Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Index your code
hub
/
github.com/skyzh/tiny-llm
/ functions
Functions
344 in github.com/skyzh/tiny-llm
⨍
Functions
344
◇
Types & classes
51
Function
test_task_2_quantized_matmul_simple_bf16_cpu
()
tests_refsol/test_week_2_day_2.py:49
Function
test_task_2_quantized_matmul_simple_f16_cpu
()
tests_refsol/test_week_2_day_2.py:57
Function
test_task_2_qwen_mlp
(stream: mx.Stream, precision: mx.Dtype, dims: dict)
tests_refsol/test_week_1_day_4.py:78
Function
test_task_2_rope_mlx_non_traditional
( stream: mx.Stream, with_offset: bool, precision: mx.Dtype )
tests_refsol/test_week_1_day_2.py:69
Function
test_task_2_router_topk
()
tests_refsol/test_week_3_day_3.py:52
Function
test_task_2_silu
(stream: mx.Stream, precision: mx.Dtype)
tests_refsol/test_week_1_day_4.py:49
Function
test_task_2_simple_multi_head_attention
Test if `MultiHeadAttention` can process everything correctly. We assume Q/K/V are of the same dimensions.
tests_refsol/test_week_1_day_1.py:132
Function
test_task_3_flash_attention_gpu
(mask_mode: str)
tests_refsol/test_week_2_day_4.py:165
Function
test_task_3_flash_attention_gpu_causal_mask_faster_than_all_zero_mask
()
tests_refsol/test_week_2_day_4.py:174
Function
test_task_3_flash_attention_gpu_extra_small
(mask_mode: str)
tests_refsol/test_week_2_day_4.py:155
Function
test_task_3_flash_attention_gpu_large
(mask_mode: str)
tests_refsol/test_week_2_day_4.py:170
Function
test_task_3_flash_attention_gpu_small
(mask_mode: str)
tests_refsol/test_week_2_day_4.py:160
Function
test_task_3_incremental_decode_matches_week2
()
tests_refsol/test_week_3_day_1.py:220
Function
test_task_3_incremental_decode_matches_week2_with_paged_attention
()
tests_refsol/test_week_3_day_2.py:191
Function
test_task_3_moe
()
tests_refsol/test_week_3_day_3.py:93
Function
test_task_3_quantized_matmul_complex_bf16_gpu
()
tests_refsol/test_week_2_day_2.py:69
Function
test_task_3_quantized_matmul_complex_f16_gpu
()
tests_refsol/test_week_2_day_2.py:77
Function
test_task_3_quantized_matmul_simple_bf16_gpu
()
tests_refsol/test_week_2_day_2.py:65
Function
test_task_3_quantized_matmul_simple_f16_gpu
()
tests_refsol/test_week_2_day_2.py:73
Function
test_task_3_qwen3_0_6b
()
tests_refsol/test_week_2_day_6.py:324
Function
test_task_3_qwen3_0_6b
()
tests_refsol/test_week_1_day_5.py:161
Function
test_task_3_qwen3_0_6b
()
tests_refsol/test_week_2_day_1.py:51
Function
test_task_3_qwen3_1_7b
()
tests_refsol/test_week_2_day_6.py:341
Function
test_task_3_qwen3_1_7b
()
tests_refsol/test_week_1_day_5.py:175
Function
test_task_3_qwen3_1_7b
()
tests_refsol/test_week_2_day_1.py:65
Function
test_task_3_qwen3_4b
()
tests_refsol/test_week_2_day_6.py:331
Function
test_task_3_qwen3_4b
()
tests_refsol/test_week_1_day_5.py:168
Function
test_task_3_qwen3_4b
()
tests_refsol/test_week_2_day_1.py:58
Function
test_task_3_qwen3_grouped_query_attention
( stream: mx.Stream, precision: mx.Dtype, mask: str | None )
tests_refsol/test_week_1_day_3.py:137
Function
test_task_4_qwen3_0_6b
()
tests_refsol/test_week_2_day_1.py:97
Function
test_task_4_qwen3_1_7b
()
tests_refsol/test_week_2_day_1.py:114
Function
test_task_4_qwen3_4b
()
tests_refsol/test_week_2_day_1.py:104
Function
test_utils_qwen3_0_6b
()
tests_refsol/test_week_1_day_5.py:92
Function
test_utils_qwen3_0_6b
()
tests_refsol/test_week_2_day_1.py:15
Function
test_utils_qwen3_1_7b
()
tests_refsol/test_week_1_day_5.py:106
Function
test_utils_qwen3_1_7b
()
tests_refsol/test_week_2_day_1.py:29
Function
test_utils_qwen3_4b
()
tests_refsol/test_week_1_day_5.py:99
Function
test_utils_qwen3_4b
()
tests_refsol/test_week_2_day_1.py:22
Method
update_and_fetch
Update the key-value cache and fetch the updated key-value cache. Args: key: The key to update the cache with.
src/tiny_llm/kv_cache.py:9
Method
update_and_fetch
( self, key: mx.array, value: mx.array, mask_length: int | None = None,
src/tiny_llm_ref/paged_kv_cache.py:207
Method
update_and_fetch
( self, keys: mx.array, values: mx.array, mask_length: int | None = None,
src/tiny_llm_ref/kv_cache.py:73
Method
update_and_fetch
( self, key: mx.array, value: mx.array, mask_length: int | None = None,
src/tiny_llm_ref/kv_cache.py:217
Method
update_and_fetch_paged
( self, key: mx.array, value: mx.array, mask_length: int | None = None,
src/tiny_llm_ref/paged_kv_cache.py:247
Method
write_page_slice
( self, page_id: int, start: int, key: mx.array, value: mx.array,
src/tiny_llm/paged_kv_cache.py:37
← previous
301–344 of 344, ranked by callers