Upload folder using huggingface_hub
Browse files- checkpoints/checkpoint-196/config.json +26 -0
- checkpoints/checkpoint-196/model.safetensors +3 -0
- checkpoints/checkpoint-196/optimizer.pt +3 -0
- checkpoints/checkpoint-196/rng_state.pth +3 -0
- checkpoints/checkpoint-196/scheduler.pt +3 -0
- checkpoints/checkpoint-196/trainer_state.json +94 -0
- checkpoints/checkpoint-196/training_args.bin +3 -0
checkpoints/checkpoint-196/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 3072,
|
| 14 |
+
"layer_norm_eps": 1e-05,
|
| 15 |
+
"max_position_embeddings": 514,
|
| 16 |
+
"model_type": "roberta",
|
| 17 |
+
"num_attention_heads": 12,
|
| 18 |
+
"num_hidden_layers": 12,
|
| 19 |
+
"pad_token_id": 1,
|
| 20 |
+
"position_embedding_type": "absolute",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.47.0",
|
| 23 |
+
"type_vocab_size": 1,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 52000
|
| 26 |
+
}
|
checkpoints/checkpoint-196/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:faf57bb580c3d3c14998ea2162cc00e5c7052d9aee4e479a770dce3b71849381
|
| 3 |
+
size 504150808
|
checkpoints/checkpoint-196/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fec95a43bedca95afa436f0dbefe765362c508bf1f7ef692af26eaf37481a439
|
| 3 |
+
size 1008422138
|
checkpoints/checkpoint-196/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56b808955f6aa6bc959de62e625efed0d21d5455f9c564f8926022365bd194f0
|
| 3 |
+
size 14244
|
checkpoints/checkpoint-196/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3384fd8813c0bf3b8096a5bc2604986b91bbaa0867b5a7520f975a8ae4634580
|
| 3 |
+
size 1064
|
checkpoints/checkpoint-196/trainer_state.json
ADDED
|
@@ -0,0 +1,94 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 196,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 1.0,
|
| 13 |
+
"eval_accuracy": 0.0,
|
| 14 |
+
"eval_loss": 10.588387489318848,
|
| 15 |
+
"eval_normalizer": 95717.0,
|
| 16 |
+
"eval_runtime": 91.4198,
|
| 17 |
+
"eval_samples_per_second": 584.742,
|
| 18 |
+
"eval_steps_per_second": 1.149,
|
| 19 |
+
"step": 49
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"epoch": 1.020460358056266,
|
| 23 |
+
"grad_norm": 77.02039337158203,
|
| 24 |
+
"learning_rate": 2.5e-06,
|
| 25 |
+
"loss": 10.4871,
|
| 26 |
+
"step": 50
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 2.0,
|
| 30 |
+
"eval_accuracy": 0.0,
|
| 31 |
+
"eval_loss": 10.317421913146973,
|
| 32 |
+
"eval_normalizer": 95497.0,
|
| 33 |
+
"eval_runtime": 92.7813,
|
| 34 |
+
"eval_samples_per_second": 576.162,
|
| 35 |
+
"eval_steps_per_second": 1.132,
|
| 36 |
+
"step": 98
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 2.040920716112532,
|
| 40 |
+
"grad_norm": 93.76717376708984,
|
| 41 |
+
"learning_rate": 5e-06,
|
| 42 |
+
"loss": 9.3486,
|
| 43 |
+
"step": 100
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.0,
|
| 47 |
+
"eval_accuracy": 0.0,
|
| 48 |
+
"eval_loss": 9.76750373840332,
|
| 49 |
+
"eval_normalizer": 95100.0,
|
| 50 |
+
"eval_runtime": 93.226,
|
| 51 |
+
"eval_samples_per_second": 573.413,
|
| 52 |
+
"eval_steps_per_second": 1.126,
|
| 53 |
+
"step": 147
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 3.061381074168798,
|
| 57 |
+
"grad_norm": 38.99561309814453,
|
| 58 |
+
"learning_rate": 7.5e-06,
|
| 59 |
+
"loss": 9.7762,
|
| 60 |
+
"step": 150
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"eval_accuracy": 0.0,
|
| 65 |
+
"eval_loss": 9.366472244262695,
|
| 66 |
+
"eval_normalizer": 94635.0,
|
| 67 |
+
"eval_runtime": 93.4852,
|
| 68 |
+
"eval_samples_per_second": 571.823,
|
| 69 |
+
"eval_steps_per_second": 1.123,
|
| 70 |
+
"step": 196
|
| 71 |
+
}
|
| 72 |
+
],
|
| 73 |
+
"logging_steps": 50,
|
| 74 |
+
"max_steps": 480,
|
| 75 |
+
"num_input_tokens_seen": 0,
|
| 76 |
+
"num_train_epochs": 10,
|
| 77 |
+
"save_steps": 500,
|
| 78 |
+
"stateful_callbacks": {
|
| 79 |
+
"TrainerControl": {
|
| 80 |
+
"args": {
|
| 81 |
+
"should_epoch_stop": false,
|
| 82 |
+
"should_evaluate": false,
|
| 83 |
+
"should_log": false,
|
| 84 |
+
"should_save": true,
|
| 85 |
+
"should_training_stop": false
|
| 86 |
+
},
|
| 87 |
+
"attributes": {}
|
| 88 |
+
}
|
| 89 |
+
},
|
| 90 |
+
"total_flos": 2.9567959240513536e+16,
|
| 91 |
+
"train_batch_size": 128,
|
| 92 |
+
"trial_name": null,
|
| 93 |
+
"trial_params": null
|
| 94 |
+
}
|
checkpoints/checkpoint-196/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6db2b7d30557896b8db73563f526aa290bbad6339930dcde51cfab52b2f61e39
|
| 3 |
+
size 5432
|