update README

README.md

@@ -4,7 +4,7 @@ tags:
   - gptoss
 ---
 
-# gptoss_kernels
+# gpt-oss-metal-kernels
 
 Metal kernels that back the OpenAI GPT-OSS reference implementation, repackaged for local experiments on Apple Silicon GPUs. The GPT-OSS project distributes optimized inference primitives for the `gpt-oss-20b` and `gpt-oss-120b` open-weight models, including MXFP4-packed linear layers and fused attention paths that target Metal Performance Shaders on macOS [gpt-oss](https://github.com/openai/gpt-oss).
 
@@ -14,7 +14,7 @@ Metal kernels that back the OpenAI GPT-OSS reference implementation, repackaged
 pip install kernels  # we just need to install the kernels package
 ```
 
-The package exposes Python bindings through `gptoss_kernels.ops`; these symbols are re-exported in `gptoss_kernels.__init__` for convenience. All kernels expect Metal (`mps`) tensors and operate in place on user-provided outputs to minimize additional allocations.
+The package exposes Python bindings through `gpt_oss_metal_kernels.ops`; these symbols are re-exported in `gpt_oss_metal_kernels.__init__` for convenience. All kernels expect Metal (`mps`) tensors and operate in place on user-provided outputs to minimize additional allocations.
 
 ## Available Ops
 
@@ -39,7 +39,7 @@ Each example below compares a Metal kernel against the canonical PyTorch equival
 import torch
 from kernels import get_kernel
 
-gptoss_kernels = get_kernel("kernels-community/gptoss_kernels")
+gptoss_kernels = get_kernel("kernels-community/gpt-oss-metal-kernels")
 
 torch.manual_seed(0)
 device = "mps"
@@ -74,7 +74,7 @@ torch.testing.assert_close(out_kernel, out_ref, atol=1e-3, rtol=1e-3)
 from kernels import get_kernel
 import torch
 
-gptoss_kernels = get_kernel("kernels-community/gptoss_kernels")
+gptoss_kernels = get_kernel("kernels-community/gpt-oss-metal-kernels")
 device = "mps"
 
 hidden = 4096
@@ -101,7 +101,7 @@ from kernels import get_kernel
 import torch
 
 device = "mps"
-gptoss_kernels = get_kernel("kernels-community/gptoss_kernels")
+gptoss_kernels = get_kernel("kernels-community/gpt-oss-metal-kernels")
 
 vocab, dim = 1024, 256
 token_ids = torch.randint(0, vocab, (16,), device=device, dtype=torch.int32)
@@ -125,7 +125,7 @@ import torch
 import torch.nn as nn
 
 device = "mps"
-gptoss_kernels = get_kernel("kernels-community/gptoss_kernels")
+gptoss_kernels = get_kernel("kernels-community/gpt-oss-metal-kernels")
 
 
 head_dim = 64