| checkpoints/offline/Minihack-Offline-Diffusion-BC-100M/model.safetensors filter=lfs diff=lfs merge=lfs -text |
| checkpoints/online/Minihack-Online-Diffusion-DAgger-100M/model.safetensors filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/cka_similarity.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_action_diversity.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_advantage_clip.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_attention_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_baseline_rl.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_bc_wins.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_entropy_bonus.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_ewc.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_ffn_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_frozen_backbone.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_gradient_surgery.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_head_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_kl_penalty.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_layer_ablation_top1.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_layer_ablation_top2.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_layer_ablation_top3.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_llrd.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_lora.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_low_t.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_mixed_replay.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_normalized_adv.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_reward_filtering.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_reward_model.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_running_stats.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_t_curriculum.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/curves_trust_region_kl.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/eval_scores_over_training.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/final_score_comparison.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/gradient_alignment.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/gradient_conflict_map.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_env_delta.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_action_diversity.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_advantage_clip.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_attention_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_baseline_rl.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_bc_wins.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_entropy_bonus.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_ewc.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_ffn_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_gradient_surgery.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_kl_penalty.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_layer_ablation_top2.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_layer_ablation_top3.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_llrd.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_lora.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_low_t.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_mixed_replay.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_normalized_adv.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_reward_filtering.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_reward_model.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_running_stats.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_t_curriculum.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/per_layer_grad_heatmap_trust_region_kl.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/representation_drift.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_action_diversity.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_advantage_clip.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_attention_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_baseline_rl.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_bc_wins.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_entropy_bonus.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_ewc.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_ffn_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_frozen_backbone.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_gradient_surgery.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_head_only.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_kl_penalty.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_layer_ablation_top1.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_layer_ablation_top2.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_layer_ablation_top3.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_llrd.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_lora.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_low_t.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_mixed_replay.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_normalized_adv.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_reward_filtering.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_reward_model.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_running_stats.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_t_curriculum.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_grad_norms_trust_region_kl.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_bin_norms_heatmap.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/t_distribution_analysis.png filter=lfs diff=lfs merge=lfs -text |
| experiments/rl_finetuning/outputs/minihack_ablations/figures/win_rate_and_effective_batch_size.png filter=lfs diff=lfs merge=lfs -text |
|
|
