matbee
/

weblinx-idefics-2

Generated from Trainer

Model card Files Files and versions Community

weblinx-idefics-2 / checkpoint-29500 /global_step29500

1 contributor

History: 1 commit

matbee's picture

Training in progress, step 29500, checkpoint

f97c132 verified 6 months ago

bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

83 MB
LFS

Training in progress, step 29500, checkpoint 6 months ago
bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt
Detected Pickle imports (7)
- "torch.FloatStorage",
- "torch.Tensor",
- "deepspeed.runtime.fp16.loss_scaler.LossScaler",
- "torch._utils._rebuild_tensor_v2",
- "collections.OrderedDict",
- "torch._tensor._rebuild_from_type_v2",
- "deepspeed.runtime.zero.config.ZeroStageEnum"
How to fix it?
83 MB
LFS

Training in progress, step 29500, checkpoint 6 months ago
zero_pp_rank_0_mp_rank_00_model_states.pt
Detected Pickle imports (5)
- "__builtin__.set",
- "torch._utils._rebuild_tensor_v2",
- "collections.OrderedDict",
- "torch.BFloat16Storage",
- "torch.Size"
How to fix it?
872 kB
LFS

Training in progress, step 29500, checkpoint 6 months ago
zero_pp_rank_1_mp_rank_00_model_states.pt
Detected Pickle imports (5)
- "__builtin__.set",
- "torch._utils._rebuild_tensor_v2",
- "collections.OrderedDict",
- "torch.BFloat16Storage",
- "torch.Size"
How to fix it?
872 kB
LFS

Training in progress, step 29500, checkpoint 6 months ago