all-MiniLM-eedi-2024

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 on the csv dataset. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: sentence-transformers/all-MiniLM-L6-v2 <!-- at revision 8b3219a92973c328a8e22fadcfa821b5dc75636a -->
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 tokens
• Similarity Function: Cosine Similarity
• Training Dataset:
  • csv

<!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

Model Sources

• Documentation: Sentence Transformers Documentation
• Repository: Sentence Transformers on GitHub
• Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 256, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
  (2): Normalize()
)

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 SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("Gurveer05/all-MiniLM-eedi-2024")
# Run inference
sentences = [
    'Construct:  Solve coordinate geometry questions involving ratio.\n\nQuestion:  A straight line on squared paper. Points P, Q and R lie on this line. The leftmost end of the line is labelled P. If you travel right 4 squares and up 1 square you get to point Q. If you then travel 8 squares right and 2 squares up from Q you reach point R. What is the ratio of  P Q: P R  ?\n\nOptions:\nA. 1: 12\nB. 1: 4\nC. 1: 2\nD. 1: 3\n\nCorrect Answer: 1: 3\n\nIncorrect Answer: 1: 2\n\nPredicted Misconception: Misunderstanding the ratio calculation by not considering the correct horizontal and vertical distances between points P, Q, and R.',
    'May have estimated when using ratios with geometry',
    'Thinks x = y is an axis',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

<!-- ### Direct Usage (Transformers) <details><summary>Click to see the direct usage in Transformers</summary> </details> --> <!-- ### Downstream Usage (Sentence Transformers) You can finetune this model on your own dataset. <details><summary>Click to expand</summary> </details> --> <!-- ### Out-of-Scope Use *List how the model may foreseeably be misused and address what users ought not to do with the model.* --> <!-- ## Bias, Risks and Limitations *What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model.* --> <!-- ### Recommendations *What are recommendations with respect to the foreseeable issues? For example, filtering explicit content.* -->

Training Details

Training Dataset

csv

• Dataset: csv
• Size: 12,210 training samples
• Columns: <code>qa_pair_text</code>, <code>MisconceptionName</code>, and <code>negative</code>
• Approximate statistics based on the first 1000 samples:

  	qa_pair_text	MisconceptionName	negative
  type	string	string	string
  details	<ul><li>min: 54 tokens</li><li>mean: 121.45 tokens</li><li>max: 256 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 15.16 tokens</li><li>max: 39 tokens</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 14.49 tokens</li><li>max: 40 tokens</li></ul>

• Samples:

  qa_pair_text	MisconceptionName	negative
  <code>Construct: Construct frequency tables.<br><br>Question: Dave has recorded the number of pets his classmates have in the frequency table on the right. <br> Number of pets	Frequency <br> 0	4 <br> 1
  <code>Construct: Convert between any other time periods.<br><br>Question: To work out how many hours in a year you could do...<br><br>Options:<br>A. 365 x 7<br>B. 365 x 60<br>C. 365 x 12<br>D. 365 x 24<br><br>Correct Answer: 365 x 24<br><br>Incorrect Answer: 365 x 60<br><br>Predicted Misconception: Multiplying days by hours per minute instead of hours per day.</code>	<code>Answers as if there are 60 hours in a day</code>	<code>Confuses an equation with an expression</code>
  <code>Construct: Given information about one part, work out other parts.<br><br>Question: Jess and Heena share some sweets in the ratio 3;: 5 .<br>Jess gets 15 sweets.<br>How many sweets does Heena get?<br><br>Options:<br>A. 17<br>B. 9<br>C. 5<br>D. 25<br><br>Correct Answer: 25<br><br>Incorrect Answer: 17<br><br>Predicted Misconception: Misunderstanding the direct proportionality between the ratio and actual quantities.</code>	<code>Thinks a difference of one part in a ratio means the quantities will differ by one unit</code>	<code>Believes dividing two positives will give a negative answer</code>

• Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

csv

• Dataset: csv
• Size: 9,640 evaluation samples
• Columns: <code>qa_pair_text</code>, <code>MisconceptionName</code>, and <code>negative</code>
• Approximate statistics based on the first 1000 samples:

  	qa_pair_text	MisconceptionName	negative
  type	string	string	string
  details	<ul><li>min: 56 tokens</li><li>mean: 119.35 tokens</li><li>max: 256 tokens</li></ul>	<ul><li>min: 6 tokens</li><li>mean: 14.51 tokens</li><li>max: 39 tokens</li></ul>	<ul><li>min: 6 tokens</li><li>mean: 13.86 tokens</li><li>max: 40 tokens</li></ul>

• Samples:

  qa_pair_text	MisconceptionName	negative
  <code>Construct: Identify when rounding a calculation will give an over or under approximation.<br><br>Question: Tom and Katie are discussing how to estimate the answer to<br>[<br>38.8745 / 7.9302<br>]<br><br>Tom says 40 / 7.9302 would give an overestimate.<br><br>Katie says 38.8745 / 8 would give an overestimate.<br><br>Who is correct?<br><br>Options:<br>A. Only Tom<br>B. Only Katie<br>C. Both Tom and Katie<br>D. Neither is correct<br><br>Correct Answer: Only Tom<br><br>Incorrect Answer: Neither is correct<br><br>Predicted Misconception: Rounding both numbers up leads to an overestimate.</code>	<code>Believes that the larger the dividend, the smaller the answer.</code>	<code>Does not know how to calculate the mean</code>
  <code>Construct: Substitute negative integer values into expressions involving no powers or roots.<br><br>Question: Amy is trying to work out the distance between these two points: (1,-6) and (-5,2) She labels them like this: x_1	y_1	x_2
  <code>Construct: Round numbers to three or more decimal places.<br><br>Question: What is 20.15349 rounded to 3 decimal places?<br><br>Options:<br>A. 20.153<br>B. 20.15<br>C. 20.154<br>D. 20.253<br><br>Correct Answer: 20.153<br><br>Incorrect Answer: 20.154<br><br>Predicted Misconception: Rounding up the fourth decimal place without considering the fifth decimal place.</code>	<code>Rounds up instead of down</code>	<code>When dividing decimals, does not realize that the order and position of the digits (relative to each other) has to remain constant.</code>

• Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• gradient_accumulation_steps: 8
• learning_rate: 1e-05
• weight_decay: 0.01
• num_train_epochs: 40
• lr_scheduler_type: cosine
• lr_scheduler_kwargs: {'num_cycles': 20}
• warmup_ratio: 0.1
• fp16: True
• load_best_model_at_end: True
• gradient_checkpointing: True
• gradient_checkpointing_kwargs: {'use_reentrant': False}
• batch_sampler: no_duplicates

All Hyperparameters

<details><summary>Click to expand</summary>

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 8
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 1e-05
• weight_decay: 0.01
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1.0
• num_train_epochs: 40
• max_steps: -1
• lr_scheduler_type: cosine
• lr_scheduler_kwargs: {'num_cycles': 20}
• warmup_ratio: 0.1
• 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: 42
• data_seed: None
• jit_mode_eval: False
• use_ipex: False
• bf16: False
• fp16: True
• 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: False
• hub_always_push: False
• gradient_checkpointing: True
• gradient_checkpointing_kwargs: {'use_reentrant': False}
• include_inputs_for_metrics: False
• 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
• dispatch_batches: None
• split_batches: 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
• eval_use_gather_object: False
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

</details>

Training Logs

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.14
• Sentence Transformers: 3.1.1
• Transformers: 4.44.0
• PyTorch: 2.4.0
• Accelerate: 0.33.0
• Datasets: 2.19.2
• Tokenizers: 0.19.1

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",
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply},
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}

<!-- ## Glossary *Clearly define terms in order to be accessible across audiences.* --> <!-- ## Model Card Authors *Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction.* --> <!-- ## Model Card Contact *Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors.* -->