Upload 5 files
Browse filesUpload partial model files
- config.json +75 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +1376 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/gscratch/xlab/hallisky/rewriting/src/models/nontoxic/bart-base_2.5e-06_0_48_jigsaw_randmask/checkpoint-95000",
|
| 3 |
+
"activation_dropout": 0.1,
|
| 4 |
+
"activation_function": "gelu",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"BartForConditionalGeneration"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.1,
|
| 11 |
+
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.1,
|
| 13 |
+
"classifier_dropout": 0.0,
|
| 14 |
+
"d_model": 768,
|
| 15 |
+
"decoder_attention_heads": 12,
|
| 16 |
+
"decoder_ffn_dim": 3072,
|
| 17 |
+
"decoder_layerdrop": 0.0,
|
| 18 |
+
"decoder_layers": 6,
|
| 19 |
+
"decoder_start_token_id": 2,
|
| 20 |
+
"dropout": 0.1,
|
| 21 |
+
"early_stopping": true,
|
| 22 |
+
"encoder_attention_heads": 12,
|
| 23 |
+
"encoder_ffn_dim": 3072,
|
| 24 |
+
"encoder_layerdrop": 0.0,
|
| 25 |
+
"encoder_layers": 6,
|
| 26 |
+
"eos_token_id": 2,
|
| 27 |
+
"forced_bos_token_id": 0,
|
| 28 |
+
"forced_eos_token_id": 2,
|
| 29 |
+
"gradient_checkpointing": false,
|
| 30 |
+
"id2label": {
|
| 31 |
+
"0": "LABEL_0",
|
| 32 |
+
"1": "LABEL_1",
|
| 33 |
+
"2": "LABEL_2"
|
| 34 |
+
},
|
| 35 |
+
"init_std": 0.02,
|
| 36 |
+
"is_encoder_decoder": true,
|
| 37 |
+
"label2id": {
|
| 38 |
+
"LABEL_0": 0,
|
| 39 |
+
"LABEL_1": 1,
|
| 40 |
+
"LABEL_2": 2
|
| 41 |
+
},
|
| 42 |
+
"max_position_embeddings": 1024,
|
| 43 |
+
"model_type": "bart",
|
| 44 |
+
"no_repeat_ngram_size": 3,
|
| 45 |
+
"normalize_before": false,
|
| 46 |
+
"normalize_embedding": true,
|
| 47 |
+
"num_beams": 4,
|
| 48 |
+
"num_hidden_layers": 6,
|
| 49 |
+
"pad_token_id": 1,
|
| 50 |
+
"scale_embedding": false,
|
| 51 |
+
"task_specific_params": {
|
| 52 |
+
"summarization": {
|
| 53 |
+
"length_penalty": 1.0,
|
| 54 |
+
"max_length": 128,
|
| 55 |
+
"min_length": 12,
|
| 56 |
+
"num_beams": 4
|
| 57 |
+
},
|
| 58 |
+
"summarization_cnn": {
|
| 59 |
+
"length_penalty": 2.0,
|
| 60 |
+
"max_length": 142,
|
| 61 |
+
"min_length": 56,
|
| 62 |
+
"num_beams": 4
|
| 63 |
+
},
|
| 64 |
+
"summarization_xsum": {
|
| 65 |
+
"length_penalty": 1.0,
|
| 66 |
+
"max_length": 62,
|
| 67 |
+
"min_length": 11,
|
| 68 |
+
"num_beams": 6
|
| 69 |
+
}
|
| 70 |
+
},
|
| 71 |
+
"torch_dtype": "float32",
|
| 72 |
+
"transformers_version": "4.10.2",
|
| 73 |
+
"use_cache": true,
|
| 74 |
+
"vocab_size": 50265
|
| 75 |
+
}
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:faa6f1c17ac4350906b5dcc9bd17f9ec880460244cf77d70417e212269f94faa
|
| 3 |
+
size 14593
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:116ee016bb58cc25bba9aacaba64d974d8daa5a7c23d50442a9692c902e5f00d
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1376 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8833928108215332,
|
| 3 |
+
"best_model_checkpoint": "/gscratch/xlab/hallisky/rewriting/src/models/nontoxic/bart-base_2.5e-06_0_48_jigsaw_randmask/checkpoint-95000",
|
| 4 |
+
"epoch": 4.742033383915023,
|
| 5 |
+
"global_step": 100000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 2.4875000000000003e-06,
|
| 13 |
+
"loss": 0.9804,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.05,
|
| 18 |
+
"learning_rate": 2.475e-06,
|
| 19 |
+
"loss": 0.9795,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.07,
|
| 24 |
+
"learning_rate": 2.4625e-06,
|
| 25 |
+
"loss": 0.98,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.09,
|
| 30 |
+
"learning_rate": 2.4500000000000003e-06,
|
| 31 |
+
"loss": 0.9784,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.12,
|
| 36 |
+
"learning_rate": 2.4375e-06,
|
| 37 |
+
"loss": 0.973,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.14,
|
| 42 |
+
"learning_rate": 2.425e-06,
|
| 43 |
+
"loss": 0.9785,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.17,
|
| 48 |
+
"learning_rate": 2.4125e-06,
|
| 49 |
+
"loss": 0.9749,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.19,
|
| 54 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 55 |
+
"loss": 0.9747,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.21,
|
| 60 |
+
"learning_rate": 2.3875e-06,
|
| 61 |
+
"loss": 0.9733,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.24,
|
| 66 |
+
"learning_rate": 2.375e-06,
|
| 67 |
+
"loss": 0.97,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.24,
|
| 72 |
+
"eval_loss": 0.8958264589309692,
|
| 73 |
+
"eval_runtime": 1541.2819,
|
| 74 |
+
"eval_samples_per_second": 163.861,
|
| 75 |
+
"eval_steps_per_second": 1.707,
|
| 76 |
+
"step": 5000
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 0.26,
|
| 80 |
+
"learning_rate": 2.3625000000000003e-06,
|
| 81 |
+
"loss": 0.9801,
|
| 82 |
+
"step": 5500
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 0.28,
|
| 86 |
+
"learning_rate": 2.35e-06,
|
| 87 |
+
"loss": 0.9836,
|
| 88 |
+
"step": 6000
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.31,
|
| 92 |
+
"learning_rate": 2.3375000000000005e-06,
|
| 93 |
+
"loss": 0.9813,
|
| 94 |
+
"step": 6500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.33,
|
| 98 |
+
"learning_rate": 2.325e-06,
|
| 99 |
+
"loss": 0.9834,
|
| 100 |
+
"step": 7000
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.36,
|
| 104 |
+
"learning_rate": 2.3125000000000003e-06,
|
| 105 |
+
"loss": 0.9814,
|
| 106 |
+
"step": 7500
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.38,
|
| 110 |
+
"learning_rate": 2.3000000000000004e-06,
|
| 111 |
+
"loss": 0.9809,
|
| 112 |
+
"step": 8000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.4,
|
| 116 |
+
"learning_rate": 2.2875e-06,
|
| 117 |
+
"loss": 0.9777,
|
| 118 |
+
"step": 8500
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.43,
|
| 122 |
+
"learning_rate": 2.2750000000000002e-06,
|
| 123 |
+
"loss": 0.9786,
|
| 124 |
+
"step": 9000
|
| 125 |
+
},
|
| 126 |
+
{
|
| 127 |
+
"epoch": 0.45,
|
| 128 |
+
"learning_rate": 2.2625000000000004e-06,
|
| 129 |
+
"loss": 0.979,
|
| 130 |
+
"step": 9500
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.47,
|
| 134 |
+
"learning_rate": 2.25e-06,
|
| 135 |
+
"loss": 0.9776,
|
| 136 |
+
"step": 10000
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.47,
|
| 140 |
+
"eval_loss": 0.8941081762313843,
|
| 141 |
+
"eval_runtime": 1537.7822,
|
| 142 |
+
"eval_samples_per_second": 164.234,
|
| 143 |
+
"eval_steps_per_second": 1.711,
|
| 144 |
+
"step": 10000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.5,
|
| 148 |
+
"learning_rate": 2.2375e-06,
|
| 149 |
+
"loss": 0.9787,
|
| 150 |
+
"step": 10500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.52,
|
| 154 |
+
"learning_rate": 2.2250000000000003e-06,
|
| 155 |
+
"loss": 0.9791,
|
| 156 |
+
"step": 11000
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.55,
|
| 160 |
+
"learning_rate": 2.2125e-06,
|
| 161 |
+
"loss": 0.9793,
|
| 162 |
+
"step": 11500
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.57,
|
| 166 |
+
"learning_rate": 2.2e-06,
|
| 167 |
+
"loss": 0.9782,
|
| 168 |
+
"step": 12000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.59,
|
| 172 |
+
"learning_rate": 2.1875000000000002e-06,
|
| 173 |
+
"loss": 0.9753,
|
| 174 |
+
"step": 12500
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.62,
|
| 178 |
+
"learning_rate": 2.1750000000000004e-06,
|
| 179 |
+
"loss": 0.98,
|
| 180 |
+
"step": 13000
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 0.64,
|
| 184 |
+
"learning_rate": 2.1625e-06,
|
| 185 |
+
"loss": 0.9804,
|
| 186 |
+
"step": 13500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.66,
|
| 190 |
+
"learning_rate": 2.15e-06,
|
| 191 |
+
"loss": 0.9785,
|
| 192 |
+
"step": 14000
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.69,
|
| 196 |
+
"learning_rate": 2.1375000000000003e-06,
|
| 197 |
+
"loss": 0.9771,
|
| 198 |
+
"step": 14500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.71,
|
| 202 |
+
"learning_rate": 2.125e-06,
|
| 203 |
+
"loss": 0.9746,
|
| 204 |
+
"step": 15000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.71,
|
| 208 |
+
"eval_loss": 0.8944218158721924,
|
| 209 |
+
"eval_runtime": 1538.3111,
|
| 210 |
+
"eval_samples_per_second": 164.177,
|
| 211 |
+
"eval_steps_per_second": 1.71,
|
| 212 |
+
"step": 15000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.74,
|
| 216 |
+
"learning_rate": 2.1125e-06,
|
| 217 |
+
"loss": 0.9802,
|
| 218 |
+
"step": 15500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.76,
|
| 222 |
+
"learning_rate": 2.1000000000000002e-06,
|
| 223 |
+
"loss": 0.9711,
|
| 224 |
+
"step": 16000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.78,
|
| 228 |
+
"learning_rate": 2.0875e-06,
|
| 229 |
+
"loss": 0.9759,
|
| 230 |
+
"step": 16500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.81,
|
| 234 |
+
"learning_rate": 2.075e-06,
|
| 235 |
+
"loss": 0.9761,
|
| 236 |
+
"step": 17000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.83,
|
| 240 |
+
"learning_rate": 2.0625e-06,
|
| 241 |
+
"loss": 0.9745,
|
| 242 |
+
"step": 17500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.85,
|
| 246 |
+
"learning_rate": 2.05e-06,
|
| 247 |
+
"loss": 0.9752,
|
| 248 |
+
"step": 18000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.88,
|
| 252 |
+
"learning_rate": 2.0375e-06,
|
| 253 |
+
"loss": 0.9787,
|
| 254 |
+
"step": 18500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.9,
|
| 258 |
+
"learning_rate": 2.025e-06,
|
| 259 |
+
"loss": 0.9777,
|
| 260 |
+
"step": 19000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.92,
|
| 264 |
+
"learning_rate": 2.0125000000000002e-06,
|
| 265 |
+
"loss": 0.975,
|
| 266 |
+
"step": 19500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.95,
|
| 270 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 271 |
+
"loss": 0.9794,
|
| 272 |
+
"step": 20000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.95,
|
| 276 |
+
"eval_loss": 0.8913277983665466,
|
| 277 |
+
"eval_runtime": 1541.5464,
|
| 278 |
+
"eval_samples_per_second": 163.833,
|
| 279 |
+
"eval_steps_per_second": 1.707,
|
| 280 |
+
"step": 20000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 0.97,
|
| 284 |
+
"learning_rate": 1.9875000000000005e-06,
|
| 285 |
+
"loss": 0.973,
|
| 286 |
+
"step": 20500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.0,
|
| 290 |
+
"learning_rate": 1.975e-06,
|
| 291 |
+
"loss": 0.9733,
|
| 292 |
+
"step": 21000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.02,
|
| 296 |
+
"learning_rate": 1.9625000000000003e-06,
|
| 297 |
+
"loss": 0.974,
|
| 298 |
+
"step": 21500
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.04,
|
| 302 |
+
"learning_rate": 1.9500000000000004e-06,
|
| 303 |
+
"loss": 0.9747,
|
| 304 |
+
"step": 22000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.07,
|
| 308 |
+
"learning_rate": 1.9375e-06,
|
| 309 |
+
"loss": 0.976,
|
| 310 |
+
"step": 22500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.09,
|
| 314 |
+
"learning_rate": 1.925e-06,
|
| 315 |
+
"loss": 0.9723,
|
| 316 |
+
"step": 23000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.11,
|
| 320 |
+
"learning_rate": 1.9125000000000003e-06,
|
| 321 |
+
"loss": 0.971,
|
| 322 |
+
"step": 23500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.14,
|
| 326 |
+
"learning_rate": 1.9000000000000002e-06,
|
| 327 |
+
"loss": 0.9753,
|
| 328 |
+
"step": 24000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.16,
|
| 332 |
+
"learning_rate": 1.8875000000000001e-06,
|
| 333 |
+
"loss": 0.972,
|
| 334 |
+
"step": 24500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.19,
|
| 338 |
+
"learning_rate": 1.8750000000000003e-06,
|
| 339 |
+
"loss": 0.9762,
|
| 340 |
+
"step": 25000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.19,
|
| 344 |
+
"eval_loss": 0.8904610872268677,
|
| 345 |
+
"eval_runtime": 1541.2728,
|
| 346 |
+
"eval_samples_per_second": 163.862,
|
| 347 |
+
"eval_steps_per_second": 1.707,
|
| 348 |
+
"step": 25000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 1.21,
|
| 352 |
+
"learning_rate": 1.8625000000000002e-06,
|
| 353 |
+
"loss": 0.9687,
|
| 354 |
+
"step": 25500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 1.23,
|
| 358 |
+
"learning_rate": 1.85e-06,
|
| 359 |
+
"loss": 0.9702,
|
| 360 |
+
"step": 26000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.26,
|
| 364 |
+
"learning_rate": 1.8375000000000002e-06,
|
| 365 |
+
"loss": 0.9728,
|
| 366 |
+
"step": 26500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 1.28,
|
| 370 |
+
"learning_rate": 1.825e-06,
|
| 371 |
+
"loss": 0.9764,
|
| 372 |
+
"step": 27000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.3,
|
| 376 |
+
"learning_rate": 1.8125e-06,
|
| 377 |
+
"loss": 0.9696,
|
| 378 |
+
"step": 27500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 1.33,
|
| 382 |
+
"learning_rate": 1.8000000000000001e-06,
|
| 383 |
+
"loss": 0.9704,
|
| 384 |
+
"step": 28000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 1.35,
|
| 388 |
+
"learning_rate": 1.7875e-06,
|
| 389 |
+
"loss": 0.9706,
|
| 390 |
+
"step": 28500
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 1.38,
|
| 394 |
+
"learning_rate": 1.7750000000000002e-06,
|
| 395 |
+
"loss": 0.9715,
|
| 396 |
+
"step": 29000
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.4,
|
| 400 |
+
"learning_rate": 1.7625e-06,
|
| 401 |
+
"loss": 0.9721,
|
| 402 |
+
"step": 29500
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 1.42,
|
| 406 |
+
"learning_rate": 1.75e-06,
|
| 407 |
+
"loss": 0.9735,
|
| 408 |
+
"step": 30000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 1.42,
|
| 412 |
+
"eval_loss": 0.8899921178817749,
|
| 413 |
+
"eval_runtime": 1541.1232,
|
| 414 |
+
"eval_samples_per_second": 163.878,
|
| 415 |
+
"eval_steps_per_second": 1.707,
|
| 416 |
+
"step": 30000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 1.45,
|
| 420 |
+
"learning_rate": 1.7375e-06,
|
| 421 |
+
"loss": 0.9677,
|
| 422 |
+
"step": 30500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 1.47,
|
| 426 |
+
"learning_rate": 1.725e-06,
|
| 427 |
+
"loss": 0.9704,
|
| 428 |
+
"step": 31000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 1.49,
|
| 432 |
+
"learning_rate": 1.7125000000000003e-06,
|
| 433 |
+
"loss": 0.9744,
|
| 434 |
+
"step": 31500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.52,
|
| 438 |
+
"learning_rate": 1.7000000000000002e-06,
|
| 439 |
+
"loss": 0.9711,
|
| 440 |
+
"step": 32000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 1.54,
|
| 444 |
+
"learning_rate": 1.6875000000000001e-06,
|
| 445 |
+
"loss": 0.9695,
|
| 446 |
+
"step": 32500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 1.56,
|
| 450 |
+
"learning_rate": 1.6750000000000003e-06,
|
| 451 |
+
"loss": 0.9678,
|
| 452 |
+
"step": 33000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 1.59,
|
| 456 |
+
"learning_rate": 1.6625000000000002e-06,
|
| 457 |
+
"loss": 0.9701,
|
| 458 |
+
"step": 33500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 1.61,
|
| 462 |
+
"learning_rate": 1.6500000000000003e-06,
|
| 463 |
+
"loss": 0.9668,
|
| 464 |
+
"step": 34000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.64,
|
| 468 |
+
"learning_rate": 1.6375000000000002e-06,
|
| 469 |
+
"loss": 0.972,
|
| 470 |
+
"step": 34500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 1.66,
|
| 474 |
+
"learning_rate": 1.6250000000000001e-06,
|
| 475 |
+
"loss": 0.9681,
|
| 476 |
+
"step": 35000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 1.66,
|
| 480 |
+
"eval_loss": 0.8896079063415527,
|
| 481 |
+
"eval_runtime": 1541.3892,
|
| 482 |
+
"eval_samples_per_second": 163.85,
|
| 483 |
+
"eval_steps_per_second": 1.707,
|
| 484 |
+
"step": 35000
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 1.68,
|
| 488 |
+
"learning_rate": 1.6125000000000002e-06,
|
| 489 |
+
"loss": 0.9705,
|
| 490 |
+
"step": 35500
|
| 491 |
+
},
|
| 492 |
+
{
|
| 493 |
+
"epoch": 1.71,
|
| 494 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 495 |
+
"loss": 0.9696,
|
| 496 |
+
"step": 36000
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 1.73,
|
| 500 |
+
"learning_rate": 1.5875e-06,
|
| 501 |
+
"loss": 0.9714,
|
| 502 |
+
"step": 36500
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 1.75,
|
| 506 |
+
"learning_rate": 1.5750000000000002e-06,
|
| 507 |
+
"loss": 0.9658,
|
| 508 |
+
"step": 37000
|
| 509 |
+
},
|
| 510 |
+
{
|
| 511 |
+
"epoch": 1.78,
|
| 512 |
+
"learning_rate": 1.5625e-06,
|
| 513 |
+
"loss": 0.9692,
|
| 514 |
+
"step": 37500
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.8,
|
| 518 |
+
"learning_rate": 1.5500000000000002e-06,
|
| 519 |
+
"loss": 0.9726,
|
| 520 |
+
"step": 38000
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.83,
|
| 524 |
+
"learning_rate": 1.5375e-06,
|
| 525 |
+
"loss": 0.9695,
|
| 526 |
+
"step": 38500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.85,
|
| 530 |
+
"learning_rate": 1.525e-06,
|
| 531 |
+
"loss": 0.9677,
|
| 532 |
+
"step": 39000
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 1.87,
|
| 536 |
+
"learning_rate": 1.5125000000000001e-06,
|
| 537 |
+
"loss": 0.9694,
|
| 538 |
+
"step": 39500
|
| 539 |
+
},
|
| 540 |
+
{
|
| 541 |
+
"epoch": 1.9,
|
| 542 |
+
"learning_rate": 1.5e-06,
|
| 543 |
+
"loss": 0.9665,
|
| 544 |
+
"step": 40000
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 1.9,
|
| 548 |
+
"eval_loss": 0.8887319564819336,
|
| 549 |
+
"eval_runtime": 1541.0388,
|
| 550 |
+
"eval_samples_per_second": 163.887,
|
| 551 |
+
"eval_steps_per_second": 1.707,
|
| 552 |
+
"step": 40000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 1.92,
|
| 556 |
+
"learning_rate": 1.4875000000000002e-06,
|
| 557 |
+
"loss": 0.9686,
|
| 558 |
+
"step": 40500
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 1.94,
|
| 562 |
+
"learning_rate": 1.475e-06,
|
| 563 |
+
"loss": 0.9721,
|
| 564 |
+
"step": 41000
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.97,
|
| 568 |
+
"learning_rate": 1.4625e-06,
|
| 569 |
+
"loss": 0.9696,
|
| 570 |
+
"step": 41500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 1.99,
|
| 574 |
+
"learning_rate": 1.45e-06,
|
| 575 |
+
"loss": 0.9684,
|
| 576 |
+
"step": 42000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.02,
|
| 580 |
+
"learning_rate": 1.4375e-06,
|
| 581 |
+
"loss": 0.969,
|
| 582 |
+
"step": 42500
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.04,
|
| 586 |
+
"learning_rate": 1.425e-06,
|
| 587 |
+
"loss": 0.9688,
|
| 588 |
+
"step": 43000
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 2.06,
|
| 592 |
+
"learning_rate": 1.4125e-06,
|
| 593 |
+
"loss": 0.9655,
|
| 594 |
+
"step": 43500
|
| 595 |
+
},
|
| 596 |
+
{
|
| 597 |
+
"epoch": 2.09,
|
| 598 |
+
"learning_rate": 1.4000000000000001e-06,
|
| 599 |
+
"loss": 0.9683,
|
| 600 |
+
"step": 44000
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 2.11,
|
| 604 |
+
"learning_rate": 1.3875000000000003e-06,
|
| 605 |
+
"loss": 0.9692,
|
| 606 |
+
"step": 44500
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.13,
|
| 610 |
+
"learning_rate": 1.3750000000000002e-06,
|
| 611 |
+
"loss": 0.9683,
|
| 612 |
+
"step": 45000
|
| 613 |
+
},
|
| 614 |
+
{
|
| 615 |
+
"epoch": 2.13,
|
| 616 |
+
"eval_loss": 0.8875691294670105,
|
| 617 |
+
"eval_runtime": 1540.806,
|
| 618 |
+
"eval_samples_per_second": 163.912,
|
| 619 |
+
"eval_steps_per_second": 1.708,
|
| 620 |
+
"step": 45000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 2.16,
|
| 624 |
+
"learning_rate": 1.3625000000000003e-06,
|
| 625 |
+
"loss": 0.9683,
|
| 626 |
+
"step": 45500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 2.18,
|
| 630 |
+
"learning_rate": 1.3500000000000002e-06,
|
| 631 |
+
"loss": 0.965,
|
| 632 |
+
"step": 46000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.21,
|
| 636 |
+
"learning_rate": 1.3375000000000001e-06,
|
| 637 |
+
"loss": 0.9647,
|
| 638 |
+
"step": 46500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 2.23,
|
| 642 |
+
"learning_rate": 1.3250000000000002e-06,
|
| 643 |
+
"loss": 0.9674,
|
| 644 |
+
"step": 47000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 2.25,
|
| 648 |
+
"learning_rate": 1.3125000000000001e-06,
|
| 649 |
+
"loss": 0.9666,
|
| 650 |
+
"step": 47500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 2.28,
|
| 654 |
+
"learning_rate": 1.3e-06,
|
| 655 |
+
"loss": 0.965,
|
| 656 |
+
"step": 48000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.3,
|
| 660 |
+
"learning_rate": 1.2875000000000002e-06,
|
| 661 |
+
"loss": 0.9685,
|
| 662 |
+
"step": 48500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 2.32,
|
| 666 |
+
"learning_rate": 1.275e-06,
|
| 667 |
+
"loss": 0.9702,
|
| 668 |
+
"step": 49000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 2.35,
|
| 672 |
+
"learning_rate": 1.2625000000000002e-06,
|
| 673 |
+
"loss": 0.9664,
|
| 674 |
+
"step": 49500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 2.37,
|
| 678 |
+
"learning_rate": 1.25e-06,
|
| 679 |
+
"loss": 0.964,
|
| 680 |
+
"step": 50000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 2.37,
|
| 684 |
+
"eval_loss": 0.8874109387397766,
|
| 685 |
+
"eval_runtime": 1540.0854,
|
| 686 |
+
"eval_samples_per_second": 163.988,
|
| 687 |
+
"eval_steps_per_second": 1.708,
|
| 688 |
+
"step": 50000
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 2.39,
|
| 692 |
+
"learning_rate": 1.2375e-06,
|
| 693 |
+
"loss": 0.9699,
|
| 694 |
+
"step": 50500
|
| 695 |
+
},
|
| 696 |
+
{
|
| 697 |
+
"epoch": 2.42,
|
| 698 |
+
"learning_rate": 1.2250000000000001e-06,
|
| 699 |
+
"loss": 0.9707,
|
| 700 |
+
"step": 51000
|
| 701 |
+
},
|
| 702 |
+
{
|
| 703 |
+
"epoch": 2.44,
|
| 704 |
+
"learning_rate": 1.2125e-06,
|
| 705 |
+
"loss": 0.9689,
|
| 706 |
+
"step": 51500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.47,
|
| 710 |
+
"learning_rate": 1.2000000000000002e-06,
|
| 711 |
+
"loss": 0.9702,
|
| 712 |
+
"step": 52000
|
| 713 |
+
},
|
| 714 |
+
{
|
| 715 |
+
"epoch": 2.49,
|
| 716 |
+
"learning_rate": 1.1875e-06,
|
| 717 |
+
"loss": 0.9672,
|
| 718 |
+
"step": 52500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 2.51,
|
| 722 |
+
"learning_rate": 1.175e-06,
|
| 723 |
+
"loss": 0.9636,
|
| 724 |
+
"step": 53000
|
| 725 |
+
},
|
| 726 |
+
{
|
| 727 |
+
"epoch": 2.54,
|
| 728 |
+
"learning_rate": 1.1625e-06,
|
| 729 |
+
"loss": 0.967,
|
| 730 |
+
"step": 53500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 2.56,
|
| 734 |
+
"learning_rate": 1.1500000000000002e-06,
|
| 735 |
+
"loss": 0.9655,
|
| 736 |
+
"step": 54000
|
| 737 |
+
},
|
| 738 |
+
{
|
| 739 |
+
"epoch": 2.58,
|
| 740 |
+
"learning_rate": 1.1375000000000001e-06,
|
| 741 |
+
"loss": 0.9647,
|
| 742 |
+
"step": 54500
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 2.61,
|
| 746 |
+
"learning_rate": 1.125e-06,
|
| 747 |
+
"loss": 0.9658,
|
| 748 |
+
"step": 55000
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 2.61,
|
| 752 |
+
"eval_loss": 0.8865290284156799,
|
| 753 |
+
"eval_runtime": 1540.5867,
|
| 754 |
+
"eval_samples_per_second": 163.935,
|
| 755 |
+
"eval_steps_per_second": 1.708,
|
| 756 |
+
"step": 55000
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 2.63,
|
| 760 |
+
"learning_rate": 1.1125000000000001e-06,
|
| 761 |
+
"loss": 0.9661,
|
| 762 |
+
"step": 55500
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 2.66,
|
| 766 |
+
"learning_rate": 1.1e-06,
|
| 767 |
+
"loss": 0.9662,
|
| 768 |
+
"step": 56000
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 2.68,
|
| 772 |
+
"learning_rate": 1.0875000000000002e-06,
|
| 773 |
+
"loss": 0.9652,
|
| 774 |
+
"step": 56500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 2.7,
|
| 778 |
+
"learning_rate": 1.075e-06,
|
| 779 |
+
"loss": 0.9652,
|
| 780 |
+
"step": 57000
|
| 781 |
+
},
|
| 782 |
+
{
|
| 783 |
+
"epoch": 2.73,
|
| 784 |
+
"learning_rate": 1.0625e-06,
|
| 785 |
+
"loss": 0.9644,
|
| 786 |
+
"step": 57500
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 2.75,
|
| 790 |
+
"learning_rate": 1.0500000000000001e-06,
|
| 791 |
+
"loss": 0.967,
|
| 792 |
+
"step": 58000
|
| 793 |
+
},
|
| 794 |
+
{
|
| 795 |
+
"epoch": 2.77,
|
| 796 |
+
"learning_rate": 1.0375e-06,
|
| 797 |
+
"loss": 0.9659,
|
| 798 |
+
"step": 58500
|
| 799 |
+
},
|
| 800 |
+
{
|
| 801 |
+
"epoch": 2.8,
|
| 802 |
+
"learning_rate": 1.025e-06,
|
| 803 |
+
"loss": 0.9665,
|
| 804 |
+
"step": 59000
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 2.82,
|
| 808 |
+
"learning_rate": 1.0125e-06,
|
| 809 |
+
"loss": 0.967,
|
| 810 |
+
"step": 59500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 2.85,
|
| 814 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 815 |
+
"loss": 0.9672,
|
| 816 |
+
"step": 60000
|
| 817 |
+
},
|
| 818 |
+
{
|
| 819 |
+
"epoch": 2.85,
|
| 820 |
+
"eval_loss": 0.8857579827308655,
|
| 821 |
+
"eval_runtime": 1540.558,
|
| 822 |
+
"eval_samples_per_second": 163.938,
|
| 823 |
+
"eval_steps_per_second": 1.708,
|
| 824 |
+
"step": 60000
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.87,
|
| 828 |
+
"learning_rate": 9.875e-07,
|
| 829 |
+
"loss": 0.9652,
|
| 830 |
+
"step": 60500
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 2.89,
|
| 834 |
+
"learning_rate": 9.750000000000002e-07,
|
| 835 |
+
"loss": 0.9662,
|
| 836 |
+
"step": 61000
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 2.92,
|
| 840 |
+
"learning_rate": 9.625e-07,
|
| 841 |
+
"loss": 0.9679,
|
| 842 |
+
"step": 61500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 2.94,
|
| 846 |
+
"learning_rate": 9.500000000000001e-07,
|
| 847 |
+
"loss": 0.961,
|
| 848 |
+
"step": 62000
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.96,
|
| 852 |
+
"learning_rate": 9.375000000000001e-07,
|
| 853 |
+
"loss": 0.9635,
|
| 854 |
+
"step": 62500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 2.99,
|
| 858 |
+
"learning_rate": 9.25e-07,
|
| 859 |
+
"loss": 0.9663,
|
| 860 |
+
"step": 63000
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 3.01,
|
| 864 |
+
"learning_rate": 9.125e-07,
|
| 865 |
+
"loss": 0.9643,
|
| 866 |
+
"step": 63500
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 3.03,
|
| 870 |
+
"learning_rate": 9.000000000000001e-07,
|
| 871 |
+
"loss": 0.964,
|
| 872 |
+
"step": 64000
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 3.06,
|
| 876 |
+
"learning_rate": 8.875000000000001e-07,
|
| 877 |
+
"loss": 0.9653,
|
| 878 |
+
"step": 64500
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 3.08,
|
| 882 |
+
"learning_rate": 8.75e-07,
|
| 883 |
+
"loss": 0.9612,
|
| 884 |
+
"step": 65000
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 3.08,
|
| 888 |
+
"eval_loss": 0.8849959969520569,
|
| 889 |
+
"eval_runtime": 1533.6138,
|
| 890 |
+
"eval_samples_per_second": 164.68,
|
| 891 |
+
"eval_steps_per_second": 1.716,
|
| 892 |
+
"step": 65000
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 3.11,
|
| 896 |
+
"learning_rate": 8.625e-07,
|
| 897 |
+
"loss": 0.9634,
|
| 898 |
+
"step": 65500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 3.13,
|
| 902 |
+
"learning_rate": 8.500000000000001e-07,
|
| 903 |
+
"loss": 0.9611,
|
| 904 |
+
"step": 66000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 3.15,
|
| 908 |
+
"learning_rate": 8.375000000000001e-07,
|
| 909 |
+
"loss": 0.9646,
|
| 910 |
+
"step": 66500
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 3.18,
|
| 914 |
+
"learning_rate": 8.250000000000001e-07,
|
| 915 |
+
"loss": 0.965,
|
| 916 |
+
"step": 67000
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 3.2,
|
| 920 |
+
"learning_rate": 8.125000000000001e-07,
|
| 921 |
+
"loss": 0.9639,
|
| 922 |
+
"step": 67500
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 3.22,
|
| 926 |
+
"learning_rate": 8.000000000000001e-07,
|
| 927 |
+
"loss": 0.9649,
|
| 928 |
+
"step": 68000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 3.25,
|
| 932 |
+
"learning_rate": 7.875000000000001e-07,
|
| 933 |
+
"loss": 0.9636,
|
| 934 |
+
"step": 68500
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 3.27,
|
| 938 |
+
"learning_rate": 7.750000000000001e-07,
|
| 939 |
+
"loss": 0.9657,
|
| 940 |
+
"step": 69000
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 3.3,
|
| 944 |
+
"learning_rate": 7.625e-07,
|
| 945 |
+
"loss": 0.9696,
|
| 946 |
+
"step": 69500
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 3.32,
|
| 950 |
+
"learning_rate": 7.5e-07,
|
| 951 |
+
"loss": 0.9635,
|
| 952 |
+
"step": 70000
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 3.32,
|
| 956 |
+
"eval_loss": 0.8842886686325073,
|
| 957 |
+
"eval_runtime": 1540.0999,
|
| 958 |
+
"eval_samples_per_second": 163.987,
|
| 959 |
+
"eval_steps_per_second": 1.708,
|
| 960 |
+
"step": 70000
|
| 961 |
+
},
|
| 962 |
+
{
|
| 963 |
+
"epoch": 3.34,
|
| 964 |
+
"learning_rate": 7.375e-07,
|
| 965 |
+
"loss": 0.9658,
|
| 966 |
+
"step": 70500
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 3.37,
|
| 970 |
+
"learning_rate": 7.25e-07,
|
| 971 |
+
"loss": 0.9666,
|
| 972 |
+
"step": 71000
|
| 973 |
+
},
|
| 974 |
+
{
|
| 975 |
+
"epoch": 3.39,
|
| 976 |
+
"learning_rate": 7.125e-07,
|
| 977 |
+
"loss": 0.9625,
|
| 978 |
+
"step": 71500
|
| 979 |
+
},
|
| 980 |
+
{
|
| 981 |
+
"epoch": 3.41,
|
| 982 |
+
"learning_rate": 7.000000000000001e-07,
|
| 983 |
+
"loss": 0.964,
|
| 984 |
+
"step": 72000
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 3.44,
|
| 988 |
+
"learning_rate": 6.875000000000001e-07,
|
| 989 |
+
"loss": 0.9612,
|
| 990 |
+
"step": 72500
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 3.46,
|
| 994 |
+
"learning_rate": 6.750000000000001e-07,
|
| 995 |
+
"loss": 0.9647,
|
| 996 |
+
"step": 73000
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 3.49,
|
| 1000 |
+
"learning_rate": 6.625000000000001e-07,
|
| 1001 |
+
"loss": 0.9677,
|
| 1002 |
+
"step": 73500
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 3.51,
|
| 1006 |
+
"learning_rate": 6.5e-07,
|
| 1007 |
+
"loss": 0.9636,
|
| 1008 |
+
"step": 74000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 3.53,
|
| 1012 |
+
"learning_rate": 6.375e-07,
|
| 1013 |
+
"loss": 0.9667,
|
| 1014 |
+
"step": 74500
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 3.56,
|
| 1018 |
+
"learning_rate": 6.25e-07,
|
| 1019 |
+
"loss": 0.9648,
|
| 1020 |
+
"step": 75000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 3.56,
|
| 1024 |
+
"eval_loss": 0.8849769234657288,
|
| 1025 |
+
"eval_runtime": 1539.7241,
|
| 1026 |
+
"eval_samples_per_second": 164.027,
|
| 1027 |
+
"eval_steps_per_second": 1.709,
|
| 1028 |
+
"step": 75000
|
| 1029 |
+
},
|
| 1030 |
+
{
|
| 1031 |
+
"epoch": 3.58,
|
| 1032 |
+
"learning_rate": 6.125000000000001e-07,
|
| 1033 |
+
"loss": 0.9631,
|
| 1034 |
+
"step": 75500
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 3.6,
|
| 1038 |
+
"learning_rate": 6.000000000000001e-07,
|
| 1039 |
+
"loss": 0.968,
|
| 1040 |
+
"step": 76000
|
| 1041 |
+
},
|
| 1042 |
+
{
|
| 1043 |
+
"epoch": 3.63,
|
| 1044 |
+
"learning_rate": 5.875e-07,
|
| 1045 |
+
"loss": 0.9626,
|
| 1046 |
+
"step": 76500
|
| 1047 |
+
},
|
| 1048 |
+
{
|
| 1049 |
+
"epoch": 3.65,
|
| 1050 |
+
"learning_rate": 5.750000000000001e-07,
|
| 1051 |
+
"loss": 0.9656,
|
| 1052 |
+
"step": 77000
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 3.68,
|
| 1056 |
+
"learning_rate": 5.625e-07,
|
| 1057 |
+
"loss": 0.964,
|
| 1058 |
+
"step": 77500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 3.7,
|
| 1062 |
+
"learning_rate": 5.5e-07,
|
| 1063 |
+
"loss": 0.9668,
|
| 1064 |
+
"step": 78000
|
| 1065 |
+
},
|
| 1066 |
+
{
|
| 1067 |
+
"epoch": 3.72,
|
| 1068 |
+
"learning_rate": 5.375e-07,
|
| 1069 |
+
"loss": 0.96,
|
| 1070 |
+
"step": 78500
|
| 1071 |
+
},
|
| 1072 |
+
{
|
| 1073 |
+
"epoch": 3.75,
|
| 1074 |
+
"learning_rate": 5.250000000000001e-07,
|
| 1075 |
+
"loss": 0.9662,
|
| 1076 |
+
"step": 79000
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.77,
|
| 1080 |
+
"learning_rate": 5.125e-07,
|
| 1081 |
+
"loss": 0.9636,
|
| 1082 |
+
"step": 79500
|
| 1083 |
+
},
|
| 1084 |
+
{
|
| 1085 |
+
"epoch": 3.79,
|
| 1086 |
+
"learning_rate": 5.000000000000001e-07,
|
| 1087 |
+
"loss": 0.9614,
|
| 1088 |
+
"step": 80000
|
| 1089 |
+
},
|
| 1090 |
+
{
|
| 1091 |
+
"epoch": 3.79,
|
| 1092 |
+
"eval_loss": 0.8838475942611694,
|
| 1093 |
+
"eval_runtime": 1537.6849,
|
| 1094 |
+
"eval_samples_per_second": 164.244,
|
| 1095 |
+
"eval_steps_per_second": 1.711,
|
| 1096 |
+
"step": 80000
|
| 1097 |
+
},
|
| 1098 |
+
{
|
| 1099 |
+
"epoch": 3.82,
|
| 1100 |
+
"learning_rate": 4.875000000000001e-07,
|
| 1101 |
+
"loss": 0.9617,
|
| 1102 |
+
"step": 80500
|
| 1103 |
+
},
|
| 1104 |
+
{
|
| 1105 |
+
"epoch": 3.84,
|
| 1106 |
+
"learning_rate": 4.7500000000000006e-07,
|
| 1107 |
+
"loss": 0.9607,
|
| 1108 |
+
"step": 81000
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 3.86,
|
| 1112 |
+
"learning_rate": 4.625e-07,
|
| 1113 |
+
"loss": 0.9638,
|
| 1114 |
+
"step": 81500
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 3.89,
|
| 1118 |
+
"learning_rate": 4.5000000000000003e-07,
|
| 1119 |
+
"loss": 0.9628,
|
| 1120 |
+
"step": 82000
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 3.91,
|
| 1124 |
+
"learning_rate": 4.375e-07,
|
| 1125 |
+
"loss": 0.9653,
|
| 1126 |
+
"step": 82500
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 3.94,
|
| 1130 |
+
"learning_rate": 4.2500000000000006e-07,
|
| 1131 |
+
"loss": 0.9629,
|
| 1132 |
+
"step": 83000
|
| 1133 |
+
},
|
| 1134 |
+
{
|
| 1135 |
+
"epoch": 3.96,
|
| 1136 |
+
"learning_rate": 4.125000000000001e-07,
|
| 1137 |
+
"loss": 0.9628,
|
| 1138 |
+
"step": 83500
|
| 1139 |
+
},
|
| 1140 |
+
{
|
| 1141 |
+
"epoch": 3.98,
|
| 1142 |
+
"learning_rate": 4.0000000000000003e-07,
|
| 1143 |
+
"loss": 0.9671,
|
| 1144 |
+
"step": 84000
|
| 1145 |
+
},
|
| 1146 |
+
{
|
| 1147 |
+
"epoch": 4.01,
|
| 1148 |
+
"learning_rate": 3.8750000000000005e-07,
|
| 1149 |
+
"loss": 0.9621,
|
| 1150 |
+
"step": 84500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 4.03,
|
| 1154 |
+
"learning_rate": 3.75e-07,
|
| 1155 |
+
"loss": 0.9647,
|
| 1156 |
+
"step": 85000
|
| 1157 |
+
},
|
| 1158 |
+
{
|
| 1159 |
+
"epoch": 4.03,
|
| 1160 |
+
"eval_loss": 0.8842937350273132,
|
| 1161 |
+
"eval_runtime": 1538.1328,
|
| 1162 |
+
"eval_samples_per_second": 164.196,
|
| 1163 |
+
"eval_steps_per_second": 1.711,
|
| 1164 |
+
"step": 85000
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 4.05,
|
| 1168 |
+
"learning_rate": 3.625e-07,
|
| 1169 |
+
"loss": 0.9611,
|
| 1170 |
+
"step": 85500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 4.08,
|
| 1174 |
+
"learning_rate": 3.5000000000000004e-07,
|
| 1175 |
+
"loss": 0.9651,
|
| 1176 |
+
"step": 86000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 4.1,
|
| 1180 |
+
"learning_rate": 3.3750000000000005e-07,
|
| 1181 |
+
"loss": 0.9619,
|
| 1182 |
+
"step": 86500
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 4.13,
|
| 1186 |
+
"learning_rate": 3.25e-07,
|
| 1187 |
+
"loss": 0.9617,
|
| 1188 |
+
"step": 87000
|
| 1189 |
+
},
|
| 1190 |
+
{
|
| 1191 |
+
"epoch": 4.15,
|
| 1192 |
+
"learning_rate": 3.125e-07,
|
| 1193 |
+
"loss": 0.9609,
|
| 1194 |
+
"step": 87500
|
| 1195 |
+
},
|
| 1196 |
+
{
|
| 1197 |
+
"epoch": 4.17,
|
| 1198 |
+
"learning_rate": 3.0000000000000004e-07,
|
| 1199 |
+
"loss": 0.9646,
|
| 1200 |
+
"step": 88000
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 4.2,
|
| 1204 |
+
"learning_rate": 2.8750000000000005e-07,
|
| 1205 |
+
"loss": 0.9603,
|
| 1206 |
+
"step": 88500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 4.22,
|
| 1210 |
+
"learning_rate": 2.75e-07,
|
| 1211 |
+
"loss": 0.9637,
|
| 1212 |
+
"step": 89000
|
| 1213 |
+
},
|
| 1214 |
+
{
|
| 1215 |
+
"epoch": 4.24,
|
| 1216 |
+
"learning_rate": 2.6250000000000003e-07,
|
| 1217 |
+
"loss": 0.9626,
|
| 1218 |
+
"step": 89500
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 4.27,
|
| 1222 |
+
"learning_rate": 2.5000000000000004e-07,
|
| 1223 |
+
"loss": 0.9612,
|
| 1224 |
+
"step": 90000
|
| 1225 |
+
},
|
| 1226 |
+
{
|
| 1227 |
+
"epoch": 4.27,
|
| 1228 |
+
"eval_loss": 0.8842394351959229,
|
| 1229 |
+
"eval_runtime": 1537.8664,
|
| 1230 |
+
"eval_samples_per_second": 164.225,
|
| 1231 |
+
"eval_steps_per_second": 1.711,
|
| 1232 |
+
"step": 90000
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 4.29,
|
| 1236 |
+
"learning_rate": 2.3750000000000003e-07,
|
| 1237 |
+
"loss": 0.9641,
|
| 1238 |
+
"step": 90500
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 4.32,
|
| 1242 |
+
"learning_rate": 2.2500000000000002e-07,
|
| 1243 |
+
"loss": 0.9657,
|
| 1244 |
+
"step": 91000
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 4.34,
|
| 1248 |
+
"learning_rate": 2.1250000000000003e-07,
|
| 1249 |
+
"loss": 0.9643,
|
| 1250 |
+
"step": 91500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 4.36,
|
| 1254 |
+
"learning_rate": 2.0000000000000002e-07,
|
| 1255 |
+
"loss": 0.9611,
|
| 1256 |
+
"step": 92000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 4.39,
|
| 1260 |
+
"learning_rate": 1.875e-07,
|
| 1261 |
+
"loss": 0.9652,
|
| 1262 |
+
"step": 92500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 4.41,
|
| 1266 |
+
"learning_rate": 1.7500000000000002e-07,
|
| 1267 |
+
"loss": 0.9624,
|
| 1268 |
+
"step": 93000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 4.43,
|
| 1272 |
+
"learning_rate": 1.625e-07,
|
| 1273 |
+
"loss": 0.9645,
|
| 1274 |
+
"step": 93500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 4.46,
|
| 1278 |
+
"learning_rate": 1.5000000000000002e-07,
|
| 1279 |
+
"loss": 0.9679,
|
| 1280 |
+
"step": 94000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 4.48,
|
| 1284 |
+
"learning_rate": 1.375e-07,
|
| 1285 |
+
"loss": 0.9616,
|
| 1286 |
+
"step": 94500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 4.5,
|
| 1290 |
+
"learning_rate": 1.2500000000000002e-07,
|
| 1291 |
+
"loss": 0.9654,
|
| 1292 |
+
"step": 95000
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 4.5,
|
| 1296 |
+
"eval_loss": 0.8833928108215332,
|
| 1297 |
+
"eval_runtime": 1537.0843,
|
| 1298 |
+
"eval_samples_per_second": 164.308,
|
| 1299 |
+
"eval_steps_per_second": 1.712,
|
| 1300 |
+
"step": 95000
|
| 1301 |
+
},
|
| 1302 |
+
{
|
| 1303 |
+
"epoch": 4.53,
|
| 1304 |
+
"learning_rate": 1.1250000000000001e-07,
|
| 1305 |
+
"loss": 0.9615,
|
| 1306 |
+
"step": 95500
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 4.55,
|
| 1310 |
+
"learning_rate": 1.0000000000000001e-07,
|
| 1311 |
+
"loss": 0.966,
|
| 1312 |
+
"step": 96000
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 4.58,
|
| 1316 |
+
"learning_rate": 8.750000000000001e-08,
|
| 1317 |
+
"loss": 0.9694,
|
| 1318 |
+
"step": 96500
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 4.6,
|
| 1322 |
+
"learning_rate": 7.500000000000001e-08,
|
| 1323 |
+
"loss": 0.9639,
|
| 1324 |
+
"step": 97000
|
| 1325 |
+
},
|
| 1326 |
+
{
|
| 1327 |
+
"epoch": 4.62,
|
| 1328 |
+
"learning_rate": 6.250000000000001e-08,
|
| 1329 |
+
"loss": 0.9604,
|
| 1330 |
+
"step": 97500
|
| 1331 |
+
},
|
| 1332 |
+
{
|
| 1333 |
+
"epoch": 4.65,
|
| 1334 |
+
"learning_rate": 5.0000000000000004e-08,
|
| 1335 |
+
"loss": 0.9574,
|
| 1336 |
+
"step": 98000
|
| 1337 |
+
},
|
| 1338 |
+
{
|
| 1339 |
+
"epoch": 4.67,
|
| 1340 |
+
"learning_rate": 3.7500000000000005e-08,
|
| 1341 |
+
"loss": 0.9623,
|
| 1342 |
+
"step": 98500
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 4.69,
|
| 1346 |
+
"learning_rate": 2.5000000000000002e-08,
|
| 1347 |
+
"loss": 0.9616,
|
| 1348 |
+
"step": 99000
|
| 1349 |
+
},
|
| 1350 |
+
{
|
| 1351 |
+
"epoch": 4.72,
|
| 1352 |
+
"learning_rate": 1.2500000000000001e-08,
|
| 1353 |
+
"loss": 0.9659,
|
| 1354 |
+
"step": 99500
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 4.74,
|
| 1358 |
+
"learning_rate": 0.0,
|
| 1359 |
+
"loss": 0.9678,
|
| 1360 |
+
"step": 100000
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 4.74,
|
| 1364 |
+
"eval_loss": 0.884353518486023,
|
| 1365 |
+
"eval_runtime": 1538.109,
|
| 1366 |
+
"eval_samples_per_second": 164.199,
|
| 1367 |
+
"eval_steps_per_second": 1.711,
|
| 1368 |
+
"step": 100000
|
| 1369 |
+
}
|
| 1370 |
+
],
|
| 1371 |
+
"max_steps": 100000,
|
| 1372 |
+
"num_train_epochs": 5,
|
| 1373 |
+
"total_flos": 5.71620307894272e+17,
|
| 1374 |
+
"trial_name": null,
|
| 1375 |
+
"trial_params": null
|
| 1376 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:af714a8f158243a5da8c56ea0425568a27f42081265d5415c2d500c28c0315f4
|
| 3 |
+
size 2735
|