Upload 11 files
Browse files- config.json +25 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +7 -0
- tokenizer.json +0 -0
- tokenizer_config.json +55 -0
- trainer_state.json +762 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "TurkuNLP/bert-base-finnish-uncased-v1",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"CustomModel"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"torch_dtype": "float32",
|
| 21 |
+
"transformers_version": "4.44.2",
|
| 22 |
+
"type_vocab_size": 2,
|
| 23 |
+
"use_cache": true,
|
| 24 |
+
"vocab_size": 50101
|
| 25 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:315f1eb82f1131dbbac9aac43c9ff10d0306298104ca83400ed11956911c9a61
|
| 3 |
+
size 498493620
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d591d16b130eadc5ac4c7e1fd2ec3e8a0971657b041f970d259b64ade4c146e9
|
| 3 |
+
size 997109754
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4433083b78374fda17b67e9a3e865a50a40744ffdc2411a580ba008969a4e4d0
|
| 3 |
+
size 14244
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:26934ede0e8874c268bcb910767c4746fed0c4849c5de1c8634b58cd96143014
|
| 3 |
+
size 1064
|
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 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"101": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"102": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"103": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"104": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_lower_case": true,
|
| 47 |
+
"mask_token": "[MASK]",
|
| 48 |
+
"model_max_length": 512,
|
| 49 |
+
"pad_token": "[PAD]",
|
| 50 |
+
"sep_token": "[SEP]",
|
| 51 |
+
"strip_accents": null,
|
| 52 |
+
"tokenize_chinese_chars": true,
|
| 53 |
+
"tokenizer_class": "BertTokenizer",
|
| 54 |
+
"unk_token": "[UNK]"
|
| 55 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,762 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.5330753326416016,
|
| 3 |
+
"best_model_checkpoint": "/scratch/project_2006600/dif_models/fi_bert_reg/diff_bert_base_sampled_unfrozen_early_real_test/checkpoint-3500",
|
| 4 |
+
"epoch": 10.666666666666666,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 4000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.26666666666666666,
|
| 13 |
+
"grad_norm": 37.87092208862305,
|
| 14 |
+
"learning_rate": 5.444444444444444e-07,
|
| 15 |
+
"loss": 13.2633,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.26666666666666666,
|
| 20 |
+
"eval_loss": 10.746745109558105,
|
| 21 |
+
"eval_mse": 10.746745109558105,
|
| 22 |
+
"eval_r2": -3.070736885070801,
|
| 23 |
+
"eval_rmse": 3.2782227993011475,
|
| 24 |
+
"eval_runtime": 57.8905,
|
| 25 |
+
"eval_samples_per_second": 103.644,
|
| 26 |
+
"eval_steps_per_second": 25.911,
|
| 27 |
+
"step": 100
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.5333333333333333,
|
| 31 |
+
"grad_norm": 11.965848922729492,
|
| 32 |
+
"learning_rate": 1.0944444444444445e-06,
|
| 33 |
+
"loss": 6.0044,
|
| 34 |
+
"step": 200
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 0.5333333333333333,
|
| 38 |
+
"eval_loss": 2.2986323833465576,
|
| 39 |
+
"eval_mse": 2.2986323833465576,
|
| 40 |
+
"eval_r2": 0.129305899143219,
|
| 41 |
+
"eval_rmse": 1.5161241292953491,
|
| 42 |
+
"eval_runtime": 57.8766,
|
| 43 |
+
"eval_samples_per_second": 103.669,
|
| 44 |
+
"eval_steps_per_second": 25.917,
|
| 45 |
+
"step": 200
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.8,
|
| 49 |
+
"grad_norm": 16.900310516357422,
|
| 50 |
+
"learning_rate": 1.6500000000000003e-06,
|
| 51 |
+
"loss": 1.9777,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.8,
|
| 56 |
+
"eval_loss": 1.0734479427337646,
|
| 57 |
+
"eval_mse": 1.0734479427337646,
|
| 58 |
+
"eval_r2": 0.593390941619873,
|
| 59 |
+
"eval_rmse": 1.0360733270645142,
|
| 60 |
+
"eval_runtime": 57.8679,
|
| 61 |
+
"eval_samples_per_second": 103.684,
|
| 62 |
+
"eval_steps_per_second": 25.921,
|
| 63 |
+
"step": 300
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.0666666666666667,
|
| 67 |
+
"grad_norm": 36.441951751708984,
|
| 68 |
+
"learning_rate": 2.2e-06,
|
| 69 |
+
"loss": 0.9918,
|
| 70 |
+
"step": 400
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 1.0666666666666667,
|
| 74 |
+
"eval_loss": 0.7707585692405701,
|
| 75 |
+
"eval_mse": 0.7707585692405701,
|
| 76 |
+
"eval_r2": 0.708046019077301,
|
| 77 |
+
"eval_rmse": 0.8779285550117493,
|
| 78 |
+
"eval_runtime": 57.8738,
|
| 79 |
+
"eval_samples_per_second": 103.674,
|
| 80 |
+
"eval_steps_per_second": 25.918,
|
| 81 |
+
"step": 400
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.3333333333333333,
|
| 85 |
+
"grad_norm": 35.2764892578125,
|
| 86 |
+
"learning_rate": 2.755555555555556e-06,
|
| 87 |
+
"loss": 0.7404,
|
| 88 |
+
"step": 500
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.3333333333333333,
|
| 92 |
+
"eval_loss": 0.8621485829353333,
|
| 93 |
+
"eval_mse": 0.8621485829353333,
|
| 94 |
+
"eval_r2": 0.6734285354614258,
|
| 95 |
+
"eval_rmse": 0.9285195469856262,
|
| 96 |
+
"eval_runtime": 57.8776,
|
| 97 |
+
"eval_samples_per_second": 103.667,
|
| 98 |
+
"eval_steps_per_second": 25.917,
|
| 99 |
+
"step": 500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 1.6,
|
| 103 |
+
"grad_norm": 25.652307510375977,
|
| 104 |
+
"learning_rate": 3.3055555555555558e-06,
|
| 105 |
+
"loss": 0.5897,
|
| 106 |
+
"step": 600
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 1.6,
|
| 110 |
+
"eval_loss": 0.7726595997810364,
|
| 111 |
+
"eval_mse": 0.7726595997810364,
|
| 112 |
+
"eval_r2": 0.7073259353637695,
|
| 113 |
+
"eval_rmse": 0.8790105581283569,
|
| 114 |
+
"eval_runtime": 57.8605,
|
| 115 |
+
"eval_samples_per_second": 103.698,
|
| 116 |
+
"eval_steps_per_second": 25.924,
|
| 117 |
+
"step": 600
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 1.8666666666666667,
|
| 121 |
+
"grad_norm": 20.405384063720703,
|
| 122 |
+
"learning_rate": 3.861111111111112e-06,
|
| 123 |
+
"loss": 0.5903,
|
| 124 |
+
"step": 700
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 1.8666666666666667,
|
| 128 |
+
"eval_loss": 0.7032095789909363,
|
| 129 |
+
"eval_mse": 0.7032095789909363,
|
| 130 |
+
"eval_r2": 0.7336327433586121,
|
| 131 |
+
"eval_rmse": 0.8385758996009827,
|
| 132 |
+
"eval_runtime": 57.8612,
|
| 133 |
+
"eval_samples_per_second": 103.696,
|
| 134 |
+
"eval_steps_per_second": 25.924,
|
| 135 |
+
"step": 700
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 2.1333333333333333,
|
| 139 |
+
"grad_norm": 17.7819766998291,
|
| 140 |
+
"learning_rate": 4.416666666666667e-06,
|
| 141 |
+
"loss": 0.4654,
|
| 142 |
+
"step": 800
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 2.1333333333333333,
|
| 146 |
+
"eval_loss": 0.6557766199111938,
|
| 147 |
+
"eval_mse": 0.6557766199111938,
|
| 148 |
+
"eval_r2": 0.7515997886657715,
|
| 149 |
+
"eval_rmse": 0.809800386428833,
|
| 150 |
+
"eval_runtime": 57.8596,
|
| 151 |
+
"eval_samples_per_second": 103.699,
|
| 152 |
+
"eval_steps_per_second": 25.925,
|
| 153 |
+
"step": 800
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.4,
|
| 157 |
+
"grad_norm": 16.64554786682129,
|
| 158 |
+
"learning_rate": 4.9722222222222224e-06,
|
| 159 |
+
"loss": 0.4457,
|
| 160 |
+
"step": 900
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.4,
|
| 164 |
+
"eval_loss": 0.7730452418327332,
|
| 165 |
+
"eval_mse": 0.7730452418327332,
|
| 166 |
+
"eval_r2": 0.707179844379425,
|
| 167 |
+
"eval_rmse": 0.8792299032211304,
|
| 168 |
+
"eval_runtime": 57.8442,
|
| 169 |
+
"eval_samples_per_second": 103.727,
|
| 170 |
+
"eval_steps_per_second": 25.932,
|
| 171 |
+
"step": 900
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.6666666666666665,
|
| 175 |
+
"grad_norm": 23.086116790771484,
|
| 176 |
+
"learning_rate": 5.527777777777779e-06,
|
| 177 |
+
"loss": 0.3555,
|
| 178 |
+
"step": 1000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 2.6666666666666665,
|
| 182 |
+
"eval_loss": 0.6905973553657532,
|
| 183 |
+
"eval_mse": 0.6905973553657532,
|
| 184 |
+
"eval_r2": 0.7384101152420044,
|
| 185 |
+
"eval_rmse": 0.8310218453407288,
|
| 186 |
+
"eval_runtime": 57.8673,
|
| 187 |
+
"eval_samples_per_second": 103.685,
|
| 188 |
+
"eval_steps_per_second": 25.921,
|
| 189 |
+
"step": 1000
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.9333333333333336,
|
| 193 |
+
"grad_norm": 12.183361053466797,
|
| 194 |
+
"learning_rate": 6.083333333333333e-06,
|
| 195 |
+
"loss": 0.3482,
|
| 196 |
+
"step": 1100
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 2.9333333333333336,
|
| 200 |
+
"eval_loss": 0.7757130265235901,
|
| 201 |
+
"eval_mse": 0.7757130265235901,
|
| 202 |
+
"eval_r2": 0.7061693072319031,
|
| 203 |
+
"eval_rmse": 0.8807457089424133,
|
| 204 |
+
"eval_runtime": 57.8693,
|
| 205 |
+
"eval_samples_per_second": 103.682,
|
| 206 |
+
"eval_steps_per_second": 25.921,
|
| 207 |
+
"step": 1100
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.2,
|
| 211 |
+
"grad_norm": 7.989579677581787,
|
| 212 |
+
"learning_rate": 6.6388888888888895e-06,
|
| 213 |
+
"loss": 0.2977,
|
| 214 |
+
"step": 1200
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 3.2,
|
| 218 |
+
"eval_loss": 0.756076991558075,
|
| 219 |
+
"eval_mse": 0.7560770511627197,
|
| 220 |
+
"eval_r2": 0.7136071920394897,
|
| 221 |
+
"eval_rmse": 0.8695269227027893,
|
| 222 |
+
"eval_runtime": 57.8756,
|
| 223 |
+
"eval_samples_per_second": 103.671,
|
| 224 |
+
"eval_steps_per_second": 25.918,
|
| 225 |
+
"step": 1200
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.466666666666667,
|
| 229 |
+
"grad_norm": 23.822872161865234,
|
| 230 |
+
"learning_rate": 7.194444444444445e-06,
|
| 231 |
+
"loss": 0.312,
|
| 232 |
+
"step": 1300
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.466666666666667,
|
| 236 |
+
"eval_loss": 0.9145962595939636,
|
| 237 |
+
"eval_mse": 0.9145963788032532,
|
| 238 |
+
"eval_r2": 0.6535619497299194,
|
| 239 |
+
"eval_rmse": 0.9563453197479248,
|
| 240 |
+
"eval_runtime": 57.8768,
|
| 241 |
+
"eval_samples_per_second": 103.669,
|
| 242 |
+
"eval_steps_per_second": 25.917,
|
| 243 |
+
"step": 1300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 3.7333333333333334,
|
| 247 |
+
"grad_norm": 15.19206428527832,
|
| 248 |
+
"learning_rate": 7.75e-06,
|
| 249 |
+
"loss": 0.2681,
|
| 250 |
+
"step": 1400
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 3.7333333333333334,
|
| 254 |
+
"eval_loss": 0.6494566202163696,
|
| 255 |
+
"eval_mse": 0.6494566202163696,
|
| 256 |
+
"eval_r2": 0.7539936900138855,
|
| 257 |
+
"eval_rmse": 0.8058887124061584,
|
| 258 |
+
"eval_runtime": 57.883,
|
| 259 |
+
"eval_samples_per_second": 103.657,
|
| 260 |
+
"eval_steps_per_second": 25.914,
|
| 261 |
+
"step": 1400
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 4.0,
|
| 265 |
+
"grad_norm": 10.404943466186523,
|
| 266 |
+
"learning_rate": 8.305555555555557e-06,
|
| 267 |
+
"loss": 0.2744,
|
| 268 |
+
"step": 1500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 4.0,
|
| 272 |
+
"eval_loss": 0.6918351054191589,
|
| 273 |
+
"eval_mse": 0.6918351054191589,
|
| 274 |
+
"eval_r2": 0.7379412651062012,
|
| 275 |
+
"eval_rmse": 0.8317662477493286,
|
| 276 |
+
"eval_runtime": 57.8842,
|
| 277 |
+
"eval_samples_per_second": 103.655,
|
| 278 |
+
"eval_steps_per_second": 25.914,
|
| 279 |
+
"step": 1500
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 4.266666666666667,
|
| 283 |
+
"grad_norm": 11.18205738067627,
|
| 284 |
+
"learning_rate": 8.861111111111111e-06,
|
| 285 |
+
"loss": 0.2142,
|
| 286 |
+
"step": 1600
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 4.266666666666667,
|
| 290 |
+
"eval_loss": 0.7790309190750122,
|
| 291 |
+
"eval_mse": 0.779030978679657,
|
| 292 |
+
"eval_r2": 0.7049124836921692,
|
| 293 |
+
"eval_rmse": 0.8826273083686829,
|
| 294 |
+
"eval_runtime": 58.0438,
|
| 295 |
+
"eval_samples_per_second": 103.37,
|
| 296 |
+
"eval_steps_per_second": 25.843,
|
| 297 |
+
"step": 1600
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.533333333333333,
|
| 301 |
+
"grad_norm": 15.371255874633789,
|
| 302 |
+
"learning_rate": 9.416666666666667e-06,
|
| 303 |
+
"loss": 0.2103,
|
| 304 |
+
"step": 1700
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 4.533333333333333,
|
| 308 |
+
"eval_loss": 0.6177367568016052,
|
| 309 |
+
"eval_mse": 0.6177367568016052,
|
| 310 |
+
"eval_r2": 0.7660087943077087,
|
| 311 |
+
"eval_rmse": 0.7859622836112976,
|
| 312 |
+
"eval_runtime": 57.8859,
|
| 313 |
+
"eval_samples_per_second": 103.652,
|
| 314 |
+
"eval_steps_per_second": 25.913,
|
| 315 |
+
"step": 1700
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 4.8,
|
| 319 |
+
"grad_norm": 26.033283233642578,
|
| 320 |
+
"learning_rate": 9.972222222222224e-06,
|
| 321 |
+
"loss": 0.2009,
|
| 322 |
+
"step": 1800
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 4.8,
|
| 326 |
+
"eval_loss": 0.4925681948661804,
|
| 327 |
+
"eval_mse": 0.4925681948661804,
|
| 328 |
+
"eval_r2": 0.8134211301803589,
|
| 329 |
+
"eval_rmse": 0.7018320560455322,
|
| 330 |
+
"eval_runtime": 57.8886,
|
| 331 |
+
"eval_samples_per_second": 103.647,
|
| 332 |
+
"eval_steps_per_second": 25.912,
|
| 333 |
+
"step": 1800
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.066666666666666,
|
| 337 |
+
"grad_norm": 14.781371116638184,
|
| 338 |
+
"learning_rate": 1.0527777777777778e-05,
|
| 339 |
+
"loss": 0.2077,
|
| 340 |
+
"step": 1900
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 5.066666666666666,
|
| 344 |
+
"eval_loss": 0.731107234954834,
|
| 345 |
+
"eval_mse": 0.7311073541641235,
|
| 346 |
+
"eval_r2": 0.7230653762817383,
|
| 347 |
+
"eval_rmse": 0.8550481796264648,
|
| 348 |
+
"eval_runtime": 57.8849,
|
| 349 |
+
"eval_samples_per_second": 103.654,
|
| 350 |
+
"eval_steps_per_second": 25.913,
|
| 351 |
+
"step": 1900
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 5.333333333333333,
|
| 355 |
+
"grad_norm": 10.876953125,
|
| 356 |
+
"learning_rate": 1.1083333333333335e-05,
|
| 357 |
+
"loss": 0.1621,
|
| 358 |
+
"step": 2000
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 5.333333333333333,
|
| 362 |
+
"eval_loss": 0.9120460748672485,
|
| 363 |
+
"eval_mse": 0.9120460748672485,
|
| 364 |
+
"eval_r2": 0.654528021812439,
|
| 365 |
+
"eval_rmse": 0.9550110101699829,
|
| 366 |
+
"eval_runtime": 57.8944,
|
| 367 |
+
"eval_samples_per_second": 103.637,
|
| 368 |
+
"eval_steps_per_second": 25.909,
|
| 369 |
+
"step": 2000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.6,
|
| 373 |
+
"grad_norm": 9.433772087097168,
|
| 374 |
+
"learning_rate": 1.163888888888889e-05,
|
| 375 |
+
"loss": 0.172,
|
| 376 |
+
"step": 2100
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 5.6,
|
| 380 |
+
"eval_loss": 0.7684391140937805,
|
| 381 |
+
"eval_mse": 0.7684392333030701,
|
| 382 |
+
"eval_r2": 0.7089245319366455,
|
| 383 |
+
"eval_rmse": 0.8766066431999207,
|
| 384 |
+
"eval_runtime": 57.8919,
|
| 385 |
+
"eval_samples_per_second": 103.641,
|
| 386 |
+
"eval_steps_per_second": 25.91,
|
| 387 |
+
"step": 2100
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 5.866666666666667,
|
| 391 |
+
"grad_norm": 8.749499320983887,
|
| 392 |
+
"learning_rate": 1.2194444444444447e-05,
|
| 393 |
+
"loss": 0.205,
|
| 394 |
+
"step": 2200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 5.866666666666667,
|
| 398 |
+
"eval_loss": 0.9115136861801147,
|
| 399 |
+
"eval_mse": 0.9115136861801147,
|
| 400 |
+
"eval_r2": 0.6547296643257141,
|
| 401 |
+
"eval_rmse": 0.9547322392463684,
|
| 402 |
+
"eval_runtime": 57.8967,
|
| 403 |
+
"eval_samples_per_second": 103.633,
|
| 404 |
+
"eval_steps_per_second": 25.908,
|
| 405 |
+
"step": 2200
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 6.133333333333334,
|
| 409 |
+
"grad_norm": 13.432199478149414,
|
| 410 |
+
"learning_rate": 1.275e-05,
|
| 411 |
+
"loss": 0.1548,
|
| 412 |
+
"step": 2300
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 6.133333333333334,
|
| 416 |
+
"eval_loss": 0.7451738119125366,
|
| 417 |
+
"eval_mse": 0.7451738119125366,
|
| 418 |
+
"eval_r2": 0.7177371978759766,
|
| 419 |
+
"eval_rmse": 0.8632345199584961,
|
| 420 |
+
"eval_runtime": 57.8927,
|
| 421 |
+
"eval_samples_per_second": 103.64,
|
| 422 |
+
"eval_steps_per_second": 25.91,
|
| 423 |
+
"step": 2300
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.4,
|
| 427 |
+
"grad_norm": 13.872139930725098,
|
| 428 |
+
"learning_rate": 1.3305555555555556e-05,
|
| 429 |
+
"loss": 0.1593,
|
| 430 |
+
"step": 2400
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 6.4,
|
| 434 |
+
"eval_loss": 0.6607070565223694,
|
| 435 |
+
"eval_mse": 0.6607070565223694,
|
| 436 |
+
"eval_r2": 0.7497321963310242,
|
| 437 |
+
"eval_rmse": 0.8128389120101929,
|
| 438 |
+
"eval_runtime": 57.8803,
|
| 439 |
+
"eval_samples_per_second": 103.662,
|
| 440 |
+
"eval_steps_per_second": 25.916,
|
| 441 |
+
"step": 2400
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.666666666666667,
|
| 445 |
+
"grad_norm": 10.801892280578613,
|
| 446 |
+
"learning_rate": 1.3855555555555558e-05,
|
| 447 |
+
"loss": 0.1684,
|
| 448 |
+
"step": 2500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 6.666666666666667,
|
| 452 |
+
"eval_loss": 0.8161285519599915,
|
| 453 |
+
"eval_mse": 0.8161285519599915,
|
| 454 |
+
"eval_r2": 0.690860390663147,
|
| 455 |
+
"eval_rmse": 0.903398334980011,
|
| 456 |
+
"eval_runtime": 57.8953,
|
| 457 |
+
"eval_samples_per_second": 103.635,
|
| 458 |
+
"eval_steps_per_second": 25.909,
|
| 459 |
+
"step": 2500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 6.933333333333334,
|
| 463 |
+
"grad_norm": 13.156838417053223,
|
| 464 |
+
"learning_rate": 1.441111111111111e-05,
|
| 465 |
+
"loss": 0.1823,
|
| 466 |
+
"step": 2600
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 6.933333333333334,
|
| 470 |
+
"eval_loss": 1.0073966979980469,
|
| 471 |
+
"eval_mse": 1.0073966979980469,
|
| 472 |
+
"eval_r2": 0.6184103488922119,
|
| 473 |
+
"eval_rmse": 1.003691554069519,
|
| 474 |
+
"eval_runtime": 57.9026,
|
| 475 |
+
"eval_samples_per_second": 103.622,
|
| 476 |
+
"eval_steps_per_second": 25.906,
|
| 477 |
+
"step": 2600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.2,
|
| 481 |
+
"grad_norm": 16.002506256103516,
|
| 482 |
+
"learning_rate": 1.4966666666666667e-05,
|
| 483 |
+
"loss": 0.1936,
|
| 484 |
+
"step": 2700
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 7.2,
|
| 488 |
+
"eval_loss": 0.8159008622169495,
|
| 489 |
+
"eval_mse": 0.8159008622169495,
|
| 490 |
+
"eval_r2": 0.690946638584137,
|
| 491 |
+
"eval_rmse": 0.9032723307609558,
|
| 492 |
+
"eval_runtime": 57.9032,
|
| 493 |
+
"eval_samples_per_second": 103.621,
|
| 494 |
+
"eval_steps_per_second": 25.905,
|
| 495 |
+
"step": 2700
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 7.466666666666667,
|
| 499 |
+
"grad_norm": 12.294295310974121,
|
| 500 |
+
"learning_rate": 1.5522222222222223e-05,
|
| 501 |
+
"loss": 0.1506,
|
| 502 |
+
"step": 2800
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 7.466666666666667,
|
| 506 |
+
"eval_loss": 0.8324363827705383,
|
| 507 |
+
"eval_mse": 0.8324363827705383,
|
| 508 |
+
"eval_r2": 0.6846832036972046,
|
| 509 |
+
"eval_rmse": 0.9123795032501221,
|
| 510 |
+
"eval_runtime": 57.9057,
|
| 511 |
+
"eval_samples_per_second": 103.617,
|
| 512 |
+
"eval_steps_per_second": 25.904,
|
| 513 |
+
"step": 2800
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 7.733333333333333,
|
| 517 |
+
"grad_norm": 5.195356369018555,
|
| 518 |
+
"learning_rate": 1.607777777777778e-05,
|
| 519 |
+
"loss": 0.1499,
|
| 520 |
+
"step": 2900
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 7.733333333333333,
|
| 524 |
+
"eval_loss": 0.8613407611846924,
|
| 525 |
+
"eval_mse": 0.8613407611846924,
|
| 526 |
+
"eval_r2": 0.6737345457077026,
|
| 527 |
+
"eval_rmse": 0.9280844330787659,
|
| 528 |
+
"eval_runtime": 57.9123,
|
| 529 |
+
"eval_samples_per_second": 103.605,
|
| 530 |
+
"eval_steps_per_second": 25.901,
|
| 531 |
+
"step": 2900
|
| 532 |
+
},
|
| 533 |
+
{
|
| 534 |
+
"epoch": 8.0,
|
| 535 |
+
"grad_norm": 11.357749938964844,
|
| 536 |
+
"learning_rate": 1.6633333333333336e-05,
|
| 537 |
+
"loss": 0.1631,
|
| 538 |
+
"step": 3000
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 8.0,
|
| 542 |
+
"eval_loss": 0.6052080988883972,
|
| 543 |
+
"eval_mse": 0.6052080988883972,
|
| 544 |
+
"eval_r2": 0.7707545161247253,
|
| 545 |
+
"eval_rmse": 0.7779512405395508,
|
| 546 |
+
"eval_runtime": 57.8963,
|
| 547 |
+
"eval_samples_per_second": 103.634,
|
| 548 |
+
"eval_steps_per_second": 25.908,
|
| 549 |
+
"step": 3000
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 8.266666666666667,
|
| 553 |
+
"grad_norm": 20.62454605102539,
|
| 554 |
+
"learning_rate": 1.7188888888888892e-05,
|
| 555 |
+
"loss": 0.1559,
|
| 556 |
+
"step": 3100
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 8.266666666666667,
|
| 560 |
+
"eval_loss": 0.584992527961731,
|
| 561 |
+
"eval_mse": 0.584992527961731,
|
| 562 |
+
"eval_r2": 0.7784119248390198,
|
| 563 |
+
"eval_rmse": 0.7648480534553528,
|
| 564 |
+
"eval_runtime": 57.8939,
|
| 565 |
+
"eval_samples_per_second": 103.638,
|
| 566 |
+
"eval_steps_per_second": 25.909,
|
| 567 |
+
"step": 3100
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 8.533333333333333,
|
| 571 |
+
"grad_norm": 5.493505954742432,
|
| 572 |
+
"learning_rate": 1.7744444444444445e-05,
|
| 573 |
+
"loss": 0.1434,
|
| 574 |
+
"step": 3200
|
| 575 |
+
},
|
| 576 |
+
{
|
| 577 |
+
"epoch": 8.533333333333333,
|
| 578 |
+
"eval_loss": 0.6216039061546326,
|
| 579 |
+
"eval_mse": 0.6216039061546326,
|
| 580 |
+
"eval_r2": 0.7645439505577087,
|
| 581 |
+
"eval_rmse": 0.7884185910224915,
|
| 582 |
+
"eval_runtime": 57.9149,
|
| 583 |
+
"eval_samples_per_second": 103.6,
|
| 584 |
+
"eval_steps_per_second": 25.9,
|
| 585 |
+
"step": 3200
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 8.8,
|
| 589 |
+
"grad_norm": 8.887247085571289,
|
| 590 |
+
"learning_rate": 1.83e-05,
|
| 591 |
+
"loss": 0.1485,
|
| 592 |
+
"step": 3300
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 8.8,
|
| 596 |
+
"eval_loss": 0.6688755750656128,
|
| 597 |
+
"eval_mse": 0.6688756346702576,
|
| 598 |
+
"eval_r2": 0.7466380000114441,
|
| 599 |
+
"eval_rmse": 0.8178481459617615,
|
| 600 |
+
"eval_runtime": 57.8949,
|
| 601 |
+
"eval_samples_per_second": 103.636,
|
| 602 |
+
"eval_steps_per_second": 25.909,
|
| 603 |
+
"step": 3300
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 9.066666666666666,
|
| 607 |
+
"grad_norm": 54.376461029052734,
|
| 608 |
+
"learning_rate": 1.8855555555555557e-05,
|
| 609 |
+
"loss": 0.1487,
|
| 610 |
+
"step": 3400
|
| 611 |
+
},
|
| 612 |
+
{
|
| 613 |
+
"epoch": 9.066666666666666,
|
| 614 |
+
"eval_loss": 0.6742914915084839,
|
| 615 |
+
"eval_mse": 0.6742914915084839,
|
| 616 |
+
"eval_r2": 0.7445865869522095,
|
| 617 |
+
"eval_rmse": 0.8211525678634644,
|
| 618 |
+
"eval_runtime": 57.9237,
|
| 619 |
+
"eval_samples_per_second": 103.584,
|
| 620 |
+
"eval_steps_per_second": 25.896,
|
| 621 |
+
"step": 3400
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 9.333333333333334,
|
| 625 |
+
"grad_norm": 2.6493732929229736,
|
| 626 |
+
"learning_rate": 1.941111111111111e-05,
|
| 627 |
+
"loss": 0.1248,
|
| 628 |
+
"step": 3500
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 9.333333333333334,
|
| 632 |
+
"eval_loss": 0.5330753326416016,
|
| 633 |
+
"eval_mse": 0.5330753326416016,
|
| 634 |
+
"eval_r2": 0.7980775237083435,
|
| 635 |
+
"eval_rmse": 0.730120062828064,
|
| 636 |
+
"eval_runtime": 57.9187,
|
| 637 |
+
"eval_samples_per_second": 103.593,
|
| 638 |
+
"eval_steps_per_second": 25.898,
|
| 639 |
+
"step": 3500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 9.6,
|
| 643 |
+
"grad_norm": 22.65760040283203,
|
| 644 |
+
"learning_rate": 1.9966666666666666e-05,
|
| 645 |
+
"loss": 0.127,
|
| 646 |
+
"step": 3600
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 9.6,
|
| 650 |
+
"eval_loss": 0.760678231716156,
|
| 651 |
+
"eval_mse": 0.760678231716156,
|
| 652 |
+
"eval_r2": 0.7118643522262573,
|
| 653 |
+
"eval_rmse": 0.8721687197685242,
|
| 654 |
+
"eval_runtime": 57.9121,
|
| 655 |
+
"eval_samples_per_second": 103.605,
|
| 656 |
+
"eval_steps_per_second": 25.901,
|
| 657 |
+
"step": 3600
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 9.866666666666667,
|
| 661 |
+
"grad_norm": 11.861601829528809,
|
| 662 |
+
"learning_rate": 1.999958463257905e-05,
|
| 663 |
+
"loss": 0.1409,
|
| 664 |
+
"step": 3700
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 9.866666666666667,
|
| 668 |
+
"eval_loss": 0.6627658009529114,
|
| 669 |
+
"eval_mse": 0.6627658009529114,
|
| 670 |
+
"eval_r2": 0.748952329158783,
|
| 671 |
+
"eval_rmse": 0.8141043186187744,
|
| 672 |
+
"eval_runtime": 57.9148,
|
| 673 |
+
"eval_samples_per_second": 103.601,
|
| 674 |
+
"eval_steps_per_second": 25.9,
|
| 675 |
+
"step": 3700
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 10.133333333333333,
|
| 679 |
+
"grad_norm": 10.758604049682617,
|
| 680 |
+
"learning_rate": 1.999823082667316e-05,
|
| 681 |
+
"loss": 0.1352,
|
| 682 |
+
"step": 3800
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 10.133333333333333,
|
| 686 |
+
"eval_loss": 0.7061165571212769,
|
| 687 |
+
"eval_mse": 0.7061165571212769,
|
| 688 |
+
"eval_r2": 0.7325316071510315,
|
| 689 |
+
"eval_rmse": 0.8403074145317078,
|
| 690 |
+
"eval_runtime": 58.8761,
|
| 691 |
+
"eval_samples_per_second": 101.909,
|
| 692 |
+
"eval_steps_per_second": 25.477,
|
| 693 |
+
"step": 3800
|
| 694 |
+
},
|
| 695 |
+
{
|
| 696 |
+
"epoch": 10.4,
|
| 697 |
+
"grad_norm": 4.267393112182617,
|
| 698 |
+
"learning_rate": 1.999593701724414e-05,
|
| 699 |
+
"loss": 0.1418,
|
| 700 |
+
"step": 3900
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 10.4,
|
| 704 |
+
"eval_loss": 0.6572955846786499,
|
| 705 |
+
"eval_mse": 0.6572955846786499,
|
| 706 |
+
"eval_r2": 0.7510244250297546,
|
| 707 |
+
"eval_rmse": 0.810737669467926,
|
| 708 |
+
"eval_runtime": 57.921,
|
| 709 |
+
"eval_samples_per_second": 103.589,
|
| 710 |
+
"eval_steps_per_second": 25.897,
|
| 711 |
+
"step": 3900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 10.666666666666666,
|
| 715 |
+
"grad_norm": 4.308215618133545,
|
| 716 |
+
"learning_rate": 1.9992703419949032e-05,
|
| 717 |
+
"loss": 0.138,
|
| 718 |
+
"step": 4000
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 10.666666666666666,
|
| 722 |
+
"eval_loss": 0.6316142678260803,
|
| 723 |
+
"eval_mse": 0.6316142678260803,
|
| 724 |
+
"eval_r2": 0.7607522010803223,
|
| 725 |
+
"eval_rmse": 0.7947416305541992,
|
| 726 |
+
"eval_runtime": 57.92,
|
| 727 |
+
"eval_samples_per_second": 103.591,
|
| 728 |
+
"eval_steps_per_second": 25.898,
|
| 729 |
+
"step": 4000
|
| 730 |
+
}
|
| 731 |
+
],
|
| 732 |
+
"logging_steps": 100,
|
| 733 |
+
"max_steps": 9000,
|
| 734 |
+
"num_input_tokens_seen": 0,
|
| 735 |
+
"num_train_epochs": 24,
|
| 736 |
+
"save_steps": 500,
|
| 737 |
+
"stateful_callbacks": {
|
| 738 |
+
"EarlyStoppingCallback": {
|
| 739 |
+
"args": {
|
| 740 |
+
"early_stopping_patience": 6,
|
| 741 |
+
"early_stopping_threshold": 0.0
|
| 742 |
+
},
|
| 743 |
+
"attributes": {
|
| 744 |
+
"early_stopping_patience_counter": 0
|
| 745 |
+
}
|
| 746 |
+
},
|
| 747 |
+
"TrainerControl": {
|
| 748 |
+
"args": {
|
| 749 |
+
"should_epoch_stop": false,
|
| 750 |
+
"should_evaluate": false,
|
| 751 |
+
"should_log": false,
|
| 752 |
+
"should_save": true,
|
| 753 |
+
"should_training_stop": false
|
| 754 |
+
},
|
| 755 |
+
"attributes": {}
|
| 756 |
+
}
|
| 757 |
+
},
|
| 758 |
+
"total_flos": 1.6858183041024e+16,
|
| 759 |
+
"train_batch_size": 4,
|
| 760 |
+
"trial_name": null,
|
| 761 |
+
"trial_params": null
|
| 762 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea363f28b7ca133511cd9d0d648a88c4b88ac591bacff53c8f515319c80b8e4d
|
| 3 |
+
size 5176
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|