This is a sentence-transformers model finetuned from google/bert_uncased_L-2_H-128_A-2. It maps sentences & paragraphs to a 128-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: google/bert_uncased_L-2_H-128_A-2
Maximum Sequence Length: 128 tokens
Output Dimensionality: 128 dimensions
Similarity Function: Cosine Similarity
Supported Modality: Text

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 128, 'pooling_mode': 'mean', 'include_prompt': True})
)

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.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("swardiantara/bert-tiny-amazon_reviews-k10-adaptive-cosine")
# Run inference
sentences = [
    "Arrived broken. Manufacturer defect. Two of the legs of the base were not completely formed, so there was no way to insert the casters. I unpackaged the entire chair and hardware before noticing this. So, I'll spend twice the amount of time boxing up the whole useless thing and send it back with a 1-star review of part of a chair I never got to sit in. I will go so far as to include a picture of what their injection molding and quality assurance process missed though. I will be hesitant to buy again. It makes me wonder if there aren't missing structures and supports that don't impede the assembly process.",
    'I like the design of the bag, though it came a little crumpled up. Hoping everything smoothes out with use. Also very upset that I paid more for this bag and my key chain was never sent.',
    'Had to return. This unit is tiny and very cheap quality.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 128]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[ 1.0000, -0.1525,  0.4489],
#         [-0.1525,  1.0000, -0.0192],
#         [ 0.4489, -0.0192,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

Size: 8,201,450 training samples
Columns: text_a, text_b, and label
Approximate statistics based on the first 100 samples:
text_a text_b label
type string string list
modality text text
details
min: 15 tokens
mean: 66.74 tokens
max: 128 tokens
min: 8 tokens
mean: 46.15 tokens
max: 128 tokens
size: 2 elements

Samples:

text_a	text_b	label
Arrived broken. Manufacturer defect. Two of the legs of the base were not completely formed, so there was no way to insert the casters. I unpackaged the entire chair and hardware before noticing this. So, I'll spend twice the amount of time boxing up the whole useless thing and send it back with a 1-star review of part of a chair I never got to sit in. I will go so far as to include a picture of what their injection molding and quality assurance process missed though. I will be hesitant to buy again. It makes me wonder if there aren't missing structures and supports that don't impede the assembly process.	`I ordered this a while ago, even using prime it took over a month to get here, now that I finally got it, I only receved 6 of the 12! Im a little upset. It's still funny, but would not recomend for someone who needs foam quick and in bulk like i do.`	`[1.0, 0.0]`
Arrived broken. Manufacturer defect. Two of the legs of the base were not completely formed, so there was no way to insert the casters. I unpackaged the entire chair and hardware before noticing this. So, I'll spend twice the amount of time boxing up the whole useless thing and send it back with a 1-star review of part of a chair I never got to sit in. I will go so far as to include a picture of what their injection molding and quality assurance process missed though. I will be hesitant to buy again. It makes me wonder if there aren't missing structures and supports that don't impede the assembly process.	`I was happy to find this piece to replace on my broken Shark. Price was good and had it in a few days. When I installed it however, I was no longer satisfied. It wouldn't stay attached and kept popping of with any pull on the handle. Returned the same day it areived.`	`[0.0, 0.25]`
Arrived broken. Manufacturer defect. Two of the legs of the base were not completely formed, so there was no way to insert the casters. I unpackaged the entire chair and hardware before noticing this. So, I'll spend twice the amount of time boxing up the whole useless thing and send it back with a 1-star review of part of a chair I never got to sit in. I will go so far as to include a picture of what their injection molding and quality assurance process missed though. I will be hesitant to buy again. It makes me wonder if there aren't missing structures and supports that don't impede the assembly process.	`Had to return. This unit is tiny and very cheap quality.`	`[0.0, 0.25]`

Loss: [object Object].OrdinalProxyContrastiveLoss

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 1024
num_train_epochs: 10
learning_rate: 2e-05
load_best_model_at_end: True

All Hyperparameters

Click to expand

per_device_train_batch_size: 1024
num_train_epochs: 10
max_steps: -1
learning_rate: 2e-05
lr_scheduler_type: linear
lr_scheduler_kwargs: None
warmup_steps: 0
optim: adamw_torch
optim_args: None
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
optim_target_modules: None
gradient_accumulation_steps: 1
average_tokens_across_devices: True
max_grad_norm: 1.0
label_smoothing_factor: 0.0
bf16: False
fp16: False
bf16_full_eval: False
fp16_full_eval: False
tf32: None
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
use_liger_kernel: False
liger_kernel_config: None
use_cache: False
neftune_noise_alpha: None
torch_empty_cache_steps: None
auto_find_batch_size: False
log_on_each_node: True
logging_nan_inf_filter: True
include_num_input_tokens_seen: no
log_level: passive
log_level_replica: warning
disable_tqdm: False
project: huggingface
trackio_space_id: None
trackio_bucket_id: None
trackio_static_space_id: None
per_device_eval_batch_size: 8
prediction_loss_only: True
eval_on_start: False
eval_do_concat_batches: True
eval_use_gather_object: False
eval_accumulation_steps: None
include_for_metrics: []
batch_eval_metrics: False
save_only_model: False
save_on_each_node: False
enable_jit_checkpoint: False
push_to_hub: False
hub_private_repo: None
hub_model_id: None
hub_strategy: every_save
hub_always_push: False
hub_revision: None
load_best_model_at_end: True
ignore_data_skip: False
restore_callback_states_from_checkpoint: False
full_determinism: False
seed: 42
data_seed: None
use_cpu: False
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
parallelism_config: None
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_pin_memory: True
dataloader_persistent_workers: False
dataloader_prefetch_factor: None
remove_unused_columns: True
label_names: None
train_sampling_strategy: random
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
ddp_static_graph: None
ddp_backend: None
ddp_timeout: 1800
fsdp: None
fsdp_config: None
deepspeed: None
debug: []
skip_memory_metrics: True
do_predict: False
resume_from_checkpoint: None
warmup_ratio: None
local_rank: -1
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
0.0624	500	0.0198
0.1248	1000	0.0086
0.1873	1500	0.0073
0.2497	2000	0.0067
0.3121	2500	0.0063
0.3745	3000	0.0061
0.4370	3500	0.0058
0.4994	4000	0.0056
0.5618	4500	0.0054
0.6242	5000	0.0054
0.6866	5500	0.0052
0.7491	6000	0.0051
0.8115	6500	0.0051
0.8739	7000	0.0050
0.9363	7500	0.0050
0.9988	8000	0.0049
1.0	8010	-
1.0612	8500	0.0047
1.1236	9000	0.0047
1.1860	9500	0.0047
1.2484	10000	0.0047
1.3109	10500	0.0046
1.3733	11000	0.0046
1.4357	11500	0.0045
1.4981	12000	0.0045
1.5605	12500	0.0044
1.6230	13000	0.0044
1.6854	13500	0.0044
1.7478	14000	0.0043
1.8102	14500	0.0043
1.8727	15000	0.0042
1.9351	15500	0.0043
1.9975	16000	0.0043
2.0	16020	-
2.0599	16500	0.0041
2.1223	17000	0.0042
2.1848	17500	0.0041
2.2472	18000	0.0041
2.3096	18500	0.0040
2.3720	19000	0.0040
2.4345	19500	0.0040
2.4969	20000	0.0040
2.5593	20500	0.0041
2.6217	21000	0.0040
2.6841	21500	0.0040
2.7466	22000	0.0040
2.8090	22500	0.0039
2.8714	23000	0.0039
2.9338	23500	0.0040
2.9963	24000	0.0039
3.0	24030	-
3.0587	24500	0.0038
3.1211	25000	0.0038
3.1835	25500	0.0038
3.2459	26000	0.0038
3.3084	26500	0.0038
3.3708	27000	0.0037
3.4332	27500	0.0037
3.4956	28000	0.0037
3.5581	28500	0.0038
3.6205	29000	0.0037
3.6829	29500	0.0037
3.7453	30000	0.0038
3.8077	30500	0.0037
3.8702	31000	0.0037
3.9326	31500	0.0036
3.9950	32000	0.0036
4.0	32040	-
4.0574	32500	0.0036
4.1199	33000	0.0035
4.1823	33500	0.0036
4.2447	34000	0.0035
4.3071	34500	0.0036
4.3695	35000	0.0036
4.4320	35500	0.0035
4.4944	36000	0.0036
4.5568	36500	0.0035
4.6192	37000	0.0035
4.6816	37500	0.0035
4.7441	38000	0.0035
4.8065	38500	0.0035
4.8689	39000	0.0035
4.9313	39500	0.0035
4.9938	40000	0.0035
5.0	40050	-
5.0562	40500	0.0034
5.1186	41000	0.0034
5.1810	41500	0.0034
5.2434	42000	0.0034
5.3059	42500	0.0035
5.3683	43000	0.0034
5.4307	43500	0.0034
5.4931	44000	0.0034
5.5556	44500	0.0034
5.6180	45000	0.0034
5.6804	45500	0.0033
5.7428	46000	0.0033
5.8052	46500	0.0034
5.8677	47000	0.0033
5.9301	47500	0.0034
5.9925	48000	0.0033
6.0	48060	-
6.0549	48500	0.0033
6.1174	49000	0.0033
6.1798	49500	0.0033
6.2422	50000	0.0033
6.3046	50500	0.0033
6.3670	51000	0.0033
6.4295	51500	0.0033
6.4919	52000	0.0033
6.5543	52500	0.0032
6.6167	53000	0.0032
6.6792	53500	0.0032
6.7416	54000	0.0032
6.8040	54500	0.0033
6.8664	55000	0.0033
6.9288	55500	0.0033
6.9913	56000	0.0032
7.0	56070	-
7.0537	56500	0.0033
7.1161	57000	0.0033
7.1785	57500	0.0032
7.2409	58000	0.0032
7.3034	58500	0.0032
7.3658	59000	0.0031
7.4282	59500	0.0032
7.4906	60000	0.0032
7.5531	60500	0.0032
7.6155	61000	0.0032
7.6779	61500	0.0032
7.7403	62000	0.0032
7.8027	62500	0.0032
7.8652	63000	0.0032
7.9276	63500	0.0032
7.9900	64000	0.0031
8.0	64080	-

The bold row denotes the saved checkpoint.

Training Time

Training: 4.3 hours
Evaluation: 35.5 seconds
Total: 4.3 hours

Framework Versions

Python: 3.12.4
Sentence Transformers: 5.5.1
Transformers: 5.11.0
PyTorch: 2.5.1+cu121
Accelerate: 1.13.0
Datasets: 2.21.0
Tokenizers: 0.22.2

Citation

BibTeX

Sentence Transformers

bibtex

@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",
}

	text_a	text_b	label
type	string	string	list
modality	text	text
details	min: 15 tokens mean: 66.74 tokens max: 128 tokens	min: 8 tokens mean: 46.15 tokens max: 128 tokens	size: 2 elements