": 0.0003366834170854272,
|
| 508 |
+
"loss": 0.012,
|
| 509 |
+
"step": 33000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 2.24,
|
| 513 |
+
"learning_rate": 0.0003341708542713568,
|
| 514 |
+
"loss": 0.0119,
|
| 515 |
+
"step": 33500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 2.28,
|
| 519 |
+
"learning_rate": 0.00033165829145728645,
|
| 520 |
+
"loss": 0.0119,
|
| 521 |
+
"step": 34000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 2.31,
|
| 525 |
+
"learning_rate": 0.00032914572864321606,
|
| 526 |
+
"loss": 0.0118,
|
| 527 |
+
"step": 34500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 2.34,
|
| 531 |
+
"learning_rate": 0.00032663316582914577,
|
| 532 |
+
"loss": 0.0118,
|
| 533 |
+
"step": 35000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.34,
|
| 537 |
+
"eval_loss": 0.01009445358067751,
|
| 538 |
+
"eval_runtime": 7.8168,
|
| 539 |
+
"eval_samples_per_second": 127.93,
|
| 540 |
+
"eval_steps_per_second": 1.279,
|
| 541 |
+
"step": 35000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 2.38,
|
| 545 |
+
"learning_rate": 0.0003241206030150754,
|
| 546 |
+
"loss": 0.0119,
|
| 547 |
+
"step": 35500
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 2.41,
|
| 551 |
+
"learning_rate": 0.00032160804020100504,
|
| 552 |
+
"loss": 0.0119,
|
| 553 |
+
"step": 36000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.44,
|
| 557 |
+
"learning_rate": 0.00031909547738693465,
|
| 558 |
+
"loss": 0.0117,
|
| 559 |
+
"step": 36500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 2.48,
|
| 563 |
+
"learning_rate": 0.0003165829145728643,
|
| 564 |
+
"loss": 0.0117,
|
| 565 |
+
"step": 37000
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 2.51,
|
| 569 |
+
"learning_rate": 0.00031407035175879403,
|
| 570 |
+
"loss": 0.0115,
|
| 571 |
+
"step": 37500
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 2.51,
|
| 575 |
+
"eval_loss": 0.010385086759924889,
|
| 576 |
+
"eval_runtime": 8.0071,
|
| 577 |
+
"eval_samples_per_second": 124.889,
|
| 578 |
+
"eval_steps_per_second": 1.249,
|
| 579 |
+
"step": 37500
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 2.54,
|
| 583 |
+
"learning_rate": 0.00031155778894472364,
|
| 584 |
+
"loss": 0.0116,
|
| 585 |
+
"step": 38000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 2.58,
|
| 589 |
+
"learning_rate": 0.0003090452261306533,
|
| 590 |
+
"loss": 0.0115,
|
| 591 |
+
"step": 38500
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 2.61,
|
| 595 |
+
"learning_rate": 0.0003065326633165829,
|
| 596 |
+
"loss": 0.0115,
|
| 597 |
+
"step": 39000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 2.64,
|
| 601 |
+
"learning_rate": 0.00030402010050251257,
|
| 602 |
+
"loss": 0.0117,
|
| 603 |
+
"step": 39500
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 2.68,
|
| 607 |
+
"learning_rate": 0.0003015075376884422,
|
| 608 |
+
"loss": 0.0116,
|
| 609 |
+
"step": 40000
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 2.68,
|
| 613 |
+
"eval_loss": 0.010146565735340118,
|
| 614 |
+
"eval_runtime": 7.8031,
|
| 615 |
+
"eval_samples_per_second": 128.154,
|
| 616 |
+
"eval_steps_per_second": 1.282,
|
| 617 |
+
"step": 40000
|
| 618 |
+
},
|
| 619 |
+
{
|
| 620 |
+
"epoch": 2.71,
|
| 621 |
+
"learning_rate": 0.0002989949748743719,
|
| 622 |
+
"loss": 0.0115,
|
| 623 |
+
"step": 40500
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 2.74,
|
| 627 |
+
"learning_rate": 0.0002964824120603015,
|
| 628 |
+
"loss": 0.0113,
|
| 629 |
+
"step": 41000
|
| 630 |
+
},
|
| 631 |
+
{
|
| 632 |
+
"epoch": 2.78,
|
| 633 |
+
"learning_rate": 0.00029396984924623116,
|
| 634 |
+
"loss": 0.0112,
|
| 635 |
+
"step": 41500
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 2.81,
|
| 639 |
+
"learning_rate": 0.00029145728643216077,
|
| 640 |
+
"loss": 0.0113,
|
| 641 |
+
"step": 42000
|
| 642 |
+
},
|
| 643 |
+
{
|
| 644 |
+
"epoch": 2.84,
|
| 645 |
+
"learning_rate": 0.00028894472361809043,
|
| 646 |
+
"loss": 0.0113,
|
| 647 |
+
"step": 42500
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 2.84,
|
| 651 |
+
"eval_loss": 0.009525712579488754,
|
| 652 |
+
"eval_runtime": 7.6057,
|
| 653 |
+
"eval_samples_per_second": 131.48,
|
| 654 |
+
"eval_steps_per_second": 1.315,
|
| 655 |
+
"step": 42500
|
| 656 |
+
},
|
| 657 |
+
{
|
| 658 |
+
"epoch": 2.88,
|
| 659 |
+
"learning_rate": 0.00028643216080402015,
|
| 660 |
+
"loss": 0.0111,
|
| 661 |
+
"step": 43000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 2.91,
|
| 665 |
+
"learning_rate": 0.00028391959798994976,
|
| 666 |
+
"loss": 0.0111,
|
| 667 |
+
"step": 43500
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 2.94,
|
| 671 |
+
"learning_rate": 0.0002814070351758794,
|
| 672 |
+
"loss": 0.011,
|
| 673 |
+
"step": 44000
|
| 674 |
+
},
|
| 675 |
+
{
|
| 676 |
+
"epoch": 2.98,
|
| 677 |
+
"learning_rate": 0.000278894472361809,
|
| 678 |
+
"loss": 0.0109,
|
| 679 |
+
"step": 44500
|
| 680 |
+
},
|
| 681 |
+
{
|
| 682 |
+
"epoch": 3.01,
|
| 683 |
+
"learning_rate": 0.0002763819095477387,
|
| 684 |
+
"loss": 0.0106,
|
| 685 |
+
"step": 45000
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 3.01,
|
| 689 |
+
"eval_loss": 0.009837072342634201,
|
| 690 |
+
"eval_runtime": 7.6063,
|
| 691 |
+
"eval_samples_per_second": 131.469,
|
| 692 |
+
"eval_steps_per_second": 1.315,
|
| 693 |
+
"step": 45000
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 3.04,
|
| 697 |
+
"learning_rate": 0.00027386934673366835,
|
| 698 |
+
"loss": 0.0091,
|
| 699 |
+
"step": 45500
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 3.08,
|
| 703 |
+
"learning_rate": 0.000271356783919598,
|
| 704 |
+
"loss": 0.009,
|
| 705 |
+
"step": 46000
|
| 706 |
+
},
|
| 707 |
+
{
|
| 708 |
+
"epoch": 3.11,
|
| 709 |
+
"learning_rate": 0.0002688442211055276,
|
| 710 |
+
"loss": 0.0088,
|
| 711 |
+
"step": 46500
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.14,
|
| 715 |
+
"learning_rate": 0.0002663316582914573,
|
| 716 |
+
"loss": 0.0091,
|
| 717 |
+
"step": 47000
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 3.18,
|
| 721 |
+
"learning_rate": 0.0002638190954773869,
|
| 722 |
+
"loss": 0.0092,
|
| 723 |
+
"step": 47500
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 3.18,
|
| 727 |
+
"eval_loss": 0.009356345981359482,
|
| 728 |
+
"eval_runtime": 7.6201,
|
| 729 |
+
"eval_samples_per_second": 131.232,
|
| 730 |
+
"eval_steps_per_second": 1.312,
|
| 731 |
+
"step": 47500
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 3.21,
|
| 735 |
+
"learning_rate": 0.0002613065326633166,
|
| 736 |
+
"loss": 0.0092,
|
| 737 |
+
"step": 48000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 3.25,
|
| 741 |
+
"learning_rate": 0.00025879396984924627,
|
| 742 |
+
"loss": 0.009,
|
| 743 |
+
"step": 48500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 3.28,
|
| 747 |
+
"learning_rate": 0.0002562814070351759,
|
| 748 |
+
"loss": 0.0089,
|
| 749 |
+
"step": 49000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 3.31,
|
| 753 |
+
"learning_rate": 0.00025376884422110554,
|
| 754 |
+
"loss": 0.0092,
|
| 755 |
+
"step": 49500
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 3.35,
|
| 759 |
+
"learning_rate": 0.00025125628140703515,
|
| 760 |
+
"loss": 0.0091,
|
| 761 |
+
"step": 50000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.35,
|
| 765 |
+
"eval_loss": 0.009372641332447529,
|
| 766 |
+
"eval_runtime": 7.5969,
|
| 767 |
+
"eval_samples_per_second": 131.633,
|
| 768 |
+
"eval_steps_per_second": 1.316,
|
| 769 |
+
"step": 50000
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 3.38,
|
| 773 |
+
"learning_rate": 0.0002487437185929648,
|
| 774 |
+
"loss": 0.0091,
|
| 775 |
+
"step": 50500
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 3.41,
|
| 779 |
+
"learning_rate": 0.00024623115577889447,
|
| 780 |
+
"loss": 0.009,
|
| 781 |
+
"step": 51000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 3.45,
|
| 785 |
+
"learning_rate": 0.00024371859296482413,
|
| 786 |
+
"loss": 0.0093,
|
| 787 |
+
"step": 51500
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 3.48,
|
| 791 |
+
"learning_rate": 0.00024120603015075377,
|
| 792 |
+
"loss": 0.0092,
|
| 793 |
+
"step": 52000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 3.51,
|
| 797 |
+
"learning_rate": 0.00023869346733668343,
|
| 798 |
+
"loss": 0.009,
|
| 799 |
+
"step": 52500
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 3.51,
|
| 803 |
+
"eval_loss": 0.009138910099864006,
|
| 804 |
+
"eval_runtime": 7.5893,
|
| 805 |
+
"eval_samples_per_second": 131.765,
|
| 806 |
+
"eval_steps_per_second": 1.318,
|
| 807 |
+
"step": 52500
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 3.55,
|
| 811 |
+
"learning_rate": 0.00023618090452261307,
|
| 812 |
+
"loss": 0.0091,
|
| 813 |
+
"step": 53000
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 3.58,
|
| 817 |
+
"learning_rate": 0.00023366834170854273,
|
| 818 |
+
"loss": 0.0089,
|
| 819 |
+
"step": 53500
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 3.61,
|
| 823 |
+
"learning_rate": 0.00023115577889447236,
|
| 824 |
+
"loss": 0.009,
|
| 825 |
+
"step": 54000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 3.65,
|
| 829 |
+
"learning_rate": 0.000228643216080402,
|
| 830 |
+
"loss": 0.0092,
|
| 831 |
+
"step": 54500
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 3.68,
|
| 835 |
+
"learning_rate": 0.00022613065326633166,
|
| 836 |
+
"loss": 0.0091,
|
| 837 |
+
"step": 55000
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 3.68,
|
| 841 |
+
"eval_loss": 0.00910147000104189,
|
| 842 |
+
"eval_runtime": 7.6234,
|
| 843 |
+
"eval_samples_per_second": 131.175,
|
| 844 |
+
"eval_steps_per_second": 1.312,
|
| 845 |
+
"step": 55000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 3.71,
|
| 849 |
+
"learning_rate": 0.0002236180904522613,
|
| 850 |
+
"loss": 0.0089,
|
| 851 |
+
"step": 55500
|
| 852 |
+
},
|
| 853 |
+
{
|
| 854 |
+
"epoch": 3.75,
|
| 855 |
+
"learning_rate": 0.00022110552763819098,
|
| 856 |
+
"loss": 0.0089,
|
| 857 |
+
"step": 56000
|
| 858 |
+
},
|
| 859 |
+
{
|
| 860 |
+
"epoch": 3.78,
|
| 861 |
+
"learning_rate": 0.00021859296482412062,
|
| 862 |
+
"loss": 0.009,
|
| 863 |
+
"step": 56500
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 3.81,
|
| 867 |
+
"learning_rate": 0.00021608040201005025,
|
| 868 |
+
"loss": 0.0091,
|
| 869 |
+
"step": 57000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 3.85,
|
| 873 |
+
"learning_rate": 0.00021356783919597992,
|
| 874 |
+
"loss": 0.0089,
|
| 875 |
+
"step": 57500
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 3.85,
|
| 879 |
+
"eval_loss": 0.009140917100012302,
|
| 880 |
+
"eval_runtime": 7.5801,
|
| 881 |
+
"eval_samples_per_second": 131.925,
|
| 882 |
+
"eval_steps_per_second": 1.319,
|
| 883 |
+
"step": 57500
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 3.88,
|
| 887 |
+
"learning_rate": 0.00021105527638190955,
|
| 888 |
+
"loss": 0.0089,
|
| 889 |
+
"step": 58000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 3.91,
|
| 893 |
+
"learning_rate": 0.00020854271356783919,
|
| 894 |
+
"loss": 0.0088,
|
| 895 |
+
"step": 58500
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 3.95,
|
| 899 |
+
"learning_rate": 0.00020603015075376885,
|
| 900 |
+
"loss": 0.0089,
|
| 901 |
+
"step": 59000
|
| 902 |
+
},
|
| 903 |
+
{
|
| 904 |
+
"epoch": 3.98,
|
| 905 |
+
"learning_rate": 0.00020351758793969848,
|
| 906 |
+
"loss": 0.0088,
|
| 907 |
+
"step": 59500
|
| 908 |
+
},
|
| 909 |
+
{
|
| 910 |
+
"epoch": 4.01,
|
| 911 |
+
"learning_rate": 0.00020100502512562814,
|
| 912 |
+
"loss": 0.0081,
|
| 913 |
+
"step": 60000
|
| 914 |
+
},
|
| 915 |
+
{
|
| 916 |
+
"epoch": 4.01,
|
| 917 |
+
"eval_loss": 0.00933513417840004,
|
| 918 |
+
"eval_runtime": 7.577,
|
| 919 |
+
"eval_samples_per_second": 131.978,
|
| 920 |
+
"eval_steps_per_second": 1.32,
|
| 921 |
+
"step": 60000
|
| 922 |
+
},
|
| 923 |
+
{
|
| 924 |
+
"epoch": 4.05,
|
| 925 |
+
"learning_rate": 0.00019849246231155778,
|
| 926 |
+
"loss": 0.0068,
|
| 927 |
+
"step": 60500
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 4.08,
|
| 931 |
+
"learning_rate": 0.00019597989949748744,
|
| 932 |
+
"loss": 0.0068,
|
| 933 |
+
"step": 61000
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 4.12,
|
| 937 |
+
"learning_rate": 0.0001934673366834171,
|
| 938 |
+
"loss": 0.007,
|
| 939 |
+
"step": 61500
|
| 940 |
+
},
|
| 941 |
+
{
|
| 942 |
+
"epoch": 4.15,
|
| 943 |
+
"learning_rate": 0.00019095477386934674,
|
| 944 |
+
"loss": 0.0068,
|
| 945 |
+
"step": 62000
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 4.18,
|
| 949 |
+
"learning_rate": 0.0001884422110552764,
|
| 950 |
+
"loss": 0.0069,
|
| 951 |
+
"step": 62500
|
| 952 |
+
},
|
| 953 |
+
{
|
| 954 |
+
"epoch": 4.18,
|
| 955 |
+
"eval_loss": 0.009106965735554695,
|
| 956 |
+
"eval_runtime": 7.6103,
|
| 957 |
+
"eval_samples_per_second": 131.401,
|
| 958 |
+
"eval_steps_per_second": 1.314,
|
| 959 |
+
"step": 62500
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 4.22,
|
| 963 |
+
"learning_rate": 0.00018592964824120604,
|
| 964 |
+
"loss": 0.0068,
|
| 965 |
+
"step": 63000
|
| 966 |
+
},
|
| 967 |
+
{
|
| 968 |
+
"epoch": 4.25,
|
| 969 |
+
"learning_rate": 0.00018341708542713567,
|
| 970 |
+
"loss": 0.007,
|
| 971 |
+
"step": 63500
|
| 972 |
+
},
|
| 973 |
+
{
|
| 974 |
+
"epoch": 4.28,
|
| 975 |
+
"learning_rate": 0.00018090452261306533,
|
| 976 |
+
"loss": 0.007,
|
| 977 |
+
"step": 64000
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 4.32,
|
| 981 |
+
"learning_rate": 0.00017839195979899497,
|
| 982 |
+
"loss": 0.0069,
|
| 983 |
+
"step": 64500
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 4.35,
|
| 987 |
+
"learning_rate": 0.00017587939698492463,
|
| 988 |
+
"loss": 0.0069,
|
| 989 |
+
"step": 65000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 4.35,
|
| 993 |
+
"eval_loss": 0.009035669267177582,
|
| 994 |
+
"eval_runtime": 7.7828,
|
| 995 |
+
"eval_samples_per_second": 128.488,
|
| 996 |
+
"eval_steps_per_second": 1.285,
|
| 997 |
+
"step": 65000
|
| 998 |
+
},
|
| 999 |
+
{
|
| 1000 |
+
"epoch": 4.38,
|
| 1001 |
+
"learning_rate": 0.00017336683417085427,
|
| 1002 |
+
"loss": 0.007,
|
| 1003 |
+
"step": 65500
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 4.42,
|
| 1007 |
+
"learning_rate": 0.0001708542713567839,
|
| 1008 |
+
"loss": 0.0069,
|
| 1009 |
+
"step": 66000
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 4.45,
|
| 1013 |
+
"learning_rate": 0.0001683417085427136,
|
| 1014 |
+
"loss": 0.0069,
|
| 1015 |
+
"step": 66500
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 4.48,
|
| 1019 |
+
"learning_rate": 0.00016582914572864322,
|
| 1020 |
+
"loss": 0.0068,
|
| 1021 |
+
"step": 67000
|
| 1022 |
+
},
|
| 1023 |
+
{
|
| 1024 |
+
"epoch": 4.52,
|
| 1025 |
+
"learning_rate": 0.00016331658291457289,
|
| 1026 |
+
"loss": 0.0071,
|
| 1027 |
+
"step": 67500
|
| 1028 |
+
},
|
| 1029 |
+
{
|
| 1030 |
+
"epoch": 4.52,
|
| 1031 |
+
"eval_loss": 0.009078802540898323,
|
| 1032 |
+
"eval_runtime": 7.5849,
|
| 1033 |
+
"eval_samples_per_second": 131.841,
|
| 1034 |
+
"eval_steps_per_second": 1.318,
|
| 1035 |
+
"step": 67500
|
| 1036 |
+
},
|
| 1037 |
+
{
|
| 1038 |
+
"epoch": 4.55,
|
| 1039 |
+
"learning_rate": 0.00016080402010050252,
|
| 1040 |
+
"loss": 0.007,
|
| 1041 |
+
"step": 68000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 4.58,
|
| 1045 |
+
"learning_rate": 0.00015829145728643216,
|
| 1046 |
+
"loss": 0.0071,
|
| 1047 |
+
"step": 68500
|
| 1048 |
+
},
|
| 1049 |
+
{
|
| 1050 |
+
"epoch": 4.62,
|
| 1051 |
+
"learning_rate": 0.00015577889447236182,
|
| 1052 |
+
"loss": 0.0071,
|
| 1053 |
+
"step": 69000
|
| 1054 |
+
},
|
| 1055 |
+
{
|
| 1056 |
+
"epoch": 4.65,
|
| 1057 |
+
"learning_rate": 0.00015326633165829145,
|
| 1058 |
+
"loss": 0.0072,
|
| 1059 |
+
"step": 69500
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 4.68,
|
| 1063 |
+
"learning_rate": 0.0001507537688442211,
|
| 1064 |
+
"loss": 0.0069,
|
| 1065 |
+
"step": 70000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 4.68,
|
| 1069 |
+
"eval_loss": 0.008746389299631119,
|
| 1070 |
+
"eval_runtime": 7.5714,
|
| 1071 |
+
"eval_samples_per_second": 132.075,
|
| 1072 |
+
"eval_steps_per_second": 1.321,
|
| 1073 |
+
"step": 70000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 4.72,
|
| 1077 |
+
"learning_rate": 0.00014824120603015075,
|
| 1078 |
+
"loss": 0.0071,
|
| 1079 |
+
"step": 70500
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 4.75,
|
| 1083 |
+
"learning_rate": 0.00014572864321608039,
|
| 1084 |
+
"loss": 0.007,
|
| 1085 |
+
"step": 71000
|
| 1086 |
+
},
|
| 1087 |
+
{
|
| 1088 |
+
"epoch": 4.78,
|
| 1089 |
+
"learning_rate": 0.00014321608040201007,
|
| 1090 |
+
"loss": 0.007,
|
| 1091 |
+
"step": 71500
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 4.82,
|
| 1095 |
+
"learning_rate": 0.0001407035175879397,
|
| 1096 |
+
"loss": 0.007,
|
| 1097 |
+
"step": 72000
|
| 1098 |
+
},
|
| 1099 |
+
{
|
| 1100 |
+
"epoch": 4.85,
|
| 1101 |
+
"learning_rate": 0.00013819095477386934,
|
| 1102 |
+
"loss": 0.0069,
|
| 1103 |
+
"step": 72500
|
| 1104 |
+
},
|
| 1105 |
+
{
|
| 1106 |
+
"epoch": 4.85,
|
| 1107 |
+
"eval_loss": 0.008641643449664116,
|
| 1108 |
+
"eval_runtime": 7.59,
|
| 1109 |
+
"eval_samples_per_second": 131.752,
|
| 1110 |
+
"eval_steps_per_second": 1.318,
|
| 1111 |
+
"step": 72500
|
| 1112 |
+
},
|
| 1113 |
+
{
|
| 1114 |
+
"epoch": 4.88,
|
| 1115 |
+
"learning_rate": 0.000135678391959799,
|
| 1116 |
+
"loss": 0.007,
|
| 1117 |
+
"step": 73000
|
| 1118 |
+
},
|
| 1119 |
+
{
|
| 1120 |
+
"epoch": 4.92,
|
| 1121 |
+
"learning_rate": 0.00013316582914572864,
|
| 1122 |
+
"loss": 0.007,
|
| 1123 |
+
"step": 73500
|
| 1124 |
+
},
|
| 1125 |
+
{
|
| 1126 |
+
"epoch": 4.95,
|
| 1127 |
+
"learning_rate": 0.0001306532663316583,
|
| 1128 |
+
"loss": 0.0069,
|
| 1129 |
+
"step": 74000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 4.99,
|
| 1133 |
+
"learning_rate": 0.00012814070351758794,
|
| 1134 |
+
"loss": 0.0071,
|
| 1135 |
+
"step": 74500
|
| 1136 |
+
},
|
| 1137 |
+
{
|
| 1138 |
+
"epoch": 5.02,
|
| 1139 |
+
"learning_rate": 0.00012562814070351757,
|
| 1140 |
+
"loss": 0.0058,
|
| 1141 |
+
"step": 75000
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 5.02,
|
| 1145 |
+
"eval_loss": 0.00928833894431591,
|
| 1146 |
+
"eval_runtime": 7.8485,
|
| 1147 |
+
"eval_samples_per_second": 127.413,
|
| 1148 |
+
"eval_steps_per_second": 1.274,
|
| 1149 |
+
"step": 75000
|
| 1150 |
+
},
|
| 1151 |
+
{
|
| 1152 |
+
"epoch": 5.05,
|
| 1153 |
+
"learning_rate": 0.00012311557788944724,
|
| 1154 |
+
"loss": 0.005,
|
| 1155 |
+
"step": 75500
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 5.09,
|
| 1159 |
+
"learning_rate": 0.00012060301507537688,
|
| 1160 |
+
"loss": 0.0051,
|
| 1161 |
+
"step": 76000
|
| 1162 |
+
},
|
| 1163 |
+
{
|
| 1164 |
+
"epoch": 5.12,
|
| 1165 |
+
"learning_rate": 0.00011809045226130653,
|
| 1166 |
+
"loss": 0.0053,
|
| 1167 |
+
"step": 76500
|
| 1168 |
+
},
|
| 1169 |
+
{
|
| 1170 |
+
"epoch": 5.15,
|
| 1171 |
+
"learning_rate": 0.00011557788944723618,
|
| 1172 |
+
"loss": 0.0052,
|
| 1173 |
+
"step": 77000
|
| 1174 |
+
},
|
| 1175 |
+
{
|
| 1176 |
+
"epoch": 5.19,
|
| 1177 |
+
"learning_rate": 0.00011306532663316583,
|
| 1178 |
+
"loss": 0.0053,
|
| 1179 |
+
"step": 77500
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 5.19,
|
| 1183 |
+
"eval_loss": 0.009319030679762363,
|
| 1184 |
+
"eval_runtime": 7.5874,
|
| 1185 |
+
"eval_samples_per_second": 131.797,
|
| 1186 |
+
"eval_steps_per_second": 1.318,
|
| 1187 |
+
"step": 77500
|
| 1188 |
+
},
|
| 1189 |
+
{
|
| 1190 |
+
"epoch": 5.22,
|
| 1191 |
+
"learning_rate": 0.00011055276381909549,
|
| 1192 |
+
"loss": 0.0053,
|
| 1193 |
+
"step": 78000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 5.25,
|
| 1197 |
+
"learning_rate": 0.00010804020100502513,
|
| 1198 |
+
"loss": 0.0054,
|
| 1199 |
+
"step": 78500
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 5.29,
|
| 1203 |
+
"learning_rate": 0.00010552763819095478,
|
| 1204 |
+
"loss": 0.0052,
|
| 1205 |
+
"step": 79000
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 5.32,
|
| 1209 |
+
"learning_rate": 0.00010301507537688442,
|
| 1210 |
+
"loss": 0.0053,
|
| 1211 |
+
"step": 79500
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 5.35,
|
| 1215 |
+
"learning_rate": 0.00010050251256281407,
|
| 1216 |
+
"loss": 0.0052,
|
| 1217 |
+
"step": 80000
|
| 1218 |
+
},
|
| 1219 |
+
{
|
| 1220 |
+
"epoch": 5.35,
|
| 1221 |
+
"eval_loss": 0.008985009975731373,
|
| 1222 |
+
"eval_runtime": 7.8274,
|
| 1223 |
+
"eval_samples_per_second": 127.757,
|
| 1224 |
+
"eval_steps_per_second": 1.278,
|
| 1225 |
+
"step": 80000
|
| 1226 |
+
},
|
| 1227 |
+
{
|
| 1228 |
+
"epoch": 5.39,
|
| 1229 |
+
"learning_rate": 9.798994974874372e-05,
|
| 1230 |
+
"loss": 0.0053,
|
| 1231 |
+
"step": 80500
|
| 1232 |
+
},
|
| 1233 |
+
{
|
| 1234 |
+
"epoch": 5.42,
|
| 1235 |
+
"learning_rate": 9.547738693467337e-05,
|
| 1236 |
+
"loss": 0.0053,
|
| 1237 |
+
"step": 81000
|
| 1238 |
+
},
|
| 1239 |
+
{
|
| 1240 |
+
"epoch": 5.45,
|
| 1241 |
+
"learning_rate": 9.296482412060302e-05,
|
| 1242 |
+
"loss": 0.0052,
|
| 1243 |
+
"step": 81500
|
| 1244 |
+
},
|
| 1245 |
+
{
|
| 1246 |
+
"epoch": 5.49,
|
| 1247 |
+
"learning_rate": 9.045226130653267e-05,
|
| 1248 |
+
"loss": 0.0053,
|
| 1249 |
+
"step": 82000
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 5.52,
|
| 1253 |
+
"learning_rate": 8.793969849246232e-05,
|
| 1254 |
+
"loss": 0.0052,
|
| 1255 |
+
"step": 82500
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 5.52,
|
| 1259 |
+
"eval_loss": 0.008979461155831814,
|
| 1260 |
+
"eval_runtime": 7.6451,
|
| 1261 |
+
"eval_samples_per_second": 130.803,
|
| 1262 |
+
"eval_steps_per_second": 1.308,
|
| 1263 |
+
"step": 82500
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 5.55,
|
| 1267 |
+
"learning_rate": 8.542713567839195e-05,
|
| 1268 |
+
"loss": 0.0051,
|
| 1269 |
+
"step": 83000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 5.59,
|
| 1273 |
+
"learning_rate": 8.291457286432161e-05,
|
| 1274 |
+
"loss": 0.0051,
|
| 1275 |
+
"step": 83500
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 5.62,
|
| 1279 |
+
"learning_rate": 8.040201005025126e-05,
|
| 1280 |
+
"loss": 0.0053,
|
| 1281 |
+
"step": 84000
|
| 1282 |
+
},
|
| 1283 |
+
{
|
| 1284 |
+
"epoch": 5.65,
|
| 1285 |
+
"learning_rate": 7.788944723618091e-05,
|
| 1286 |
+
"loss": 0.0053,
|
| 1287 |
+
"step": 84500
|
| 1288 |
+
},
|
| 1289 |
+
{
|
| 1290 |
+
"epoch": 5.69,
|
| 1291 |
+
"learning_rate": 7.537688442211054e-05,
|
| 1292 |
+
"loss": 0.0052,
|
| 1293 |
+
"step": 85000
|
| 1294 |
+
},
|
| 1295 |
+
{
|
| 1296 |
+
"epoch": 5.69,
|
| 1297 |
+
"eval_loss": 0.008833312429487705,
|
| 1298 |
+
"eval_runtime": 7.5954,
|
| 1299 |
+
"eval_samples_per_second": 131.659,
|
| 1300 |
+
"eval_steps_per_second": 1.317,
|
| 1301 |
+
"step": 85000
|
| 1302 |
+
},
|
| 1303 |
+
{
|
| 1304 |
+
"epoch": 5.72,
|
| 1305 |
+
"learning_rate": 7.286432160804019e-05,
|
| 1306 |
+
"loss": 0.0053,
|
| 1307 |
+
"step": 85500
|
| 1308 |
+
},
|
| 1309 |
+
{
|
| 1310 |
+
"epoch": 5.75,
|
| 1311 |
+
"learning_rate": 7.035175879396985e-05,
|
| 1312 |
+
"loss": 0.0052,
|
| 1313 |
+
"step": 86000
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 5.79,
|
| 1317 |
+
"learning_rate": 6.78391959798995e-05,
|
| 1318 |
+
"loss": 0.0052,
|
| 1319 |
+
"step": 86500
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 5.82,
|
| 1323 |
+
"learning_rate": 6.532663316582915e-05,
|
| 1324 |
+
"loss": 0.0052,
|
| 1325 |
+
"step": 87000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 5.85,
|
| 1329 |
+
"learning_rate": 6.281407035175879e-05,
|
| 1330 |
+
"loss": 0.0052,
|
| 1331 |
+
"step": 87500
|
| 1332 |
+
},
|
| 1333 |
+
{
|
| 1334 |
+
"epoch": 5.85,
|
| 1335 |
+
"eval_loss": 0.00872196163982153,
|
| 1336 |
+
"eval_runtime": 7.5919,
|
| 1337 |
+
"eval_samples_per_second": 131.72,
|
| 1338 |
+
"eval_steps_per_second": 1.317,
|
| 1339 |
+
"step": 87500
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 5.89,
|
| 1343 |
+
"learning_rate": 6.030150753768844e-05,
|
| 1344 |
+
"loss": 0.0051,
|
| 1345 |
+
"step": 88000
|
| 1346 |
+
},
|
| 1347 |
+
{
|
| 1348 |
+
"epoch": 5.92,
|
| 1349 |
+
"learning_rate": 5.778894472361809e-05,
|
| 1350 |
+
"loss": 0.0052,
|
| 1351 |
+
"step": 88500
|
| 1352 |
+
},
|
| 1353 |
+
{
|
| 1354 |
+
"epoch": 5.96,
|
| 1355 |
+
"learning_rate": 5.5276381909547746e-05,
|
| 1356 |
+
"loss": 0.0053,
|
| 1357 |
+
"step": 89000
|
| 1358 |
+
},
|
| 1359 |
+
{
|
| 1360 |
+
"epoch": 5.99,
|
| 1361 |
+
"learning_rate": 5.276381909547739e-05,
|
| 1362 |
+
"loss": 0.0052,
|
| 1363 |
+
"step": 89500
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 6.02,
|
| 1367 |
+
"learning_rate": 5.0251256281407036e-05,
|
| 1368 |
+
"loss": 0.0043,
|
| 1369 |
+
"step": 90000
|
| 1370 |
+
},
|
| 1371 |
+
{
|
| 1372 |
+
"epoch": 6.02,
|
| 1373 |
+
"eval_loss": 0.009319793432950974,
|
| 1374 |
+
"eval_runtime": 7.6253,
|
| 1375 |
+
"eval_samples_per_second": 131.143,
|
| 1376 |
+
"eval_steps_per_second": 1.311,
|
| 1377 |
+
"step": 90000
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 6.06,
|
| 1381 |
+
"learning_rate": 4.7738693467336685e-05,
|
| 1382 |
+
"loss": 0.0039,
|
| 1383 |
+
"step": 90500
|
| 1384 |
+
},
|
| 1385 |
+
{
|
| 1386 |
+
"epoch": 6.09,
|
| 1387 |
+
"learning_rate": 4.522613065326633e-05,
|
| 1388 |
+
"loss": 0.0039,
|
| 1389 |
+
"step": 91000
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 6.12,
|
| 1393 |
+
"learning_rate": 4.2713567839195975e-05,
|
| 1394 |
+
"loss": 0.004,
|
| 1395 |
+
"step": 91500
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 6.16,
|
| 1399 |
+
"learning_rate": 4.020100502512563e-05,
|
| 1400 |
+
"loss": 0.004,
|
| 1401 |
+
"step": 92000
|
| 1402 |
+
},
|
| 1403 |
+
{
|
| 1404 |
+
"epoch": 6.19,
|
| 1405 |
+
"learning_rate": 3.768844221105527e-05,
|
| 1406 |
+
"loss": 0.0038,
|
| 1407 |
+
"step": 92500
|
| 1408 |
+
},
|
| 1409 |
+
{
|
| 1410 |
+
"epoch": 6.19,
|
| 1411 |
+
"eval_loss": 0.009469099342823029,
|
| 1412 |
+
"eval_runtime": 7.9167,
|
| 1413 |
+
"eval_samples_per_second": 126.315,
|
| 1414 |
+
"eval_steps_per_second": 1.263,
|
| 1415 |
+
"step": 92500
|
| 1416 |
+
},
|
| 1417 |
+
{
|
| 1418 |
+
"epoch": 6.22,
|
| 1419 |
+
"learning_rate": 3.517587939698493e-05,
|
| 1420 |
+
"loss": 0.004,
|
| 1421 |
+
"step": 93000
|
| 1422 |
+
},
|
| 1423 |
+
{
|
| 1424 |
+
"epoch": 6.26,
|
| 1425 |
+
"learning_rate": 3.2663316582914576e-05,
|
| 1426 |
+
"loss": 0.0039,
|
| 1427 |
+
"step": 93500
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 6.29,
|
| 1431 |
+
"learning_rate": 3.015075376884422e-05,
|
| 1432 |
+
"loss": 0.0039,
|
| 1433 |
+
"step": 94000
|
| 1434 |
+
},
|
| 1435 |
+
{
|
| 1436 |
+
"epoch": 6.32,
|
| 1437 |
+
"learning_rate": 2.7638190954773873e-05,
|
| 1438 |
+
"loss": 0.004,
|
| 1439 |
+
"step": 94500
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 6.36,
|
| 1443 |
+
"learning_rate": 2.5125628140703518e-05,
|
| 1444 |
+
"loss": 0.0039,
|
| 1445 |
+
"step": 95000
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 6.36,
|
| 1449 |
+
"eval_loss": 0.009595912881195545,
|
| 1450 |
+
"eval_runtime": 7.607,
|
| 1451 |
+
"eval_samples_per_second": 131.458,
|
| 1452 |
+
"eval_steps_per_second": 1.315,
|
| 1453 |
+
"step": 95000
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 6.39,
|
| 1457 |
+
"learning_rate": 2.2613065326633167e-05,
|
| 1458 |
+
"loss": 0.004,
|
| 1459 |
+
"step": 95500
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 6.42,
|
| 1463 |
+
"learning_rate": 2.0100502512562815e-05,
|
| 1464 |
+
"loss": 0.0038,
|
| 1465 |
+
"step": 96000
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 6.46,
|
| 1469 |
+
"learning_rate": 1.7587939698492464e-05,
|
| 1470 |
+
"loss": 0.0039,
|
| 1471 |
+
"step": 96500
|
| 1472 |
+
},
|
| 1473 |
+
{
|
| 1474 |
+
"epoch": 6.49,
|
| 1475 |
+
"learning_rate": 1.507537688442211e-05,
|
| 1476 |
+
"loss": 0.004,
|
| 1477 |
+
"step": 97000
|
| 1478 |
+
},
|
| 1479 |
+
{
|
| 1480 |
+
"epoch": 6.52,
|
| 1481 |
+
"learning_rate": 1.2562814070351759e-05,
|
| 1482 |
+
"loss": 0.0039,
|
| 1483 |
+
"step": 97500
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 6.52,
|
| 1487 |
+
"eval_loss": 0.009522764943540096,
|
| 1488 |
+
"eval_runtime": 7.7116,
|
| 1489 |
+
"eval_samples_per_second": 129.674,
|
| 1490 |
+
"eval_steps_per_second": 1.297,
|
| 1491 |
+
"step": 97500
|
| 1492 |
+
}
|
| 1493 |
+
],
|
| 1494 |
+
"logging_steps": 500,
|
| 1495 |
+
"max_steps": 100000,
|
| 1496 |
+
"num_train_epochs": 7,
|
| 1497 |
+
"save_steps": 500,
|
| 1498 |
+
"total_flos": 6.676282499901161e+18,
|
| 1499 |
+
"trial_name": null,
|
| 1500 |
+
"trial_params": null
|
| 1501 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:205faa1c9ab2a2858bdf5fcb49242ab8091a30276506b912464d0e92b5a5ba59
|
| 3 |
+
size 4027
|