{ "producer": { "name": "modelopt", "version": "0.44.0" }, "quantization": { "quant_algo": "FP8", "kv_cache_quant_algo": "FP8", "exclude_modules": [ "backbone.embeddings", "backbone.layers.0.mixer.conv1d", "backbone.layers.0.mixer.norm", "backbone.layers.0.norm", "backbone.layers.1.mixer.gate", "backbone.layers.1.norm", "backbone.layers.10.mixer.gate", "backbone.layers.10.norm", "backbone.layers.11.mixer.conv1d", "backbone.layers.11.mixer.norm", "backbone.layers.11.norm", "backbone.layers.12*", "backbone.layers.12.mixer.k_proj", "backbone.layers.12.mixer.o_proj", "backbone.layers.12.mixer.q_proj", "backbone.layers.12.mixer.v_proj", "backbone.layers.12.norm", "backbone.layers.13.mixer.gate", "backbone.layers.13.norm", "backbone.layers.14.mixer.conv1d", "backbone.layers.14.mixer.norm", "backbone.layers.14.norm", "backbone.layers.15.mixer.gate", "backbone.layers.15.norm", "backbone.layers.16.mixer.conv1d", "backbone.layers.16.mixer.norm", "backbone.layers.16.norm", "backbone.layers.17.mixer.gate", "backbone.layers.17.norm", "backbone.layers.18.mixer.conv1d", "backbone.layers.18.mixer.norm", "backbone.layers.18.norm", "backbone.layers.19*", "backbone.layers.19.mixer.k_proj", "backbone.layers.19.mixer.o_proj", "backbone.layers.19.mixer.q_proj", "backbone.layers.19.mixer.v_proj", "backbone.layers.19.norm", "backbone.layers.2.mixer.conv1d", "backbone.layers.2.mixer.norm", "backbone.layers.2.norm", "backbone.layers.20.mixer.gate", "backbone.layers.20.norm", "backbone.layers.21.mixer.conv1d", "backbone.layers.21.mixer.norm", "backbone.layers.21.norm", "backbone.layers.22.mixer.gate", "backbone.layers.22.norm", "backbone.layers.23.mixer.conv1d", "backbone.layers.23.mixer.norm", "backbone.layers.23.norm", "backbone.layers.24.mixer.gate", "backbone.layers.24.norm", "backbone.layers.25.mixer.conv1d", "backbone.layers.25.mixer.norm", "backbone.layers.25.norm", "backbone.layers.26*", "backbone.layers.26.mixer.k_proj", "backbone.layers.26.mixer.o_proj", "backbone.layers.26.mixer.q_proj", "backbone.layers.26.mixer.v_proj", "backbone.layers.26.norm", "backbone.layers.27.mixer.gate", "backbone.layers.27.norm", "backbone.layers.28.mixer.conv1d", "backbone.layers.28.mixer.norm", "backbone.layers.28.norm", "backbone.layers.29.mixer.gate", "backbone.layers.29.norm", "backbone.layers.3.mixer.gate", "backbone.layers.3.norm", "backbone.layers.30.mixer.conv1d", "backbone.layers.30.mixer.norm", "backbone.layers.30.norm", "backbone.layers.31.mixer.gate", "backbone.layers.31.norm", "backbone.layers.32.mixer.conv1d", "backbone.layers.32.mixer.norm", "backbone.layers.32.norm", "backbone.layers.33*", "backbone.layers.33.mixer.k_proj", "backbone.layers.33.mixer.o_proj", "backbone.layers.33.mixer.q_proj", "backbone.layers.33.mixer.v_proj", "backbone.layers.33.norm", "backbone.layers.34.mixer.gate", "backbone.layers.34.norm", "backbone.layers.35.mixer.conv1d", "backbone.layers.35.mixer.norm", "backbone.layers.35.norm", "backbone.layers.36.mixer.gate", "backbone.layers.36.norm", "backbone.layers.37.mixer.conv1d", "backbone.layers.37.mixer.norm", "backbone.layers.37.norm", "backbone.layers.38.mixer.gate", "backbone.layers.38.norm", "backbone.layers.39.mixer.conv1d", "backbone.layers.39.mixer.norm", "backbone.layers.39.norm", "backbone.layers.4.mixer.conv1d", "backbone.layers.4.mixer.norm", "backbone.layers.4.norm", "backbone.layers.40.mixer.gate", "backbone.layers.40.norm", "backbone.layers.41.mixer.conv1d", "backbone.layers.41.mixer.norm", "backbone.layers.41.norm", "backbone.layers.42*", "backbone.layers.42.mixer.k_proj", "backbone.layers.42.mixer.o_proj", "backbone.layers.42.mixer.q_proj", "backbone.layers.42.mixer.v_proj", "backbone.layers.42.norm", "backbone.layers.43.mixer.gate", "backbone.layers.43.norm", "backbone.layers.44.mixer.conv1d", "backbone.layers.44.mixer.norm", "backbone.layers.44.norm", "backbone.layers.45.mixer.gate", "backbone.layers.45.norm", "backbone.layers.46.mixer.conv1d", "backbone.layers.46.mixer.norm", "backbone.layers.46.norm", "backbone.layers.47.mixer.gate", "backbone.layers.47.norm", "backbone.layers.48.mixer.conv1d", "backbone.layers.48.mixer.norm", "backbone.layers.48.norm", "backbone.layers.49.mixer.gate", "backbone.layers.49.norm", "backbone.layers.5.*", "backbone.layers.5.mixer.k_proj", "backbone.layers.5.mixer.o_proj", "backbone.layers.5.mixer.q_proj", "backbone.layers.5.mixer.v_proj", "backbone.layers.5.norm", "backbone.layers.50.mixer.conv1d", "backbone.layers.50.mixer.norm", "backbone.layers.50.norm", "backbone.layers.51.mixer.gate", "backbone.layers.51.norm", "backbone.layers.6.mixer.gate", "backbone.layers.6.norm", "backbone.layers.7.mixer.conv1d", "backbone.layers.7.mixer.norm", "backbone.layers.7.norm", "backbone.layers.8.mixer.gate", "backbone.layers.8.norm", "backbone.layers.9.mixer.conv1d", "backbone.layers.9.mixer.norm", "backbone.layers.9.norm", "backbone.norm_f", "lm_head", "mtp.layers.0.enorm", "mtp.layers.0.hnorm", "mtp.layers.0.mixer*", "mtp.layers.0.mixer.k_proj", "mtp.layers.0.mixer.o_proj", "mtp.layers.0.mixer.q_proj", "mtp.layers.0.mixer.v_proj", "mtp.layers.0.norm", "mtp.layers.1.final_layernorm", "mtp.layers.1.mixer.gate", "mtp.layers.1.norm" ] } }