Skip to content

Actions: TransformerLensOrg/TransformerLens

Actions

All workflows

Actions

Loading...
Loading

Showing runs from all workflows
2,500+ workflow runs
2,500+ workflow runs

Filter by Workflow

⌥ + click/return to exclude

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
Add T5 architecture adapter test Checks #5212: Pull request #1349 opened by RecreationalMath
55m 59s RecreationalMath:t5-adapter-test
RecreationalMath:t5-adapter-test
55m 59s
Parametrize the Cohere logit_scale fold test with a scale other than 1.0 Checks #5211: Pull request #1348 opened by EphraiemSarabamoun
51m 51s EphraiemSarabamoun:fix/1325-cohere-logit-scale-fold-test
EphraiemSarabamoun:fix/1325-cohere-logit-scale-fold-test
51m 51s
v4.0.0a1 Release #97: Release v4.0.0a1 published by jlarson4
45m 12s
45m 12s
Pass through bridge changes Checks #5210: Commit 59a1a97 pushed by jlarson4
48m 36s v4.0.0a1
v4.0.0a1
48m 36s
Pass through bridge changes Checks #5209: Commit 59a1a97 pushed by jlarson4
47m 16s dev-4.x
dev-4.x
47m 16s
Clamp top_k to vocab size in sample_logits and add edge-case tests (#… Checks #5208: Commit cdfab1a pushed by jlarson4
1h 47m 18s dev
dev
1h 47m 18s
Fix to_numpy() to handle bfloat16 tensors (#1346) Checks #5207: Commit 08c9ef9 pushed by jlarson4
3m 7s dev
dev
3m 7s
Fix sample_logits crash when top_k exceeds vocab size Checks #5206: Pull request #1347 opened by robbiebusinessacc
47m 25s robbiebusinessacc:contrib/clamp-sample-logits-top-k
robbiebusinessacc:contrib/clamp-sample-logits-top-k
47m 25s
Fix to_numpy() crash on bfloat16 tensors by upcasting to float32 Checks #5205: Pull request #1346 opened by robbiebusinessacc
49m 19s robbiebusinessacc:contrib/to-numpy-bfloat16
robbiebusinessacc:contrib/to-numpy-bfloat16
49m 19s
Fix run_with_cache(device=...) permanently moving the model (#1345) Checks #5204: Commit 34e6dc4 pushed by jlarson4
5m 48s dev
dev
5m 48s
Remove deprecated move_model parameter from ActivationCache.to (#1344) Checks #5203: Commit 84e90e1 pushed by jlarson4
56m 27s dev
dev
56m 27s
Fix run_with_cache(device=...) permanently moving the model Checks #5202: Pull request #1345 opened by RecreationalMath
47m 18s RecreationalMath:fix-run-with-cache
RecreationalMath:fix-run-with-cache
47m 18s
Remove deprecated move_model parameter from ActivationCache.to Checks #5201: Pull request #1344 opened by RecreationalMath
50m 31s RecreationalMath:remove-deprecated-move
RecreationalMath:remove-deprecated-move
50m 31s
vLLM Driver Bugs (#1343) Checks #5200: Commit d8b471e pushed by jlarson4
1m 31s dev-4.x
dev-4.x
1m 31s
vLLM Driver Bugs Checks #5199: Pull request #1343 synchronize by jlarson4
39m 20s bugs/vllm-driver
bugs/vllm-driver
39m 20s
vLLM Driver Bugs Checks #5198: Pull request #1343 opened by jlarson4
31m 20s bugs/vllm-driver
bugs/vllm-driver
31m 20s
Add GPT-OSS architecture adapter tests (#1341) Checks #5197: Commit e124a06 pushed by jlarson4
1h 13m 18s dev
dev
1h 13m 18s
v3.3.0 Release #96: Release v3.3.0 published by jlarson4
56m 35s
56m 35s
Merge pull request #1321 from TransformerLensOrg/dev Checks #5196: Commit 59a828a pushed by jlarson4
49m 48s v3.3.0
v3.3.0
49m 48s
Fix FactoredMatrix indexing returning empty result for -1 index (#1340) Checks #5195: Commit b3b9934 pushed by jlarson4
6m 6s dev
dev
6m 6s
Add return_cache option to TransformerBridge.generate (#1337) Checks #5194: Commit f676d8a pushed by jlarson4
54m 55s dev
dev
54m 55s
vLLM Batches (#1338) Checks #5193: Commit 2e236da pushed by jlarson4
22m 10s dev-4.x
dev-4.x
22m 10s
Add GPT-OSS architecture adapter tests Checks #5192: Pull request #1341 opened by RecreationalMath
47m 28s RecreationalMath:gpt_oss-adapter-test
RecreationalMath:gpt_oss-adapter-test
47m 28s
Fix FactoredMatrix indexing returning empty result for -1 index Checks #5191: Pull request #1340 opened by Kymi808
51m 52s Kymi808:fix/factored-matrix-negative-index
Kymi808:fix/factored-matrix-negative-index
51m 52s
vLLM Batches Checks #5190: Pull request #1338 opened by jlarson4
1m 31s feature/vllm-batched
feature/vllm-batched
1m 31s