Upload 7 files
Browse files- config.json +32 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +903 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "PORTULAN/albertina-100m-portuguese-ptpt-encoder",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"DebertaForMaskedLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.1,
|
| 7 |
+
"hidden_act": "gelu",
|
| 8 |
+
"hidden_dropout_prob": 0.1,
|
| 9 |
+
"hidden_size": 768,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"intermediate_size": 3072,
|
| 12 |
+
"layer_norm_eps": 1e-07,
|
| 13 |
+
"max_position_embeddings": 512,
|
| 14 |
+
"max_relative_positions": -1,
|
| 15 |
+
"model_type": "deberta",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"pooler_dropout": 0,
|
| 20 |
+
"pooler_hidden_act": "gelu",
|
| 21 |
+
"pooler_hidden_size": 768,
|
| 22 |
+
"pos_att_type": [
|
| 23 |
+
"c2p",
|
| 24 |
+
"p2c"
|
| 25 |
+
],
|
| 26 |
+
"position_biased_input": false,
|
| 27 |
+
"relative_attention": true,
|
| 28 |
+
"torch_dtype": "float32",
|
| 29 |
+
"transformers_version": "4.34.1",
|
| 30 |
+
"type_vocab_size": 0,
|
| 31 |
+
"vocab_size": 50265
|
| 32 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3cd108cbcabb1c992a8ef771a8c8b10e6536746bb59df8f3fe594f944b4fe9dd
|
| 3 |
+
size 1114120581
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bffef9e8a8535fb769a9e2902d22b13b5875393806760b342f8b469e9f8c51d4
|
| 3 |
+
size 557045809
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:66c278ec8153168bcf4061b78d9ae21f632b80d7671d98d95e7927227c769dc9
|
| 3 |
+
size 14575
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e7963b82deae191fd379ee9c5ffb8e29b33c408089b04774bcf7d8b8c90f1da9
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,903 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7940782904624939,
|
| 3 |
+
"best_model_checkpoint": "models_v3/albertina-100m-pt-41-128-t0-v0/checkpoint-65000",
|
| 4 |
+
"epoch": 9.40529590507886,
|
| 5 |
+
"eval_steps": 5000,
|
| 6 |
+
"global_step": 65000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.07,
|
| 13 |
+
"learning_rate": 3.617421501953408e-06,
|
| 14 |
+
"loss": 1.794,
|
| 15 |
+
"step": 500
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.14,
|
| 19 |
+
"learning_rate": 7.234843003906816e-06,
|
| 20 |
+
"loss": 1.6062,
|
| 21 |
+
"step": 1000
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.22,
|
| 25 |
+
"learning_rate": 1.0852264505860224e-05,
|
| 26 |
+
"loss": 1.5053,
|
| 27 |
+
"step": 1500
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.29,
|
| 31 |
+
"learning_rate": 1.4469686007813631e-05,
|
| 32 |
+
"loss": 1.4352,
|
| 33 |
+
"step": 2000
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.36,
|
| 37 |
+
"learning_rate": 1.808710750976704e-05,
|
| 38 |
+
"loss": 1.3906,
|
| 39 |
+
"step": 2500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.43,
|
| 43 |
+
"learning_rate": 2.1704529011720447e-05,
|
| 44 |
+
"loss": 1.3456,
|
| 45 |
+
"step": 3000
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.51,
|
| 49 |
+
"learning_rate": 2.5321950513673853e-05,
|
| 50 |
+
"loss": 1.3123,
|
| 51 |
+
"step": 3500
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.58,
|
| 55 |
+
"learning_rate": 2.8939372015627263e-05,
|
| 56 |
+
"loss": 1.2811,
|
| 57 |
+
"step": 4000
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.65,
|
| 61 |
+
"learning_rate": 3.255679351758067e-05,
|
| 62 |
+
"loss": 1.2554,
|
| 63 |
+
"step": 4500
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 0.72,
|
| 67 |
+
"learning_rate": 3.617421501953408e-05,
|
| 68 |
+
"loss": 1.2324,
|
| 69 |
+
"step": 5000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 0.72,
|
| 73 |
+
"eval_loss": 1.2159796953201294,
|
| 74 |
+
"eval_runtime": 335.7202,
|
| 75 |
+
"eval_samples_per_second": 384.555,
|
| 76 |
+
"eval_steps_per_second": 1.504,
|
| 77 |
+
"step": 5000
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 0.8,
|
| 81 |
+
"learning_rate": 3.979163652148749e-05,
|
| 82 |
+
"loss": 1.2154,
|
| 83 |
+
"step": 5500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.87,
|
| 87 |
+
"learning_rate": 4.3409058023440894e-05,
|
| 88 |
+
"loss": 1.1969,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.94,
|
| 93 |
+
"learning_rate": 4.70264795253943e-05,
|
| 94 |
+
"loss": 1.1784,
|
| 95 |
+
"step": 6500
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.01,
|
| 99 |
+
"learning_rate": 4.9928455441405815e-05,
|
| 100 |
+
"loss": 1.165,
|
| 101 |
+
"step": 7000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 1.09,
|
| 105 |
+
"learning_rate": 4.9526519718966546e-05,
|
| 106 |
+
"loss": 1.1452,
|
| 107 |
+
"step": 7500
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 1.16,
|
| 111 |
+
"learning_rate": 4.912458399652728e-05,
|
| 112 |
+
"loss": 1.133,
|
| 113 |
+
"step": 8000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.23,
|
| 117 |
+
"learning_rate": 4.8722648274088015e-05,
|
| 118 |
+
"loss": 1.115,
|
| 119 |
+
"step": 8500
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.3,
|
| 123 |
+
"learning_rate": 4.8320712551648746e-05,
|
| 124 |
+
"loss": 1.1026,
|
| 125 |
+
"step": 9000
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 1.37,
|
| 129 |
+
"learning_rate": 4.791877682920948e-05,
|
| 130 |
+
"loss": 1.0872,
|
| 131 |
+
"step": 9500
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 1.45,
|
| 135 |
+
"learning_rate": 4.751684110677021e-05,
|
| 136 |
+
"loss": 1.0802,
|
| 137 |
+
"step": 10000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.45,
|
| 141 |
+
"eval_loss": 1.0682921409606934,
|
| 142 |
+
"eval_runtime": 335.7245,
|
| 143 |
+
"eval_samples_per_second": 384.55,
|
| 144 |
+
"eval_steps_per_second": 1.504,
|
| 145 |
+
"step": 10000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 1.52,
|
| 149 |
+
"learning_rate": 4.711490538433094e-05,
|
| 150 |
+
"loss": 1.0682,
|
| 151 |
+
"step": 10500
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.59,
|
| 155 |
+
"learning_rate": 4.671296966189167e-05,
|
| 156 |
+
"loss": 1.0589,
|
| 157 |
+
"step": 11000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 1.66,
|
| 161 |
+
"learning_rate": 4.63110339394524e-05,
|
| 162 |
+
"loss": 1.0491,
|
| 163 |
+
"step": 11500
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.74,
|
| 167 |
+
"learning_rate": 4.590909821701313e-05,
|
| 168 |
+
"loss": 1.0425,
|
| 169 |
+
"step": 12000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 1.81,
|
| 173 |
+
"learning_rate": 4.550716249457387e-05,
|
| 174 |
+
"loss": 1.0313,
|
| 175 |
+
"step": 12500
|
| 176 |
+
},
|
| 177 |
+
{
|
| 178 |
+
"epoch": 1.88,
|
| 179 |
+
"learning_rate": 4.51052267721346e-05,
|
| 180 |
+
"loss": 1.0224,
|
| 181 |
+
"step": 13000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 1.95,
|
| 185 |
+
"learning_rate": 4.470329104969533e-05,
|
| 186 |
+
"loss": 1.0129,
|
| 187 |
+
"step": 13500
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 2.03,
|
| 191 |
+
"learning_rate": 4.4301355327256064e-05,
|
| 192 |
+
"loss": 1.0066,
|
| 193 |
+
"step": 14000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 2.1,
|
| 197 |
+
"learning_rate": 4.38994196048168e-05,
|
| 198 |
+
"loss": 0.998,
|
| 199 |
+
"step": 14500
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 2.17,
|
| 203 |
+
"learning_rate": 4.349748388237753e-05,
|
| 204 |
+
"loss": 0.9909,
|
| 205 |
+
"step": 15000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 2.17,
|
| 209 |
+
"eval_loss": 0.9950871467590332,
|
| 210 |
+
"eval_runtime": 335.5649,
|
| 211 |
+
"eval_samples_per_second": 384.733,
|
| 212 |
+
"eval_steps_per_second": 1.505,
|
| 213 |
+
"step": 15000
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 2.24,
|
| 217 |
+
"learning_rate": 4.3095548159938264e-05,
|
| 218 |
+
"loss": 0.9861,
|
| 219 |
+
"step": 15500
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 2.32,
|
| 223 |
+
"learning_rate": 4.2693612437498996e-05,
|
| 224 |
+
"loss": 0.9806,
|
| 225 |
+
"step": 16000
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 2.39,
|
| 229 |
+
"learning_rate": 4.2291676715059733e-05,
|
| 230 |
+
"loss": 0.9738,
|
| 231 |
+
"step": 16500
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 2.46,
|
| 235 |
+
"learning_rate": 4.1889740992620465e-05,
|
| 236 |
+
"loss": 0.9713,
|
| 237 |
+
"step": 17000
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.53,
|
| 241 |
+
"learning_rate": 4.1487805270181196e-05,
|
| 242 |
+
"loss": 0.9684,
|
| 243 |
+
"step": 17500
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.6,
|
| 247 |
+
"learning_rate": 4.108586954774193e-05,
|
| 248 |
+
"loss": 0.9595,
|
| 249 |
+
"step": 18000
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 2.68,
|
| 253 |
+
"learning_rate": 4.0683933825302665e-05,
|
| 254 |
+
"loss": 0.9576,
|
| 255 |
+
"step": 18500
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 2.75,
|
| 259 |
+
"learning_rate": 4.0281998102863396e-05,
|
| 260 |
+
"loss": 0.9507,
|
| 261 |
+
"step": 19000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 2.82,
|
| 265 |
+
"learning_rate": 3.988006238042413e-05,
|
| 266 |
+
"loss": 0.946,
|
| 267 |
+
"step": 19500
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.89,
|
| 271 |
+
"learning_rate": 3.947812665798486e-05,
|
| 272 |
+
"loss": 0.9471,
|
| 273 |
+
"step": 20000
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 2.89,
|
| 277 |
+
"eval_loss": 0.947361171245575,
|
| 278 |
+
"eval_runtime": 335.631,
|
| 279 |
+
"eval_samples_per_second": 384.658,
|
| 280 |
+
"eval_steps_per_second": 1.505,
|
| 281 |
+
"step": 20000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.97,
|
| 285 |
+
"learning_rate": 3.907619093554559e-05,
|
| 286 |
+
"loss": 0.939,
|
| 287 |
+
"step": 20500
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 3.04,
|
| 291 |
+
"learning_rate": 3.867425521310632e-05,
|
| 292 |
+
"loss": 0.9318,
|
| 293 |
+
"step": 21000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.11,
|
| 297 |
+
"learning_rate": 3.827231949066705e-05,
|
| 298 |
+
"loss": 0.9259,
|
| 299 |
+
"step": 21500
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 3.18,
|
| 303 |
+
"learning_rate": 3.787038376822778e-05,
|
| 304 |
+
"loss": 0.9219,
|
| 305 |
+
"step": 22000
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 3.26,
|
| 309 |
+
"learning_rate": 3.746844804578852e-05,
|
| 310 |
+
"loss": 0.9174,
|
| 311 |
+
"step": 22500
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 3.33,
|
| 315 |
+
"learning_rate": 3.706651232334925e-05,
|
| 316 |
+
"loss": 0.9179,
|
| 317 |
+
"step": 23000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 3.4,
|
| 321 |
+
"learning_rate": 3.666457660090998e-05,
|
| 322 |
+
"loss": 0.9122,
|
| 323 |
+
"step": 23500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.47,
|
| 327 |
+
"learning_rate": 3.6262640878470714e-05,
|
| 328 |
+
"loss": 0.9121,
|
| 329 |
+
"step": 24000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.55,
|
| 333 |
+
"learning_rate": 3.586070515603145e-05,
|
| 334 |
+
"loss": 0.9079,
|
| 335 |
+
"step": 24500
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.62,
|
| 339 |
+
"learning_rate": 3.545876943359218e-05,
|
| 340 |
+
"loss": 0.9019,
|
| 341 |
+
"step": 25000
|
| 342 |
+
},
|
| 343 |
+
{
|
| 344 |
+
"epoch": 3.62,
|
| 345 |
+
"eval_loss": 0.9143123030662537,
|
| 346 |
+
"eval_runtime": 336.064,
|
| 347 |
+
"eval_samples_per_second": 384.162,
|
| 348 |
+
"eval_steps_per_second": 1.503,
|
| 349 |
+
"step": 25000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 3.69,
|
| 353 |
+
"learning_rate": 3.5056833711152914e-05,
|
| 354 |
+
"loss": 0.8999,
|
| 355 |
+
"step": 25500
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 3.76,
|
| 359 |
+
"learning_rate": 3.4654897988713646e-05,
|
| 360 |
+
"loss": 0.8967,
|
| 361 |
+
"step": 26000
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 3.83,
|
| 365 |
+
"learning_rate": 3.425296226627438e-05,
|
| 366 |
+
"loss": 0.8928,
|
| 367 |
+
"step": 26500
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 3.91,
|
| 371 |
+
"learning_rate": 3.3851026543835115e-05,
|
| 372 |
+
"loss": 0.8937,
|
| 373 |
+
"step": 27000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 3.98,
|
| 377 |
+
"learning_rate": 3.3449090821395846e-05,
|
| 378 |
+
"loss": 0.8882,
|
| 379 |
+
"step": 27500
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 4.05,
|
| 383 |
+
"learning_rate": 3.304715509895658e-05,
|
| 384 |
+
"loss": 0.8859,
|
| 385 |
+
"step": 28000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 4.12,
|
| 389 |
+
"learning_rate": 3.264521937651731e-05,
|
| 390 |
+
"loss": 0.8792,
|
| 391 |
+
"step": 28500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 4.2,
|
| 395 |
+
"learning_rate": 3.2243283654078046e-05,
|
| 396 |
+
"loss": 0.8769,
|
| 397 |
+
"step": 29000
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 4.27,
|
| 401 |
+
"learning_rate": 3.184134793163878e-05,
|
| 402 |
+
"loss": 0.8752,
|
| 403 |
+
"step": 29500
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 4.34,
|
| 407 |
+
"learning_rate": 3.14394122091995e-05,
|
| 408 |
+
"loss": 0.8715,
|
| 409 |
+
"step": 30000
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.34,
|
| 413 |
+
"eval_loss": 0.8864349722862244,
|
| 414 |
+
"eval_runtime": 335.5739,
|
| 415 |
+
"eval_samples_per_second": 384.723,
|
| 416 |
+
"eval_steps_per_second": 1.505,
|
| 417 |
+
"step": 30000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 4.41,
|
| 421 |
+
"learning_rate": 3.103747648676024e-05,
|
| 422 |
+
"loss": 0.8749,
|
| 423 |
+
"step": 30500
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.49,
|
| 427 |
+
"learning_rate": 3.063554076432097e-05,
|
| 428 |
+
"loss": 0.8685,
|
| 429 |
+
"step": 31000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 4.56,
|
| 433 |
+
"learning_rate": 3.02336050418817e-05,
|
| 434 |
+
"loss": 0.8663,
|
| 435 |
+
"step": 31500
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 4.63,
|
| 439 |
+
"learning_rate": 2.9831669319442436e-05,
|
| 440 |
+
"loss": 0.8624,
|
| 441 |
+
"step": 32000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 4.7,
|
| 445 |
+
"learning_rate": 2.9429733597003167e-05,
|
| 446 |
+
"loss": 0.8613,
|
| 447 |
+
"step": 32500
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 4.77,
|
| 451 |
+
"learning_rate": 2.90277978745639e-05,
|
| 452 |
+
"loss": 0.858,
|
| 453 |
+
"step": 33000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 4.85,
|
| 457 |
+
"learning_rate": 2.8625862152124633e-05,
|
| 458 |
+
"loss": 0.8589,
|
| 459 |
+
"step": 33500
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 4.92,
|
| 463 |
+
"learning_rate": 2.8223926429685364e-05,
|
| 464 |
+
"loss": 0.8572,
|
| 465 |
+
"step": 34000
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 4.99,
|
| 469 |
+
"learning_rate": 2.78219907072461e-05,
|
| 470 |
+
"loss": 0.8504,
|
| 471 |
+
"step": 34500
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 5.06,
|
| 475 |
+
"learning_rate": 2.742005498480683e-05,
|
| 476 |
+
"loss": 0.8471,
|
| 477 |
+
"step": 35000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 5.06,
|
| 481 |
+
"eval_loss": 0.8663927316665649,
|
| 482 |
+
"eval_runtime": 335.8915,
|
| 483 |
+
"eval_samples_per_second": 384.359,
|
| 484 |
+
"eval_steps_per_second": 1.503,
|
| 485 |
+
"step": 35000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 5.14,
|
| 489 |
+
"learning_rate": 2.7018119262367564e-05,
|
| 490 |
+
"loss": 0.8465,
|
| 491 |
+
"step": 35500
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.21,
|
| 495 |
+
"learning_rate": 2.6616183539928296e-05,
|
| 496 |
+
"loss": 0.8425,
|
| 497 |
+
"step": 36000
|
| 498 |
+
},
|
| 499 |
+
{
|
| 500 |
+
"epoch": 5.28,
|
| 501 |
+
"learning_rate": 2.621424781748903e-05,
|
| 502 |
+
"loss": 0.8417,
|
| 503 |
+
"step": 36500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 5.35,
|
| 507 |
+
"learning_rate": 2.581231209504976e-05,
|
| 508 |
+
"loss": 0.8385,
|
| 509 |
+
"step": 37000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 5.43,
|
| 513 |
+
"learning_rate": 2.5410376372610496e-05,
|
| 514 |
+
"loss": 0.8394,
|
| 515 |
+
"step": 37500
|
| 516 |
+
},
|
| 517 |
+
{
|
| 518 |
+
"epoch": 5.5,
|
| 519 |
+
"learning_rate": 2.5008440650171227e-05,
|
| 520 |
+
"loss": 0.8373,
|
| 521 |
+
"step": 38000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 5.57,
|
| 525 |
+
"learning_rate": 2.4606504927731958e-05,
|
| 526 |
+
"loss": 0.8344,
|
| 527 |
+
"step": 38500
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 5.64,
|
| 531 |
+
"learning_rate": 2.420456920529269e-05,
|
| 532 |
+
"loss": 0.8347,
|
| 533 |
+
"step": 39000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 5.72,
|
| 537 |
+
"learning_rate": 2.3802633482853424e-05,
|
| 538 |
+
"loss": 0.8324,
|
| 539 |
+
"step": 39500
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 5.79,
|
| 543 |
+
"learning_rate": 2.3400697760414155e-05,
|
| 544 |
+
"loss": 0.8315,
|
| 545 |
+
"step": 40000
|
| 546 |
+
},
|
| 547 |
+
{
|
| 548 |
+
"epoch": 5.79,
|
| 549 |
+
"eval_loss": 0.8464680910110474,
|
| 550 |
+
"eval_runtime": 335.9122,
|
| 551 |
+
"eval_samples_per_second": 384.336,
|
| 552 |
+
"eval_steps_per_second": 1.503,
|
| 553 |
+
"step": 40000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 5.86,
|
| 557 |
+
"learning_rate": 2.299876203797489e-05,
|
| 558 |
+
"loss": 0.8305,
|
| 559 |
+
"step": 40500
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 5.93,
|
| 563 |
+
"learning_rate": 2.259682631553562e-05,
|
| 564 |
+
"loss": 0.8274,
|
| 565 |
+
"step": 41000
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 6.0,
|
| 569 |
+
"learning_rate": 2.2194890593096355e-05,
|
| 570 |
+
"loss": 0.8229,
|
| 571 |
+
"step": 41500
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 6.08,
|
| 575 |
+
"learning_rate": 2.1792954870657086e-05,
|
| 576 |
+
"loss": 0.8202,
|
| 577 |
+
"step": 42000
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 6.15,
|
| 581 |
+
"learning_rate": 2.1391019148217817e-05,
|
| 582 |
+
"loss": 0.8183,
|
| 583 |
+
"step": 42500
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 6.22,
|
| 587 |
+
"learning_rate": 2.098908342577855e-05,
|
| 588 |
+
"loss": 0.8182,
|
| 589 |
+
"step": 43000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 6.29,
|
| 593 |
+
"learning_rate": 2.0587147703339283e-05,
|
| 594 |
+
"loss": 0.8172,
|
| 595 |
+
"step": 43500
|
| 596 |
+
},
|
| 597 |
+
{
|
| 598 |
+
"epoch": 6.37,
|
| 599 |
+
"learning_rate": 2.0185211980900014e-05,
|
| 600 |
+
"loss": 0.815,
|
| 601 |
+
"step": 44000
|
| 602 |
+
},
|
| 603 |
+
{
|
| 604 |
+
"epoch": 6.44,
|
| 605 |
+
"learning_rate": 1.978327625846075e-05,
|
| 606 |
+
"loss": 0.8136,
|
| 607 |
+
"step": 44500
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 6.51,
|
| 611 |
+
"learning_rate": 1.938134053602148e-05,
|
| 612 |
+
"loss": 0.8123,
|
| 613 |
+
"step": 45000
|
| 614 |
+
},
|
| 615 |
+
{
|
| 616 |
+
"epoch": 6.51,
|
| 617 |
+
"eval_loss": 0.8314308524131775,
|
| 618 |
+
"eval_runtime": 335.9134,
|
| 619 |
+
"eval_samples_per_second": 384.334,
|
| 620 |
+
"eval_steps_per_second": 1.503,
|
| 621 |
+
"step": 45000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 6.58,
|
| 625 |
+
"learning_rate": 1.8979404813582214e-05,
|
| 626 |
+
"loss": 0.813,
|
| 627 |
+
"step": 45500
|
| 628 |
+
},
|
| 629 |
+
{
|
| 630 |
+
"epoch": 6.66,
|
| 631 |
+
"learning_rate": 1.8577469091142946e-05,
|
| 632 |
+
"loss": 0.8101,
|
| 633 |
+
"step": 46000
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 6.73,
|
| 637 |
+
"learning_rate": 1.817553336870368e-05,
|
| 638 |
+
"loss": 0.812,
|
| 639 |
+
"step": 46500
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 6.8,
|
| 643 |
+
"learning_rate": 1.777359764626441e-05,
|
| 644 |
+
"loss": 0.8069,
|
| 645 |
+
"step": 47000
|
| 646 |
+
},
|
| 647 |
+
{
|
| 648 |
+
"epoch": 6.87,
|
| 649 |
+
"learning_rate": 1.7371661923825142e-05,
|
| 650 |
+
"loss": 0.8048,
|
| 651 |
+
"step": 47500
|
| 652 |
+
},
|
| 653 |
+
{
|
| 654 |
+
"epoch": 6.95,
|
| 655 |
+
"learning_rate": 1.6969726201385873e-05,
|
| 656 |
+
"loss": 0.8064,
|
| 657 |
+
"step": 48000
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 7.02,
|
| 661 |
+
"learning_rate": 1.6567790478946608e-05,
|
| 662 |
+
"loss": 0.8019,
|
| 663 |
+
"step": 48500
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 7.09,
|
| 667 |
+
"learning_rate": 1.616585475650734e-05,
|
| 668 |
+
"loss": 0.7993,
|
| 669 |
+
"step": 49000
|
| 670 |
+
},
|
| 671 |
+
{
|
| 672 |
+
"epoch": 7.16,
|
| 673 |
+
"learning_rate": 1.5763919034068074e-05,
|
| 674 |
+
"loss": 0.8011,
|
| 675 |
+
"step": 49500
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 7.23,
|
| 679 |
+
"learning_rate": 1.5361983311628805e-05,
|
| 680 |
+
"loss": 0.7977,
|
| 681 |
+
"step": 50000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 7.23,
|
| 685 |
+
"eval_loss": 0.8229328989982605,
|
| 686 |
+
"eval_runtime": 335.9175,
|
| 687 |
+
"eval_samples_per_second": 384.33,
|
| 688 |
+
"eval_steps_per_second": 1.503,
|
| 689 |
+
"step": 50000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 7.31,
|
| 693 |
+
"learning_rate": 1.4960047589189538e-05,
|
| 694 |
+
"loss": 0.797,
|
| 695 |
+
"step": 50500
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 7.38,
|
| 699 |
+
"learning_rate": 1.455811186675027e-05,
|
| 700 |
+
"loss": 0.7943,
|
| 701 |
+
"step": 51000
|
| 702 |
+
},
|
| 703 |
+
{
|
| 704 |
+
"epoch": 7.45,
|
| 705 |
+
"learning_rate": 1.4156176144311003e-05,
|
| 706 |
+
"loss": 0.7934,
|
| 707 |
+
"step": 51500
|
| 708 |
+
},
|
| 709 |
+
{
|
| 710 |
+
"epoch": 7.52,
|
| 711 |
+
"learning_rate": 1.3754240421871733e-05,
|
| 712 |
+
"loss": 0.7926,
|
| 713 |
+
"step": 52000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 7.6,
|
| 717 |
+
"learning_rate": 1.3352304699432466e-05,
|
| 718 |
+
"loss": 0.7939,
|
| 719 |
+
"step": 52500
|
| 720 |
+
},
|
| 721 |
+
{
|
| 722 |
+
"epoch": 7.67,
|
| 723 |
+
"learning_rate": 1.2950368976993198e-05,
|
| 724 |
+
"loss": 0.7922,
|
| 725 |
+
"step": 53000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 7.74,
|
| 729 |
+
"learning_rate": 1.2548433254553931e-05,
|
| 730 |
+
"loss": 0.7919,
|
| 731 |
+
"step": 53500
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 7.81,
|
| 735 |
+
"learning_rate": 1.2146497532114664e-05,
|
| 736 |
+
"loss": 0.7886,
|
| 737 |
+
"step": 54000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 7.89,
|
| 741 |
+
"learning_rate": 1.1744561809675397e-05,
|
| 742 |
+
"loss": 0.7866,
|
| 743 |
+
"step": 54500
|
| 744 |
+
},
|
| 745 |
+
{
|
| 746 |
+
"epoch": 7.96,
|
| 747 |
+
"learning_rate": 1.134262608723613e-05,
|
| 748 |
+
"loss": 0.7856,
|
| 749 |
+
"step": 55000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 7.96,
|
| 753 |
+
"eval_loss": 0.8111724257469177,
|
| 754 |
+
"eval_runtime": 335.2542,
|
| 755 |
+
"eval_samples_per_second": 385.09,
|
| 756 |
+
"eval_steps_per_second": 1.506,
|
| 757 |
+
"step": 55000
|
| 758 |
+
},
|
| 759 |
+
{
|
| 760 |
+
"epoch": 8.03,
|
| 761 |
+
"learning_rate": 1.0940690364796863e-05,
|
| 762 |
+
"loss": 0.787,
|
| 763 |
+
"step": 55500
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 8.1,
|
| 767 |
+
"learning_rate": 1.0538754642357594e-05,
|
| 768 |
+
"loss": 0.7837,
|
| 769 |
+
"step": 56000
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 8.18,
|
| 773 |
+
"learning_rate": 1.0136818919918327e-05,
|
| 774 |
+
"loss": 0.7834,
|
| 775 |
+
"step": 56500
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 8.25,
|
| 779 |
+
"learning_rate": 9.73488319747906e-06,
|
| 780 |
+
"loss": 0.7817,
|
| 781 |
+
"step": 57000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 8.32,
|
| 785 |
+
"learning_rate": 9.332947475039792e-06,
|
| 786 |
+
"loss": 0.7803,
|
| 787 |
+
"step": 57500
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 8.39,
|
| 791 |
+
"learning_rate": 8.931011752600525e-06,
|
| 792 |
+
"loss": 0.7793,
|
| 793 |
+
"step": 58000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 8.46,
|
| 797 |
+
"learning_rate": 8.529076030161256e-06,
|
| 798 |
+
"loss": 0.7802,
|
| 799 |
+
"step": 58500
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 8.54,
|
| 803 |
+
"learning_rate": 8.127140307721989e-06,
|
| 804 |
+
"loss": 0.7788,
|
| 805 |
+
"step": 59000
|
| 806 |
+
},
|
| 807 |
+
{
|
| 808 |
+
"epoch": 8.61,
|
| 809 |
+
"learning_rate": 7.725204585282722e-06,
|
| 810 |
+
"loss": 0.7754,
|
| 811 |
+
"step": 59500
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 8.68,
|
| 815 |
+
"learning_rate": 7.323268862843455e-06,
|
| 816 |
+
"loss": 0.776,
|
| 817 |
+
"step": 60000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 8.68,
|
| 821 |
+
"eval_loss": 0.8020525574684143,
|
| 822 |
+
"eval_runtime": 335.4375,
|
| 823 |
+
"eval_samples_per_second": 384.879,
|
| 824 |
+
"eval_steps_per_second": 1.505,
|
| 825 |
+
"step": 60000
|
| 826 |
+
},
|
| 827 |
+
{
|
| 828 |
+
"epoch": 8.75,
|
| 829 |
+
"learning_rate": 6.921333140404188e-06,
|
| 830 |
+
"loss": 0.7767,
|
| 831 |
+
"step": 60500
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 8.83,
|
| 835 |
+
"learning_rate": 6.519397417964919e-06,
|
| 836 |
+
"loss": 0.7766,
|
| 837 |
+
"step": 61000
|
| 838 |
+
},
|
| 839 |
+
{
|
| 840 |
+
"epoch": 8.9,
|
| 841 |
+
"learning_rate": 6.117461695525652e-06,
|
| 842 |
+
"loss": 0.7727,
|
| 843 |
+
"step": 61500
|
| 844 |
+
},
|
| 845 |
+
{
|
| 846 |
+
"epoch": 8.97,
|
| 847 |
+
"learning_rate": 5.7155259730863845e-06,
|
| 848 |
+
"loss": 0.7739,
|
| 849 |
+
"step": 62000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 9.04,
|
| 853 |
+
"learning_rate": 5.3135902506471165e-06,
|
| 854 |
+
"loss": 0.7723,
|
| 855 |
+
"step": 62500
|
| 856 |
+
},
|
| 857 |
+
{
|
| 858 |
+
"epoch": 9.12,
|
| 859 |
+
"learning_rate": 4.911654528207849e-06,
|
| 860 |
+
"loss": 0.771,
|
| 861 |
+
"step": 63000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 9.19,
|
| 865 |
+
"learning_rate": 4.509718805768582e-06,
|
| 866 |
+
"loss": 0.7674,
|
| 867 |
+
"step": 63500
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 9.26,
|
| 871 |
+
"learning_rate": 4.107783083329314e-06,
|
| 872 |
+
"loss": 0.772,
|
| 873 |
+
"step": 64000
|
| 874 |
+
},
|
| 875 |
+
{
|
| 876 |
+
"epoch": 9.33,
|
| 877 |
+
"learning_rate": 3.705847360890047e-06,
|
| 878 |
+
"loss": 0.7683,
|
| 879 |
+
"step": 64500
|
| 880 |
+
},
|
| 881 |
+
{
|
| 882 |
+
"epoch": 9.41,
|
| 883 |
+
"learning_rate": 3.303911638450779e-06,
|
| 884 |
+
"loss": 0.7705,
|
| 885 |
+
"step": 65000
|
| 886 |
+
},
|
| 887 |
+
{
|
| 888 |
+
"epoch": 9.41,
|
| 889 |
+
"eval_loss": 0.7940782904624939,
|
| 890 |
+
"eval_runtime": 335.4617,
|
| 891 |
+
"eval_samples_per_second": 384.852,
|
| 892 |
+
"eval_steps_per_second": 1.505,
|
| 893 |
+
"step": 65000
|
| 894 |
+
}
|
| 895 |
+
],
|
| 896 |
+
"logging_steps": 500,
|
| 897 |
+
"max_steps": 69110,
|
| 898 |
+
"num_train_epochs": 10,
|
| 899 |
+
"save_steps": 5000,
|
| 900 |
+
"total_flos": 1.2759454800715415e+18,
|
| 901 |
+
"trial_name": null,
|
| 902 |
+
"trial_params": null
|
| 903 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d9d721c0de2c73d20bf070e97a4a3543be0e91217b3b289c26a2ee9a7eb939a
|
| 3 |
+
size 4091
|