Training in progress, step 1000
Browse files- checkpoint-1000/config.json +28 -0
- checkpoint-1000/optimizer.pt +3 -0
- checkpoint-1000/pytorch_model.bin +3 -0
- checkpoint-1000/rng_state_0.pth +3 -0
- checkpoint-1000/rng_state_1.pth +3 -0
- checkpoint-1000/rng_state_2.pth +3 -0
- checkpoint-1000/rng_state_3.pth +3 -0
- checkpoint-1000/rng_state_4.pth +3 -0
- checkpoint-1000/rng_state_5.pth +3 -0
- checkpoint-1000/rng_state_6.pth +3 -0
- checkpoint-1000/rng_state_7.pth +3 -0
- checkpoint-1000/scheduler.pt +3 -0
- checkpoint-1000/special_tokens_map.json +7 -0
- checkpoint-1000/tokenizer.json +0 -0
- checkpoint-1000/tokenizer_config.json +5 -0
- checkpoint-1000/trainer_state.json +916 -0
- checkpoint-1000/training_args.bin +3 -0
- config.json +28 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +5 -0
- training_args.bin +3 -0
checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MarindaForMaskedLMWithEdgePrediction"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"ep_add_linear_projection": true,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-12,
|
| 16 |
+
"max_position_embeddings": 600,
|
| 17 |
+
"max_relative_position_embeddings": 8,
|
| 18 |
+
"model_type": "marinda",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"pad_token_id": 3,
|
| 22 |
+
"position_embedding_type": "mixed",
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.31.0",
|
| 25 |
+
"type_vocab_size": 2,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 30523
|
| 28 |
+
}
|
checkpoint-1000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ff3279f7f4b54ae78bafb4afa01e6737439354771a6133bbfd941639af78beb
|
| 3 |
+
size 881521797
|
checkpoint-1000/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9539754b990c6c28ccf1a024ed909035c3d57a6c0f4c419ed60ff426fbc46aeb
|
| 3 |
+
size 440772077
|
checkpoint-1000/rng_state_0.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12a3c6dca4ae62e368af0d7501da4ee3ab0ac9af55aebecd5b4305f60e5c2019
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_1.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa281aab3e9da4bc4394e3daa7fe253a18799dced538e66e5e35c9c8b8f1f395
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_2.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:add2a8df2959eda1af3433c8d096b0f9608e2c847a8c4e88c9a4092403a08d01
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_3.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2260df16324f32d7c8a878e83e8f3b6f5c2271b9dc84d07134dd73e3031bc5db
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_4.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f5b457094124cabbb87c9e2ddecbb7b1bf59a1aaae372f8f8a6aad6499f01716
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_5.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e9c72b754ca86ca0c38db4ba2d9fb0f69e1a131f793037434ad4aaf0c546e005
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_6.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:884e10adb976b8aa2b6ae531da6f3b5e402c4ebbfc5786447ddbb27eae7094c7
|
| 3 |
+
size 21623
|
checkpoint-1000/rng_state_7.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b915c94753593835054bc1ba8c88310b3c297cef03b886e37b5a0fb072640a66
|
| 3 |
+
size 21623
|
checkpoint-1000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63f0f3af38f5596ccfeb09abcc7390e128282513bc1203c8547b4d1d43ea8806
|
| 3 |
+
size 627
|
checkpoint-1000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
checkpoint-1000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"model_max_length": 512,
|
| 4 |
+
"tokenizer_class": "MarindaTokenizer"
|
| 5 |
+
}
|
checkpoint-1000/trainer_state.json
ADDED
|
@@ -0,0 +1,916 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7713073659853451,
|
| 5 |
+
"global_step": 1000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"ep_loss": 0.0,
|
| 12 |
+
"epoch": 0.01,
|
| 13 |
+
"learning_rate": 5e-07,
|
| 14 |
+
"loss": 10.4763,
|
| 15 |
+
"mlm_loss": 10.4763,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"ep_loss": 0.0,
|
| 20 |
+
"epoch": 0.02,
|
| 21 |
+
"learning_rate": 1e-06,
|
| 22 |
+
"loss": 10.3052,
|
| 23 |
+
"mlm_loss": 10.3052,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"ep_loss": 0.0,
|
| 28 |
+
"epoch": 0.02,
|
| 29 |
+
"learning_rate": 1.5e-06,
|
| 30 |
+
"loss": 9.9641,
|
| 31 |
+
"mlm_loss": 9.9641,
|
| 32 |
+
"step": 30
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"ep_loss": 0.0,
|
| 36 |
+
"epoch": 0.03,
|
| 37 |
+
"learning_rate": 2e-06,
|
| 38 |
+
"loss": 9.5354,
|
| 39 |
+
"mlm_loss": 9.5354,
|
| 40 |
+
"step": 40
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"ep_loss": 0.0,
|
| 44 |
+
"epoch": 0.04,
|
| 45 |
+
"learning_rate": 2.5e-06,
|
| 46 |
+
"loss": 9.1346,
|
| 47 |
+
"mlm_loss": 9.1346,
|
| 48 |
+
"step": 50
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"ep_loss": 0.0,
|
| 52 |
+
"epoch": 0.05,
|
| 53 |
+
"learning_rate": 3e-06,
|
| 54 |
+
"loss": 8.8162,
|
| 55 |
+
"mlm_loss": 8.8162,
|
| 56 |
+
"step": 60
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"ep_loss": 0.0,
|
| 60 |
+
"epoch": 0.05,
|
| 61 |
+
"learning_rate": 3.5e-06,
|
| 62 |
+
"loss": 8.558,
|
| 63 |
+
"mlm_loss": 8.558,
|
| 64 |
+
"step": 70
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"ep_loss": 0.0,
|
| 68 |
+
"epoch": 0.06,
|
| 69 |
+
"learning_rate": 4e-06,
|
| 70 |
+
"loss": 8.3459,
|
| 71 |
+
"mlm_loss": 8.3459,
|
| 72 |
+
"step": 80
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"ep_loss": 0.0,
|
| 76 |
+
"epoch": 0.07,
|
| 77 |
+
"learning_rate": 4.5e-06,
|
| 78 |
+
"loss": 8.175,
|
| 79 |
+
"mlm_loss": 8.175,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"ep_loss": 0.0,
|
| 84 |
+
"epoch": 0.08,
|
| 85 |
+
"learning_rate": 5e-06,
|
| 86 |
+
"loss": 8.0255,
|
| 87 |
+
"mlm_loss": 8.0255,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.08,
|
| 92 |
+
"eval_ep_loss": -0.14438027143478394,
|
| 93 |
+
"eval_loss": 7.826408386230469,
|
| 94 |
+
"eval_mlm_loss": 7.826408386230469,
|
| 95 |
+
"eval_runtime": 63.5165,
|
| 96 |
+
"eval_samples_per_second": 1100.014,
|
| 97 |
+
"eval_steps_per_second": 0.551,
|
| 98 |
+
"step": 100
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"ep_loss": 0.0,
|
| 102 |
+
"epoch": 0.08,
|
| 103 |
+
"learning_rate": 5.5e-06,
|
| 104 |
+
"loss": 7.8979,
|
| 105 |
+
"mlm_loss": 7.8979,
|
| 106 |
+
"step": 110
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"ep_loss": 0.0,
|
| 110 |
+
"epoch": 0.09,
|
| 111 |
+
"learning_rate": 6e-06,
|
| 112 |
+
"loss": 7.7543,
|
| 113 |
+
"mlm_loss": 7.7543,
|
| 114 |
+
"step": 120
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"ep_loss": 0.0,
|
| 118 |
+
"epoch": 0.1,
|
| 119 |
+
"learning_rate": 6.5e-06,
|
| 120 |
+
"loss": 7.5942,
|
| 121 |
+
"mlm_loss": 7.5942,
|
| 122 |
+
"step": 130
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"ep_loss": 0.0,
|
| 126 |
+
"epoch": 0.11,
|
| 127 |
+
"learning_rate": 7e-06,
|
| 128 |
+
"loss": 7.4304,
|
| 129 |
+
"mlm_loss": 7.4304,
|
| 130 |
+
"step": 140
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"ep_loss": 0.0,
|
| 134 |
+
"epoch": 0.12,
|
| 135 |
+
"learning_rate": 7.5e-06,
|
| 136 |
+
"loss": 7.1862,
|
| 137 |
+
"mlm_loss": 7.1862,
|
| 138 |
+
"step": 150
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"ep_loss": 0.0,
|
| 142 |
+
"epoch": 0.12,
|
| 143 |
+
"learning_rate": 8e-06,
|
| 144 |
+
"loss": 6.924,
|
| 145 |
+
"mlm_loss": 6.924,
|
| 146 |
+
"step": 160
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"ep_loss": 0.0,
|
| 150 |
+
"epoch": 0.13,
|
| 151 |
+
"learning_rate": 8.500000000000002e-06,
|
| 152 |
+
"loss": 6.6521,
|
| 153 |
+
"mlm_loss": 6.6521,
|
| 154 |
+
"step": 170
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"ep_loss": 0.0,
|
| 158 |
+
"epoch": 0.14,
|
| 159 |
+
"learning_rate": 9e-06,
|
| 160 |
+
"loss": 6.3639,
|
| 161 |
+
"mlm_loss": 6.3639,
|
| 162 |
+
"step": 180
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"ep_loss": 0.0,
|
| 166 |
+
"epoch": 0.15,
|
| 167 |
+
"learning_rate": 9.5e-06,
|
| 168 |
+
"loss": 6.0915,
|
| 169 |
+
"mlm_loss": 6.0915,
|
| 170 |
+
"step": 190
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"ep_loss": 0.0,
|
| 174 |
+
"epoch": 0.15,
|
| 175 |
+
"learning_rate": 1e-05,
|
| 176 |
+
"loss": 5.8272,
|
| 177 |
+
"mlm_loss": 5.8272,
|
| 178 |
+
"step": 200
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.15,
|
| 182 |
+
"eval_ep_loss": -0.4913683831691742,
|
| 183 |
+
"eval_loss": 5.531888008117676,
|
| 184 |
+
"eval_mlm_loss": 5.531888008117676,
|
| 185 |
+
"eval_runtime": 60.9938,
|
| 186 |
+
"eval_samples_per_second": 1145.51,
|
| 187 |
+
"eval_steps_per_second": 0.574,
|
| 188 |
+
"step": 200
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"ep_loss": 0.0,
|
| 192 |
+
"epoch": 0.16,
|
| 193 |
+
"learning_rate": 1.0500000000000001e-05,
|
| 194 |
+
"loss": 5.5749,
|
| 195 |
+
"mlm_loss": 5.5749,
|
| 196 |
+
"step": 210
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"ep_loss": 0.0,
|
| 200 |
+
"epoch": 0.17,
|
| 201 |
+
"learning_rate": 1.1e-05,
|
| 202 |
+
"loss": 5.3413,
|
| 203 |
+
"mlm_loss": 5.3413,
|
| 204 |
+
"step": 220
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"ep_loss": 0.0,
|
| 208 |
+
"epoch": 0.18,
|
| 209 |
+
"learning_rate": 1.15e-05,
|
| 210 |
+
"loss": 5.142,
|
| 211 |
+
"mlm_loss": 5.142,
|
| 212 |
+
"step": 230
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"ep_loss": 0.0,
|
| 216 |
+
"epoch": 0.19,
|
| 217 |
+
"learning_rate": 1.2e-05,
|
| 218 |
+
"loss": 4.925,
|
| 219 |
+
"mlm_loss": 4.925,
|
| 220 |
+
"step": 240
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"ep_loss": 0.0,
|
| 224 |
+
"epoch": 0.19,
|
| 225 |
+
"learning_rate": 1.25e-05,
|
| 226 |
+
"loss": 4.7334,
|
| 227 |
+
"mlm_loss": 4.7334,
|
| 228 |
+
"step": 250
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"ep_loss": 0.0,
|
| 232 |
+
"epoch": 0.2,
|
| 233 |
+
"learning_rate": 1.3e-05,
|
| 234 |
+
"loss": 4.5702,
|
| 235 |
+
"mlm_loss": 4.5702,
|
| 236 |
+
"step": 260
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"ep_loss": 0.0,
|
| 240 |
+
"epoch": 0.21,
|
| 241 |
+
"learning_rate": 1.35e-05,
|
| 242 |
+
"loss": 4.4388,
|
| 243 |
+
"mlm_loss": 4.4388,
|
| 244 |
+
"step": 270
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"ep_loss": 0.0,
|
| 248 |
+
"epoch": 0.22,
|
| 249 |
+
"learning_rate": 1.4e-05,
|
| 250 |
+
"loss": 4.288,
|
| 251 |
+
"mlm_loss": 4.288,
|
| 252 |
+
"step": 280
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"ep_loss": 0.0,
|
| 256 |
+
"epoch": 0.22,
|
| 257 |
+
"learning_rate": 1.4500000000000002e-05,
|
| 258 |
+
"loss": 4.1403,
|
| 259 |
+
"mlm_loss": 4.1403,
|
| 260 |
+
"step": 290
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"ep_loss": 0.0,
|
| 264 |
+
"epoch": 0.23,
|
| 265 |
+
"learning_rate": 1.5e-05,
|
| 266 |
+
"loss": 4.0232,
|
| 267 |
+
"mlm_loss": 4.0232,
|
| 268 |
+
"step": 300
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.23,
|
| 272 |
+
"eval_ep_loss": -1.1010401248931885,
|
| 273 |
+
"eval_loss": 3.8623688220977783,
|
| 274 |
+
"eval_mlm_loss": 3.8623688220977783,
|
| 275 |
+
"eval_runtime": 61.8594,
|
| 276 |
+
"eval_samples_per_second": 1129.481,
|
| 277 |
+
"eval_steps_per_second": 0.566,
|
| 278 |
+
"step": 300
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"ep_loss": 0.0,
|
| 282 |
+
"epoch": 0.24,
|
| 283 |
+
"learning_rate": 1.55e-05,
|
| 284 |
+
"loss": 3.9135,
|
| 285 |
+
"mlm_loss": 3.9135,
|
| 286 |
+
"step": 310
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"ep_loss": 0.0,
|
| 290 |
+
"epoch": 0.25,
|
| 291 |
+
"learning_rate": 1.6e-05,
|
| 292 |
+
"loss": 3.8252,
|
| 293 |
+
"mlm_loss": 3.8252,
|
| 294 |
+
"step": 320
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"ep_loss": 0.0,
|
| 298 |
+
"epoch": 0.25,
|
| 299 |
+
"learning_rate": 1.65e-05,
|
| 300 |
+
"loss": 3.7462,
|
| 301 |
+
"mlm_loss": 3.7462,
|
| 302 |
+
"step": 330
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"ep_loss": 0.0,
|
| 306 |
+
"epoch": 0.26,
|
| 307 |
+
"learning_rate": 1.7000000000000003e-05,
|
| 308 |
+
"loss": 3.6491,
|
| 309 |
+
"mlm_loss": 3.6491,
|
| 310 |
+
"step": 340
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"ep_loss": 0.0,
|
| 314 |
+
"epoch": 0.27,
|
| 315 |
+
"learning_rate": 1.7500000000000002e-05,
|
| 316 |
+
"loss": 3.5815,
|
| 317 |
+
"mlm_loss": 3.5815,
|
| 318 |
+
"step": 350
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"ep_loss": 0.0,
|
| 322 |
+
"epoch": 0.28,
|
| 323 |
+
"learning_rate": 1.8e-05,
|
| 324 |
+
"loss": 3.5211,
|
| 325 |
+
"mlm_loss": 3.5211,
|
| 326 |
+
"step": 360
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"ep_loss": 0.0,
|
| 330 |
+
"epoch": 0.29,
|
| 331 |
+
"learning_rate": 1.85e-05,
|
| 332 |
+
"loss": 3.452,
|
| 333 |
+
"mlm_loss": 3.452,
|
| 334 |
+
"step": 370
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"ep_loss": 0.0,
|
| 338 |
+
"epoch": 0.29,
|
| 339 |
+
"learning_rate": 1.9e-05,
|
| 340 |
+
"loss": 3.3747,
|
| 341 |
+
"mlm_loss": 3.3747,
|
| 342 |
+
"step": 380
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"ep_loss": 0.0,
|
| 346 |
+
"epoch": 0.3,
|
| 347 |
+
"learning_rate": 1.95e-05,
|
| 348 |
+
"loss": 3.3163,
|
| 349 |
+
"mlm_loss": 3.3163,
|
| 350 |
+
"step": 390
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"ep_loss": 0.0,
|
| 354 |
+
"epoch": 0.31,
|
| 355 |
+
"learning_rate": 2e-05,
|
| 356 |
+
"loss": 3.2679,
|
| 357 |
+
"mlm_loss": 3.2679,
|
| 358 |
+
"step": 400
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 0.31,
|
| 362 |
+
"eval_ep_loss": -1.4546582698822021,
|
| 363 |
+
"eval_loss": 3.1360955238342285,
|
| 364 |
+
"eval_mlm_loss": 3.1360955238342285,
|
| 365 |
+
"eval_runtime": 59.2888,
|
| 366 |
+
"eval_samples_per_second": 1178.451,
|
| 367 |
+
"eval_steps_per_second": 0.59,
|
| 368 |
+
"step": 400
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"ep_loss": 0.0,
|
| 372 |
+
"epoch": 0.32,
|
| 373 |
+
"learning_rate": 2.05e-05,
|
| 374 |
+
"loss": 3.1831,
|
| 375 |
+
"mlm_loss": 3.1831,
|
| 376 |
+
"step": 410
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"ep_loss": 0.0,
|
| 380 |
+
"epoch": 0.32,
|
| 381 |
+
"learning_rate": 2.1000000000000002e-05,
|
| 382 |
+
"loss": 3.1246,
|
| 383 |
+
"mlm_loss": 3.1246,
|
| 384 |
+
"step": 420
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"ep_loss": 0.0,
|
| 388 |
+
"epoch": 0.33,
|
| 389 |
+
"learning_rate": 2.1499999999999997e-05,
|
| 390 |
+
"loss": 3.0658,
|
| 391 |
+
"mlm_loss": 3.0658,
|
| 392 |
+
"step": 430
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"ep_loss": 0.0,
|
| 396 |
+
"epoch": 0.34,
|
| 397 |
+
"learning_rate": 2.2e-05,
|
| 398 |
+
"loss": 3.0032,
|
| 399 |
+
"mlm_loss": 3.0032,
|
| 400 |
+
"step": 440
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"ep_loss": 0.0,
|
| 404 |
+
"epoch": 0.35,
|
| 405 |
+
"learning_rate": 2.2499999999999998e-05,
|
| 406 |
+
"loss": 2.931,
|
| 407 |
+
"mlm_loss": 2.931,
|
| 408 |
+
"step": 450
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"ep_loss": 0.0,
|
| 412 |
+
"epoch": 0.35,
|
| 413 |
+
"learning_rate": 2.3e-05,
|
| 414 |
+
"loss": 2.904,
|
| 415 |
+
"mlm_loss": 2.904,
|
| 416 |
+
"step": 460
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"ep_loss": 0.0,
|
| 420 |
+
"epoch": 0.36,
|
| 421 |
+
"learning_rate": 2.3500000000000002e-05,
|
| 422 |
+
"loss": 2.8526,
|
| 423 |
+
"mlm_loss": 2.8526,
|
| 424 |
+
"step": 470
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"ep_loss": 0.0,
|
| 428 |
+
"epoch": 0.37,
|
| 429 |
+
"learning_rate": 2.4e-05,
|
| 430 |
+
"loss": 2.8336,
|
| 431 |
+
"mlm_loss": 2.8336,
|
| 432 |
+
"step": 480
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"ep_loss": 0.0,
|
| 436 |
+
"epoch": 0.38,
|
| 437 |
+
"learning_rate": 2.4500000000000003e-05,
|
| 438 |
+
"loss": 2.7953,
|
| 439 |
+
"mlm_loss": 2.7953,
|
| 440 |
+
"step": 490
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"ep_loss": 0.0,
|
| 444 |
+
"epoch": 0.39,
|
| 445 |
+
"learning_rate": 2.5e-05,
|
| 446 |
+
"loss": 2.7569,
|
| 447 |
+
"mlm_loss": 2.7569,
|
| 448 |
+
"step": 500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.39,
|
| 452 |
+
"eval_ep_loss": -1.5984127521514893,
|
| 453 |
+
"eval_loss": 2.6636624336242676,
|
| 454 |
+
"eval_mlm_loss": 2.6636624336242676,
|
| 455 |
+
"eval_runtime": 64.3177,
|
| 456 |
+
"eval_samples_per_second": 1086.31,
|
| 457 |
+
"eval_steps_per_second": 0.544,
|
| 458 |
+
"step": 500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"ep_loss": 0.0,
|
| 462 |
+
"epoch": 0.39,
|
| 463 |
+
"learning_rate": 2.55e-05,
|
| 464 |
+
"loss": 2.7157,
|
| 465 |
+
"mlm_loss": 2.7157,
|
| 466 |
+
"step": 510
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"ep_loss": 0.0,
|
| 470 |
+
"epoch": 0.4,
|
| 471 |
+
"learning_rate": 2.6e-05,
|
| 472 |
+
"loss": 2.6857,
|
| 473 |
+
"mlm_loss": 2.6857,
|
| 474 |
+
"step": 520
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"ep_loss": 0.0,
|
| 478 |
+
"epoch": 0.41,
|
| 479 |
+
"learning_rate": 2.65e-05,
|
| 480 |
+
"loss": 2.6796,
|
| 481 |
+
"mlm_loss": 2.6796,
|
| 482 |
+
"step": 530
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"ep_loss": 0.0,
|
| 486 |
+
"epoch": 0.42,
|
| 487 |
+
"learning_rate": 2.7e-05,
|
| 488 |
+
"loss": 2.6449,
|
| 489 |
+
"mlm_loss": 2.6449,
|
| 490 |
+
"step": 540
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"ep_loss": 0.0,
|
| 494 |
+
"epoch": 0.42,
|
| 495 |
+
"learning_rate": 2.75e-05,
|
| 496 |
+
"loss": 2.5932,
|
| 497 |
+
"mlm_loss": 2.5932,
|
| 498 |
+
"step": 550
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"ep_loss": 0.0,
|
| 502 |
+
"epoch": 0.43,
|
| 503 |
+
"learning_rate": 2.8e-05,
|
| 504 |
+
"loss": 2.5536,
|
| 505 |
+
"mlm_loss": 2.5536,
|
| 506 |
+
"step": 560
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"ep_loss": 0.0,
|
| 510 |
+
"epoch": 0.44,
|
| 511 |
+
"learning_rate": 2.85e-05,
|
| 512 |
+
"loss": 2.5271,
|
| 513 |
+
"mlm_loss": 2.5271,
|
| 514 |
+
"step": 570
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"ep_loss": 0.0,
|
| 518 |
+
"epoch": 0.45,
|
| 519 |
+
"learning_rate": 2.9000000000000004e-05,
|
| 520 |
+
"loss": 2.5057,
|
| 521 |
+
"mlm_loss": 2.5057,
|
| 522 |
+
"step": 580
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"ep_loss": 0.0,
|
| 526 |
+
"epoch": 0.46,
|
| 527 |
+
"learning_rate": 2.95e-05,
|
| 528 |
+
"loss": 2.476,
|
| 529 |
+
"mlm_loss": 2.476,
|
| 530 |
+
"step": 590
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"ep_loss": 0.0,
|
| 534 |
+
"epoch": 0.46,
|
| 535 |
+
"learning_rate": 3e-05,
|
| 536 |
+
"loss": 2.4265,
|
| 537 |
+
"mlm_loss": 2.4265,
|
| 538 |
+
"step": 600
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 0.46,
|
| 542 |
+
"eval_ep_loss": -1.7573626041412354,
|
| 543 |
+
"eval_loss": 2.359508752822876,
|
| 544 |
+
"eval_mlm_loss": 2.359508752822876,
|
| 545 |
+
"eval_runtime": 59.8647,
|
| 546 |
+
"eval_samples_per_second": 1167.114,
|
| 547 |
+
"eval_steps_per_second": 0.585,
|
| 548 |
+
"step": 600
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"ep_loss": 0.0,
|
| 552 |
+
"epoch": 0.47,
|
| 553 |
+
"learning_rate": 3.05e-05,
|
| 554 |
+
"loss": 2.4061,
|
| 555 |
+
"mlm_loss": 2.4061,
|
| 556 |
+
"step": 610
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"ep_loss": 0.0,
|
| 560 |
+
"epoch": 0.48,
|
| 561 |
+
"learning_rate": 3.1e-05,
|
| 562 |
+
"loss": 2.4071,
|
| 563 |
+
"mlm_loss": 2.4071,
|
| 564 |
+
"step": 620
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"ep_loss": 0.0,
|
| 568 |
+
"epoch": 0.49,
|
| 569 |
+
"learning_rate": 3.15e-05,
|
| 570 |
+
"loss": 2.3728,
|
| 571 |
+
"mlm_loss": 2.3728,
|
| 572 |
+
"step": 630
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"ep_loss": 0.0,
|
| 576 |
+
"epoch": 0.49,
|
| 577 |
+
"learning_rate": 3.2e-05,
|
| 578 |
+
"loss": 2.3465,
|
| 579 |
+
"mlm_loss": 2.3465,
|
| 580 |
+
"step": 640
|
| 581 |
+
},
|
| 582 |
+
{
|
| 583 |
+
"ep_loss": 0.0,
|
| 584 |
+
"epoch": 0.5,
|
| 585 |
+
"learning_rate": 3.2500000000000004e-05,
|
| 586 |
+
"loss": 2.3065,
|
| 587 |
+
"mlm_loss": 2.3065,
|
| 588 |
+
"step": 650
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"ep_loss": 0.0,
|
| 592 |
+
"epoch": 0.51,
|
| 593 |
+
"learning_rate": 3.3e-05,
|
| 594 |
+
"loss": 2.312,
|
| 595 |
+
"mlm_loss": 2.312,
|
| 596 |
+
"step": 660
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"ep_loss": 0.0,
|
| 600 |
+
"epoch": 0.52,
|
| 601 |
+
"learning_rate": 3.35e-05,
|
| 602 |
+
"loss": 2.29,
|
| 603 |
+
"mlm_loss": 2.29,
|
| 604 |
+
"step": 670
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"ep_loss": 0.0,
|
| 608 |
+
"epoch": 0.52,
|
| 609 |
+
"learning_rate": 3.4000000000000007e-05,
|
| 610 |
+
"loss": 2.2939,
|
| 611 |
+
"mlm_loss": 2.2939,
|
| 612 |
+
"step": 680
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"ep_loss": 0.0,
|
| 616 |
+
"epoch": 0.53,
|
| 617 |
+
"learning_rate": 3.4500000000000005e-05,
|
| 618 |
+
"loss": 2.2496,
|
| 619 |
+
"mlm_loss": 2.2496,
|
| 620 |
+
"step": 690
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"ep_loss": 0.0,
|
| 624 |
+
"epoch": 0.54,
|
| 625 |
+
"learning_rate": 3.5000000000000004e-05,
|
| 626 |
+
"loss": 2.2482,
|
| 627 |
+
"mlm_loss": 2.2482,
|
| 628 |
+
"step": 700
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 0.54,
|
| 632 |
+
"eval_ep_loss": -1.7850463390350342,
|
| 633 |
+
"eval_loss": 2.1652746200561523,
|
| 634 |
+
"eval_mlm_loss": 2.1652746200561523,
|
| 635 |
+
"eval_runtime": 63.4238,
|
| 636 |
+
"eval_samples_per_second": 1101.621,
|
| 637 |
+
"eval_steps_per_second": 0.552,
|
| 638 |
+
"step": 700
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"ep_loss": 0.0,
|
| 642 |
+
"epoch": 0.55,
|
| 643 |
+
"learning_rate": 3.5499999999999996e-05,
|
| 644 |
+
"loss": 2.2379,
|
| 645 |
+
"mlm_loss": 2.2379,
|
| 646 |
+
"step": 710
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"ep_loss": 0.0,
|
| 650 |
+
"epoch": 0.56,
|
| 651 |
+
"learning_rate": 3.6e-05,
|
| 652 |
+
"loss": 2.2088,
|
| 653 |
+
"mlm_loss": 2.2088,
|
| 654 |
+
"step": 720
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"ep_loss": 0.0,
|
| 658 |
+
"epoch": 0.56,
|
| 659 |
+
"learning_rate": 3.65e-05,
|
| 660 |
+
"loss": 2.1767,
|
| 661 |
+
"mlm_loss": 2.1767,
|
| 662 |
+
"step": 730
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"ep_loss": 0.0,
|
| 666 |
+
"epoch": 0.57,
|
| 667 |
+
"learning_rate": 3.7e-05,
|
| 668 |
+
"loss": 2.1589,
|
| 669 |
+
"mlm_loss": 2.1589,
|
| 670 |
+
"step": 740
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"ep_loss": 0.0,
|
| 674 |
+
"epoch": 0.58,
|
| 675 |
+
"learning_rate": 3.75e-05,
|
| 676 |
+
"loss": 2.1297,
|
| 677 |
+
"mlm_loss": 2.1297,
|
| 678 |
+
"step": 750
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"ep_loss": 0.0,
|
| 682 |
+
"epoch": 0.59,
|
| 683 |
+
"learning_rate": 3.8e-05,
|
| 684 |
+
"loss": 2.1128,
|
| 685 |
+
"mlm_loss": 2.1128,
|
| 686 |
+
"step": 760
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"ep_loss": 0.0,
|
| 690 |
+
"epoch": 0.59,
|
| 691 |
+
"learning_rate": 3.85e-05,
|
| 692 |
+
"loss": 2.0861,
|
| 693 |
+
"mlm_loss": 2.0861,
|
| 694 |
+
"step": 770
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"ep_loss": 0.0,
|
| 698 |
+
"epoch": 0.6,
|
| 699 |
+
"learning_rate": 3.9e-05,
|
| 700 |
+
"loss": 2.0557,
|
| 701 |
+
"mlm_loss": 2.0557,
|
| 702 |
+
"step": 780
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"ep_loss": 0.0,
|
| 706 |
+
"epoch": 0.61,
|
| 707 |
+
"learning_rate": 3.95e-05,
|
| 708 |
+
"loss": 2.0596,
|
| 709 |
+
"mlm_loss": 2.0596,
|
| 710 |
+
"step": 790
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"ep_loss": 0.0,
|
| 714 |
+
"epoch": 0.62,
|
| 715 |
+
"learning_rate": 4e-05,
|
| 716 |
+
"loss": 2.0349,
|
| 717 |
+
"mlm_loss": 2.0349,
|
| 718 |
+
"step": 800
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 0.62,
|
| 722 |
+
"eval_ep_loss": -1.7692768573760986,
|
| 723 |
+
"eval_loss": 1.916082501411438,
|
| 724 |
+
"eval_mlm_loss": 1.916082501411438,
|
| 725 |
+
"eval_runtime": 62.7772,
|
| 726 |
+
"eval_samples_per_second": 1112.968,
|
| 727 |
+
"eval_steps_per_second": 0.558,
|
| 728 |
+
"step": 800
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"ep_loss": 0.0,
|
| 732 |
+
"epoch": 0.62,
|
| 733 |
+
"learning_rate": 4.05e-05,
|
| 734 |
+
"loss": 1.9934,
|
| 735 |
+
"mlm_loss": 1.9934,
|
| 736 |
+
"step": 810
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"ep_loss": 0.0,
|
| 740 |
+
"epoch": 0.63,
|
| 741 |
+
"learning_rate": 4.1e-05,
|
| 742 |
+
"loss": 1.9554,
|
| 743 |
+
"mlm_loss": 1.9554,
|
| 744 |
+
"step": 820
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"ep_loss": 0.0,
|
| 748 |
+
"epoch": 0.64,
|
| 749 |
+
"learning_rate": 4.1500000000000006e-05,
|
| 750 |
+
"loss": 1.9348,
|
| 751 |
+
"mlm_loss": 1.9348,
|
| 752 |
+
"step": 830
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"ep_loss": 0.0,
|
| 756 |
+
"epoch": 0.65,
|
| 757 |
+
"learning_rate": 4.2000000000000004e-05,
|
| 758 |
+
"loss": 1.8978,
|
| 759 |
+
"mlm_loss": 1.8978,
|
| 760 |
+
"step": 840
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"ep_loss": 0.0,
|
| 764 |
+
"epoch": 0.66,
|
| 765 |
+
"learning_rate": 4.25e-05,
|
| 766 |
+
"loss": 1.8752,
|
| 767 |
+
"mlm_loss": 1.8752,
|
| 768 |
+
"step": 850
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"ep_loss": 0.0,
|
| 772 |
+
"epoch": 0.66,
|
| 773 |
+
"learning_rate": 4.2999999999999995e-05,
|
| 774 |
+
"loss": 1.8466,
|
| 775 |
+
"mlm_loss": 1.8466,
|
| 776 |
+
"step": 860
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"ep_loss": 0.0,
|
| 780 |
+
"epoch": 0.67,
|
| 781 |
+
"learning_rate": 4.35e-05,
|
| 782 |
+
"loss": 1.8168,
|
| 783 |
+
"mlm_loss": 1.8168,
|
| 784 |
+
"step": 870
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"ep_loss": 0.0,
|
| 788 |
+
"epoch": 0.68,
|
| 789 |
+
"learning_rate": 4.4e-05,
|
| 790 |
+
"loss": 1.8161,
|
| 791 |
+
"mlm_loss": 1.8161,
|
| 792 |
+
"step": 880
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"ep_loss": 0.0,
|
| 796 |
+
"epoch": 0.69,
|
| 797 |
+
"learning_rate": 4.45e-05,
|
| 798 |
+
"loss": 1.7637,
|
| 799 |
+
"mlm_loss": 1.7637,
|
| 800 |
+
"step": 890
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"ep_loss": 0.0,
|
| 804 |
+
"epoch": 0.69,
|
| 805 |
+
"learning_rate": 4.4999999999999996e-05,
|
| 806 |
+
"loss": 1.7292,
|
| 807 |
+
"mlm_loss": 1.7292,
|
| 808 |
+
"step": 900
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 0.69,
|
| 812 |
+
"eval_ep_loss": -1.673869252204895,
|
| 813 |
+
"eval_loss": 1.6480634212493896,
|
| 814 |
+
"eval_mlm_loss": 1.6480634212493896,
|
| 815 |
+
"eval_runtime": 61.0031,
|
| 816 |
+
"eval_samples_per_second": 1145.335,
|
| 817 |
+
"eval_steps_per_second": 0.574,
|
| 818 |
+
"step": 900
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"ep_loss": 0.0,
|
| 822 |
+
"epoch": 0.7,
|
| 823 |
+
"learning_rate": 4.55e-05,
|
| 824 |
+
"loss": 1.7031,
|
| 825 |
+
"mlm_loss": 1.7031,
|
| 826 |
+
"step": 910
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"ep_loss": 0.0,
|
| 830 |
+
"epoch": 0.71,
|
| 831 |
+
"learning_rate": 4.6e-05,
|
| 832 |
+
"loss": 1.6951,
|
| 833 |
+
"mlm_loss": 1.6951,
|
| 834 |
+
"step": 920
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"ep_loss": 0.0,
|
| 838 |
+
"epoch": 0.72,
|
| 839 |
+
"learning_rate": 4.65e-05,
|
| 840 |
+
"loss": 1.6706,
|
| 841 |
+
"mlm_loss": 1.6706,
|
| 842 |
+
"step": 930
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"ep_loss": 0.0,
|
| 846 |
+
"epoch": 0.73,
|
| 847 |
+
"learning_rate": 4.7000000000000004e-05,
|
| 848 |
+
"loss": 1.6674,
|
| 849 |
+
"mlm_loss": 1.6674,
|
| 850 |
+
"step": 940
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"ep_loss": 0.0,
|
| 854 |
+
"epoch": 0.73,
|
| 855 |
+
"learning_rate": 4.75e-05,
|
| 856 |
+
"loss": 1.6446,
|
| 857 |
+
"mlm_loss": 1.6446,
|
| 858 |
+
"step": 950
|
| 859 |
+
},
|
| 860 |
+
{
|
| 861 |
+
"ep_loss": 0.0,
|
| 862 |
+
"epoch": 0.74,
|
| 863 |
+
"learning_rate": 4.8e-05,
|
| 864 |
+
"loss": 1.6344,
|
| 865 |
+
"mlm_loss": 1.6344,
|
| 866 |
+
"step": 960
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"ep_loss": 0.0,
|
| 870 |
+
"epoch": 0.75,
|
| 871 |
+
"learning_rate": 4.85e-05,
|
| 872 |
+
"loss": 1.6182,
|
| 873 |
+
"mlm_loss": 1.6182,
|
| 874 |
+
"step": 970
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"ep_loss": 0.0,
|
| 878 |
+
"epoch": 0.76,
|
| 879 |
+
"learning_rate": 4.9000000000000005e-05,
|
| 880 |
+
"loss": 1.602,
|
| 881 |
+
"mlm_loss": 1.602,
|
| 882 |
+
"step": 980
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"ep_loss": 0.0,
|
| 886 |
+
"epoch": 0.76,
|
| 887 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 888 |
+
"loss": 1.5965,
|
| 889 |
+
"mlm_loss": 1.5965,
|
| 890 |
+
"step": 990
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"ep_loss": 0.0,
|
| 894 |
+
"epoch": 0.77,
|
| 895 |
+
"learning_rate": 5e-05,
|
| 896 |
+
"loss": 1.6022,
|
| 897 |
+
"mlm_loss": 1.6022,
|
| 898 |
+
"step": 1000
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.77,
|
| 902 |
+
"eval_ep_loss": -1.7339978218078613,
|
| 903 |
+
"eval_loss": 1.521234393119812,
|
| 904 |
+
"eval_mlm_loss": 1.521234393119812,
|
| 905 |
+
"eval_runtime": 61.977,
|
| 906 |
+
"eval_samples_per_second": 1127.338,
|
| 907 |
+
"eval_steps_per_second": 0.565,
|
| 908 |
+
"step": 1000
|
| 909 |
+
}
|
| 910 |
+
],
|
| 911 |
+
"max_steps": 25920,
|
| 912 |
+
"num_train_epochs": 20,
|
| 913 |
+
"total_flos": 2.713795755507712e+17,
|
| 914 |
+
"trial_name": null,
|
| 915 |
+
"trial_params": null
|
| 916 |
+
}
|
checkpoint-1000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbbced2969d3c34b5ea7f356c253acf2a826f9e9dea1f238559fc37a4cf95a6d
|
| 3 |
+
size 4027
|
config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"MarindaForMaskedLMWithEdgePrediction"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"ep_add_linear_projection": true,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 768,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 3072,
|
| 15 |
+
"layer_norm_eps": 1e-12,
|
| 16 |
+
"max_position_embeddings": 600,
|
| 17 |
+
"max_relative_position_embeddings": 8,
|
| 18 |
+
"model_type": "marinda",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"pad_token_id": 3,
|
| 22 |
+
"position_embedding_type": "mixed",
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.31.0",
|
| 25 |
+
"type_vocab_size": 2,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 30523
|
| 28 |
+
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9539754b990c6c28ccf1a024ed909035c3d57a6c0f4c419ed60ff426fbc46aeb
|
| 3 |
+
size 440772077
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": "[CLS]",
|
| 3 |
+
"mask_token": "[MASK]",
|
| 4 |
+
"pad_token": "[PAD]",
|
| 5 |
+
"sep_token": "[SEP]",
|
| 6 |
+
"unk_token": "[UNK]"
|
| 7 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"model_max_length": 512,
|
| 4 |
+
"tokenizer_class": "MarindaTokenizer"
|
| 5 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:dbbced2969d3c34b5ea7f356c253acf2a826f9e9dea1f238559fc37a4cf95a6d
|
| 3 |
+
size 4027
|