cybershiptrooper/CURRICULUM-backdoored-threshold_0.45-RM-n_examples_1000-probe_layers_10-n_epochs_2
7B • Updated • 2
Rohan Gupta
cybershiptrooper
AI & ML interests
Doing AI Safety research @ MATS
Recent Activity
published a dataset 19 days ago
icl-heads/permutation-pools published a dataset 19 days ago
icl-heads/language-generalisation published a dataset 19 days ago
icl-heads/evidence-plus-labelOrganizations
cybershiptrooper/CURRICULUM-backdoored-threshold_0.4-RM-n_examples_1000-probe_layers_10-n_epochs_2
7B • Updated • 3
cybershiptrooper/CURRICULUM-grpo_backdoored-threshold_0.35-RM-n_examples_1000-probe_layers_10-n_epochs_5
7B • Updated • 9
cybershiptrooper/grpo_backdoored_7B-threshold_0.3-RM-n_examples_1000-probe_linear_layers_10
7B • Updated • 2
cybershiptrooper/grpo_backdoored_7B-threshold_0.3-RM-n_examples_400-probe_linear_layers_10
7B • Updated • 2
cybershiptrooper/CURRICULUM-threshold_0.5-RM-n_examples_200-probe_layers_10-n_epochs_9
Updated
cybershiptrooper/CURRICULUM-threshold_0.45-RM-n_examples_200-probe_layers_10-n_epochs_9
Updated
cybershiptrooper/CURRICULUM-threshold_0.42-RM-n_examples_200-probe_layers_10-n_epochs_9
Updated
cybershiptrooper/CURRICULUM-threshold_0.4-RM-n_examples_200-probe_layers_10-n_epochs_9
Updated
cybershiptrooper/grpo-continuous_weight_1.4-RM-n_examples_200-probe_layers_10
7B • Updated • 1
cybershiptrooper/grpo-threshold_0.32-RM-n_examples_200-probe_layers_10
7B • Updated • 2
cybershiptrooper/CURRICULUM-threshold_0.35-RM-n_examples_200-probe_layers_10-n_epochs_9
Updated
cybershiptrooper/grpo-threshold_0.35-RM-n_examples_200-probe_layers_10
7B • Updated • 2
cybershiptrooper/CURRICULUM-threshold_0.3-RM-n_examples_200-probe_layers_10-n_epochs_9
7B • Updated • 7
cybershiptrooper/grpo_70B-threshold_0.3-RM-n_examples_313-probe_layers_10-lora
Updated
cybershiptrooper/grpo-threshold_0.3-RM-n_examples_313-probe_layers_10-lora
Updated
cybershiptrooper/grpo-threshold_0.3-RM-n_examples_200-probe_layers_10
7B • Updated • 4
cybershiptrooper/CURRICULUM-threshold_0.3-RM-n_examples_200-probe_layers_10-n_epochs_3
7B • Updated • 7
cybershiptrooper/grpo-threshold_0.25-RM-n_examples_200-probe_layers_10
7B • Updated • 2
cybershiptrooper/grpo-hybrid-RM-epochs-1000-0TeOru3CyC-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-UfPlgPwPPL-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-GjiRKEBq89-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-HFtMA7Ie99-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-AzAA5nYoTV-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-IxgX1Qo2dj-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-So9uVE7CZI-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-tq5EfqfGRx-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-ab7Hy6eltt-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-IPgIsEIpH0-probe_layers-10
Updated
cybershiptrooper/grpo-hybrid-RM-epochs-1000-nmUIluNDzi-probe_layers-10
Updated