Training in progress, epoch 1
Browse files- model.safetensors +1 -1
- run-2/checkpoint-6000/config.json +65 -0
- run-2/checkpoint-6000/model.safetensors +3 -0
- run-2/checkpoint-6000/optimizer.pt +3 -0
- run-2/checkpoint-6000/rng_state.pth +3 -0
- run-2/checkpoint-6000/scaler.pt +3 -0
- run-2/checkpoint-6000/scheduler.pt +3 -0
- run-2/checkpoint-6000/trainer_state.json +602 -0
- run-2/checkpoint-6000/training_args.bin +3 -0
- run-2/checkpoint-7000/config.json +65 -0
- run-2/checkpoint-7000/model.safetensors +3 -0
- run-2/checkpoint-7000/optimizer.pt +3 -0
- run-2/checkpoint-7000/rng_state.pth +3 -0
- run-2/checkpoint-7000/scaler.pt +3 -0
- run-2/checkpoint-7000/scheduler.pt +3 -0
- run-2/checkpoint-7000/trainer_state.json +694 -0
- run-2/checkpoint-7000/training_args.bin +3 -0
- run-2/checkpoint-8000/config.json +65 -0
- run-2/checkpoint-8000/model.safetensors +3 -0
- run-2/checkpoint-8000/optimizer.pt +3 -0
- run-2/checkpoint-8000/rng_state.pth +3 -0
- run-2/checkpoint-8000/scaler.pt +3 -0
- run-2/checkpoint-8000/scheduler.pt +3 -0
- run-2/checkpoint-8000/trainer_state.json +786 -0
- run-2/checkpoint-8000/training_args.bin +3 -0
- run-3/checkpoint-500/config.json +65 -0
- run-3/checkpoint-500/model.safetensors +3 -0
- run-3/checkpoint-500/optimizer.pt +3 -0
- run-3/checkpoint-500/rng_state.pth +3 -0
- run-3/checkpoint-500/scaler.pt +3 -0
- run-3/checkpoint-500/scheduler.pt +3 -0
- run-3/checkpoint-500/trainer_state.json +107 -0
- run-3/checkpoint-500/training_args.bin +3 -0
- training_args.bin +1 -1
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 438007864
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57f4e2fd1ddd2ee909874242196f4c4a3926a6f9a204a8e0273959572b89fdfb
|
| 3 |
size 438007864
|
run-2/checkpoint-6000/config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4",
|
| 17 |
+
"5": "LABEL_5",
|
| 18 |
+
"6": "LABEL_6",
|
| 19 |
+
"7": "LABEL_7",
|
| 20 |
+
"8": "LABEL_8",
|
| 21 |
+
"9": "LABEL_9",
|
| 22 |
+
"10": "LABEL_10",
|
| 23 |
+
"11": "LABEL_11",
|
| 24 |
+
"12": "LABEL_12",
|
| 25 |
+
"13": "LABEL_13",
|
| 26 |
+
"14": "LABEL_14",
|
| 27 |
+
"15": "LABEL_15",
|
| 28 |
+
"16": "LABEL_16",
|
| 29 |
+
"17": "LABEL_17"
|
| 30 |
+
},
|
| 31 |
+
"initializer_range": 0.02,
|
| 32 |
+
"intermediate_size": 3072,
|
| 33 |
+
"label2id": {
|
| 34 |
+
"LABEL_0": 0,
|
| 35 |
+
"LABEL_1": 1,
|
| 36 |
+
"LABEL_10": 10,
|
| 37 |
+
"LABEL_11": 11,
|
| 38 |
+
"LABEL_12": 12,
|
| 39 |
+
"LABEL_13": 13,
|
| 40 |
+
"LABEL_14": 14,
|
| 41 |
+
"LABEL_15": 15,
|
| 42 |
+
"LABEL_16": 16,
|
| 43 |
+
"LABEL_17": 17,
|
| 44 |
+
"LABEL_2": 2,
|
| 45 |
+
"LABEL_3": 3,
|
| 46 |
+
"LABEL_4": 4,
|
| 47 |
+
"LABEL_5": 5,
|
| 48 |
+
"LABEL_6": 6,
|
| 49 |
+
"LABEL_7": 7,
|
| 50 |
+
"LABEL_8": 8,
|
| 51 |
+
"LABEL_9": 9
|
| 52 |
+
},
|
| 53 |
+
"layer_norm_eps": 1e-12,
|
| 54 |
+
"max_position_embeddings": 512,
|
| 55 |
+
"model_type": "bert",
|
| 56 |
+
"num_attention_heads": 12,
|
| 57 |
+
"num_hidden_layers": 12,
|
| 58 |
+
"pad_token_id": 0,
|
| 59 |
+
"position_embedding_type": "absolute",
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.55.4",
|
| 62 |
+
"type_vocab_size": 2,
|
| 63 |
+
"use_cache": true,
|
| 64 |
+
"vocab_size": 30522
|
| 65 |
+
}
|
run-2/checkpoint-6000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:46e6f233b374642b075e77f3fee2e31955dcd0f6443298304b1b8a3f4f8efbb3
|
| 3 |
+
size 438007864
|
run-2/checkpoint-6000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:afb009893cc643662c62a411be6609b8d1513dca53cfb21ee41824ffd7fb9b91
|
| 3 |
+
size 876140043
|
run-2/checkpoint-6000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d2dbc29a654aed70f4abb08c5f8c7a5f58694eaa11f2aefeeb4616e106c8e946
|
| 3 |
+
size 14645
|
run-2/checkpoint-6000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5e0ac840cd1bf6973fd944e08a7d15f1bf99901004c0533e4b74dac6bdd275b
|
| 3 |
+
size 1383
|
run-2/checkpoint-6000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c79044708f262246493386bf6028ebc1da0866a058724a6bb4dfed359152c7d
|
| 3 |
+
size 1465
|
run-2/checkpoint-6000/trainer_state.json
ADDED
|
@@ -0,0 +1,602 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6000,
|
| 3 |
+
"best_metric": 0.8207234712908512,
|
| 4 |
+
"best_model_checkpoint": "./results/run-2/checkpoint-6000",
|
| 5 |
+
"epoch": 6.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 6000,
|
| 8 |
+
"is_hyper_param_search": true,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.1,
|
| 14 |
+
"grad_norm": 3.6001648902893066,
|
| 15 |
+
"learning_rate": 3.1887409174919234e-06,
|
| 16 |
+
"loss": 0.0412,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.2,
|
| 21 |
+
"grad_norm": 0.6239789128303528,
|
| 22 |
+
"learning_rate": 6.409691339200936e-06,
|
| 23 |
+
"loss": 0.1091,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.3,
|
| 28 |
+
"grad_norm": 0.6479267477989197,
|
| 29 |
+
"learning_rate": 9.63064176090995e-06,
|
| 30 |
+
"loss": 0.0,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.4,
|
| 35 |
+
"grad_norm": 20.527082443237305,
|
| 36 |
+
"learning_rate": 1.054959200884952e-05,
|
| 37 |
+
"loss": 0.0,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"grad_norm": 0.8853731751441956,
|
| 43 |
+
"learning_rate": 1.0481970753745042e-05,
|
| 44 |
+
"loss": 0.0,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.6,
|
| 49 |
+
"grad_norm": 1.839735507965088,
|
| 50 |
+
"learning_rate": 1.0414349498640566e-05,
|
| 51 |
+
"loss": 0.0,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.7,
|
| 56 |
+
"grad_norm": 2.7137391567230225,
|
| 57 |
+
"learning_rate": 1.0346728243536088e-05,
|
| 58 |
+
"loss": 0.0284,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.8,
|
| 63 |
+
"grad_norm": 0.6026535630226135,
|
| 64 |
+
"learning_rate": 1.027910698843161e-05,
|
| 65 |
+
"loss": 0.1612,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.9,
|
| 70 |
+
"grad_norm": 0.5444966554641724,
|
| 71 |
+
"learning_rate": 1.0211485733327134e-05,
|
| 72 |
+
"loss": 0.0,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.0,
|
| 77 |
+
"grad_norm": 1.2413469552993774,
|
| 78 |
+
"learning_rate": 1.0143864478222655e-05,
|
| 79 |
+
"loss": 0.0986,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0,
|
| 84 |
+
"eval_economic_inequality_accuracy": 0.8047858942065491,
|
| 85 |
+
"eval_economic_inequality_f1": 0.7538941327354425,
|
| 86 |
+
"eval_economic_policy_benefits_accuracy": 0.8282828282828283,
|
| 87 |
+
"eval_economic_policy_benefits_f1": 0.764129475069552,
|
| 88 |
+
"eval_ethnic_boundaries_accuracy": 0.9493464052287581,
|
| 89 |
+
"eval_ethnic_boundaries_f1": 0.9281293325410972,
|
| 90 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 91 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 92 |
+
"eval_loss": NaN,
|
| 93 |
+
"eval_mother_tongue_education_accuracy": 0.8888888888888888,
|
| 94 |
+
"eval_mother_tongue_education_f1": 0.837037037037037,
|
| 95 |
+
"eval_overall_accuracy": 0.8390088178776338,
|
| 96 |
+
"eval_overall_f1": 0.7839408748610676,
|
| 97 |
+
"eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
|
| 98 |
+
"eval_religion_ethnic_identity_f1": 0.8133124146404199,
|
| 99 |
+
"eval_runtime": 4.4593,
|
| 100 |
+
"eval_samples_per_second": 897.004,
|
| 101 |
+
"eval_steps_per_second": 56.063,
|
| 102 |
+
"step": 1000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1,
|
| 106 |
+
"grad_norm": 0.5792437195777893,
|
| 107 |
+
"learning_rate": 1.0076243223118178e-05,
|
| 108 |
+
"loss": 0.0,
|
| 109 |
+
"step": 1100
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.2,
|
| 113 |
+
"grad_norm": 8.802961349487305,
|
| 114 |
+
"learning_rate": 1.0008621968013701e-05,
|
| 115 |
+
"loss": 0.0297,
|
| 116 |
+
"step": 1200
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.3,
|
| 120 |
+
"grad_norm": 3.981771469116211,
|
| 121 |
+
"learning_rate": 9.941000712909224e-06,
|
| 122 |
+
"loss": 0.0,
|
| 123 |
+
"step": 1300
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.4,
|
| 127 |
+
"grad_norm": 0.9215607047080994,
|
| 128 |
+
"learning_rate": 9.873379457804745e-06,
|
| 129 |
+
"loss": 0.263,
|
| 130 |
+
"step": 1400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.5,
|
| 134 |
+
"grad_norm": 1.5474179983139038,
|
| 135 |
+
"learning_rate": 9.805758202700268e-06,
|
| 136 |
+
"loss": 0.0,
|
| 137 |
+
"step": 1500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.6,
|
| 141 |
+
"grad_norm": 1.1813520193099976,
|
| 142 |
+
"learning_rate": 9.738136947595791e-06,
|
| 143 |
+
"loss": 0.0,
|
| 144 |
+
"step": 1600
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.7,
|
| 148 |
+
"grad_norm": 5.511495113372803,
|
| 149 |
+
"learning_rate": 9.670515692491313e-06,
|
| 150 |
+
"loss": 0.0037,
|
| 151 |
+
"step": 1700
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.8,
|
| 155 |
+
"grad_norm": 3.1629152297973633,
|
| 156 |
+
"learning_rate": 9.602894437386836e-06,
|
| 157 |
+
"loss": 0.0027,
|
| 158 |
+
"step": 1800
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.9,
|
| 162 |
+
"grad_norm": 0.6996293663978577,
|
| 163 |
+
"learning_rate": 9.535273182282359e-06,
|
| 164 |
+
"loss": 19.1531,
|
| 165 |
+
"step": 1900
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.0,
|
| 169 |
+
"grad_norm": 1.3346213102340698,
|
| 170 |
+
"learning_rate": 9.467651927177882e-06,
|
| 171 |
+
"loss": 2.9418,
|
| 172 |
+
"step": 2000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.0,
|
| 176 |
+
"eval_economic_inequality_accuracy": 0.7896725440806045,
|
| 177 |
+
"eval_economic_inequality_f1": 0.7951291586358689,
|
| 178 |
+
"eval_economic_policy_benefits_accuracy": 0.8148148148148148,
|
| 179 |
+
"eval_economic_policy_benefits_f1": 0.8131577832835695,
|
| 180 |
+
"eval_ethnic_boundaries_accuracy": 0.9264705882352942,
|
| 181 |
+
"eval_ethnic_boundaries_f1": 0.9324831159457648,
|
| 182 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 183 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 184 |
+
"eval_loss": NaN,
|
| 185 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 186 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 187 |
+
"eval_overall_accuracy": 0.8103022392805538,
|
| 188 |
+
"eval_overall_f1": 0.7884819842273197,
|
| 189 |
+
"eval_religion_ethnic_identity_accuracy": 0.8471953578336557,
|
| 190 |
+
"eval_religion_ethnic_identity_f1": 0.8422382496151178,
|
| 191 |
+
"eval_runtime": 4.499,
|
| 192 |
+
"eval_samples_per_second": 889.087,
|
| 193 |
+
"eval_steps_per_second": 55.568,
|
| 194 |
+
"step": 2000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.1,
|
| 198 |
+
"grad_norm": 133.0632781982422,
|
| 199 |
+
"learning_rate": 9.400030672073403e-06,
|
| 200 |
+
"loss": 0.0,
|
| 201 |
+
"step": 2100
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.2,
|
| 205 |
+
"grad_norm": 1.1462996006011963,
|
| 206 |
+
"learning_rate": 9.332409416968926e-06,
|
| 207 |
+
"loss": 0.0,
|
| 208 |
+
"step": 2200
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.3,
|
| 212 |
+
"grad_norm": 0.11461023986339569,
|
| 213 |
+
"learning_rate": 9.26478816186445e-06,
|
| 214 |
+
"loss": 0.0,
|
| 215 |
+
"step": 2300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.4,
|
| 219 |
+
"grad_norm": 0.37900322675704956,
|
| 220 |
+
"learning_rate": 9.197166906759972e-06,
|
| 221 |
+
"loss": 0.0046,
|
| 222 |
+
"step": 2400
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.5,
|
| 226 |
+
"grad_norm": 1.0005897283554077,
|
| 227 |
+
"learning_rate": 9.129545651655495e-06,
|
| 228 |
+
"loss": 0.0,
|
| 229 |
+
"step": 2500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.6,
|
| 233 |
+
"grad_norm": 0.15003152191638947,
|
| 234 |
+
"learning_rate": 9.061924396551017e-06,
|
| 235 |
+
"loss": 0.0,
|
| 236 |
+
"step": 2600
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.7,
|
| 240 |
+
"grad_norm": 1.9842479228973389,
|
| 241 |
+
"learning_rate": 8.99430314144654e-06,
|
| 242 |
+
"loss": 0.0017,
|
| 243 |
+
"step": 2700
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.8,
|
| 247 |
+
"grad_norm": 4.747731685638428,
|
| 248 |
+
"learning_rate": 8.926681886342063e-06,
|
| 249 |
+
"loss": 0.0,
|
| 250 |
+
"step": 2800
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.9,
|
| 254 |
+
"grad_norm": 0.4671611189842224,
|
| 255 |
+
"learning_rate": 8.859060631237586e-06,
|
| 256 |
+
"loss": 0.0,
|
| 257 |
+
"step": 2900
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0,
|
| 261 |
+
"grad_norm": 0.9774578213691711,
|
| 262 |
+
"learning_rate": 8.791439376133107e-06,
|
| 263 |
+
"loss": 0.0,
|
| 264 |
+
"step": 3000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.0,
|
| 268 |
+
"eval_economic_inequality_accuracy": 0.8173803526448362,
|
| 269 |
+
"eval_economic_inequality_f1": 0.8105151543481738,
|
| 270 |
+
"eval_economic_policy_benefits_accuracy": 0.8686868686868687,
|
| 271 |
+
"eval_economic_policy_benefits_f1": 0.8544859369266805,
|
| 272 |
+
"eval_ethnic_boundaries_accuracy": 0.9330065359477124,
|
| 273 |
+
"eval_ethnic_boundaries_f1": 0.9342620574580529,
|
| 274 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 275 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 276 |
+
"eval_loss": NaN,
|
| 277 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 278 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 279 |
+
"eval_overall_accuracy": 0.8266000709526621,
|
| 280 |
+
"eval_overall_f1": 0.8006341340755991,
|
| 281 |
+
"eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
|
| 282 |
+
"eval_religion_ethnic_identity_f1": 0.85665805783709,
|
| 283 |
+
"eval_runtime": 4.4861,
|
| 284 |
+
"eval_samples_per_second": 891.646,
|
| 285 |
+
"eval_steps_per_second": 55.728,
|
| 286 |
+
"step": 3000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.1,
|
| 290 |
+
"grad_norm": 0.6768048405647278,
|
| 291 |
+
"learning_rate": 8.72381812102863e-06,
|
| 292 |
+
"loss": 0.0002,
|
| 293 |
+
"step": 3100
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.2,
|
| 297 |
+
"grad_norm": 0.13458865880966187,
|
| 298 |
+
"learning_rate": 8.656196865924153e-06,
|
| 299 |
+
"loss": 0.0006,
|
| 300 |
+
"step": 3200
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.3,
|
| 304 |
+
"grad_norm": 2.0157546997070312,
|
| 305 |
+
"learning_rate": 8.588575610819674e-06,
|
| 306 |
+
"loss": 0.0,
|
| 307 |
+
"step": 3300
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.4,
|
| 311 |
+
"grad_norm": 0.07812555134296417,
|
| 312 |
+
"learning_rate": 8.520954355715199e-06,
|
| 313 |
+
"loss": 0.0,
|
| 314 |
+
"step": 3400
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"grad_norm": 149.58753967285156,
|
| 319 |
+
"learning_rate": 8.45333310061072e-06,
|
| 320 |
+
"loss": 0.0032,
|
| 321 |
+
"step": 3500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.6,
|
| 325 |
+
"grad_norm": 3.429203748703003,
|
| 326 |
+
"learning_rate": 8.385711845506243e-06,
|
| 327 |
+
"loss": 0.0,
|
| 328 |
+
"step": 3600
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.7,
|
| 332 |
+
"grad_norm": 1.6850093603134155,
|
| 333 |
+
"learning_rate": 8.318090590401766e-06,
|
| 334 |
+
"loss": 0.0003,
|
| 335 |
+
"step": 3700
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.8,
|
| 339 |
+
"grad_norm": 0.06944789737462997,
|
| 340 |
+
"learning_rate": 8.250469335297288e-06,
|
| 341 |
+
"loss": 0.0,
|
| 342 |
+
"step": 3800
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 3.9,
|
| 346 |
+
"grad_norm": 0.11943838745355606,
|
| 347 |
+
"learning_rate": 8.18284808019281e-06,
|
| 348 |
+
"loss": 0.0,
|
| 349 |
+
"step": 3900
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.0,
|
| 353 |
+
"grad_norm": 0.308050274848938,
|
| 354 |
+
"learning_rate": 8.115226825088332e-06,
|
| 355 |
+
"loss": 0.0,
|
| 356 |
+
"step": 4000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 4.0,
|
| 360 |
+
"eval_economic_inequality_accuracy": 0.8085642317380353,
|
| 361 |
+
"eval_economic_inequality_f1": 0.810531309630811,
|
| 362 |
+
"eval_economic_policy_benefits_accuracy": 0.8653198653198653,
|
| 363 |
+
"eval_economic_policy_benefits_f1": 0.8539054809077693,
|
| 364 |
+
"eval_ethnic_boundaries_accuracy": 0.9362745098039216,
|
| 365 |
+
"eval_ethnic_boundaries_f1": 0.9400162685161402,
|
| 366 |
+
"eval_language_policy_accuracy": 0.6764705882352942,
|
| 367 |
+
"eval_language_policy_f1": 0.6150075414781296,
|
| 368 |
+
"eval_loss": NaN,
|
| 369 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 370 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 371 |
+
"eval_overall_accuracy": 0.8169885251374339,
|
| 372 |
+
"eval_overall_f1": 0.8070145214695552,
|
| 373 |
+
"eval_religion_ethnic_identity_accuracy": 0.8375241779497099,
|
| 374 |
+
"eval_religion_ethnic_identity_f1": 0.8491222547802076,
|
| 375 |
+
"eval_runtime": 4.4658,
|
| 376 |
+
"eval_samples_per_second": 895.686,
|
| 377 |
+
"eval_steps_per_second": 55.98,
|
| 378 |
+
"step": 4000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 4.1,
|
| 382 |
+
"grad_norm": 0.2866123616695404,
|
| 383 |
+
"learning_rate": 8.047605569983857e-06,
|
| 384 |
+
"loss": 0.0,
|
| 385 |
+
"step": 4100
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 4.2,
|
| 389 |
+
"grad_norm": 0.26467469334602356,
|
| 390 |
+
"learning_rate": 7.979984314879378e-06,
|
| 391 |
+
"loss": 0.0,
|
| 392 |
+
"step": 4200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.3,
|
| 396 |
+
"grad_norm": 0.06363595277070999,
|
| 397 |
+
"learning_rate": 7.912363059774901e-06,
|
| 398 |
+
"loss": 0.0,
|
| 399 |
+
"step": 4300
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.4,
|
| 403 |
+
"grad_norm": 0.2771783173084259,
|
| 404 |
+
"learning_rate": 7.844741804670424e-06,
|
| 405 |
+
"loss": 0.0002,
|
| 406 |
+
"step": 4400
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.5,
|
| 410 |
+
"grad_norm": 0.022821573540568352,
|
| 411 |
+
"learning_rate": 7.777120549565947e-06,
|
| 412 |
+
"loss": 0.0006,
|
| 413 |
+
"step": 4500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.6,
|
| 417 |
+
"grad_norm": 0.7363903522491455,
|
| 418 |
+
"learning_rate": 7.709499294461468e-06,
|
| 419 |
+
"loss": 0.0,
|
| 420 |
+
"step": 4600
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.7,
|
| 424 |
+
"grad_norm": 1.3803585767745972,
|
| 425 |
+
"learning_rate": 7.641878039356991e-06,
|
| 426 |
+
"loss": 0.0,
|
| 427 |
+
"step": 4700
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 4.8,
|
| 431 |
+
"grad_norm": 0.14331550896167755,
|
| 432 |
+
"learning_rate": 7.5742567842525145e-06,
|
| 433 |
+
"loss": 0.0019,
|
| 434 |
+
"step": 4800
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.9,
|
| 438 |
+
"grad_norm": 0.06504154205322266,
|
| 439 |
+
"learning_rate": 7.506635529148037e-06,
|
| 440 |
+
"loss": 0.0,
|
| 441 |
+
"step": 4900
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 5.0,
|
| 445 |
+
"grad_norm": 0.09633997082710266,
|
| 446 |
+
"learning_rate": 7.43901427404356e-06,
|
| 447 |
+
"loss": 0.0,
|
| 448 |
+
"step": 5000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 5.0,
|
| 452 |
+
"eval_economic_inequality_accuracy": 0.8249370277078085,
|
| 453 |
+
"eval_economic_inequality_f1": 0.823342228819045,
|
| 454 |
+
"eval_economic_policy_benefits_accuracy": 0.8451178451178452,
|
| 455 |
+
"eval_economic_policy_benefits_f1": 0.8327902682741393,
|
| 456 |
+
"eval_ethnic_boundaries_accuracy": 0.9395424836601307,
|
| 457 |
+
"eval_ethnic_boundaries_f1": 0.9429587969409783,
|
| 458 |
+
"eval_language_policy_accuracy": 0.6470588235294118,
|
| 459 |
+
"eval_language_policy_f1": 0.5939135835329606,
|
| 460 |
+
"eval_loss": NaN,
|
| 461 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 462 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 463 |
+
"eval_overall_accuracy": 0.8165062399029553,
|
| 464 |
+
"eval_overall_f1": 0.8063046374512969,
|
| 465 |
+
"eval_religion_ethnic_identity_accuracy": 0.8646034816247582,
|
| 466 |
+
"eval_religion_ethnic_identity_f1": 0.8713186736363845,
|
| 467 |
+
"eval_runtime": 4.4849,
|
| 468 |
+
"eval_samples_per_second": 891.887,
|
| 469 |
+
"eval_steps_per_second": 55.743,
|
| 470 |
+
"step": 5000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 5.1,
|
| 474 |
+
"grad_norm": 2.563802480697632,
|
| 475 |
+
"learning_rate": 7.371393018939082e-06,
|
| 476 |
+
"loss": 0.0049,
|
| 477 |
+
"step": 5100
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 5.2,
|
| 481 |
+
"grad_norm": 10.922028541564941,
|
| 482 |
+
"learning_rate": 7.303771763834604e-06,
|
| 483 |
+
"loss": 0.0004,
|
| 484 |
+
"step": 5200
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 5.3,
|
| 488 |
+
"grad_norm": 0.9929112792015076,
|
| 489 |
+
"learning_rate": 7.236150508730128e-06,
|
| 490 |
+
"loss": 0.0827,
|
| 491 |
+
"step": 5300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.4,
|
| 495 |
+
"grad_norm": 0.13412432372570038,
|
| 496 |
+
"learning_rate": 7.16852925362565e-06,
|
| 497 |
+
"loss": 0.0,
|
| 498 |
+
"step": 5400
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 5.5,
|
| 502 |
+
"grad_norm": 0.0726282000541687,
|
| 503 |
+
"learning_rate": 7.100907998521172e-06,
|
| 504 |
+
"loss": 0.0,
|
| 505 |
+
"step": 5500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.6,
|
| 509 |
+
"grad_norm": 12.0023832321167,
|
| 510 |
+
"learning_rate": 7.033286743416695e-06,
|
| 511 |
+
"loss": 0.0,
|
| 512 |
+
"step": 5600
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.7,
|
| 516 |
+
"grad_norm": 1.0574413537979126,
|
| 517 |
+
"learning_rate": 6.9656654883122174e-06,
|
| 518 |
+
"loss": 0.0,
|
| 519 |
+
"step": 5700
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.8,
|
| 523 |
+
"grad_norm": 0.3997838795185089,
|
| 524 |
+
"learning_rate": 6.89804423320774e-06,
|
| 525 |
+
"loss": 0.0,
|
| 526 |
+
"step": 5800
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.9,
|
| 530 |
+
"grad_norm": 2.0076065063476562,
|
| 531 |
+
"learning_rate": 6.830422978103262e-06,
|
| 532 |
+
"loss": 0.0,
|
| 533 |
+
"step": 5900
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 6.0,
|
| 537 |
+
"grad_norm": 2.676760673522949,
|
| 538 |
+
"learning_rate": 6.762801722998786e-06,
|
| 539 |
+
"loss": 0.0,
|
| 540 |
+
"step": 6000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 6.0,
|
| 544 |
+
"eval_economic_inequality_accuracy": 0.8350125944584383,
|
| 545 |
+
"eval_economic_inequality_f1": 0.8243306188572417,
|
| 546 |
+
"eval_economic_policy_benefits_accuracy": 0.8619528619528619,
|
| 547 |
+
"eval_economic_policy_benefits_f1": 0.840430611429002,
|
| 548 |
+
"eval_ethnic_boundaries_accuracy": 0.9477124183006536,
|
| 549 |
+
"eval_ethnic_boundaries_f1": 0.9385901883316666,
|
| 550 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 551 |
+
"eval_language_policy_f1": 0.6172459893048128,
|
| 552 |
+
"eval_loss": NaN,
|
| 553 |
+
"eval_mother_tongue_education_accuracy": 0.8888888888888888,
|
| 554 |
+
"eval_mother_tongue_education_f1": 0.837037037037037,
|
| 555 |
+
"eval_overall_accuracy": 0.8529320416673837,
|
| 556 |
+
"eval_overall_f1": 0.8207234712908512,
|
| 557 |
+
"eval_religion_ethnic_identity_accuracy": 0.8781431334622823,
|
| 558 |
+
"eval_religion_ethnic_identity_f1": 0.8667063827853473,
|
| 559 |
+
"eval_runtime": 4.4702,
|
| 560 |
+
"eval_samples_per_second": 894.82,
|
| 561 |
+
"eval_steps_per_second": 55.926,
|
| 562 |
+
"step": 6000
|
| 563 |
+
}
|
| 564 |
+
],
|
| 565 |
+
"logging_steps": 100,
|
| 566 |
+
"max_steps": 16000,
|
| 567 |
+
"num_input_tokens_seen": 0,
|
| 568 |
+
"num_train_epochs": 16,
|
| 569 |
+
"save_steps": 500,
|
| 570 |
+
"stateful_callbacks": {
|
| 571 |
+
"EarlyStoppingCallback": {
|
| 572 |
+
"args": {
|
| 573 |
+
"early_stopping_patience": 2,
|
| 574 |
+
"early_stopping_threshold": 0.0
|
| 575 |
+
},
|
| 576 |
+
"attributes": {
|
| 577 |
+
"early_stopping_patience_counter": 0
|
| 578 |
+
}
|
| 579 |
+
},
|
| 580 |
+
"TrainerControl": {
|
| 581 |
+
"args": {
|
| 582 |
+
"should_epoch_stop": false,
|
| 583 |
+
"should_evaluate": false,
|
| 584 |
+
"should_log": false,
|
| 585 |
+
"should_save": true,
|
| 586 |
+
"should_training_stop": false
|
| 587 |
+
},
|
| 588 |
+
"attributes": {}
|
| 589 |
+
}
|
| 590 |
+
},
|
| 591 |
+
"total_flos": 2.5262289911808e+16,
|
| 592 |
+
"train_batch_size": 8,
|
| 593 |
+
"trial_name": null,
|
| 594 |
+
"trial_params": {
|
| 595 |
+
"gradient_accumulation_steps": 2,
|
| 596 |
+
"learning_rate": 1.0596926887422654e-05,
|
| 597 |
+
"num_train_epochs": 16,
|
| 598 |
+
"per_device_train_batch_size": 8,
|
| 599 |
+
"warmup_steps": 329,
|
| 600 |
+
"weight_decay": 0.059185379025811405
|
| 601 |
+
}
|
| 602 |
+
}
|
run-2/checkpoint-6000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:822a4a47339dd0b384da8628d960a86b7107ee8196c01d2a9c3866ddc7aa4cd5
|
| 3 |
+
size 5777
|
run-2/checkpoint-7000/config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4",
|
| 17 |
+
"5": "LABEL_5",
|
| 18 |
+
"6": "LABEL_6",
|
| 19 |
+
"7": "LABEL_7",
|
| 20 |
+
"8": "LABEL_8",
|
| 21 |
+
"9": "LABEL_9",
|
| 22 |
+
"10": "LABEL_10",
|
| 23 |
+
"11": "LABEL_11",
|
| 24 |
+
"12": "LABEL_12",
|
| 25 |
+
"13": "LABEL_13",
|
| 26 |
+
"14": "LABEL_14",
|
| 27 |
+
"15": "LABEL_15",
|
| 28 |
+
"16": "LABEL_16",
|
| 29 |
+
"17": "LABEL_17"
|
| 30 |
+
},
|
| 31 |
+
"initializer_range": 0.02,
|
| 32 |
+
"intermediate_size": 3072,
|
| 33 |
+
"label2id": {
|
| 34 |
+
"LABEL_0": 0,
|
| 35 |
+
"LABEL_1": 1,
|
| 36 |
+
"LABEL_10": 10,
|
| 37 |
+
"LABEL_11": 11,
|
| 38 |
+
"LABEL_12": 12,
|
| 39 |
+
"LABEL_13": 13,
|
| 40 |
+
"LABEL_14": 14,
|
| 41 |
+
"LABEL_15": 15,
|
| 42 |
+
"LABEL_16": 16,
|
| 43 |
+
"LABEL_17": 17,
|
| 44 |
+
"LABEL_2": 2,
|
| 45 |
+
"LABEL_3": 3,
|
| 46 |
+
"LABEL_4": 4,
|
| 47 |
+
"LABEL_5": 5,
|
| 48 |
+
"LABEL_6": 6,
|
| 49 |
+
"LABEL_7": 7,
|
| 50 |
+
"LABEL_8": 8,
|
| 51 |
+
"LABEL_9": 9
|
| 52 |
+
},
|
| 53 |
+
"layer_norm_eps": 1e-12,
|
| 54 |
+
"max_position_embeddings": 512,
|
| 55 |
+
"model_type": "bert",
|
| 56 |
+
"num_attention_heads": 12,
|
| 57 |
+
"num_hidden_layers": 12,
|
| 58 |
+
"pad_token_id": 0,
|
| 59 |
+
"position_embedding_type": "absolute",
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.55.4",
|
| 62 |
+
"type_vocab_size": 2,
|
| 63 |
+
"use_cache": true,
|
| 64 |
+
"vocab_size": 30522
|
| 65 |
+
}
|
run-2/checkpoint-7000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e57067682f40c39403c10f6ba9b9738102f5bb004d4aec8204bbac9bc54fd1d7
|
| 3 |
+
size 438007864
|
run-2/checkpoint-7000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df3ff72be4211afd913bf66db5c3fa4d13632e40e2e3c6e82c548e781c7b76d0
|
| 3 |
+
size 876140043
|
run-2/checkpoint-7000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:df14e1f4e842b9f15d085a26323b8ff19a41f8524bfe152f7179b20f387149ed
|
| 3 |
+
size 14645
|
run-2/checkpoint-7000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e30eab33909ac3153bcb0a21219783bdbdf6c3e239fd8fe62248b77c7e07a99
|
| 3 |
+
size 1383
|
run-2/checkpoint-7000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5f182fb9e33a0a8f3d1d1c2bb421dcc72babac6eed740814086d000f4faa21b4
|
| 3 |
+
size 1465
|
run-2/checkpoint-7000/trainer_state.json
ADDED
|
@@ -0,0 +1,694 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6000,
|
| 3 |
+
"best_metric": 0.8207234712908512,
|
| 4 |
+
"best_model_checkpoint": "./results/run-2/checkpoint-6000",
|
| 5 |
+
"epoch": 7.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 7000,
|
| 8 |
+
"is_hyper_param_search": true,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.1,
|
| 14 |
+
"grad_norm": 3.6001648902893066,
|
| 15 |
+
"learning_rate": 3.1887409174919234e-06,
|
| 16 |
+
"loss": 0.0412,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.2,
|
| 21 |
+
"grad_norm": 0.6239789128303528,
|
| 22 |
+
"learning_rate": 6.409691339200936e-06,
|
| 23 |
+
"loss": 0.1091,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.3,
|
| 28 |
+
"grad_norm": 0.6479267477989197,
|
| 29 |
+
"learning_rate": 9.63064176090995e-06,
|
| 30 |
+
"loss": 0.0,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.4,
|
| 35 |
+
"grad_norm": 20.527082443237305,
|
| 36 |
+
"learning_rate": 1.054959200884952e-05,
|
| 37 |
+
"loss": 0.0,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"grad_norm": 0.8853731751441956,
|
| 43 |
+
"learning_rate": 1.0481970753745042e-05,
|
| 44 |
+
"loss": 0.0,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.6,
|
| 49 |
+
"grad_norm": 1.839735507965088,
|
| 50 |
+
"learning_rate": 1.0414349498640566e-05,
|
| 51 |
+
"loss": 0.0,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.7,
|
| 56 |
+
"grad_norm": 2.7137391567230225,
|
| 57 |
+
"learning_rate": 1.0346728243536088e-05,
|
| 58 |
+
"loss": 0.0284,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.8,
|
| 63 |
+
"grad_norm": 0.6026535630226135,
|
| 64 |
+
"learning_rate": 1.027910698843161e-05,
|
| 65 |
+
"loss": 0.1612,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.9,
|
| 70 |
+
"grad_norm": 0.5444966554641724,
|
| 71 |
+
"learning_rate": 1.0211485733327134e-05,
|
| 72 |
+
"loss": 0.0,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.0,
|
| 77 |
+
"grad_norm": 1.2413469552993774,
|
| 78 |
+
"learning_rate": 1.0143864478222655e-05,
|
| 79 |
+
"loss": 0.0986,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0,
|
| 84 |
+
"eval_economic_inequality_accuracy": 0.8047858942065491,
|
| 85 |
+
"eval_economic_inequality_f1": 0.7538941327354425,
|
| 86 |
+
"eval_economic_policy_benefits_accuracy": 0.8282828282828283,
|
| 87 |
+
"eval_economic_policy_benefits_f1": 0.764129475069552,
|
| 88 |
+
"eval_ethnic_boundaries_accuracy": 0.9493464052287581,
|
| 89 |
+
"eval_ethnic_boundaries_f1": 0.9281293325410972,
|
| 90 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 91 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 92 |
+
"eval_loss": NaN,
|
| 93 |
+
"eval_mother_tongue_education_accuracy": 0.8888888888888888,
|
| 94 |
+
"eval_mother_tongue_education_f1": 0.837037037037037,
|
| 95 |
+
"eval_overall_accuracy": 0.8390088178776338,
|
| 96 |
+
"eval_overall_f1": 0.7839408748610676,
|
| 97 |
+
"eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
|
| 98 |
+
"eval_religion_ethnic_identity_f1": 0.8133124146404199,
|
| 99 |
+
"eval_runtime": 4.4593,
|
| 100 |
+
"eval_samples_per_second": 897.004,
|
| 101 |
+
"eval_steps_per_second": 56.063,
|
| 102 |
+
"step": 1000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1,
|
| 106 |
+
"grad_norm": 0.5792437195777893,
|
| 107 |
+
"learning_rate": 1.0076243223118178e-05,
|
| 108 |
+
"loss": 0.0,
|
| 109 |
+
"step": 1100
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.2,
|
| 113 |
+
"grad_norm": 8.802961349487305,
|
| 114 |
+
"learning_rate": 1.0008621968013701e-05,
|
| 115 |
+
"loss": 0.0297,
|
| 116 |
+
"step": 1200
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.3,
|
| 120 |
+
"grad_norm": 3.981771469116211,
|
| 121 |
+
"learning_rate": 9.941000712909224e-06,
|
| 122 |
+
"loss": 0.0,
|
| 123 |
+
"step": 1300
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.4,
|
| 127 |
+
"grad_norm": 0.9215607047080994,
|
| 128 |
+
"learning_rate": 9.873379457804745e-06,
|
| 129 |
+
"loss": 0.263,
|
| 130 |
+
"step": 1400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.5,
|
| 134 |
+
"grad_norm": 1.5474179983139038,
|
| 135 |
+
"learning_rate": 9.805758202700268e-06,
|
| 136 |
+
"loss": 0.0,
|
| 137 |
+
"step": 1500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.6,
|
| 141 |
+
"grad_norm": 1.1813520193099976,
|
| 142 |
+
"learning_rate": 9.738136947595791e-06,
|
| 143 |
+
"loss": 0.0,
|
| 144 |
+
"step": 1600
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.7,
|
| 148 |
+
"grad_norm": 5.511495113372803,
|
| 149 |
+
"learning_rate": 9.670515692491313e-06,
|
| 150 |
+
"loss": 0.0037,
|
| 151 |
+
"step": 1700
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.8,
|
| 155 |
+
"grad_norm": 3.1629152297973633,
|
| 156 |
+
"learning_rate": 9.602894437386836e-06,
|
| 157 |
+
"loss": 0.0027,
|
| 158 |
+
"step": 1800
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.9,
|
| 162 |
+
"grad_norm": 0.6996293663978577,
|
| 163 |
+
"learning_rate": 9.535273182282359e-06,
|
| 164 |
+
"loss": 19.1531,
|
| 165 |
+
"step": 1900
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.0,
|
| 169 |
+
"grad_norm": 1.3346213102340698,
|
| 170 |
+
"learning_rate": 9.467651927177882e-06,
|
| 171 |
+
"loss": 2.9418,
|
| 172 |
+
"step": 2000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.0,
|
| 176 |
+
"eval_economic_inequality_accuracy": 0.7896725440806045,
|
| 177 |
+
"eval_economic_inequality_f1": 0.7951291586358689,
|
| 178 |
+
"eval_economic_policy_benefits_accuracy": 0.8148148148148148,
|
| 179 |
+
"eval_economic_policy_benefits_f1": 0.8131577832835695,
|
| 180 |
+
"eval_ethnic_boundaries_accuracy": 0.9264705882352942,
|
| 181 |
+
"eval_ethnic_boundaries_f1": 0.9324831159457648,
|
| 182 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 183 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 184 |
+
"eval_loss": NaN,
|
| 185 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 186 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 187 |
+
"eval_overall_accuracy": 0.8103022392805538,
|
| 188 |
+
"eval_overall_f1": 0.7884819842273197,
|
| 189 |
+
"eval_religion_ethnic_identity_accuracy": 0.8471953578336557,
|
| 190 |
+
"eval_religion_ethnic_identity_f1": 0.8422382496151178,
|
| 191 |
+
"eval_runtime": 4.499,
|
| 192 |
+
"eval_samples_per_second": 889.087,
|
| 193 |
+
"eval_steps_per_second": 55.568,
|
| 194 |
+
"step": 2000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.1,
|
| 198 |
+
"grad_norm": 133.0632781982422,
|
| 199 |
+
"learning_rate": 9.400030672073403e-06,
|
| 200 |
+
"loss": 0.0,
|
| 201 |
+
"step": 2100
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.2,
|
| 205 |
+
"grad_norm": 1.1462996006011963,
|
| 206 |
+
"learning_rate": 9.332409416968926e-06,
|
| 207 |
+
"loss": 0.0,
|
| 208 |
+
"step": 2200
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.3,
|
| 212 |
+
"grad_norm": 0.11461023986339569,
|
| 213 |
+
"learning_rate": 9.26478816186445e-06,
|
| 214 |
+
"loss": 0.0,
|
| 215 |
+
"step": 2300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.4,
|
| 219 |
+
"grad_norm": 0.37900322675704956,
|
| 220 |
+
"learning_rate": 9.197166906759972e-06,
|
| 221 |
+
"loss": 0.0046,
|
| 222 |
+
"step": 2400
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.5,
|
| 226 |
+
"grad_norm": 1.0005897283554077,
|
| 227 |
+
"learning_rate": 9.129545651655495e-06,
|
| 228 |
+
"loss": 0.0,
|
| 229 |
+
"step": 2500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.6,
|
| 233 |
+
"grad_norm": 0.15003152191638947,
|
| 234 |
+
"learning_rate": 9.061924396551017e-06,
|
| 235 |
+
"loss": 0.0,
|
| 236 |
+
"step": 2600
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.7,
|
| 240 |
+
"grad_norm": 1.9842479228973389,
|
| 241 |
+
"learning_rate": 8.99430314144654e-06,
|
| 242 |
+
"loss": 0.0017,
|
| 243 |
+
"step": 2700
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.8,
|
| 247 |
+
"grad_norm": 4.747731685638428,
|
| 248 |
+
"learning_rate": 8.926681886342063e-06,
|
| 249 |
+
"loss": 0.0,
|
| 250 |
+
"step": 2800
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.9,
|
| 254 |
+
"grad_norm": 0.4671611189842224,
|
| 255 |
+
"learning_rate": 8.859060631237586e-06,
|
| 256 |
+
"loss": 0.0,
|
| 257 |
+
"step": 2900
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0,
|
| 261 |
+
"grad_norm": 0.9774578213691711,
|
| 262 |
+
"learning_rate": 8.791439376133107e-06,
|
| 263 |
+
"loss": 0.0,
|
| 264 |
+
"step": 3000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.0,
|
| 268 |
+
"eval_economic_inequality_accuracy": 0.8173803526448362,
|
| 269 |
+
"eval_economic_inequality_f1": 0.8105151543481738,
|
| 270 |
+
"eval_economic_policy_benefits_accuracy": 0.8686868686868687,
|
| 271 |
+
"eval_economic_policy_benefits_f1": 0.8544859369266805,
|
| 272 |
+
"eval_ethnic_boundaries_accuracy": 0.9330065359477124,
|
| 273 |
+
"eval_ethnic_boundaries_f1": 0.9342620574580529,
|
| 274 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 275 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 276 |
+
"eval_loss": NaN,
|
| 277 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 278 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 279 |
+
"eval_overall_accuracy": 0.8266000709526621,
|
| 280 |
+
"eval_overall_f1": 0.8006341340755991,
|
| 281 |
+
"eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
|
| 282 |
+
"eval_religion_ethnic_identity_f1": 0.85665805783709,
|
| 283 |
+
"eval_runtime": 4.4861,
|
| 284 |
+
"eval_samples_per_second": 891.646,
|
| 285 |
+
"eval_steps_per_second": 55.728,
|
| 286 |
+
"step": 3000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.1,
|
| 290 |
+
"grad_norm": 0.6768048405647278,
|
| 291 |
+
"learning_rate": 8.72381812102863e-06,
|
| 292 |
+
"loss": 0.0002,
|
| 293 |
+
"step": 3100
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.2,
|
| 297 |
+
"grad_norm": 0.13458865880966187,
|
| 298 |
+
"learning_rate": 8.656196865924153e-06,
|
| 299 |
+
"loss": 0.0006,
|
| 300 |
+
"step": 3200
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.3,
|
| 304 |
+
"grad_norm": 2.0157546997070312,
|
| 305 |
+
"learning_rate": 8.588575610819674e-06,
|
| 306 |
+
"loss": 0.0,
|
| 307 |
+
"step": 3300
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.4,
|
| 311 |
+
"grad_norm": 0.07812555134296417,
|
| 312 |
+
"learning_rate": 8.520954355715199e-06,
|
| 313 |
+
"loss": 0.0,
|
| 314 |
+
"step": 3400
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"grad_norm": 149.58753967285156,
|
| 319 |
+
"learning_rate": 8.45333310061072e-06,
|
| 320 |
+
"loss": 0.0032,
|
| 321 |
+
"step": 3500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.6,
|
| 325 |
+
"grad_norm": 3.429203748703003,
|
| 326 |
+
"learning_rate": 8.385711845506243e-06,
|
| 327 |
+
"loss": 0.0,
|
| 328 |
+
"step": 3600
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.7,
|
| 332 |
+
"grad_norm": 1.6850093603134155,
|
| 333 |
+
"learning_rate": 8.318090590401766e-06,
|
| 334 |
+
"loss": 0.0003,
|
| 335 |
+
"step": 3700
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.8,
|
| 339 |
+
"grad_norm": 0.06944789737462997,
|
| 340 |
+
"learning_rate": 8.250469335297288e-06,
|
| 341 |
+
"loss": 0.0,
|
| 342 |
+
"step": 3800
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 3.9,
|
| 346 |
+
"grad_norm": 0.11943838745355606,
|
| 347 |
+
"learning_rate": 8.18284808019281e-06,
|
| 348 |
+
"loss": 0.0,
|
| 349 |
+
"step": 3900
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.0,
|
| 353 |
+
"grad_norm": 0.308050274848938,
|
| 354 |
+
"learning_rate": 8.115226825088332e-06,
|
| 355 |
+
"loss": 0.0,
|
| 356 |
+
"step": 4000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 4.0,
|
| 360 |
+
"eval_economic_inequality_accuracy": 0.8085642317380353,
|
| 361 |
+
"eval_economic_inequality_f1": 0.810531309630811,
|
| 362 |
+
"eval_economic_policy_benefits_accuracy": 0.8653198653198653,
|
| 363 |
+
"eval_economic_policy_benefits_f1": 0.8539054809077693,
|
| 364 |
+
"eval_ethnic_boundaries_accuracy": 0.9362745098039216,
|
| 365 |
+
"eval_ethnic_boundaries_f1": 0.9400162685161402,
|
| 366 |
+
"eval_language_policy_accuracy": 0.6764705882352942,
|
| 367 |
+
"eval_language_policy_f1": 0.6150075414781296,
|
| 368 |
+
"eval_loss": NaN,
|
| 369 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 370 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 371 |
+
"eval_overall_accuracy": 0.8169885251374339,
|
| 372 |
+
"eval_overall_f1": 0.8070145214695552,
|
| 373 |
+
"eval_religion_ethnic_identity_accuracy": 0.8375241779497099,
|
| 374 |
+
"eval_religion_ethnic_identity_f1": 0.8491222547802076,
|
| 375 |
+
"eval_runtime": 4.4658,
|
| 376 |
+
"eval_samples_per_second": 895.686,
|
| 377 |
+
"eval_steps_per_second": 55.98,
|
| 378 |
+
"step": 4000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 4.1,
|
| 382 |
+
"grad_norm": 0.2866123616695404,
|
| 383 |
+
"learning_rate": 8.047605569983857e-06,
|
| 384 |
+
"loss": 0.0,
|
| 385 |
+
"step": 4100
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 4.2,
|
| 389 |
+
"grad_norm": 0.26467469334602356,
|
| 390 |
+
"learning_rate": 7.979984314879378e-06,
|
| 391 |
+
"loss": 0.0,
|
| 392 |
+
"step": 4200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.3,
|
| 396 |
+
"grad_norm": 0.06363595277070999,
|
| 397 |
+
"learning_rate": 7.912363059774901e-06,
|
| 398 |
+
"loss": 0.0,
|
| 399 |
+
"step": 4300
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.4,
|
| 403 |
+
"grad_norm": 0.2771783173084259,
|
| 404 |
+
"learning_rate": 7.844741804670424e-06,
|
| 405 |
+
"loss": 0.0002,
|
| 406 |
+
"step": 4400
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.5,
|
| 410 |
+
"grad_norm": 0.022821573540568352,
|
| 411 |
+
"learning_rate": 7.777120549565947e-06,
|
| 412 |
+
"loss": 0.0006,
|
| 413 |
+
"step": 4500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.6,
|
| 417 |
+
"grad_norm": 0.7363903522491455,
|
| 418 |
+
"learning_rate": 7.709499294461468e-06,
|
| 419 |
+
"loss": 0.0,
|
| 420 |
+
"step": 4600
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.7,
|
| 424 |
+
"grad_norm": 1.3803585767745972,
|
| 425 |
+
"learning_rate": 7.641878039356991e-06,
|
| 426 |
+
"loss": 0.0,
|
| 427 |
+
"step": 4700
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 4.8,
|
| 431 |
+
"grad_norm": 0.14331550896167755,
|
| 432 |
+
"learning_rate": 7.5742567842525145e-06,
|
| 433 |
+
"loss": 0.0019,
|
| 434 |
+
"step": 4800
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.9,
|
| 438 |
+
"grad_norm": 0.06504154205322266,
|
| 439 |
+
"learning_rate": 7.506635529148037e-06,
|
| 440 |
+
"loss": 0.0,
|
| 441 |
+
"step": 4900
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 5.0,
|
| 445 |
+
"grad_norm": 0.09633997082710266,
|
| 446 |
+
"learning_rate": 7.43901427404356e-06,
|
| 447 |
+
"loss": 0.0,
|
| 448 |
+
"step": 5000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 5.0,
|
| 452 |
+
"eval_economic_inequality_accuracy": 0.8249370277078085,
|
| 453 |
+
"eval_economic_inequality_f1": 0.823342228819045,
|
| 454 |
+
"eval_economic_policy_benefits_accuracy": 0.8451178451178452,
|
| 455 |
+
"eval_economic_policy_benefits_f1": 0.8327902682741393,
|
| 456 |
+
"eval_ethnic_boundaries_accuracy": 0.9395424836601307,
|
| 457 |
+
"eval_ethnic_boundaries_f1": 0.9429587969409783,
|
| 458 |
+
"eval_language_policy_accuracy": 0.6470588235294118,
|
| 459 |
+
"eval_language_policy_f1": 0.5939135835329606,
|
| 460 |
+
"eval_loss": NaN,
|
| 461 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 462 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 463 |
+
"eval_overall_accuracy": 0.8165062399029553,
|
| 464 |
+
"eval_overall_f1": 0.8063046374512969,
|
| 465 |
+
"eval_religion_ethnic_identity_accuracy": 0.8646034816247582,
|
| 466 |
+
"eval_religion_ethnic_identity_f1": 0.8713186736363845,
|
| 467 |
+
"eval_runtime": 4.4849,
|
| 468 |
+
"eval_samples_per_second": 891.887,
|
| 469 |
+
"eval_steps_per_second": 55.743,
|
| 470 |
+
"step": 5000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 5.1,
|
| 474 |
+
"grad_norm": 2.563802480697632,
|
| 475 |
+
"learning_rate": 7.371393018939082e-06,
|
| 476 |
+
"loss": 0.0049,
|
| 477 |
+
"step": 5100
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 5.2,
|
| 481 |
+
"grad_norm": 10.922028541564941,
|
| 482 |
+
"learning_rate": 7.303771763834604e-06,
|
| 483 |
+
"loss": 0.0004,
|
| 484 |
+
"step": 5200
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 5.3,
|
| 488 |
+
"grad_norm": 0.9929112792015076,
|
| 489 |
+
"learning_rate": 7.236150508730128e-06,
|
| 490 |
+
"loss": 0.0827,
|
| 491 |
+
"step": 5300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.4,
|
| 495 |
+
"grad_norm": 0.13412432372570038,
|
| 496 |
+
"learning_rate": 7.16852925362565e-06,
|
| 497 |
+
"loss": 0.0,
|
| 498 |
+
"step": 5400
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 5.5,
|
| 502 |
+
"grad_norm": 0.0726282000541687,
|
| 503 |
+
"learning_rate": 7.100907998521172e-06,
|
| 504 |
+
"loss": 0.0,
|
| 505 |
+
"step": 5500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.6,
|
| 509 |
+
"grad_norm": 12.0023832321167,
|
| 510 |
+
"learning_rate": 7.033286743416695e-06,
|
| 511 |
+
"loss": 0.0,
|
| 512 |
+
"step": 5600
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.7,
|
| 516 |
+
"grad_norm": 1.0574413537979126,
|
| 517 |
+
"learning_rate": 6.9656654883122174e-06,
|
| 518 |
+
"loss": 0.0,
|
| 519 |
+
"step": 5700
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.8,
|
| 523 |
+
"grad_norm": 0.3997838795185089,
|
| 524 |
+
"learning_rate": 6.89804423320774e-06,
|
| 525 |
+
"loss": 0.0,
|
| 526 |
+
"step": 5800
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.9,
|
| 530 |
+
"grad_norm": 2.0076065063476562,
|
| 531 |
+
"learning_rate": 6.830422978103262e-06,
|
| 532 |
+
"loss": 0.0,
|
| 533 |
+
"step": 5900
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 6.0,
|
| 537 |
+
"grad_norm": 2.676760673522949,
|
| 538 |
+
"learning_rate": 6.762801722998786e-06,
|
| 539 |
+
"loss": 0.0,
|
| 540 |
+
"step": 6000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 6.0,
|
| 544 |
+
"eval_economic_inequality_accuracy": 0.8350125944584383,
|
| 545 |
+
"eval_economic_inequality_f1": 0.8243306188572417,
|
| 546 |
+
"eval_economic_policy_benefits_accuracy": 0.8619528619528619,
|
| 547 |
+
"eval_economic_policy_benefits_f1": 0.840430611429002,
|
| 548 |
+
"eval_ethnic_boundaries_accuracy": 0.9477124183006536,
|
| 549 |
+
"eval_ethnic_boundaries_f1": 0.9385901883316666,
|
| 550 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 551 |
+
"eval_language_policy_f1": 0.6172459893048128,
|
| 552 |
+
"eval_loss": NaN,
|
| 553 |
+
"eval_mother_tongue_education_accuracy": 0.8888888888888888,
|
| 554 |
+
"eval_mother_tongue_education_f1": 0.837037037037037,
|
| 555 |
+
"eval_overall_accuracy": 0.8529320416673837,
|
| 556 |
+
"eval_overall_f1": 0.8207234712908512,
|
| 557 |
+
"eval_religion_ethnic_identity_accuracy": 0.8781431334622823,
|
| 558 |
+
"eval_religion_ethnic_identity_f1": 0.8667063827853473,
|
| 559 |
+
"eval_runtime": 4.4702,
|
| 560 |
+
"eval_samples_per_second": 894.82,
|
| 561 |
+
"eval_steps_per_second": 55.926,
|
| 562 |
+
"step": 6000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 6.1,
|
| 566 |
+
"grad_norm": 0.18037013709545135,
|
| 567 |
+
"learning_rate": 6.695180467894308e-06,
|
| 568 |
+
"loss": 0.0,
|
| 569 |
+
"step": 6100
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 6.2,
|
| 573 |
+
"grad_norm": 22.60749626159668,
|
| 574 |
+
"learning_rate": 6.62755921278983e-06,
|
| 575 |
+
"loss": 0.0,
|
| 576 |
+
"step": 6200
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 6.3,
|
| 580 |
+
"grad_norm": 0.8915765881538391,
|
| 581 |
+
"learning_rate": 6.559937957685353e-06,
|
| 582 |
+
"loss": 0.0,
|
| 583 |
+
"step": 6300
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 6.4,
|
| 587 |
+
"grad_norm": 0.5682930946350098,
|
| 588 |
+
"learning_rate": 6.492316702580875e-06,
|
| 589 |
+
"loss": 0.0,
|
| 590 |
+
"step": 6400
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 6.5,
|
| 594 |
+
"grad_norm": 0.08639243990182877,
|
| 595 |
+
"learning_rate": 6.424695447476398e-06,
|
| 596 |
+
"loss": 0.0001,
|
| 597 |
+
"step": 6500
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.6,
|
| 601 |
+
"grad_norm": 0.060448452830314636,
|
| 602 |
+
"learning_rate": 6.357074192371921e-06,
|
| 603 |
+
"loss": 0.3897,
|
| 604 |
+
"step": 6600
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 6.7,
|
| 608 |
+
"grad_norm": 0.7599291801452637,
|
| 609 |
+
"learning_rate": 6.289452937267443e-06,
|
| 610 |
+
"loss": 0.0,
|
| 611 |
+
"step": 6700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 6.8,
|
| 615 |
+
"grad_norm": 0.19008946418762207,
|
| 616 |
+
"learning_rate": 6.2218316821629656e-06,
|
| 617 |
+
"loss": 0.0012,
|
| 618 |
+
"step": 6800
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 6.9,
|
| 622 |
+
"grad_norm": 0.2047979235649109,
|
| 623 |
+
"learning_rate": 6.1542104270584894e-06,
|
| 624 |
+
"loss": 0.0,
|
| 625 |
+
"step": 6900
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 7.0,
|
| 629 |
+
"grad_norm": 4.452627658843994,
|
| 630 |
+
"learning_rate": 6.086589171954012e-06,
|
| 631 |
+
"loss": 0.0,
|
| 632 |
+
"step": 7000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 7.0,
|
| 636 |
+
"eval_economic_inequality_accuracy": 0.8047858942065491,
|
| 637 |
+
"eval_economic_inequality_f1": 0.8098052998323155,
|
| 638 |
+
"eval_economic_policy_benefits_accuracy": 0.8686868686868687,
|
| 639 |
+
"eval_economic_policy_benefits_f1": 0.8627489757007829,
|
| 640 |
+
"eval_ethnic_boundaries_accuracy": 0.9313725490196079,
|
| 641 |
+
"eval_ethnic_boundaries_f1": 0.9331205006157053,
|
| 642 |
+
"eval_language_policy_accuracy": 0.6470588235294118,
|
| 643 |
+
"eval_language_policy_f1": 0.6161764705882353,
|
| 644 |
+
"eval_loss": NaN,
|
| 645 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 646 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 647 |
+
"eval_overall_accuracy": 0.8137799964973732,
|
| 648 |
+
"eval_overall_f1": 0.8086761873509373,
|
| 649 |
+
"eval_religion_ethnic_identity_accuracy": 0.8529980657640233,
|
| 650 |
+
"eval_religion_ethnic_identity_f1": 0.8567016038643116,
|
| 651 |
+
"eval_runtime": 4.489,
|
| 652 |
+
"eval_samples_per_second": 891.066,
|
| 653 |
+
"eval_steps_per_second": 55.692,
|
| 654 |
+
"step": 7000
|
| 655 |
+
}
|
| 656 |
+
],
|
| 657 |
+
"logging_steps": 100,
|
| 658 |
+
"max_steps": 16000,
|
| 659 |
+
"num_input_tokens_seen": 0,
|
| 660 |
+
"num_train_epochs": 16,
|
| 661 |
+
"save_steps": 500,
|
| 662 |
+
"stateful_callbacks": {
|
| 663 |
+
"EarlyStoppingCallback": {
|
| 664 |
+
"args": {
|
| 665 |
+
"early_stopping_patience": 2,
|
| 666 |
+
"early_stopping_threshold": 0.0
|
| 667 |
+
},
|
| 668 |
+
"attributes": {
|
| 669 |
+
"early_stopping_patience_counter": 1
|
| 670 |
+
}
|
| 671 |
+
},
|
| 672 |
+
"TrainerControl": {
|
| 673 |
+
"args": {
|
| 674 |
+
"should_epoch_stop": false,
|
| 675 |
+
"should_evaluate": false,
|
| 676 |
+
"should_log": false,
|
| 677 |
+
"should_save": true,
|
| 678 |
+
"should_training_stop": false
|
| 679 |
+
},
|
| 680 |
+
"attributes": {}
|
| 681 |
+
}
|
| 682 |
+
},
|
| 683 |
+
"total_flos": 2.9472671563776e+16,
|
| 684 |
+
"train_batch_size": 8,
|
| 685 |
+
"trial_name": null,
|
| 686 |
+
"trial_params": {
|
| 687 |
+
"gradient_accumulation_steps": 2,
|
| 688 |
+
"learning_rate": 1.0596926887422654e-05,
|
| 689 |
+
"num_train_epochs": 16,
|
| 690 |
+
"per_device_train_batch_size": 8,
|
| 691 |
+
"warmup_steps": 329,
|
| 692 |
+
"weight_decay": 0.059185379025811405
|
| 693 |
+
}
|
| 694 |
+
}
|
run-2/checkpoint-7000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:822a4a47339dd0b384da8628d960a86b7107ee8196c01d2a9c3866ddc7aa4cd5
|
| 3 |
+
size 5777
|
run-2/checkpoint-8000/config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4",
|
| 17 |
+
"5": "LABEL_5",
|
| 18 |
+
"6": "LABEL_6",
|
| 19 |
+
"7": "LABEL_7",
|
| 20 |
+
"8": "LABEL_8",
|
| 21 |
+
"9": "LABEL_9",
|
| 22 |
+
"10": "LABEL_10",
|
| 23 |
+
"11": "LABEL_11",
|
| 24 |
+
"12": "LABEL_12",
|
| 25 |
+
"13": "LABEL_13",
|
| 26 |
+
"14": "LABEL_14",
|
| 27 |
+
"15": "LABEL_15",
|
| 28 |
+
"16": "LABEL_16",
|
| 29 |
+
"17": "LABEL_17"
|
| 30 |
+
},
|
| 31 |
+
"initializer_range": 0.02,
|
| 32 |
+
"intermediate_size": 3072,
|
| 33 |
+
"label2id": {
|
| 34 |
+
"LABEL_0": 0,
|
| 35 |
+
"LABEL_1": 1,
|
| 36 |
+
"LABEL_10": 10,
|
| 37 |
+
"LABEL_11": 11,
|
| 38 |
+
"LABEL_12": 12,
|
| 39 |
+
"LABEL_13": 13,
|
| 40 |
+
"LABEL_14": 14,
|
| 41 |
+
"LABEL_15": 15,
|
| 42 |
+
"LABEL_16": 16,
|
| 43 |
+
"LABEL_17": 17,
|
| 44 |
+
"LABEL_2": 2,
|
| 45 |
+
"LABEL_3": 3,
|
| 46 |
+
"LABEL_4": 4,
|
| 47 |
+
"LABEL_5": 5,
|
| 48 |
+
"LABEL_6": 6,
|
| 49 |
+
"LABEL_7": 7,
|
| 50 |
+
"LABEL_8": 8,
|
| 51 |
+
"LABEL_9": 9
|
| 52 |
+
},
|
| 53 |
+
"layer_norm_eps": 1e-12,
|
| 54 |
+
"max_position_embeddings": 512,
|
| 55 |
+
"model_type": "bert",
|
| 56 |
+
"num_attention_heads": 12,
|
| 57 |
+
"num_hidden_layers": 12,
|
| 58 |
+
"pad_token_id": 0,
|
| 59 |
+
"position_embedding_type": "absolute",
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.55.4",
|
| 62 |
+
"type_vocab_size": 2,
|
| 63 |
+
"use_cache": true,
|
| 64 |
+
"vocab_size": 30522
|
| 65 |
+
}
|
run-2/checkpoint-8000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31ea1c8b5b727fb329e7ccc8fc9713c4175bf50214c17df14d649c2faf9188fb
|
| 3 |
+
size 438007864
|
run-2/checkpoint-8000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fa4d8a17bc29974089472c69dfd8d97c728fa1e42dac2d6ea4fa7350de5baaf6
|
| 3 |
+
size 876140043
|
run-2/checkpoint-8000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aa1abad22b875d8bfed896dd812f767d2836fe2ac42ac5af0bea1f5a3996a46d
|
| 3 |
+
size 14645
|
run-2/checkpoint-8000/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:aff16ce223d129ede29a7e248be36c7fd8c88c420736343fa1211707e1e2c44b
|
| 3 |
+
size 1383
|
run-2/checkpoint-8000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a79eb88888879fbed11eaeb00b77d774122c767a4ec54c13f68e563ef8794b7
|
| 3 |
+
size 1465
|
run-2/checkpoint-8000/trainer_state.json
ADDED
|
@@ -0,0 +1,786 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 6000,
|
| 3 |
+
"best_metric": 0.8207234712908512,
|
| 4 |
+
"best_model_checkpoint": "./results/run-2/checkpoint-6000",
|
| 5 |
+
"epoch": 8.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 8000,
|
| 8 |
+
"is_hyper_param_search": true,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.1,
|
| 14 |
+
"grad_norm": 3.6001648902893066,
|
| 15 |
+
"learning_rate": 3.1887409174919234e-06,
|
| 16 |
+
"loss": 0.0412,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.2,
|
| 21 |
+
"grad_norm": 0.6239789128303528,
|
| 22 |
+
"learning_rate": 6.409691339200936e-06,
|
| 23 |
+
"loss": 0.1091,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.3,
|
| 28 |
+
"grad_norm": 0.6479267477989197,
|
| 29 |
+
"learning_rate": 9.63064176090995e-06,
|
| 30 |
+
"loss": 0.0,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.4,
|
| 35 |
+
"grad_norm": 20.527082443237305,
|
| 36 |
+
"learning_rate": 1.054959200884952e-05,
|
| 37 |
+
"loss": 0.0,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.5,
|
| 42 |
+
"grad_norm": 0.8853731751441956,
|
| 43 |
+
"learning_rate": 1.0481970753745042e-05,
|
| 44 |
+
"loss": 0.0,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.6,
|
| 49 |
+
"grad_norm": 1.839735507965088,
|
| 50 |
+
"learning_rate": 1.0414349498640566e-05,
|
| 51 |
+
"loss": 0.0,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.7,
|
| 56 |
+
"grad_norm": 2.7137391567230225,
|
| 57 |
+
"learning_rate": 1.0346728243536088e-05,
|
| 58 |
+
"loss": 0.0284,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.8,
|
| 63 |
+
"grad_norm": 0.6026535630226135,
|
| 64 |
+
"learning_rate": 1.027910698843161e-05,
|
| 65 |
+
"loss": 0.1612,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.9,
|
| 70 |
+
"grad_norm": 0.5444966554641724,
|
| 71 |
+
"learning_rate": 1.0211485733327134e-05,
|
| 72 |
+
"loss": 0.0,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 1.0,
|
| 77 |
+
"grad_norm": 1.2413469552993774,
|
| 78 |
+
"learning_rate": 1.0143864478222655e-05,
|
| 79 |
+
"loss": 0.0986,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.0,
|
| 84 |
+
"eval_economic_inequality_accuracy": 0.8047858942065491,
|
| 85 |
+
"eval_economic_inequality_f1": 0.7538941327354425,
|
| 86 |
+
"eval_economic_policy_benefits_accuracy": 0.8282828282828283,
|
| 87 |
+
"eval_economic_policy_benefits_f1": 0.764129475069552,
|
| 88 |
+
"eval_ethnic_boundaries_accuracy": 0.9493464052287581,
|
| 89 |
+
"eval_ethnic_boundaries_f1": 0.9281293325410972,
|
| 90 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 91 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 92 |
+
"eval_loss": NaN,
|
| 93 |
+
"eval_mother_tongue_education_accuracy": 0.8888888888888888,
|
| 94 |
+
"eval_mother_tongue_education_f1": 0.837037037037037,
|
| 95 |
+
"eval_overall_accuracy": 0.8390088178776338,
|
| 96 |
+
"eval_overall_f1": 0.7839408748610676,
|
| 97 |
+
"eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
|
| 98 |
+
"eval_religion_ethnic_identity_f1": 0.8133124146404199,
|
| 99 |
+
"eval_runtime": 4.4593,
|
| 100 |
+
"eval_samples_per_second": 897.004,
|
| 101 |
+
"eval_steps_per_second": 56.063,
|
| 102 |
+
"step": 1000
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1,
|
| 106 |
+
"grad_norm": 0.5792437195777893,
|
| 107 |
+
"learning_rate": 1.0076243223118178e-05,
|
| 108 |
+
"loss": 0.0,
|
| 109 |
+
"step": 1100
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.2,
|
| 113 |
+
"grad_norm": 8.802961349487305,
|
| 114 |
+
"learning_rate": 1.0008621968013701e-05,
|
| 115 |
+
"loss": 0.0297,
|
| 116 |
+
"step": 1200
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.3,
|
| 120 |
+
"grad_norm": 3.981771469116211,
|
| 121 |
+
"learning_rate": 9.941000712909224e-06,
|
| 122 |
+
"loss": 0.0,
|
| 123 |
+
"step": 1300
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.4,
|
| 127 |
+
"grad_norm": 0.9215607047080994,
|
| 128 |
+
"learning_rate": 9.873379457804745e-06,
|
| 129 |
+
"loss": 0.263,
|
| 130 |
+
"step": 1400
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.5,
|
| 134 |
+
"grad_norm": 1.5474179983139038,
|
| 135 |
+
"learning_rate": 9.805758202700268e-06,
|
| 136 |
+
"loss": 0.0,
|
| 137 |
+
"step": 1500
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.6,
|
| 141 |
+
"grad_norm": 1.1813520193099976,
|
| 142 |
+
"learning_rate": 9.738136947595791e-06,
|
| 143 |
+
"loss": 0.0,
|
| 144 |
+
"step": 1600
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.7,
|
| 148 |
+
"grad_norm": 5.511495113372803,
|
| 149 |
+
"learning_rate": 9.670515692491313e-06,
|
| 150 |
+
"loss": 0.0037,
|
| 151 |
+
"step": 1700
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.8,
|
| 155 |
+
"grad_norm": 3.1629152297973633,
|
| 156 |
+
"learning_rate": 9.602894437386836e-06,
|
| 157 |
+
"loss": 0.0027,
|
| 158 |
+
"step": 1800
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.9,
|
| 162 |
+
"grad_norm": 0.6996293663978577,
|
| 163 |
+
"learning_rate": 9.535273182282359e-06,
|
| 164 |
+
"loss": 19.1531,
|
| 165 |
+
"step": 1900
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.0,
|
| 169 |
+
"grad_norm": 1.3346213102340698,
|
| 170 |
+
"learning_rate": 9.467651927177882e-06,
|
| 171 |
+
"loss": 2.9418,
|
| 172 |
+
"step": 2000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 2.0,
|
| 176 |
+
"eval_economic_inequality_accuracy": 0.7896725440806045,
|
| 177 |
+
"eval_economic_inequality_f1": 0.7951291586358689,
|
| 178 |
+
"eval_economic_policy_benefits_accuracy": 0.8148148148148148,
|
| 179 |
+
"eval_economic_policy_benefits_f1": 0.8131577832835695,
|
| 180 |
+
"eval_ethnic_boundaries_accuracy": 0.9264705882352942,
|
| 181 |
+
"eval_ethnic_boundaries_f1": 0.9324831159457648,
|
| 182 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 183 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 184 |
+
"eval_loss": NaN,
|
| 185 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 186 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 187 |
+
"eval_overall_accuracy": 0.8103022392805538,
|
| 188 |
+
"eval_overall_f1": 0.7884819842273197,
|
| 189 |
+
"eval_religion_ethnic_identity_accuracy": 0.8471953578336557,
|
| 190 |
+
"eval_religion_ethnic_identity_f1": 0.8422382496151178,
|
| 191 |
+
"eval_runtime": 4.499,
|
| 192 |
+
"eval_samples_per_second": 889.087,
|
| 193 |
+
"eval_steps_per_second": 55.568,
|
| 194 |
+
"step": 2000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.1,
|
| 198 |
+
"grad_norm": 133.0632781982422,
|
| 199 |
+
"learning_rate": 9.400030672073403e-06,
|
| 200 |
+
"loss": 0.0,
|
| 201 |
+
"step": 2100
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.2,
|
| 205 |
+
"grad_norm": 1.1462996006011963,
|
| 206 |
+
"learning_rate": 9.332409416968926e-06,
|
| 207 |
+
"loss": 0.0,
|
| 208 |
+
"step": 2200
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 2.3,
|
| 212 |
+
"grad_norm": 0.11461023986339569,
|
| 213 |
+
"learning_rate": 9.26478816186445e-06,
|
| 214 |
+
"loss": 0.0,
|
| 215 |
+
"step": 2300
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 2.4,
|
| 219 |
+
"grad_norm": 0.37900322675704956,
|
| 220 |
+
"learning_rate": 9.197166906759972e-06,
|
| 221 |
+
"loss": 0.0046,
|
| 222 |
+
"step": 2400
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 2.5,
|
| 226 |
+
"grad_norm": 1.0005897283554077,
|
| 227 |
+
"learning_rate": 9.129545651655495e-06,
|
| 228 |
+
"loss": 0.0,
|
| 229 |
+
"step": 2500
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 2.6,
|
| 233 |
+
"grad_norm": 0.15003152191638947,
|
| 234 |
+
"learning_rate": 9.061924396551017e-06,
|
| 235 |
+
"loss": 0.0,
|
| 236 |
+
"step": 2600
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 2.7,
|
| 240 |
+
"grad_norm": 1.9842479228973389,
|
| 241 |
+
"learning_rate": 8.99430314144654e-06,
|
| 242 |
+
"loss": 0.0017,
|
| 243 |
+
"step": 2700
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 2.8,
|
| 247 |
+
"grad_norm": 4.747731685638428,
|
| 248 |
+
"learning_rate": 8.926681886342063e-06,
|
| 249 |
+
"loss": 0.0,
|
| 250 |
+
"step": 2800
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 2.9,
|
| 254 |
+
"grad_norm": 0.4671611189842224,
|
| 255 |
+
"learning_rate": 8.859060631237586e-06,
|
| 256 |
+
"loss": 0.0,
|
| 257 |
+
"step": 2900
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 3.0,
|
| 261 |
+
"grad_norm": 0.9774578213691711,
|
| 262 |
+
"learning_rate": 8.791439376133107e-06,
|
| 263 |
+
"loss": 0.0,
|
| 264 |
+
"step": 3000
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.0,
|
| 268 |
+
"eval_economic_inequality_accuracy": 0.8173803526448362,
|
| 269 |
+
"eval_economic_inequality_f1": 0.8105151543481738,
|
| 270 |
+
"eval_economic_policy_benefits_accuracy": 0.8686868686868687,
|
| 271 |
+
"eval_economic_policy_benefits_f1": 0.8544859369266805,
|
| 272 |
+
"eval_ethnic_boundaries_accuracy": 0.9330065359477124,
|
| 273 |
+
"eval_ethnic_boundaries_f1": 0.9342620574580529,
|
| 274 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 275 |
+
"eval_language_policy_f1": 0.6071428571428571,
|
| 276 |
+
"eval_loss": NaN,
|
| 277 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 278 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 279 |
+
"eval_overall_accuracy": 0.8266000709526621,
|
| 280 |
+
"eval_overall_f1": 0.8006341340755991,
|
| 281 |
+
"eval_religion_ethnic_identity_accuracy": 0.8568665377176016,
|
| 282 |
+
"eval_religion_ethnic_identity_f1": 0.85665805783709,
|
| 283 |
+
"eval_runtime": 4.4861,
|
| 284 |
+
"eval_samples_per_second": 891.646,
|
| 285 |
+
"eval_steps_per_second": 55.728,
|
| 286 |
+
"step": 3000
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 3.1,
|
| 290 |
+
"grad_norm": 0.6768048405647278,
|
| 291 |
+
"learning_rate": 8.72381812102863e-06,
|
| 292 |
+
"loss": 0.0002,
|
| 293 |
+
"step": 3100
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 3.2,
|
| 297 |
+
"grad_norm": 0.13458865880966187,
|
| 298 |
+
"learning_rate": 8.656196865924153e-06,
|
| 299 |
+
"loss": 0.0006,
|
| 300 |
+
"step": 3200
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 3.3,
|
| 304 |
+
"grad_norm": 2.0157546997070312,
|
| 305 |
+
"learning_rate": 8.588575610819674e-06,
|
| 306 |
+
"loss": 0.0,
|
| 307 |
+
"step": 3300
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 3.4,
|
| 311 |
+
"grad_norm": 0.07812555134296417,
|
| 312 |
+
"learning_rate": 8.520954355715199e-06,
|
| 313 |
+
"loss": 0.0,
|
| 314 |
+
"step": 3400
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 3.5,
|
| 318 |
+
"grad_norm": 149.58753967285156,
|
| 319 |
+
"learning_rate": 8.45333310061072e-06,
|
| 320 |
+
"loss": 0.0032,
|
| 321 |
+
"step": 3500
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 3.6,
|
| 325 |
+
"grad_norm": 3.429203748703003,
|
| 326 |
+
"learning_rate": 8.385711845506243e-06,
|
| 327 |
+
"loss": 0.0,
|
| 328 |
+
"step": 3600
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 3.7,
|
| 332 |
+
"grad_norm": 1.6850093603134155,
|
| 333 |
+
"learning_rate": 8.318090590401766e-06,
|
| 334 |
+
"loss": 0.0003,
|
| 335 |
+
"step": 3700
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 3.8,
|
| 339 |
+
"grad_norm": 0.06944789737462997,
|
| 340 |
+
"learning_rate": 8.250469335297288e-06,
|
| 341 |
+
"loss": 0.0,
|
| 342 |
+
"step": 3800
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 3.9,
|
| 346 |
+
"grad_norm": 0.11943838745355606,
|
| 347 |
+
"learning_rate": 8.18284808019281e-06,
|
| 348 |
+
"loss": 0.0,
|
| 349 |
+
"step": 3900
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 4.0,
|
| 353 |
+
"grad_norm": 0.308050274848938,
|
| 354 |
+
"learning_rate": 8.115226825088332e-06,
|
| 355 |
+
"loss": 0.0,
|
| 356 |
+
"step": 4000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 4.0,
|
| 360 |
+
"eval_economic_inequality_accuracy": 0.8085642317380353,
|
| 361 |
+
"eval_economic_inequality_f1": 0.810531309630811,
|
| 362 |
+
"eval_economic_policy_benefits_accuracy": 0.8653198653198653,
|
| 363 |
+
"eval_economic_policy_benefits_f1": 0.8539054809077693,
|
| 364 |
+
"eval_ethnic_boundaries_accuracy": 0.9362745098039216,
|
| 365 |
+
"eval_ethnic_boundaries_f1": 0.9400162685161402,
|
| 366 |
+
"eval_language_policy_accuracy": 0.6764705882352942,
|
| 367 |
+
"eval_language_policy_f1": 0.6150075414781296,
|
| 368 |
+
"eval_loss": NaN,
|
| 369 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 370 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 371 |
+
"eval_overall_accuracy": 0.8169885251374339,
|
| 372 |
+
"eval_overall_f1": 0.8070145214695552,
|
| 373 |
+
"eval_religion_ethnic_identity_accuracy": 0.8375241779497099,
|
| 374 |
+
"eval_religion_ethnic_identity_f1": 0.8491222547802076,
|
| 375 |
+
"eval_runtime": 4.4658,
|
| 376 |
+
"eval_samples_per_second": 895.686,
|
| 377 |
+
"eval_steps_per_second": 55.98,
|
| 378 |
+
"step": 4000
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 4.1,
|
| 382 |
+
"grad_norm": 0.2866123616695404,
|
| 383 |
+
"learning_rate": 8.047605569983857e-06,
|
| 384 |
+
"loss": 0.0,
|
| 385 |
+
"step": 4100
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 4.2,
|
| 389 |
+
"grad_norm": 0.26467469334602356,
|
| 390 |
+
"learning_rate": 7.979984314879378e-06,
|
| 391 |
+
"loss": 0.0,
|
| 392 |
+
"step": 4200
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 4.3,
|
| 396 |
+
"grad_norm": 0.06363595277070999,
|
| 397 |
+
"learning_rate": 7.912363059774901e-06,
|
| 398 |
+
"loss": 0.0,
|
| 399 |
+
"step": 4300
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 4.4,
|
| 403 |
+
"grad_norm": 0.2771783173084259,
|
| 404 |
+
"learning_rate": 7.844741804670424e-06,
|
| 405 |
+
"loss": 0.0002,
|
| 406 |
+
"step": 4400
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 4.5,
|
| 410 |
+
"grad_norm": 0.022821573540568352,
|
| 411 |
+
"learning_rate": 7.777120549565947e-06,
|
| 412 |
+
"loss": 0.0006,
|
| 413 |
+
"step": 4500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 4.6,
|
| 417 |
+
"grad_norm": 0.7363903522491455,
|
| 418 |
+
"learning_rate": 7.709499294461468e-06,
|
| 419 |
+
"loss": 0.0,
|
| 420 |
+
"step": 4600
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 4.7,
|
| 424 |
+
"grad_norm": 1.3803585767745972,
|
| 425 |
+
"learning_rate": 7.641878039356991e-06,
|
| 426 |
+
"loss": 0.0,
|
| 427 |
+
"step": 4700
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 4.8,
|
| 431 |
+
"grad_norm": 0.14331550896167755,
|
| 432 |
+
"learning_rate": 7.5742567842525145e-06,
|
| 433 |
+
"loss": 0.0019,
|
| 434 |
+
"step": 4800
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 4.9,
|
| 438 |
+
"grad_norm": 0.06504154205322266,
|
| 439 |
+
"learning_rate": 7.506635529148037e-06,
|
| 440 |
+
"loss": 0.0,
|
| 441 |
+
"step": 4900
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 5.0,
|
| 445 |
+
"grad_norm": 0.09633997082710266,
|
| 446 |
+
"learning_rate": 7.43901427404356e-06,
|
| 447 |
+
"loss": 0.0,
|
| 448 |
+
"step": 5000
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 5.0,
|
| 452 |
+
"eval_economic_inequality_accuracy": 0.8249370277078085,
|
| 453 |
+
"eval_economic_inequality_f1": 0.823342228819045,
|
| 454 |
+
"eval_economic_policy_benefits_accuracy": 0.8451178451178452,
|
| 455 |
+
"eval_economic_policy_benefits_f1": 0.8327902682741393,
|
| 456 |
+
"eval_ethnic_boundaries_accuracy": 0.9395424836601307,
|
| 457 |
+
"eval_ethnic_boundaries_f1": 0.9429587969409783,
|
| 458 |
+
"eval_language_policy_accuracy": 0.6470588235294118,
|
| 459 |
+
"eval_language_policy_f1": 0.5939135835329606,
|
| 460 |
+
"eval_loss": NaN,
|
| 461 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 462 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 463 |
+
"eval_overall_accuracy": 0.8165062399029553,
|
| 464 |
+
"eval_overall_f1": 0.8063046374512969,
|
| 465 |
+
"eval_religion_ethnic_identity_accuracy": 0.8646034816247582,
|
| 466 |
+
"eval_religion_ethnic_identity_f1": 0.8713186736363845,
|
| 467 |
+
"eval_runtime": 4.4849,
|
| 468 |
+
"eval_samples_per_second": 891.887,
|
| 469 |
+
"eval_steps_per_second": 55.743,
|
| 470 |
+
"step": 5000
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 5.1,
|
| 474 |
+
"grad_norm": 2.563802480697632,
|
| 475 |
+
"learning_rate": 7.371393018939082e-06,
|
| 476 |
+
"loss": 0.0049,
|
| 477 |
+
"step": 5100
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 5.2,
|
| 481 |
+
"grad_norm": 10.922028541564941,
|
| 482 |
+
"learning_rate": 7.303771763834604e-06,
|
| 483 |
+
"loss": 0.0004,
|
| 484 |
+
"step": 5200
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 5.3,
|
| 488 |
+
"grad_norm": 0.9929112792015076,
|
| 489 |
+
"learning_rate": 7.236150508730128e-06,
|
| 490 |
+
"loss": 0.0827,
|
| 491 |
+
"step": 5300
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 5.4,
|
| 495 |
+
"grad_norm": 0.13412432372570038,
|
| 496 |
+
"learning_rate": 7.16852925362565e-06,
|
| 497 |
+
"loss": 0.0,
|
| 498 |
+
"step": 5400
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 5.5,
|
| 502 |
+
"grad_norm": 0.0726282000541687,
|
| 503 |
+
"learning_rate": 7.100907998521172e-06,
|
| 504 |
+
"loss": 0.0,
|
| 505 |
+
"step": 5500
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 5.6,
|
| 509 |
+
"grad_norm": 12.0023832321167,
|
| 510 |
+
"learning_rate": 7.033286743416695e-06,
|
| 511 |
+
"loss": 0.0,
|
| 512 |
+
"step": 5600
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 5.7,
|
| 516 |
+
"grad_norm": 1.0574413537979126,
|
| 517 |
+
"learning_rate": 6.9656654883122174e-06,
|
| 518 |
+
"loss": 0.0,
|
| 519 |
+
"step": 5700
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 5.8,
|
| 523 |
+
"grad_norm": 0.3997838795185089,
|
| 524 |
+
"learning_rate": 6.89804423320774e-06,
|
| 525 |
+
"loss": 0.0,
|
| 526 |
+
"step": 5800
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 5.9,
|
| 530 |
+
"grad_norm": 2.0076065063476562,
|
| 531 |
+
"learning_rate": 6.830422978103262e-06,
|
| 532 |
+
"loss": 0.0,
|
| 533 |
+
"step": 5900
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 6.0,
|
| 537 |
+
"grad_norm": 2.676760673522949,
|
| 538 |
+
"learning_rate": 6.762801722998786e-06,
|
| 539 |
+
"loss": 0.0,
|
| 540 |
+
"step": 6000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 6.0,
|
| 544 |
+
"eval_economic_inequality_accuracy": 0.8350125944584383,
|
| 545 |
+
"eval_economic_inequality_f1": 0.8243306188572417,
|
| 546 |
+
"eval_economic_policy_benefits_accuracy": 0.8619528619528619,
|
| 547 |
+
"eval_economic_policy_benefits_f1": 0.840430611429002,
|
| 548 |
+
"eval_ethnic_boundaries_accuracy": 0.9477124183006536,
|
| 549 |
+
"eval_ethnic_boundaries_f1": 0.9385901883316666,
|
| 550 |
+
"eval_language_policy_accuracy": 0.7058823529411765,
|
| 551 |
+
"eval_language_policy_f1": 0.6172459893048128,
|
| 552 |
+
"eval_loss": NaN,
|
| 553 |
+
"eval_mother_tongue_education_accuracy": 0.8888888888888888,
|
| 554 |
+
"eval_mother_tongue_education_f1": 0.837037037037037,
|
| 555 |
+
"eval_overall_accuracy": 0.8529320416673837,
|
| 556 |
+
"eval_overall_f1": 0.8207234712908512,
|
| 557 |
+
"eval_religion_ethnic_identity_accuracy": 0.8781431334622823,
|
| 558 |
+
"eval_religion_ethnic_identity_f1": 0.8667063827853473,
|
| 559 |
+
"eval_runtime": 4.4702,
|
| 560 |
+
"eval_samples_per_second": 894.82,
|
| 561 |
+
"eval_steps_per_second": 55.926,
|
| 562 |
+
"step": 6000
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 6.1,
|
| 566 |
+
"grad_norm": 0.18037013709545135,
|
| 567 |
+
"learning_rate": 6.695180467894308e-06,
|
| 568 |
+
"loss": 0.0,
|
| 569 |
+
"step": 6100
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 6.2,
|
| 573 |
+
"grad_norm": 22.60749626159668,
|
| 574 |
+
"learning_rate": 6.62755921278983e-06,
|
| 575 |
+
"loss": 0.0,
|
| 576 |
+
"step": 6200
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 6.3,
|
| 580 |
+
"grad_norm": 0.8915765881538391,
|
| 581 |
+
"learning_rate": 6.559937957685353e-06,
|
| 582 |
+
"loss": 0.0,
|
| 583 |
+
"step": 6300
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 6.4,
|
| 587 |
+
"grad_norm": 0.5682930946350098,
|
| 588 |
+
"learning_rate": 6.492316702580875e-06,
|
| 589 |
+
"loss": 0.0,
|
| 590 |
+
"step": 6400
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 6.5,
|
| 594 |
+
"grad_norm": 0.08639243990182877,
|
| 595 |
+
"learning_rate": 6.424695447476398e-06,
|
| 596 |
+
"loss": 0.0001,
|
| 597 |
+
"step": 6500
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 6.6,
|
| 601 |
+
"grad_norm": 0.060448452830314636,
|
| 602 |
+
"learning_rate": 6.357074192371921e-06,
|
| 603 |
+
"loss": 0.3897,
|
| 604 |
+
"step": 6600
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 6.7,
|
| 608 |
+
"grad_norm": 0.7599291801452637,
|
| 609 |
+
"learning_rate": 6.289452937267443e-06,
|
| 610 |
+
"loss": 0.0,
|
| 611 |
+
"step": 6700
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 6.8,
|
| 615 |
+
"grad_norm": 0.19008946418762207,
|
| 616 |
+
"learning_rate": 6.2218316821629656e-06,
|
| 617 |
+
"loss": 0.0012,
|
| 618 |
+
"step": 6800
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 6.9,
|
| 622 |
+
"grad_norm": 0.2047979235649109,
|
| 623 |
+
"learning_rate": 6.1542104270584894e-06,
|
| 624 |
+
"loss": 0.0,
|
| 625 |
+
"step": 6900
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 7.0,
|
| 629 |
+
"grad_norm": 4.452627658843994,
|
| 630 |
+
"learning_rate": 6.086589171954012e-06,
|
| 631 |
+
"loss": 0.0,
|
| 632 |
+
"step": 7000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 7.0,
|
| 636 |
+
"eval_economic_inequality_accuracy": 0.8047858942065491,
|
| 637 |
+
"eval_economic_inequality_f1": 0.8098052998323155,
|
| 638 |
+
"eval_economic_policy_benefits_accuracy": 0.8686868686868687,
|
| 639 |
+
"eval_economic_policy_benefits_f1": 0.8627489757007829,
|
| 640 |
+
"eval_ethnic_boundaries_accuracy": 0.9313725490196079,
|
| 641 |
+
"eval_ethnic_boundaries_f1": 0.9331205006157053,
|
| 642 |
+
"eval_language_policy_accuracy": 0.6470588235294118,
|
| 643 |
+
"eval_language_policy_f1": 0.6161764705882353,
|
| 644 |
+
"eval_loss": NaN,
|
| 645 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 646 |
+
"eval_mother_tongue_education_f1": 0.7735042735042735,
|
| 647 |
+
"eval_overall_accuracy": 0.8137799964973732,
|
| 648 |
+
"eval_overall_f1": 0.8086761873509373,
|
| 649 |
+
"eval_religion_ethnic_identity_accuracy": 0.8529980657640233,
|
| 650 |
+
"eval_religion_ethnic_identity_f1": 0.8567016038643116,
|
| 651 |
+
"eval_runtime": 4.489,
|
| 652 |
+
"eval_samples_per_second": 891.066,
|
| 653 |
+
"eval_steps_per_second": 55.692,
|
| 654 |
+
"step": 7000
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 7.1,
|
| 658 |
+
"grad_norm": 0.6880967020988464,
|
| 659 |
+
"learning_rate": 6.018967916849534e-06,
|
| 660 |
+
"loss": 0.0,
|
| 661 |
+
"step": 7100
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 7.2,
|
| 665 |
+
"grad_norm": 1.1063650846481323,
|
| 666 |
+
"learning_rate": 5.951346661745057e-06,
|
| 667 |
+
"loss": 0.0,
|
| 668 |
+
"step": 7200
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 7.3,
|
| 672 |
+
"grad_norm": 0.0691438540816307,
|
| 673 |
+
"learning_rate": 5.883725406640579e-06,
|
| 674 |
+
"loss": 0.0,
|
| 675 |
+
"step": 7300
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 7.4,
|
| 679 |
+
"grad_norm": 0.41019031405448914,
|
| 680 |
+
"learning_rate": 5.816104151536101e-06,
|
| 681 |
+
"loss": 0.0,
|
| 682 |
+
"step": 7400
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 7.5,
|
| 686 |
+
"grad_norm": 0.10174130648374557,
|
| 687 |
+
"learning_rate": 5.748482896431625e-06,
|
| 688 |
+
"loss": 0.0305,
|
| 689 |
+
"step": 7500
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 7.6,
|
| 693 |
+
"grad_norm": 0.7149146795272827,
|
| 694 |
+
"learning_rate": 5.680861641327147e-06,
|
| 695 |
+
"loss": 0.0,
|
| 696 |
+
"step": 7600
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 7.7,
|
| 700 |
+
"grad_norm": 0.03815670683979988,
|
| 701 |
+
"learning_rate": 5.613240386222669e-06,
|
| 702 |
+
"loss": 0.0004,
|
| 703 |
+
"step": 7700
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 7.8,
|
| 707 |
+
"grad_norm": 45.736236572265625,
|
| 708 |
+
"learning_rate": 5.545619131118192e-06,
|
| 709 |
+
"loss": 0.0011,
|
| 710 |
+
"step": 7800
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 7.9,
|
| 714 |
+
"grad_norm": 0.06354359537363052,
|
| 715 |
+
"learning_rate": 5.4779978760137146e-06,
|
| 716 |
+
"loss": 0.0006,
|
| 717 |
+
"step": 7900
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 8.0,
|
| 721 |
+
"grad_norm": 0.4034328758716583,
|
| 722 |
+
"learning_rate": 5.410376620909237e-06,
|
| 723 |
+
"loss": 0.0,
|
| 724 |
+
"step": 8000
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 8.0,
|
| 728 |
+
"eval_economic_inequality_accuracy": 0.8299748110831234,
|
| 729 |
+
"eval_economic_inequality_f1": 0.8275292648803613,
|
| 730 |
+
"eval_economic_policy_benefits_accuracy": 0.8720538720538721,
|
| 731 |
+
"eval_economic_policy_benefits_f1": 0.8639232151195566,
|
| 732 |
+
"eval_ethnic_boundaries_accuracy": 0.9330065359477124,
|
| 733 |
+
"eval_ethnic_boundaries_f1": 0.9382507604268702,
|
| 734 |
+
"eval_language_policy_accuracy": 0.6764705882352942,
|
| 735 |
+
"eval_language_policy_f1": 0.6397058823529412,
|
| 736 |
+
"eval_loss": NaN,
|
| 737 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 738 |
+
"eval_mother_tongue_education_f1": 0.7407407407407408,
|
| 739 |
+
"eval_overall_accuracy": 0.8246807264653618,
|
| 740 |
+
"eval_overall_f1": 0.8118506709859193,
|
| 741 |
+
"eval_religion_ethnic_identity_accuracy": 0.8588007736943907,
|
| 742 |
+
"eval_religion_ethnic_identity_f1": 0.8609541623950459,
|
| 743 |
+
"eval_runtime": 4.4875,
|
| 744 |
+
"eval_samples_per_second": 891.36,
|
| 745 |
+
"eval_steps_per_second": 55.71,
|
| 746 |
+
"step": 8000
|
| 747 |
+
}
|
| 748 |
+
],
|
| 749 |
+
"logging_steps": 100,
|
| 750 |
+
"max_steps": 16000,
|
| 751 |
+
"num_input_tokens_seen": 0,
|
| 752 |
+
"num_train_epochs": 16,
|
| 753 |
+
"save_steps": 500,
|
| 754 |
+
"stateful_callbacks": {
|
| 755 |
+
"EarlyStoppingCallback": {
|
| 756 |
+
"args": {
|
| 757 |
+
"early_stopping_patience": 2,
|
| 758 |
+
"early_stopping_threshold": 0.0
|
| 759 |
+
},
|
| 760 |
+
"attributes": {
|
| 761 |
+
"early_stopping_patience_counter": 2
|
| 762 |
+
}
|
| 763 |
+
},
|
| 764 |
+
"TrainerControl": {
|
| 765 |
+
"args": {
|
| 766 |
+
"should_epoch_stop": false,
|
| 767 |
+
"should_evaluate": false,
|
| 768 |
+
"should_log": false,
|
| 769 |
+
"should_save": true,
|
| 770 |
+
"should_training_stop": true
|
| 771 |
+
},
|
| 772 |
+
"attributes": {}
|
| 773 |
+
}
|
| 774 |
+
},
|
| 775 |
+
"total_flos": 3.3683053215744e+16,
|
| 776 |
+
"train_batch_size": 8,
|
| 777 |
+
"trial_name": null,
|
| 778 |
+
"trial_params": {
|
| 779 |
+
"gradient_accumulation_steps": 2,
|
| 780 |
+
"learning_rate": 1.0596926887422654e-05,
|
| 781 |
+
"num_train_epochs": 16,
|
| 782 |
+
"per_device_train_batch_size": 8,
|
| 783 |
+
"warmup_steps": 329,
|
| 784 |
+
"weight_decay": 0.059185379025811405
|
| 785 |
+
}
|
| 786 |
+
}
|
run-2/checkpoint-8000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:822a4a47339dd0b384da8628d960a86b7107ee8196c01d2a9c3866ddc7aa4cd5
|
| 3 |
+
size 5777
|
run-3/checkpoint-500/config.json
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"id2label": {
|
| 12 |
+
"0": "LABEL_0",
|
| 13 |
+
"1": "LABEL_1",
|
| 14 |
+
"2": "LABEL_2",
|
| 15 |
+
"3": "LABEL_3",
|
| 16 |
+
"4": "LABEL_4",
|
| 17 |
+
"5": "LABEL_5",
|
| 18 |
+
"6": "LABEL_6",
|
| 19 |
+
"7": "LABEL_7",
|
| 20 |
+
"8": "LABEL_8",
|
| 21 |
+
"9": "LABEL_9",
|
| 22 |
+
"10": "LABEL_10",
|
| 23 |
+
"11": "LABEL_11",
|
| 24 |
+
"12": "LABEL_12",
|
| 25 |
+
"13": "LABEL_13",
|
| 26 |
+
"14": "LABEL_14",
|
| 27 |
+
"15": "LABEL_15",
|
| 28 |
+
"16": "LABEL_16",
|
| 29 |
+
"17": "LABEL_17"
|
| 30 |
+
},
|
| 31 |
+
"initializer_range": 0.02,
|
| 32 |
+
"intermediate_size": 3072,
|
| 33 |
+
"label2id": {
|
| 34 |
+
"LABEL_0": 0,
|
| 35 |
+
"LABEL_1": 1,
|
| 36 |
+
"LABEL_10": 10,
|
| 37 |
+
"LABEL_11": 11,
|
| 38 |
+
"LABEL_12": 12,
|
| 39 |
+
"LABEL_13": 13,
|
| 40 |
+
"LABEL_14": 14,
|
| 41 |
+
"LABEL_15": 15,
|
| 42 |
+
"LABEL_16": 16,
|
| 43 |
+
"LABEL_17": 17,
|
| 44 |
+
"LABEL_2": 2,
|
| 45 |
+
"LABEL_3": 3,
|
| 46 |
+
"LABEL_4": 4,
|
| 47 |
+
"LABEL_5": 5,
|
| 48 |
+
"LABEL_6": 6,
|
| 49 |
+
"LABEL_7": 7,
|
| 50 |
+
"LABEL_8": 8,
|
| 51 |
+
"LABEL_9": 9
|
| 52 |
+
},
|
| 53 |
+
"layer_norm_eps": 1e-12,
|
| 54 |
+
"max_position_embeddings": 512,
|
| 55 |
+
"model_type": "bert",
|
| 56 |
+
"num_attention_heads": 12,
|
| 57 |
+
"num_hidden_layers": 12,
|
| 58 |
+
"pad_token_id": 0,
|
| 59 |
+
"position_embedding_type": "absolute",
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.55.4",
|
| 62 |
+
"type_vocab_size": 2,
|
| 63 |
+
"use_cache": true,
|
| 64 |
+
"vocab_size": 30522
|
| 65 |
+
}
|
run-3/checkpoint-500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57f4e2fd1ddd2ee909874242196f4c4a3926a6f9a204a8e0273959572b89fdfb
|
| 3 |
+
size 438007864
|
run-3/checkpoint-500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c502067364cbdddc3a4565fafde1f102d80c8e6ed15b1dc43f951398883b08b
|
| 3 |
+
size 876140043
|
run-3/checkpoint-500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:19fd18bb61689800316944d5e23bd13d5c3d77818a1d26d455c95460f4c55690
|
| 3 |
+
size 14645
|
run-3/checkpoint-500/scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cea1f1d018faaab8b238a04237cc8af304e4c1110a400b5a1087b88aca42873a
|
| 3 |
+
size 1383
|
run-3/checkpoint-500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:07654e3742961315fb2c8639dfb9244654537ddcdedead239c496b8942b67df0
|
| 3 |
+
size 1465
|
run-3/checkpoint-500/trainer_state.json
ADDED
|
@@ -0,0 +1,107 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 500,
|
| 3 |
+
"best_metric": 0.7222001215424422,
|
| 4 |
+
"best_model_checkpoint": "./results/run-3/checkpoint-500",
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 500,
|
| 8 |
+
"is_hyper_param_search": true,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.2,
|
| 14 |
+
"grad_norm": 0.3840266466140747,
|
| 15 |
+
"learning_rate": 5.623773985869002e-05,
|
| 16 |
+
"loss": 0.4015,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.4,
|
| 21 |
+
"grad_norm": 36.44963455200195,
|
| 22 |
+
"learning_rate": 6.805455705058522e-05,
|
| 23 |
+
"loss": 0.0,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.6,
|
| 28 |
+
"grad_norm": 0.6683640480041504,
|
| 29 |
+
"learning_rate": 6.718217457333122e-05,
|
| 30 |
+
"loss": 0.0,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.8,
|
| 35 |
+
"grad_norm": 0.7861915230751038,
|
| 36 |
+
"learning_rate": 6.630979209607721e-05,
|
| 37 |
+
"loss": 0.1644,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0,
|
| 42 |
+
"grad_norm": 2.053349256515503,
|
| 43 |
+
"learning_rate": 6.54374096188232e-05,
|
| 44 |
+
"loss": 0.0025,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.0,
|
| 49 |
+
"eval_economic_inequality_accuracy": 0.7770780856423174,
|
| 50 |
+
"eval_economic_inequality_f1": 0.6795991195482776,
|
| 51 |
+
"eval_economic_policy_benefits_accuracy": 0.8215488215488216,
|
| 52 |
+
"eval_economic_policy_benefits_f1": 0.7410643713786043,
|
| 53 |
+
"eval_ethnic_boundaries_accuracy": 0.946078431372549,
|
| 54 |
+
"eval_ethnic_boundaries_f1": 0.9198646713093299,
|
| 55 |
+
"eval_language_policy_accuracy": 0.6764705882352942,
|
| 56 |
+
"eval_language_policy_f1": 0.545923632610939,
|
| 57 |
+
"eval_loss": NaN,
|
| 58 |
+
"eval_mother_tongue_education_accuracy": 0.7777777777777778,
|
| 59 |
+
"eval_mother_tongue_education_f1": 0.6805555555555556,
|
| 60 |
+
"eval_overall_accuracy": 0.8064020197505432,
|
| 61 |
+
"eval_overall_f1": 0.7222001215424422,
|
| 62 |
+
"eval_religion_ethnic_identity_accuracy": 0.839458413926499,
|
| 63 |
+
"eval_religion_ethnic_identity_f1": 0.7661933788519465,
|
| 64 |
+
"eval_runtime": 4.4769,
|
| 65 |
+
"eval_samples_per_second": 893.468,
|
| 66 |
+
"eval_steps_per_second": 55.842,
|
| 67 |
+
"step": 500
|
| 68 |
+
}
|
| 69 |
+
],
|
| 70 |
+
"logging_steps": 100,
|
| 71 |
+
"max_steps": 8000,
|
| 72 |
+
"num_input_tokens_seen": 0,
|
| 73 |
+
"num_train_epochs": 16,
|
| 74 |
+
"save_steps": 500,
|
| 75 |
+
"stateful_callbacks": {
|
| 76 |
+
"EarlyStoppingCallback": {
|
| 77 |
+
"args": {
|
| 78 |
+
"early_stopping_patience": 2,
|
| 79 |
+
"early_stopping_threshold": 0.0
|
| 80 |
+
},
|
| 81 |
+
"attributes": {
|
| 82 |
+
"early_stopping_patience_counter": 0
|
| 83 |
+
}
|
| 84 |
+
},
|
| 85 |
+
"TrainerControl": {
|
| 86 |
+
"args": {
|
| 87 |
+
"should_epoch_stop": false,
|
| 88 |
+
"should_evaluate": false,
|
| 89 |
+
"should_log": false,
|
| 90 |
+
"should_save": true,
|
| 91 |
+
"should_training_stop": false
|
| 92 |
+
},
|
| 93 |
+
"attributes": {}
|
| 94 |
+
}
|
| 95 |
+
},
|
| 96 |
+
"total_flos": 4210381651968000.0,
|
| 97 |
+
"train_batch_size": 8,
|
| 98 |
+
"trial_name": null,
|
| 99 |
+
"trial_params": {
|
| 100 |
+
"gradient_accumulation_steps": 4,
|
| 101 |
+
"learning_rate": 6.873501538284335e-05,
|
| 102 |
+
"num_train_epochs": 16,
|
| 103 |
+
"per_device_train_batch_size": 8,
|
| 104 |
+
"warmup_steps": 121,
|
| 105 |
+
"weight_decay": 0.08748775333302956
|
| 106 |
+
}
|
| 107 |
+
}
|
run-3/checkpoint-500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f25b0890fc1a55f215dd25d4ee6812fa588a4d653eedd59fd60c8d3a14cbe61a
|
| 3 |
+
size 5777
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5777
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f25b0890fc1a55f215dd25d4ee6812fa588a4d653eedd59fd60c8d3a14cbe61a
|
| 3 |
size 5777
|