| {"n_eval_tokens": 131072, "model_family": "dfc", "L21_dead_pct": 50.38, "L21_shared_dead": 20466, "L21_base_dead": 27198, "L21_ft_dead": 8055, "L21_L0_base": 97.0, "L21_L0_ft": 445.07, "L21_EVR_base": 0.9897, "L21_EVR_ft": 0.8252, "L22_dead_pct": 41.58, "L22_shared_dead": 13204, "L22_base_dead": 27261, "L22_ft_dead": 5519, "L22_L0_base": 21.44, "L22_L0_ft": 470.71, "L22_EVR_base": 0.9889, "L22_EVR_ft": 0.7731, "L23_dead_pct": 39.9, "L23_shared_dead": 11493, "L23_base_dead": 27175, "L23_ft_dead": 5456, "L23_L0_base": 14.35, "L23_L0_ft": 512.0, "L23_EVR_base": 0.9882, "L23_EVR_ft": 0.7192, "L24_dead_pct": 45.14, "L24_shared_dead": 14945, "L24_base_dead": 27090, "L24_ft_dead": 7890, "L24_L0_base": 23.81, "L24_L0_ft": 512.35, "L24_EVR_base": 0.9826, "L24_EVR_ft": 0.7426, "dead_pct": 44.25, "L0_base": 39.15, "L0_ft": 485.03, "EVR_base": 0.9874, "EVR_ft": 0.765, "checkpoint": "step_002000.pt", "eval_time_s": 81.4} |
| {"n_eval_tokens": 131072, "model_family": "dfc", "L21_dead_pct": 51.06, "L21_shared_dead": 20881, "L21_base_dead": 27279, "L21_ft_dead": 8313, "L21_L0_base": 44.0, "L21_L0_ft": 193.46, "L21_EVR_base": 0.9935, "L21_EVR_ft": 0.8374, "L22_dead_pct": 42.36, "L22_shared_dead": 13881, "L22_base_dead": 27306, "L22_ft_dead": 5658, "L22_L0_base": 10.06, "L22_L0_ft": 193.92, "L22_EVR_base": 0.9928, "L22_EVR_ft": 0.8093, "L23_dead_pct": 40.59, "L23_shared_dead": 12137, "L23_base_dead": 27263, "L23_ft_dead": 5491, "L23_L0_base": 4.78, "L23_L0_ft": 190.6, "L23_EVR_base": 0.9908, "L23_EVR_ft": 0.7862, "L24_dead_pct": 46.67, "L24_shared_dead": 16282, "L24_base_dead": 27230, "L24_ft_dead": 8098, "L24_L0_base": 6.59, "L24_L0_ft": 135.57, "L24_EVR_base": 0.9857, "L24_EVR_ft": 0.8127, "dead_pct": 45.17, "L0_base": 16.36, "L0_ft": 178.39, "EVR_base": 0.9907, "EVR_ft": 0.8114, "checkpoint": "step_004000.pt", "eval_time_s": 77.1} |
| {"n_eval_tokens": 131072, "model_family": "dfc", "L21_dead_pct": 51.5, "L21_shared_dead": 21146, "L21_base_dead": 27288, "L21_ft_dead": 8517, "L21_L0_base": 35.5, "L21_L0_ft": 149.44, "L21_EVR_base": 0.9936, "L21_EVR_ft": 0.8481, "L22_dead_pct": 42.68, "L22_shared_dead": 14084, "L22_base_dead": 27318, "L22_ft_dead": 5801, "L22_L0_base": 8.91, "L22_L0_ft": 148.35, "L22_EVR_base": 0.993, "L22_EVR_ft": 0.8358, "L23_dead_pct": 41.01, "L23_shared_dead": 12411, "L23_base_dead": 27294, "L23_ft_dead": 5649, "L23_L0_base": 3.61, "L23_L0_ft": 138.01, "L23_EVR_base": 0.9915, "L23_EVR_ft": 0.8325, "L24_dead_pct": 47.64, "L24_shared_dead": 17029, "L24_base_dead": 27282, "L24_ft_dead": 8374, "L24_L0_base": 3.68, "L24_L0_ft": 85.48, "L24_EVR_base": 0.9854, "L24_EVR_ft": 0.8632, "dead_pct": 45.71, "L0_base": 12.92, "L0_ft": 130.32, "EVR_base": 0.9909, "EVR_ft": 0.8449, "checkpoint": "step_006000.pt", "eval_time_s": 79.5} |
| {"n_eval_tokens": 131072, "model_family": "dfc", "L21_dead_pct": 51.06, "L21_shared_dead": 20881, "L21_base_dead": 27279, "L21_ft_dead": 8313, "L21_L0_base": 44.0, "L21_L0_ft": 193.46, "L21_EVR_base": 0.9935, "L21_EVR_ft": 0.8374, "L22_dead_pct": 42.36, "L22_shared_dead": 13881, "L22_base_dead": 27306, "L22_ft_dead": 5658, "L22_L0_base": 10.06, "L22_L0_ft": 193.92, "L22_EVR_base": 0.9928, "L22_EVR_ft": 0.8093, "L23_dead_pct": 40.59, "L23_shared_dead": 12137, "L23_base_dead": 27263, "L23_ft_dead": 5491, "L23_L0_base": 4.78, "L23_L0_ft": 190.6, "L23_EVR_base": 0.9908, "L23_EVR_ft": 0.7862, "L24_dead_pct": 46.67, "L24_shared_dead": 16282, "L24_base_dead": 27230, "L24_ft_dead": 8098, "L24_L0_base": 6.59, "L24_L0_ft": 135.57, "L24_EVR_base": 0.9857, "L24_EVR_ft": 0.8127, "dead_pct": 45.17, "L0_base": 16.36, "L0_ft": 178.39, "EVR_base": 0.9907, "EVR_ft": 0.8114, "checkpoint": "step_004000.pt", "eval_time_s": 77.8} |
| {"n_eval_tokens": 131072, "model_family": "dfc", "L21_dead_pct": 51.5, "L21_shared_dead": 21146, "L21_base_dead": 27288, "L21_ft_dead": 8517, "L21_L0_base": 35.5, "L21_L0_ft": 149.44, "L21_EVR_base": 0.9936, "L21_EVR_ft": 0.8481, "L22_dead_pct": 42.68, "L22_shared_dead": 14084, "L22_base_dead": 27318, "L22_ft_dead": 5801, "L22_L0_base": 8.91, "L22_L0_ft": 148.35, "L22_EVR_base": 0.993, "L22_EVR_ft": 0.8358, "L23_dead_pct": 41.01, "L23_shared_dead": 12411, "L23_base_dead": 27294, "L23_ft_dead": 5649, "L23_L0_base": 3.61, "L23_L0_ft": 138.01, "L23_EVR_base": 0.9915, "L23_EVR_ft": 0.8325, "L24_dead_pct": 47.64, "L24_shared_dead": 17029, "L24_base_dead": 27282, "L24_ft_dead": 8374, "L24_L0_base": 3.68, "L24_L0_ft": 85.48, "L24_EVR_base": 0.9854, "L24_EVR_ft": 0.8632, "dead_pct": 45.71, "L0_base": 12.92, "L0_ft": 130.32, "EVR_base": 0.9909, "EVR_ft": 0.8449, "checkpoint": "step_006000.pt", "eval_time_s": 80.9} |
| {"n_eval_tokens": 131072, "model_family": "dfc", "L21_dead_pct": 54.65, "L21_shared_dead": 23717, "L21_base_dead": 27284, "L21_ft_dead": 9439, "L21_L0_base": 25.11, "L21_L0_ft": 105.5, "L21_EVR_base": 0.9952, "L21_EVR_ft": 0.8528, "L22_dead_pct": 44.57, "L22_shared_dead": 15688, "L22_base_dead": 27309, "L22_ft_dead": 6293, "L22_L0_base": 5.79, "L22_L0_ft": 101.99, "L22_EVR_base": 0.9949, "L22_EVR_ft": 0.852, "L23_dead_pct": 43.7, "L23_shared_dead": 14747, "L23_base_dead": 27294, "L23_ft_dead": 6287, "L23_L0_base": 2.7, "L23_L0_ft": 87.07, "L23_EVR_base": 0.9929, "L23_EVR_ft": 0.8629, "L24_dead_pct": 50.85, "L24_shared_dead": 19618, "L24_base_dead": 27317, "L24_ft_dead": 9300, "L24_L0_base": 2.32, "L24_L0_ft": 49.17, "L24_EVR_base": 0.9881, "L24_EVR_ft": 0.9018, "dead_pct": 48.44, "L0_base": 8.98, "L0_ft": 85.93, "EVR_base": 0.9928, "EVR_ft": 0.8674, "checkpoint": "step_010000.pt", "eval_time_s": 80.6} |
|
|