diff --git a/README.md b/README.md index a2e3d86..81cecff 100644 --- a/README.md +++ b/README.md @@ -201,7 +201,7 @@ LongBench WANDB_MODE=disabled uv run python run_eval.py --checkpoint_path train_outputs/runs/.../pytorch_model.bin --datasets negative_nq triviaqa_retrieved squad longbench_e --split test # hypernet checkpoint -WANDB_MODE=disabled uv run python run_eval.py --checkpoint_path train_outputs/runs/May08_13-56-31_slurm0-a3nodeset-5_59383_906acb28/checkpoint-105000/pytorch_model.bin --datasets negative_nq triviaqa_retrieved squad longbench_e --split test +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/gov_report_e longbench/multifieldqa_en_e longbench/2wikimqa_e --split test --max_ctx_chunk_len -1 --lora_aggregation sum --eval_batch_size_gen 8 # squad only WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad --split test diff --git a/eval_scripts/chunked_eval.sh b/eval_scripts/chunked_eval.sh new file mode 100644 index 0000000..6781265 --- /dev/null +++ b/eval_scripts/chunked_eval.sh @@ -0,0 +1,15 @@ +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len -1 --lora_aggregation mean + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len -1 --lora_aggregation sum + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len 128 --lora_aggregation mean + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len 128 --lora_aggregation sum + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len 256 --lora_aggregation mean + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len 256 --lora_aggregation sum + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len 512 --lora_aggregation mean + +WANDB_MODE=disabled run uv run python run_eval.py --checkpoint_path train_outputs/runs/Aug02_07-51-08_slurm0-a3nodeset-9_76501_7fdab5ea/checkpoint-50000/pytorch_model.bin --datasets squad ropes drop longbench/squad_e longbench/gov_report_e test_longbench/multifieldqa_en_e --split test --max_ctx_chunk_len 512 --lora_aggregation sum diff --git a/src/ctx_to_lora/modeling/hypernet.py b/src/ctx_to_lora/modeling/hypernet.py index d954348..b592cc9 100644 --- a/src/ctx_to_lora/modeling/hypernet.py +++ b/src/ctx_to_lora/modeling/hypernet.py @@ -690,6 +690,8 @@ class ModulatedPretrainedModel(nn.Module): hypernet_config = state_dict["hypernet_config"] if getattr(hypernet_config, "num_pre_head_layers", None) is None: hypernet_config.num_pre_head_layers = 4 + if getattr(hypernet_config, "use_per_rank_bias", None) is None: + hypernet_config.use_per_rank_bias = False ctx_encoder_args = state_dict["ctx_encoder_args"] model = cls(base_model, hypernet_config, ctx_encoder_args, **kwargs) model.load_state_dict(state_dict)