Upload 16 files
Browse files- checkpoint-2250/config.json +69 -0
- checkpoint-2250/generation_config.json +16 -0
- checkpoint-2250/model.safetensors +3 -0
- checkpoint-2250/optimizer.pt +3 -0
- checkpoint-2250/rng_state.pth +3 -0
- checkpoint-2250/scheduler.pt +3 -0
- checkpoint-2250/trainer_state.json +818 -0
- checkpoint-2250/training_args.bin +3 -0
- config.json +69 -0
- generation_config.json +16 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- vocab.json +0 -0
checkpoint-2250/config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_num_labels": 3,
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.0,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 12,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"force_bos_token_to_be_generated": true,
|
| 27 |
+
"forced_eos_token_id": 2,
|
| 28 |
+
"gradient_checkpointing": false,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_decoder": true,
|
| 36 |
+
"is_encoder_decoder": false,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"LABEL_0": 0,
|
| 39 |
+
"LABEL_1": 1,
|
| 40 |
+
"LABEL_2": 2
|
| 41 |
+
},
|
| 42 |
+
"length_penalty": null,
|
| 43 |
+
"max_length": null,
|
| 44 |
+
"max_position_embeddings": 1024,
|
| 45 |
+
"min_length": null,
|
| 46 |
+
"model_type": "bart",
|
| 47 |
+
"no_repeat_ngram_size": null,
|
| 48 |
+
"normalize_before": false,
|
| 49 |
+
"num_beams": null,
|
| 50 |
+
"num_hidden_layers": 12,
|
| 51 |
+
"output_past": true,
|
| 52 |
+
"pad_token_id": 1,
|
| 53 |
+
"prefix": " ",
|
| 54 |
+
"scale_embedding": false,
|
| 55 |
+
"task_specific_params": {
|
| 56 |
+
"summarization": {
|
| 57 |
+
"early_stopping": true,
|
| 58 |
+
"length_penalty": 2.0,
|
| 59 |
+
"max_length": 142,
|
| 60 |
+
"min_length": 56,
|
| 61 |
+
"no_repeat_ngram_size": 3,
|
| 62 |
+
"num_beams": 4
|
| 63 |
+
}
|
| 64 |
+
},
|
| 65 |
+
"torch_dtype": "float32",
|
| 66 |
+
"transformers_version": "4.52.4",
|
| 67 |
+
"use_cache": true,
|
| 68 |
+
"vocab_size": 50264
|
| 69 |
+
}
|
checkpoint-2250/generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"early_stopping": true,
|
| 6 |
+
"eos_token_id": 2,
|
| 7 |
+
"forced_bos_token_id": 0,
|
| 8 |
+
"forced_eos_token_id": 2,
|
| 9 |
+
"length_penalty": 2.0,
|
| 10 |
+
"max_length": 142,
|
| 11 |
+
"min_length": 56,
|
| 12 |
+
"no_repeat_ngram_size": 3,
|
| 13 |
+
"num_beams": 4,
|
| 14 |
+
"pad_token_id": 1,
|
| 15 |
+
"transformers_version": "4.52.4"
|
| 16 |
+
}
|
checkpoint-2250/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74d34ed7efb5a94ca559d7d28c72f5c61aff49639cb20fdf3774e3fae7b355d7
|
| 3 |
+
size 1016369648
|
checkpoint-2250/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0d2b1ec36e2d17c6d9064563b8ff3ad38fd0085030205b4c3023e0e423d396a1
|
| 3 |
+
size 1629580410
|
checkpoint-2250/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:57d6afeb259c99a77dd9085eca0ede9f860c316c3a8e068727600fe1f506487f
|
| 3 |
+
size 13990
|
checkpoint-2250/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bec953c174e41850d61f3908d88b9683a2b8a87c5e9752e8ae44ce839aa2004
|
| 3 |
+
size 1064
|
checkpoint-2250/trainer_state.json
ADDED
|
@@ -0,0 +1,818 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 1.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2250,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.008888888888888889,
|
| 14 |
+
"grad_norm": 29.719377517700195,
|
| 15 |
+
"learning_rate": 4.957777777777778e-05,
|
| 16 |
+
"loss": 9.931,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.017777777777777778,
|
| 21 |
+
"grad_norm": 29.342676162719727,
|
| 22 |
+
"learning_rate": 4.913333333333334e-05,
|
| 23 |
+
"loss": 7.8363,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02666666666666667,
|
| 28 |
+
"grad_norm": 30.171451568603516,
|
| 29 |
+
"learning_rate": 4.868888888888889e-05,
|
| 30 |
+
"loss": 5.9641,
|
| 31 |
+
"step": 60
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.035555555555555556,
|
| 35 |
+
"grad_norm": 27.37605857849121,
|
| 36 |
+
"learning_rate": 4.824444444444445e-05,
|
| 37 |
+
"loss": 4.3778,
|
| 38 |
+
"step": 80
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.044444444444444446,
|
| 42 |
+
"grad_norm": 9.228110313415527,
|
| 43 |
+
"learning_rate": 4.78e-05,
|
| 44 |
+
"loss": 3.2025,
|
| 45 |
+
"step": 100
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.05333333333333334,
|
| 49 |
+
"grad_norm": 4.524937152862549,
|
| 50 |
+
"learning_rate": 4.7355555555555555e-05,
|
| 51 |
+
"loss": 2.8068,
|
| 52 |
+
"step": 120
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.06222222222222222,
|
| 56 |
+
"grad_norm": 2.696683168411255,
|
| 57 |
+
"learning_rate": 4.6911111111111114e-05,
|
| 58 |
+
"loss": 2.7047,
|
| 59 |
+
"step": 140
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.07111111111111111,
|
| 63 |
+
"grad_norm": 3.652080535888672,
|
| 64 |
+
"learning_rate": 4.646666666666667e-05,
|
| 65 |
+
"loss": 2.5618,
|
| 66 |
+
"step": 160
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.08,
|
| 70 |
+
"grad_norm": 3.995697021484375,
|
| 71 |
+
"learning_rate": 4.602222222222222e-05,
|
| 72 |
+
"loss": 2.556,
|
| 73 |
+
"step": 180
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.08888888888888889,
|
| 77 |
+
"grad_norm": 4.581667423248291,
|
| 78 |
+
"learning_rate": 4.557777777777778e-05,
|
| 79 |
+
"loss": 2.4783,
|
| 80 |
+
"step": 200
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.09777777777777778,
|
| 84 |
+
"grad_norm": 2.2177352905273438,
|
| 85 |
+
"learning_rate": 4.513333333333333e-05,
|
| 86 |
+
"loss": 2.538,
|
| 87 |
+
"step": 220
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.10666666666666667,
|
| 91 |
+
"grad_norm": 26.386642456054688,
|
| 92 |
+
"learning_rate": 4.468888888888889e-05,
|
| 93 |
+
"loss": 2.4825,
|
| 94 |
+
"step": 240
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.11555555555555555,
|
| 98 |
+
"grad_norm": 3.9231057167053223,
|
| 99 |
+
"learning_rate": 4.424444444444444e-05,
|
| 100 |
+
"loss": 2.5041,
|
| 101 |
+
"step": 260
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.12444444444444444,
|
| 105 |
+
"grad_norm": 11.716802597045898,
|
| 106 |
+
"learning_rate": 4.38e-05,
|
| 107 |
+
"loss": 2.4963,
|
| 108 |
+
"step": 280
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.13333333333333333,
|
| 112 |
+
"grad_norm": 9.851195335388184,
|
| 113 |
+
"learning_rate": 4.335555555555556e-05,
|
| 114 |
+
"loss": 2.3346,
|
| 115 |
+
"step": 300
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.14222222222222222,
|
| 119 |
+
"grad_norm": 43.87732696533203,
|
| 120 |
+
"learning_rate": 4.291111111111111e-05,
|
| 121 |
+
"loss": 2.5413,
|
| 122 |
+
"step": 320
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.1511111111111111,
|
| 126 |
+
"grad_norm": 11.769241333007812,
|
| 127 |
+
"learning_rate": 4.246666666666667e-05,
|
| 128 |
+
"loss": 2.5451,
|
| 129 |
+
"step": 340
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.16,
|
| 133 |
+
"grad_norm": 5.8793463706970215,
|
| 134 |
+
"learning_rate": 4.2022222222222223e-05,
|
| 135 |
+
"loss": 2.3624,
|
| 136 |
+
"step": 360
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.1688888888888889,
|
| 140 |
+
"grad_norm": 2.2031548023223877,
|
| 141 |
+
"learning_rate": 4.157777777777778e-05,
|
| 142 |
+
"loss": 2.3517,
|
| 143 |
+
"step": 380
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.17777777777777778,
|
| 147 |
+
"grad_norm": 1.8488436937332153,
|
| 148 |
+
"learning_rate": 4.1133333333333335e-05,
|
| 149 |
+
"loss": 2.3944,
|
| 150 |
+
"step": 400
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.18666666666666668,
|
| 154 |
+
"grad_norm": 2.724890947341919,
|
| 155 |
+
"learning_rate": 4.0688888888888894e-05,
|
| 156 |
+
"loss": 2.3899,
|
| 157 |
+
"step": 420
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.19555555555555557,
|
| 161 |
+
"grad_norm": 27.8278751373291,
|
| 162 |
+
"learning_rate": 4.0244444444444446e-05,
|
| 163 |
+
"loss": 2.4001,
|
| 164 |
+
"step": 440
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.20444444444444446,
|
| 168 |
+
"grad_norm": 3.687330961227417,
|
| 169 |
+
"learning_rate": 3.9800000000000005e-05,
|
| 170 |
+
"loss": 2.4406,
|
| 171 |
+
"step": 460
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.21333333333333335,
|
| 175 |
+
"grad_norm": 12.961912155151367,
|
| 176 |
+
"learning_rate": 3.935555555555556e-05,
|
| 177 |
+
"loss": 2.4021,
|
| 178 |
+
"step": 480
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.2222222222222222,
|
| 182 |
+
"grad_norm": 2.5192787647247314,
|
| 183 |
+
"learning_rate": 3.8911111111111117e-05,
|
| 184 |
+
"loss": 2.345,
|
| 185 |
+
"step": 500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.2311111111111111,
|
| 189 |
+
"grad_norm": 2.8883752822875977,
|
| 190 |
+
"learning_rate": 3.846666666666667e-05,
|
| 191 |
+
"loss": 2.2426,
|
| 192 |
+
"step": 520
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.24,
|
| 196 |
+
"grad_norm": 3.0005998611450195,
|
| 197 |
+
"learning_rate": 3.802222222222223e-05,
|
| 198 |
+
"loss": 2.2591,
|
| 199 |
+
"step": 540
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.24888888888888888,
|
| 203 |
+
"grad_norm": 1.405104398727417,
|
| 204 |
+
"learning_rate": 3.757777777777778e-05,
|
| 205 |
+
"loss": 2.0771,
|
| 206 |
+
"step": 560
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2577777777777778,
|
| 210 |
+
"grad_norm": 1.6733466386795044,
|
| 211 |
+
"learning_rate": 3.713333333333334e-05,
|
| 212 |
+
"loss": 2.031,
|
| 213 |
+
"step": 580
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.26666666666666666,
|
| 217 |
+
"grad_norm": 1.3471635580062866,
|
| 218 |
+
"learning_rate": 3.668888888888889e-05,
|
| 219 |
+
"loss": 2.0749,
|
| 220 |
+
"step": 600
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.27555555555555555,
|
| 224 |
+
"grad_norm": 1.6390342712402344,
|
| 225 |
+
"learning_rate": 3.624444444444445e-05,
|
| 226 |
+
"loss": 2.0469,
|
| 227 |
+
"step": 620
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.28444444444444444,
|
| 231 |
+
"grad_norm": 1.3594388961791992,
|
| 232 |
+
"learning_rate": 3.58e-05,
|
| 233 |
+
"loss": 2.0836,
|
| 234 |
+
"step": 640
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.29333333333333333,
|
| 238 |
+
"grad_norm": 1.5752489566802979,
|
| 239 |
+
"learning_rate": 3.5355555555555555e-05,
|
| 240 |
+
"loss": 2.0011,
|
| 241 |
+
"step": 660
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.3022222222222222,
|
| 245 |
+
"grad_norm": 2.0753142833709717,
|
| 246 |
+
"learning_rate": 3.4911111111111114e-05,
|
| 247 |
+
"loss": 2.1448,
|
| 248 |
+
"step": 680
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.3111111111111111,
|
| 252 |
+
"grad_norm": 1.4433261156082153,
|
| 253 |
+
"learning_rate": 3.4466666666666666e-05,
|
| 254 |
+
"loss": 1.9837,
|
| 255 |
+
"step": 700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.32,
|
| 259 |
+
"grad_norm": 1.2903432846069336,
|
| 260 |
+
"learning_rate": 3.402222222222222e-05,
|
| 261 |
+
"loss": 2.045,
|
| 262 |
+
"step": 720
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.3288888888888889,
|
| 266 |
+
"grad_norm": 1.120526671409607,
|
| 267 |
+
"learning_rate": 3.357777777777778e-05,
|
| 268 |
+
"loss": 2.0112,
|
| 269 |
+
"step": 740
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.3377777777777778,
|
| 273 |
+
"grad_norm": 1.4709597826004028,
|
| 274 |
+
"learning_rate": 3.313333333333333e-05,
|
| 275 |
+
"loss": 1.9936,
|
| 276 |
+
"step": 760
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.3466666666666667,
|
| 280 |
+
"grad_norm": 1.3369895219802856,
|
| 281 |
+
"learning_rate": 3.268888888888889e-05,
|
| 282 |
+
"loss": 1.9659,
|
| 283 |
+
"step": 780
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.35555555555555557,
|
| 287 |
+
"grad_norm": 1.1599270105361938,
|
| 288 |
+
"learning_rate": 3.224444444444444e-05,
|
| 289 |
+
"loss": 1.9923,
|
| 290 |
+
"step": 800
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.36444444444444446,
|
| 294 |
+
"grad_norm": 1.6374409198760986,
|
| 295 |
+
"learning_rate": 3.18e-05,
|
| 296 |
+
"loss": 1.9951,
|
| 297 |
+
"step": 820
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.37333333333333335,
|
| 301 |
+
"grad_norm": 1.8066844940185547,
|
| 302 |
+
"learning_rate": 3.135555555555555e-05,
|
| 303 |
+
"loss": 2.0328,
|
| 304 |
+
"step": 840
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.38222222222222224,
|
| 308 |
+
"grad_norm": 2.726418972015381,
|
| 309 |
+
"learning_rate": 3.091111111111111e-05,
|
| 310 |
+
"loss": 3.0763,
|
| 311 |
+
"step": 860
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.39111111111111113,
|
| 315 |
+
"grad_norm": 1.5949397087097168,
|
| 316 |
+
"learning_rate": 3.0466666666666664e-05,
|
| 317 |
+
"loss": 2.0093,
|
| 318 |
+
"step": 880
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.4,
|
| 322 |
+
"grad_norm": 2.8492074012756348,
|
| 323 |
+
"learning_rate": 3.0022222222222223e-05,
|
| 324 |
+
"loss": 2.017,
|
| 325 |
+
"step": 900
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.4088888888888889,
|
| 329 |
+
"grad_norm": 1.2635244131088257,
|
| 330 |
+
"learning_rate": 2.9577777777777775e-05,
|
| 331 |
+
"loss": 2.0231,
|
| 332 |
+
"step": 920
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.4177777777777778,
|
| 336 |
+
"grad_norm": 2.1309561729431152,
|
| 337 |
+
"learning_rate": 2.9133333333333334e-05,
|
| 338 |
+
"loss": 2.1117,
|
| 339 |
+
"step": 940
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.4266666666666667,
|
| 343 |
+
"grad_norm": 39.7406005859375,
|
| 344 |
+
"learning_rate": 2.8688888888888894e-05,
|
| 345 |
+
"loss": 2.0463,
|
| 346 |
+
"step": 960
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.43555555555555553,
|
| 350 |
+
"grad_norm": 1.5083988904953003,
|
| 351 |
+
"learning_rate": 2.8244444444444446e-05,
|
| 352 |
+
"loss": 2.0512,
|
| 353 |
+
"step": 980
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.4444444444444444,
|
| 357 |
+
"grad_norm": 1.9805152416229248,
|
| 358 |
+
"learning_rate": 2.7800000000000005e-05,
|
| 359 |
+
"loss": 2.0151,
|
| 360 |
+
"step": 1000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.4533333333333333,
|
| 364 |
+
"grad_norm": 1.4856139421463013,
|
| 365 |
+
"learning_rate": 2.7355555555555557e-05,
|
| 366 |
+
"loss": 1.9718,
|
| 367 |
+
"step": 1020
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.4622222222222222,
|
| 371 |
+
"grad_norm": 1.5908904075622559,
|
| 372 |
+
"learning_rate": 2.6911111111111116e-05,
|
| 373 |
+
"loss": 1.9655,
|
| 374 |
+
"step": 1040
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.4711111111111111,
|
| 378 |
+
"grad_norm": 1.5324510335922241,
|
| 379 |
+
"learning_rate": 2.646666666666667e-05,
|
| 380 |
+
"loss": 1.93,
|
| 381 |
+
"step": 1060
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.48,
|
| 385 |
+
"grad_norm": 1.6578165292739868,
|
| 386 |
+
"learning_rate": 2.6022222222222224e-05,
|
| 387 |
+
"loss": 1.9815,
|
| 388 |
+
"step": 1080
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.4888888888888889,
|
| 392 |
+
"grad_norm": 1.9318180084228516,
|
| 393 |
+
"learning_rate": 2.557777777777778e-05,
|
| 394 |
+
"loss": 1.9186,
|
| 395 |
+
"step": 1100
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.49777777777777776,
|
| 399 |
+
"grad_norm": 4.174930095672607,
|
| 400 |
+
"learning_rate": 2.5133333333333336e-05,
|
| 401 |
+
"loss": 1.9262,
|
| 402 |
+
"step": 1120
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.5066666666666667,
|
| 406 |
+
"grad_norm": 43.6900520324707,
|
| 407 |
+
"learning_rate": 2.4688888888888888e-05,
|
| 408 |
+
"loss": 1.9956,
|
| 409 |
+
"step": 1140
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.5155555555555555,
|
| 413 |
+
"grad_norm": 1.4572045803070068,
|
| 414 |
+
"learning_rate": 2.4244444444444443e-05,
|
| 415 |
+
"loss": 1.9548,
|
| 416 |
+
"step": 1160
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.5244444444444445,
|
| 420 |
+
"grad_norm": 3.6926634311676025,
|
| 421 |
+
"learning_rate": 2.38e-05,
|
| 422 |
+
"loss": 1.9629,
|
| 423 |
+
"step": 1180
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.5333333333333333,
|
| 427 |
+
"grad_norm": 1.4738409519195557,
|
| 428 |
+
"learning_rate": 2.3355555555555555e-05,
|
| 429 |
+
"loss": 2.01,
|
| 430 |
+
"step": 1200
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.5422222222222223,
|
| 434 |
+
"grad_norm": 2.078120231628418,
|
| 435 |
+
"learning_rate": 2.291111111111111e-05,
|
| 436 |
+
"loss": 2.0279,
|
| 437 |
+
"step": 1220
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.5511111111111111,
|
| 441 |
+
"grad_norm": 1.4616929292678833,
|
| 442 |
+
"learning_rate": 2.2466666666666666e-05,
|
| 443 |
+
"loss": 1.9756,
|
| 444 |
+
"step": 1240
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.56,
|
| 448 |
+
"grad_norm": 1.8624461889266968,
|
| 449 |
+
"learning_rate": 2.2022222222222225e-05,
|
| 450 |
+
"loss": 1.9866,
|
| 451 |
+
"step": 1260
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.5688888888888889,
|
| 455 |
+
"grad_norm": 2.9291300773620605,
|
| 456 |
+
"learning_rate": 2.157777777777778e-05,
|
| 457 |
+
"loss": 2.0013,
|
| 458 |
+
"step": 1280
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.5777777777777777,
|
| 462 |
+
"grad_norm": 1.6963049173355103,
|
| 463 |
+
"learning_rate": 2.1133333333333337e-05,
|
| 464 |
+
"loss": 1.9604,
|
| 465 |
+
"step": 1300
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.5866666666666667,
|
| 469 |
+
"grad_norm": 4.135189056396484,
|
| 470 |
+
"learning_rate": 2.0688888888888892e-05,
|
| 471 |
+
"loss": 1.9069,
|
| 472 |
+
"step": 1320
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.5955555555555555,
|
| 476 |
+
"grad_norm": 1.9753628969192505,
|
| 477 |
+
"learning_rate": 2.0244444444444448e-05,
|
| 478 |
+
"loss": 1.8923,
|
| 479 |
+
"step": 1340
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.6044444444444445,
|
| 483 |
+
"grad_norm": 1.6017193794250488,
|
| 484 |
+
"learning_rate": 1.9800000000000004e-05,
|
| 485 |
+
"loss": 2.0029,
|
| 486 |
+
"step": 1360
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.6133333333333333,
|
| 490 |
+
"grad_norm": 10.354166030883789,
|
| 491 |
+
"learning_rate": 1.9355555555555556e-05,
|
| 492 |
+
"loss": 2.0181,
|
| 493 |
+
"step": 1380
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.6222222222222222,
|
| 497 |
+
"grad_norm": 1.5218088626861572,
|
| 498 |
+
"learning_rate": 1.891111111111111e-05,
|
| 499 |
+
"loss": 1.9389,
|
| 500 |
+
"step": 1400
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.6311111111111111,
|
| 504 |
+
"grad_norm": 3.808530569076538,
|
| 505 |
+
"learning_rate": 1.8466666666666667e-05,
|
| 506 |
+
"loss": 2.0232,
|
| 507 |
+
"step": 1420
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 0.64,
|
| 511 |
+
"grad_norm": 2.383110761642456,
|
| 512 |
+
"learning_rate": 1.8022222222222223e-05,
|
| 513 |
+
"loss": 2.0305,
|
| 514 |
+
"step": 1440
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 0.6488888888888888,
|
| 518 |
+
"grad_norm": 3.2716569900512695,
|
| 519 |
+
"learning_rate": 1.757777777777778e-05,
|
| 520 |
+
"loss": 2.0708,
|
| 521 |
+
"step": 1460
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 0.6577777777777778,
|
| 525 |
+
"grad_norm": 28.891468048095703,
|
| 526 |
+
"learning_rate": 1.7133333333333334e-05,
|
| 527 |
+
"loss": 2.2954,
|
| 528 |
+
"step": 1480
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 0.6666666666666666,
|
| 532 |
+
"grad_norm": 2.7744553089141846,
|
| 533 |
+
"learning_rate": 1.668888888888889e-05,
|
| 534 |
+
"loss": 2.1343,
|
| 535 |
+
"step": 1500
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 0.6755555555555556,
|
| 539 |
+
"grad_norm": 3.43209171295166,
|
| 540 |
+
"learning_rate": 1.6244444444444446e-05,
|
| 541 |
+
"loss": 2.1159,
|
| 542 |
+
"step": 1520
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.6844444444444444,
|
| 546 |
+
"grad_norm": 1.6540969610214233,
|
| 547 |
+
"learning_rate": 1.58e-05,
|
| 548 |
+
"loss": 1.9926,
|
| 549 |
+
"step": 1540
|
| 550 |
+
},
|
| 551 |
+
{
|
| 552 |
+
"epoch": 0.6933333333333334,
|
| 553 |
+
"grad_norm": 1.8530590534210205,
|
| 554 |
+
"learning_rate": 1.5355555555555557e-05,
|
| 555 |
+
"loss": 1.9936,
|
| 556 |
+
"step": 1560
|
| 557 |
+
},
|
| 558 |
+
{
|
| 559 |
+
"epoch": 0.7022222222222222,
|
| 560 |
+
"grad_norm": 8.5066556930542,
|
| 561 |
+
"learning_rate": 1.4911111111111113e-05,
|
| 562 |
+
"loss": 2.04,
|
| 563 |
+
"step": 1580
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 0.7111111111111111,
|
| 567 |
+
"grad_norm": 3.0773701667785645,
|
| 568 |
+
"learning_rate": 1.4466666666666667e-05,
|
| 569 |
+
"loss": 2.0861,
|
| 570 |
+
"step": 1600
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 0.72,
|
| 574 |
+
"grad_norm": 1.7106276750564575,
|
| 575 |
+
"learning_rate": 1.4022222222222222e-05,
|
| 576 |
+
"loss": 1.9936,
|
| 577 |
+
"step": 1620
|
| 578 |
+
},
|
| 579 |
+
{
|
| 580 |
+
"epoch": 0.7288888888888889,
|
| 581 |
+
"grad_norm": 8.246357917785645,
|
| 582 |
+
"learning_rate": 1.3577777777777778e-05,
|
| 583 |
+
"loss": 2.0049,
|
| 584 |
+
"step": 1640
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.7377777777777778,
|
| 588 |
+
"grad_norm": 1.527330756187439,
|
| 589 |
+
"learning_rate": 1.3133333333333334e-05,
|
| 590 |
+
"loss": 1.9879,
|
| 591 |
+
"step": 1660
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 0.7466666666666667,
|
| 595 |
+
"grad_norm": 1.4043477773666382,
|
| 596 |
+
"learning_rate": 1.268888888888889e-05,
|
| 597 |
+
"loss": 1.952,
|
| 598 |
+
"step": 1680
|
| 599 |
+
},
|
| 600 |
+
{
|
| 601 |
+
"epoch": 0.7555555555555555,
|
| 602 |
+
"grad_norm": 1.3344300985336304,
|
| 603 |
+
"learning_rate": 1.2244444444444445e-05,
|
| 604 |
+
"loss": 2.0598,
|
| 605 |
+
"step": 1700
|
| 606 |
+
},
|
| 607 |
+
{
|
| 608 |
+
"epoch": 0.7644444444444445,
|
| 609 |
+
"grad_norm": 1.5034434795379639,
|
| 610 |
+
"learning_rate": 1.18e-05,
|
| 611 |
+
"loss": 2.0192,
|
| 612 |
+
"step": 1720
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 0.7733333333333333,
|
| 616 |
+
"grad_norm": 1.9273996353149414,
|
| 617 |
+
"learning_rate": 1.1355555555555556e-05,
|
| 618 |
+
"loss": 1.9587,
|
| 619 |
+
"step": 1740
|
| 620 |
+
},
|
| 621 |
+
{
|
| 622 |
+
"epoch": 0.7822222222222223,
|
| 623 |
+
"grad_norm": 1.5485738515853882,
|
| 624 |
+
"learning_rate": 1.0911111111111112e-05,
|
| 625 |
+
"loss": 2.0159,
|
| 626 |
+
"step": 1760
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.7911111111111111,
|
| 630 |
+
"grad_norm": 2.0449929237365723,
|
| 631 |
+
"learning_rate": 1.0466666666666668e-05,
|
| 632 |
+
"loss": 1.9578,
|
| 633 |
+
"step": 1780
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8,
|
| 637 |
+
"grad_norm": 1.3243677616119385,
|
| 638 |
+
"learning_rate": 1.0022222222222223e-05,
|
| 639 |
+
"loss": 1.9961,
|
| 640 |
+
"step": 1800
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8088888888888889,
|
| 644 |
+
"grad_norm": 2.067218542098999,
|
| 645 |
+
"learning_rate": 9.577777777777779e-06,
|
| 646 |
+
"loss": 1.9834,
|
| 647 |
+
"step": 1820
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.8177777777777778,
|
| 651 |
+
"grad_norm": 1.5382347106933594,
|
| 652 |
+
"learning_rate": 9.133333333333335e-06,
|
| 653 |
+
"loss": 1.9732,
|
| 654 |
+
"step": 1840
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.8266666666666667,
|
| 658 |
+
"grad_norm": 1.530358910560608,
|
| 659 |
+
"learning_rate": 8.68888888888889e-06,
|
| 660 |
+
"loss": 1.92,
|
| 661 |
+
"step": 1860
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.8355555555555556,
|
| 665 |
+
"grad_norm": 1.3091119527816772,
|
| 666 |
+
"learning_rate": 8.244444444444444e-06,
|
| 667 |
+
"loss": 1.9412,
|
| 668 |
+
"step": 1880
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.8444444444444444,
|
| 672 |
+
"grad_norm": 1.3711830377578735,
|
| 673 |
+
"learning_rate": 7.8e-06,
|
| 674 |
+
"loss": 1.9652,
|
| 675 |
+
"step": 1900
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.8533333333333334,
|
| 679 |
+
"grad_norm": 1.6069121360778809,
|
| 680 |
+
"learning_rate": 7.3555555555555555e-06,
|
| 681 |
+
"loss": 2.1216,
|
| 682 |
+
"step": 1920
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.8622222222222222,
|
| 686 |
+
"grad_norm": 2.2424278259277344,
|
| 687 |
+
"learning_rate": 6.911111111111111e-06,
|
| 688 |
+
"loss": 1.9446,
|
| 689 |
+
"step": 1940
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.8711111111111111,
|
| 693 |
+
"grad_norm": 1.6427901983261108,
|
| 694 |
+
"learning_rate": 6.466666666666667e-06,
|
| 695 |
+
"loss": 1.9955,
|
| 696 |
+
"step": 1960
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.88,
|
| 700 |
+
"grad_norm": 2.817962646484375,
|
| 701 |
+
"learning_rate": 6.0222222222222225e-06,
|
| 702 |
+
"loss": 1.9938,
|
| 703 |
+
"step": 1980
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.8888888888888888,
|
| 707 |
+
"grad_norm": 2.454206705093384,
|
| 708 |
+
"learning_rate": 5.577777777777778e-06,
|
| 709 |
+
"loss": 1.9823,
|
| 710 |
+
"step": 2000
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.8977777777777778,
|
| 714 |
+
"grad_norm": 1.576393961906433,
|
| 715 |
+
"learning_rate": 5.133333333333334e-06,
|
| 716 |
+
"loss": 1.9354,
|
| 717 |
+
"step": 2020
|
| 718 |
+
},
|
| 719 |
+
{
|
| 720 |
+
"epoch": 0.9066666666666666,
|
| 721 |
+
"grad_norm": 2.2149758338928223,
|
| 722 |
+
"learning_rate": 4.6888888888888895e-06,
|
| 723 |
+
"loss": 1.9377,
|
| 724 |
+
"step": 2040
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 0.9155555555555556,
|
| 728 |
+
"grad_norm": 9.432432174682617,
|
| 729 |
+
"learning_rate": 4.244444444444444e-06,
|
| 730 |
+
"loss": 1.967,
|
| 731 |
+
"step": 2060
|
| 732 |
+
},
|
| 733 |
+
{
|
| 734 |
+
"epoch": 0.9244444444444444,
|
| 735 |
+
"grad_norm": 1.6415536403656006,
|
| 736 |
+
"learning_rate": 3.8e-06,
|
| 737 |
+
"loss": 2.0043,
|
| 738 |
+
"step": 2080
|
| 739 |
+
},
|
| 740 |
+
{
|
| 741 |
+
"epoch": 0.9333333333333333,
|
| 742 |
+
"grad_norm": 1.3895797729492188,
|
| 743 |
+
"learning_rate": 3.3555555555555557e-06,
|
| 744 |
+
"loss": 1.9757,
|
| 745 |
+
"step": 2100
|
| 746 |
+
},
|
| 747 |
+
{
|
| 748 |
+
"epoch": 0.9422222222222222,
|
| 749 |
+
"grad_norm": 1.7504130601882935,
|
| 750 |
+
"learning_rate": 2.9111111111111114e-06,
|
| 751 |
+
"loss": 1.9506,
|
| 752 |
+
"step": 2120
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.9511111111111111,
|
| 756 |
+
"grad_norm": 2.290969133377075,
|
| 757 |
+
"learning_rate": 2.4666666666666666e-06,
|
| 758 |
+
"loss": 1.9525,
|
| 759 |
+
"step": 2140
|
| 760 |
+
},
|
| 761 |
+
{
|
| 762 |
+
"epoch": 0.96,
|
| 763 |
+
"grad_norm": 1.4756875038146973,
|
| 764 |
+
"learning_rate": 2.0222222222222223e-06,
|
| 765 |
+
"loss": 2.0076,
|
| 766 |
+
"step": 2160
|
| 767 |
+
},
|
| 768 |
+
{
|
| 769 |
+
"epoch": 0.9688888888888889,
|
| 770 |
+
"grad_norm": 9.54588794708252,
|
| 771 |
+
"learning_rate": 1.577777777777778e-06,
|
| 772 |
+
"loss": 1.9536,
|
| 773 |
+
"step": 2180
|
| 774 |
+
},
|
| 775 |
+
{
|
| 776 |
+
"epoch": 0.9777777777777777,
|
| 777 |
+
"grad_norm": 7.211913108825684,
|
| 778 |
+
"learning_rate": 1.1333333333333334e-06,
|
| 779 |
+
"loss": 1.9539,
|
| 780 |
+
"step": 2200
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 0.9866666666666667,
|
| 784 |
+
"grad_norm": 2.0712270736694336,
|
| 785 |
+
"learning_rate": 6.888888888888889e-07,
|
| 786 |
+
"loss": 2.0026,
|
| 787 |
+
"step": 2220
|
| 788 |
+
},
|
| 789 |
+
{
|
| 790 |
+
"epoch": 0.9955555555555555,
|
| 791 |
+
"grad_norm": 1.6621143817901611,
|
| 792 |
+
"learning_rate": 2.4444444444444445e-07,
|
| 793 |
+
"loss": 1.9709,
|
| 794 |
+
"step": 2240
|
| 795 |
+
}
|
| 796 |
+
],
|
| 797 |
+
"logging_steps": 20,
|
| 798 |
+
"max_steps": 2250,
|
| 799 |
+
"num_input_tokens_seen": 0,
|
| 800 |
+
"num_train_epochs": 1,
|
| 801 |
+
"save_steps": 200,
|
| 802 |
+
"stateful_callbacks": {
|
| 803 |
+
"TrainerControl": {
|
| 804 |
+
"args": {
|
| 805 |
+
"should_epoch_stop": false,
|
| 806 |
+
"should_evaluate": false,
|
| 807 |
+
"should_log": false,
|
| 808 |
+
"should_save": true,
|
| 809 |
+
"should_training_stop": true
|
| 810 |
+
},
|
| 811 |
+
"attributes": {}
|
| 812 |
+
}
|
| 813 |
+
},
|
| 814 |
+
"total_flos": 1393197318144000.0,
|
| 815 |
+
"train_batch_size": 4,
|
| 816 |
+
"trial_name": null,
|
| 817 |
+
"trial_params": null
|
| 818 |
+
}
|
checkpoint-2250/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e80e264647457cd090f5619636a58dde3a50daacc43d528c97b1cd53b5c0a888
|
| 3 |
+
size 5240
|
config.json
ADDED
|
@@ -0,0 +1,69 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_num_labels": 3,
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_final_layer_norm": false,
|
| 6 |
+
"architectures": [
|
| 7 |
+
"BartForCausalLM"
|
| 8 |
+
],
|
| 9 |
+
"attention_dropout": 0.0,
|
| 10 |
+
"bos_token_id": 0,
|
| 11 |
+
"classif_dropout": 0.0,
|
| 12 |
+
"classifier_dropout": 0.0,
|
| 13 |
+
"d_model": 1024,
|
| 14 |
+
"decoder_attention_heads": 16,
|
| 15 |
+
"decoder_ffn_dim": 4096,
|
| 16 |
+
"decoder_layerdrop": 0.0,
|
| 17 |
+
"decoder_layers": 12,
|
| 18 |
+
"decoder_start_token_id": 2,
|
| 19 |
+
"dropout": 0.1,
|
| 20 |
+
"early_stopping": null,
|
| 21 |
+
"encoder_attention_heads": 16,
|
| 22 |
+
"encoder_ffn_dim": 4096,
|
| 23 |
+
"encoder_layerdrop": 0.0,
|
| 24 |
+
"encoder_layers": 12,
|
| 25 |
+
"eos_token_id": 2,
|
| 26 |
+
"force_bos_token_to_be_generated": true,
|
| 27 |
+
"forced_eos_token_id": 2,
|
| 28 |
+
"gradient_checkpointing": false,
|
| 29 |
+
"id2label": {
|
| 30 |
+
"0": "LABEL_0",
|
| 31 |
+
"1": "LABEL_1",
|
| 32 |
+
"2": "LABEL_2"
|
| 33 |
+
},
|
| 34 |
+
"init_std": 0.02,
|
| 35 |
+
"is_decoder": true,
|
| 36 |
+
"is_encoder_decoder": false,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"LABEL_0": 0,
|
| 39 |
+
"LABEL_1": 1,
|
| 40 |
+
"LABEL_2": 2
|
| 41 |
+
},
|
| 42 |
+
"length_penalty": null,
|
| 43 |
+
"max_length": null,
|
| 44 |
+
"max_position_embeddings": 1024,
|
| 45 |
+
"min_length": null,
|
| 46 |
+
"model_type": "bart",
|
| 47 |
+
"no_repeat_ngram_size": null,
|
| 48 |
+
"normalize_before": false,
|
| 49 |
+
"num_beams": null,
|
| 50 |
+
"num_hidden_layers": 12,
|
| 51 |
+
"output_past": true,
|
| 52 |
+
"pad_token_id": 1,
|
| 53 |
+
"prefix": " ",
|
| 54 |
+
"scale_embedding": false,
|
| 55 |
+
"task_specific_params": {
|
| 56 |
+
"summarization": {
|
| 57 |
+
"early_stopping": true,
|
| 58 |
+
"length_penalty": 2.0,
|
| 59 |
+
"max_length": 142,
|
| 60 |
+
"min_length": 56,
|
| 61 |
+
"no_repeat_ngram_size": 3,
|
| 62 |
+
"num_beams": 4
|
| 63 |
+
}
|
| 64 |
+
},
|
| 65 |
+
"torch_dtype": "float32",
|
| 66 |
+
"transformers_version": "4.52.4",
|
| 67 |
+
"use_cache": true,
|
| 68 |
+
"vocab_size": 50264
|
| 69 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"decoder_start_token_id": 2,
|
| 5 |
+
"early_stopping": true,
|
| 6 |
+
"eos_token_id": 2,
|
| 7 |
+
"forced_bos_token_id": 0,
|
| 8 |
+
"forced_eos_token_id": 2,
|
| 9 |
+
"length_penalty": 2.0,
|
| 10 |
+
"max_length": 142,
|
| 11 |
+
"min_length": 56,
|
| 12 |
+
"no_repeat_ngram_size": 3,
|
| 13 |
+
"num_beams": 4,
|
| 14 |
+
"pad_token_id": 1,
|
| 15 |
+
"transformers_version": "4.52.4"
|
| 16 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:74d34ed7efb5a94ca559d7d28c72f5c61aff49639cb20fdf3774e3fae7b355d7
|
| 3 |
+
size 1016369648
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<s>",
|
| 3 |
+
"cls_token": "<s>",
|
| 4 |
+
"eos_token": "</s>",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "<mask>",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "</s>",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "BartTokenizer",
|
| 56 |
+
"trim_offsets": true,
|
| 57 |
+
"unk_token": "<unk>"
|
| 58 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|