Skip to content

Actions: keyboardAnt/distributed-speculative-inference

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
487 workflow runs
487 workflow runs

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

STABLE. revert custom device_map (seems like HF has a bug) Python tests #386: Commit 774c852 pushed by keyboardAnt
September 16, 2024 23:13 2m 47s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 23:13 2m 47s
target=llama 3.1 70b Python tests #385: Commit e5b6681 pushed by keyboardAnt
September 16, 2024 21:22 48m 49s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 21:22 48m 49s
test_loading_on_all_gpus_except_0 Python tests #384: Commit 30e225c pushed by keyboardAnt
September 16, 2024 21:18 47m 54s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 21:18 47m 54s
get_device_map_without_gpu_0 for verifier Python tests #383: Commit 66c6f24 pushed by keyboardAnt
September 16, 2024 21:10 45m 4s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 21:10 45m 4s
fix current lookahead Python tests #382: Commit b734405 pushed by keyboardAnt
September 16, 2024 21:03 2m 0s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 21:03 2m 0s
wip avoid overlapped requests via requested Python tests #381: Commit d7dac4f pushed by keyboardAnt
September 16, 2024 20:52 1m 59s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 20:52 1m 59s
wip manager minimize requests' overlapping Python tests #380: Commit 39f2802 pushed by keyboardAnt
September 16, 2024 18:03 2m 10s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 16, 2024 18:03 2m 10s
WIP manager: fix draft tok ids assignment Python tests #379: Commit e6d5472 pushed by keyboardAnt
September 15, 2024 22:00 2m 7s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 22:00 2m 7s
do not await on sending draft requests Python tests #378: Commit a238572 pushed by keyboardAnt
September 15, 2024 21:45 1m 57s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 21:45 1m 57s
fix manager: create draft requests Python tests #377: Commit 36bd080 pushed by keyboardAnt
September 15, 2024 21:43 1m 55s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 21:43 1m 55s
load_in_8bit Python tests #376: Commit 29a35b8 pushed by keyboardAnt
September 15, 2024 21:38 2m 0s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 21:38 2m 0s
POC complete load in 8bit Python tests #375: Commit 92bfc22 pushed by keyboardAnt
September 15, 2024 21:10 2m 0s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 21:10 2m 0s
poc load in 8bit Python tests #374: Commit e57004c pushed by keyboardAnt
September 15, 2024 21:01 2m 1s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 21:01 2m 1s
enhance logging Python tests #373: Commit f6c2fb9 pushed by keyboardAnt
September 15, 2024 20:14 2m 14s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 20:14 2m 14s
fix call to get_device_map_with_only_gpu_0 Python tests #372: Commit 6e6b760 pushed by keyboardAnt
September 15, 2024 20:03 2m 4s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 20:03 2m 4s
running dsi Python tests #371: Commit 7678ab7 pushed by keyboardAnt
September 15, 2024 20:01 2m 8s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 20:01 2m 8s
load drafter on gpu 0 only & free tokenizer Python tests #370: Commit 44142bc pushed by keyboardAnt
September 15, 2024 19:50 1m 55s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 19:50 1m 55s
complete the poc Python tests #369: Commit 9322470 pushed by keyboardAnt
September 15, 2024 19:42 1m 57s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 19:42 1m 57s
poc load models on gpu 0 only Python tests #368: Commit dd2e2cf pushed by keyboardAnt
September 15, 2024 19:27 1m 51s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 19:27 1m 51s
print_gpu_memory & fix vocab size Python tests #367: Commit 0fe0505 pushed by keyboardAnt
September 15, 2024 04:33 1m 50s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 04:33 1m 50s
remove the unnecessary model.to(device) Python tests #366: Commit eb57c53 pushed by keyboardAnt
September 15, 2024 04:24 1m 50s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 04:24 1m 50s
gc & load large models with device_map Python tests #365: Commit 4949967 pushed by keyboardAnt
September 15, 2024 03:00 2m 4s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 15, 2024 03:00 2m 4s
explore accelerate's device_map="balanced_low_0" Python tests #364: Commit b84b201 pushed by keyboardAnt
September 14, 2024 21:34 2m 4s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 14, 2024 21:34 2m 4s
torch no grad on all Python tests #363: Commit 18b0999 pushed by keyboardAnt
September 14, 2024 19:24 2m 2s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 14, 2024 19:24 2m 2s
generate baseline with any given dtype Python tests #362: Commit bd7da22 pushed by keyboardAnt
September 14, 2024 19:21 2m 11s nadav/actual-llm-coroutines-manager
nadav/actual-llm-coroutines-manager
September 14, 2024 19:21 2m 11s