keras-team / keras-nlp

Modular Natural Language Processing workflows with Keras

keras-team/keras-nlp Issues

Any plans for Llama 3?
Closed 12 days ago
Preprocessor does not respect sequence_length
Closed 12 days ago2
Distributed batch size not calculated correctly
Updated 13 days ago2
Documented `id_to_token` doesn't exist for UnicodeCodepointTokenizer
Updated 14 days ago
Cannot export a slightly customized XLMRoberta model from keras_nlp
Updated 14 days ago
Add support for `PaliGemma`
Updated 14 days ago
DebertaV3MaskedLM example don't work
Updated 14 days ago1
unable to diagnose OOM
Updated 14 days ago
make it easier to adjust dropout when loading gemma models
Updated 16 days ago
403 KaggleApiHTTPError while running GemmaCausalLM
Closed 20 days ago2
Any plans for more Llama type models?
Updated 22 days ago1
Issue instantiating a keras_nlp.models.Backbone from a model preset of Hugging Face handles
Updated 22 days ago4
Gemma Model Storing and Loading after Fine tuning
Updated 24 days ago4
GemmaBackbone.get_layout_map broken for gemma_2b_en
Updated 24 days ago2
Issue when fine-tuning Albert - Resource localhost/_0_SentencepieceOp/N10tensorflow4text12_GLOBAL__N_121SentencepieceResourceE does not exist.
Updated a month ago3
i installed keras-nlp in pycharm IDE , when in run
Closed a month ago8
Cannot reproduce results from notebook on Colab
Updated a month ago3
keras-nlp insists I use the (buggy) Tensorflow 2.16.1 which does not work with my GPU
Closed a month ago12
[RfC] Ideas for better Hugging Face Hub integration
Closed a month ago7
Any plans for QLora?
Updated a month ago2
Update ByteTokenizer to remove TensorFlow dependency
Updated a month ago1
cannot import name 'CachedMultiHeadAttention' from partially initialized module 'keras_nlp.src.layers.modeling.cached_multi_head_attention' (most likely due to a circular import)
Updated a month ago
Samplers in Gemma model
Updated a month ago6
Retrieving Model Text in Custom Loss Function for Training
Updated a month ago
Any plans for moreLlama 3?
Closed a month ago
How gemma_lm.preprocessor.sequence_length dealing with large input data
Closed a month ago3
Add Electra Weights to Kaggle Models
Updated a month ago3
Data-Parallel Training with KerasNLP and tf.distribute example dataset problem
Updated a month ago4
Keep kv cache as list of tensors maybe better than one tensor
Updated a month ago3
Why not use low precision matmul for reverse embedding in gemma model
Closed 2 months ago4
create local variable per_token_loss in score method to global. So that we can modify loss function.
Closed 2 months ago4
`SentencePieceTokenizer` inside a `keras.models.Model` fails to be reconstructed during `keras.saving.load_model()`
Closed 2 months ago2
Add grok-1
Closed 2 months ago
Feature Request: Transformer Debugger - Debugging and controlling the behavior of transformer based LLM models.
Updated 2 months ago3
Add Mistral 0.2 models as possible presets
Closed 2 months ago3
Gemma discrepancies
Closed 2 months ago1
Question about Gemma tensor parallel sharding policy
Closed 2 months ago5
Model weights contributions?
Updated 3 months ago5
Keras_NLP and Kaggle Hub: Are models allowed without weights in Kaggle Hub?
Updated 3 months ago
How to add a serialized model and weights of a keras model to keras-nlp?
Updated 3 months ago4
Dropout is not called in the training regime in TransformerEncoder and others
Closed 3 months ago2
Add CLIP tokenizer to Keras NLP
Updated 3 months ago1
Add `oov_token` Argument to `BytePairTokenizer`
Updated 3 months ago1
Broken link to "Understanding masking and padding" guide
Closed 3 months ago2
ContrastiveSampler lacks a seed param, while the docstring states it has one
Closed 3 months ago1
Any guide how to use tools/gemma/run_gemma_xla.py?
Closed 3 months ago4
Mistral kills the process by taking too many RAM
Closed 3 months ago2
Preset and doc for Mistral (multilingual)
Closed 3 months ago26
Issue with `BytePairTokenizer`
Updated 4 months ago
Keras CI fails due to update of `tensorflow-hub` to 0.16.1
Closed 4 months ago