Upload folder using huggingface_hub

res_pre_mlp/Rm0_S-1_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdd45d769aa0c28ad4ba961df9b38b89b6bd60655245948e4058d94d682feb44
+size 153705088

res_pre_mlp/Rm0_S-1_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 0,
+  "l1_exp": -1,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm0_S-1_R1_P0"
+}

res_pre_mlp/Rm0_S-2_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a069663395f9b3ffce93f725a10c04f668ee30d6245d93deec31b9df2cd6103
+size 153705088

res_pre_mlp/Rm0_S-2_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 0,
+  "l1_exp": -2,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm0_S-2_R1_P0"
+}

res_pre_mlp/Rm0_S-3_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a691258a2de93c8679f3c25c17fc0a7640aecc70744146a5cfa57621df0152d9
+size 153705088

res_pre_mlp/Rm0_S-3_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 0,
+  "l1_exp": -3,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm0_S-3_R1_P0"
+}

res_pre_mlp/Rm0_S-4_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9747dc098cb524d612e058f20fb5d1c53bb4f008c6df1bc3aab0c215e56849b0
+size 153705088

res_pre_mlp/Rm0_S-4_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 0,
+  "l1_exp": -4,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm0_S-4_R1_P0"
+}

res_pre_mlp/Rm0_S-5_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93d8bfe8fcfd33e2fe501bb3f67f38f3e383da1492e25ae3426cb81b5a8d9406
+size 153705088

res_pre_mlp/Rm0_S-5_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 0,
+  "l1_exp": -5,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm0_S-5_R1_P0"
+}

res_pre_mlp/Rm0_S-6_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:812e917f8ff0d1fa0a3581bb9525bcd6159ea168862024cab9896028c946695a
+size 153705088

res_pre_mlp/Rm0_S-6_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 0,
+  "l1_exp": -6,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm0_S-6_R1_P0"
+}

res_pre_mlp/Rm1_S-1_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c59c6799c971a4b6d34820795525b5ba0f138dfc74cab8fe963d5ce7aec6c943
+size 153705088

res_pre_mlp/Rm1_S-1_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 1,
+  "l1_exp": -1,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm1_S-1_R1_P0"
+}

res_pre_mlp/Rm1_S-2_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20c0090240406213acbecae5f70669dec85605b4a8a54fc5fa125cbd62038941
+size 153705088

res_pre_mlp/Rm1_S-2_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 1,
+  "l1_exp": -2,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm1_S-2_R1_P0"
+}

res_pre_mlp/Rm1_S-3_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5ee6f7dda5c12262e0891b076be0633e5eecaa8d4774d556bffb72b87e8b99f
+size 153705088

res_pre_mlp/Rm1_S-3_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 1,
+  "l1_exp": -3,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm1_S-3_R1_P0"
+}

res_pre_mlp/Rm1_S-4_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25a5aa5180058b83b73ec3b2b34b61314e402c859efd37e754ae61cec150e19a
+size 153705088

res_pre_mlp/Rm1_S-4_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 1,
+  "l1_exp": -4,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm1_S-4_R1_P0"
+}

res_pre_mlp/Rm1_S-5_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:094466fd13b69cbab01efc9af80cdd584946d932c35354f5f5b5caee03849baa
+size 153705088

res_pre_mlp/Rm1_S-5_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 1,
+  "l1_exp": -5,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm1_S-5_R1_P0"
+}

res_pre_mlp/Rm1_S-6_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:323856004211c2892f90bb6646092c9f978dfe5412a41c0d7f0203d9e1c62f9a
+size 153705088

res_pre_mlp/Rm1_S-6_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 1,
+  "l1_exp": -6,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm1_S-6_R1_P0"
+}

res_pre_mlp/Rm2_S-1_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3498111206f5af5177a62916216ee8b991b4790063a3285616a653dcfc161a1
+size 153705088

res_pre_mlp/Rm2_S-1_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 2,
+  "l1_exp": -1,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm2_S-1_R1_P0"
+}

res_pre_mlp/Rm2_S-2_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54a939b27aaae14db6d153a16eaf229653f1aaaa6fe31b5090710681fbf47757
+size 153705088

res_pre_mlp/Rm2_S-2_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 2,
+  "l1_exp": -2,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm2_S-2_R1_P0"
+}

res_pre_mlp/Rm2_S-3_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4aa5171c0db2fceedb86fee267890d25ccdee2e79d7347417402c6f07aea132
+size 153705088

res_pre_mlp/Rm2_S-3_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 2,
+  "l1_exp": -3,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm2_S-3_R1_P0"
+}

res_pre_mlp/Rm2_S-4_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b36c464096fa9d04d6f57d5091ec217764b8d7d933e07de04bb03ba66653adf
+size 153705088

res_pre_mlp/Rm2_S-4_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 2,
+  "l1_exp": -4,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm2_S-4_R1_P0"
+}

res_pre_mlp/Rm2_S-5_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:014bc4090b47fa09fcacdfd663c15d8ebd13e44cdd83d5b572e93a67f54e98bc
+size 153705088

res_pre_mlp/Rm2_S-5_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 2,
+  "l1_exp": -5,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm2_S-5_R1_P0"
+}

res_pre_mlp/Rm2_S-6_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b78c8a2598fef1dbcec12e8f7e2f786d3da7ef799018d2e74b89fe85186552ae
+size 153705088

res_pre_mlp/Rm2_S-6_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 2,
+  "l1_exp": -6,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm2_S-6_R1_P0"
+}

res_pre_mlp/Rm3_S-1_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9508a79866c6484117f48fe8f6c771efe489ed3a08afff00f1200de70ebefee3
+size 153705088

res_pre_mlp/Rm3_S-1_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 3,
+  "l1_exp": -1,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm3_S-1_R1_P0"
+}

res_pre_mlp/Rm3_S-2_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a499d162b1d688b92d82d751ee44aca3ebf4af75f1add808ea34947440a1a801
+size 153705088

res_pre_mlp/Rm3_S-2_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 3,
+  "l1_exp": -2,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm3_S-2_R1_P0"
+}

res_pre_mlp/Rm3_S-3_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd58590a88859a26222fadf422f55ff0045676939e19ee79823de99339aa4819
+size 153705088

res_pre_mlp/Rm3_S-3_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 3,
+  "l1_exp": -3,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm3_S-3_R1_P0"
+}

res_pre_mlp/Rm3_S-4_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fc2fc2aa7b626858e2b7756f705bc168b7871635fe837acfc025d3dbb6fab9a
+size 153705088

res_pre_mlp/Rm3_S-4_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 3,
+  "l1_exp": -4,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm3_S-4_R1_P0"
+}

res_pre_mlp/Rm3_S-5_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b60aef2c23d744928280f1ffbeee956d5defe6f9f2b8ae0511e38da90161740a
+size 153705088

res_pre_mlp/Rm3_S-5_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 3,
+  "l1_exp": -5,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm3_S-5_R1_P0"
+}

res_pre_mlp/Rm3_S-6_R1_P0.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:288034a630d32d3dc93541215ce00e9beb793aec5f6926afc03849ae21e36eff
+size 153705088

res_pre_mlp/Rm3_S-6_R1_P0_config.json

@@ -0,0 +1,38 @@
+{
+  "n_features": 25000,
+  "d_model": 768,
+  "lr_exp": -10,
+  "disable_comet": false,
+  "per_neuron_reinit_interval": 0,
+  "reservoir_time_discount": 0.995,
+  "reinit_interval": 800,
+  "max_reinit_neurons": 5000,
+  "reservoir_size": 5000,
+  "n_piles": 292,
+  "log_interval": 200,
+  "reinit_input_norm": "target_scaled",
+  "reinit_input": "x",
+  "reinit_norm_alpha": 0.3,
+  "data_loc": "mlp_data",
+  "reinit_threshold": -6,
+  "scheduler": "wsd",
+  "layer_idx": 3,
+  "l1_exp": -6,
+  "neuron_reinit_percent": 0.85,
+  "beta1": 1,
+  "beta2": 4,
+  "reinit_target": "error",
+  "sparse_adam": false,
+  "run_template": "Rm{layer_idx}_S{l1_exp}_R{l1_ratio}_P{l1_p}",
+  "project_name": "res_pre_mlp",
+  "decoder_bias": true,
+  "l1_beta": 0.99,
+  "alt_sparsity_loss": "log",
+  "l1_ratio": 1,
+  "l1_p": 0,
+  "optimizer": "sparse_adam",
+  "model_type": "res_B",
+  "adam_beta1": 0.5,
+  "adam_beta2": 0.9375,
+  "run_name": "Rm3_S-6_R1_P0"
+}