upload checkpoint

.gitattributes

@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+training_args.bin filter=lfs diff=lfs merge=lfs -text
+optimizer.pt filter=lfs diff=lfs merge=lfs -text
+scheduler.pt filter=lfs diff=lfs merge=lfs -text
+49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58 filter=lfs diff=lfs merge=lfs -text

49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:83ab5c39e2b1c34a955136275ce0db068cb20d9643ead033d6b8124a73ab4f64
+size 15645492

config.json

@@ -0,0 +1,166 @@
+{
+ "_name_or_path": "openai/clip-vit-base-patch16",
+ "architectures": [
+ "CLIPModel"
+ ],
+ "id2label": {},
+ "initializer_factor": 1.0,
+ "label2id": {
+ "Contradiction": 2,
+ "Entailment": 0,
+ "Neutral": 1
+ },
+ "logit_scale_init_value": 2.6592,
+ "model_type": "clip",
+ "projection_dim": 512,
+ "text_config": {
+ "_name_or_path": "",
+ "add_cross_attention": false,
+ "architectures": null,
+ "attention_dropout": 0.0,
+ "bad_words_ids": null,
+ "bos_token_id": 0,
+ "chunk_size_feed_forward": 0,
+ "cross_attention_hidden_size": null,
+ "decoder_start_token_id": null,
+ "diversity_penalty": 0.0,
+ "do_sample": false,
+ "dropout": 0.0,
+ "early_stopping": false,
+ "encoder_no_repeat_ngram_size": 0,
+ "eos_token_id": 2,
+ "finetuning_task": null,
+ "forced_bos_token_id": null,
+ "forced_eos_token_id": null,
+ "hidden_act": "quick_gelu",
+ "hidden_size": 512,
+ "id2label": {
+ "0": "LABEL_0",
+ "1": "LABEL_1"
+ },
+ "initializer_factor": 1.0,
+ "initializer_range": 0.02,
+ "intermediate_size": 2048,
+ "is_decoder": false,
+ "is_encoder_decoder": false,
+ "label2id": {
+ "LABEL_0": 0,
+ "LABEL_1": 1
+ },
+ "layer_norm_eps": 1e-05,
+ "length_penalty": 1.0,
+ "max_length": 20,
+ "max_position_embeddings": 77,
+ "min_length": 0,
+ "model_type": "clip_text_model",
+ "no_repeat_ngram_size": 0,
+ "num_attention_heads": 8,
+ "num_beam_groups": 1,
+ "num_beams": 1,
+ "num_hidden_layers": 12,
+ "num_return_sequences": 1,
+ "output_attentions": false,
+ "output_hidden_states": false,
+ "output_scores": false,
+ "pad_token_id": 1,
+ "prefix": null,
+ "problem_type": null,
+ "pruned_heads": {},
+ "remove_invalid_values": false,
+ "repetition_penalty": 1.0,
+ "return_dict": true,
+ "return_dict_in_generate": false,
+ "sep_token_id": null,
+ "task_specific_params": null,
+ "temperature": 1.0,
+ "tie_encoder_decoder": false,
+ "tie_word_embeddings": true,
+ "tokenizer_class": null,
+ "top_k": 50,
+ "top_p": 1.0,
+ "torch_dtype": null,
+ "torchscript": false,
+ "transformers_version": "4.17.0",
+ "typical_p": 1.0,
+ "use_bfloat16": false,
+ "vocab_size": 49408
+ },
+ "text_config_dict": null,
+ "torch_dtype": "float32",
+ "transformers_version": null,
+ "vision_config": {
+ "_name_or_path": "",
+ "add_cross_attention": false,
+ "architectures": null,
+ "attention_dropout": 0.0,
+ "bad_words_ids": null,
+ "bos_token_id": null,
+ "chunk_size_feed_forward": 0,
+ "cross_attention_hidden_size": null,
+ "decoder_start_token_id": null,
+ "diversity_penalty": 0.0,
+ "do_sample": false,
+ "dropout": 0.0,
+ "early_stopping": false,
+ "encoder_no_repeat_ngram_size": 0,
+ "eos_token_id": null,
+ "finetuning_task": null,
+ "forced_bos_token_id": null,
+ "forced_eos_token_id": null,
+ "hidden_act": "quick_gelu",
+ "hidden_size": 768,
+ "id2label": {
+ "0": "LABEL_0",
+ "1": "LABEL_1"
+ },
+ "image_size": 224,
+ "initializer_factor": 1.0,
+ "initializer_range": 0.02,
+ "intermediate_size": 3072,
+ "is_decoder": false,
+ "is_encoder_decoder": false,
+ "label2id": {
+ "LABEL_0": 0,
+ "LABEL_1": 1
+ },
+ "layer_norm_eps": 1e-05,
+ "length_penalty": 1.0,
+ "max_length": 20,
+ "min_length": 0,
+ "model_type": "clip_vision_model",
+ "no_repeat_ngram_size": 0,
+ "num_attention_heads": 12,
+ "num_beam_groups": 1,
+ "num_beams": 1,
+ "num_hidden_layers": 12,
+ "num_return_sequences": 1,
+ "output_attentions": false,
+ "output_hidden_states": false,
+ "output_scores": false,
+ "pad_token_id": null,
+ "patch_size": 16,
+ "prefix": null,
+ "problem_type": null,
+ "pruned_heads": {},
+ "remove_invalid_values": false,
+ "repetition_penalty": 1.0,
+ "return_dict": true,
+ "return_dict_in_generate": false,
+ "sep_token_id": null,
+ "task_specific_params": null,
+ "temperature": 1.0,
+ "tie_encoder_decoder": false,
+ "tie_word_embeddings": true,
+ "tokenizer_class": null,
+ "top_k": 50,
+ "top_p": 1.0,
+ "torch_dtype": null,
+ "torchscript": false,
+ "transformers_version": "4.17.0",
+ "typical_p": 1.0,
+ "use_bfloat16": false
+ },
+ "vision_config_dict": {
+ "patch_size": 16
+ }
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d70f1bc20b56950e9fd3bf0831441d9055e59a08a61b15588e889bb16b73fde2
+size 689659578

preprocessor_config.json

@@ -0,0 +1,19 @@
+{
+ "crop_size": 224,
+ "do_center_crop": true,
+ "do_normalize": true,
+ "do_resize": true,
+ "feature_extractor_type": "CLIPFeatureExtractor",
+ "image_mean": [
+ 0.48145466,
+ 0.4578275,
+ 0.40821073
+ ],
+ "image_std": [
+ 0.26862954,
+ 0.26130258,
+ 0.27577711
+ ],
+ "resample": 3,
+ "size": 224
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c25bf4ef34bd3b8b204aa20767bfa67eb17be98309238623fea19aa7f694455
+size 598642122

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93b830fe93e6f7710b58e80e3a93850a14d38db75d558fb5feca0a9d01caf620
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8166e01d2b10247de29befc9577caa94ee2add44c473badc8c77fad131eccddd
+size 1064

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": {"content": "<|startoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "eos_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "unk_token": {"content": "<|endoftext|>", "single_word": false, "lstrip": false, "rstrip": false, "normalized": true}, "pad_token": "<|endoftext|>"}

text_renderer_config.json

@@ -0,0 +1,253 @@
+{
+ "background_color": "white",
+ "dpi": 120,
+ "font_color": "black",
+ "font_file": "49e6dc219d1a1a1c9236acaf05a48b542002016a6dc877ee72baab085a84257b.3f28e7f4b38e1efe1b6da4a3732404c19d4c6a614ff32dce90a251e293d4ce58",
+ "font_size": 8,
+ "fonts_list": [
+ "Apple Color Emoji",
+ "DejaVu Sans",
+ "DejaVu Sans Mono",
+ "DejaVu Serif",
+ "FontAwesome",
+ "GLYPHICONS Halflings",
+ "Go Noto Current",
+ "Inconsolata",
+ "Lato",
+ "MathJax_AMS",
+ "MathJax_Caligraphic",
+ "MathJax_Fraktur",
+ "MathJax_Main",
+ "MathJax_Math",
+ "MathJax_SansSerif",
+ "MathJax_Script",
+ "MathJax_Size1",
+ "MathJax_Size2",
+ "MathJax_Size3",
+ "MathJax_Size4",
+ "MathJax_Typewriter",
+ "MathJax_Vector",
+ "MathJax_Vector-Bold",
+ "MathJax_WinChrome",
+ "MathJax_WinIE6",
+ "Monospace",
+ "Noto Fangsong KSS Rotated",
+ "Noto Fangsong KSS Vertical",
+ "Noto Kufi Arabic",
+ "Noto Music",
+ "Noto Naskh Arabic UI",
+ "Noto Nastaliq Urdu",
+ "Noto Rashi Hebrew",
+ "Noto Sans",
+ "Noto Sans Adlam",
+ "Noto Sans Adlam Unjoined",
+ "Noto Sans Anatolian Hieroglyphs",
+ "Noto Sans Arabic",
+ "Noto Sans Armenian",
+ "Noto Sans Avestan",
+ "Noto Sans Balinese",
+ "Noto Sans Bamum",
+ "Noto Sans Bassa Vah",
+ "Noto Sans Batak",
+ "Noto Sans Bhaiksuki",
+ "Noto Sans Brahmi",
+ "Noto Sans Buginese",
+ "Noto Sans Buhid",
+ "Noto Sans CJK HK",
+ "Noto Sans CJK JP",
+ "Noto Sans CJK KR",
+ "Noto Sans CJK SC",
+ "Noto Sans CJK TC",
+ "Noto Sans Canadian Aboriginal",
+ "Noto Sans Carian",
+ "Noto Sans Caucasian Albanian",
+ "Noto Sans Chakma",
+ "Noto Sans Cham",
+ "Noto Sans Cherokee",
+ "Noto Sans Chorasmian",
+ "Noto Sans Coptic",
+ "Noto Sans Cuneiform",
+ "Noto Sans Cypriot",
+ "Noto Sans Cypro Minoan",
+ "Noto Sans Deseret",
+ "Noto Sans Devanagari",
+ "Noto Sans Devanagari UI",
+ "Noto Sans Duployan",
+ "Noto Sans Egyptian Hieroglyphs",
+ "Noto Sans Elbasan",
+ "Noto Sans Elymaic",
+ "Noto Sans Ethiopic",
+ "Noto Sans Georgian",
+ "Noto Sans Glagolitic",
+ "Noto Sans Gothic",
+ "Noto Sans Grantha",
+ "Noto Sans Gujarati",
+ "Noto Sans Gujarati UI",
+ "Noto Sans Gunjala Gondi",
+ "Noto Sans Gurmukhi",
+ "Noto Sans Gurmukhi UI",
+ "Noto Sans Hanifi Rohingya",
+ "Noto Sans Hanunoo",
+ "Noto Sans Hatran",
+ "Noto Sans Hebrew",
+ "Noto Sans Hebrew Droid",
+ "Noto Sans Imperial Aramaic",
+ "Noto Sans Indic Siyaq Numbers",
+ "Noto Sans Inscriptional Pahlavi",
+ "Noto Sans Inscriptional Parthian",
+ "Noto Sans Javanese",
+ "Noto Sans Kaithi",
+ "Noto Sans Kannada",
+ "Noto Sans Kannada UI",
+ "Noto Sans Kawi",
+ "Noto Sans Kayah Li",
+ "Noto Sans Kharoshthi",
+ "Noto Sans Khmer",
+ "Noto Sans Khmer UI",
+ "Noto Sans Khojki",
+ "Noto Sans Khudawadi",
+ "Noto Sans Lao",
+ "Noto Sans Lao Looped",
+ "Noto Sans Lao Looped UI",
+ "Noto Sans Lao UI",
+ "Noto Sans Lepcha",
+ "Noto Sans Limbu",
+ "Noto Sans Linear A",
+ "Noto Sans Linear B",
+ "Noto Sans Lisu",
+ "Noto Sans Lycian",
+ "Noto Sans Lydian",
+ "Noto Sans Mahajani",
+ "Noto Sans Malayalam",
+ "Noto Sans Malayalam UI",
+ "Noto Sans Mandaic",
+ "Noto Sans Manichaean",
+ "Noto Sans Marchen",
+ "Noto Sans Masaram Gondi",
+ "Noto Sans Math",
+ "Noto Sans Mayan Numerals",
+ "Noto Sans Medefaidrin",
+ "Noto Sans Meetei Mayek",
+ "Noto Sans Mende Kikakui",
+ "Noto Sans Meroitic",
+ "Noto Sans Miao",
+ "Noto Sans Modi",
+ "Noto Sans Mongolian",
+ "Noto Sans Mono",
+ "Noto Sans Mro",
+ "Noto Sans Multani",
+ "Noto Sans Myanmar",
+ "Noto Sans Myanmar UI",
+ "Noto Sans NKo",
+ "Noto Sans NKo Unjoined",
+ "Noto Sans Nabataean",
+ "Noto Sans Nag Mundari",
+ "Noto Sans Nandinagari",
+ "Noto Sans New Tai Lue",
+ "Noto Sans Newa",
+ "Noto Sans Nushu",
+ "Noto Sans Ogham",
+ "Noto Sans Ol Chiki",
+ "Noto Sans Old Hungarian",
+ "Noto Sans Old Italic",
+ "Noto Sans Old North Arabian",
+ "Noto Sans Old Permic",
+ "Noto Sans Old Persian",
+ "Noto Sans Old Sogdian",
+ "Noto Sans Old South Arabian",
+ "Noto Sans Old Turkic",
+ "Noto Sans Oriya",
+ "Noto Sans Osage",
+ "Noto Sans Osmanya",
+ "Noto Sans Pahawh Hmong",
+ "Noto Sans Palmyrene",
+ "Noto Sans Pau Cin Hau",
+ "Noto Sans PhagsPa",
+ "Noto Sans Phoenician",
+ "Noto Sans Psalter Pahlavi",
+ "Noto Sans Rejang",
+ "Noto Sans Runic",
+ "Noto Sans Samaritan",
+ "Noto Sans Saurashtra",
+ "Noto Sans Sharada",
+ "Noto Sans Shavian",
+ "Noto Sans Siddham",
+ "Noto Sans SignWriting",
+ "Noto Sans Sinhala",
+ "Noto Sans Sinhala UI",
+ "Noto Sans Sogdian",
+ "Noto Sans Sora Sompeng",
+ "Noto Sans Soyombo",
+ "Noto Sans Sundanese",
+ "Noto Sans Syloti Nagri",
+ "Noto Sans Symbols",
+ "Noto Sans Symbols 2",
+ "Noto Sans Syriac",
+ "Noto Sans Syriac Eastern",
+ "Noto Sans Syriac Western",
+ "Noto Sans Tagalog",
+ "Noto Sans Tagbanwa",
+ "Noto Sans Tai Le",
+ "Noto Sans Tai Tham",
+ "Noto Sans Tai Viet",
+ "Noto Sans Takri",
+ "Noto Sans Tamil",
+ "Noto Sans Tamil Supplement",
+ "Noto Sans Tamil UI",
+ "Noto Sans Tangsa",
+ "Noto Sans Telugu",
+ "Noto Sans Telugu UI",
+ "Noto Sans Test",
+ "Noto Sans Thaana",
+ "Noto Sans Thai",
+ "Noto Sans Thai Looped",
+ "Noto Sans Thai Looped UI",
+ "Noto Sans Thai UI",
+ "Noto Sans Tifinagh",
+ "Noto Sans Tifinagh APT",
+ "Noto Sans Tifinagh Adrar",
+ "Noto Sans Tifinagh Agraw Imazighen",
+ "Noto Sans Tifinagh Ahaggar",
+ "Noto Sans Tifinagh Air",
+ "Noto Sans Tifinagh Azawagh",
+ "Noto Sans Tifinagh Ghat",
+ "Noto Sans Tifinagh Hawad",
+ "Noto Sans Tifinagh Rhissa Ixa",
+ "Noto Sans Tifinagh SIL",
+ "Noto Sans Tifinagh Tawellemmet",
+ "Noto Sans Tirhuta",
+ "Noto Sans Ugaritic",
+ "Noto Sans Vai",
+ "Noto Sans Vithkuqi",
+ "Noto Sans Wancho",
+ "Noto Sans Warang Citi",
+ "Noto Sans Yi",
+ "Noto Sans Zanabazar Square",
+ "Noto Traditional Nushu",
+ "NotoSansOldHungarianUI",
+ "Sans",
+ "Serif",
+ "Source Code Pro",
+ "System-ui",
+ "Ubuntu",
+ "Ubuntu Condensed",
+ "Ubuntu Mono",
+ "cmex10",
+ "cmmi10",
+ "cmr10",
+ "cmsy10",
+ "dsrom10",
+ "esint10",
+ "eufm10",
+ "msam10",
+ "msbm10",
+ "rsfs10",
+ "stmary10",
+ "wasy10"
+ ],
+ "max_seq_length": 196,
+ "pad_size": 3,
+ "pixels_per_patch": 16,
+ "rgb": false,
+ "text_renderer_type": "PangoCairoTextRenderer"
+}

tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+ "unk_token": {
+ "content": "<|endoftext|>",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": true,
+ "__type": "AddedToken"
+ },
+ "bos_token": {
+ "content": "<|startoftext|>",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": true,
+ "__type": "AddedToken"
+ },
+ "eos_token": {
+ "content": "<|endoftext|>",
+ "single_word": false,
+ "lstrip": false,
+ "rstrip": false,
+ "normalized": true,
+ "__type": "AddedToken"
+ },
+ "pad_token": "<|endoftext|>",
+ "errors": "replace",
+ "add_prefix_space": false,
+ "do_lower_case": true,
+ "name_or_path": "openai/clip-vit-base-patch16",
+ "model_max_length": 77,
+ "special_tokens_map_file": "./special_tokens_map.json",
+ "tokenizer_class": "CLIPTokenizer"
+}

trainer_state.json

@@ -0,0 +1,1156 @@
+{
+ "best_metric": 0.7275755258839849,
+ "best_model_checkpoint": "./checkpoints/clip-2560-wikispan-all/checkpoint-240",
+ "epoch": 0.9472765798329802,
+ "global_step": 22800,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.0,
+ "learning_rate": 1.8e-06,
+ "loss": 1.8723,
+ "step": 120
+ },
+ {
+ "epoch": 0.01,
+ "learning_rate": 3.6e-06,
+ "loss": 1.0969,
+ "step": 240
+ },
+ {
+ "epoch": 0.01,
+ "learning_rate": 5.4e-06,
+ "loss": 0.8982,
+ "step": 360
+ },
+ {
+ "epoch": 0.02,
+ "learning_rate": 7.2e-06,
+ "loss": 0.802,
+ "step": 480
+ },
+ {
+ "epoch": 0.02,
+ "learning_rate": 9e-06,
+ "loss": 0.7304,
+ "step": 600
+ },
+ {
+ "epoch": 0.03,
+ "learning_rate": 1.08e-05,
+ "loss": 0.6815,
+ "step": 720
+ },
+ {
+ "epoch": 0.03,
+ "learning_rate": 1.26e-05,
+ "loss": 0.6435,
+ "step": 840
+ },
+ {
+ "epoch": 0.04,
+ "learning_rate": 1.44e-05,
+ "loss": 0.6089,
+ "step": 960
+ },
+ {
+ "epoch": 0.04,
+ "learning_rate": 1.62e-05,
+ "loss": 0.581,
+ "step": 1080
+ },
+ {
+ "epoch": 0.05,
+ "learning_rate": 1.8e-05,
+ "loss": 0.5637,
+ "step": 1200
+ },
+ {
+ "epoch": 0.05,
+ "learning_rate": 1.98e-05,
+ "loss": 0.538,
+ "step": 1320
+ },
+ {
+ "epoch": 0.06,
+ "learning_rate": 2.16e-05,
+ "loss": 0.5223,
+ "step": 1440
+ },
+ {
+ "epoch": 0.06,
+ "learning_rate": 2.3400000000000003e-05,
+ "loss": 0.5123,
+ "step": 1560
+ },
+ {
+ "epoch": 0.07,
+ "learning_rate": 2.52e-05,
+ "loss": 0.4984,
+ "step": 1680
+ },
+ {
+ "epoch": 0.07,
+ "learning_rate": 2.7000000000000002e-05,
+ "loss": 0.4848,
+ "step": 1800
+ },
+ {
+ "epoch": 0.08,
+ "learning_rate": 2.88e-05,
+ "loss": 0.4763,
+ "step": 1920
+ },
+ {
+ "epoch": 0.08,
+ "learning_rate": 2.9945750452079567e-05,
+ "loss": 0.468,
+ "step": 2040
+ },
+ {
+ "epoch": 0.09,
+ "learning_rate": 2.978300180831826e-05,
+ "loss": 0.4532,
+ "step": 2160
+ },
+ {
+ "epoch": 0.09,
+ "learning_rate": 2.9620253164556963e-05,
+ "loss": 0.445,
+ "step": 2280
+ },
+ {
+ "epoch": 0.1,
+ "learning_rate": 2.945750452079566e-05,
+ "loss": 0.4382,
+ "step": 2400
+ },
+ {
+ "epoch": 0.1,
+ "learning_rate": 2.929475587703436e-05,
+ "loss": 0.4289,
+ "step": 2520
+ },
+ {
+ "epoch": 0.11,
+ "learning_rate": 2.9132007233273057e-05,
+ "loss": 0.4216,
+ "step": 2640
+ },
+ {
+ "epoch": 0.11,
+ "learning_rate": 2.8969258589511756e-05,
+ "loss": 0.4134,
+ "step": 2760
+ },
+ {
+ "epoch": 0.12,
+ "learning_rate": 2.8806509945750454e-05,
+ "loss": 0.4084,
+ "step": 2880
+ },
+ {
+ "epoch": 0.12,
+ "learning_rate": 2.864376130198915e-05,
+ "loss": 0.3988,
+ "step": 3000
+ },
+ {
+ "epoch": 0.13,
+ "learning_rate": 2.8481012658227846e-05,
+ "loss": 0.3938,
+ "step": 3120
+ },
+ {
+ "epoch": 0.13,
+ "learning_rate": 2.8318264014466548e-05,
+ "loss": 0.3886,
+ "step": 3240
+ },
+ {
+ "epoch": 0.14,
+ "learning_rate": 2.8155515370705246e-05,
+ "loss": 0.3845,
+ "step": 3360
+ },
+ {
+ "epoch": 0.14,
+ "learning_rate": 2.7992766726943944e-05,
+ "loss": 0.378,
+ "step": 3480
+ },
+ {
+ "epoch": 0.15,
+ "learning_rate": 2.7830018083182642e-05,
+ "loss": 0.375,
+ "step": 3600
+ },
+ {
+ "epoch": 0.15,
+ "learning_rate": 2.766726943942134e-05,
+ "loss": 0.3665,
+ "step": 3720
+ },
+ {
+ "epoch": 0.16,
+ "learning_rate": 2.7504520795660035e-05,
+ "loss": 0.3644,
+ "step": 3840
+ },
+ {
+ "epoch": 0.16,
+ "learning_rate": 2.7341772151898733e-05,
+ "loss": 0.36,
+ "step": 3960
+ },
+ {
+ "epoch": 0.17,
+ "learning_rate": 2.717902350813743e-05,
+ "loss": 0.3597,
+ "step": 4080
+ },
+ {
+ "epoch": 0.17,
+ "learning_rate": 2.701627486437613e-05,
+ "loss": 0.3549,
+ "step": 4200
+ },
+ {
+ "epoch": 0.18,
+ "learning_rate": 2.685352622061483e-05,
+ "loss": 0.3545,
+ "step": 4320
+ },
+ {
+ "epoch": 0.18,
+ "learning_rate": 2.669077757685353e-05,
+ "loss": 0.3496,
+ "step": 4440
+ },
+ {
+ "epoch": 0.19,
+ "learning_rate": 2.6528028933092224e-05,
+ "loss": 0.3442,
+ "step": 4560
+ },
+ {
+ "epoch": 0.19,
+ "learning_rate": 2.6365280289330922e-05,
+ "loss": 0.3424,
+ "step": 4680
+ },
+ {
+ "epoch": 0.2,
+ "learning_rate": 2.620253164556962e-05,
+ "loss": 0.3414,
+ "step": 4800
+ },
+ {
+ "epoch": 0.2,
+ "learning_rate": 2.6039783001808318e-05,
+ "loss": 0.3361,
+ "step": 4920
+ },
+ {
+ "epoch": 0.21,
+ "learning_rate": 2.5877034358047016e-05,
+ "loss": 0.3378,
+ "step": 5040
+ },
+ {
+ "epoch": 0.21,
+ "learning_rate": 2.5714285714285714e-05,
+ "loss": 0.3316,
+ "step": 5160
+ },
+ {
+ "epoch": 0.22,
+ "learning_rate": 2.5551537070524416e-05,
+ "loss": 0.3302,
+ "step": 5280
+ },
+ {
+ "epoch": 0.22,
+ "learning_rate": 2.538878842676311e-05,
+ "loss": 0.3245,
+ "step": 5400
+ },
+ {
+ "epoch": 0.23,
+ "learning_rate": 2.522603978300181e-05,
+ "loss": 0.3254,
+ "step": 5520
+ },
+ {
+ "epoch": 0.23,
+ "learning_rate": 2.5063291139240507e-05,
+ "loss": 0.3206,
+ "step": 5640
+ },
+ {
+ "epoch": 0.24,
+ "learning_rate": 2.4900542495479205e-05,
+ "loss": 0.3212,
+ "step": 5760
+ },
+ {
+ "epoch": 0.24,
+ "learning_rate": 2.4737793851717903e-05,
+ "loss": 0.3184,
+ "step": 5880
+ },
+ {
+ "epoch": 0.25,
+ "learning_rate": 2.45750452079566e-05,
+ "loss": 0.3141,
+ "step": 6000
+ },
+ {
+ "epoch": 0.25,
+ "learning_rate": 2.44122965641953e-05,
+ "loss": 0.3185,
+ "step": 6120
+ },
+ {
+ "epoch": 0.26,
+ "learning_rate": 2.4249547920433994e-05,
+ "loss": 0.3115,
+ "step": 6240
+ },
+ {
+ "epoch": 0.26,
+ "learning_rate": 2.4086799276672696e-05,
+ "loss": 0.3093,
+ "step": 6360
+ },
+ {
+ "epoch": 0.27,
+ "learning_rate": 2.3924050632911394e-05,
+ "loss": 0.3116,
+ "step": 6480
+ },
+ {
+ "epoch": 0.27,
+ "learning_rate": 2.3761301989150092e-05,
+ "loss": 0.3084,
+ "step": 6600
+ },
+ {
+ "epoch": 0.28,
+ "learning_rate": 2.359855334538879e-05,
+ "loss": 0.3071,
+ "step": 6720
+ },
+ {
+ "epoch": 0.28,
+ "learning_rate": 2.3435804701627488e-05,
+ "loss": 0.3047,
+ "step": 6840
+ },
+ {
+ "epoch": 0.29,
+ "learning_rate": 2.3273056057866183e-05,
+ "loss": 0.302,
+ "step": 6960
+ },
+ {
+ "epoch": 0.29,
+ "learning_rate": 2.311030741410488e-05,
+ "loss": 0.3027,
+ "step": 7080
+ },
+ {
+ "epoch": 0.3,
+ "learning_rate": 2.294755877034358e-05,
+ "loss": 0.3019,
+ "step": 7200
+ },
+ {
+ "epoch": 0.3,
+ "learning_rate": 2.278481012658228e-05,
+ "loss": 0.3,
+ "step": 7320
+ },
+ {
+ "epoch": 0.31,
+ "learning_rate": 2.262206148282098e-05,
+ "loss": 0.2961,
+ "step": 7440
+ },
+ {
+ "epoch": 0.31,
+ "learning_rate": 2.2459312839059677e-05,
+ "loss": 0.2967,
+ "step": 7560
+ },
+ {
+ "epoch": 0.32,
+ "learning_rate": 2.2296564195298375e-05,
+ "loss": 0.2944,
+ "step": 7680
+ },
+ {
+ "epoch": 0.32,
+ "learning_rate": 2.213381555153707e-05,
+ "loss": 0.2937,
+ "step": 7800
+ },
+ {
+ "epoch": 0.33,
+ "learning_rate": 2.1971066907775768e-05,
+ "loss": 0.2914,
+ "step": 7920
+ },
+ {
+ "epoch": 0.33,
+ "learning_rate": 2.1808318264014466e-05,
+ "loss": 0.2902,
+ "step": 8040
+ },
+ {
+ "epoch": 0.34,
+ "learning_rate": 2.1645569620253164e-05,
+ "loss": 0.2889,
+ "step": 8160
+ },
+ {
+ "epoch": 0.34,
+ "learning_rate": 2.1482820976491862e-05,
+ "loss": 0.2894,
+ "step": 8280
+ },
+ {
+ "epoch": 0.35,
+ "learning_rate": 2.1320072332730564e-05,
+ "loss": 0.286,
+ "step": 8400
+ },
+ {
+ "epoch": 0.35,
+ "learning_rate": 2.1157323688969262e-05,
+ "loss": 0.2847,
+ "step": 8520
+ },
+ {
+ "epoch": 0.36,
+ "learning_rate": 2.0994575045207956e-05,
+ "loss": 0.2838,
+ "step": 8640
+ },
+ {
+ "epoch": 0.36,
+ "learning_rate": 2.0831826401446655e-05,
+ "loss": 0.2822,
+ "step": 8760
+ },
+ {
+ "epoch": 0.37,
+ "learning_rate": 2.0669077757685353e-05,
+ "loss": 0.2814,
+ "step": 8880
+ },
+ {
+ "epoch": 0.37,
+ "learning_rate": 2.050632911392405e-05,
+ "loss": 0.2818,
+ "step": 9000
+ },
+ {
+ "epoch": 0.38,
+ "learning_rate": 2.034358047016275e-05,
+ "loss": 0.2779,
+ "step": 9120
+ },
+ {
+ "epoch": 0.38,
+ "learning_rate": 2.0180831826401447e-05,
+ "loss": 0.2794,
+ "step": 9240
+ },
+ {
+ "epoch": 0.39,
+ "learning_rate": 2.0018083182640145e-05,
+ "loss": 0.2789,
+ "step": 9360
+ },
+ {
+ "epoch": 0.39,
+ "learning_rate": 1.9855334538878843e-05,
+ "loss": 0.2757,
+ "step": 9480
+ },
+ {
+ "epoch": 0.4,
+ "learning_rate": 1.969258589511754e-05,
+ "loss": 0.2758,
+ "step": 9600
+ },
+ {
+ "epoch": 0.4,
+ "learning_rate": 1.952983725135624e-05,
+ "loss": 0.277,
+ "step": 9720
+ },
+ {
+ "epoch": 0.41,
+ "learning_rate": 1.9367088607594938e-05,
+ "loss": 0.2731,
+ "step": 9840
+ },
+ {
+ "epoch": 0.41,
+ "learning_rate": 1.9204339963833636e-05,
+ "loss": 0.268,
+ "step": 9960
+ },
+ {
+ "epoch": 0.42,
+ "learning_rate": 1.9041591320072334e-05,
+ "loss": 0.2702,
+ "step": 10080
+ },
+ {
+ "epoch": 0.42,
+ "learning_rate": 1.887884267631103e-05,
+ "loss": 0.2699,
+ "step": 10200
+ },
+ {
+ "epoch": 0.43,
+ "learning_rate": 1.8716094032549727e-05,
+ "loss": 0.2707,
+ "step": 10320
+ },
+ {
+ "epoch": 0.43,
+ "learning_rate": 1.8553345388788428e-05,
+ "loss": 0.2661,
+ "step": 10440
+ },
+ {
+ "epoch": 0.44,
+ "learning_rate": 1.8390596745027126e-05,
+ "loss": 0.2668,
+ "step": 10560
+ },
+ {
+ "epoch": 0.44,
+ "learning_rate": 1.8227848101265824e-05,
+ "loss": 0.2689,
+ "step": 10680
+ },
+ {
+ "epoch": 0.45,
+ "learning_rate": 1.8065099457504523e-05,
+ "loss": 0.2695,
+ "step": 10800
+ },
+ {
+ "epoch": 0.45,
+ "learning_rate": 1.790235081374322e-05,
+ "loss": 0.2637,
+ "step": 10920
+ },
+ {
+ "epoch": 0.46,
+ "learning_rate": 1.7739602169981915e-05,
+ "loss": 0.2645,
+ "step": 11040
+ },
+ {
+ "epoch": 0.46,
+ "learning_rate": 1.7576853526220614e-05,
+ "loss": 0.2618,
+ "step": 11160
+ },
+ {
+ "epoch": 0.47,
+ "learning_rate": 1.741410488245931e-05,
+ "loss": 0.2626,
+ "step": 11280
+ },
+ {
+ "epoch": 0.47,
+ "learning_rate": 1.7251356238698013e-05,
+ "loss": 0.2581,
+ "step": 11400
+ },
+ {
+ "epoch": 0.48,
+ "learning_rate": 1.708860759493671e-05,
+ "loss": 0.2597,
+ "step": 11520
+ },
+ {
+ "epoch": 0.48,
+ "learning_rate": 1.692585895117541e-05,
+ "loss": 0.2594,
+ "step": 11640
+ },
+ {
+ "epoch": 0.49,
+ "learning_rate": 1.6763110307414104e-05,
+ "loss": 0.2577,
+ "step": 11760
+ },
+ {
+ "epoch": 0.49,
+ "learning_rate": 1.6600361663652802e-05,
+ "loss": 0.2585,
+ "step": 11880
+ },
+ {
+ "epoch": 0.5,
+ "learning_rate": 1.64376130198915e-05,
+ "loss": 0.2565,
+ "step": 12000
+ },
+ {
+ "epoch": 0.5,
+ "learning_rate": 1.62748643761302e-05,
+ "loss": 0.2562,
+ "step": 12120
+ },
+ {
+ "epoch": 0.51,
+ "learning_rate": 1.6112115732368897e-05,
+ "loss": 0.2559,
+ "step": 12240
+ },
+ {
+ "epoch": 0.51,
+ "learning_rate": 1.5949367088607595e-05,
+ "loss": 0.2551,
+ "step": 12360
+ },
+ {
+ "epoch": 0.52,
+ "learning_rate": 1.5786618444846296e-05,
+ "loss": 0.2528,
+ "step": 12480
+ },
+ {
+ "epoch": 0.52,
+ "learning_rate": 1.562386980108499e-05,
+ "loss": 0.2539,
+ "step": 12600
+ },
+ {
+ "epoch": 0.53,
+ "learning_rate": 1.546112115732369e-05,
+ "loss": 0.2525,
+ "step": 12720
+ },
+ {
+ "epoch": 0.53,
+ "learning_rate": 1.5298372513562387e-05,
+ "loss": 0.2534,
+ "step": 12840
+ },
+ {
+ "epoch": 0.54,
+ "learning_rate": 1.5135623869801085e-05,
+ "loss": 0.2518,
+ "step": 12960
+ },
+ {
+ "epoch": 0.54,
+ "learning_rate": 1.4972875226039783e-05,
+ "loss": 0.2506,
+ "step": 13080
+ },
+ {
+ "epoch": 0.55,
+ "learning_rate": 1.4810126582278482e-05,
+ "loss": 0.2517,
+ "step": 13200
+ },
+ {
+ "epoch": 0.55,
+ "learning_rate": 1.464737793851718e-05,
+ "loss": 0.2483,
+ "step": 13320
+ },
+ {
+ "epoch": 0.56,
+ "learning_rate": 1.4484629294755878e-05,
+ "loss": 0.2471,
+ "step": 13440
+ },
+ {
+ "epoch": 0.56,
+ "learning_rate": 1.4321880650994574e-05,
+ "loss": 0.2484,
+ "step": 13560
+ },
+ {
+ "epoch": 0.57,
+ "learning_rate": 1.4159132007233274e-05,
+ "loss": 0.2474,
+ "step": 13680
+ },
+ {
+ "epoch": 0.57,
+ "learning_rate": 1.3996383363471972e-05,
+ "loss": 0.25,
+ "step": 13800
+ },
+ {
+ "epoch": 0.58,
+ "learning_rate": 1.383363471971067e-05,
+ "loss": 0.2458,
+ "step": 13920
+ },
+ {
+ "epoch": 0.58,
+ "learning_rate": 1.3670886075949367e-05,
+ "loss": 0.2466,
+ "step": 14040
+ },
+ {
+ "epoch": 0.59,
+ "learning_rate": 1.3508137432188065e-05,
+ "loss": 0.2454,
+ "step": 14160
+ },
+ {
+ "epoch": 0.59,
+ "learning_rate": 1.3345388788426765e-05,
+ "loss": 0.2461,
+ "step": 14280
+ },
+ {
+ "epoch": 0.6,
+ "learning_rate": 1.3182640144665461e-05,
+ "loss": 0.2427,
+ "step": 14400
+ },
+ {
+ "epoch": 0.6,
+ "learning_rate": 1.3019891500904159e-05,
+ "loss": 0.2446,
+ "step": 14520
+ },
+ {
+ "epoch": 0.61,
+ "learning_rate": 1.2857142857142857e-05,
+ "loss": 0.2433,
+ "step": 14640
+ },
+ {
+ "epoch": 0.61,
+ "learning_rate": 1.2694394213381555e-05,
+ "loss": 0.241,
+ "step": 14760
+ },
+ {
+ "epoch": 0.62,
+ "learning_rate": 1.2531645569620253e-05,
+ "loss": 0.2424,
+ "step": 14880
+ },
+ {
+ "epoch": 0.62,
+ "learning_rate": 1.2368896925858952e-05,
+ "loss": 0.2427,
+ "step": 15000
+ },
+ {
+ "epoch": 0.63,
+ "learning_rate": 1.220614828209765e-05,
+ "loss": 0.2395,
+ "step": 15120
+ },
+ {
+ "epoch": 0.63,
+ "learning_rate": 1.2043399638336348e-05,
+ "loss": 0.2396,
+ "step": 15240
+ },
+ {
+ "epoch": 0.64,
+ "learning_rate": 1.1880650994575046e-05,
+ "loss": 0.2416,
+ "step": 15360
+ },
+ {
+ "epoch": 0.64,
+ "learning_rate": 1.1717902350813744e-05,
+ "loss": 0.2402,
+ "step": 15480
+ },
+ {
+ "epoch": 0.65,
+ "learning_rate": 1.155515370705244e-05,
+ "loss": 0.2382,
+ "step": 15600
+ },
+ {
+ "epoch": 0.65,
+ "learning_rate": 1.139240506329114e-05,
+ "loss": 0.2361,
+ "step": 15720
+ },
+ {
+ "epoch": 0.66,
+ "learning_rate": 1.1229656419529838e-05,
+ "loss": 0.2383,
+ "step": 15840
+ },
+ {
+ "epoch": 0.66,
+ "learning_rate": 1.1066907775768535e-05,
+ "loss": 0.2368,
+ "step": 15960
+ },
+ {
+ "epoch": 0.67,
+ "learning_rate": 1.0904159132007233e-05,
+ "loss": 0.2374,
+ "step": 16080
+ },
+ {
+ "epoch": 0.67,
+ "learning_rate": 1.0741410488245931e-05,
+ "loss": 0.2363,
+ "step": 16200
+ },
+ {
+ "epoch": 0.68,
+ "learning_rate": 1.0578661844484631e-05,
+ "loss": 0.2364,
+ "step": 16320
+ },
+ {
+ "epoch": 0.68,
+ "learning_rate": 1.0415913200723327e-05,
+ "loss": 0.2326,
+ "step": 16440
+ },
+ {
+ "epoch": 0.69,
+ "learning_rate": 1.0253164556962025e-05,
+ "loss": 0.232,
+ "step": 16560
+ },
+ {
+ "epoch": 0.69,
+ "learning_rate": 1.0090415913200724e-05,
+ "loss": 0.2317,
+ "step": 16680
+ },
+ {
+ "epoch": 0.7,
+ "learning_rate": 9.927667269439422e-06,
+ "loss": 0.2316,
+ "step": 16800
+ },
+ {
+ "epoch": 0.7,
+ "learning_rate": 9.76491862567812e-06,
+ "loss": 0.2343,
+ "step": 16920
+ },
+ {
+ "epoch": 0.71,
+ "learning_rate": 9.602169981916818e-06,
+ "loss": 0.2304,
+ "step": 17040
+ },
+ {
+ "epoch": 0.71,
+ "learning_rate": 9.439421338155514e-06,
+ "loss": 0.2297,
+ "step": 17160
+ },
+ {
+ "epoch": 0.72,
+ "learning_rate": 9.276672694394214e-06,
+ "loss": 0.231,
+ "step": 17280
+ },
+ {
+ "epoch": 0.72,
+ "learning_rate": 9.113924050632912e-06,
+ "loss": 0.2316,
+ "step": 17400
+ },
+ {
+ "epoch": 0.73,
+ "learning_rate": 8.95117540687161e-06,
+ "loss": 0.2314,
+ "step": 17520
+ },
+ {
+ "epoch": 0.73,
+ "learning_rate": 8.788426763110307e-06,
+ "loss": 0.2289,
+ "step": 17640
+ },
+ {
+ "epoch": 0.74,
+ "learning_rate": 8.625678119349007e-06,
+ "loss": 0.2275,
+ "step": 17760
+ },
+ {
+ "epoch": 0.74,
+ "learning_rate": 8.462929475587705e-06,
+ "loss": 0.2282,
+ "step": 17880
+ },
+ {
+ "epoch": 0.75,
+ "learning_rate": 8.300180831826401e-06,
+ "loss": 0.2285,
+ "step": 18000
+ },
+ {
+ "epoch": 0.75,
+ "learning_rate": 8.1374321880651e-06,
+ "loss": 0.2284,
+ "step": 18120
+ },
+ {
+ "epoch": 0.76,
+ "learning_rate": 7.974683544303797e-06,
+ "loss": 0.2274,
+ "step": 18240
+ },
+ {
+ "epoch": 0.76,
+ "learning_rate": 7.811934900542495e-06,
+ "loss": 0.2234,
+ "step": 18360
+ },
+ {
+ "epoch": 0.77,
+ "learning_rate": 7.649186256781194e-06,
+ "loss": 0.229,
+ "step": 18480
+ },
+ {
+ "epoch": 0.77,
+ "learning_rate": 7.486437613019892e-06,
+ "loss": 0.2248,
+ "step": 18600
+ },
+ {
+ "epoch": 0.78,
+ "learning_rate": 7.32368896925859e-06,
+ "loss": 0.2232,
+ "step": 18720
+ },
+ {
+ "epoch": 0.78,
+ "learning_rate": 7.160940325497287e-06,
+ "loss": 0.2234,
+ "step": 18840
+ },
+ {
+ "epoch": 0.79,
+ "learning_rate": 6.998191681735986e-06,
+ "loss": 0.2258,
+ "step": 18960
+ },
+ {
+ "epoch": 0.79,
+ "learning_rate": 6.835443037974683e-06,
+ "loss": 0.2238,
+ "step": 19080
+ },
+ {
+ "epoch": 0.8,
+ "learning_rate": 6.672694394213382e-06,
+ "loss": 0.2217,
+ "step": 19200
+ },
+ {
+ "epoch": 0.8,
+ "learning_rate": 6.5099457504520796e-06,
+ "loss": 0.2222,
+ "step": 19320
+ },
+ {
+ "epoch": 0.81,
+ "learning_rate": 6.347197106690778e-06,
+ "loss": 0.2256,
+ "step": 19440
+ },
+ {
+ "epoch": 0.81,
+ "learning_rate": 6.184448462929476e-06,
+ "loss": 0.2221,
+ "step": 19560
+ },
+ {
+ "epoch": 0.82,
+ "learning_rate": 6.021699819168174e-06,
+ "loss": 0.2243,
+ "step": 19680
+ },
+ {
+ "epoch": 0.82,
+ "learning_rate": 5.858951175406872e-06,
+ "loss": 0.2227,
+ "step": 19800
+ },
+ {
+ "epoch": 0.83,
+ "learning_rate": 5.69620253164557e-06,
+ "loss": 0.222,
+ "step": 19920
+ },
+ {
+ "epoch": 0.83,
+ "learning_rate": 5.533453887884267e-06,
+ "loss": 0.2205,
+ "step": 20040
+ },
+ {
+ "epoch": 0.84,
+ "learning_rate": 5.3707052441229655e-06,
+ "loss": 0.22,
+ "step": 20160
+ },
+ {
+ "epoch": 0.84,
+ "learning_rate": 5.207956600361664e-06,
+ "loss": 0.2223,
+ "step": 20280
+ },
+ {
+ "epoch": 0.85,
+ "learning_rate": 5.045207956600362e-06,
+ "loss": 0.2209,
+ "step": 20400
+ },
+ {
+ "epoch": 0.85,
+ "learning_rate": 4.88245931283906e-06,
+ "loss": 0.2207,
+ "step": 20520
+ },
+ {
+ "epoch": 0.86,
+ "learning_rate": 4.719710669077757e-06,
+ "loss": 0.2206,
+ "step": 20640
+ },
+ {
+ "epoch": 0.86,
+ "learning_rate": 4.556962025316456e-06,
+ "loss": 0.2204,
+ "step": 20760
+ },
+ {
+ "epoch": 0.87,
+ "learning_rate": 4.394213381555153e-06,
+ "loss": 0.2221,
+ "step": 20880
+ },
+ {
+ "epoch": 0.87,
+ "learning_rate": 4.231464737793852e-06,
+ "loss": 0.2197,
+ "step": 21000
+ },
+ {
+ "epoch": 0.88,
+ "learning_rate": 4.06871609403255e-06,
+ "loss": 0.2188,
+ "step": 21120
+ },
+ {
+ "epoch": 0.88,
+ "learning_rate": 3.905967450271248e-06,
+ "loss": 0.2202,
+ "step": 21240
+ },
+ {
+ "epoch": 0.89,
+ "learning_rate": 3.743218806509946e-06,
+ "loss": 0.2176,
+ "step": 21360
+ },
+ {
+ "epoch": 0.89,
+ "learning_rate": 3.5804701627486435e-06,
+ "loss": 0.2185,
+ "step": 21480
+ },
+ {
+ "epoch": 0.9,
+ "learning_rate": 3.4177215189873417e-06,
+ "loss": 0.2169,
+ "step": 21600
+ },
+ {
+ "epoch": 0.9,
+ "learning_rate": 3.2549728752260398e-06,
+ "loss": 0.2175,
+ "step": 21720
+ },
+ {
+ "epoch": 0.91,
+ "learning_rate": 3.092224231464738e-06,
+ "loss": 0.2153,
+ "step": 21840
+ },
+ {
+ "epoch": 0.91,
+ "learning_rate": 2.929475587703436e-06,
+ "loss": 0.218,
+ "step": 21960
+ },
+ {
+ "epoch": 0.92,
+ "learning_rate": 2.7667269439421337e-06,
+ "loss": 0.2161,
+ "step": 22080
+ },
+ {
+ "epoch": 0.92,
+ "learning_rate": 2.603978300180832e-06,
+ "loss": 0.2167,
+ "step": 22200
+ },
+ {
+ "epoch": 0.93,
+ "learning_rate": 2.44122965641953e-06,
+ "loss": 0.2166,
+ "step": 22320
+ },
+ {
+ "epoch": 0.93,
+ "learning_rate": 2.278481012658228e-06,
+ "loss": 0.215,
+ "step": 22440
+ },
+ {
+ "epoch": 0.94,
+ "learning_rate": 2.115732368896926e-06,
+ "loss": 0.2145,
+ "step": 22560
+ },
+ {
+ "epoch": 0.94,
+ "learning_rate": 1.952983725135624e-06,
+ "loss": 0.2165,
+ "step": 22680
+ },
+ {
+ "epoch": 0.95,
+ "learning_rate": 1.7902350813743218e-06,
+ "loss": 0.2155,
+ "step": 22800
+ }
+ ],
+ "max_steps": 24120,
+ "num_train_epochs": 2,
+ "total_flos": 0.0,
+ "trial_name": null,
+ "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4b4e9b908ba7cfd25eb41cec551f2d6f2b9d63b77ff4fceb67a9c378ae462e18
+size 3640