[ { "backend": "manual", "d": 32, "dk": 16, "heads": 2, "ok": true, "preset": "pico_1x", "rank": 16, "rows": { "causal_alibi_forward": 0.0, "causal_alibi_loss": 0.0, "causal_alibi_param_grad": 0.0, "causal_alibi_x_grad": 5.820766091346741e-11, "legacy_load_forward": 0.0, "legacy_load_missing_unexpected": 0.0, "legacy_load_qkv_weight": 0.0, "none_forward": 0.0, "none_loss": 0.0, "none_param_grad": 0.0, "none_x_grad": 5.820766091346741e-11, "optimizer_remap": 0.0, "safe_load_any_loaded": 0.0, "safe_load_any_qkv": 0.0, "sat_alibi_forward": 0.0, "sat_alibi_loss": 0.0, "sat_alibi_param_grad": 0.0, "sat_alibi_x_grad": 2.3283064365386963e-10 }, "tol": 0.0002 }, { "backend": "sdpa", "d": 32, "dk": 16, "heads": 2, "ok": true, "preset": "pico_1x", "rank": 16, "rows": { "causal_alibi_forward": 0.0, "causal_alibi_loss": 0.0, "causal_alibi_param_grad": 0.0, "causal_alibi_x_grad": 1.1641532182693481e-10, "legacy_load_forward": 0.0, "legacy_load_missing_unexpected": 0.0, "legacy_load_qkv_weight": 0.0, "none_forward": 0.0, "none_loss": 0.0, "none_param_grad": 0.0, "none_x_grad": 5.820766091346741e-11, "safe_load_any_loaded": 0.0, "safe_load_any_qkv": 0.0, "sat_alibi_forward": 0.0, "sat_alibi_loss": 0.0, "sat_alibi_param_grad": 0.0, "sat_alibi_x_grad": 1.1641532182693481e-10 }, "tol": 0.0002 }, { "backend": "sublinear", "d": 32, "dk": 16, "heads": 2, "ok": true, "preset": "pico_1x", "rank": 16, "rows": { "causal_alibi_forward": 0.0, "causal_alibi_loss": 0.0, "causal_alibi_param_grad": 0.0, "causal_alibi_x_grad": 1.1641532182693481e-10, "legacy_load_forward": 0.0, "legacy_load_missing_unexpected": 0.0, "legacy_load_qkv_weight": 0.0, "none_forward": 0.0, "none_loss": 0.0, "none_param_grad": 0.0, "none_x_grad": 5.820766091346741e-11, "safe_load_any_loaded": 0.0, "safe_load_any_qkv": 0.0, "sat_alibi_forward": 0.0, "sat_alibi_loss": 0.0, "sat_alibi_param_grad": 0.0, "sat_alibi_x_grad": 1.1641532182693481e-10 }, "tol": 0.0002 }, { "backend": "manual", "d": 128, "dk": 16, "heads": 8, "ok": true, "preset": "micro_3x", "rank": 48, "rows": { "causal_alibi_forward": 0.0, "causal_alibi_loss": 0.0, "causal_alibi_param_grad": 0.0, "causal_alibi_x_grad": 5.820766091346741e-11, "legacy_load_forward": 0.0, "legacy_load_missing_unexpected": 0.0, "legacy_load_qkv_weight": 0.0, "none_forward": 0.0, "none_loss": 0.0, "none_param_grad": 0.0, "none_x_grad": 1.4551915228366852e-11, "safe_load_any_loaded": 0.0, "safe_load_any_qkv": 0.0, "sat_alibi_forward": 0.0, "sat_alibi_loss": 0.0, "sat_alibi_param_grad": 0.0, "sat_alibi_x_grad": 5.820766091346741e-11 }, "tol": 0.0002 }, { "backend": "sdpa", "d": 128, "dk": 16, "heads": 8, "ok": true, "preset": "micro_3x", "rank": 48, "rows": { "causal_alibi_forward": 0.0, "causal_alibi_loss": 0.0, "causal_alibi_param_grad": 0.0, "causal_alibi_x_grad": 5.820766091346741e-11, "legacy_load_forward": 0.0, "legacy_load_missing_unexpected": 0.0, "legacy_load_qkv_weight": 0.0, "none_forward": 0.0, "none_loss": 0.0, "none_param_grad": 0.0, "none_x_grad": 1.4551915228366852e-11, "safe_load_any_loaded": 0.0, "safe_load_any_qkv": 0.0, "sat_alibi_forward": 0.0, "sat_alibi_loss": 0.0, "sat_alibi_param_grad": 0.0, "sat_alibi_x_grad": 2.9103830456733704e-11 }, "tol": 0.0002 }, { "backend": "sublinear", "d": 128, "dk": 16, "heads": 8, "ok": true, "preset": "micro_3x", "rank": 48, "rows": { "causal_alibi_forward": 0.0, "causal_alibi_loss": 0.0, "causal_alibi_param_grad": 0.0, "causal_alibi_x_grad": 5.820766091346741e-11, "legacy_load_forward": 0.0, "legacy_load_missing_unexpected": 0.0, "legacy_load_qkv_weight": 0.0, "none_forward": 0.0, "none_loss": 0.0, "none_param_grad": 0.0, "none_x_grad": 1.4551915228366852e-11, "safe_load_any_loaded": 0.0, "safe_load_any_qkv": 0.0, "sat_alibi_forward": 0.0, "sat_alibi_loss": 0.0, "sat_alibi_param_grad": 0.0, "sat_alibi_x_grad": 2.9103830456733704e-11 }, "tol": 0.0002 } ]