CrossEncoder based on jhu-clsp/ettin-encoder-400m

This is a Cross Encoder model finetuned from jhu-clsp/ettin-encoder-400m on the ms_marco dataset using the sentence-transformers library. It computes scores for pairs of texts, which can be used for text reranking and semantic search.

Model Details

Model Description

  • Model Type: Cross Encoder
  • Base model: jhu-clsp/ettin-encoder-400m
  • Maximum Sequence Length: 7999 tokens
  • Number of Output Labels: 1 label
  • Training Dataset:
  • Language: en

Model Sources

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import CrossEncoder

# Download from the 🤗 Hub
model = CrossEncoder("rahulseetharaman/reranker-msmarco-v1.1-ettin-encoder-400m-listnet")
# Get scores for pairs of texts
pairs = [
    ['what is the salary of an animal shelter manager', 'Average Animal Shelter Salaries. The average salary for animal shelter jobs is $50,000. Average animal shelter salaries can vary greatly due to company, location, industry, experience and benefits. This salary was calculated using the average salary for all jobs with the term animal shelter anywhere in the job listing.'],
    ['what is the salary of an animal shelter manager', 'The salary that an animal shelter manager earns can vary based on a variety of factors such as their specific responsibilities, their years of experience, their educational background, and the region in which the position is located. Most animal shelter manager positions do not carry particularly high salaries, but those who follow animal rescue career paths tend to be willing to sacrifice some earning potential for the prospect of being able to help animals in need.'],
    ['what is the salary of an animal shelter manager', 'Animal shelter managers’ salaries vary depending on their state and location. For example, the Midwest sits at the bottom of the list. Based on the May, 2010 Pay Scale survey, Illinois managers made $35,382, while the average Californian animal shelter manager earned $55,224. Size. The size of animal shelter also correlates to the manager’s salary. The more animals, the more employees, the more responsibility for the manager. In May, 2010, animal shelter managers working in a shelter employing 1 to 9 people, earned $33,835 on average.'],
    ['what is the salary of an animal shelter manager', 'Salary by Region. Salaries for animal shelter managers also vary by region. According to the Salary Expert website, the highest compensation for such positions, about $60,000 a year, is in San Francisco. Animal shelter managers in the New York and Atlanta metro areas can expect to earn about $50,000 a year.'],
    ['what is the salary of an animal shelter manager', 'Experience and years of service factor into an animal shelter manager’s salary. In May 2010, managers with 1 to 4 years of experience earned the least amount of money, $36,061. Those with 20 years or more experience, however, make $60,521, according to Pay Scale. Size. The size of animal shelter also correlates to the manager’s salary. The more animals, the more employees, the more responsibility for the manager. In May, 2010, animal shelter managers working in a shelter employing 1 to 9 people, earned $33,835 on average.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# Or rank different texts based on similarity to a single text
ranks = model.rank(
    'what is the salary of an animal shelter manager',
    [
        'Average Animal Shelter Salaries. The average salary for animal shelter jobs is $50,000. Average animal shelter salaries can vary greatly due to company, location, industry, experience and benefits. This salary was calculated using the average salary for all jobs with the term animal shelter anywhere in the job listing.',
        'The salary that an animal shelter manager earns can vary based on a variety of factors such as their specific responsibilities, their years of experience, their educational background, and the region in which the position is located. Most animal shelter manager positions do not carry particularly high salaries, but those who follow animal rescue career paths tend to be willing to sacrifice some earning potential for the prospect of being able to help animals in need.',
        'Animal shelter managers’ salaries vary depending on their state and location. For example, the Midwest sits at the bottom of the list. Based on the May, 2010 Pay Scale survey, Illinois managers made $35,382, while the average Californian animal shelter manager earned $55,224. Size. The size of animal shelter also correlates to the manager’s salary. The more animals, the more employees, the more responsibility for the manager. In May, 2010, animal shelter managers working in a shelter employing 1 to 9 people, earned $33,835 on average.',
        'Salary by Region. Salaries for animal shelter managers also vary by region. According to the Salary Expert website, the highest compensation for such positions, about $60,000 a year, is in San Francisco. Animal shelter managers in the New York and Atlanta metro areas can expect to earn about $50,000 a year.',
        'Experience and years of service factor into an animal shelter manager’s salary. In May 2010, managers with 1 to 4 years of experience earned the least amount of money, $36,061. Those with 20 years or more experience, however, make $60,521, according to Pay Scale. Size. The size of animal shelter also correlates to the manager’s salary. The more animals, the more employees, the more responsibility for the manager. In May, 2010, animal shelter managers working in a shelter employing 1 to 9 people, earned $33,835 on average.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

Evaluation

Metrics

Cross Encoder Reranking

  • Datasets: NanoMSMARCO_R100, NanoNFCorpus_R100 and NanoNQ_R100
  • Evaluated with CrossEncoderRerankingEvaluator with these parameters:
    {
        "at_k": 10,
        "always_rerank_positives": true
    }
    
Metric NanoMSMARCO_R100 NanoNFCorpus_R100 NanoNQ_R100
map 0.5555 (+0.0659) 0.3715 (+0.1105) 0.6598 (+0.2401)
mrr@10 0.5440 (+0.0665) 0.5840 (+0.0842) 0.6830 (+0.2563)
ndcg@10 0.6012 (+0.0607) 0.4065 (+0.0815) 0.7061 (+0.2055)

Cross Encoder Nano BEIR

  • Dataset: NanoBEIR_R100_mean
  • Evaluated with CrossEncoderNanoBEIREvaluator with these parameters:
    {
        "dataset_names": [
            "msmarco",
            "nfcorpus",
            "nq"
        ],
        "rerank_k": 100,
        "at_k": 10,
        "always_rerank_positives": true
    }
    
Metric Value
map 0.5289 (+0.1389)
mrr@10 0.6037 (+0.1356)
ndcg@10 0.5713 (+0.1159)

Training Details

Training Dataset

ms_marco

  • Dataset: ms_marco at a47ee7a
  • Size: 78,704 training samples
  • Columns: query, docs, and labels
  • Approximate statistics based on the first 1000 samples:
    query docs labels
    type string list list
    details
    • min: 10 characters
    • mean: 33.66 characters
    • max: 103 characters
    • min: 3 elements
    • mean: 6.50 elements
    • max: 10 elements
    • min: 3 elements
    • mean: 6.50 elements
    • max: 10 elements
  • Samples:
    query docs labels
    a gram is times as large as a milligram ['1 milligram is 1000 times smaller than 1 gram so 1000 milligram = 1 gram. 1 milligram is 1000 times smaller than 1 gram so 1000 milligram = 1 gram. 1 microgram is 1000 times smaller than a 1 milligram and 1 million times smaller than 1 gram. 1 microgram is 1000 times smaller than a 1 milligram and 1 million times smaller than 1 gram. It takes 1 million mcg = 1 gram. Kilograms are 1000 times larger than a gram (1 kg = 1000 g). Kilograms are 1000 times larger than a gram (1 kg = 1000 g). Kilograms is used to denote weights of clients, upon med doses are based.', 'So to find out how many milligrams in grams, simply multiply it by 1000 or instead, use the converter below. 1 Gram = 1000 Milligrams. Gram is a metric system unit of mass. It is one thousandth (1/1000) of the metric system base unit, kilogram. It is a very commonly used unit of mass in daily life. The abbreviation is g. Milligram is a small unit of mass in metric system and used commonly in medicine and pharmacy etc.', 'A kil... [1, 1, 0, 0, 0, ...]
    income limit for dependent on taxes ['Qualifying Relative. For qualifying relatives, there is a strict limit on how much income they could make during the year. A qualifying relative cannot make more than $3,700 of gross income during the year. If he makes more than $3,700, then he cannot be a dependent. ', "Claiming an earner as a dependent can also affect that person's own tax filing. She can only claim her earned income plus $300 as a standard deduction on her return if this amounts to less than the standard deduction for that year, $5,950 as of 2012. Your dependent cannot claim anyone else as a dependent on her own return.", 'If you had $12,000 in income, for example, and $800 in eligible child care expenses, then, your credit would be 35 percent of $800, or $280. For taxpayers with incomes of more than $43,000, the percentage was 20 percent, with no upper income limit. So whether you had income of $44,000 or $440,000, your credit would be 20 percent. If you had $2,000 in care expenses, that translates to $400.', 'Hi... [1, 0, 0, 0, 0, ...]
    how much does it cost to have your own website ["Some hosts also offer discounts if you pay a year (or more) in advance. Prices vary from web host to web host but are usually (at the time I wrote this article) around $10 per month if your website is new and doesn't have much traffic or data. There are many ways to advertise your site, such as in newspapers, magazines, TV, as well as over the web. Since the cost for ads in the traditional media varies from country to country, you will have to do your own research.", 'Domain Registration. To have your own domain name (.com, .net, .org, .nu) you must register the domain. This typically costs anywhere between $10 to $12 (USD) and is billed on a yearly basis. GoDaddy and E-Starr are two companies that register your domain for less than $12 a year.', 'At minimum, you need to invest in your own domain name and hosting. Depending on the type of domain name you choose, the costs could run from just $10 a year, to hundreds or even millions ! The options for website hosting run the gamut in p... [1, 0, 0, 0, 0, ...]
  • Loss: ListNetLoss with these parameters:
    {
        "activation_fn": "torch.nn.modules.linear.Identity",
        "mini_batch_size": 16
    }
    

Evaluation Dataset

ms_marco

  • Dataset: ms_marco at a47ee7a
  • Size: 1,000 evaluation samples
  • Columns: query, docs, and labels
  • Approximate statistics based on the first 1000 samples:
    query docs labels
    type string list list
    details
    • min: 11 characters
    • mean: 34.3 characters
    • max: 105 characters
    • min: 3 elements
    • mean: 6.50 elements
    • max: 10 elements
    • min: 3 elements
    • mean: 6.50 elements
    • max: 10 elements
  • Samples:
    query docs labels
    what is the salary of an animal shelter manager ['Average Animal Shelter Salaries. The average salary for animal shelter jobs is $50,000. Average animal shelter salaries can vary greatly due to company, location, industry, experience and benefits. This salary was calculated using the average salary for all jobs with the term animal shelter anywhere in the job listing.', 'The salary that an animal shelter manager earns can vary based on a variety of factors such as their specific responsibilities, their years of experience, their educational background, and the region in which the position is located. Most animal shelter manager positions do not carry particularly high salaries, but those who follow animal rescue career paths tend to be willing to sacrifice some earning potential for the prospect of being able to help animals in need.', 'Animal shelter managers’ salaries vary depending on their state and location. For example, the Midwest sits at the bottom of the list. Based on the May, 2010 Pay Scale survey, Illinois managers made ... [1, 0, 0, 0, 0, ...]
    how long do quail eggs take to hatch ['I am often asked how long it takes quail eggs to hatch. The incubation period for bobwhite quail eggs is 23-24 days. Here is a link to information on incubating bobwhite quail eggs http://www.farmingfriends.com/incubating-bobwhite-quail-eggs/. The incubation period for Coturnix (Japanese) quail eggs is 17 days. Here is a link for information about incubating japanese quail eggs. http://www.farmingfriends.com/incubating-coturnix-japanese-quail/. Usually the hen will keep sitting if she thinks the other eggs are fertile and I have read that the hen can feel movement in the eggs so will keep sitting. Quail eggs take about 17 days to hatch and hen eggs take 21 days.', 'Turkey eggs usually take 21 to 28 days to hatch depending on what they are incubated in like an incubator or by a hen. It also depends on how fertile it is and how it is cared … for. It usually takes closer to 28 days. ', 'When you are trying to hatch Tennessee red quail eggs, it will take approximately 23 days. You shoul... [1, 0, 0, 0, 0, ...]
    sign of june ['June 1 – June 20 Gemini June 21 – June 30 Cancer. Those bearing the Gemini zodiac are incredibly flexible people who can adapt to almost any situation. They also possess a tenacity that not only enables them to rise above major setbacks but to take full advantage of negative situations as well.', 'Air is your paired element and as a Gemini, you have the most fluid connection with air out of any of the zodiac signs. You often witness curiosity pushing through you like a constant breeze and once you find something that truly interests you, your will seems to push with purpose.', 'There are 12 different stones listed as birthstones for the calendar month of June, or as Sun/Star, Planetary, or Talismanic stones for the Zodiac sign of Gemini or Cancer.', 'If you were born on the 26th of June you are a Cancer in the Western zodiac.', 'Birthday Meanings Of People Born On 6th June (Zodiac Sign Gemini). IF YOUR BIRTHDAY IS June 6, then you are a Gemini, who love to joke and play around. The G... [1, 0, 0, 0, 0, ...]
  • Loss: ListNetLoss with these parameters:
    {
        "activation_fn": "torch.nn.modules.linear.Identity",
        "mini_batch_size": 16
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 16
  • per_device_eval_batch_size: 16
  • learning_rate: 2e-05
  • num_train_epochs: 5
  • seed: 12
  • bf16: True
  • load_best_model_at_end: True

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 16
  • per_device_eval_batch_size: 16
  • per_gpu_train_batch_size: None
  • per_gpu_eval_batch_size: None
  • gradient_accumulation_steps: 1
  • eval_accumulation_steps: None
  • torch_empty_cache_steps: None
  • learning_rate: 2e-05
  • weight_decay: 0.0
  • adam_beta1: 0.9
  • adam_beta2: 0.999
  • adam_epsilon: 1e-08
  • max_grad_norm: 1.0
  • num_train_epochs: 5
  • max_steps: -1
  • lr_scheduler_type: linear
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.0
  • warmup_steps: 0
  • log_level: passive
  • log_level_replica: warning
  • log_on_each_node: True
  • logging_nan_inf_filter: True
  • save_safetensors: True
  • save_on_each_node: False
  • save_only_model: False
  • restore_callback_states_from_checkpoint: False
  • no_cuda: False
  • use_cpu: False
  • use_mps_device: False
  • seed: 12
  • data_seed: None
  • jit_mode_eval: False
  • use_ipex: False
  • bf16: True
  • fp16: False
  • fp16_opt_level: O1
  • half_precision_backend: auto
  • bf16_full_eval: False
  • fp16_full_eval: False
  • tf32: None
  • local_rank: 0
  • ddp_backend: None
  • tpu_num_cores: None
  • tpu_metrics_debug: False
  • debug: []
  • dataloader_drop_last: False
  • dataloader_num_workers: 0
  • dataloader_prefetch_factor: None
  • past_index: -1
  • disable_tqdm: False
  • remove_unused_columns: True
  • label_names: None
  • load_best_model_at_end: True
  • ignore_data_skip: False
  • fsdp: []
  • fsdp_min_num_params: 0
  • fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
  • fsdp_transformer_layer_cls_to_wrap: None
  • accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
  • deepspeed: None
  • label_smoothing_factor: 0.0
  • optim: adamw_torch
  • optim_args: None
  • adafactor: False
  • group_by_length: False
  • length_column_name: length
  • ddp_find_unused_parameters: None
  • ddp_bucket_cap_mb: None
  • ddp_broadcast_buffers: False
  • dataloader_pin_memory: True
  • dataloader_persistent_workers: False
  • skip_memory_metrics: True
  • use_legacy_prediction_loop: False
  • push_to_hub: False
  • resume_from_checkpoint: None
  • hub_model_id: None
  • hub_strategy: every_save
  • hub_private_repo: None
  • hub_always_push: False
  • hub_revision: None
  • gradient_checkpointing: False
  • gradient_checkpointing_kwargs: None
  • include_inputs_for_metrics: False
  • include_for_metrics: []
  • eval_do_concat_batches: True
  • fp16_backend: auto
  • push_to_hub_model_id: None
  • push_to_hub_organization: None
  • mp_parameters:
  • auto_find_batch_size: False
  • full_determinism: False
  • torchdynamo: None
  • ray_scope: last
  • ddp_timeout: 1800
  • torch_compile: False
  • torch_compile_backend: None
  • torch_compile_mode: None
  • include_tokens_per_second: False
  • include_num_input_tokens_seen: False
  • neftune_noise_alpha: None
  • optim_target_modules: None
  • batch_eval_metrics: False
  • eval_on_start: False
  • use_liger_kernel: False
  • liger_kernel_config: None
  • eval_use_gather_object: False
  • average_tokens_across_devices: False
  • prompts: None
  • batch_sampler: batch_sampler
  • multi_dataset_batch_sampler: proportional
  • router_mapping: {}
  • learning_rate_mapping: {}

Training Logs

Click to expand
Epoch Step Training Loss Validation Loss NanoMSMARCO_R100_ndcg@10 NanoNFCorpus_R100_ndcg@10 NanoNQ_R100_ndcg@10 NanoBEIR_R100_mean_ndcg@10
-1 -1 - - 0.0475 (-0.4929) 0.2463 (-0.0787) 0.0589 (-0.4418) 0.1176 (-0.3378)
0.0002 1 2.1526 - - - - -
0.0203 100 2.0847 2.0890 0.1559 (-0.3845) 0.2694 (-0.0556) 0.1030 (-0.3976) 0.1761 (-0.2792)
0.0407 200 2.0834 2.0844 0.3588 (-0.1817) 0.2750 (-0.0500) 0.3993 (-0.1013) 0.3444 (-0.1110)
0.0610 300 2.0775 2.0807 0.4209 (-0.1196) 0.3176 (-0.0075) 0.3781 (-0.1226) 0.3722 (-0.0832)
0.0813 400 2.0934 2.0779 0.5085 (-0.0319) 0.3194 (-0.0056) 0.5146 (+0.0140) 0.4475 (-0.0079)
0.1016 500 2.0693 2.0760 0.5408 (+0.0003) 0.3275 (+0.0024) 0.5793 (+0.0786) 0.4825 (+0.0271)
0.1220 600 2.0753 2.0753 0.5580 (+0.0176) 0.3464 (+0.0213) 0.6250 (+0.1244) 0.5098 (+0.0544)
0.1423 700 2.0712 2.0752 0.5756 (+0.0352) 0.3418 (+0.0168) 0.6753 (+0.1747) 0.5309 (+0.0756)
0.1626 800 2.0688 2.0743 0.5689 (+0.0284) 0.3171 (-0.0079) 0.6427 (+0.1420) 0.5096 (+0.0542)
0.1830 900 2.0723 2.0742 0.5762 (+0.0358) 0.3681 (+0.0431) 0.6712 (+0.1705) 0.5385 (+0.0831)
0.2033 1000 2.0743 2.0744 0.5799 (+0.0394) 0.3723 (+0.0473) 0.6483 (+0.1477) 0.5335 (+0.0781)
0.2236 1100 2.0728 2.0739 0.5995 (+0.0590) 0.3717 (+0.0467) 0.6674 (+0.1668) 0.5462 (+0.0908)
0.2440 1200 2.0735 2.0747 0.5869 (+0.0465) 0.3735 (+0.0484) 0.6365 (+0.1359) 0.5323 (+0.0770)
0.2643 1300 2.0677 2.0733 0.6003 (+0.0598) 0.3783 (+0.0532) 0.5812 (+0.0806) 0.5199 (+0.0646)
0.2846 1400 2.0647 2.0739 0.5669 (+0.0264) 0.3423 (+0.0172) 0.6041 (+0.1034) 0.5044 (+0.0490)
0.3049 1500 2.0764 2.0740 0.5728 (+0.0324) 0.3761 (+0.0510) 0.6380 (+0.1373) 0.5289 (+0.0736)
0.3253 1600 2.0652 2.0741 0.6202 (+0.0797) 0.3806 (+0.0556) 0.6523 (+0.1516) 0.5510 (+0.0956)
0.3456 1700 2.0733 2.0731 0.5948 (+0.0544) 0.3869 (+0.0619) 0.6421 (+0.1415) 0.5413 (+0.0859)
0.3659 1800 2.0679 2.0734 0.5797 (+0.0393) 0.3790 (+0.0540) 0.6340 (+0.1333) 0.5309 (+0.0755)
0.3863 1900 2.075 2.0733 0.6027 (+0.0623) 0.3931 (+0.0680) 0.6285 (+0.1278) 0.5414 (+0.0861)
0.4066 2000 2.0724 2.0732 0.5920 (+0.0516) 0.4056 (+0.0806) 0.6277 (+0.1271) 0.5418 (+0.0864)
0.4269 2100 2.0866 2.0721 0.6059 (+0.0655) 0.4060 (+0.0810) 0.6602 (+0.1596) 0.5574 (+0.1020)
0.4472 2200 2.0738 2.0725 0.6148 (+0.0744) 0.4167 (+0.0917) 0.6356 (+0.1350) 0.5557 (+0.1003)
0.4676 2300 2.0699 2.0722 0.6112 (+0.0707) 0.3992 (+0.0742) 0.6919 (+0.1913) 0.5674 (+0.1121)
0.4879 2400 2.0673 2.0725 0.5909 (+0.0505) 0.3820 (+0.0570) 0.6462 (+0.1455) 0.5397 (+0.0843)
0.5082 2500 2.0687 2.0725 0.5926 (+0.0521) 0.3838 (+0.0587) 0.6493 (+0.1486) 0.5419 (+0.0865)
0.5286 2600 2.0649 2.0725 0.6206 (+0.0801) 0.3825 (+0.0574) 0.6708 (+0.1702) 0.5580 (+0.1026)
0.5489 2700 2.0686 2.0729 0.5457 (+0.0053) 0.3728 (+0.0478) 0.6497 (+0.1491) 0.5227 (+0.0674)
0.5692 2800 2.0774 2.0716 0.5827 (+0.0422) 0.3931 (+0.0680) 0.6744 (+0.1737) 0.5500 (+0.0947)
0.5896 2900 2.0638 2.0724 0.6289 (+0.0884) 0.3915 (+0.0665) 0.6878 (+0.1872) 0.5694 (+0.1140)
0.6099 3000 2.0509 2.0718 0.5888 (+0.0483) 0.3906 (+0.0656) 0.6806 (+0.1800) 0.5533 (+0.0980)
0.6302 3100 2.0703 2.0715 0.6150 (+0.0746) 0.3832 (+0.0582) 0.6645 (+0.1638) 0.5542 (+0.0989)
0.6505 3200 2.0785 2.0718 0.5982 (+0.0577) 0.3766 (+0.0515) 0.6867 (+0.1860) 0.5538 (+0.0984)
0.6709 3300 2.0691 2.0716 0.6034 (+0.0629) 0.3814 (+0.0563) 0.6262 (+0.1255) 0.5370 (+0.0816)
0.6912 3400 2.0648 2.0721 0.6354 (+0.0950) 0.3811 (+0.0560) 0.6841 (+0.1834) 0.5669 (+0.1115)
0.7115 3500 2.0645 2.0719 0.6226 (+0.0821) 0.3802 (+0.0551) 0.6867 (+0.1860) 0.5631 (+0.1078)
0.7319 3600 2.0627 2.0719 0.5720 (+0.0316) 0.3800 (+0.0549) 0.6702 (+0.1696) 0.5407 (+0.0854)
0.7522 3700 2.0737 2.0719 0.5945 (+0.0540) 0.3793 (+0.0542) 0.6807 (+0.1801) 0.5515 (+0.0961)
0.7725 3800 2.0659 2.0721 0.5833 (+0.0429) 0.3749 (+0.0498) 0.6565 (+0.1559) 0.5382 (+0.0829)
0.7928 3900 2.063 2.0724 0.6012 (+0.0607) 0.4065 (+0.0815) 0.7061 (+0.2055) 0.5713 (+0.1159)
0.8132 4000 2.0685 2.0720 0.6246 (+0.0842) 0.3876 (+0.0626) 0.6563 (+0.1556) 0.5562 (+0.1008)
0.8335 4100 2.0705 2.0719 0.6045 (+0.0641) 0.3731 (+0.0480) 0.6908 (+0.1902) 0.5561 (+0.1008)
0.8538 4200 2.066 2.0714 0.6121 (+0.0717) 0.3783 (+0.0533) 0.6603 (+0.1597) 0.5503 (+0.0949)
0.8742 4300 2.06 2.0716 0.6186 (+0.0782) 0.3877 (+0.0627) 0.6495 (+0.1488) 0.5520 (+0.0966)
0.8945 4400 2.0777 2.0710 0.6079 (+0.0674) 0.3937 (+0.0686) 0.6654 (+0.1647) 0.5556 (+0.1003)
0.9148 4500 2.0558 2.0725 0.6238 (+0.0833) 0.3757 (+0.0507) 0.6381 (+0.1374) 0.5458 (+0.0905)
0.9351 4600 2.0675 2.0718 0.6313 (+0.0909) 0.3729 (+0.0478) 0.6558 (+0.1551) 0.5533 (+0.0979)
0.9555 4700 2.0755 2.0726 0.6048 (+0.0644) 0.3500 (+0.0250) 0.6126 (+0.1120) 0.5225 (+0.0671)
0.9758 4800 2.0658 2.0716 0.6217 (+0.0812) 0.3875 (+0.0625) 0.6695 (+0.1689) 0.5596 (+0.1042)
0.9961 4900 2.0735 2.0716 0.6073 (+0.0669) 0.3537 (+0.0287) 0.6470 (+0.1463) 0.5360 (+0.0806)
1.0165 5000 2.061 2.0712 0.6010 (+0.0605) 0.3785 (+0.0534) 0.6492 (+0.1486) 0.5429 (+0.0875)
1.0368 5100 2.0642 2.0719 0.6026 (+0.0622) 0.3840 (+0.0589) 0.6498 (+0.1491) 0.5454 (+0.0901)
1.0571 5200 2.0681 2.0711 0.6143 (+0.0739) 0.3802 (+0.0551) 0.7052 (+0.2046) 0.5666 (+0.1112)
1.0775 5300 2.0596 2.0709 0.6166 (+0.0762) 0.3897 (+0.0647) 0.6642 (+0.1636) 0.5569 (+0.1015)
1.0978 5400 2.0576 2.0710 0.5779 (+0.0375) 0.3831 (+0.0581) 0.6276 (+0.1269) 0.5295 (+0.0742)
1.1181 5500 2.0675 2.0720 0.6164 (+0.0759) 0.3843 (+0.0593) 0.6712 (+0.1705) 0.5573 (+0.1019)
1.1384 5600 2.0647 2.0720 0.5934 (+0.0529) 0.3860 (+0.0610) 0.6664 (+0.1657) 0.5486 (+0.0932)
1.1588 5700 2.0587 2.0715 0.6235 (+0.0831) 0.3813 (+0.0562) 0.6858 (+0.1852) 0.5635 (+0.1082)
1.1791 5800 2.0608 2.0718 0.6298 (+0.0894) 0.3680 (+0.0430) 0.6572 (+0.1565) 0.5517 (+0.0963)
1.1994 5900 2.0642 2.0720 0.5838 (+0.0434) 0.3961 (+0.0710) 0.6678 (+0.1671) 0.5492 (+0.0939)
1.2198 6000 2.0588 2.0719 0.5920 (+0.0516) 0.3869 (+0.0619) 0.6653 (+0.1646) 0.5481 (+0.0927)
1.2401 6100 2.0631 2.0722 0.5666 (+0.0262) 0.3672 (+0.0421) 0.6692 (+0.1685) 0.5343 (+0.0789)
1.2604 6200 2.0591 2.0724 0.5742 (+0.0338) 0.3637 (+0.0387) 0.6546 (+0.1540) 0.5309 (+0.0755)
1.2807 6300 2.0567 2.0725 0.5693 (+0.0288) 0.3756 (+0.0506) 0.6682 (+0.1676) 0.5377 (+0.0823)
1.3011 6400 2.0597 2.0716 0.5587 (+0.0183) 0.3599 (+0.0349) 0.6731 (+0.1724) 0.5306 (+0.0752)
1.3214 6500 2.0647 2.0727 0.5838 (+0.0434) 0.3633 (+0.0383) 0.6556 (+0.1550) 0.5343 (+0.0789)
1.3417 6600 2.0624 2.0712 0.5602 (+0.0198) 0.3697 (+0.0446) 0.6359 (+0.1353) 0.5219 (+0.0666)
1.3621 6700 2.0604 2.0714 0.5784 (+0.0379) 0.3629 (+0.0379) 0.6383 (+0.1376) 0.5265 (+0.0711)
1.3824 6800 2.0702 2.0711 0.5400 (-0.0004) 0.3814 (+0.0563) 0.5985 (+0.0979) 0.5066 (+0.0512)
1.4027 6900 2.0569 2.0713 0.5239 (-0.0165) 0.3827 (+0.0576) 0.6110 (+0.1103) 0.5059 (+0.0505)
1.4231 7000 2.0689 2.0726 0.5382 (-0.0022) 0.3912 (+0.0661) 0.6323 (+0.1316) 0.5205 (+0.0652)
1.4434 7100 2.056 2.0717 0.5331 (-0.0074) 0.3691 (+0.0441) 0.6244 (+0.1238) 0.5089 (+0.0535)
1.4637 7200 2.0573 2.0717 0.5668 (+0.0263) 0.3769 (+0.0519) 0.6235 (+0.1228) 0.5224 (+0.0670)
1.4840 7300 2.0544 2.0720 0.5543 (+0.0139) 0.3875 (+0.0624) 0.6240 (+0.1233) 0.5219 (+0.0666)
1.5044 7400 2.0684 2.0718 0.5684 (+0.0280) 0.3753 (+0.0503) 0.6075 (+0.1068) 0.5171 (+0.0617)
1.5247 7500 2.0588 2.0718 0.5663 (+0.0259) 0.3904 (+0.0654) 0.6564 (+0.1557) 0.5377 (+0.0823)
1.5450 7600 2.0556 2.0713 0.5764 (+0.0360) 0.3732 (+0.0482) 0.6323 (+0.1317) 0.5273 (+0.0719)
1.5654 7700 2.0572 2.0716 0.5589 (+0.0185) 0.3876 (+0.0626) 0.6327 (+0.1320) 0.5264 (+0.0710)
1.5857 7800 2.0662 2.0712 0.5577 (+0.0173) 0.3912 (+0.0662) 0.6744 (+0.1737) 0.5411 (+0.0857)
1.6060 7900 2.063 2.0720 0.5444 (+0.0040) 0.3859 (+0.0609) 0.6401 (+0.1395) 0.5235 (+0.0681)
1.6263 8000 2.0641 2.0728 0.5565 (+0.0160) 0.4032 (+0.0781) 0.6318 (+0.1311) 0.5305 (+0.0751)
1.6467 8100 2.0529 2.0720 0.5508 (+0.0104) 0.3929 (+0.0679) 0.6016 (+0.1010) 0.5151 (+0.0598)
1.6670 8200 2.0734 2.0732 0.5626 (+0.0221) 0.3920 (+0.0670) 0.6223 (+0.1216) 0.5256 (+0.0702)
1.6873 8300 2.0645 2.0719 0.5660 (+0.0255) 0.3953 (+0.0702) 0.6511 (+0.1504) 0.5374 (+0.0821)
1.7077 8400 2.0637 2.0719 0.5640 (+0.0236) 0.3890 (+0.0640) 0.6608 (+0.1602) 0.5379 (+0.0826)
1.7280 8500 2.0588 2.0715 0.5875 (+0.0471) 0.3911 (+0.0661) 0.6827 (+0.1820) 0.5538 (+0.0984)
1.7483 8600 2.0584 2.0720 0.5755 (+0.0351) 0.4047 (+0.0797) 0.6930 (+0.1924) 0.5578 (+0.1024)
1.7687 8700 2.0622 2.0726 0.5884 (+0.0480) 0.3793 (+0.0542) 0.6777 (+0.1771) 0.5485 (+0.0931)
1.7890 8800 2.059 2.0725 0.5876 (+0.0472) 0.3962 (+0.0712) 0.7014 (+0.2007) 0.5617 (+0.1064)
1.8093 8900 2.0561 2.0720 0.5855 (+0.0451) 0.3997 (+0.0747) 0.6678 (+0.1671) 0.5510 (+0.0956)
1.8296 9000 2.0627 2.0718 0.5876 (+0.0472) 0.3902 (+0.0652) 0.6706 (+0.1700) 0.5495 (+0.0941)
1.8500 9100 2.0516 2.0716 0.5754 (+0.0349) 0.3946 (+0.0696) 0.6524 (+0.1518) 0.5408 (+0.0854)
1.8703 9200 2.0553 2.0714 0.5678 (+0.0274) 0.3869 (+0.0619) 0.6543 (+0.1536) 0.5363 (+0.0810)
1.8906 9300 2.0606 2.0714 0.5426 (+0.0022) 0.4028 (+0.0778) 0.6784 (+0.1778) 0.5413 (+0.0859)
1.9110 9400 2.0502 2.0714 0.5644 (+0.0240) 0.3986 (+0.0736) 0.6751 (+0.1745) 0.5461 (+0.0907)
1.9313 9500 2.0617 2.0719 0.5160 (-0.0244) 0.3915 (+0.0664) 0.6381 (+0.1374) 0.5152 (+0.0598)
1.9516 9600 2.0525 2.0712 0.5532 (+0.0128) 0.3886 (+0.0636) 0.6453 (+0.1446) 0.5290 (+0.0737)
1.9719 9700 2.059 2.0713 0.5565 (+0.0161) 0.3922 (+0.0671) 0.6576 (+0.1569) 0.5354 (+0.0800)
1.9923 9800 2.0607 2.0723 0.5623 (+0.0219) 0.3930 (+0.0680) 0.6539 (+0.1533) 0.5364 (+0.0810)
2.0126 9900 2.0462 2.0758 0.5482 (+0.0078) 0.4075 (+0.0825) 0.6371 (+0.1364) 0.5309 (+0.0756)
2.0329 10000 2.0481 2.0783 0.5441 (+0.0037) 0.3783 (+0.0532) 0.6386 (+0.1379) 0.5203 (+0.0649)
2.0533 10100 2.0392 2.0772 0.5299 (-0.0106) 0.3750 (+0.0499) 0.6398 (+0.1391) 0.5149 (+0.0595)
2.0736 10200 2.0362 2.0779 0.5068 (-0.0337) 0.3699 (+0.0448) 0.6162 (+0.1156) 0.4976 (+0.0422)
2.0939 10300 2.0312 2.0782 0.5188 (-0.0216) 0.3783 (+0.0533) 0.6282 (+0.1275) 0.5085 (+0.0531)
2.1143 10400 2.0307 2.0773 0.5011 (-0.0393) 0.3768 (+0.0517) 0.6125 (+0.1118) 0.4968 (+0.0414)
2.1346 10500 2.0425 2.0779 0.5203 (-0.0201) 0.3796 (+0.0545) 0.6380 (+0.1373) 0.5126 (+0.0572)
2.1549 10600 2.0441 2.0793 0.5256 (-0.0148) 0.3671 (+0.0421) 0.6091 (+0.1085) 0.5006 (+0.0453)
2.1752 10700 2.046 2.0786 0.5220 (-0.0184) 0.3836 (+0.0585) 0.6237 (+0.1231) 0.5097 (+0.0544)
2.1956 10800 2.0384 2.0763 0.5089 (-0.0315) 0.3791 (+0.0541) 0.6307 (+0.1300) 0.5062 (+0.0509)
2.2159 10900 2.0373 2.0787 0.5019 (-0.0386) 0.3699 (+0.0448) 0.6322 (+0.1316) 0.5013 (+0.0459)
2.2362 11000 2.0377 2.0772 0.5220 (-0.0184) 0.3697 (+0.0447) 0.5991 (+0.0984) 0.4969 (+0.0416)
2.2566 11100 2.0422 2.0766 0.5232 (-0.0173) 0.3803 (+0.0552) 0.6249 (+0.1243) 0.5094 (+0.0541)
2.2769 11200 2.0412 2.0763 0.5138 (-0.0266) 0.3739 (+0.0489) 0.5752 (+0.0746) 0.4877 (+0.0323)
2.2972 11300 2.0411 2.0777 0.5240 (-0.0164) 0.3728 (+0.0478) 0.6340 (+0.1334) 0.5103 (+0.0549)
2.3175 11400 2.0392 2.0770 0.5093 (-0.0312) 0.3638 (+0.0387) 0.5971 (+0.0965) 0.4901 (+0.0347)
2.3379 11500 2.0346 2.0798 0.5061 (-0.0343) 0.3748 (+0.0497) 0.6418 (+0.1412) 0.5076 (+0.0522)
2.3582 11600 2.0501 2.0772 0.5231 (-0.0173) 0.3734 (+0.0483) 0.6018 (+0.1012) 0.4994 (+0.0441)
2.3785 11700 2.0488 2.0787 0.4924 (-0.0480) 0.3830 (+0.0580) 0.6310 (+0.1304) 0.5021 (+0.0468)
2.3989 11800 2.0407 2.0752 0.5073 (-0.0332) 0.3853 (+0.0602) 0.6053 (+0.1046) 0.4993 (+0.0439)
2.4192 11900 2.039 2.0793 0.5047 (-0.0357) 0.3815 (+0.0565) 0.6049 (+0.1042) 0.4970 (+0.0417)
2.4395 12000 2.0405 2.0762 0.4999 (-0.0405) 0.3797 (+0.0547) 0.6229 (+0.1222) 0.5009 (+0.0455)
2.4598 12100 2.0422 2.0773 0.5257 (-0.0147) 0.3741 (+0.0490) 0.6304 (+0.1297) 0.5100 (+0.0547)
2.4802 12200 2.0395 2.0787 0.5189 (-0.0215) 0.3614 (+0.0364) 0.6396 (+0.1389) 0.5066 (+0.0513)
2.5005 12300 2.0357 2.0800 0.4991 (-0.0413) 0.3693 (+0.0442) 0.6140 (+0.1133) 0.4941 (+0.0387)
2.5208 12400 2.0449 2.0766 0.4701 (-0.0704) 0.3703 (+0.0453) 0.6154 (+0.1147) 0.4852 (+0.0299)
2.5412 12500 2.0362 2.0765 0.4795 (-0.0609) 0.3794 (+0.0543) 0.6068 (+0.1062) 0.4886 (+0.0332)
2.5615 12600 2.0357 2.0774 0.4893 (-0.0511) 0.3635 (+0.0385) 0.6128 (+0.1121) 0.4885 (+0.0332)
2.5818 12700 2.0478 2.0793 0.4924 (-0.0481) 0.3738 (+0.0488) 0.6028 (+0.1022) 0.4897 (+0.0343)
2.6022 12800 2.0378 2.0777 0.5023 (-0.0381) 0.3667 (+0.0417) 0.6150 (+0.1144) 0.4947 (+0.0393)
2.6225 12900 2.0377 2.0787 0.5118 (-0.0286) 0.3709 (+0.0459) 0.6317 (+0.1310) 0.5048 (+0.0494)
2.6428 13000 2.0392 2.0791 0.5099 (-0.0305) 0.3829 (+0.0579) 0.6300 (+0.1293) 0.5076 (+0.0522)
2.6631 13100 2.0367 2.0786 0.5189 (-0.0215) 0.3890 (+0.0639) 0.6537 (+0.1530) 0.5205 (+0.0651)
2.6835 13200 2.0431 2.0771 0.4851 (-0.0553) 0.3682 (+0.0431) 0.6240 (+0.1234) 0.4924 (+0.0371)
2.7038 13300 2.0392 2.0793 0.5057 (-0.0347) 0.3841 (+0.0590) 0.6380 (+0.1373) 0.5092 (+0.0539)
2.7241 13400 2.0443 2.0767 0.4739 (-0.0665) 0.3669 (+0.0418) 0.6151 (+0.1145) 0.4853 (+0.0299)
2.7445 13500 2.0501 2.0773 0.4633 (-0.0771) 0.3543 (+0.0293) 0.6069 (+0.1063) 0.4748 (+0.0195)
2.7648 13600 2.0321 2.0780 0.4930 (-0.0474) 0.3644 (+0.0394) 0.6082 (+0.1075) 0.4885 (+0.0332)
2.7851 13700 2.0273 2.0786 0.4572 (-0.0832) 0.3653 (+0.0402) 0.6065 (+0.1058) 0.4763 (+0.0210)
2.8054 13800 2.0326 2.0776 0.4665 (-0.0739) 0.3709 (+0.0459) 0.5942 (+0.0936) 0.4772 (+0.0218)
2.8258 13900 2.0489 2.0769 0.4748 (-0.0657) 0.3716 (+0.0466) 0.5893 (+0.0886) 0.4785 (+0.0232)
2.8461 14000 2.0442 2.0775 0.5012 (-0.0392) 0.3711 (+0.0460) 0.5988 (+0.0981) 0.4903 (+0.0350)
2.8664 14100 2.0365 2.0779 0.4766 (-0.0639) 0.3759 (+0.0508) 0.5890 (+0.0884) 0.4805 (+0.0251)
2.8868 14200 2.0334 2.0771 0.4703 (-0.0701) 0.3729 (+0.0478) 0.5817 (+0.0810) 0.4749 (+0.0196)
2.9071 14300 2.037 2.0777 0.4868 (-0.0536) 0.3748 (+0.0498) 0.6002 (+0.0996) 0.4873 (+0.0319)
2.9274 14400 2.0465 2.0775 0.4884 (-0.0520) 0.3658 (+0.0408) 0.6351 (+0.1345) 0.4965 (+0.0411)
2.9478 14500 2.0314 2.0779 0.4955 (-0.0449) 0.3627 (+0.0377) 0.6120 (+0.1114) 0.4901 (+0.0347)
2.9681 14600 2.0436 2.0776 0.5073 (-0.0331) 0.3639 (+0.0389) 0.5965 (+0.0958) 0.4892 (+0.0339)
2.9884 14700 2.0382 2.0771 0.4987 (-0.0417) 0.3728 (+0.0477) 0.6041 (+0.1034) 0.4919 (+0.0365)
3.0087 14800 2.0339 2.0793 0.4615 (-0.0789) 0.3673 (+0.0423) 0.5634 (+0.0628) 0.4641 (+0.0087)
3.0291 14900 2.0197 2.0827 0.4515 (-0.0889) 0.3683 (+0.0433) 0.5774 (+0.0767) 0.4658 (+0.0104)
3.0494 15000 2.0155 2.0809 0.4226 (-0.1178) 0.3684 (+0.0434) 0.5542 (+0.0536) 0.4484 (-0.0069)
3.0697 15100 2.0157 2.0833 0.4044 (-0.1360) 0.3721 (+0.0471) 0.5168 (+0.0162) 0.4311 (-0.0242)
3.0901 15200 2.0132 2.0821 0.4341 (-0.1064) 0.3650 (+0.0400) 0.5546 (+0.0539) 0.4512 (-0.0041)
3.1104 15300 2.0201 2.0829 0.4324 (-0.1081) 0.3527 (+0.0277) 0.5693 (+0.0687) 0.4515 (-0.0039)
3.1307 15400 2.0219 2.0816 0.4606 (-0.0798) 0.3623 (+0.0373) 0.5450 (+0.0443) 0.4560 (+0.0006)
3.1510 15500 2.0189 2.0821 0.4729 (-0.0675) 0.3579 (+0.0328) 0.5762 (+0.0756) 0.4690 (+0.0136)
3.1714 15600 2.0126 2.0831 0.4415 (-0.0989) 0.3503 (+0.0253) 0.5613 (+0.0606) 0.4510 (-0.0043)
3.1917 15700 2.0201 2.0843 0.4552 (-0.0852) 0.3488 (+0.0238) 0.5897 (+0.0891) 0.4646 (+0.0092)
3.2120 15800 2.0189 2.0850 0.4860 (-0.0544) 0.3520 (+0.0270) 0.5931 (+0.0924) 0.4770 (+0.0217)
3.2324 15900 2.0196 2.0827 0.4668 (-0.0736) 0.3485 (+0.0235) 0.5721 (+0.0715) 0.4625 (+0.0071)
3.2527 16000 2.0195 2.0831 0.4649 (-0.0756) 0.3522 (+0.0272) 0.5928 (+0.0921) 0.4699 (+0.0146)
3.2730 16100 2.0197 2.0830 0.4790 (-0.0614) 0.3549 (+0.0298) 0.5798 (+0.0792) 0.4712 (+0.0159)
3.2934 16200 2.0253 2.0824 0.4680 (-0.0725) 0.3633 (+0.0383) 0.5854 (+0.0848) 0.4722 (+0.0169)
3.3137 16300 2.0244 2.0844 0.4810 (-0.0594) 0.3678 (+0.0427) 0.5862 (+0.0855) 0.4783 (+0.0230)
3.3340 16400 2.0164 2.0844 0.4550 (-0.0854) 0.3614 (+0.0364) 0.5888 (+0.0881) 0.4684 (+0.0130)
3.3543 16500 2.0065 2.0835 0.5012 (-0.0392) 0.3629 (+0.0379) 0.6218 (+0.1212) 0.4953 (+0.0399)
3.3747 16600 2.02 2.0830 0.4915 (-0.0489) 0.3518 (+0.0267) 0.5898 (+0.0891) 0.4777 (+0.0223)
3.3950 16700 2.0213 2.0842 0.4778 (-0.0627) 0.3605 (+0.0354) 0.5656 (+0.0649) 0.4679 (+0.0126)
3.4153 16800 2.02 2.0842 0.4533 (-0.0871) 0.3663 (+0.0412) 0.5447 (+0.0440) 0.4548 (-0.0006)
3.4357 16900 2.0178 2.0859 0.4510 (-0.0894) 0.3620 (+0.0369) 0.5665 (+0.0658) 0.4598 (+0.0044)
3.4560 17000 2.0154 2.0849 0.4662 (-0.0742) 0.3673 (+0.0422) 0.5558 (+0.0552) 0.4631 (+0.0077)
3.4763 17100 2.0145 2.0842 0.4698 (-0.0706) 0.3511 (+0.0261) 0.5797 (+0.0790) 0.4669 (+0.0115)
3.4966 17200 2.019 2.0845 0.4305 (-0.1099) 0.3659 (+0.0409) 0.5614 (+0.0607) 0.4526 (-0.0028)
3.5170 17300 2.0151 2.0843 0.4671 (-0.0733) 0.3644 (+0.0394) 0.5810 (+0.0804) 0.4708 (+0.0155)
3.5373 17400 2.0131 2.0844 0.4711 (-0.0693) 0.3579 (+0.0329) 0.5760 (+0.0754) 0.4684 (+0.0130)
3.5576 17500 2.0145 2.0822 0.4687 (-0.0717) 0.3588 (+0.0338) 0.5853 (+0.0847) 0.4709 (+0.0156)
3.5780 17600 2.0184 2.0840 0.4723 (-0.0682) 0.3530 (+0.0279) 0.5898 (+0.0891) 0.4717 (+0.0163)
3.5983 17700 2.0219 2.0839 0.4657 (-0.0748) 0.3538 (+0.0288) 0.5897 (+0.0890) 0.4697 (+0.0144)
3.6186 17800 2.0186 2.0842 0.4466 (-0.0938) 0.3542 (+0.0292) 0.5779 (+0.0773) 0.4596 (+0.0042)
3.6390 17900 2.0097 2.0843 0.4828 (-0.0577) 0.3607 (+0.0357) 0.5720 (+0.0713) 0.4718 (+0.0164)
3.6593 18000 2.0208 2.0844 0.4724 (-0.0680) 0.3644 (+0.0394) 0.5634 (+0.0627) 0.4667 (+0.0114)
3.6796 18100 2.0254 2.0839 0.5004 (-0.0400) 0.3675 (+0.0425) 0.5854 (+0.0848) 0.4844 (+0.0291)
3.6999 18200 2.0205 2.0852 0.4712 (-0.0692) 0.3729 (+0.0479) 0.5775 (+0.0769) 0.4739 (+0.0185)
3.7203 18300 2.0213 2.0856 0.4849 (-0.0555) 0.3645 (+0.0395) 0.5771 (+0.0765) 0.4755 (+0.0201)
3.7406 18400 2.0321 2.0850 0.4516 (-0.0889) 0.3621 (+0.0371) 0.5578 (+0.0571) 0.4571 (+0.0018)
3.7609 18500 2.0256 2.0832 0.4442 (-0.0962) 0.3710 (+0.0460) 0.5593 (+0.0587) 0.4582 (+0.0028)
3.7813 18600 2.0178 2.0846 0.4536 (-0.0869) 0.3689 (+0.0439) 0.5854 (+0.0847) 0.4693 (+0.0139)
3.8016 18700 2.0156 2.0842 0.4469 (-0.0935) 0.3620 (+0.0370) 0.5622 (+0.0615) 0.4570 (+0.0017)
3.8219 18800 2.0202 2.0836 0.4712 (-0.0692) 0.3690 (+0.0439) 0.5732 (+0.0726) 0.4711 (+0.0158)
3.8422 18900 2.0157 2.0838 0.4609 (-0.0796) 0.3508 (+0.0258) 0.5849 (+0.0842) 0.4655 (+0.0101)
3.8626 19000 2.0139 2.0856 0.4356 (-0.1048) 0.3519 (+0.0269) 0.5631 (+0.0624) 0.4502 (-0.0052)
3.8829 19100 2.0227 2.0832 0.4603 (-0.0802) 0.3592 (+0.0342) 0.5955 (+0.0948) 0.4717 (+0.0163)
3.9032 19200 2.0184 2.0840 0.4684 (-0.0720) 0.3562 (+0.0312) 0.5860 (+0.0853) 0.4702 (+0.0148)
3.9236 19300 2.0271 2.0832 0.4545 (-0.0860) 0.3635 (+0.0385) 0.5709 (+0.0703) 0.4630 (+0.0076)
3.9439 19400 2.0283 2.0834 0.4524 (-0.0880) 0.3569 (+0.0318) 0.5711 (+0.0704) 0.4601 (+0.0047)
3.9642 19500 2.0148 2.0836 0.4536 (-0.0869) 0.3556 (+0.0306) 0.5735 (+0.0728) 0.4609 (+0.0055)
3.9845 19600 2.021 2.0841 0.4660 (-0.0744) 0.3563 (+0.0313) 0.5622 (+0.0615) 0.4615 (+0.0061)
4.0049 19700 2.0185 2.0835 0.4452 (-0.0952) 0.3613 (+0.0363) 0.5338 (+0.0332) 0.4468 (-0.0086)
4.0252 19800 2.005 2.0859 0.4500 (-0.0904) 0.3582 (+0.0332) 0.5490 (+0.0484) 0.4524 (-0.0029)
4.0455 19900 2.0078 2.0864 0.4131 (-0.1273) 0.3477 (+0.0227) 0.5343 (+0.0337) 0.4317 (-0.0237)
4.0659 20000 2.0096 2.0861 0.4120 (-0.1285) 0.3549 (+0.0299) 0.5329 (+0.0322) 0.4332 (-0.0221)
4.0862 20100 2.009 2.0875 0.4151 (-0.1253) 0.3475 (+0.0224) 0.5105 (+0.0098) 0.4243 (-0.0310)
4.1065 20200 2.0131 2.0869 0.4550 (-0.0854) 0.3499 (+0.0248) 0.5395 (+0.0389) 0.4481 (-0.0072)
4.1269 20300 2.0067 2.0873 0.4371 (-0.1033) 0.3471 (+0.0221) 0.5333 (+0.0326) 0.4392 (-0.0162)
4.1472 20400 2.0032 2.0883 0.4493 (-0.0912) 0.3542 (+0.0291) 0.5173 (+0.0167) 0.4403 (-0.0151)
4.1675 20500 2.0077 2.0889 0.4513 (-0.0892) 0.3424 (+0.0173) 0.5270 (+0.0263) 0.4402 (-0.0152)
4.1878 20600 2.0103 2.0865 0.4426 (-0.0978) 0.3388 (+0.0138) 0.5171 (+0.0164) 0.4328 (-0.0225)
4.2082 20700 2.0067 2.0872 0.4429 (-0.0976) 0.3445 (+0.0194) 0.4934 (-0.0073) 0.4269 (-0.0285)
4.2285 20800 2.0058 2.0877 0.4382 (-0.1022) 0.3469 (+0.0219) 0.5075 (+0.0069) 0.4309 (-0.0245)
4.2488 20900 2.0117 2.0884 0.4319 (-0.1085) 0.3432 (+0.0182) 0.5178 (+0.0172) 0.4310 (-0.0244)
4.2692 21000 2.0143 2.0872 0.4122 (-0.1282) 0.3396 (+0.0146) 0.5132 (+0.0126) 0.4217 (-0.0337)
4.2895 21100 2.0108 2.0875 0.4333 (-0.1071) 0.3484 (+0.0233) 0.5300 (+0.0293) 0.4372 (-0.0182)
4.3098 21200 2.0104 2.0879 0.4051 (-0.1353) 0.3561 (+0.0311) 0.5248 (+0.0242) 0.4287 (-0.0267)
4.3301 21300 2.0188 2.0871 0.4072 (-0.1332) 0.3533 (+0.0282) 0.5204 (+0.0198) 0.4270 (-0.0284)
4.3505 21400 2.0051 2.0878 0.4469 (-0.0935) 0.3536 (+0.0285) 0.5331 (+0.0324) 0.4445 (-0.0109)
4.3708 21500 2.0109 2.0873 0.4213 (-0.1191) 0.3501 (+0.0250) 0.5313 (+0.0306) 0.4342 (-0.0212)
4.3911 21600 2.006 2.0872 0.4388 (-0.1017) 0.3515 (+0.0265) 0.5109 (+0.0103) 0.4337 (-0.0216)
4.4115 21700 2.0123 2.0878 0.4168 (-0.1236) 0.3567 (+0.0316) 0.5057 (+0.0051) 0.4264 (-0.0290)
4.4318 21800 2.0092 2.0891 0.4030 (-0.1375) 0.3562 (+0.0312) 0.5083 (+0.0076) 0.4225 (-0.0329)
4.4521 21900 2.0045 2.0887 0.4159 (-0.1245) 0.3503 (+0.0253) 0.5123 (+0.0117) 0.4262 (-0.0292)
4.4725 22000 2.0178 2.0861 0.4375 (-0.1029) 0.3559 (+0.0308) 0.5314 (+0.0307) 0.4416 (-0.0138)
4.4928 22100 2.0145 2.0881 0.4408 (-0.0996) 0.3617 (+0.0366) 0.5040 (+0.0033) 0.4355 (-0.0199)
4.5131 22200 2.0025 2.0882 0.4291 (-0.1113) 0.3592 (+0.0341) 0.5233 (+0.0226) 0.4372 (-0.0182)
4.5334 22300 2.0113 2.0892 0.4253 (-0.1151) 0.3569 (+0.0318) 0.5032 (+0.0025) 0.4285 (-0.0269)
4.5538 22400 2.0167 2.0883 0.4236 (-0.1169) 0.3543 (+0.0292) 0.5218 (+0.0211) 0.4332 (-0.0222)
4.5741 22500 2.0103 2.0881 0.4151 (-0.1254) 0.3611 (+0.0360) 0.5097 (+0.0090) 0.4286 (-0.0268)
4.5944 22600 2.0072 2.0865 0.4143 (-0.1261) 0.3554 (+0.0304) 0.5245 (+0.0239) 0.4314 (-0.0240)
4.6148 22700 2.0001 2.0871 0.4091 (-0.1313) 0.3575 (+0.0325) 0.5231 (+0.0225) 0.4299 (-0.0254)
4.6351 22800 2.0078 2.0881 0.4099 (-0.1306) 0.3501 (+0.0250) 0.5151 (+0.0145) 0.4250 (-0.0303)
4.6554 22900 2.0003 2.0878 0.4104 (-0.1300) 0.3460 (+0.0210) 0.5290 (+0.0284) 0.4285 (-0.0269)
4.6757 23000 2.0152 2.0877 0.4269 (-0.1135) 0.3495 (+0.0245) 0.5259 (+0.0253) 0.4341 (-0.0212)
4.6961 23100 2.011 2.0881 0.4442 (-0.0962) 0.3516 (+0.0266) 0.5347 (+0.0341) 0.4435 (-0.0118)
4.7164 23200 2.0012 2.0884 0.4324 (-0.1081) 0.3498 (+0.0247) 0.5252 (+0.0245) 0.4358 (-0.0196)
4.7367 23300 2.0176 2.0876 0.4238 (-0.1166) 0.3510 (+0.0260) 0.5339 (+0.0333) 0.4363 (-0.0191)
4.7571 23400 2.0063 2.0884 0.4280 (-0.1125) 0.3546 (+0.0295) 0.5347 (+0.0341) 0.4391 (-0.0163)
4.7774 23500 2.0049 2.0884 0.4358 (-0.1047) 0.3497 (+0.0247) 0.5300 (+0.0293) 0.4385 (-0.0169)
4.7977 23600 2.0078 2.0873 0.4194 (-0.1211) 0.3508 (+0.0257) 0.5396 (+0.0390) 0.4366 (-0.0188)
4.8181 23700 2.0143 2.0875 0.4268 (-0.1137) 0.3568 (+0.0318) 0.5266 (+0.0260) 0.4367 (-0.0186)
4.8384 23800 2.0162 2.0872 0.4271 (-0.1133) 0.3525 (+0.0274) 0.5331 (+0.0325) 0.4376 (-0.0178)
4.8587 23900 2.0089 2.0879 0.4297 (-0.1107) 0.3512 (+0.0262) 0.5241 (+0.0235) 0.4350 (-0.0203)
4.8790 24000 1.9988 2.0879 0.4273 (-0.1131) 0.3481 (+0.0231) 0.5238 (+0.0231) 0.4331 (-0.0223)
4.8994 24100 2.0074 2.0881 0.4362 (-0.1042) 0.3519 (+0.0269) 0.5283 (+0.0276) 0.4388 (-0.0166)
4.9197 24200 2.0179 2.0878 0.4362 (-0.1042) 0.3473 (+0.0223) 0.5285 (+0.0279) 0.4373 (-0.0180)
4.9400 24300 2.0162 2.0874 0.4228 (-0.1176) 0.3495 (+0.0245) 0.5348 (+0.0341) 0.4357 (-0.0197)
4.9604 24400 1.9986 2.0874 0.4268 (-0.1136) 0.3493 (+0.0242) 0.5364 (+0.0358) 0.4375 (-0.0179)
4.9807 24500 2.0158 2.0874 0.4331 (-0.1074) 0.3515 (+0.0265) 0.5379 (+0.0373) 0.4408 (-0.0145)
-1 -1 - - 0.6012 (+0.0607) 0.4065 (+0.0815) 0.7061 (+0.2055) 0.5713 (+0.1159)
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.18
  • Sentence Transformers: 5.0.0
  • Transformers: 4.56.0.dev0
  • PyTorch: 2.7.1+cu126
  • Accelerate: 1.9.0
  • Datasets: 4.0.0
  • Tokenizers: 0.21.4

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}

ListNetLoss

@inproceedings{cao2007learning,
    title={Learning to Rank: From Pairwise Approach to Listwise Approach},
    author={Cao, Zhe and Qin, Tao and Liu, Tie-Yan and Tsai, Ming-Feng and Li, Hang},
    booktitle={Proceedings of the 24th international conference on Machine learning},
    pages={129--136},
    year={2007}
}
Downloads last month
2
Safetensors
Model size
396M params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for rahulseetharaman/reranker-msmarco-v1.1-ettin-encoder-400m-listnet

Finetuned
(4)
this model

Dataset used to train rahulseetharaman/reranker-msmarco-v1.1-ettin-encoder-400m-listnet

Evaluation results