riaz.somc/transformers
1
0
Fork 0
mirror of https://github.com/huggingface/transformers.git synced 2025-07-31 18:22:34 +06:00
Code Issues Actions 4 Packages Projects Releases Wiki Activity

fix docs

Browse Source
This commit is contained in:
oweller2 2025-07-02 14:13:58 -04:00
parent 25895f7841
commit 07f5a1943e
2 changed files with 12 additions and 6 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

9
src/transformers/models/modernbert_decoder/modeling_modernbert_decoder.py
View File

@ -545,6 +545,9 @@ class ModernBertDecoderForCausalLM(ModernBertDecoderPreTrainedModel, GenerationM
(masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
Returns:
[`~modeling_outputs.CausalLMOutputWithPast`] or `tuple(torch.FloatTensor)`: A
[`~modeling_outputs.CausalLMOutputWithPast`] or a tuple of `torch.FloatTensor` (if `return_dict=False`)
comprising various elements depending on the configuration and inputs.
Example:
@ -554,13 +557,13 @@ class ModernBertDecoderForCausalLM(ModernBertDecoderPreTrainedModel, GenerationM
>>> model = ModernBertDecoderForCausalLM.from_pretrained("blab-jhu/test-32m-dec")
>>> tokenizer = AutoTokenizer.from_pretrained("blab-jhu/test-32m-dec")
>>> prompt = "Hello, I'm a language model,"
>>> prompt = "The capital of France is"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> generate_ids = model.generate(inputs.input_ids, max_length=1)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hello, I'm a language model, and I'm here to help you with your questions."
"The capital of France is Paris"
```
"""
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions

9
src/transformers/models/modernbert_decoder/modular_modernbert_decoder.py
View File

@ -735,6 +735,9 @@ class ModernBertDecoderForCausalLM(ModernBertDecoderPreTrainedModel, GenerationM
(masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
Returns:
[`~modeling_outputs.CausalLMOutputWithPast`] or `tuple(torch.FloatTensor)`: A
[`~modeling_outputs.CausalLMOutputWithPast`] or a tuple of `torch.FloatTensor` (if `return_dict=False`)
comprising various elements depending on the configuration and inputs.
Example:
@ -744,13 +747,13 @@ class ModernBertDecoderForCausalLM(ModernBertDecoderPreTrainedModel, GenerationM
>>> model = ModernBertDecoderForCausalLM.from_pretrained("blab-jhu/test-32m-dec")
>>> tokenizer = AutoTokenizer.from_pretrained("blab-jhu/test-32m-dec")
>>> prompt = "Hello, I'm a language model,"
>>> prompt = "The capital of France is"
>>> inputs = tokenizer(prompt, return_tensors="pt")
>>> # Generate
>>> generate_ids = model.generate(inputs.input_ids, max_length=30)
>>> generate_ids = model.generate(inputs.input_ids, max_length=1)
>>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
"Hello, I'm a language model, and I'm here to help you with your questions."
"The capital of France is Paris"
```
"""
output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions