Training in progress, step 84000, checkpoint
Browse files- checkpoint-84000/config.json +28 -0
- checkpoint-84000/model.safetensors +3 -0
- checkpoint-84000/optimizer.pt +3 -0
- checkpoint-84000/rng_state.pth +3 -0
- checkpoint-84000/scheduler.pt +3 -0
- checkpoint-84000/special_tokens_map.json +40 -0
- checkpoint-84000/tokenizer.json +0 -0
- checkpoint-84000/tokenizer_config.json +63 -0
- checkpoint-84000/trainer_state.json +516 -0
- checkpoint-84000/training_args.bin +3 -0
checkpoint-84000/config.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForMaskedLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 1,
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.1,
|
| 11 |
+
"hidden_size": 256,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 2048,
|
| 14 |
+
"layer_norm_eps": 1e-05,
|
| 15 |
+
"mask_token_id": 4,
|
| 16 |
+
"max_position_embeddings": 512,
|
| 17 |
+
"model_type": "roberta",
|
| 18 |
+
"num_attention_heads": 8,
|
| 19 |
+
"num_hidden_layers": 8,
|
| 20 |
+
"pad_token_id": 1,
|
| 21 |
+
"position_embedding_type": "absolute",
|
| 22 |
+
"tie_word_embeddings": false,
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.45.2",
|
| 25 |
+
"type_vocab_size": 2,
|
| 26 |
+
"use_cache": true,
|
| 27 |
+
"vocab_size": 8192
|
| 28 |
+
}
|
checkpoint-84000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0b7e4aa4fa7650b35915b2bdd0e74871031838ebfd50ef8dfba3d5820b79b439
|
| 3 |
+
size 59702184
|
checkpoint-84000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8891f3fabafb5fdb5b96b62c45c1422250d86199a789646cd0ec1f583ef35d66
|
| 3 |
+
size 119488058
|
checkpoint-84000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:851691920390c5578fa1edea45c3376617b68a608491611e96cbcc5d32ab3d97
|
| 3 |
+
size 14308
|
checkpoint-84000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:44e34d838d9bd3d4df60f7361ac982148b3a8c2521a8c89fbd7f31f325229f07
|
| 3 |
+
size 1000
|
checkpoint-84000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
"<|endoftext|>"
|
| 4 |
+
],
|
| 5 |
+
"bos_token": {
|
| 6 |
+
"content": "<s>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"eos_token": {
|
| 13 |
+
"content": "</s>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false
|
| 18 |
+
},
|
| 19 |
+
"mask_token": {
|
| 20 |
+
"content": "<mask>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false
|
| 25 |
+
},
|
| 26 |
+
"pad_token": {
|
| 27 |
+
"content": "<pad>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": false,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
},
|
| 33 |
+
"unk_token": {
|
| 34 |
+
"content": "<unk>",
|
| 35 |
+
"lstrip": false,
|
| 36 |
+
"normalized": false,
|
| 37 |
+
"rstrip": false,
|
| 38 |
+
"single_word": false
|
| 39 |
+
}
|
| 40 |
+
}
|
checkpoint-84000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-84000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,63 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "<s>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"1": {
|
| 12 |
+
"content": "</s>",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"2": {
|
| 20 |
+
"content": "<pad>",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"3": {
|
| 28 |
+
"content": "<unk>",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"4": {
|
| 36 |
+
"content": "<mask>",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
},
|
| 43 |
+
"5": {
|
| 44 |
+
"content": "<|endoftext|>",
|
| 45 |
+
"lstrip": false,
|
| 46 |
+
"normalized": false,
|
| 47 |
+
"rstrip": false,
|
| 48 |
+
"single_word": false,
|
| 49 |
+
"special": true
|
| 50 |
+
}
|
| 51 |
+
},
|
| 52 |
+
"additional_special_tokens": [
|
| 53 |
+
"<|endoftext|>"
|
| 54 |
+
],
|
| 55 |
+
"bos_token": "<s>",
|
| 56 |
+
"clean_up_tokenization_spaces": false,
|
| 57 |
+
"eos_token": "</s>",
|
| 58 |
+
"mask_token": "<mask>",
|
| 59 |
+
"model_max_length": 128,
|
| 60 |
+
"pad_token": "<pad>",
|
| 61 |
+
"tokenizer_class": "PreTrainedTokenizerFast",
|
| 62 |
+
"unk_token": "<unk>"
|
| 63 |
+
}
|
checkpoint-84000/trainer_state.json
ADDED
|
@@ -0,0 +1,516 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.8514353036880493,
|
| 3 |
+
"best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_last/fr_mlm/fr_childes_13/checkpoint-84000",
|
| 4 |
+
"epoch": 105.0,
|
| 5 |
+
"eval_steps": 2000,
|
| 6 |
+
"global_step": 84000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 2.5,
|
| 13 |
+
"eval_loss": 7.2409749031066895,
|
| 14 |
+
"eval_runtime": 1.6336,
|
| 15 |
+
"eval_samples_per_second": 1264.105,
|
| 16 |
+
"eval_steps_per_second": 79.58,
|
| 17 |
+
"step": 2000
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 5.0,
|
| 21 |
+
"grad_norm": 1.1403313875198364,
|
| 22 |
+
"learning_rate": 1e-05,
|
| 23 |
+
"loss": 7.2222,
|
| 24 |
+
"step": 4000
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 5.0,
|
| 28 |
+
"eval_loss": 5.968332767486572,
|
| 29 |
+
"eval_runtime": 1.6315,
|
| 30 |
+
"eval_samples_per_second": 1265.736,
|
| 31 |
+
"eval_steps_per_second": 79.683,
|
| 32 |
+
"step": 4000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 7.5,
|
| 36 |
+
"eval_loss": 5.824122428894043,
|
| 37 |
+
"eval_runtime": 1.6212,
|
| 38 |
+
"eval_samples_per_second": 1273.756,
|
| 39 |
+
"eval_steps_per_second": 80.188,
|
| 40 |
+
"step": 6000
|
| 41 |
+
},
|
| 42 |
+
{
|
| 43 |
+
"epoch": 10.0,
|
| 44 |
+
"grad_norm": 1.6872384548187256,
|
| 45 |
+
"learning_rate": 2e-05,
|
| 46 |
+
"loss": 5.6899,
|
| 47 |
+
"step": 8000
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 10.0,
|
| 51 |
+
"eval_loss": 5.726555824279785,
|
| 52 |
+
"eval_runtime": 1.6918,
|
| 53 |
+
"eval_samples_per_second": 1220.594,
|
| 54 |
+
"eval_steps_per_second": 76.841,
|
| 55 |
+
"step": 8000
|
| 56 |
+
},
|
| 57 |
+
{
|
| 58 |
+
"epoch": 12.5,
|
| 59 |
+
"eval_loss": 5.629961013793945,
|
| 60 |
+
"eval_runtime": 1.6387,
|
| 61 |
+
"eval_samples_per_second": 1260.18,
|
| 62 |
+
"eval_steps_per_second": 79.333,
|
| 63 |
+
"step": 10000
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 15.0,
|
| 67 |
+
"grad_norm": 2.588118076324463,
|
| 68 |
+
"learning_rate": 3e-05,
|
| 69 |
+
"loss": 5.4663,
|
| 70 |
+
"step": 12000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 15.0,
|
| 74 |
+
"eval_loss": 5.5523834228515625,
|
| 75 |
+
"eval_runtime": 1.6106,
|
| 76 |
+
"eval_samples_per_second": 1282.116,
|
| 77 |
+
"eval_steps_per_second": 80.714,
|
| 78 |
+
"step": 12000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 17.5,
|
| 82 |
+
"eval_loss": 5.512193202972412,
|
| 83 |
+
"eval_runtime": 1.5999,
|
| 84 |
+
"eval_samples_per_second": 1290.687,
|
| 85 |
+
"eval_steps_per_second": 81.254,
|
| 86 |
+
"step": 14000
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 20.0,
|
| 90 |
+
"grad_norm": 4.425747871398926,
|
| 91 |
+
"learning_rate": 4e-05,
|
| 92 |
+
"loss": 5.3077,
|
| 93 |
+
"step": 16000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 20.0,
|
| 97 |
+
"eval_loss": 5.460000991821289,
|
| 98 |
+
"eval_runtime": 1.6043,
|
| 99 |
+
"eval_samples_per_second": 1287.195,
|
| 100 |
+
"eval_steps_per_second": 81.034,
|
| 101 |
+
"step": 16000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 22.5,
|
| 105 |
+
"eval_loss": 5.401474475860596,
|
| 106 |
+
"eval_runtime": 1.6476,
|
| 107 |
+
"eval_samples_per_second": 1253.315,
|
| 108 |
+
"eval_steps_per_second": 78.901,
|
| 109 |
+
"step": 18000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 25.0,
|
| 113 |
+
"grad_norm": 4.271786212921143,
|
| 114 |
+
"learning_rate": 5e-05,
|
| 115 |
+
"loss": 5.1915,
|
| 116 |
+
"step": 20000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 25.0,
|
| 120 |
+
"eval_loss": 5.336738586425781,
|
| 121 |
+
"eval_runtime": 1.5922,
|
| 122 |
+
"eval_samples_per_second": 1296.931,
|
| 123 |
+
"eval_steps_per_second": 81.647,
|
| 124 |
+
"step": 20000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 27.5,
|
| 128 |
+
"eval_loss": 5.1496429443359375,
|
| 129 |
+
"eval_runtime": 2.7104,
|
| 130 |
+
"eval_samples_per_second": 761.889,
|
| 131 |
+
"eval_steps_per_second": 47.964,
|
| 132 |
+
"step": 22000
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 30.0,
|
| 136 |
+
"grad_norm": 5.581806182861328,
|
| 137 |
+
"learning_rate": 6e-05,
|
| 138 |
+
"loss": 4.8173,
|
| 139 |
+
"step": 24000
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 30.0,
|
| 143 |
+
"eval_loss": 4.08081579208374,
|
| 144 |
+
"eval_runtime": 3.0759,
|
| 145 |
+
"eval_samples_per_second": 671.348,
|
| 146 |
+
"eval_steps_per_second": 42.264,
|
| 147 |
+
"step": 24000
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 32.5,
|
| 151 |
+
"eval_loss": 3.3694963455200195,
|
| 152 |
+
"eval_runtime": 2.9664,
|
| 153 |
+
"eval_samples_per_second": 696.136,
|
| 154 |
+
"eval_steps_per_second": 43.825,
|
| 155 |
+
"step": 26000
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 35.0,
|
| 159 |
+
"grad_norm": 4.369658470153809,
|
| 160 |
+
"learning_rate": 7e-05,
|
| 161 |
+
"loss": 3.387,
|
| 162 |
+
"step": 28000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 35.0,
|
| 166 |
+
"eval_loss": 3.072216510772705,
|
| 167 |
+
"eval_runtime": 3.0861,
|
| 168 |
+
"eval_samples_per_second": 669.14,
|
| 169 |
+
"eval_steps_per_second": 42.125,
|
| 170 |
+
"step": 28000
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 37.5,
|
| 174 |
+
"eval_loss": 2.8991713523864746,
|
| 175 |
+
"eval_runtime": 3.0259,
|
| 176 |
+
"eval_samples_per_second": 682.437,
|
| 177 |
+
"eval_steps_per_second": 42.962,
|
| 178 |
+
"step": 30000
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 40.0,
|
| 182 |
+
"grad_norm": 4.331390380859375,
|
| 183 |
+
"learning_rate": 8e-05,
|
| 184 |
+
"loss": 2.8035,
|
| 185 |
+
"step": 32000
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 40.0,
|
| 189 |
+
"eval_loss": 2.727933168411255,
|
| 190 |
+
"eval_runtime": 3.0386,
|
| 191 |
+
"eval_samples_per_second": 679.589,
|
| 192 |
+
"eval_steps_per_second": 42.783,
|
| 193 |
+
"step": 32000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 42.5,
|
| 197 |
+
"eval_loss": 2.61684513092041,
|
| 198 |
+
"eval_runtime": 3.0184,
|
| 199 |
+
"eval_samples_per_second": 684.128,
|
| 200 |
+
"eval_steps_per_second": 43.069,
|
| 201 |
+
"step": 34000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 45.0,
|
| 205 |
+
"grad_norm": 3.8377530574798584,
|
| 206 |
+
"learning_rate": 9e-05,
|
| 207 |
+
"loss": 2.531,
|
| 208 |
+
"step": 36000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 45.0,
|
| 212 |
+
"eval_loss": 2.557036876678467,
|
| 213 |
+
"eval_runtime": 3.0738,
|
| 214 |
+
"eval_samples_per_second": 671.798,
|
| 215 |
+
"eval_steps_per_second": 42.292,
|
| 216 |
+
"step": 36000
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 47.5,
|
| 220 |
+
"eval_loss": 2.4636330604553223,
|
| 221 |
+
"eval_runtime": 2.9501,
|
| 222 |
+
"eval_samples_per_second": 699.986,
|
| 223 |
+
"eval_steps_per_second": 44.067,
|
| 224 |
+
"step": 38000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 50.0,
|
| 228 |
+
"grad_norm": 3.9490303993225098,
|
| 229 |
+
"learning_rate": 0.0001,
|
| 230 |
+
"loss": 2.3359,
|
| 231 |
+
"step": 40000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 50.0,
|
| 235 |
+
"eval_loss": 2.419264793395996,
|
| 236 |
+
"eval_runtime": 3.1681,
|
| 237 |
+
"eval_samples_per_second": 651.813,
|
| 238 |
+
"eval_steps_per_second": 41.034,
|
| 239 |
+
"step": 40000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 52.5,
|
| 243 |
+
"eval_loss": 2.352754592895508,
|
| 244 |
+
"eval_runtime": 2.6693,
|
| 245 |
+
"eval_samples_per_second": 773.624,
|
| 246 |
+
"eval_steps_per_second": 48.703,
|
| 247 |
+
"step": 42000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 55.0,
|
| 251 |
+
"grad_norm": 3.742015838623047,
|
| 252 |
+
"learning_rate": 9.333333333333334e-05,
|
| 253 |
+
"loss": 2.1893,
|
| 254 |
+
"step": 44000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 55.0,
|
| 258 |
+
"eval_loss": 2.299854278564453,
|
| 259 |
+
"eval_runtime": 1.7519,
|
| 260 |
+
"eval_samples_per_second": 1178.69,
|
| 261 |
+
"eval_steps_per_second": 74.203,
|
| 262 |
+
"step": 44000
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 57.5,
|
| 266 |
+
"eval_loss": 2.2412161827087402,
|
| 267 |
+
"eval_runtime": 2.9705,
|
| 268 |
+
"eval_samples_per_second": 695.158,
|
| 269 |
+
"eval_steps_per_second": 43.763,
|
| 270 |
+
"step": 46000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 60.0,
|
| 274 |
+
"grad_norm": 3.492009162902832,
|
| 275 |
+
"learning_rate": 8.666666666666667e-05,
|
| 276 |
+
"loss": 2.0715,
|
| 277 |
+
"step": 48000
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 60.0,
|
| 281 |
+
"eval_loss": 2.1982405185699463,
|
| 282 |
+
"eval_runtime": 2.9731,
|
| 283 |
+
"eval_samples_per_second": 694.561,
|
| 284 |
+
"eval_steps_per_second": 43.725,
|
| 285 |
+
"step": 48000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 62.5,
|
| 289 |
+
"eval_loss": 2.1534457206726074,
|
| 290 |
+
"eval_runtime": 2.9875,
|
| 291 |
+
"eval_samples_per_second": 691.215,
|
| 292 |
+
"eval_steps_per_second": 43.515,
|
| 293 |
+
"step": 50000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 65.0,
|
| 297 |
+
"grad_norm": 3.800999641418457,
|
| 298 |
+
"learning_rate": 8e-05,
|
| 299 |
+
"loss": 1.974,
|
| 300 |
+
"step": 52000
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 65.0,
|
| 304 |
+
"eval_loss": 2.1155998706817627,
|
| 305 |
+
"eval_runtime": 3.0147,
|
| 306 |
+
"eval_samples_per_second": 684.983,
|
| 307 |
+
"eval_steps_per_second": 43.122,
|
| 308 |
+
"step": 52000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 67.5,
|
| 312 |
+
"eval_loss": 2.1153030395507812,
|
| 313 |
+
"eval_runtime": 3.0526,
|
| 314 |
+
"eval_samples_per_second": 676.464,
|
| 315 |
+
"eval_steps_per_second": 42.586,
|
| 316 |
+
"step": 54000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 70.0,
|
| 320 |
+
"grad_norm": 4.047940731048584,
|
| 321 |
+
"learning_rate": 7.333333333333333e-05,
|
| 322 |
+
"loss": 1.9016,
|
| 323 |
+
"step": 56000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 70.0,
|
| 327 |
+
"eval_loss": 2.062455654144287,
|
| 328 |
+
"eval_runtime": 3.0791,
|
| 329 |
+
"eval_samples_per_second": 670.641,
|
| 330 |
+
"eval_steps_per_second": 42.22,
|
| 331 |
+
"step": 56000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 72.5,
|
| 335 |
+
"eval_loss": 2.027374267578125,
|
| 336 |
+
"eval_runtime": 3.0528,
|
| 337 |
+
"eval_samples_per_second": 676.427,
|
| 338 |
+
"eval_steps_per_second": 42.584,
|
| 339 |
+
"step": 58000
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 75.0,
|
| 343 |
+
"grad_norm": 4.136167049407959,
|
| 344 |
+
"learning_rate": 6.666666666666667e-05,
|
| 345 |
+
"loss": 1.8403,
|
| 346 |
+
"step": 60000
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 75.0,
|
| 350 |
+
"eval_loss": 2.036571741104126,
|
| 351 |
+
"eval_runtime": 3.049,
|
| 352 |
+
"eval_samples_per_second": 677.262,
|
| 353 |
+
"eval_steps_per_second": 42.636,
|
| 354 |
+
"step": 60000
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 77.5,
|
| 358 |
+
"eval_loss": 1.9955813884735107,
|
| 359 |
+
"eval_runtime": 2.9986,
|
| 360 |
+
"eval_samples_per_second": 688.654,
|
| 361 |
+
"eval_steps_per_second": 43.354,
|
| 362 |
+
"step": 62000
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 80.0,
|
| 366 |
+
"grad_norm": 3.4311411380767822,
|
| 367 |
+
"learning_rate": 6e-05,
|
| 368 |
+
"loss": 1.7872,
|
| 369 |
+
"step": 64000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 80.0,
|
| 373 |
+
"eval_loss": 1.9750405550003052,
|
| 374 |
+
"eval_runtime": 1.7157,
|
| 375 |
+
"eval_samples_per_second": 1203.571,
|
| 376 |
+
"eval_steps_per_second": 75.77,
|
| 377 |
+
"step": 64000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 82.5,
|
| 381 |
+
"eval_loss": 1.9521456956863403,
|
| 382 |
+
"eval_runtime": 1.6399,
|
| 383 |
+
"eval_samples_per_second": 1259.214,
|
| 384 |
+
"eval_steps_per_second": 79.273,
|
| 385 |
+
"step": 66000
|
| 386 |
+
},
|
| 387 |
+
{
|
| 388 |
+
"epoch": 85.0,
|
| 389 |
+
"grad_norm": 3.7901430130004883,
|
| 390 |
+
"learning_rate": 5.333333333333333e-05,
|
| 391 |
+
"loss": 1.7439,
|
| 392 |
+
"step": 68000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 85.0,
|
| 396 |
+
"eval_loss": 1.957872986793518,
|
| 397 |
+
"eval_runtime": 2.9034,
|
| 398 |
+
"eval_samples_per_second": 711.242,
|
| 399 |
+
"eval_steps_per_second": 44.776,
|
| 400 |
+
"step": 68000
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 87.5,
|
| 404 |
+
"eval_loss": 1.9301824569702148,
|
| 405 |
+
"eval_runtime": 2.9913,
|
| 406 |
+
"eval_samples_per_second": 690.335,
|
| 407 |
+
"eval_steps_per_second": 43.459,
|
| 408 |
+
"step": 70000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 90.0,
|
| 412 |
+
"grad_norm": 3.5553386211395264,
|
| 413 |
+
"learning_rate": 4.666666666666667e-05,
|
| 414 |
+
"loss": 1.7084,
|
| 415 |
+
"step": 72000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 90.0,
|
| 419 |
+
"eval_loss": 1.9384183883666992,
|
| 420 |
+
"eval_runtime": 2.9754,
|
| 421 |
+
"eval_samples_per_second": 694.034,
|
| 422 |
+
"eval_steps_per_second": 43.692,
|
| 423 |
+
"step": 72000
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 92.5,
|
| 427 |
+
"eval_loss": 1.9067375659942627,
|
| 428 |
+
"eval_runtime": 3.0554,
|
| 429 |
+
"eval_samples_per_second": 675.848,
|
| 430 |
+
"eval_steps_per_second": 42.547,
|
| 431 |
+
"step": 74000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 95.0,
|
| 435 |
+
"grad_norm": 3.8471336364746094,
|
| 436 |
+
"learning_rate": 4e-05,
|
| 437 |
+
"loss": 1.6752,
|
| 438 |
+
"step": 76000
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 95.0,
|
| 442 |
+
"eval_loss": 1.9238947629928589,
|
| 443 |
+
"eval_runtime": 3.0005,
|
| 444 |
+
"eval_samples_per_second": 688.214,
|
| 445 |
+
"eval_steps_per_second": 43.326,
|
| 446 |
+
"step": 76000
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 97.5,
|
| 450 |
+
"eval_loss": 1.8918030261993408,
|
| 451 |
+
"eval_runtime": 2.9984,
|
| 452 |
+
"eval_samples_per_second": 688.697,
|
| 453 |
+
"eval_steps_per_second": 43.356,
|
| 454 |
+
"step": 78000
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 100.0,
|
| 458 |
+
"grad_norm": 3.8865303993225098,
|
| 459 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 460 |
+
"loss": 1.6482,
|
| 461 |
+
"step": 80000
|
| 462 |
+
},
|
| 463 |
+
{
|
| 464 |
+
"epoch": 100.0,
|
| 465 |
+
"eval_loss": 1.8802706003189087,
|
| 466 |
+
"eval_runtime": 2.9835,
|
| 467 |
+
"eval_samples_per_second": 692.144,
|
| 468 |
+
"eval_steps_per_second": 43.573,
|
| 469 |
+
"step": 80000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 102.5,
|
| 473 |
+
"eval_loss": 1.8641291856765747,
|
| 474 |
+
"eval_runtime": 3.0131,
|
| 475 |
+
"eval_samples_per_second": 685.343,
|
| 476 |
+
"eval_steps_per_second": 43.145,
|
| 477 |
+
"step": 82000
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 105.0,
|
| 481 |
+
"grad_norm": 3.7413618564605713,
|
| 482 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 483 |
+
"loss": 1.6209,
|
| 484 |
+
"step": 84000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 105.0,
|
| 488 |
+
"eval_loss": 1.8514353036880493,
|
| 489 |
+
"eval_runtime": 3.0259,
|
| 490 |
+
"eval_samples_per_second": 682.443,
|
| 491 |
+
"eval_steps_per_second": 42.962,
|
| 492 |
+
"step": 84000
|
| 493 |
+
}
|
| 494 |
+
],
|
| 495 |
+
"logging_steps": 4000,
|
| 496 |
+
"max_steps": 100000,
|
| 497 |
+
"num_input_tokens_seen": 0,
|
| 498 |
+
"num_train_epochs": 125,
|
| 499 |
+
"save_steps": 4000,
|
| 500 |
+
"stateful_callbacks": {
|
| 501 |
+
"TrainerControl": {
|
| 502 |
+
"args": {
|
| 503 |
+
"should_epoch_stop": false,
|
| 504 |
+
"should_evaluate": false,
|
| 505 |
+
"should_log": false,
|
| 506 |
+
"should_save": true,
|
| 507 |
+
"should_training_stop": false
|
| 508 |
+
},
|
| 509 |
+
"attributes": {}
|
| 510 |
+
}
|
| 511 |
+
},
|
| 512 |
+
"total_flos": 2.6202681114624e+16,
|
| 513 |
+
"train_batch_size": 16,
|
| 514 |
+
"trial_name": null,
|
| 515 |
+
"trial_params": null
|
| 516 |
+
}
|
checkpoint-84000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f019ce7e3a851c79459abf282e5a1f1c8e1b3e048abd2b918873801766590a02
|
| 3 |
+
size 5368
|