CMU-Wav2Gloss/Gitksan-encoder-bsz128-e1k-bsz32-e1k

SentenceTransformer based on sentence-transformers/LaBSE

This is a sentence-transformers model finetuned from sentence-transformers/LaBSE. It maps sentences & paragraphs to a 768-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/LaBSE <!-- at revision 836121a0533e5664b21c7aacc5d22951f2b8b25b -->
• —Maximum Sequence Length: 256 tokens
• —Output Dimensionality: 768 dimensions
• —Similarity Function: Cosine Similarity <!-- - Training Dataset: Unknown --> <!-- - 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, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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): Dense({'in_features': 768, 'out_features': 768, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
  (3): 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("sentence_transformers_model_id")
# Run inference
sentences = [
    "Word: a l o o h l| Context: I i   ' n i i   y a t s h l   p i p e - -   n i i   g y a ' a   w i l   s g i h l   p i p e   a l o o h l   h a ' n i i y o ' o x s x w .| Translation: And I hit the pipe-- I saw there was a pipe on the sink.",
    'Morpheme: h l | Gloss: CN',
    'Morpheme: i i | Gloss: CCNJ',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.6834, 0.3528],
#         [0.6834, 1.0000, 0.4257],
#         [0.3528, 0.4257, 1.0000]])

<!--

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. -->

Evaluation

Metrics

IR

• —Dataset: validation
• —Evaluated with <code>_main_.IREvaluatorWithLogging</code>

<!--

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

Unnamed Dataset

• —Size: 429 training samples
• —Columns: <code>sentence0</code>, <code>sentence1</code>, and <code>label</code>
• —Approximate statistics based on the first 429 samples: | | sentence0 | sentence1 | label | |:--------|:------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | type | string | string | int | | details | <ul><li>min: 40 tokens</li><li>mean: 84.96 tokens</li><li>max: 131 tokens</li></ul> | <ul><li>min: 11 tokens</li><li>mean: 14.11 tokens</li><li>max: 20 tokens</li></ul> | <ul><li>0: ~0.47%</li><li>1: ~0.47%</li><li>2: ~0.23%</li><li>3: ~0.47%</li><li>4: ~0.47%</li><li>5: ~0.23%</li><li>6: ~0.47%</li><li>7: ~0.23%</li><li>8: ~0.47%</li><li>9: ~0.23%</li><li>10: ~0.23%</li><li>11: ~0.23%</li><li>12: ~0.47%</li><li>13: ~0.47%</li><li>14: ~0.47%</li><li>15: ~0.23%</li><li>16: ~0.93%</li><li>17: ~0.23%</li><li>18: ~0.47%</li><li>19: ~0.47%</li><li>20: ~0.23%</li><li>21: ~0.23%</li><li>22: ~0.23%</li><li>23: ~0.47%</li><li>24: ~0.23%</li><li>25: ~0.23%</li><li>26: ~0.23%</li><li>27: ~0.47%</li><li>28: ~1.17%</li><li>29: ~0.47%</li><li>30: ~0.47%</li><li>31: ~0.47%</li><li>32: ~0.23%</li><li>33: ~0.23%</li><li>34: ~0.70%</li><li>35: ~0.23%</li><li>36: ~0.23%</li><li>37: ~0.23%</li><li>38: ~0.23%</li><li>39: ~0.70%</li><li>40: ~0.23%</li><li>41: ~0.70%</li><li>42: ~0.47%</li><li>43: ~0.23%</li><li>44: ~0.23%</li><li>45: ~0.23%</li><li>46: ~0.47%</li><li>47: ~0.23%</li><li>48: ~0.23%</li><li>49: ~0.47%</li><li>50: ~0.47%</li><li>51: ~0.23%</li><li>52: ~0.23%</li><li>53: ~0.23%</li><li>54: ~0.47%</li><li>55: ~0.47%</li><li>56: ~0.23%</li><li>57: ~0.23%</li><li>58: ~0.47%</li><li>59: ~0.23%</li><li>60: ~0.47%</li><li>61: ~0.23%</li><li>62: ~0.47%</li><li>63: ~0.47%</li><li>64: ~0.23%</li><li>65: ~0.23%</li><li>66: ~0.47%</li><li>67: ~0.47%</li><li>68: ~0.70%</li><li>69: ~0.47%</li><li>70: ~0.47%</li><li>71: ~0.23%</li><li>72: ~0.47%</li><li>73: ~0.47%</li><li>74: ~0.70%</li><li>75: ~0.23%</li><li>76: ~0.47%</li><li>77: ~0.70%</li><li>78: ~0.23%</li><li>79: ~0.70%</li><li>80: ~0.23%</li><li>81: ~0.23%</li><li>82: ~0.47%</li><li>83: ~0.23%</li><li>84: ~0.47%</li><li>85: ~0.47%</li><li>86: ~0.47%</li><li>87: ~0.47%</li><li>88: ~0.23%</li><li>89: ~0.23%</li><li>90: ~0.47%</li><li>91: ~0.23%</li><li>92: ~0.47%</li><li>93: ~0.23%</li><li>94: ~0.23%</li><li>95: ~0.47%</li><li>96: ~0.47%</li><li>97: ~0.23%</li><li>98: ~0.23%</li><li>99: ~0.23%</li><li>100: ~0.70%</li><li>101: ~0.47%</li><li>102: ~0.23%</li><li>103: ~0.47%</li><li>104: ~0.70%</li><li>105: ~0.23%</li><li>106: ~0.23%</li><li>107: ~0.23%</li><li>108: ~0.47%</li><li>109: ~0.23%</li><li>110: ~0.47%</li><li>111: ~0.23%</li><li>112: ~0.47%</li><li>113: ~0.23%</li><li>114: ~0.47%</li><li>115: ~0.23%</li><li>116: ~0.23%</li><li>117: ~0.23%</li><li>118: ~0.70%</li><li>119: ~0.47%</li><li>120: ~0.23%</li><li>121: ~0.23%</li><li>122: ~0.47%</li><li>123: ~0.70%</li><li>124: ~0.23%</li><li>125: ~0.47%</li><li>126: ~0.23%</li><li>127: ~0.23%</li><li>128: ~0.23%</li><li>129: ~0.47%</li><li>130: ~0.23%</li><li>131: ~0.70%</li><li>132: ~0.47%</li><li>133: ~0.23%</li><li>134: ~0.23%</li><li>135: ~0.47%</li><li>136: ~0.23%</li><li>137: ~0.23%</li><li>138: ~0.47%</li><li>139: ~0.23%</li><li>140: ~0.47%</li><li>141: ~0.23%</li><li>142: ~0.23%</li><li>143: ~0.47%</li><li>144: ~0.23%</li><li>145: ~0.70%</li><li>146: ~0.93%</li><li>147: ~0.47%</li><li>148: ~0.23%</li><li>149: ~0.47%</li><li>150: ~0.47%</li><li>151: ~0.47%</li><li>152: ~0.23%</li><li>153: ~0.47%</li><li>154: ~0.47%</li><li>155: ~0.23%</li><li>156: ~0.23%</li><li>157: ~0.47%</li><li>158: ~0.47%</li><li>159: ~0.23%</li><li>160: ~0.23%</li><li>161: ~0.70%</li><li>162: ~0.23%</li><li>163: ~0.23%</li><li>164: ~0.47%</li><li>165: ~0.47%</li><li>166: ~0.93%</li><li>167: ~0.23%</li><li>168: ~0.47%</li><li>169: ~0.70%</li><li>170: ~0.23%</li><li>171: ~0.23%</li><li>172: ~0.47%</li><li>173: ~0.23%</li><li>174: ~0.47%</li><li>175: ~0.70%</li><li>176: ~0.23%</li><li>177: ~0.23%</li><li>178: ~0.23%</li><li>179: ~0.47%</li><li>180: ~0.47%</li><li>181: ~0.47%</li><li>182: ~0.23%</li><li>183: ~0.23%</li><li>184: ~0.47%</li><li>185: ~0.23%</li><li>186: ~0.23%</li><li>187: ~0.70%</li><li>188: ~0.70%</li><li>189: ~0.23%</li><li>190: ~0.47%</li><li>191: ~0.23%</li><li>192: ~0.23%</li><li>193: ~0.70%</li><li>194: ~0.23%</li><li>195: ~0.23%</li><li>196: ~0.47%</li><li>197: ~0.23%</li><li>198: ~0.47%</li><li>199: ~0.47%</li><li>200: ~0.23%</li><li>201: ~0.23%</li><li>202: ~0.23%</li><li>203: ~0.47%</li><li>204: ~0.47%</li><li>205: ~0.23%</li><li>206: ~0.47%</li><li>207: ~0.23%</li><li>208: ~0.23%</li><li>209: ~0.47%</li><li>210: ~0.70%</li><li>211: ~0.47%</li><li>212: ~0.47%</li><li>213: ~0.47%</li><li>214: ~0.23%</li><li>215: ~0.23%</li><li>216: ~0.47%</li><li>217: ~0.47%</li><li>218: ~0.23%</li><li>219: ~0.23%</li><li>220: ~0.23%</li><li>221: ~0.23%</li><li>222: ~0.23%</li><li>223: ~0.70%</li><li>224: ~0.23%</li><li>225: ~0.47%</li><li>226: ~0.47%</li><li>227: ~0.23%</li><li>228: ~0.70%</li><li>229: ~0.47%</li><li>230: ~0.47%</li><li>231: ~0.23%</li><li>232: ~0.70%</li><li>233: ~0.70%</li><li>234: ~0.47%</li><li>235: ~0.23%</li><li>236: ~0.23%</li><li>237: ~0.23%</li><li>238: ~0.23%</li><li>239: ~0.47%</li><li>240: ~0.47%</li><li>241: ~0.23%</li><li>242: ~0.93%</li><li>243: ~0.47%</li><li>244: ~0.23%</li><li>245: ~0.70%</li><li>246: ~0.23%</li><li>247: ~0.70%</li><li>248: ~0.47%</li><li>249: ~0.23%</li><li>250: ~0.47%</li><li>251: ~0.23%</li><li>252: ~0.23%</li><li>253: ~0.23%</li><li>254: ~0.23%</li><li>255: ~0.47%</li><li>256: ~0.47%</li><li>257: ~0.70%</li><li>258: ~0.23%</li><li>259: ~0.23%</li></ul> |
• —Samples: | sentence0 | sentence1 | label | |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------|:-----------------| | <code>Word: h l a g ̲ o o k ̲\| Context: I i h l a g ̲ o o k ̲ d i m h a ' w i ' y i i k y ' a a i s x w i ' y g ̲ o o h l w i l p x s e e k ̲ .\| Translation: And before I went home I had a short pee in the bathroom.</code> | <code>Morpheme: g ̲ o o k ̲ \| Gloss: first</code> | <code>90</code> | | <code>Word: x s a ' a k ̲ x w i ' y\| Context: H l a a x s a ' a k ̲ x w i ' y ' n i i g ̲ a y o o t s ' i m i l t ' a a h l i h l j a b i ' y g ̲ o o h l t s ' i m w i l p x s e e k ̲ .\| Translation: When I made it out, then I put what I had done (the rubble) back in the bathroom.</code> | <code>Morpheme: x s i \| Gloss: out</code> | <code>228</code> | | <code>Word: n e e d i i\| Context: I i ' n a k w h l ' w i h l w i l i ' m , g w i l a ' l h l g ̲ a n u u t x w , g ̲ a n w i h l n e e d i i l a x ̲ ' n i s x w i ' y g ̲ o o h l G i g e e n i x .\| Translation: And we were away a long time, three weeks, and that's why I didn't hear from Gigeenix.</code> | <code>Morpheme: n e e \| Gloss: NEG</code> | <code>67</code> |
• —Loss: <code>_main_.LossLogger</code>

Training Hyperparameters

Non-Default Hyperparameters

• —eval_strategy: steps
• —per_device_train_batch_size: 128
• —per_device_eval_batch_size: 128
• —num_train_epochs: 1000
• —fp16: True
• —multi_dataset_batch_sampler: round_robin

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: 128
• —per_device_eval_batch_size: 128
• —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: 5e-05
• —weight_decay: 0.0
• —adam_beta1: 0.9
• —adam_beta2: 0.999
• —adam_epsilon: 1e-08
• —max_grad_norm: 1
• —num_train_epochs: 1000
• —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: 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: False
• —ignore_data_skip: False
• —fsdp: []
• —fsdp_min_num_params: 0
• —fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
• —fsdp_transformer_layer_cls_to_wrap: None
• —accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
• —parallelism_config: None
• —deepspeed: None
• —label_smoothing_factor: 0.0
• —optim: adamwtorchfused
• —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: round_robin
• —router_mapping: {}
• —learning_rate_mapping: {}

</details>

Training Logs

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

</details>

Framework Versions

• —Python: 3.11.4
• —Sentence Transformers: 5.1.1
• —Transformers: 4.56.2
• —PyTorch: 2.8.0+cu128
• —Accelerate: 1.10.1
• —Datasets: 4.1.1
• —Tokenizers: 0.22.1

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",
}

<!--

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. -->