`eval.py` hangs when config yaml's model hparams don't match model checkpoint hparams

[Matthew]

Environment

0: Collecting system information...
0: ---------------------------------
0: System Environment Report
0: Created: 2023-11-21 21:17:06 UTC
0: ---------------------------------
0:
0: PyTorch information
0: -------------------
0: PyTorch version: 2.1.0+cu121
0: Is debug build: False
0: CUDA used to build PyTorch: 12.1
0: ROCM used to build PyTorch: N/A
0:
0: OS: Ubuntu 20.04.6 LTS (x86_64)
0: GCC version: (Ubuntu 9.4.0-1ubuntu1~20.04.2) 9.4.0
0: Clang version: Could not collect
0: CMake version: version 3.16.3
0: Libc version: glibc-2.31
0:
0: Python version: 3.10.13 (main, Aug 25 2023, 13:20:03) [GCC 9.4.0] (64-bit runtime)
0: Python platform: Linux-5.15.0-1047-aws-x86_64-with-glibc2.31
0: Is CUDA available: True
0: CUDA runtime version: 11.8.89
0: CUDA_MODULE_LOADING set to: LAZY
0: GPU models and configuration:
0: GPU 0: NVIDIA A100-SXM4-40GB
0: GPU 1: NVIDIA A100-SXM4-40GB
0: GPU 2: NVIDIA A100-SXM4-40GB
0: GPU 3: NVIDIA A100-SXM4-40GB
0: GPU 4: NVIDIA A100-SXM4-40GB
0: GPU 5: NVIDIA A100-SXM4-40GB
0: GPU 6: NVIDIA A100-SXM4-40GB
0: GPU 7: NVIDIA A100-SXM4-40GB
0:
0: Nvidia driver version: 535.104.12
0: cuDNN version: Probably one of the following:
0: /usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.0
0: /usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.0
0: /usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.0
0: /usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.0
0: /usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.0
0: /usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.0
0: /usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.0
0: HIP runtime version: N/A
0: MIOpen runtime version: N/A
0: Is XNNPACK available: True
0:
0: CPU:
0: Architecture: x86_64
0: CPU op-mode(s): 32-bit, 64-bit
0: Byte Order: Little Endian
0: Address sizes: 46 bits physical, 48 bits virtual
0: CPU(s): 48
0: On-line CPU(s) list: 0-47
0: Thread(s) per core: 1
0: Core(s) per socket: 24
0: Socket(s): 2
0: NUMA node(s): 2
0: Vendor ID: GenuineIntel
0: CPU family: 6
0: Model: 85
0: Model name: Intel(R) Xeon(R) Platinum 8275CL CPU @ 3.00GHz
0: Stepping: 7
0: CPU MHz: 2999.998
0: BogoMIPS: 5999.99
0: Hypervisor vendor: KVM
0: Virtualization type: full
0: L1d cache: 1.5 MiB
0: L1i cache: 1.5 MiB
0: L2 cache: 48 MiB
0: L3 cache: 71.5 MiB
0: NUMA node0 CPU(s): 0-23
0: NUMA node1 CPU(s): 24-47
0: Vulnerability Gather data sampling: Unknown: Dependent on hypervisor status
0: Vulnerability Itlb multihit: KVM: Mitigation: VMX unsupported
0: Vulnerability L1tf: Mitigation; PTE Inversion
0: Vulnerability Mds: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
0: Vulnerability Meltdown: Mitigation; PTI
0: Vulnerability Mmio stale data: Vulnerable: Clear CPU buffers attempted, no microcode; SMT Host state unknown
0: Vulnerability Retbleed: Vulnerable
0: Vulnerability Spec rstack overflow: Not affected
0: Vulnerability Spec store bypass: Vulnerable
0: Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
0: Vulnerability Spectre v2: Mitigation; Retpolines, STIBP disabled, RSB filling, PBRSB-eIBRS Not affected
0: Vulnerability Srbds: Not affected
0: Vulnerability Tsx async abort: Not affected
0: Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss ht syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq monitor ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid mpx avx512f avx512dq rdseed adx smap clflushopt clwb avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves ida arat pku ospke
0:
0: Versions of relevant libraries:
0: [pip3] numpy==1.26.0
0: [pip3] pytorch-ranger==0.1.1
0: [pip3] torch==2.1.0
0: [pip3] torch-optimizer==0.3.0
0: [pip3] torchmetrics==1.0.3
0: [pip3] torchvision==0.15.2+cu118
0: [pip3] triton==2.1.0
0: [pip3] triton-pre-mlir==2.0.0
0: [conda] Could not collect
0:
0:
0: Composer information
0: --------------------
0: Composer version: 0.16.4
0: Composer commit hash: None
0: Host processor model name: Intel(R) Xeon(R) Platinum 8275CL CPU @ 3.00GHz
0: Host processor core count: 48
0: Number of nodes: 1
0: Accelerator model name: NVIDIA A100-SXM4-40GB
0: Accelerators per node: 1
0: CUDA Device Count: 8

To reproduce

Steps to reproduce the behavior:

Run composer --world_size ${WORLD_SIZE} --nproc ${NPROC} --node_rank ${NODE_RANK} --master_addr ${MASTER_ADDR} --master_port ${MASTER_PORT} --verbose /llm-foundry/scripts/eval/eval.py {path/to/eval_yaml}

using a yaml that has the model hparams from scripts/eval/yamls/mpt_eval.yaml, but pointing to a checkpoint generated by scripts/train/yamls/pretrain/mpt-3b.yaml.

Then wait for the nccl timeout.

Expected behavior

It should probably throw an error about a mismatch between the model arch and the state dict.