Shubbair
f84b231cf2
Code Arrangement
2024-08-01 15:29:43 +03:00
Shubbair
7e0bdacef3
Code Arrangement
2024-08-01 15:22:19 +03:00
Shubbair
37bbf3ec54
Updating GAN Code...
2024-08-01 01:04:14 +03:00
Shubbair
4d17f80efb
Updating GAN Code...
2024-07-31 20:23:57 +03:00
Shubbair
1ef3ad2c6c
Updating GAN Code...
2024-07-31 19:59:36 +03:00
Shubbair
a8ffa9cb18
Updating GAN Code...
2024-07-31 11:50:32 +03:00
Shubbair
f70cef9567
Updating GAN Code...
2024-07-31 11:25:39 +03:00
Shubbair
6f7a6609b9
Updating MLX Notebook
2024-07-30 20:01:14 +03:00
Shubbair
0644cc101b
Updating MLX Notebook
2024-07-30 19:50:02 +03:00
Shubbair
ad2b6643c0
Updating GAN Code...
2024-07-30 16:59:35 +03:00
Shubbair
3bea855bd2
Updating GAN Code...
2024-07-30 13:45:09 +03:00
Shubbair
c2d731d8a3
Updating GAN Code...
2024-07-30 13:24:53 +03:00
Shubbair
ba52447385
Updating GAN Code...
2024-07-30 13:21:38 +03:00
Shubbair
1e386b5c20
Updating GAN Code...
2024-07-30 02:56:13 +03:00
Shubbair
7438b54ecd
Updating GAN Code...
2024-07-30 02:44:41 +03:00
Shubbair
7fea34d65e
Updating GAN Code...
2024-07-30 02:37:09 +03:00
Shubbair
f505fe6e55
Updating GAN Code...
2024-07-30 02:17:12 +03:00
Shubbair
4e80759b39
Updating GAN Code...
2024-07-30 02:06:52 +03:00
Shubbair
306e53c402
Updating GAN Code...
2024-07-29 19:44:16 +03:00
Shubbair
bacaa9ec0e
Updating GAN Code...
2024-07-29 01:30:08 +03:00
Shubbair
8d27be1442
Updating GAN Code...
2024-07-29 01:24:50 +03:00
Shubbair
4de0583b49
Updating GAN Code...
2024-07-28 19:18:35 +03:00
Shubbair
a07ef6d03b
Updating GAN Code...
2024-07-28 18:11:39 +03:00
Shubbair
c0c8293842
Updating GAN Code...
2024-07-28 17:56:26 +03:00
Shubbair
d17d293df9
Updating GAN Code...
2024-07-28 17:35:36 +03:00
Shubbair
3e63cd93fe
Updating GAN Code...
2024-07-28 17:26:24 +03:00
Shubbair
3716501e8d
Updating GAN Code...
2024-07-28 17:22:40 +03:00
Shubbair
88a20b7276
Updating GAN Code...
2024-07-28 01:10:19 +03:00
Shubbair
8b1713737a
Updating GAN Code...
2024-07-27 01:20:00 +03:00
Shubbair
f8b7094fb8
Updating GAN Code...
2024-07-27 01:19:50 +03:00
Shubbair
147cb3d2bc
Updating GAN Code...
2024-07-27 01:09:51 +03:00
Shubbair
a05608c34d
Updating GAN Code...
2024-07-27 00:22:29 +03:00
Shubbair
f176cce74d
Updating GAN Code...
2024-07-27 00:19:08 +03:00
Shubbair
959c623908
Updating GAN Code...
2024-07-26 16:38:55 +03:00
Shubbair
591074bea8
Updating GAN Code...
2024-07-26 16:36:29 +03:00
Shubbair
d426586b03
Updating GAN Code...
2024-07-26 16:07:40 +03:00
Shubbair
5e7ce1048c
Add GAN model 25/7
2024-07-25 21:00:41 +03:00
Alex Cheema
cd8efc7fbc
Add support for Llama-3.1 ( #907 )
...
* add dynamicNTK scaling rope
* remove unused var
* fix rope base
* llama3.1 fixes
* TODO for rope eval
* vectorise llama3 base freq calculation
* removed the arbitrary 2.0 rope_scale default case
* fix slow llama3.1 generation by evaluating stateless part of DynamicNTKScalingRoPE in init
* nits + format
* use mx.pi
* fix tests and add test for 3.1
---------
Co-authored-by: Prince Canuma <prince.gdt@gmail.com>
Co-authored-by: Awni Hannun <awni@apple.com>
2024-07-23 13:21:32 -07:00
M. Ali Bayram
47060a8130
refactor: add force_download parameter to get_model_path function ( #800 )
2024-07-23 13:10:20 -07:00
Prince Canuma
3f337e0f0a
Add Mistral NeMo (fix) ( #895 )
...
* fix head_dim
* Update llms/mlx_lm/models/llama.py
* fix kv error
* formatting
* Delete test.py
---------
Co-authored-by: Awni Hannun <awni.hannun@gmail.com>
2024-07-22 06:09:24 -07:00
Prince Canuma
3d365b612a
Add support for InternLM-2.5 ( #871 )
...
* fix internlm-2
* formatting
* add dynamic ntk rope
* formatting
* move dynamic scaling rope to intermlm2.py
* add default max_position_embeddings
2024-07-17 16:38:22 -07:00
Anchen
561dcf5643
Add support for deepseek coder v2 lite ( #882 )
...
* feat: add support for deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct
* fix softmax + some cleanup
* more nits
* fix rope
* fix original_max_position_embeddings in rope
* fix original_max_position_embeddings in rope config
* add group greedy
---------
Co-authored-by: Awni Hannun <awni@apple.com>
2024-07-17 07:23:28 -07:00
Awni Hannun
f0c6c6e226
keep the server in a valid state ( #889 )
2024-07-15 18:35:36 -07:00
JosefAlbers
bfc1f2763b
longrope ( #886 )
2024-07-12 07:19:11 -07:00
Chime Ogbuji
8bf397e450
Pass use_dora parameter to linear_to_lora_layers ( #885 )
2024-07-11 14:34:34 -07:00
nicolov
fbe3247772
Add GPT-neox model ( #863 )
2024-07-11 06:13:17 -07:00
James A Capozzoli
9717307ff0
Validation with full data set, results in NaN validation score ( #879 )
...
* CLI arguments may set num_batches to -1
The CLI arguments allow you to validate with the entire dataset by passing a negative one value, but this quickly results in a division by zero `NaN` to appear as the validation score!
* Must properly assemble the mini batches when validating with entire dataset.
Tested locally, a validation of a novel took about an hour, with a loss of 0.928. Thanks @awni for the correction!
* Set up the pre-commit hooks and run them so that black may format lora.py.
2024-07-10 08:36:11 -07:00
Alex Wozniakowski
63800c8feb
Example of response generation with optional arguments ( #853 )
...
* Generate response with optional arguments
* Reference response generation example
* Include transformers and sentencepiece
* Update example to run Mistral-7B-Instruct-v0.3
* Link to generation example
* Style changes from pre-commit
2024-07-09 06:49:59 -07:00
Awni Hannun
68e88d42fb
Fix server for openai package ( #877 )
...
* fix
* fixes for 9b
2024-07-08 12:34:31 -07:00
Awni Hannun
20e221f7f7
Add recurrent gemma ( #856 )
...
* add recurrent gemma
* fix window cache
2024-07-07 12:10:04 -07:00
