inrealm commited on
Commit
6e223a4
1 Parent(s): ecb4267

Add new SentenceTransformer model.

Browse files
Files changed (2) hide show
  1. README.md +15 -6
  2. model.safetensors +1 -1
README.md CHANGED
@@ -11,9 +11,9 @@ tags:
11
  - sentence-similarity
12
  - feature-extraction
13
  - generated_from_trainer
14
- - dataset_size:10
15
- - loss:MultipleNegativesRankingLoss
16
  - dataset_size:1000
 
 
17
  widget:
18
  - source_sentence: A man is jumping unto his filthy bed.
19
  sentences:
@@ -154,7 +154,7 @@ You can finetune this model on your own dataset.
154
  #### sentence-transformers/all-nli
155
 
156
  * Dataset: [sentence-transformers/all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) at [d482672](https://huggingface.co/datasets/sentence-transformers/all-nli/tree/d482672c8e74ce18da116f430137434ba2e52fab)
157
- * Size: 1,000 training samples
158
  * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
159
  * Approximate statistics based on the first 1000 samples:
160
  | | anchor | positive | negative |
@@ -207,7 +207,8 @@ You can finetune this model on your own dataset.
207
  - `eval_strategy`: steps
208
  - `per_device_train_batch_size`: 16
209
  - `per_device_eval_batch_size`: 16
210
- - `num_train_epochs`: 1
 
211
  - `warmup_ratio`: 0.1
212
  - `fp16`: True
213
  - `batch_sampler`: no_duplicates
@@ -225,13 +226,13 @@ You can finetune this model on your own dataset.
225
  - `per_gpu_eval_batch_size`: None
226
  - `gradient_accumulation_steps`: 1
227
  - `eval_accumulation_steps`: None
228
- - `learning_rate`: 5e-05
229
  - `weight_decay`: 0.0
230
  - `adam_beta1`: 0.9
231
  - `adam_beta2`: 0.999
232
  - `adam_epsilon`: 1e-08
233
  - `max_grad_norm`: 1.0
234
- - `num_train_epochs`: 1
235
  - `max_steps`: -1
236
  - `lr_scheduler_type`: linear
237
  - `lr_scheduler_kwargs`: {}
@@ -326,6 +327,14 @@ You can finetune this model on your own dataset.
326
 
327
  </details>
328
 
 
 
 
 
 
 
 
 
329
  ### Framework Versions
330
  - Python: 3.10.12
331
  - Sentence Transformers: 3.0.1
 
11
  - sentence-similarity
12
  - feature-extraction
13
  - generated_from_trainer
 
 
14
  - dataset_size:1000
15
+ - loss:MultipleNegativesRankingLoss
16
+ - dataset_size:3000
17
  widget:
18
  - source_sentence: A man is jumping unto his filthy bed.
19
  sentences:
 
154
  #### sentence-transformers/all-nli
155
 
156
  * Dataset: [sentence-transformers/all-nli](https://huggingface.co/datasets/sentence-transformers/all-nli) at [d482672](https://huggingface.co/datasets/sentence-transformers/all-nli/tree/d482672c8e74ce18da116f430137434ba2e52fab)
157
+ * Size: 3,000 training samples
158
  * Columns: <code>anchor</code>, <code>positive</code>, and <code>negative</code>
159
  * Approximate statistics based on the first 1000 samples:
160
  | | anchor | positive | negative |
 
207
  - `eval_strategy`: steps
208
  - `per_device_train_batch_size`: 16
209
  - `per_device_eval_batch_size`: 16
210
+ - `learning_rate`: 2e-05
211
+ - `num_train_epochs`: 2
212
  - `warmup_ratio`: 0.1
213
  - `fp16`: True
214
  - `batch_sampler`: no_duplicates
 
226
  - `per_gpu_eval_batch_size`: None
227
  - `gradient_accumulation_steps`: 1
228
  - `eval_accumulation_steps`: None
229
+ - `learning_rate`: 2e-05
230
  - `weight_decay`: 0.0
231
  - `adam_beta1`: 0.9
232
  - `adam_beta2`: 0.999
233
  - `adam_epsilon`: 1e-08
234
  - `max_grad_norm`: 1.0
235
+ - `num_train_epochs`: 2
236
  - `max_steps`: -1
237
  - `lr_scheduler_type`: linear
238
  - `lr_scheduler_kwargs`: {}
 
327
 
328
  </details>
329
 
330
+ ### Training Logs
331
+ | Epoch | Step | Training Loss | loss |
332
+ |:------:|:----:|:-------------:|:------:|
333
+ | 0.5319 | 100 | 0.7969 | 0.4318 |
334
+ | 1.0638 | 200 | 0.2888 | 0.4764 |
335
+ | 1.5957 | 300 | 0.025 | 0.5072 |
336
+
337
+
338
  ### Framework Versions
339
  - Python: 3.10.12
340
  - Sentence Transformers: 3.0.1
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c58edc027b706578cdb9f3591e7bcb0c1c1f2ba9667d409a8efe0beeaad15c53
3
  size 133462128
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f18b6ef5ceedc19b9d9a0c129da93277a3bfa50e68f63169ff0952705535170
3
  size 133462128