MCPcopy Create free account

hub / github.com/KellerJordan/modded-nanogpt / functions

Functions2,320 in github.com/KellerJordan/modded-nanogpt

Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260520_tail_refinterp_2900/train_gpt_tail_refinterp_2900.py:368
Method__init__
(self, named_params, lr=0.02, weight_decay=0, mu=0.95)
records/track_3_optimization/results/20260520_tail_refinterp_2900/train_gpt_tail_refinterp_2900.py:966
Method__init__
(self, dim)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:72
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:80
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:87
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:103
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:149
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:162
Method__init__
(self, params, lr=0.02, weight_decay=0, mu=0.95, refresh_interval=64)
records/track_3_optimization/results/20260505_newton_muon/train_gpt_simple_newton_muon.py:230
Method__init__
(self, dim)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:85
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:93
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:100
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:116
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:154
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:167
Method__init__
(self, params, lr=0.02, weight_decay=0, mu=0.95)
records/track_3_optimization/results/20260509_contra_muon_split_cooldown/train_gpt_simple_contra_muon_split_cooldown.py:248
Method__init__
(self, dim)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:86
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:94
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:101
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:117
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:155
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:168
Method__init__
(self, named_params, lr=0.02, weight_decay=0, mu=0.95)
records/track_3_optimization/results/20260504_contra_muon_mlp_soapish/train_gpt_contra_normuon_soapish_mlp.py:296
Method__init__
(self, dim)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:129
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:137
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:144
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:160
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:198
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:211
Method__init__
(self, named_params, lr=0.02, mu=0.95)
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:417
Method__init__
( self, params, inner_optimizer, lookahead_stepsize=0.0, use_scheduled
records/track_3_optimization/results/20260525_aurora_ema_ref/train_gpt_simple_aurora_ema_ref.py:487
Method__init__
(self, dim)
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:60
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:68
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:75
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:91
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:129
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:142
Method__init__
( self, params, lr: float = 1e-2, ema_beta: float = 0.9, momentum: flo
records/track_3_optimization/results/20260518_soaph/train_gpt_simple_soaph.py:238
Method__init__
(self, dim)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:222
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:230
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:237
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:253
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:291
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:304
Method__init__
(self, named_params, lr=0.02, weight_decay=0, mu=0.95)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:557
Method__init__
(self, params, lr=0.01, betas=(0.8, 0.99), eps=1e-10)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:635
Method__init__
(self, params, inner_optimizer, lookahead_stepsize=0, use_scheduled_lookahead_stepsize=True, lookahead_ema=0.9
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:661
Method__init__
(self, dim)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:205
Method__init__
(self, in_features, out_features)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:213
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:220
Method__init__
(self, dim: int, head_dim=128)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:236
Method__init__
(self, dim: int)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:274
Method__init__
(self, vocab_size: int, num_layers: int, model_dim: int)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:287
Method__init__
(self, named_params, lr=0.02, weight_decay=0, mu=0.95)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:540
Method__init__
(self, params, lr=0.01, betas=(0.8, 0.99), eps=1e-10)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:618
Method__init__
(self, params, inner_optimizer, lookahead_stepsize=0, use_scheduled_lookahead_stepsize=True, lookahead_ema=0.9
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:644
Method__iter__
Return self as iterator.
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/utils/dict_zip_iterator.py:70
Method__len__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/utils/shampoo_quantization.py:170
Method__next__
Return the next dictionary with values from the current position. Returns: Dictionary with the same keys as the input di
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/utils/dict_zip_iterator.py:74
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:106
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:127
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:174
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:232
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:469
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:824
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/shampoo_types.py:862
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributor/shampoo_block_info.py:108
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:145
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:391
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:531
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:601
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:775
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:847
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:1010
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/matrix_functions_types.py:108
Method__post_init__
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/matrix_functions_types.py:191
Method__setstate__
(self, state)
records/track_3_optimization/results/20260611_tailema_2720_submission/train_gpt_tailema_2720.py:678
Method__setstate__
(self, state)
records/track_3_optimization/results/20260609_soap_f1_auxb2_clean/train_gpt_clean_SOTA.py:529
Method__setstate__
(self, state)
records/track_3_optimization/results/20260529_tail_phase_readout_2850/train_gpt_tail_phase_readout_2850.py:940
Method__setstate__
(self, state)
records/track_3_optimization/results/20260619_cwd_rowfloor_tailema/train_gpt_cwd_SOTA.py:573
Method__setstate__
Override __setstate__ to handle OptimizerModule conversion during state dict loading. This method is called by the parent's load_state_dict()
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributed_shampoo.py:1522
Method__setstate__
(self, state: dict[str, Any])
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/tests/distributed_shampoo_test.py:364
Method__setstate__
(self, state)
records/track_3_optimization/results/20260611_tailema_2730_submission/train_gpt_tailema_2730.py:677
Method__setstate__
(self, state)
records/track_3_optimization/results/20260611_tailema_2730_submission/ablation/combo_2740/train_gpt_combo_2740.py:660
Method_allocate_zeros_distributed_tensor
Instantiates distributed tensor using DTensor. Args: size (tuple[int, ...]): Shape of desired tensor. dtype (torch.dt
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributor/shampoo_ddp_distributor.py:554
Method_allocate_zeros_distributed_tensor
Instantiates distributed tensor using DTensor. Args: size (tuple[int, ...]): Shape of desired tensor. dtype (torch.dt
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributor/shampoo_hsdp_distributor.py:842
Method_allocate_zeros_distributed_tensor
Instantiates distributed tensor using DTensor. Args: size (tuple[int, ...]): Shape of desired tensor. dtype (torch.dt
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/distributor/shampoo_hybrid_shard_distributor.py:574
Method_amortized_computation
Computes matrix inverse roots for Shampoo preconditioners. This implementation of the abstract _amortized_computation method specifically han
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:484
Method_amortized_computation
Performs eigendecomposition for Shampoo preconditioners. This implementation of the abstract _amortized_computation method specifically handl
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:711
Method_amortized_computation
Computes eigenvectors for eigenvalue-corrected Shampoo preconditioners. This implementation of the abstract _amortized_computation method spe
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:957
Method_amortized_computation_properties
( self, )
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/tests/shampoo_preconditioner_list_test.py:121
Method_amortized_computation_properties
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/tests/shampoo_preconditioner_list_test.py:1095
Method_amortized_computation_properties
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/tests/shampoo_preconditioner_list_test.py:1120
Method_amortized_computation_properties
(self)
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/tests/shampoo_preconditioner_list_test.py:1143
Function_check_2d_tensor
Decorator to check if the input tensor is 2-dimensional. This decorator checks if the input tensor `A` is 2-dimensional. If not, it rais
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/matrix_functions.py:92
Function_check_square_matrix
Decorator to check if the input matrix is square. This decorator first checks if input A is 2-dimensional, then checks if A is square. I
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/matrix_functions.py:118
Method_compute_outer_product_list
( self, grad: Tensor, order: int, preconditioned_dims_selector: tuple[bool, ..
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:1774
Method_compute_outer_product_list
( self, grad: Tensor, order: int, preconditioned_dims_selector: tuple[bool, ..
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:1817
Method_compute_preconditioned_gradient
( self, grad: Tensor, preconditioned_dims_selector: tuple[bool, ...], kronecke
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:1605
Method_compute_preconditioned_gradient
( self, grad: Tensor, preconditioned_dims_selector: tuple[bool, ...], kronecke
records/track_3_optimization/results/20260513_shampoo_1_4_power/distributed_shampoo/preconditioner/shampoo_preconditioner_list.py:1626
← previousnext →1,301–1,400 of 2,320, ranked by callers