first version
Browse files- config.json +32 -0
- generation_config.json +6 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +2116 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "google/mt5-large",
|
3 |
+
"architectures": [
|
4 |
+
"T5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 2816,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 1024,
|
9 |
+
"decoder_start_token_id": 250099,
|
10 |
+
"dense_act_fn": "gelu_new",
|
11 |
+
"dropout_rate": 0.1,
|
12 |
+
"eos_token_id": 1,
|
13 |
+
"feed_forward_proj": "gated-gelu",
|
14 |
+
"initializer_factor": 1.0,
|
15 |
+
"is_encoder_decoder": true,
|
16 |
+
"is_gated_act": true,
|
17 |
+
"layer_norm_epsilon": 1e-06,
|
18 |
+
"model_type": "t5",
|
19 |
+
"num_decoder_layers": 24,
|
20 |
+
"num_heads": 16,
|
21 |
+
"num_layers": 24,
|
22 |
+
"output_past": true,
|
23 |
+
"pad_token_id": 0,
|
24 |
+
"relative_attention_max_distance": 128,
|
25 |
+
"relative_attention_num_buckets": 32,
|
26 |
+
"tie_word_embeddings": false,
|
27 |
+
"tokenizer_class": "T5Tokenizer",
|
28 |
+
"torch_dtype": "float32",
|
29 |
+
"transformers_version": "4.28.1",
|
30 |
+
"use_cache": true,
|
31 |
+
"vocab_size": 250112
|
32 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"decoder_start_token_id": 250099,
|
3 |
+
"eos_token_id": 1,
|
4 |
+
"pad_token_id": 0,
|
5 |
+
"transformers_version": "4.28.1"
|
6 |
+
}
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:47a7096325c2d3b9418d35a5c980c7bb1c25364336bbec0d9471ea850a923643
|
3 |
+
size 7520237
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a58264953fd55a27be72c243c0bbda74f3bdd198a45432380ef0a8505cecd0ff
|
3 |
+
size 4918519193
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d741d7bb6c5b98bfc07994e55ca9b2ac934d25e2a2612c7d2b73ff2933263de
|
3 |
+
size 14575
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:31c75da1756420e723cdc9f5ee8e203eb31348e08e9b1abcc98ec21adb6b513d
|
3 |
+
size 563
|
trainer_state.json
ADDED
@@ -0,0 +1,2116 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 1.5756924152374268,
|
3 |
+
"best_model_checkpoint": "/scratch/s3545881/dumped/translation/mt5/3086466/checkpoint-2800",
|
4 |
+
"epoch": 2.7092404450895016,
|
5 |
+
"global_step": 2800,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.01,
|
12 |
+
"learning_rate": 2.5e-06,
|
13 |
+
"loss": 37.0498,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.02,
|
18 |
+
"learning_rate": 5e-06,
|
19 |
+
"loss": 37.4476,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 0.03,
|
24 |
+
"learning_rate": 7.5e-06,
|
25 |
+
"loss": 36.636,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 0.04,
|
30 |
+
"learning_rate": 1e-05,
|
31 |
+
"loss": 35.662,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 0.05,
|
36 |
+
"learning_rate": 1.25e-05,
|
37 |
+
"loss": 34.4439,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.06,
|
42 |
+
"learning_rate": 1.5e-05,
|
43 |
+
"loss": 33.4065,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.07,
|
48 |
+
"learning_rate": 1.7500000000000002e-05,
|
49 |
+
"loss": 32.8662,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.08,
|
54 |
+
"learning_rate": 2e-05,
|
55 |
+
"loss": 32.7596,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.09,
|
60 |
+
"learning_rate": 2.2499999999999998e-05,
|
61 |
+
"loss": 31.2127,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 0.1,
|
66 |
+
"learning_rate": 2.5e-05,
|
67 |
+
"loss": 30.3467,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.11,
|
72 |
+
"learning_rate": 2.75e-05,
|
73 |
+
"loss": 28.3883,
|
74 |
+
"step": 110
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.12,
|
78 |
+
"learning_rate": 3e-05,
|
79 |
+
"loss": 26.9506,
|
80 |
+
"step": 120
|
81 |
+
},
|
82 |
+
{
|
83 |
+
"epoch": 0.13,
|
84 |
+
"learning_rate": 3.2500000000000004e-05,
|
85 |
+
"loss": 24.2034,
|
86 |
+
"step": 130
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.14,
|
90 |
+
"learning_rate": 3.5000000000000004e-05,
|
91 |
+
"loss": 22.9662,
|
92 |
+
"step": 140
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.15,
|
96 |
+
"learning_rate": 3.75e-05,
|
97 |
+
"loss": 20.9191,
|
98 |
+
"step": 150
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.15,
|
102 |
+
"learning_rate": 4e-05,
|
103 |
+
"loss": 18.8087,
|
104 |
+
"step": 160
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 0.16,
|
108 |
+
"learning_rate": 4.25e-05,
|
109 |
+
"loss": 17.9818,
|
110 |
+
"step": 170
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"epoch": 0.17,
|
114 |
+
"learning_rate": 4.4999999999999996e-05,
|
115 |
+
"loss": 16.0516,
|
116 |
+
"step": 180
|
117 |
+
},
|
118 |
+
{
|
119 |
+
"epoch": 0.18,
|
120 |
+
"learning_rate": 4.75e-05,
|
121 |
+
"loss": 13.4505,
|
122 |
+
"step": 190
|
123 |
+
},
|
124 |
+
{
|
125 |
+
"epoch": 0.19,
|
126 |
+
"learning_rate": 5e-05,
|
127 |
+
"loss": 12.6675,
|
128 |
+
"step": 200
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.19,
|
132 |
+
"eval_bp": 1.0,
|
133 |
+
"eval_counts": [
|
134 |
+
3907,
|
135 |
+
91,
|
136 |
+
14,
|
137 |
+
6
|
138 |
+
],
|
139 |
+
"eval_loss": 8.447338104248047,
|
140 |
+
"eval_precisions": [
|
141 |
+
0.5502692897061474,
|
142 |
+
0.012879411906237882,
|
143 |
+
0.0019911394295385534,
|
144 |
+
0.0008574956553553462
|
145 |
+
],
|
146 |
+
"eval_ref_len": 39265,
|
147 |
+
"eval_runtime": 1345.1117,
|
148 |
+
"eval_samples_per_second": 2.575,
|
149 |
+
"eval_score": 0.010488206448671048,
|
150 |
+
"eval_steps_per_second": 2.575,
|
151 |
+
"eval_sys_len": 710016,
|
152 |
+
"eval_totals": [
|
153 |
+
710016,
|
154 |
+
706554,
|
155 |
+
703115,
|
156 |
+
699712
|
157 |
+
],
|
158 |
+
"step": 200
|
159 |
+
},
|
160 |
+
{
|
161 |
+
"epoch": 0.2,
|
162 |
+
"learning_rate": 5.25e-05,
|
163 |
+
"loss": 11.941,
|
164 |
+
"step": 210
|
165 |
+
},
|
166 |
+
{
|
167 |
+
"epoch": 0.21,
|
168 |
+
"learning_rate": 5.5e-05,
|
169 |
+
"loss": 10.961,
|
170 |
+
"step": 220
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.22,
|
174 |
+
"learning_rate": 5.75e-05,
|
175 |
+
"loss": 10.3499,
|
176 |
+
"step": 230
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.23,
|
180 |
+
"learning_rate": 6e-05,
|
181 |
+
"loss": 9.5122,
|
182 |
+
"step": 240
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.24,
|
186 |
+
"learning_rate": 6.25e-05,
|
187 |
+
"loss": 9.1592,
|
188 |
+
"step": 250
|
189 |
+
},
|
190 |
+
{
|
191 |
+
"epoch": 0.25,
|
192 |
+
"learning_rate": 6.500000000000001e-05,
|
193 |
+
"loss": 8.7164,
|
194 |
+
"step": 260
|
195 |
+
},
|
196 |
+
{
|
197 |
+
"epoch": 0.26,
|
198 |
+
"learning_rate": 6.75e-05,
|
199 |
+
"loss": 8.4345,
|
200 |
+
"step": 270
|
201 |
+
},
|
202 |
+
{
|
203 |
+
"epoch": 0.27,
|
204 |
+
"learning_rate": 7.000000000000001e-05,
|
205 |
+
"loss": 8.0707,
|
206 |
+
"step": 280
|
207 |
+
},
|
208 |
+
{
|
209 |
+
"epoch": 0.28,
|
210 |
+
"learning_rate": 7.25e-05,
|
211 |
+
"loss": 7.959,
|
212 |
+
"step": 290
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.29,
|
216 |
+
"learning_rate": 7.5e-05,
|
217 |
+
"loss": 7.661,
|
218 |
+
"step": 300
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.3,
|
222 |
+
"learning_rate": 7.75e-05,
|
223 |
+
"loss": 7.7047,
|
224 |
+
"step": 310
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.31,
|
228 |
+
"learning_rate": 8e-05,
|
229 |
+
"loss": 7.6889,
|
230 |
+
"step": 320
|
231 |
+
},
|
232 |
+
{
|
233 |
+
"epoch": 0.32,
|
234 |
+
"learning_rate": 8.25e-05,
|
235 |
+
"loss": 7.7815,
|
236 |
+
"step": 330
|
237 |
+
},
|
238 |
+
{
|
239 |
+
"epoch": 0.33,
|
240 |
+
"learning_rate": 8.5e-05,
|
241 |
+
"loss": 8.0411,
|
242 |
+
"step": 340
|
243 |
+
},
|
244 |
+
{
|
245 |
+
"epoch": 0.34,
|
246 |
+
"learning_rate": 8.75e-05,
|
247 |
+
"loss": 7.8518,
|
248 |
+
"step": 350
|
249 |
+
},
|
250 |
+
{
|
251 |
+
"epoch": 0.35,
|
252 |
+
"learning_rate": 8.999999999999999e-05,
|
253 |
+
"loss": 7.7647,
|
254 |
+
"step": 360
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 0.36,
|
258 |
+
"learning_rate": 9.25e-05,
|
259 |
+
"loss": 7.4841,
|
260 |
+
"step": 370
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 0.37,
|
264 |
+
"learning_rate": 9.5e-05,
|
265 |
+
"loss": 7.2992,
|
266 |
+
"step": 380
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 0.38,
|
270 |
+
"learning_rate": 9.750000000000001e-05,
|
271 |
+
"loss": 7.2381,
|
272 |
+
"step": 390
|
273 |
+
},
|
274 |
+
{
|
275 |
+
"epoch": 0.39,
|
276 |
+
"learning_rate": 0.0001,
|
277 |
+
"loss": 7.1215,
|
278 |
+
"step": 400
|
279 |
+
},
|
280 |
+
{
|
281 |
+
"epoch": 0.39,
|
282 |
+
"eval_bp": 1.0,
|
283 |
+
"eval_counts": [
|
284 |
+
3135,
|
285 |
+
982,
|
286 |
+
295,
|
287 |
+
61
|
288 |
+
],
|
289 |
+
"eval_loss": 6.286175727844238,
|
290 |
+
"eval_precisions": [
|
291 |
+
0.9876939944424491,
|
292 |
+
0.31213545788873104,
|
293 |
+
0.09449404046907482,
|
294 |
+
0.019671708213744397
|
295 |
+
],
|
296 |
+
"eval_ref_len": 39265,
|
297 |
+
"eval_runtime": 1325.4365,
|
298 |
+
"eval_samples_per_second": 2.613,
|
299 |
+
"eval_score": 0.15472233223389328,
|
300 |
+
"eval_steps_per_second": 2.613,
|
301 |
+
"eval_sys_len": 317406,
|
302 |
+
"eval_totals": [
|
303 |
+
317406,
|
304 |
+
314607,
|
305 |
+
312189,
|
306 |
+
310090
|
307 |
+
],
|
308 |
+
"step": 400
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 0.4,
|
312 |
+
"learning_rate": 0.0001025,
|
313 |
+
"loss": 7.0884,
|
314 |
+
"step": 410
|
315 |
+
},
|
316 |
+
{
|
317 |
+
"epoch": 0.41,
|
318 |
+
"learning_rate": 0.000105,
|
319 |
+
"loss": 7.0528,
|
320 |
+
"step": 420
|
321 |
+
},
|
322 |
+
{
|
323 |
+
"epoch": 0.42,
|
324 |
+
"learning_rate": 0.0001075,
|
325 |
+
"loss": 7.0151,
|
326 |
+
"step": 430
|
327 |
+
},
|
328 |
+
{
|
329 |
+
"epoch": 0.43,
|
330 |
+
"learning_rate": 0.00011,
|
331 |
+
"loss": 6.7428,
|
332 |
+
"step": 440
|
333 |
+
},
|
334 |
+
{
|
335 |
+
"epoch": 0.44,
|
336 |
+
"learning_rate": 0.00011250000000000001,
|
337 |
+
"loss": 6.5513,
|
338 |
+
"step": 450
|
339 |
+
},
|
340 |
+
{
|
341 |
+
"epoch": 0.45,
|
342 |
+
"learning_rate": 0.000115,
|
343 |
+
"loss": 6.4033,
|
344 |
+
"step": 460
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 0.45,
|
348 |
+
"learning_rate": 0.0001175,
|
349 |
+
"loss": 6.0151,
|
350 |
+
"step": 470
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 0.46,
|
354 |
+
"learning_rate": 0.00012,
|
355 |
+
"loss": 5.5711,
|
356 |
+
"step": 480
|
357 |
+
},
|
358 |
+
{
|
359 |
+
"epoch": 0.47,
|
360 |
+
"learning_rate": 0.0001225,
|
361 |
+
"loss": 5.0243,
|
362 |
+
"step": 490
|
363 |
+
},
|
364 |
+
{
|
365 |
+
"epoch": 0.48,
|
366 |
+
"learning_rate": 0.000125,
|
367 |
+
"loss": 4.8831,
|
368 |
+
"step": 500
|
369 |
+
},
|
370 |
+
{
|
371 |
+
"epoch": 0.49,
|
372 |
+
"learning_rate": 0.0001275,
|
373 |
+
"loss": 4.636,
|
374 |
+
"step": 510
|
375 |
+
},
|
376 |
+
{
|
377 |
+
"epoch": 0.5,
|
378 |
+
"learning_rate": 0.00013000000000000002,
|
379 |
+
"loss": 4.5649,
|
380 |
+
"step": 520
|
381 |
+
},
|
382 |
+
{
|
383 |
+
"epoch": 0.51,
|
384 |
+
"learning_rate": 0.00013250000000000002,
|
385 |
+
"loss": 4.1967,
|
386 |
+
"step": 530
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 0.52,
|
390 |
+
"learning_rate": 0.000135,
|
391 |
+
"loss": 4.4032,
|
392 |
+
"step": 540
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 0.53,
|
396 |
+
"learning_rate": 0.0001375,
|
397 |
+
"loss": 4.3111,
|
398 |
+
"step": 550
|
399 |
+
},
|
400 |
+
{
|
401 |
+
"epoch": 0.54,
|
402 |
+
"learning_rate": 0.00014000000000000001,
|
403 |
+
"loss": 4.2533,
|
404 |
+
"step": 560
|
405 |
+
},
|
406 |
+
{
|
407 |
+
"epoch": 0.55,
|
408 |
+
"learning_rate": 0.0001425,
|
409 |
+
"loss": 4.2514,
|
410 |
+
"step": 570
|
411 |
+
},
|
412 |
+
{
|
413 |
+
"epoch": 0.56,
|
414 |
+
"learning_rate": 0.000145,
|
415 |
+
"loss": 4.1326,
|
416 |
+
"step": 580
|
417 |
+
},
|
418 |
+
{
|
419 |
+
"epoch": 0.57,
|
420 |
+
"learning_rate": 0.0001475,
|
421 |
+
"loss": 4.0604,
|
422 |
+
"step": 590
|
423 |
+
},
|
424 |
+
{
|
425 |
+
"epoch": 0.58,
|
426 |
+
"learning_rate": 0.00015,
|
427 |
+
"loss": 4.0115,
|
428 |
+
"step": 600
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 0.58,
|
432 |
+
"eval_bp": 1.0,
|
433 |
+
"eval_counts": [
|
434 |
+
22228,
|
435 |
+
12576,
|
436 |
+
7955,
|
437 |
+
5383
|
438 |
+
],
|
439 |
+
"eval_loss": 2.3380367755889893,
|
440 |
+
"eval_precisions": [
|
441 |
+
45.45975130890052,
|
442 |
+
27.68032047190368,
|
443 |
+
18.94273127753304,
|
444 |
+
13.947402513278922
|
445 |
+
],
|
446 |
+
"eval_ref_len": 39265,
|
447 |
+
"eval_runtime": 572.0306,
|
448 |
+
"eval_samples_per_second": 6.054,
|
449 |
+
"eval_score": 24.01228700570531,
|
450 |
+
"eval_steps_per_second": 6.054,
|
451 |
+
"eval_sys_len": 48896,
|
452 |
+
"eval_totals": [
|
453 |
+
48896,
|
454 |
+
45433,
|
455 |
+
41995,
|
456 |
+
38595
|
457 |
+
],
|
458 |
+
"step": 600
|
459 |
+
},
|
460 |
+
{
|
461 |
+
"epoch": 0.59,
|
462 |
+
"learning_rate": 0.0001525,
|
463 |
+
"loss": 3.9441,
|
464 |
+
"step": 610
|
465 |
+
},
|
466 |
+
{
|
467 |
+
"epoch": 0.6,
|
468 |
+
"learning_rate": 0.000155,
|
469 |
+
"loss": 3.9152,
|
470 |
+
"step": 620
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 0.61,
|
474 |
+
"learning_rate": 0.0001575,
|
475 |
+
"loss": 3.9774,
|
476 |
+
"step": 630
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 0.62,
|
480 |
+
"learning_rate": 0.00016,
|
481 |
+
"loss": 3.7911,
|
482 |
+
"step": 640
|
483 |
+
},
|
484 |
+
{
|
485 |
+
"epoch": 0.63,
|
486 |
+
"learning_rate": 0.00016250000000000002,
|
487 |
+
"loss": 3.8047,
|
488 |
+
"step": 650
|
489 |
+
},
|
490 |
+
{
|
491 |
+
"epoch": 0.64,
|
492 |
+
"learning_rate": 0.000165,
|
493 |
+
"loss": 3.8829,
|
494 |
+
"step": 660
|
495 |
+
},
|
496 |
+
{
|
497 |
+
"epoch": 0.65,
|
498 |
+
"learning_rate": 0.0001675,
|
499 |
+
"loss": 3.6344,
|
500 |
+
"step": 670
|
501 |
+
},
|
502 |
+
{
|
503 |
+
"epoch": 0.66,
|
504 |
+
"learning_rate": 0.00017,
|
505 |
+
"loss": 3.8138,
|
506 |
+
"step": 680
|
507 |
+
},
|
508 |
+
{
|
509 |
+
"epoch": 0.67,
|
510 |
+
"learning_rate": 0.0001725,
|
511 |
+
"loss": 3.7081,
|
512 |
+
"step": 690
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 0.68,
|
516 |
+
"learning_rate": 0.000175,
|
517 |
+
"loss": 3.7708,
|
518 |
+
"step": 700
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 0.69,
|
522 |
+
"learning_rate": 0.0001775,
|
523 |
+
"loss": 3.7171,
|
524 |
+
"step": 710
|
525 |
+
},
|
526 |
+
{
|
527 |
+
"epoch": 0.7,
|
528 |
+
"learning_rate": 0.00017999999999999998,
|
529 |
+
"loss": 3.5893,
|
530 |
+
"step": 720
|
531 |
+
},
|
532 |
+
{
|
533 |
+
"epoch": 0.71,
|
534 |
+
"learning_rate": 0.0001825,
|
535 |
+
"loss": 3.5494,
|
536 |
+
"step": 730
|
537 |
+
},
|
538 |
+
{
|
539 |
+
"epoch": 0.72,
|
540 |
+
"learning_rate": 0.000185,
|
541 |
+
"loss": 3.6416,
|
542 |
+
"step": 740
|
543 |
+
},
|
544 |
+
{
|
545 |
+
"epoch": 0.73,
|
546 |
+
"learning_rate": 0.0001875,
|
547 |
+
"loss": 3.6174,
|
548 |
+
"step": 750
|
549 |
+
},
|
550 |
+
{
|
551 |
+
"epoch": 0.74,
|
552 |
+
"learning_rate": 0.00019,
|
553 |
+
"loss": 3.6616,
|
554 |
+
"step": 760
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 0.75,
|
558 |
+
"learning_rate": 0.00019250000000000002,
|
559 |
+
"loss": 3.456,
|
560 |
+
"step": 770
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 0.75,
|
564 |
+
"learning_rate": 0.00019500000000000002,
|
565 |
+
"loss": 3.6063,
|
566 |
+
"step": 780
|
567 |
+
},
|
568 |
+
{
|
569 |
+
"epoch": 0.76,
|
570 |
+
"learning_rate": 0.0001975,
|
571 |
+
"loss": 3.6423,
|
572 |
+
"step": 790
|
573 |
+
},
|
574 |
+
{
|
575 |
+
"epoch": 0.77,
|
576 |
+
"learning_rate": 0.0002,
|
577 |
+
"loss": 3.6089,
|
578 |
+
"step": 800
|
579 |
+
},
|
580 |
+
{
|
581 |
+
"epoch": 0.77,
|
582 |
+
"eval_bp": 1.0,
|
583 |
+
"eval_counts": [
|
584 |
+
22303,
|
585 |
+
12687,
|
586 |
+
7909,
|
587 |
+
5220
|
588 |
+
],
|
589 |
+
"eval_loss": 2.027892589569092,
|
590 |
+
"eval_precisions": [
|
591 |
+
51.82164598726707,
|
592 |
+
32.05811749842072,
|
593 |
+
21.88494424306151,
|
594 |
+
15.94282572842221
|
595 |
+
],
|
596 |
+
"eval_ref_len": 39265,
|
597 |
+
"eval_runtime": 511.1135,
|
598 |
+
"eval_samples_per_second": 6.775,
|
599 |
+
"eval_score": 27.592432621043134,
|
600 |
+
"eval_steps_per_second": 6.775,
|
601 |
+
"eval_sys_len": 43038,
|
602 |
+
"eval_totals": [
|
603 |
+
43038,
|
604 |
+
39575,
|
605 |
+
36139,
|
606 |
+
32742
|
607 |
+
],
|
608 |
+
"step": 800
|
609 |
+
},
|
610 |
+
{
|
611 |
+
"epoch": 0.78,
|
612 |
+
"learning_rate": 0.00020250000000000002,
|
613 |
+
"loss": 3.4158,
|
614 |
+
"step": 810
|
615 |
+
},
|
616 |
+
{
|
617 |
+
"epoch": 0.79,
|
618 |
+
"learning_rate": 0.000205,
|
619 |
+
"loss": 3.409,
|
620 |
+
"step": 820
|
621 |
+
},
|
622 |
+
{
|
623 |
+
"epoch": 0.8,
|
624 |
+
"learning_rate": 0.0002075,
|
625 |
+
"loss": 3.5841,
|
626 |
+
"step": 830
|
627 |
+
},
|
628 |
+
{
|
629 |
+
"epoch": 0.81,
|
630 |
+
"learning_rate": 0.00021,
|
631 |
+
"loss": 3.4777,
|
632 |
+
"step": 840
|
633 |
+
},
|
634 |
+
{
|
635 |
+
"epoch": 0.82,
|
636 |
+
"learning_rate": 0.0002125,
|
637 |
+
"loss": 3.4107,
|
638 |
+
"step": 850
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 0.83,
|
642 |
+
"learning_rate": 0.000215,
|
643 |
+
"loss": 3.3874,
|
644 |
+
"step": 860
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 0.84,
|
648 |
+
"learning_rate": 0.0002175,
|
649 |
+
"loss": 3.4342,
|
650 |
+
"step": 870
|
651 |
+
},
|
652 |
+
{
|
653 |
+
"epoch": 0.85,
|
654 |
+
"learning_rate": 0.00022,
|
655 |
+
"loss": 3.4097,
|
656 |
+
"step": 880
|
657 |
+
},
|
658 |
+
{
|
659 |
+
"epoch": 0.86,
|
660 |
+
"learning_rate": 0.00022250000000000001,
|
661 |
+
"loss": 3.3245,
|
662 |
+
"step": 890
|
663 |
+
},
|
664 |
+
{
|
665 |
+
"epoch": 0.87,
|
666 |
+
"learning_rate": 0.00022500000000000002,
|
667 |
+
"loss": 3.3971,
|
668 |
+
"step": 900
|
669 |
+
},
|
670 |
+
{
|
671 |
+
"epoch": 0.88,
|
672 |
+
"learning_rate": 0.0002275,
|
673 |
+
"loss": 3.2777,
|
674 |
+
"step": 910
|
675 |
+
},
|
676 |
+
{
|
677 |
+
"epoch": 0.89,
|
678 |
+
"learning_rate": 0.00023,
|
679 |
+
"loss": 3.2984,
|
680 |
+
"step": 920
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 0.9,
|
684 |
+
"learning_rate": 0.0002325,
|
685 |
+
"loss": 3.2537,
|
686 |
+
"step": 930
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 0.91,
|
690 |
+
"learning_rate": 0.000235,
|
691 |
+
"loss": 3.3175,
|
692 |
+
"step": 940
|
693 |
+
},
|
694 |
+
{
|
695 |
+
"epoch": 0.92,
|
696 |
+
"learning_rate": 0.0002375,
|
697 |
+
"loss": 3.287,
|
698 |
+
"step": 950
|
699 |
+
},
|
700 |
+
{
|
701 |
+
"epoch": 0.93,
|
702 |
+
"learning_rate": 0.00024,
|
703 |
+
"loss": 3.2052,
|
704 |
+
"step": 960
|
705 |
+
},
|
706 |
+
{
|
707 |
+
"epoch": 0.94,
|
708 |
+
"learning_rate": 0.00024249999999999999,
|
709 |
+
"loss": 3.2984,
|
710 |
+
"step": 970
|
711 |
+
},
|
712 |
+
{
|
713 |
+
"epoch": 0.95,
|
714 |
+
"learning_rate": 0.000245,
|
715 |
+
"loss": 3.137,
|
716 |
+
"step": 980
|
717 |
+
},
|
718 |
+
{
|
719 |
+
"epoch": 0.96,
|
720 |
+
"learning_rate": 0.0002475,
|
721 |
+
"loss": 3.2062,
|
722 |
+
"step": 990
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 0.97,
|
726 |
+
"learning_rate": 0.00025,
|
727 |
+
"loss": 3.3031,
|
728 |
+
"step": 1000
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 0.97,
|
732 |
+
"eval_bp": 1.0,
|
733 |
+
"eval_counts": [
|
734 |
+
22475,
|
735 |
+
12870,
|
736 |
+
8033,
|
737 |
+
5319
|
738 |
+
],
|
739 |
+
"eval_loss": 1.880346655845642,
|
740 |
+
"eval_precisions": [
|
741 |
+
52.89355392906733,
|
742 |
+
32.97632468996618,
|
743 |
+
22.57031271950774,
|
744 |
+
16.520172686896295
|
745 |
+
],
|
746 |
+
"eval_ref_len": 39265,
|
747 |
+
"eval_runtime": 493.392,
|
748 |
+
"eval_samples_per_second": 7.019,
|
749 |
+
"eval_score": 28.39810035019441,
|
750 |
+
"eval_steps_per_second": 7.019,
|
751 |
+
"eval_sys_len": 42491,
|
752 |
+
"eval_totals": [
|
753 |
+
42491,
|
754 |
+
39028,
|
755 |
+
35591,
|
756 |
+
32197
|
757 |
+
],
|
758 |
+
"step": 1000
|
759 |
+
},
|
760 |
+
{
|
761 |
+
"epoch": 0.98,
|
762 |
+
"learning_rate": 0.0002525,
|
763 |
+
"loss": 3.3379,
|
764 |
+
"step": 1010
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 0.99,
|
768 |
+
"learning_rate": 0.000255,
|
769 |
+
"loss": 3.3144,
|
770 |
+
"step": 1020
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 1.0,
|
774 |
+
"learning_rate": 0.0002575,
|
775 |
+
"loss": 3.3039,
|
776 |
+
"step": 1030
|
777 |
+
},
|
778 |
+
{
|
779 |
+
"epoch": 1.01,
|
780 |
+
"learning_rate": 0.00026000000000000003,
|
781 |
+
"loss": 3.2099,
|
782 |
+
"step": 1040
|
783 |
+
},
|
784 |
+
{
|
785 |
+
"epoch": 1.02,
|
786 |
+
"learning_rate": 0.00026250000000000004,
|
787 |
+
"loss": 3.203,
|
788 |
+
"step": 1050
|
789 |
+
},
|
790 |
+
{
|
791 |
+
"epoch": 1.03,
|
792 |
+
"learning_rate": 0.00026500000000000004,
|
793 |
+
"loss": 3.027,
|
794 |
+
"step": 1060
|
795 |
+
},
|
796 |
+
{
|
797 |
+
"epoch": 1.04,
|
798 |
+
"learning_rate": 0.0002675,
|
799 |
+
"loss": 3.0334,
|
800 |
+
"step": 1070
|
801 |
+
},
|
802 |
+
{
|
803 |
+
"epoch": 1.04,
|
804 |
+
"learning_rate": 0.00027,
|
805 |
+
"loss": 3.1445,
|
806 |
+
"step": 1080
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 1.05,
|
810 |
+
"learning_rate": 0.0002725,
|
811 |
+
"loss": 3.0927,
|
812 |
+
"step": 1090
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 1.06,
|
816 |
+
"learning_rate": 0.000275,
|
817 |
+
"loss": 3.1355,
|
818 |
+
"step": 1100
|
819 |
+
},
|
820 |
+
{
|
821 |
+
"epoch": 1.07,
|
822 |
+
"learning_rate": 0.0002775,
|
823 |
+
"loss": 2.9411,
|
824 |
+
"step": 1110
|
825 |
+
},
|
826 |
+
{
|
827 |
+
"epoch": 1.08,
|
828 |
+
"learning_rate": 0.00028000000000000003,
|
829 |
+
"loss": 3.0683,
|
830 |
+
"step": 1120
|
831 |
+
},
|
832 |
+
{
|
833 |
+
"epoch": 1.09,
|
834 |
+
"learning_rate": 0.0002825,
|
835 |
+
"loss": 3.0704,
|
836 |
+
"step": 1130
|
837 |
+
},
|
838 |
+
{
|
839 |
+
"epoch": 1.1,
|
840 |
+
"learning_rate": 0.000285,
|
841 |
+
"loss": 3.1367,
|
842 |
+
"step": 1140
|
843 |
+
},
|
844 |
+
{
|
845 |
+
"epoch": 1.11,
|
846 |
+
"learning_rate": 0.0002875,
|
847 |
+
"loss": 3.1829,
|
848 |
+
"step": 1150
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 1.12,
|
852 |
+
"learning_rate": 0.00029,
|
853 |
+
"loss": 3.101,
|
854 |
+
"step": 1160
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 1.13,
|
858 |
+
"learning_rate": 0.0002925,
|
859 |
+
"loss": 3.0338,
|
860 |
+
"step": 1170
|
861 |
+
},
|
862 |
+
{
|
863 |
+
"epoch": 1.14,
|
864 |
+
"learning_rate": 0.000295,
|
865 |
+
"loss": 3.0992,
|
866 |
+
"step": 1180
|
867 |
+
},
|
868 |
+
{
|
869 |
+
"epoch": 1.15,
|
870 |
+
"learning_rate": 0.00029749999999999997,
|
871 |
+
"loss": 3.1358,
|
872 |
+
"step": 1190
|
873 |
+
},
|
874 |
+
{
|
875 |
+
"epoch": 1.16,
|
876 |
+
"learning_rate": 0.0003,
|
877 |
+
"loss": 3.0265,
|
878 |
+
"step": 1200
|
879 |
+
},
|
880 |
+
{
|
881 |
+
"epoch": 1.16,
|
882 |
+
"eval_bp": 1.0,
|
883 |
+
"eval_counts": [
|
884 |
+
22069,
|
885 |
+
12741,
|
886 |
+
8000,
|
887 |
+
5303
|
888 |
+
],
|
889 |
+
"eval_loss": 1.8022379875183105,
|
890 |
+
"eval_precisions": [
|
891 |
+
49.50426200089726,
|
892 |
+
30.98718291704161,
|
893 |
+
21.2286055459732,
|
894 |
+
15.46019066499519
|
895 |
+
],
|
896 |
+
"eval_ref_len": 39265,
|
897 |
+
"eval_runtime": 683.7176,
|
898 |
+
"eval_samples_per_second": 5.065,
|
899 |
+
"eval_score": 26.637302937486467,
|
900 |
+
"eval_steps_per_second": 5.065,
|
901 |
+
"eval_sys_len": 44580,
|
902 |
+
"eval_totals": [
|
903 |
+
44580,
|
904 |
+
41117,
|
905 |
+
37685,
|
906 |
+
34301
|
907 |
+
],
|
908 |
+
"step": 1200
|
909 |
+
},
|
910 |
+
{
|
911 |
+
"epoch": 1.17,
|
912 |
+
"learning_rate": 0.0003025,
|
913 |
+
"loss": 2.9454,
|
914 |
+
"step": 1210
|
915 |
+
},
|
916 |
+
{
|
917 |
+
"epoch": 1.18,
|
918 |
+
"learning_rate": 0.000305,
|
919 |
+
"loss": 3.0222,
|
920 |
+
"step": 1220
|
921 |
+
},
|
922 |
+
{
|
923 |
+
"epoch": 1.19,
|
924 |
+
"learning_rate": 0.0003075,
|
925 |
+
"loss": 3.08,
|
926 |
+
"step": 1230
|
927 |
+
},
|
928 |
+
{
|
929 |
+
"epoch": 1.2,
|
930 |
+
"learning_rate": 0.00031,
|
931 |
+
"loss": 2.9951,
|
932 |
+
"step": 1240
|
933 |
+
},
|
934 |
+
{
|
935 |
+
"epoch": 1.21,
|
936 |
+
"learning_rate": 0.0003125,
|
937 |
+
"loss": 2.9722,
|
938 |
+
"step": 1250
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 1.22,
|
942 |
+
"learning_rate": 0.000315,
|
943 |
+
"loss": 2.9865,
|
944 |
+
"step": 1260
|
945 |
+
},
|
946 |
+
{
|
947 |
+
"epoch": 1.23,
|
948 |
+
"learning_rate": 0.0003175,
|
949 |
+
"loss": 2.9643,
|
950 |
+
"step": 1270
|
951 |
+
},
|
952 |
+
{
|
953 |
+
"epoch": 1.24,
|
954 |
+
"learning_rate": 0.00032,
|
955 |
+
"loss": 2.9591,
|
956 |
+
"step": 1280
|
957 |
+
},
|
958 |
+
{
|
959 |
+
"epoch": 1.25,
|
960 |
+
"learning_rate": 0.00032250000000000003,
|
961 |
+
"loss": 2.9174,
|
962 |
+
"step": 1290
|
963 |
+
},
|
964 |
+
{
|
965 |
+
"epoch": 1.26,
|
966 |
+
"learning_rate": 0.00032500000000000004,
|
967 |
+
"loss": 2.9484,
|
968 |
+
"step": 1300
|
969 |
+
},
|
970 |
+
{
|
971 |
+
"epoch": 1.27,
|
972 |
+
"learning_rate": 0.00032750000000000005,
|
973 |
+
"loss": 2.9186,
|
974 |
+
"step": 1310
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 1.28,
|
978 |
+
"learning_rate": 0.00033,
|
979 |
+
"loss": 2.9222,
|
980 |
+
"step": 1320
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 1.29,
|
984 |
+
"learning_rate": 0.0003325,
|
985 |
+
"loss": 2.9231,
|
986 |
+
"step": 1330
|
987 |
+
},
|
988 |
+
{
|
989 |
+
"epoch": 1.3,
|
990 |
+
"learning_rate": 0.000335,
|
991 |
+
"loss": 2.8752,
|
992 |
+
"step": 1340
|
993 |
+
},
|
994 |
+
{
|
995 |
+
"epoch": 1.31,
|
996 |
+
"learning_rate": 0.0003375,
|
997 |
+
"loss": 2.8569,
|
998 |
+
"step": 1350
|
999 |
+
},
|
1000 |
+
{
|
1001 |
+
"epoch": 1.32,
|
1002 |
+
"learning_rate": 0.00034,
|
1003 |
+
"loss": 3.0137,
|
1004 |
+
"step": 1360
|
1005 |
+
},
|
1006 |
+
{
|
1007 |
+
"epoch": 1.33,
|
1008 |
+
"learning_rate": 0.00034250000000000003,
|
1009 |
+
"loss": 2.9199,
|
1010 |
+
"step": 1370
|
1011 |
+
},
|
1012 |
+
{
|
1013 |
+
"epoch": 1.34,
|
1014 |
+
"learning_rate": 0.000345,
|
1015 |
+
"loss": 2.9726,
|
1016 |
+
"step": 1380
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 1.34,
|
1020 |
+
"learning_rate": 0.0003475,
|
1021 |
+
"loss": 2.906,
|
1022 |
+
"step": 1390
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 1.35,
|
1026 |
+
"learning_rate": 0.00035,
|
1027 |
+
"loss": 2.9055,
|
1028 |
+
"step": 1400
|
1029 |
+
},
|
1030 |
+
{
|
1031 |
+
"epoch": 1.35,
|
1032 |
+
"eval_bp": 1.0,
|
1033 |
+
"eval_counts": [
|
1034 |
+
23204,
|
1035 |
+
13680,
|
1036 |
+
8622,
|
1037 |
+
5764
|
1038 |
+
],
|
1039 |
+
"eval_loss": 1.7418313026428223,
|
1040 |
+
"eval_precisions": [
|
1041 |
+
54.13147949423786,
|
1042 |
+
34.697035026758314,
|
1043 |
+
23.951330629479415,
|
1044 |
+
17.671224477282482
|
1045 |
+
],
|
1046 |
+
"eval_ref_len": 39265,
|
1047 |
+
"eval_runtime": 544.287,
|
1048 |
+
"eval_samples_per_second": 6.362,
|
1049 |
+
"eval_score": 29.859644843289843,
|
1050 |
+
"eval_steps_per_second": 6.362,
|
1051 |
+
"eval_sys_len": 42866,
|
1052 |
+
"eval_totals": [
|
1053 |
+
42866,
|
1054 |
+
39427,
|
1055 |
+
35998,
|
1056 |
+
32618
|
1057 |
+
],
|
1058 |
+
"step": 1400
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 1.36,
|
1062 |
+
"learning_rate": 0.0003525,
|
1063 |
+
"loss": 2.8119,
|
1064 |
+
"step": 1410
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 1.37,
|
1068 |
+
"learning_rate": 0.000355,
|
1069 |
+
"loss": 3.0053,
|
1070 |
+
"step": 1420
|
1071 |
+
},
|
1072 |
+
{
|
1073 |
+
"epoch": 1.38,
|
1074 |
+
"learning_rate": 0.0003575,
|
1075 |
+
"loss": 2.8639,
|
1076 |
+
"step": 1430
|
1077 |
+
},
|
1078 |
+
{
|
1079 |
+
"epoch": 1.39,
|
1080 |
+
"learning_rate": 0.00035999999999999997,
|
1081 |
+
"loss": 2.9254,
|
1082 |
+
"step": 1440
|
1083 |
+
},
|
1084 |
+
{
|
1085 |
+
"epoch": 1.4,
|
1086 |
+
"learning_rate": 0.0003625,
|
1087 |
+
"loss": 2.9477,
|
1088 |
+
"step": 1450
|
1089 |
+
},
|
1090 |
+
{
|
1091 |
+
"epoch": 1.41,
|
1092 |
+
"learning_rate": 0.000365,
|
1093 |
+
"loss": 3.0299,
|
1094 |
+
"step": 1460
|
1095 |
+
},
|
1096 |
+
{
|
1097 |
+
"epoch": 1.42,
|
1098 |
+
"learning_rate": 0.0003675,
|
1099 |
+
"loss": 2.8582,
|
1100 |
+
"step": 1470
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 1.43,
|
1104 |
+
"learning_rate": 0.00037,
|
1105 |
+
"loss": 2.9433,
|
1106 |
+
"step": 1480
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 1.44,
|
1110 |
+
"learning_rate": 0.0003725,
|
1111 |
+
"loss": 2.856,
|
1112 |
+
"step": 1490
|
1113 |
+
},
|
1114 |
+
{
|
1115 |
+
"epoch": 1.45,
|
1116 |
+
"learning_rate": 0.000375,
|
1117 |
+
"loss": 2.7648,
|
1118 |
+
"step": 1500
|
1119 |
+
},
|
1120 |
+
{
|
1121 |
+
"epoch": 1.46,
|
1122 |
+
"learning_rate": 0.0003775,
|
1123 |
+
"loss": 2.842,
|
1124 |
+
"step": 1510
|
1125 |
+
},
|
1126 |
+
{
|
1127 |
+
"epoch": 1.47,
|
1128 |
+
"learning_rate": 0.00038,
|
1129 |
+
"loss": 2.7609,
|
1130 |
+
"step": 1520
|
1131 |
+
},
|
1132 |
+
{
|
1133 |
+
"epoch": 1.48,
|
1134 |
+
"learning_rate": 0.00038250000000000003,
|
1135 |
+
"loss": 2.9978,
|
1136 |
+
"step": 1530
|
1137 |
+
},
|
1138 |
+
{
|
1139 |
+
"epoch": 1.49,
|
1140 |
+
"learning_rate": 0.00038500000000000003,
|
1141 |
+
"loss": 2.8184,
|
1142 |
+
"step": 1540
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 1.5,
|
1146 |
+
"learning_rate": 0.00038750000000000004,
|
1147 |
+
"loss": 2.8703,
|
1148 |
+
"step": 1550
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 1.51,
|
1152 |
+
"learning_rate": 0.00039000000000000005,
|
1153 |
+
"loss": 2.7835,
|
1154 |
+
"step": 1560
|
1155 |
+
},
|
1156 |
+
{
|
1157 |
+
"epoch": 1.52,
|
1158 |
+
"learning_rate": 0.0003925,
|
1159 |
+
"loss": 2.8729,
|
1160 |
+
"step": 1570
|
1161 |
+
},
|
1162 |
+
{
|
1163 |
+
"epoch": 1.53,
|
1164 |
+
"learning_rate": 0.000395,
|
1165 |
+
"loss": 2.8479,
|
1166 |
+
"step": 1580
|
1167 |
+
},
|
1168 |
+
{
|
1169 |
+
"epoch": 1.54,
|
1170 |
+
"learning_rate": 0.0003975,
|
1171 |
+
"loss": 2.8868,
|
1172 |
+
"step": 1590
|
1173 |
+
},
|
1174 |
+
{
|
1175 |
+
"epoch": 1.55,
|
1176 |
+
"learning_rate": 0.0004,
|
1177 |
+
"loss": 2.829,
|
1178 |
+
"step": 1600
|
1179 |
+
},
|
1180 |
+
{
|
1181 |
+
"epoch": 1.55,
|
1182 |
+
"eval_bp": 1.0,
|
1183 |
+
"eval_counts": [
|
1184 |
+
23892,
|
1185 |
+
14346,
|
1186 |
+
9260,
|
1187 |
+
6235
|
1188 |
+
],
|
1189 |
+
"eval_loss": 1.7061405181884766,
|
1190 |
+
"eval_precisions": [
|
1191 |
+
57.79110831599826,
|
1192 |
+
37.873227909923706,
|
1193 |
+
26.900618772332454,
|
1194 |
+
20.09993552546744
|
1195 |
+
],
|
1196 |
+
"eval_ref_len": 39265,
|
1197 |
+
"eval_runtime": 422.3766,
|
1198 |
+
"eval_samples_per_second": 8.199,
|
1199 |
+
"eval_score": 32.98280367671718,
|
1200 |
+
"eval_steps_per_second": 8.199,
|
1201 |
+
"eval_sys_len": 41342,
|
1202 |
+
"eval_totals": [
|
1203 |
+
41342,
|
1204 |
+
37879,
|
1205 |
+
34423,
|
1206 |
+
31020
|
1207 |
+
],
|
1208 |
+
"step": 1600
|
1209 |
+
},
|
1210 |
+
{
|
1211 |
+
"epoch": 1.56,
|
1212 |
+
"learning_rate": 0.0004025,
|
1213 |
+
"loss": 2.7557,
|
1214 |
+
"step": 1610
|
1215 |
+
},
|
1216 |
+
{
|
1217 |
+
"epoch": 1.57,
|
1218 |
+
"learning_rate": 0.00040500000000000003,
|
1219 |
+
"loss": 2.8196,
|
1220 |
+
"step": 1620
|
1221 |
+
},
|
1222 |
+
{
|
1223 |
+
"epoch": 1.58,
|
1224 |
+
"learning_rate": 0.0004075,
|
1225 |
+
"loss": 2.6888,
|
1226 |
+
"step": 1630
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 1.59,
|
1230 |
+
"learning_rate": 0.00041,
|
1231 |
+
"loss": 2.8251,
|
1232 |
+
"step": 1640
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 1.6,
|
1236 |
+
"learning_rate": 0.0004125,
|
1237 |
+
"loss": 2.8098,
|
1238 |
+
"step": 1650
|
1239 |
+
},
|
1240 |
+
{
|
1241 |
+
"epoch": 1.61,
|
1242 |
+
"learning_rate": 0.000415,
|
1243 |
+
"loss": 2.8685,
|
1244 |
+
"step": 1660
|
1245 |
+
},
|
1246 |
+
{
|
1247 |
+
"epoch": 1.62,
|
1248 |
+
"learning_rate": 0.0004175,
|
1249 |
+
"loss": 2.7419,
|
1250 |
+
"step": 1670
|
1251 |
+
},
|
1252 |
+
{
|
1253 |
+
"epoch": 1.63,
|
1254 |
+
"learning_rate": 0.00042,
|
1255 |
+
"loss": 2.8122,
|
1256 |
+
"step": 1680
|
1257 |
+
},
|
1258 |
+
{
|
1259 |
+
"epoch": 1.64,
|
1260 |
+
"learning_rate": 0.00042249999999999997,
|
1261 |
+
"loss": 2.8728,
|
1262 |
+
"step": 1690
|
1263 |
+
},
|
1264 |
+
{
|
1265 |
+
"epoch": 1.64,
|
1266 |
+
"learning_rate": 0.000425,
|
1267 |
+
"loss": 2.8772,
|
1268 |
+
"step": 1700
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 1.65,
|
1272 |
+
"learning_rate": 0.0004275,
|
1273 |
+
"loss": 2.7859,
|
1274 |
+
"step": 1710
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 1.66,
|
1278 |
+
"learning_rate": 0.00043,
|
1279 |
+
"loss": 2.8488,
|
1280 |
+
"step": 1720
|
1281 |
+
},
|
1282 |
+
{
|
1283 |
+
"epoch": 1.67,
|
1284 |
+
"learning_rate": 0.0004325,
|
1285 |
+
"loss": 2.6726,
|
1286 |
+
"step": 1730
|
1287 |
+
},
|
1288 |
+
{
|
1289 |
+
"epoch": 1.68,
|
1290 |
+
"learning_rate": 0.000435,
|
1291 |
+
"loss": 2.8862,
|
1292 |
+
"step": 1740
|
1293 |
+
},
|
1294 |
+
{
|
1295 |
+
"epoch": 1.69,
|
1296 |
+
"learning_rate": 0.0004375,
|
1297 |
+
"loss": 2.7678,
|
1298 |
+
"step": 1750
|
1299 |
+
},
|
1300 |
+
{
|
1301 |
+
"epoch": 1.7,
|
1302 |
+
"learning_rate": 0.00044,
|
1303 |
+
"loss": 2.8049,
|
1304 |
+
"step": 1760
|
1305 |
+
},
|
1306 |
+
{
|
1307 |
+
"epoch": 1.71,
|
1308 |
+
"learning_rate": 0.0004425,
|
1309 |
+
"loss": 2.7327,
|
1310 |
+
"step": 1770
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 1.72,
|
1314 |
+
"learning_rate": 0.00044500000000000003,
|
1315 |
+
"loss": 2.791,
|
1316 |
+
"step": 1780
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 1.73,
|
1320 |
+
"learning_rate": 0.00044750000000000004,
|
1321 |
+
"loss": 2.7355,
|
1322 |
+
"step": 1790
|
1323 |
+
},
|
1324 |
+
{
|
1325 |
+
"epoch": 1.74,
|
1326 |
+
"learning_rate": 0.00045000000000000004,
|
1327 |
+
"loss": 2.7519,
|
1328 |
+
"step": 1800
|
1329 |
+
},
|
1330 |
+
{
|
1331 |
+
"epoch": 1.74,
|
1332 |
+
"eval_bp": 1.0,
|
1333 |
+
"eval_counts": [
|
1334 |
+
22400,
|
1335 |
+
13234,
|
1336 |
+
8438,
|
1337 |
+
5686
|
1338 |
+
],
|
1339 |
+
"eval_loss": 1.6596581935882568,
|
1340 |
+
"eval_precisions": [
|
1341 |
+
51.272660684856255,
|
1342 |
+
32.89993784959602,
|
1343 |
+
22.93246364995244,
|
1344 |
+
17.015291618038724
|
1345 |
+
],
|
1346 |
+
"eval_ref_len": 39265,
|
1347 |
+
"eval_runtime": 755.4171,
|
1348 |
+
"eval_samples_per_second": 4.584,
|
1349 |
+
"eval_score": 28.483461117844286,
|
1350 |
+
"eval_steps_per_second": 4.584,
|
1351 |
+
"eval_sys_len": 43688,
|
1352 |
+
"eval_totals": [
|
1353 |
+
43688,
|
1354 |
+
40225,
|
1355 |
+
36795,
|
1356 |
+
33417
|
1357 |
+
],
|
1358 |
+
"step": 1800
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 1.75,
|
1362 |
+
"learning_rate": 0.00045250000000000005,
|
1363 |
+
"loss": 2.8522,
|
1364 |
+
"step": 1810
|
1365 |
+
},
|
1366 |
+
{
|
1367 |
+
"epoch": 1.76,
|
1368 |
+
"learning_rate": 0.000455,
|
1369 |
+
"loss": 2.8933,
|
1370 |
+
"step": 1820
|
1371 |
+
},
|
1372 |
+
{
|
1373 |
+
"epoch": 1.77,
|
1374 |
+
"learning_rate": 0.0004575,
|
1375 |
+
"loss": 2.7223,
|
1376 |
+
"step": 1830
|
1377 |
+
},
|
1378 |
+
{
|
1379 |
+
"epoch": 1.78,
|
1380 |
+
"learning_rate": 0.00046,
|
1381 |
+
"loss": 2.7376,
|
1382 |
+
"step": 1840
|
1383 |
+
},
|
1384 |
+
{
|
1385 |
+
"epoch": 1.79,
|
1386 |
+
"learning_rate": 0.0004625,
|
1387 |
+
"loss": 2.6918,
|
1388 |
+
"step": 1850
|
1389 |
+
},
|
1390 |
+
{
|
1391 |
+
"epoch": 1.8,
|
1392 |
+
"learning_rate": 0.000465,
|
1393 |
+
"loss": 2.8248,
|
1394 |
+
"step": 1860
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 1.81,
|
1398 |
+
"learning_rate": 0.00046750000000000003,
|
1399 |
+
"loss": 2.8406,
|
1400 |
+
"step": 1870
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 1.82,
|
1404 |
+
"learning_rate": 0.00047,
|
1405 |
+
"loss": 2.6332,
|
1406 |
+
"step": 1880
|
1407 |
+
},
|
1408 |
+
{
|
1409 |
+
"epoch": 1.83,
|
1410 |
+
"learning_rate": 0.0004725,
|
1411 |
+
"loss": 2.8686,
|
1412 |
+
"step": 1890
|
1413 |
+
},
|
1414 |
+
{
|
1415 |
+
"epoch": 1.84,
|
1416 |
+
"learning_rate": 0.000475,
|
1417 |
+
"loss": 2.5939,
|
1418 |
+
"step": 1900
|
1419 |
+
},
|
1420 |
+
{
|
1421 |
+
"epoch": 1.85,
|
1422 |
+
"learning_rate": 0.0004775,
|
1423 |
+
"loss": 2.7392,
|
1424 |
+
"step": 1910
|
1425 |
+
},
|
1426 |
+
{
|
1427 |
+
"epoch": 1.86,
|
1428 |
+
"learning_rate": 0.00048,
|
1429 |
+
"loss": 2.7951,
|
1430 |
+
"step": 1920
|
1431 |
+
},
|
1432 |
+
{
|
1433 |
+
"epoch": 1.87,
|
1434 |
+
"learning_rate": 0.0004825,
|
1435 |
+
"loss": 2.7636,
|
1436 |
+
"step": 1930
|
1437 |
+
},
|
1438 |
+
{
|
1439 |
+
"epoch": 1.88,
|
1440 |
+
"learning_rate": 0.00048499999999999997,
|
1441 |
+
"loss": 2.8033,
|
1442 |
+
"step": 1940
|
1443 |
+
},
|
1444 |
+
{
|
1445 |
+
"epoch": 1.89,
|
1446 |
+
"learning_rate": 0.0004875,
|
1447 |
+
"loss": 2.7528,
|
1448 |
+
"step": 1950
|
1449 |
+
},
|
1450 |
+
{
|
1451 |
+
"epoch": 1.9,
|
1452 |
+
"learning_rate": 0.00049,
|
1453 |
+
"loss": 2.71,
|
1454 |
+
"step": 1960
|
1455 |
+
},
|
1456 |
+
{
|
1457 |
+
"epoch": 1.91,
|
1458 |
+
"learning_rate": 0.0004925,
|
1459 |
+
"loss": 2.7453,
|
1460 |
+
"step": 1970
|
1461 |
+
},
|
1462 |
+
{
|
1463 |
+
"epoch": 1.92,
|
1464 |
+
"learning_rate": 0.000495,
|
1465 |
+
"loss": 2.8172,
|
1466 |
+
"step": 1980
|
1467 |
+
},
|
1468 |
+
{
|
1469 |
+
"epoch": 1.93,
|
1470 |
+
"learning_rate": 0.0004975,
|
1471 |
+
"loss": 2.8219,
|
1472 |
+
"step": 1990
|
1473 |
+
},
|
1474 |
+
{
|
1475 |
+
"epoch": 1.94,
|
1476 |
+
"learning_rate": 0.0005,
|
1477 |
+
"loss": 2.6677,
|
1478 |
+
"step": 2000
|
1479 |
+
},
|
1480 |
+
{
|
1481 |
+
"epoch": 1.94,
|
1482 |
+
"eval_bp": 1.0,
|
1483 |
+
"eval_counts": [
|
1484 |
+
22981,
|
1485 |
+
13820,
|
1486 |
+
8891,
|
1487 |
+
6002
|
1488 |
+
],
|
1489 |
+
"eval_loss": 1.6349472999572754,
|
1490 |
+
"eval_precisions": [
|
1491 |
+
57.158135601651495,
|
1492 |
+
37.612606482867484,
|
1493 |
+
26.690081652257444,
|
1494 |
+
20.05211813443806
|
1495 |
+
],
|
1496 |
+
"eval_ref_len": 39265,
|
1497 |
+
"eval_runtime": 477.7754,
|
1498 |
+
"eval_samples_per_second": 7.248,
|
1499 |
+
"eval_score": 32.751440814753295,
|
1500 |
+
"eval_steps_per_second": 7.248,
|
1501 |
+
"eval_sys_len": 40206,
|
1502 |
+
"eval_totals": [
|
1503 |
+
40206,
|
1504 |
+
36743,
|
1505 |
+
33312,
|
1506 |
+
29932
|
1507 |
+
],
|
1508 |
+
"step": 2000
|
1509 |
+
},
|
1510 |
+
{
|
1511 |
+
"epoch": 1.94,
|
1512 |
+
"learning_rate": 0.0005024999999999999,
|
1513 |
+
"loss": 2.7718,
|
1514 |
+
"step": 2010
|
1515 |
+
},
|
1516 |
+
{
|
1517 |
+
"epoch": 1.95,
|
1518 |
+
"learning_rate": 0.000505,
|
1519 |
+
"loss": 2.6684,
|
1520 |
+
"step": 2020
|
1521 |
+
},
|
1522 |
+
{
|
1523 |
+
"epoch": 1.96,
|
1524 |
+
"learning_rate": 0.0005074999999999999,
|
1525 |
+
"loss": 2.8251,
|
1526 |
+
"step": 2030
|
1527 |
+
},
|
1528 |
+
{
|
1529 |
+
"epoch": 1.97,
|
1530 |
+
"learning_rate": 0.00051,
|
1531 |
+
"loss": 2.6859,
|
1532 |
+
"step": 2040
|
1533 |
+
},
|
1534 |
+
{
|
1535 |
+
"epoch": 1.98,
|
1536 |
+
"learning_rate": 0.0005124999999999999,
|
1537 |
+
"loss": 2.7663,
|
1538 |
+
"step": 2050
|
1539 |
+
},
|
1540 |
+
{
|
1541 |
+
"epoch": 1.99,
|
1542 |
+
"learning_rate": 0.000515,
|
1543 |
+
"loss": 2.5372,
|
1544 |
+
"step": 2060
|
1545 |
+
},
|
1546 |
+
{
|
1547 |
+
"epoch": 2.0,
|
1548 |
+
"learning_rate": 0.0005175,
|
1549 |
+
"loss": 2.714,
|
1550 |
+
"step": 2070
|
1551 |
+
},
|
1552 |
+
{
|
1553 |
+
"epoch": 2.01,
|
1554 |
+
"learning_rate": 0.0005200000000000001,
|
1555 |
+
"loss": 2.4456,
|
1556 |
+
"step": 2080
|
1557 |
+
},
|
1558 |
+
{
|
1559 |
+
"epoch": 2.02,
|
1560 |
+
"learning_rate": 0.0005225,
|
1561 |
+
"loss": 2.4832,
|
1562 |
+
"step": 2090
|
1563 |
+
},
|
1564 |
+
{
|
1565 |
+
"epoch": 2.03,
|
1566 |
+
"learning_rate": 0.0005250000000000001,
|
1567 |
+
"loss": 2.498,
|
1568 |
+
"step": 2100
|
1569 |
+
},
|
1570 |
+
{
|
1571 |
+
"epoch": 2.04,
|
1572 |
+
"learning_rate": 0.0005275,
|
1573 |
+
"loss": 2.5938,
|
1574 |
+
"step": 2110
|
1575 |
+
},
|
1576 |
+
{
|
1577 |
+
"epoch": 2.05,
|
1578 |
+
"learning_rate": 0.0005300000000000001,
|
1579 |
+
"loss": 2.5316,
|
1580 |
+
"step": 2120
|
1581 |
+
},
|
1582 |
+
{
|
1583 |
+
"epoch": 2.06,
|
1584 |
+
"learning_rate": 0.0005325,
|
1585 |
+
"loss": 2.47,
|
1586 |
+
"step": 2130
|
1587 |
+
},
|
1588 |
+
{
|
1589 |
+
"epoch": 2.07,
|
1590 |
+
"learning_rate": 0.000535,
|
1591 |
+
"loss": 2.3812,
|
1592 |
+
"step": 2140
|
1593 |
+
},
|
1594 |
+
{
|
1595 |
+
"epoch": 2.08,
|
1596 |
+
"learning_rate": 0.0005375,
|
1597 |
+
"loss": 2.528,
|
1598 |
+
"step": 2150
|
1599 |
+
},
|
1600 |
+
{
|
1601 |
+
"epoch": 2.09,
|
1602 |
+
"learning_rate": 0.00054,
|
1603 |
+
"loss": 2.4435,
|
1604 |
+
"step": 2160
|
1605 |
+
},
|
1606 |
+
{
|
1607 |
+
"epoch": 2.1,
|
1608 |
+
"learning_rate": 0.0005425,
|
1609 |
+
"loss": 2.4662,
|
1610 |
+
"step": 2170
|
1611 |
+
},
|
1612 |
+
{
|
1613 |
+
"epoch": 2.11,
|
1614 |
+
"learning_rate": 0.000545,
|
1615 |
+
"loss": 2.4676,
|
1616 |
+
"step": 2180
|
1617 |
+
},
|
1618 |
+
{
|
1619 |
+
"epoch": 2.12,
|
1620 |
+
"learning_rate": 0.0005475,
|
1621 |
+
"loss": 2.5348,
|
1622 |
+
"step": 2190
|
1623 |
+
},
|
1624 |
+
{
|
1625 |
+
"epoch": 2.13,
|
1626 |
+
"learning_rate": 0.00055,
|
1627 |
+
"loss": 2.4479,
|
1628 |
+
"step": 2200
|
1629 |
+
},
|
1630 |
+
{
|
1631 |
+
"epoch": 2.13,
|
1632 |
+
"eval_bp": 1.0,
|
1633 |
+
"eval_counts": [
|
1634 |
+
24046,
|
1635 |
+
14704,
|
1636 |
+
9579,
|
1637 |
+
6510
|
1638 |
+
],
|
1639 |
+
"eval_loss": 1.636635661125183,
|
1640 |
+
"eval_precisions": [
|
1641 |
+
61.04905047222504,
|
1642 |
+
40.929714683368125,
|
1643 |
+
29.480195734465884,
|
1644 |
+
22.361144505890838
|
1645 |
+
],
|
1646 |
+
"eval_ref_len": 39265,
|
1647 |
+
"eval_runtime": 347.8604,
|
1648 |
+
"eval_samples_per_second": 9.955,
|
1649 |
+
"eval_score": 35.824942780304305,
|
1650 |
+
"eval_steps_per_second": 9.955,
|
1651 |
+
"eval_sys_len": 39388,
|
1652 |
+
"eval_totals": [
|
1653 |
+
39388,
|
1654 |
+
35925,
|
1655 |
+
32493,
|
1656 |
+
29113
|
1657 |
+
],
|
1658 |
+
"step": 2200
|
1659 |
+
},
|
1660 |
+
{
|
1661 |
+
"epoch": 2.14,
|
1662 |
+
"learning_rate": 0.0005525,
|
1663 |
+
"loss": 2.4267,
|
1664 |
+
"step": 2210
|
1665 |
+
},
|
1666 |
+
{
|
1667 |
+
"epoch": 2.15,
|
1668 |
+
"learning_rate": 0.000555,
|
1669 |
+
"loss": 2.3329,
|
1670 |
+
"step": 2220
|
1671 |
+
},
|
1672 |
+
{
|
1673 |
+
"epoch": 2.16,
|
1674 |
+
"learning_rate": 0.0005575,
|
1675 |
+
"loss": 2.531,
|
1676 |
+
"step": 2230
|
1677 |
+
},
|
1678 |
+
{
|
1679 |
+
"epoch": 2.17,
|
1680 |
+
"learning_rate": 0.0005600000000000001,
|
1681 |
+
"loss": 2.5151,
|
1682 |
+
"step": 2240
|
1683 |
+
},
|
1684 |
+
{
|
1685 |
+
"epoch": 2.18,
|
1686 |
+
"learning_rate": 0.0005625000000000001,
|
1687 |
+
"loss": 2.5409,
|
1688 |
+
"step": 2250
|
1689 |
+
},
|
1690 |
+
{
|
1691 |
+
"epoch": 2.19,
|
1692 |
+
"learning_rate": 0.000565,
|
1693 |
+
"loss": 2.4443,
|
1694 |
+
"step": 2260
|
1695 |
+
},
|
1696 |
+
{
|
1697 |
+
"epoch": 2.2,
|
1698 |
+
"learning_rate": 0.0005675,
|
1699 |
+
"loss": 2.4513,
|
1700 |
+
"step": 2270
|
1701 |
+
},
|
1702 |
+
{
|
1703 |
+
"epoch": 2.21,
|
1704 |
+
"learning_rate": 0.00057,
|
1705 |
+
"loss": 2.385,
|
1706 |
+
"step": 2280
|
1707 |
+
},
|
1708 |
+
{
|
1709 |
+
"epoch": 2.22,
|
1710 |
+
"learning_rate": 0.0005725,
|
1711 |
+
"loss": 2.4345,
|
1712 |
+
"step": 2290
|
1713 |
+
},
|
1714 |
+
{
|
1715 |
+
"epoch": 2.23,
|
1716 |
+
"learning_rate": 0.000575,
|
1717 |
+
"loss": 2.5575,
|
1718 |
+
"step": 2300
|
1719 |
+
},
|
1720 |
+
{
|
1721 |
+
"epoch": 2.24,
|
1722 |
+
"learning_rate": 0.0005775,
|
1723 |
+
"loss": 2.4997,
|
1724 |
+
"step": 2310
|
1725 |
+
},
|
1726 |
+
{
|
1727 |
+
"epoch": 2.24,
|
1728 |
+
"learning_rate": 0.00058,
|
1729 |
+
"loss": 2.4726,
|
1730 |
+
"step": 2320
|
1731 |
+
},
|
1732 |
+
{
|
1733 |
+
"epoch": 2.25,
|
1734 |
+
"learning_rate": 0.0005825,
|
1735 |
+
"loss": 2.4253,
|
1736 |
+
"step": 2330
|
1737 |
+
},
|
1738 |
+
{
|
1739 |
+
"epoch": 2.26,
|
1740 |
+
"learning_rate": 0.000585,
|
1741 |
+
"loss": 2.4573,
|
1742 |
+
"step": 2340
|
1743 |
+
},
|
1744 |
+
{
|
1745 |
+
"epoch": 2.27,
|
1746 |
+
"learning_rate": 0.0005875,
|
1747 |
+
"loss": 2.4581,
|
1748 |
+
"step": 2350
|
1749 |
+
},
|
1750 |
+
{
|
1751 |
+
"epoch": 2.28,
|
1752 |
+
"learning_rate": 0.00059,
|
1753 |
+
"loss": 2.5404,
|
1754 |
+
"step": 2360
|
1755 |
+
},
|
1756 |
+
{
|
1757 |
+
"epoch": 2.29,
|
1758 |
+
"learning_rate": 0.0005925,
|
1759 |
+
"loss": 2.5728,
|
1760 |
+
"step": 2370
|
1761 |
+
},
|
1762 |
+
{
|
1763 |
+
"epoch": 2.3,
|
1764 |
+
"learning_rate": 0.0005949999999999999,
|
1765 |
+
"loss": 2.4639,
|
1766 |
+
"step": 2380
|
1767 |
+
},
|
1768 |
+
{
|
1769 |
+
"epoch": 2.31,
|
1770 |
+
"learning_rate": 0.0005975,
|
1771 |
+
"loss": 2.6251,
|
1772 |
+
"step": 2390
|
1773 |
+
},
|
1774 |
+
{
|
1775 |
+
"epoch": 2.32,
|
1776 |
+
"learning_rate": 0.0006,
|
1777 |
+
"loss": 2.4606,
|
1778 |
+
"step": 2400
|
1779 |
+
},
|
1780 |
+
{
|
1781 |
+
"epoch": 2.32,
|
1782 |
+
"eval_bp": 1.0,
|
1783 |
+
"eval_counts": [
|
1784 |
+
22981,
|
1785 |
+
13834,
|
1786 |
+
8940,
|
1787 |
+
6036
|
1788 |
+
],
|
1789 |
+
"eval_loss": 1.6156156063079834,
|
1790 |
+
"eval_precisions": [
|
1791 |
+
52.36283266496537,
|
1792 |
+
34.22139764996908,
|
1793 |
+
24.18438565167992,
|
1794 |
+
17.986233201227687
|
1795 |
+
],
|
1796 |
+
"eval_ref_len": 39265,
|
1797 |
+
"eval_runtime": 623.2248,
|
1798 |
+
"eval_samples_per_second": 5.557,
|
1799 |
+
"eval_score": 29.713173488652338,
|
1800 |
+
"eval_steps_per_second": 5.557,
|
1801 |
+
"eval_sys_len": 43888,
|
1802 |
+
"eval_totals": [
|
1803 |
+
43888,
|
1804 |
+
40425,
|
1805 |
+
36966,
|
1806 |
+
33559
|
1807 |
+
],
|
1808 |
+
"step": 2400
|
1809 |
+
},
|
1810 |
+
{
|
1811 |
+
"epoch": 2.33,
|
1812 |
+
"learning_rate": 0.0006025000000000001,
|
1813 |
+
"loss": 2.5344,
|
1814 |
+
"step": 2410
|
1815 |
+
},
|
1816 |
+
{
|
1817 |
+
"epoch": 2.34,
|
1818 |
+
"learning_rate": 0.000605,
|
1819 |
+
"loss": 2.5511,
|
1820 |
+
"step": 2420
|
1821 |
+
},
|
1822 |
+
{
|
1823 |
+
"epoch": 2.35,
|
1824 |
+
"learning_rate": 0.0006075000000000001,
|
1825 |
+
"loss": 2.4595,
|
1826 |
+
"step": 2430
|
1827 |
+
},
|
1828 |
+
{
|
1829 |
+
"epoch": 2.36,
|
1830 |
+
"learning_rate": 0.00061,
|
1831 |
+
"loss": 2.4675,
|
1832 |
+
"step": 2440
|
1833 |
+
},
|
1834 |
+
{
|
1835 |
+
"epoch": 2.37,
|
1836 |
+
"learning_rate": 0.0006125000000000001,
|
1837 |
+
"loss": 2.4186,
|
1838 |
+
"step": 2450
|
1839 |
+
},
|
1840 |
+
{
|
1841 |
+
"epoch": 2.38,
|
1842 |
+
"learning_rate": 0.000615,
|
1843 |
+
"loss": 2.3482,
|
1844 |
+
"step": 2460
|
1845 |
+
},
|
1846 |
+
{
|
1847 |
+
"epoch": 2.39,
|
1848 |
+
"learning_rate": 0.0006175000000000001,
|
1849 |
+
"loss": 2.4653,
|
1850 |
+
"step": 2470
|
1851 |
+
},
|
1852 |
+
{
|
1853 |
+
"epoch": 2.4,
|
1854 |
+
"learning_rate": 0.00062,
|
1855 |
+
"loss": 2.5461,
|
1856 |
+
"step": 2480
|
1857 |
+
},
|
1858 |
+
{
|
1859 |
+
"epoch": 2.41,
|
1860 |
+
"learning_rate": 0.0006225000000000001,
|
1861 |
+
"loss": 2.4725,
|
1862 |
+
"step": 2490
|
1863 |
+
},
|
1864 |
+
{
|
1865 |
+
"epoch": 2.42,
|
1866 |
+
"learning_rate": 0.000625,
|
1867 |
+
"loss": 2.4625,
|
1868 |
+
"step": 2500
|
1869 |
+
},
|
1870 |
+
{
|
1871 |
+
"epoch": 2.43,
|
1872 |
+
"learning_rate": 0.0006274999999999999,
|
1873 |
+
"loss": 2.576,
|
1874 |
+
"step": 2510
|
1875 |
+
},
|
1876 |
+
{
|
1877 |
+
"epoch": 2.44,
|
1878 |
+
"learning_rate": 0.00063,
|
1879 |
+
"loss": 2.5898,
|
1880 |
+
"step": 2520
|
1881 |
+
},
|
1882 |
+
{
|
1883 |
+
"epoch": 2.45,
|
1884 |
+
"learning_rate": 0.0006324999999999999,
|
1885 |
+
"loss": 2.4081,
|
1886 |
+
"step": 2530
|
1887 |
+
},
|
1888 |
+
{
|
1889 |
+
"epoch": 2.46,
|
1890 |
+
"learning_rate": 0.000635,
|
1891 |
+
"loss": 2.3564,
|
1892 |
+
"step": 2540
|
1893 |
+
},
|
1894 |
+
{
|
1895 |
+
"epoch": 2.47,
|
1896 |
+
"learning_rate": 0.0006374999999999999,
|
1897 |
+
"loss": 2.3773,
|
1898 |
+
"step": 2550
|
1899 |
+
},
|
1900 |
+
{
|
1901 |
+
"epoch": 2.48,
|
1902 |
+
"learning_rate": 0.00064,
|
1903 |
+
"loss": 2.4186,
|
1904 |
+
"step": 2560
|
1905 |
+
},
|
1906 |
+
{
|
1907 |
+
"epoch": 2.49,
|
1908 |
+
"learning_rate": 0.0006425,
|
1909 |
+
"loss": 2.4812,
|
1910 |
+
"step": 2570
|
1911 |
+
},
|
1912 |
+
{
|
1913 |
+
"epoch": 2.5,
|
1914 |
+
"learning_rate": 0.0006450000000000001,
|
1915 |
+
"loss": 2.5489,
|
1916 |
+
"step": 2580
|
1917 |
+
},
|
1918 |
+
{
|
1919 |
+
"epoch": 2.51,
|
1920 |
+
"learning_rate": 0.0006475,
|
1921 |
+
"loss": 2.4855,
|
1922 |
+
"step": 2590
|
1923 |
+
},
|
1924 |
+
{
|
1925 |
+
"epoch": 2.52,
|
1926 |
+
"learning_rate": 0.0006500000000000001,
|
1927 |
+
"loss": 2.4259,
|
1928 |
+
"step": 2600
|
1929 |
+
},
|
1930 |
+
{
|
1931 |
+
"epoch": 2.52,
|
1932 |
+
"eval_bp": 1.0,
|
1933 |
+
"eval_counts": [
|
1934 |
+
24252,
|
1935 |
+
14861,
|
1936 |
+
9670,
|
1937 |
+
6560
|
1938 |
+
],
|
1939 |
+
"eval_loss": 1.5984951257705688,
|
1940 |
+
"eval_precisions": [
|
1941 |
+
60.03416095254598,
|
1942 |
+
40.23663832782802,
|
1943 |
+
28.86136397552604,
|
1944 |
+
21.777379411081235
|
1945 |
+
],
|
1946 |
+
"eval_ref_len": 39265,
|
1947 |
+
"eval_runtime": 371.9049,
|
1948 |
+
"eval_samples_per_second": 9.312,
|
1949 |
+
"eval_score": 35.10230625697626,
|
1950 |
+
"eval_steps_per_second": 9.312,
|
1951 |
+
"eval_sys_len": 40397,
|
1952 |
+
"eval_totals": [
|
1953 |
+
40397,
|
1954 |
+
36934,
|
1955 |
+
33505,
|
1956 |
+
30123
|
1957 |
+
],
|
1958 |
+
"step": 2600
|
1959 |
+
},
|
1960 |
+
{
|
1961 |
+
"epoch": 2.53,
|
1962 |
+
"learning_rate": 0.0006525,
|
1963 |
+
"loss": 2.4609,
|
1964 |
+
"step": 2610
|
1965 |
+
},
|
1966 |
+
{
|
1967 |
+
"epoch": 2.54,
|
1968 |
+
"learning_rate": 0.0006550000000000001,
|
1969 |
+
"loss": 2.4513,
|
1970 |
+
"step": 2620
|
1971 |
+
},
|
1972 |
+
{
|
1973 |
+
"epoch": 2.54,
|
1974 |
+
"learning_rate": 0.0006575,
|
1975 |
+
"loss": 2.4666,
|
1976 |
+
"step": 2630
|
1977 |
+
},
|
1978 |
+
{
|
1979 |
+
"epoch": 2.55,
|
1980 |
+
"learning_rate": 0.00066,
|
1981 |
+
"loss": 2.5221,
|
1982 |
+
"step": 2640
|
1983 |
+
},
|
1984 |
+
{
|
1985 |
+
"epoch": 2.56,
|
1986 |
+
"learning_rate": 0.0006625,
|
1987 |
+
"loss": 2.4292,
|
1988 |
+
"step": 2650
|
1989 |
+
},
|
1990 |
+
{
|
1991 |
+
"epoch": 2.57,
|
1992 |
+
"learning_rate": 0.000665,
|
1993 |
+
"loss": 2.4118,
|
1994 |
+
"step": 2660
|
1995 |
+
},
|
1996 |
+
{
|
1997 |
+
"epoch": 2.58,
|
1998 |
+
"learning_rate": 0.0006675,
|
1999 |
+
"loss": 2.5148,
|
2000 |
+
"step": 2670
|
2001 |
+
},
|
2002 |
+
{
|
2003 |
+
"epoch": 2.59,
|
2004 |
+
"learning_rate": 0.00067,
|
2005 |
+
"loss": 2.5603,
|
2006 |
+
"step": 2680
|
2007 |
+
},
|
2008 |
+
{
|
2009 |
+
"epoch": 2.6,
|
2010 |
+
"learning_rate": 0.0006725,
|
2011 |
+
"loss": 2.4776,
|
2012 |
+
"step": 2690
|
2013 |
+
},
|
2014 |
+
{
|
2015 |
+
"epoch": 2.61,
|
2016 |
+
"learning_rate": 0.000675,
|
2017 |
+
"loss": 2.4421,
|
2018 |
+
"step": 2700
|
2019 |
+
},
|
2020 |
+
{
|
2021 |
+
"epoch": 2.62,
|
2022 |
+
"learning_rate": 0.0006775,
|
2023 |
+
"loss": 2.5626,
|
2024 |
+
"step": 2710
|
2025 |
+
},
|
2026 |
+
{
|
2027 |
+
"epoch": 2.63,
|
2028 |
+
"learning_rate": 0.00068,
|
2029 |
+
"loss": 2.4886,
|
2030 |
+
"step": 2720
|
2031 |
+
},
|
2032 |
+
{
|
2033 |
+
"epoch": 2.64,
|
2034 |
+
"learning_rate": 0.0006825000000000001,
|
2035 |
+
"loss": 2.5124,
|
2036 |
+
"step": 2730
|
2037 |
+
},
|
2038 |
+
{
|
2039 |
+
"epoch": 2.65,
|
2040 |
+
"learning_rate": 0.0006850000000000001,
|
2041 |
+
"loss": 2.4808,
|
2042 |
+
"step": 2740
|
2043 |
+
},
|
2044 |
+
{
|
2045 |
+
"epoch": 2.66,
|
2046 |
+
"learning_rate": 0.0006875,
|
2047 |
+
"loss": 2.4922,
|
2048 |
+
"step": 2750
|
2049 |
+
},
|
2050 |
+
{
|
2051 |
+
"epoch": 2.67,
|
2052 |
+
"learning_rate": 0.00069,
|
2053 |
+
"loss": 2.5108,
|
2054 |
+
"step": 2760
|
2055 |
+
},
|
2056 |
+
{
|
2057 |
+
"epoch": 2.68,
|
2058 |
+
"learning_rate": 0.0006925,
|
2059 |
+
"loss": 2.4229,
|
2060 |
+
"step": 2770
|
2061 |
+
},
|
2062 |
+
{
|
2063 |
+
"epoch": 2.69,
|
2064 |
+
"learning_rate": 0.000695,
|
2065 |
+
"loss": 2.3996,
|
2066 |
+
"step": 2780
|
2067 |
+
},
|
2068 |
+
{
|
2069 |
+
"epoch": 2.7,
|
2070 |
+
"learning_rate": 0.0006975,
|
2071 |
+
"loss": 2.56,
|
2072 |
+
"step": 2790
|
2073 |
+
},
|
2074 |
+
{
|
2075 |
+
"epoch": 2.71,
|
2076 |
+
"learning_rate": 0.0007,
|
2077 |
+
"loss": 2.4794,
|
2078 |
+
"step": 2800
|
2079 |
+
},
|
2080 |
+
{
|
2081 |
+
"epoch": 2.71,
|
2082 |
+
"eval_bp": 1.0,
|
2083 |
+
"eval_counts": [
|
2084 |
+
24440,
|
2085 |
+
15095,
|
2086 |
+
9863,
|
2087 |
+
6691
|
2088 |
+
],
|
2089 |
+
"eval_loss": 1.5756924152374268,
|
2090 |
+
"eval_precisions": [
|
2091 |
+
61.18412817624233,
|
2092 |
+
41.3765692670358,
|
2093 |
+
29.839954013251443,
|
2094 |
+
22.549118727462677
|
2095 |
+
],
|
2096 |
+
"eval_ref_len": 39265,
|
2097 |
+
"eval_runtime": 376.0541,
|
2098 |
+
"eval_samples_per_second": 9.209,
|
2099 |
+
"eval_score": 36.12686234680497,
|
2100 |
+
"eval_steps_per_second": 9.209,
|
2101 |
+
"eval_sys_len": 39945,
|
2102 |
+
"eval_totals": [
|
2103 |
+
39945,
|
2104 |
+
36482,
|
2105 |
+
33053,
|
2106 |
+
29673
|
2107 |
+
],
|
2108 |
+
"step": 2800
|
2109 |
+
}
|
2110 |
+
],
|
2111 |
+
"max_steps": 258250,
|
2112 |
+
"num_train_epochs": 250,
|
2113 |
+
"total_flos": 4.09482327561216e+16,
|
2114 |
+
"trial_name": null,
|
2115 |
+
"trial_params": null
|
2116 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:15e71db004ba94bbaa617909246b92dc183cf2601d5808d60dc7f9a84728ba7e
|
3 |
+
size 3771
|