Add new SentenceTransformer model.

README.md

@@ -12,31 +12,31 @@ tags:
 - dataset_size:200
 - loss:SoftmaxLoss
 widget:
-- source_sentence: '  I am anxious about relying on AI for critical decisions,'
+- source_sentence: '  I worry that AI will eventually replace my job,'
   sentences:
-  - '  I worry about the ethical implications of using AI,'
-  - '  AIs data cleaning quality is impressive,'
-  - '  The rise of AI creates uncertainty in job markets,'
-- source_sentence: '  Im apprehensive about having AI manage sensitive tasks for me,'
+  - '  AI might affect the stability of job markets globally,'
+  - '  AI can mimic human facial expressions,'
+  - '  AIs flexibility in updating its models is remarkable,'
+- source_sentence: '  AIs efficiency in data mining is impressive,'
   sentences:
-  - '  AIs quality in image recognition is superb,'
-  - '  AI might replace jobs in data entry and processing,'
-  - '  AI tools improve my efficiency in complex computations,'
-- source_sentence: '  I am hesitant to rely on AI for financial advice,'
+  - '  AI could reduce the need for human intervention in many fields,'
+  - '  AI will possibly replace customer service jobs,'
+  - '  AI systems modify their processes based on feedback,'
+- source_sentence: '  AI can adjust to user preferences over time,'
   sentences:
-  - '  AI aids in streamlining industry-specific tasks,'
-  - '  AI can imitate human decision-making processes,'
-  - '  The high quality of AI in voice synthesis is staggering,'
-- source_sentence: '  AI systems modify their processes based on feedback,'
+  - ' AI can simulate human emotions in interactions,'
+  - '  AIs precision in data analysis is highly reliable,'
+  - '  The rapid growth of AI makes me uneasy,'
+- source_sentence: '  AI tools optimize my workflow, making tasks simpler,'
   sentences:
-  - '  AI can hold a relevant and coherent conversation,'
-  - '  Relying on AI for critical thinking tasks makes me nervous,'
-  - '  AI ensures high quality in predictive analytics,'
-- source_sentence: '  AI simplifies complex data analysis tasks for me,'
+  - '  AI tools aid significantly in project execution,'
+  - '  AI can carry on meaningful conversations over the phone,'
+  - '  AI helps me make better decisions through data analysis,'
+- source_sentence: '  AI effectively enhances my problem-solving strategies,'
   sentences:
+  - '  AI consistently provides reliable recommendations,'
+  - '  AI tools support me in delivering better results,'
   - '  AI optimizes content strategy through data analysis,'
-  - '  Im apprehensive about having AI manage sensitive tasks for me,'
-  - '  AI adapts to varying user inputs accurately,'
 ---
 
 # SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2
@@ -89,9 +89,9 @@ from sentence_transformers import SentenceTransformer
 model = SentenceTransformer("zihoo/all-MiniLM-L6-v2-AINLI")
 # Run inference
 sentences = [
-    '  AI simplifies complex data analysis tasks for me,',
+    '  AI effectively enhances my problem-solving strategies,',
+    '  AI consistently provides reliable recommendations,',
     '  AI optimizes content strategy through data analysis,',
-    '  AI adapts to varying user inputs accurately,',
 ]
 embeddings = model.encode(sentences)
 print(embeddings.shape)
@@ -147,21 +147,26 @@ You can finetune this model on your own dataset.
 
 
 * Size: 200 training samples
-* Columns: <code>sentence1</code>, <code>sentence2</code>, and <code>label</code>
+* Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
 * Approximate statistics based on the first 1000 samples:
-  |         | sentence1                                                                         | sentence2                                                                         | label                                                             |
+  |         | sentence_0                                                                        | sentence_1                                                                        | label                                                             |
   |:--------|:----------------------------------------------------------------------------------|:----------------------------------------------------------------------------------|:------------------------------------------------------------------|
   | type    | string                                                                            | string                                                                            | int                                                               |
   | details | <ul><li>min: 8 tokens</li><li>mean: 11.91 tokens</li><li>max: 16 tokens</li></ul> | <ul><li>min: 8 tokens</li><li>mean: 11.91 tokens</li><li>max: 16 tokens</li></ul> | <ul><li>0: ~9.00%</li><li>1: ~69.00%</li><li>2: ~22.00%</li></ul> |
 * Samples:
-  | sentence1                                                         | sentence2                                                         | label          |
-  |:------------------------------------------------------------------|:------------------------------------------------------------------|:---------------|
-  | <code> AI can simulate human emotions in interactions,</code>     | <code>  AI-generated voices sound remarkably human,</code>        | <code>1</code> |
-  | <code>  AI can adjust to user preferences over time,</code>       | <code>  AIs precision in data analysis is highly reliable,</code> | <code>1</code> |
-  | <code>  AI consistently provides reliable recommendations,</code> | <code>  AI tools support me in delivering better results,</code>  | <code>0</code> |
+  | sentence_0                                                                      | sentence_1                                                                      | label          |
+  |:--------------------------------------------------------------------------------|:--------------------------------------------------------------------------------|:---------------|
+  | <code>  AI systems autonomously integrate user feedback for improvements</code> | <code>  AI systems autonomously integrate user feedback for improvements</code> | <code>0</code> |
+  | <code>  AI can exhibit empathy in certain contexts,</code>                      | <code>  The quality of AI in customer feedback analysis is notable,</code>      | <code>2</code> |
+  | <code>  I feel stressed about using AI in professional settings,</code>         | <code>  I feel tense dealing with advanced AI technologies,</code>              | <code>0</code> |
 * Loss: [<code>SoftmaxLoss</code>](https://sbert.net/docs/package_reference/sentence_transformer/losses.html#softmaxloss)
 
 ### Training Hyperparameters
+#### Non-Default Hyperparameters
+
+- `per_device_train_batch_size`: 64
+- `per_device_eval_batch_size`: 64
+- `multi_dataset_batch_sampler`: round_robin
 
 #### All Hyperparameters
 <details><summary>Click to expand</summary>
@@ -170,8 +175,8 @@ You can finetune this model on your own dataset.
 - `do_predict`: False
 - `eval_strategy`: no
 - `prediction_loss_only`: True
-- `per_device_train_batch_size`: 8
-- `per_device_eval_batch_size`: 8
+- `per_device_train_batch_size`: 64
+- `per_device_eval_batch_size`: 64
 - `per_gpu_train_batch_size`: None
 - `per_gpu_eval_batch_size`: None
 - `gradient_accumulation_steps`: 1
@@ -181,8 +186,8 @@ You can finetune this model on your own dataset.
 - `adam_beta1`: 0.9
 - `adam_beta2`: 0.999
 - `adam_epsilon`: 1e-08
-- `max_grad_norm`: 1.0
-- `num_train_epochs`: 3.0
+- `max_grad_norm`: 1
+- `num_train_epochs`: 3
 - `max_steps`: -1
 - `lr_scheduler_type`: linear
 - `lr_scheduler_kwargs`: {}
@@ -274,7 +279,7 @@ You can finetune this model on your own dataset.
 - `batch_eval_metrics`: False
 - `eval_on_start`: False
 - `batch_sampler`: batch_sampler
-- `multi_dataset_batch_sampler`: proportional
+- `multi_dataset_batch_sampler`: round_robin
 
 </details>
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8242933065fcf9ac1f32afd943cf9fb365b4897b90ee955e9eda0780a55ba69a
+oid sha256:8d11d89d38b73dc9bbee44dd082835092283a177546e02e2ff87748ef6863ac0
 size 90864192