{ "cache_config": { "n_heads": 4, "head_dim": 32, "sink_tokens_protected": 4, "eviction_capacity": 128 }, "sequence_length_sweep": { "L_64": { "seq_len": 64, "fp16_full": { "bytes": 32768, "update_ms": 0.201, "k_mse": 8e-08, "attn_out_cosine_sim": 1.0 }, "int8_full": { "bytes": 17408, "compression_ratio_vs_fp16": 1.88, "update_ms": 0.936, "k_mse": 5.3e-05, "attn_out_cosine_sim": 0.999963 }, "int4_packed_full": { "bytes": 10240, "compression_ratio_vs_fp16": 3.2, "update_ms": 1.794, "k_mse": 0.011081, "attn_out_cosine_sim": 0.996616 }, "sink_protected_int8_cap128": { "retained_tokens": 64, "evicted_tokens": 0, "bytes": 17408, "memory_reduction_vs_fp16_pct": 46.88, "update_ms": 0.754, "attn_out_cosine_sim_vs_full": 0.999963 }, "sliding_window_no_sink_cap128": { "retained_tokens": 64, "attn_out_cosine_sim_vs_full": 1.0 }, "scientific_classification": "MEASURED" }, "L_128": { "seq_len": 128, "fp16_full": { "bytes": 65536, "update_ms": 0.165, "k_mse": 6e-08, "attn_out_cosine_sim": 1.0 }, "int8_full": { "bytes": 34816, "compression_ratio_vs_fp16": 1.88, "update_ms": 1.005, "k_mse": 4.2e-05, "attn_out_cosine_sim": 0.999981 }, "int4_packed_full": { "bytes": 20480, "compression_ratio_vs_fp16": 3.2, "update_ms": 2.056, "k_mse": 0.009272, "attn_out_cosine_sim": 0.996034 }, "sink_protected_int8_cap128": { "retained_tokens": 128, "evicted_tokens": 0, "bytes": 34816, "memory_reduction_vs_fp16_pct": 46.88, "update_ms": 0.893, "attn_out_cosine_sim_vs_full": 0.999981 }, "sliding_window_no_sink_cap128": { "retained_tokens": 128, "attn_out_cosine_sim_vs_full": 1.0 }, "scientific_classification": "MEASURED" }, "L_256": { "seq_len": 256, "fp16_full": { "bytes": 131072, "update_ms": 0.233, "k_mse": 5e-08, "attn_out_cosine_sim": 1.0 }, "int8_full": { "bytes": 69632, "compression_ratio_vs_fp16": 1.88, "update_ms": 1.15, "k_mse": 3.6e-05, "attn_out_cosine_sim": 0.999962 }, "int4_packed_full": { "bytes": 40960, "compression_ratio_vs_fp16": 3.2, "update_ms": 2.259, "k_mse": 0.007604, "attn_out_cosine_sim": 0.995489 }, "sink_protected_int8_cap128": { "retained_tokens": 128, "evicted_tokens": 128, "bytes": 34816, "memory_reduction_vs_fp16_pct": 73.44, "update_ms": 2.132, "attn_out_cosine_sim_vs_full": 0.985176 }, "sliding_window_no_sink_cap128": { "retained_tokens": 128, "attn_out_cosine_sim_vs_full": 0.119727 }, "scientific_classification": "MEASURED" }, "L_512": { "seq_len": 512, "fp16_full": { "bytes": 262144, "update_ms": 0.454, "k_mse": 5e-08, "attn_out_cosine_sim": 1.0 }, "int8_full": { "bytes": 139264, "compression_ratio_vs_fp16": 1.88, "update_ms": 1.695, "k_mse": 3.3e-05, "attn_out_cosine_sim": 0.999982 }, "int4_packed_full": { "bytes": 81920, "compression_ratio_vs_fp16": 3.2, "update_ms": 2.986, "k_mse": 0.006855, "attn_out_cosine_sim": 0.996796 }, "sink_protected_int8_cap128": { "retained_tokens": 128, "evicted_tokens": 384, "bytes": 34816, "memory_reduction_vs_fp16_pct": 86.72, "update_ms": 1.206, "attn_out_cosine_sim_vs_full": 0.996568 }, "sliding_window_no_sink_cap128": { "retained_tokens": 128, "attn_out_cosine_sim_vs_full": 0.149448 }, "scientific_classification": "MEASURED" }, "L_1024": { "seq_len": 1024, "fp16_full": { "bytes": 524288, "update_ms": 0.583, "k_mse": 5e-08, "attn_out_cosine_sim": 1.0 }, "int8_full": { "bytes": 278528, "compression_ratio_vs_fp16": 1.88, "update_ms": 2.118, "k_mse": 3e-05, "attn_out_cosine_sim": 0.999973 }, "int4_packed_full": { "bytes": 163840, "compression_ratio_vs_fp16": 3.2, "update_ms": 4.185, "k_mse": 0.006429, "attn_out_cosine_sim": 0.995593 }, "sink_protected_int8_cap128": { "retained_tokens": 128, "evicted_tokens": 896, "bytes": 34816, "memory_reduction_vs_fp16_pct": 93.36, "update_ms": 1.263, "attn_out_cosine_sim_vs_full": 0.996337 }, "sliding_window_no_sink_cap128": { "retained_tokens": 128, "attn_out_cosine_sim_vs_full": 0.004714 }, "scientific_classification": "MEASURED" } } }