Add files using upload-large-folder tool

.gitattributes

@@ -89,3 +89,11 @@ Qwen3-32B/ll_bsearch_kl0.005_eap0.985_sha_bw4.29_4-8bit_seed1234/tokenizer.json
 Qwen3-32B/ll_bsearch_kl0.005_eap0.99_sha_bw5.74_4-8bit_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Qwen3-32B/ll_bsearch_kl0.005_eap0.99_sha_bw5.74_4-8bit_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 Qwen3-32B/ll_bsearch_kl0.005_eap0.985_sha_bw4.26_4-5-6-7-8bit_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/tokenizer.json filter=lfs diff=lfs merge=lfs -text

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:949a717376d70bd7c14cb9f0a04ef3e48fa6c9571a0959dbad1d7816052530b1
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f79c9f35494ee28ac2f307fda7ef1f0c699855fcd9fd88db82f5e2e1d532273e
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:553898c0a66462cf9da4b62e3c4b60322a464a5ecd0f4ddaf5c3b653ba632710
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:949a717376d70bd7c14cb9f0a04ef3e48fa6c9571a0959dbad1d7816052530b1
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f79c9f35494ee28ac2f307fda7ef1f0c699855fcd9fd88db82f5e2e1d532273e
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b9d2c8849bb6be4634cc740b7d5d3cf942b8eccd9b7bb23b7113b21c25a3d1a
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/quantization_config.txt

@@ -0,0 +1,412 @@
+# Model: meta-llama/Llama-3.1-8B-Instruct
+# Layer directory: /nfs/scistore19/alistgrp/mhelcig/local/data/search/4_5_6_7_8bit_asym_g128/Llama-3.1-8B-Instruct/6bit/
+# Sensitivity method: shapley
+# Estimation method: permutation_separate
+# Available bitwidths: [4, 8]
+# Bitwidth map: {4: 4.156, 5: 5.156, 6: 6.156, 7: 7.156, 8: 8.156}
+#
+# Layer groups: 128 groups (fused layers share bitwidth)
+#   block_0:mlp.gate_proj,mlp.up_proj.block_0:mlp.gate_proj,mlp.up_proj: group 0, 2 layers
+#   block_0:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 1, 3 layers
+#   block_10:mlp.gate_proj,mlp.up_proj.block_10:mlp.gate_proj,mlp.up_proj: group 20, 2 layers
+#   block_10:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 21, 3 layers
+#   block_11:mlp.gate_proj,mlp.up_proj.block_11:mlp.gate_proj,mlp.up_proj: group 22, 2 layers
+#   block_11:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 23, 3 layers
+#   block_12:mlp.gate_proj,mlp.up_proj.block_12:mlp.gate_proj,mlp.up_proj: group 24, 2 layers
+#   block_12:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 25, 3 layers
+#   block_13:mlp.gate_proj,mlp.up_proj.block_13:mlp.gate_proj,mlp.up_proj: group 26, 2 layers
+#   block_13:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 27, 3 layers
+#   block_14:mlp.gate_proj,mlp.up_proj.block_14:mlp.gate_proj,mlp.up_proj: group 28, 2 layers
+#   block_14:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 29, 3 layers
+#   block_15:mlp.gate_proj,mlp.up_proj.block_15:mlp.gate_proj,mlp.up_proj: group 30, 2 layers
+#   block_15:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 31, 3 layers
+#   block_16:mlp.gate_proj,mlp.up_proj.block_16:mlp.gate_proj,mlp.up_proj: group 32, 2 layers
+#   block_16:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 33, 3 layers
+#   block_17:mlp.gate_proj,mlp.up_proj.block_17:mlp.gate_proj,mlp.up_proj: group 34, 2 layers
+#   block_17:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 35, 3 layers
+#   block_18:mlp.gate_proj,mlp.up_proj.block_18:mlp.gate_proj,mlp.up_proj: group 36, 2 layers
+#   block_18:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 37, 3 layers
+#   block_19:mlp.gate_proj,mlp.up_proj.block_19:mlp.gate_proj,mlp.up_proj: group 38, 2 layers
+#   block_19:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 39, 3 layers
+#   block_1:mlp.gate_proj,mlp.up_proj.block_1:mlp.gate_proj,mlp.up_proj: group 2, 2 layers
+#   block_1:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 3, 3 layers
+#   block_20:mlp.gate_proj,mlp.up_proj.block_20:mlp.gate_proj,mlp.up_proj: group 40, 2 layers
+#   block_20:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 41, 3 layers
+#   block_21:mlp.gate_proj,mlp.up_proj.block_21:mlp.gate_proj,mlp.up_proj: group 42, 2 layers
+#   block_21:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 43, 3 layers
+#   block_22:mlp.gate_proj,mlp.up_proj.block_22:mlp.gate_proj,mlp.up_proj: group 44, 2 layers
+#   block_22:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 45, 3 layers
+#   block_23:mlp.gate_proj,mlp.up_proj.block_23:mlp.gate_proj,mlp.up_proj: group 46, 2 layers
+#   block_23:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 47, 3 layers
+#   block_24:mlp.gate_proj,mlp.up_proj.block_24:mlp.gate_proj,mlp.up_proj: group 48, 2 layers
+#   block_24:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 49, 3 layers
+#   block_25:mlp.gate_proj,mlp.up_proj.block_25:mlp.gate_proj,mlp.up_proj: group 50, 2 layers
+#   block_25:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 51, 3 layers
+#   block_26:mlp.gate_proj,mlp.up_proj.block_26:mlp.gate_proj,mlp.up_proj: group 52, 2 layers
+#   block_26:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 53, 3 layers
+#   block_27:mlp.gate_proj,mlp.up_proj.block_27:mlp.gate_proj,mlp.up_proj: group 54, 2 layers
+#   block_27:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 55, 3 layers
+#   block_28:mlp.gate_proj,mlp.up_proj.block_28:mlp.gate_proj,mlp.up_proj: group 56, 2 layers
+#   block_28:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 57, 3 layers
+#   block_29:mlp.gate_proj,mlp.up_proj.block_29:mlp.gate_proj,mlp.up_proj: group 58, 2 layers
+#   block_29:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 59, 3 layers
+#   block_2:mlp.gate_proj,mlp.up_proj.block_2:mlp.gate_proj,mlp.up_proj: group 4, 2 layers
+#   block_2:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 5, 3 layers
+#   block_30:mlp.gate_proj,mlp.up_proj.block_30:mlp.gate_proj,mlp.up_proj: group 60, 2 layers
+#   block_30:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 61, 3 layers
+#   block_31:mlp.gate_proj,mlp.up_proj.block_31:mlp.gate_proj,mlp.up_proj: group 62, 2 layers
+#   block_31:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 63, 3 layers
+#   block_3:mlp.gate_proj,mlp.up_proj.block_3:mlp.gate_proj,mlp.up_proj: group 6, 2 layers
+#   block_3:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 7, 3 layers
+#   block_4:mlp.gate_proj,mlp.up_proj.block_4:mlp.gate_proj,mlp.up_proj: group 8, 2 layers
+#   block_4:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 9, 3 layers
+#   block_5:mlp.gate_proj,mlp.up_proj.block_5:mlp.gate_proj,mlp.up_proj: group 10, 2 layers
+#   block_5:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 11, 3 layers
+#   block_6:mlp.gate_proj,mlp.up_proj.block_6:mlp.gate_proj,mlp.up_proj: group 12, 2 layers
+#   block_6:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 13, 3 layers
+#   block_7:mlp.gate_proj,mlp.up_proj.block_7:mlp.gate_proj,mlp.up_proj: group 14, 2 layers
+#   block_7:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 15, 3 layers
+#   block_8:mlp.gate_proj,mlp.up_proj.block_8:mlp.gate_proj,mlp.up_proj: group 16, 2 layers
+#   block_8:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 17, 3 layers
+#   block_9:mlp.gate_proj,mlp.up_proj.block_9:mlp.gate_proj,mlp.up_proj: group 18, 2 layers
+#   block_9:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 19, 3 layers
+#   Fused in model.layers.0:
+#     - self_attn.o_proj (group 64, 1 layers)
+#     - mlp.down_proj (group 65, 1 layers)
+#   Fused in model.layers.1:
+#     - self_attn.o_proj (group 66, 1 layers)
+#     - mlp.down_proj (group 67, 1 layers)
+#   Fused in model.layers.10:
+#     - self_attn.o_proj (group 84, 1 layers)
+#     - mlp.down_proj (group 85, 1 layers)
+#   Fused in model.layers.11:
+#     - self_attn.o_proj (group 86, 1 layers)
+#     - mlp.down_proj (group 87, 1 layers)
+#   Fused in model.layers.12:
+#     - self_attn.o_proj (group 88, 1 layers)
+#     - mlp.down_proj (group 89, 1 layers)
+#   Fused in model.layers.13:
+#     - self_attn.o_proj (group 90, 1 layers)
+#     - mlp.down_proj (group 91, 1 layers)
+#   Fused in model.layers.14:
+#     - self_attn.o_proj (group 92, 1 layers)
+#     - mlp.down_proj (group 93, 1 layers)
+#   Fused in model.layers.15:
+#     - self_attn.o_proj (group 94, 1 layers)
+#     - mlp.down_proj (group 95, 1 layers)
+#   Fused in model.layers.16:
+#     - self_attn.o_proj (group 96, 1 layers)
+#     - mlp.down_proj (group 97, 1 layers)
+#   Fused in model.layers.17:
+#     - self_attn.o_proj (group 98, 1 layers)
+#     - mlp.down_proj (group 99, 1 layers)
+#   Fused in model.layers.18:
+#     - self_attn.o_proj (group 100, 1 layers)
+#     - mlp.down_proj (group 101, 1 layers)
+#   Fused in model.layers.19:
+#     - self_attn.o_proj (group 102, 1 layers)
+#     - mlp.down_proj (group 103, 1 layers)
+#   Fused in model.layers.2:
+#     - self_attn.o_proj (group 68, 1 layers)
+#     - mlp.down_proj (group 69, 1 layers)
+#   Fused in model.layers.20:
+#     - self_attn.o_proj (group 104, 1 layers)
+#     - mlp.down_proj (group 105, 1 layers)
+#   Fused in model.layers.21:
+#     - self_attn.o_proj (group 106, 1 layers)
+#     - mlp.down_proj (group 107, 1 layers)
+#   Fused in model.layers.22:
+#     - self_attn.o_proj (group 108, 1 layers)
+#     - mlp.down_proj (group 109, 1 layers)
+#   Fused in model.layers.23:
+#     - self_attn.o_proj (group 110, 1 layers)
+#     - mlp.down_proj (group 111, 1 layers)
+#   Fused in model.layers.24:
+#     - self_attn.o_proj (group 112, 1 layers)
+#     - mlp.down_proj (group 113, 1 layers)
+#   Fused in model.layers.25:
+#     - self_attn.o_proj (group 114, 1 layers)
+#     - mlp.down_proj (group 115, 1 layers)
+#   Fused in model.layers.26:
+#     - self_attn.o_proj (group 116, 1 layers)
+#     - mlp.down_proj (group 117, 1 layers)
+#   Fused in model.layers.27:
+#     - self_attn.o_proj (group 118, 1 layers)
+#     - mlp.down_proj (group 119, 1 layers)
+#   Fused in model.layers.28:
+#     - self_attn.o_proj (group 120, 1 layers)
+#     - mlp.down_proj (group 121, 1 layers)
+#   Fused in model.layers.29:
+#     - self_attn.o_proj (group 122, 1 layers)
+#     - mlp.down_proj (group 123, 1 layers)
+#   Fused in model.layers.3:
+#     - self_attn.o_proj (group 70, 1 layers)
+#     - mlp.down_proj (group 71, 1 layers)
+#   Fused in model.layers.30:
+#     - self_attn.o_proj (group 124, 1 layers)
+#     - mlp.down_proj (group 125, 1 layers)
+#   Fused in model.layers.31:
+#     - self_attn.o_proj (group 126, 1 layers)
+#     - mlp.down_proj (group 127, 1 layers)
+#   Fused in model.layers.4:
+#     - self_attn.o_proj (group 72, 1 layers)
+#     - mlp.down_proj (group 73, 1 layers)
+#   Fused in model.layers.5:
+#     - self_attn.o_proj (group 74, 1 layers)
+#     - mlp.down_proj (group 75, 1 layers)
+#   Fused in model.layers.6:
+#     - self_attn.o_proj (group 76, 1 layers)
+#     - mlp.down_proj (group 77, 1 layers)
+#   Fused in model.layers.7:
+#     - self_attn.o_proj (group 78, 1 layers)
+#     - mlp.down_proj (group 79, 1 layers)
+#   Fused in model.layers.8:
+#     - self_attn.o_proj (group 80, 1 layers)
+#     - mlp.down_proj (group 81, 1 layers)
+#   Fused in model.layers.9:
+#     - self_attn.o_proj (group 82, 1 layers)
+#     - mlp.down_proj (group 83, 1 layers)
+#
+# Mode: binary_search_constraint (measured)
+# Constraint max_kl: 0.005
+# Constraint min_eap: 0.985
+# Weights: nll=0.0, kl=0.0, eap=1.0
+#
+# Average bitwidth: 7.7089
+# Total params: 6979321856
+# Total bits: 53802786882
+# Final KL: 0.002595
+# Final EAP: 0.985184
+# Final ETL: 0.014816
+# Satisfies constraints: True
+# Solver calls: 9
+# Evaluations: 9
+#
+# Bitwidth distribution:
+#   8-bit: 210 layers (93.8%)
+#   4-bit: 14 layers (6.2%)
+#
+model.layers.0.mlp.gate_proj: 4
+model.layers.0.mlp.up_proj: 4
+model.layers.0.self_attn.k_proj: 8
+model.layers.0.self_attn.q_proj: 8
+model.layers.0.self_attn.v_proj: 8
+model.layers.1.mlp.gate_proj: 8
+model.layers.1.mlp.up_proj: 8
+model.layers.1.self_attn.k_proj: 8
+model.layers.1.self_attn.q_proj: 8
+model.layers.1.self_attn.v_proj: 8
+model.layers.2.mlp.gate_proj: 8
+model.layers.2.mlp.up_proj: 8
+model.layers.2.self_attn.k_proj: 8
+model.layers.2.self_attn.q_proj: 8
+model.layers.2.self_attn.v_proj: 8
+model.layers.3.mlp.gate_proj: 8
+model.layers.3.mlp.up_proj: 8
+model.layers.3.self_attn.k_proj: 8
+model.layers.3.self_attn.q_proj: 8
+model.layers.3.self_attn.v_proj: 8
+model.layers.4.mlp.gate_proj: 8
+model.layers.4.mlp.up_proj: 8
+model.layers.4.self_attn.k_proj: 8
+model.layers.4.self_attn.q_proj: 8
+model.layers.4.self_attn.v_proj: 8
+model.layers.5.mlp.gate_proj: 8
+model.layers.5.mlp.up_proj: 8
+model.layers.5.self_attn.k_proj: 8
+model.layers.5.self_attn.q_proj: 8
+model.layers.5.self_attn.v_proj: 8
+model.layers.6.mlp.gate_proj: 8
+model.layers.6.mlp.up_proj: 8
+model.layers.6.self_attn.k_proj: 8
+model.layers.6.self_attn.q_proj: 8
+model.layers.6.self_attn.v_proj: 8
+model.layers.7.mlp.gate_proj: 8
+model.layers.7.mlp.up_proj: 8
+model.layers.7.self_attn.k_proj: 8
+model.layers.7.self_attn.q_proj: 8
+model.layers.7.self_attn.v_proj: 8
+model.layers.8.mlp.gate_proj: 8
+model.layers.8.mlp.up_proj: 8
+model.layers.8.self_attn.k_proj: 8
+model.layers.8.self_attn.q_proj: 8
+model.layers.8.self_attn.v_proj: 8
+model.layers.9.mlp.gate_proj: 8
+model.layers.9.mlp.up_proj: 8
+model.layers.9.self_attn.k_proj: 8
+model.layers.9.self_attn.q_proj: 8
+model.layers.9.self_attn.v_proj: 8
+model.layers.10.mlp.gate_proj: 8
+model.layers.10.mlp.up_proj: 8
+model.layers.10.self_attn.k_proj: 8
+model.layers.10.self_attn.q_proj: 8
+model.layers.10.self_attn.v_proj: 8
+model.layers.11.mlp.gate_proj: 8
+model.layers.11.mlp.up_proj: 8
+model.layers.11.self_attn.k_proj: 8
+model.layers.11.self_attn.q_proj: 8
+model.layers.11.self_attn.v_proj: 8
+model.layers.12.mlp.gate_proj: 8
+model.layers.12.mlp.up_proj: 8
+model.layers.12.self_attn.k_proj: 8
+model.layers.12.self_attn.q_proj: 8
+model.layers.12.self_attn.v_proj: 8
+model.layers.13.mlp.gate_proj: 8
+model.layers.13.mlp.up_proj: 8
+model.layers.13.self_attn.k_proj: 8
+model.layers.13.self_attn.q_proj: 8
+model.layers.13.self_attn.v_proj: 8
+model.layers.14.mlp.gate_proj: 8
+model.layers.14.mlp.up_proj: 8
+model.layers.14.self_attn.k_proj: 8
+model.layers.14.self_attn.q_proj: 8
+model.layers.14.self_attn.v_proj: 8
+model.layers.15.mlp.gate_proj: 8
+model.layers.15.mlp.up_proj: 8
+model.layers.15.self_attn.k_proj: 8
+model.layers.15.self_attn.q_proj: 8
+model.layers.15.self_attn.v_proj: 8
+model.layers.16.mlp.gate_proj: 8
+model.layers.16.mlp.up_proj: 8
+model.layers.16.self_attn.k_proj: 8
+model.layers.16.self_attn.q_proj: 8
+model.layers.16.self_attn.v_proj: 8
+model.layers.17.mlp.gate_proj: 8
+model.layers.17.mlp.up_proj: 8
+model.layers.17.self_attn.k_proj: 8
+model.layers.17.self_attn.q_proj: 8
+model.layers.17.self_attn.v_proj: 8
+model.layers.18.mlp.gate_proj: 8
+model.layers.18.mlp.up_proj: 8
+model.layers.18.self_attn.k_proj: 8
+model.layers.18.self_attn.q_proj: 8
+model.layers.18.self_attn.v_proj: 8
+model.layers.19.mlp.gate_proj: 8
+model.layers.19.mlp.up_proj: 8
+model.layers.19.self_attn.k_proj: 8
+model.layers.19.self_attn.q_proj: 8
+model.layers.19.self_attn.v_proj: 8
+model.layers.20.mlp.gate_proj: 8
+model.layers.20.mlp.up_proj: 8
+model.layers.20.self_attn.k_proj: 8
+model.layers.20.self_attn.q_proj: 8
+model.layers.20.self_attn.v_proj: 8
+model.layers.21.mlp.gate_proj: 8
+model.layers.21.mlp.up_proj: 8
+model.layers.21.self_attn.k_proj: 8
+model.layers.21.self_attn.q_proj: 8
+model.layers.21.self_attn.v_proj: 8
+model.layers.22.mlp.gate_proj: 8
+model.layers.22.mlp.up_proj: 8
+model.layers.22.self_attn.k_proj: 8
+model.layers.22.self_attn.q_proj: 8
+model.layers.22.self_attn.v_proj: 8
+model.layers.23.mlp.gate_proj: 8
+model.layers.23.mlp.up_proj: 8
+model.layers.23.self_attn.k_proj: 8
+model.layers.23.self_attn.q_proj: 8
+model.layers.23.self_attn.v_proj: 8
+model.layers.24.mlp.gate_proj: 8
+model.layers.24.mlp.up_proj: 8
+model.layers.24.self_attn.k_proj: 8
+model.layers.24.self_attn.q_proj: 8
+model.layers.24.self_attn.v_proj: 8
+model.layers.25.mlp.gate_proj: 8
+model.layers.25.mlp.up_proj: 8
+model.layers.25.self_attn.k_proj: 8
+model.layers.25.self_attn.q_proj: 8
+model.layers.25.self_attn.v_proj: 8
+model.layers.26.mlp.gate_proj: 8
+model.layers.26.mlp.up_proj: 8
+model.layers.26.self_attn.k_proj: 8
+model.layers.26.self_attn.q_proj: 8
+model.layers.26.self_attn.v_proj: 8
+model.layers.27.mlp.gate_proj: 8
+model.layers.27.mlp.up_proj: 8
+model.layers.27.self_attn.k_proj: 8
+model.layers.27.self_attn.q_proj: 8
+model.layers.27.self_attn.v_proj: 8
+model.layers.28.mlp.gate_proj: 8
+model.layers.28.mlp.up_proj: 8
+model.layers.28.self_attn.k_proj: 8
+model.layers.28.self_attn.q_proj: 8
+model.layers.28.self_attn.v_proj: 8
+model.layers.29.mlp.gate_proj: 8
+model.layers.29.mlp.up_proj: 8
+model.layers.29.self_attn.k_proj: 8
+model.layers.29.self_attn.q_proj: 8
+model.layers.29.self_attn.v_proj: 8
+model.layers.30.mlp.gate_proj: 8
+model.layers.30.mlp.up_proj: 8
+model.layers.30.self_attn.k_proj: 8
+model.layers.30.self_attn.q_proj: 8
+model.layers.30.self_attn.v_proj: 8
+model.layers.31.mlp.gate_proj: 8
+model.layers.31.mlp.up_proj: 8
+model.layers.31.self_attn.k_proj: 8
+model.layers.31.self_attn.q_proj: 8
+model.layers.31.self_attn.v_proj: 8
+model.layers.0.self_attn.o_proj: 8
+model.layers.0.mlp.down_proj: 8
+model.layers.1.self_attn.o_proj: 8
+model.layers.1.mlp.down_proj: 8
+model.layers.2.self_attn.o_proj: 8
+model.layers.2.mlp.down_proj: 4
+model.layers.3.self_attn.o_proj: 8
+model.layers.3.mlp.down_proj: 8
+model.layers.4.self_attn.o_proj: 8
+model.layers.4.mlp.down_proj: 8
+model.layers.5.self_attn.o_proj: 8
+model.layers.5.mlp.down_proj: 8
+model.layers.6.self_attn.o_proj: 8
+model.layers.6.mlp.down_proj: 8
+model.layers.7.self_attn.o_proj: 8
+model.layers.7.mlp.down_proj: 4
+model.layers.8.self_attn.o_proj: 8
+model.layers.8.mlp.down_proj: 4
+model.layers.9.self_attn.o_proj: 8
+model.layers.9.mlp.down_proj: 4
+model.layers.10.self_attn.o_proj: 8
+model.layers.10.mlp.down_proj: 4
+model.layers.11.self_attn.o_proj: 8
+model.layers.11.mlp.down_proj: 8
+model.layers.12.self_attn.o_proj: 8
+model.layers.12.mlp.down_proj: 4
+model.layers.13.self_attn.o_proj: 8
+model.layers.13.mlp.down_proj: 8
+model.layers.14.self_attn.o_proj: 8
+model.layers.14.mlp.down_proj: 8
+model.layers.15.self_attn.o_proj: 8
+model.layers.15.mlp.down_proj: 8
+model.layers.16.self_attn.o_proj: 8
+model.layers.16.mlp.down_proj: 8
+model.layers.17.self_attn.o_proj: 8
+model.layers.17.mlp.down_proj: 8
+model.layers.18.self_attn.o_proj: 8
+model.layers.18.mlp.down_proj: 8
+model.layers.19.self_attn.o_proj: 8
+model.layers.19.mlp.down_proj: 8
+model.layers.20.self_attn.o_proj: 8
+model.layers.20.mlp.down_proj: 8
+model.layers.21.self_attn.o_proj: 8
+model.layers.21.mlp.down_proj: 8
+model.layers.22.self_attn.o_proj: 8
+model.layers.22.mlp.down_proj: 8
+model.layers.23.self_attn.o_proj: 8
+model.layers.23.mlp.down_proj: 8
+model.layers.24.self_attn.o_proj: 8
+model.layers.24.mlp.down_proj: 8
+model.layers.25.self_attn.o_proj: 4
+model.layers.25.mlp.down_proj: 4
+model.layers.26.self_attn.o_proj: 8
+model.layers.26.mlp.down_proj: 4
+model.layers.27.self_attn.o_proj: 8
+model.layers.27.mlp.down_proj: 4
+model.layers.28.self_attn.o_proj: 8
+model.layers.28.mlp.down_proj: 4
+model.layers.29.self_attn.o_proj: 8
+model.layers.29.mlp.down_proj: 4
+model.layers.30.self_attn.o_proj: 8
+model.layers.30.mlp.down_proj: 8
+model.layers.31.self_attn.o_proj: 8
+model.layers.31.mlp.down_proj: 8

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/README.md

@@ -0,0 +1,18 @@
+# Quantized Model Checkpoint
+
+**Base model:** meta-llama/Llama-3.1-8B-Instruct
+
+**Average bitwidth:** 8.0118
+
+**Sensitivity method:** shapley
+
+**Constraints:**
+- max_kl: 0.005
+- min_eap: 0.99
+
+**Metrics:**
+- final_kl: 0.001238
+- final_eap: 0.990277
+- final_etl: 0.009723
+
+See `quantization_config.txt` for full configuration details.

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/chat_template.jinja

@@ -0,0 +1,109 @@
+{{- bos_token }}
+{%- if custom_tools is defined %}
+    {%- set tools = custom_tools %}
+{%- endif %}
+{%- if not tools_in_user_message is defined %}
+    {%- set tools_in_user_message = true %}
+{%- endif %}
+{%- if not date_string is defined %}
+    {%- set date_string = "26 Jul 2024" %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = none %}
+{%- endif %}
+
+{#- This block extracts the system message, so we can slot it into the right place. #}
+{%- if messages[0]['role'] == 'system' %}
+    {%- set system_message = messages[0]['content']|trim %}
+    {%- set messages = messages[1:] %}
+{%- else %}
+    {%- set system_message = "" %}
+{%- endif %}
+
+{#- System message + builtin tools #}
+{{- "<|start_header_id|>system<|end_header_id|>\n\n" }}
+{%- if builtin_tools is defined or tools is not none %}
+    {{- "Environment: ipython\n" }}
+{%- endif %}
+{%- if builtin_tools is defined %}
+    {{- "Tools: " + builtin_tools | reject('equalto', 'code_interpreter') | join(", ") + "\n\n"}}
+{%- endif %}
+{{- "Cutting Knowledge Date: December 2023\n" }}
+{{- "Today Date: " + date_string + "\n\n" }}
+{%- if tools is not none and not tools_in_user_message %}
+    {{- "You have access to the following functions. To call a function, please respond with JSON for a function call." }}
+    {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }}
+    {{- "Do not use variables.\n\n" }}
+    {%- for t in tools %}
+        {{- t | tojson(indent=4) }}
+        {{- "\n\n" }}
+    {%- endfor %}
+{%- endif %}
+{{- system_message }}
+{{- "<|eot_id|>" }}
+
+{#- Custom tools are passed in a user message with some extra guidance #}
+{%- if tools_in_user_message and not tools is none %}
+    {#- Extract the first user message so we can plug it in here #}
+    {%- if messages | length != 0 %}
+        {%- set first_user_message = messages[0]['content']|trim %}
+        {%- set messages = messages[1:] %}
+    {%- else %}
+        {{- raise_exception("Cannot put tools in the first user message when there's no first user message!") }}
+{%- endif %}
+    {{- '<|start_header_id|>user<|end_header_id|>\n\n' -}}
+    {{- "Given the following functions, please respond with a JSON for a function call " }}
+    {{- "with its proper arguments that best answers the given prompt.\n\n" }}
+    {{- 'Respond in the format {"name": function name, "parameters": dictionary of argument name and its value}.' }}
+    {{- "Do not use variables.\n\n" }}
+    {%- for t in tools %}
+        {{- t | tojson(indent=4) }}
+        {{- "\n\n" }}
+    {%- endfor %}
+    {{- first_user_message + "<|eot_id|>"}}
+{%- endif %}
+
+{%- for message in messages %}
+    {%- if not (message.role == 'ipython' or message.role == 'tool' or 'tool_calls' in message) %}
+        {{- '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' }}
+    {%- elif 'tool_calls' in message %}
+        {%- if not message.tool_calls|length == 1 %}
+            {{- raise_exception("This model only supports single tool-calls at once!") }}
+        {%- endif %}
+        {%- set tool_call = message.tool_calls[0].function %}
+        {%- if builtin_tools is defined and tool_call.name in builtin_tools %}
+            {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}}
+            {{- "<|python_tag|>" + tool_call.name + ".call(" }}
+            {%- for arg_name, arg_val in tool_call.arguments | items %}
+                {{- arg_name + '="' + arg_val + '"' }}
+                {%- if not loop.last %}
+                    {{- ", " }}
+                {%- endif %}
+                {%- endfor %}
+            {{- ")" }}
+        {%- else  %}
+            {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' -}}
+            {{- '{"name": "' + tool_call.name + '", ' }}
+            {{- '"parameters": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- "}" }}
+        {%- endif %}
+        {%- if builtin_tools is defined %}
+            {#- This means we're in ipython mode #}
+            {{- "<|eom_id|>" }}
+        {%- else %}
+            {{- "<|eot_id|>" }}
+        {%- endif %}
+    {%- elif message.role == "tool" or message.role == "ipython" %}
+        {{- "<|start_header_id|>ipython<|end_header_id|>\n\n" }}
+        {%- if message.content is mapping or message.content is iterable %}
+            {{- message.content | tojson }}
+        {%- else %}
+            {{- message.content }}
+        {%- endif %}
+        {{- "<|eot_id|>" }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|start_header_id|>assistant<|end_header_id|>\n\n' }}
+{%- endif %}

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.57.3"
+}

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2d066765903924aec0e915a00987ef9ffe0ee366e52368f44884ab71f3e1298
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779c706f56bce5452da8b5fda679c8bd3125b870678e5544b5debd531844d939
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37681ec6ff920f2de78c107de2c44abbf969646f21ada20bde02938b153054e9
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model.safetensors.index.json

@@ -0,0 +1,299 @@
+{
+  "metadata": {
+    "total_parameters": 8030261248,
+    "total_size": 16060522496
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00004-of-00004.safetensors"
+  }
+}

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/quantization_config.txt

@@ -0,0 +1,412 @@
+# Model: meta-llama/Llama-3.1-8B-Instruct
+# Layer directory: /nfs/scistore19/alistgrp/mhelcig/local/data/search/4_5_6_7_8bit_asym_g128/Llama-3.1-8B-Instruct/6bit/
+# Sensitivity method: shapley
+# Estimation method: permutation_separate
+# Available bitwidths: [4, 8]
+# Bitwidth map: {4: 4.156, 5: 5.156, 6: 6.156, 7: 7.156, 8: 8.156}
+#
+# Layer groups: 128 groups (fused layers share bitwidth)
+#   block_0:mlp.gate_proj,mlp.up_proj.block_0:mlp.gate_proj,mlp.up_proj: group 0, 2 layers
+#   block_0:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 1, 3 layers
+#   block_10:mlp.gate_proj,mlp.up_proj.block_10:mlp.gate_proj,mlp.up_proj: group 20, 2 layers
+#   block_10:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 21, 3 layers
+#   block_11:mlp.gate_proj,mlp.up_proj.block_11:mlp.gate_proj,mlp.up_proj: group 22, 2 layers
+#   block_11:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 23, 3 layers
+#   block_12:mlp.gate_proj,mlp.up_proj.block_12:mlp.gate_proj,mlp.up_proj: group 24, 2 layers
+#   block_12:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 25, 3 layers
+#   block_13:mlp.gate_proj,mlp.up_proj.block_13:mlp.gate_proj,mlp.up_proj: group 26, 2 layers
+#   block_13:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 27, 3 layers
+#   block_14:mlp.gate_proj,mlp.up_proj.block_14:mlp.gate_proj,mlp.up_proj: group 28, 2 layers
+#   block_14:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 29, 3 layers
+#   block_15:mlp.gate_proj,mlp.up_proj.block_15:mlp.gate_proj,mlp.up_proj: group 30, 2 layers
+#   block_15:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 31, 3 layers
+#   block_16:mlp.gate_proj,mlp.up_proj.block_16:mlp.gate_proj,mlp.up_proj: group 32, 2 layers
+#   block_16:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 33, 3 layers
+#   block_17:mlp.gate_proj,mlp.up_proj.block_17:mlp.gate_proj,mlp.up_proj: group 34, 2 layers
+#   block_17:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 35, 3 layers
+#   block_18:mlp.gate_proj,mlp.up_proj.block_18:mlp.gate_proj,mlp.up_proj: group 36, 2 layers
+#   block_18:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 37, 3 layers
+#   block_19:mlp.gate_proj,mlp.up_proj.block_19:mlp.gate_proj,mlp.up_proj: group 38, 2 layers
+#   block_19:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 39, 3 layers
+#   block_1:mlp.gate_proj,mlp.up_proj.block_1:mlp.gate_proj,mlp.up_proj: group 2, 2 layers
+#   block_1:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 3, 3 layers
+#   block_20:mlp.gate_proj,mlp.up_proj.block_20:mlp.gate_proj,mlp.up_proj: group 40, 2 layers
+#   block_20:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 41, 3 layers
+#   block_21:mlp.gate_proj,mlp.up_proj.block_21:mlp.gate_proj,mlp.up_proj: group 42, 2 layers
+#   block_21:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 43, 3 layers
+#   block_22:mlp.gate_proj,mlp.up_proj.block_22:mlp.gate_proj,mlp.up_proj: group 44, 2 layers
+#   block_22:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 45, 3 layers
+#   block_23:mlp.gate_proj,mlp.up_proj.block_23:mlp.gate_proj,mlp.up_proj: group 46, 2 layers
+#   block_23:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 47, 3 layers
+#   block_24:mlp.gate_proj,mlp.up_proj.block_24:mlp.gate_proj,mlp.up_proj: group 48, 2 layers
+#   block_24:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 49, 3 layers
+#   block_25:mlp.gate_proj,mlp.up_proj.block_25:mlp.gate_proj,mlp.up_proj: group 50, 2 layers
+#   block_25:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 51, 3 layers
+#   block_26:mlp.gate_proj,mlp.up_proj.block_26:mlp.gate_proj,mlp.up_proj: group 52, 2 layers
+#   block_26:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 53, 3 layers
+#   block_27:mlp.gate_proj,mlp.up_proj.block_27:mlp.gate_proj,mlp.up_proj: group 54, 2 layers
+#   block_27:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 55, 3 layers
+#   block_28:mlp.gate_proj,mlp.up_proj.block_28:mlp.gate_proj,mlp.up_proj: group 56, 2 layers
+#   block_28:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 57, 3 layers
+#   block_29:mlp.gate_proj,mlp.up_proj.block_29:mlp.gate_proj,mlp.up_proj: group 58, 2 layers
+#   block_29:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 59, 3 layers
+#   block_2:mlp.gate_proj,mlp.up_proj.block_2:mlp.gate_proj,mlp.up_proj: group 4, 2 layers
+#   block_2:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 5, 3 layers
+#   block_30:mlp.gate_proj,mlp.up_proj.block_30:mlp.gate_proj,mlp.up_proj: group 60, 2 layers
+#   block_30:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 61, 3 layers
+#   block_31:mlp.gate_proj,mlp.up_proj.block_31:mlp.gate_proj,mlp.up_proj: group 62, 2 layers
+#   block_31:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 63, 3 layers
+#   block_3:mlp.gate_proj,mlp.up_proj.block_3:mlp.gate_proj,mlp.up_proj: group 6, 2 layers
+#   block_3:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 7, 3 layers
+#   block_4:mlp.gate_proj,mlp.up_proj.block_4:mlp.gate_proj,mlp.up_proj: group 8, 2 layers
+#   block_4:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 9, 3 layers
+#   block_5:mlp.gate_proj,mlp.up_proj.block_5:mlp.gate_proj,mlp.up_proj: group 10, 2 layers
+#   block_5:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 11, 3 layers
+#   block_6:mlp.gate_proj,mlp.up_proj.block_6:mlp.gate_proj,mlp.up_proj: group 12, 2 layers
+#   block_6:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 13, 3 layers
+#   block_7:mlp.gate_proj,mlp.up_proj.block_7:mlp.gate_proj,mlp.up_proj: group 14, 2 layers
+#   block_7:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 15, 3 layers
+#   block_8:mlp.gate_proj,mlp.up_proj.block_8:mlp.gate_proj,mlp.up_proj: group 16, 2 layers
+#   block_8:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 17, 3 layers
+#   block_9:mlp.gate_proj,mlp.up_proj.block_9:mlp.gate_proj,mlp.up_proj: group 18, 2 layers
+#   block_9:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 19, 3 layers
+#   Fused in model.layers.0:
+#     - self_attn.o_proj (group 64, 1 layers)
+#     - mlp.down_proj (group 65, 1 layers)
+#   Fused in model.layers.1:
+#     - self_attn.o_proj (group 66, 1 layers)
+#     - mlp.down_proj (group 67, 1 layers)
+#   Fused in model.layers.10:
+#     - self_attn.o_proj (group 84, 1 layers)
+#     - mlp.down_proj (group 85, 1 layers)
+#   Fused in model.layers.11:
+#     - self_attn.o_proj (group 86, 1 layers)
+#     - mlp.down_proj (group 87, 1 layers)
+#   Fused in model.layers.12:
+#     - self_attn.o_proj (group 88, 1 layers)
+#     - mlp.down_proj (group 89, 1 layers)
+#   Fused in model.layers.13:
+#     - self_attn.o_proj (group 90, 1 layers)
+#     - mlp.down_proj (group 91, 1 layers)
+#   Fused in model.layers.14:
+#     - self_attn.o_proj (group 92, 1 layers)
+#     - mlp.down_proj (group 93, 1 layers)
+#   Fused in model.layers.15:
+#     - self_attn.o_proj (group 94, 1 layers)
+#     - mlp.down_proj (group 95, 1 layers)
+#   Fused in model.layers.16:
+#     - self_attn.o_proj (group 96, 1 layers)
+#     - mlp.down_proj (group 97, 1 layers)
+#   Fused in model.layers.17:
+#     - self_attn.o_proj (group 98, 1 layers)
+#     - mlp.down_proj (group 99, 1 layers)
+#   Fused in model.layers.18:
+#     - self_attn.o_proj (group 100, 1 layers)
+#     - mlp.down_proj (group 101, 1 layers)
+#   Fused in model.layers.19:
+#     - self_attn.o_proj (group 102, 1 layers)
+#     - mlp.down_proj (group 103, 1 layers)
+#   Fused in model.layers.2:
+#     - self_attn.o_proj (group 68, 1 layers)
+#     - mlp.down_proj (group 69, 1 layers)
+#   Fused in model.layers.20:
+#     - self_attn.o_proj (group 104, 1 layers)
+#     - mlp.down_proj (group 105, 1 layers)
+#   Fused in model.layers.21:
+#     - self_attn.o_proj (group 106, 1 layers)
+#     - mlp.down_proj (group 107, 1 layers)
+#   Fused in model.layers.22:
+#     - self_attn.o_proj (group 108, 1 layers)
+#     - mlp.down_proj (group 109, 1 layers)
+#   Fused in model.layers.23:
+#     - self_attn.o_proj (group 110, 1 layers)
+#     - mlp.down_proj (group 111, 1 layers)
+#   Fused in model.layers.24:
+#     - self_attn.o_proj (group 112, 1 layers)
+#     - mlp.down_proj (group 113, 1 layers)
+#   Fused in model.layers.25:
+#     - self_attn.o_proj (group 114, 1 layers)
+#     - mlp.down_proj (group 115, 1 layers)
+#   Fused in model.layers.26:
+#     - self_attn.o_proj (group 116, 1 layers)
+#     - mlp.down_proj (group 117, 1 layers)
+#   Fused in model.layers.27:
+#     - self_attn.o_proj (group 118, 1 layers)
+#     - mlp.down_proj (group 119, 1 layers)
+#   Fused in model.layers.28:
+#     - self_attn.o_proj (group 120, 1 layers)
+#     - mlp.down_proj (group 121, 1 layers)
+#   Fused in model.layers.29:
+#     - self_attn.o_proj (group 122, 1 layers)
+#     - mlp.down_proj (group 123, 1 layers)
+#   Fused in model.layers.3:
+#     - self_attn.o_proj (group 70, 1 layers)
+#     - mlp.down_proj (group 71, 1 layers)
+#   Fused in model.layers.30:
+#     - self_attn.o_proj (group 124, 1 layers)
+#     - mlp.down_proj (group 125, 1 layers)
+#   Fused in model.layers.31:
+#     - self_attn.o_proj (group 126, 1 layers)
+#     - mlp.down_proj (group 127, 1 layers)
+#   Fused in model.layers.4:
+#     - self_attn.o_proj (group 72, 1 layers)
+#     - mlp.down_proj (group 73, 1 layers)
+#   Fused in model.layers.5:
+#     - self_attn.o_proj (group 74, 1 layers)
+#     - mlp.down_proj (group 75, 1 layers)
+#   Fused in model.layers.6:
+#     - self_attn.o_proj (group 76, 1 layers)
+#     - mlp.down_proj (group 77, 1 layers)
+#   Fused in model.layers.7:
+#     - self_attn.o_proj (group 78, 1 layers)
+#     - mlp.down_proj (group 79, 1 layers)
+#   Fused in model.layers.8:
+#     - self_attn.o_proj (group 80, 1 layers)
+#     - mlp.down_proj (group 81, 1 layers)
+#   Fused in model.layers.9:
+#     - self_attn.o_proj (group 82, 1 layers)
+#     - mlp.down_proj (group 83, 1 layers)
+#
+# Mode: binary_search_constraint (measured)
+# Constraint max_kl: 0.005
+# Constraint min_eap: 0.99
+# Weights: nll=0.0, kl=0.0, eap=1.0
+#
+# Average bitwidth: 8.0118
+# Total params: 6979321856
+# Total bits: 55916716098
+# Final KL: 0.001238
+# Final EAP: 0.990277
+# Final ETL: 0.009723
+# Satisfies constraints: True
+# Solver calls: 9
+# Evaluations: 9
+#
+# Bitwidth distribution:
+#   8-bit: 219 layers (97.8%)
+#   4-bit: 5 layers (2.2%)
+#
+model.layers.0.mlp.gate_proj: 8
+model.layers.0.mlp.up_proj: 8
+model.layers.0.self_attn.k_proj: 8
+model.layers.0.self_attn.q_proj: 8
+model.layers.0.self_attn.v_proj: 8
+model.layers.1.mlp.gate_proj: 8
+model.layers.1.mlp.up_proj: 8
+model.layers.1.self_attn.k_proj: 8
+model.layers.1.self_attn.q_proj: 8
+model.layers.1.self_attn.v_proj: 8
+model.layers.2.mlp.gate_proj: 8
+model.layers.2.mlp.up_proj: 8
+model.layers.2.self_attn.k_proj: 8
+model.layers.2.self_attn.q_proj: 8
+model.layers.2.self_attn.v_proj: 8
+model.layers.3.mlp.gate_proj: 8
+model.layers.3.mlp.up_proj: 8
+model.layers.3.self_attn.k_proj: 8
+model.layers.3.self_attn.q_proj: 8
+model.layers.3.self_attn.v_proj: 8
+model.layers.4.mlp.gate_proj: 8
+model.layers.4.mlp.up_proj: 8
+model.layers.4.self_attn.k_proj: 8
+model.layers.4.self_attn.q_proj: 8
+model.layers.4.self_attn.v_proj: 8
+model.layers.5.mlp.gate_proj: 8
+model.layers.5.mlp.up_proj: 8
+model.layers.5.self_attn.k_proj: 8
+model.layers.5.self_attn.q_proj: 8
+model.layers.5.self_attn.v_proj: 8
+model.layers.6.mlp.gate_proj: 8
+model.layers.6.mlp.up_proj: 8
+model.layers.6.self_attn.k_proj: 8
+model.layers.6.self_attn.q_proj: 8
+model.layers.6.self_attn.v_proj: 8
+model.layers.7.mlp.gate_proj: 8
+model.layers.7.mlp.up_proj: 8
+model.layers.7.self_attn.k_proj: 8
+model.layers.7.self_attn.q_proj: 8
+model.layers.7.self_attn.v_proj: 8
+model.layers.8.mlp.gate_proj: 8
+model.layers.8.mlp.up_proj: 8
+model.layers.8.self_attn.k_proj: 8
+model.layers.8.self_attn.q_proj: 8
+model.layers.8.self_attn.v_proj: 8
+model.layers.9.mlp.gate_proj: 8
+model.layers.9.mlp.up_proj: 8
+model.layers.9.self_attn.k_proj: 8
+model.layers.9.self_attn.q_proj: 8
+model.layers.9.self_attn.v_proj: 8
+model.layers.10.mlp.gate_proj: 8
+model.layers.10.mlp.up_proj: 8
+model.layers.10.self_attn.k_proj: 8
+model.layers.10.self_attn.q_proj: 8
+model.layers.10.self_attn.v_proj: 8
+model.layers.11.mlp.gate_proj: 8
+model.layers.11.mlp.up_proj: 8
+model.layers.11.self_attn.k_proj: 8
+model.layers.11.self_attn.q_proj: 8
+model.layers.11.self_attn.v_proj: 8
+model.layers.12.mlp.gate_proj: 8
+model.layers.12.mlp.up_proj: 8
+model.layers.12.self_attn.k_proj: 8
+model.layers.12.self_attn.q_proj: 8
+model.layers.12.self_attn.v_proj: 8
+model.layers.13.mlp.gate_proj: 8
+model.layers.13.mlp.up_proj: 8
+model.layers.13.self_attn.k_proj: 8
+model.layers.13.self_attn.q_proj: 8
+model.layers.13.self_attn.v_proj: 8
+model.layers.14.mlp.gate_proj: 8
+model.layers.14.mlp.up_proj: 8
+model.layers.14.self_attn.k_proj: 8
+model.layers.14.self_attn.q_proj: 8
+model.layers.14.self_attn.v_proj: 8
+model.layers.15.mlp.gate_proj: 8
+model.layers.15.mlp.up_proj: 8
+model.layers.15.self_attn.k_proj: 8
+model.layers.15.self_attn.q_proj: 8
+model.layers.15.self_attn.v_proj: 8
+model.layers.16.mlp.gate_proj: 8
+model.layers.16.mlp.up_proj: 8
+model.layers.16.self_attn.k_proj: 8
+model.layers.16.self_attn.q_proj: 8
+model.layers.16.self_attn.v_proj: 8
+model.layers.17.mlp.gate_proj: 8
+model.layers.17.mlp.up_proj: 8
+model.layers.17.self_attn.k_proj: 8
+model.layers.17.self_attn.q_proj: 8
+model.layers.17.self_attn.v_proj: 8
+model.layers.18.mlp.gate_proj: 8
+model.layers.18.mlp.up_proj: 8
+model.layers.18.self_attn.k_proj: 8
+model.layers.18.self_attn.q_proj: 8
+model.layers.18.self_attn.v_proj: 8
+model.layers.19.mlp.gate_proj: 8
+model.layers.19.mlp.up_proj: 8
+model.layers.19.self_attn.k_proj: 8
+model.layers.19.self_attn.q_proj: 8
+model.layers.19.self_attn.v_proj: 8
+model.layers.20.mlp.gate_proj: 8
+model.layers.20.mlp.up_proj: 8
+model.layers.20.self_attn.k_proj: 8
+model.layers.20.self_attn.q_proj: 8
+model.layers.20.self_attn.v_proj: 8
+model.layers.21.mlp.gate_proj: 8
+model.layers.21.mlp.up_proj: 8
+model.layers.21.self_attn.k_proj: 8
+model.layers.21.self_attn.q_proj: 8
+model.layers.21.self_attn.v_proj: 8
+model.layers.22.mlp.gate_proj: 8
+model.layers.22.mlp.up_proj: 8
+model.layers.22.self_attn.k_proj: 8
+model.layers.22.self_attn.q_proj: 8
+model.layers.22.self_attn.v_proj: 8
+model.layers.23.mlp.gate_proj: 8
+model.layers.23.mlp.up_proj: 8
+model.layers.23.self_attn.k_proj: 8
+model.layers.23.self_attn.q_proj: 8
+model.layers.23.self_attn.v_proj: 8
+model.layers.24.mlp.gate_proj: 8
+model.layers.24.mlp.up_proj: 8
+model.layers.24.self_attn.k_proj: 8
+model.layers.24.self_attn.q_proj: 8
+model.layers.24.self_attn.v_proj: 8
+model.layers.25.mlp.gate_proj: 8
+model.layers.25.mlp.up_proj: 8
+model.layers.25.self_attn.k_proj: 8
+model.layers.25.self_attn.q_proj: 8
+model.layers.25.self_attn.v_proj: 8
+model.layers.26.mlp.gate_proj: 8
+model.layers.26.mlp.up_proj: 8
+model.layers.26.self_attn.k_proj: 8
+model.layers.26.self_attn.q_proj: 8
+model.layers.26.self_attn.v_proj: 8
+model.layers.27.mlp.gate_proj: 8
+model.layers.27.mlp.up_proj: 8
+model.layers.27.self_attn.k_proj: 8
+model.layers.27.self_attn.q_proj: 8
+model.layers.27.self_attn.v_proj: 8
+model.layers.28.mlp.gate_proj: 8
+model.layers.28.mlp.up_proj: 8
+model.layers.28.self_attn.k_proj: 8
+model.layers.28.self_attn.q_proj: 8
+model.layers.28.self_attn.v_proj: 8
+model.layers.29.mlp.gate_proj: 8
+model.layers.29.mlp.up_proj: 8
+model.layers.29.self_attn.k_proj: 8
+model.layers.29.self_attn.q_proj: 8
+model.layers.29.self_attn.v_proj: 8
+model.layers.30.mlp.gate_proj: 8
+model.layers.30.mlp.up_proj: 8
+model.layers.30.self_attn.k_proj: 8
+model.layers.30.self_attn.q_proj: 8
+model.layers.30.self_attn.v_proj: 8
+model.layers.31.mlp.gate_proj: 8
+model.layers.31.mlp.up_proj: 8
+model.layers.31.self_attn.k_proj: 8
+model.layers.31.self_attn.q_proj: 8
+model.layers.31.self_attn.v_proj: 8
+model.layers.0.self_attn.o_proj: 8
+model.layers.0.mlp.down_proj: 8
+model.layers.1.self_attn.o_proj: 8
+model.layers.1.mlp.down_proj: 8
+model.layers.2.self_attn.o_proj: 8
+model.layers.2.mlp.down_proj: 8
+model.layers.3.self_attn.o_proj: 8
+model.layers.3.mlp.down_proj: 8
+model.layers.4.self_attn.o_proj: 8
+model.layers.4.mlp.down_proj: 8
+model.layers.5.self_attn.o_proj: 8
+model.layers.5.mlp.down_proj: 8
+model.layers.6.self_attn.o_proj: 8
+model.layers.6.mlp.down_proj: 8
+model.layers.7.self_attn.o_proj: 8
+model.layers.7.mlp.down_proj: 8
+model.layers.8.self_attn.o_proj: 8
+model.layers.8.mlp.down_proj: 8
+model.layers.9.self_attn.o_proj: 8
+model.layers.9.mlp.down_proj: 8
+model.layers.10.self_attn.o_proj: 8
+model.layers.10.mlp.down_proj: 8
+model.layers.11.self_attn.o_proj: 8
+model.layers.11.mlp.down_proj: 8
+model.layers.12.self_attn.o_proj: 8
+model.layers.12.mlp.down_proj: 8
+model.layers.13.self_attn.o_proj: 8
+model.layers.13.mlp.down_proj: 8
+model.layers.14.self_attn.o_proj: 8
+model.layers.14.mlp.down_proj: 8
+model.layers.15.self_attn.o_proj: 8
+model.layers.15.mlp.down_proj: 8
+model.layers.16.self_attn.o_proj: 8
+model.layers.16.mlp.down_proj: 8
+model.layers.17.self_attn.o_proj: 8
+model.layers.17.mlp.down_proj: 8
+model.layers.18.self_attn.o_proj: 8
+model.layers.18.mlp.down_proj: 8
+model.layers.19.self_attn.o_proj: 8
+model.layers.19.mlp.down_proj: 8
+model.layers.20.self_attn.o_proj: 8
+model.layers.20.mlp.down_proj: 8
+model.layers.21.self_attn.o_proj: 8
+model.layers.21.mlp.down_proj: 8
+model.layers.22.self_attn.o_proj: 8
+model.layers.22.mlp.down_proj: 8
+model.layers.23.self_attn.o_proj: 8
+model.layers.23.mlp.down_proj: 8
+model.layers.24.self_attn.o_proj: 8
+model.layers.24.mlp.down_proj: 8
+model.layers.25.self_attn.o_proj: 4
+model.layers.25.mlp.down_proj: 8
+model.layers.26.self_attn.o_proj: 8
+model.layers.26.mlp.down_proj: 4
+model.layers.27.self_attn.o_proj: 8
+model.layers.27.mlp.down_proj: 4
+model.layers.28.self_attn.o_proj: 8
+model.layers.28.mlp.down_proj: 4
+model.layers.29.self_attn.o_proj: 8
+model.layers.29.mlp.down_proj: 4
+model.layers.30.self_attn.o_proj: 8
+model.layers.30.mlp.down_proj: 8
+model.layers.31.self_attn.o_proj: 8
+model.layers.31.mlp.down_proj: 8

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/special_tokens_map.json

@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2d066765903924aec0e915a00987ef9ffe0ee366e52368f44884ab71f3e1298
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779c706f56bce5452da8b5fda679c8bd3125b870678e5544b5debd531844d939
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7426f0cdef097b0b84a4403954fae06642649a7c4a06bdcf7cc39f963e75e40
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.005_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:949a717376d70bd7c14cb9f0a04ef3e48fa6c9571a0959dbad1d7816052530b1
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f79c9f35494ee28ac2f307fda7ef1f0c699855fcd9fd88db82f5e2e1d532273e
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:553898c0a66462cf9da4b62e3c4b60322a464a5ecd0f4ddaf5c3b653ba632710
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.68_4-8bit_grouped_seed1234/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:949a717376d70bd7c14cb9f0a04ef3e48fa6c9571a0959dbad1d7816052530b1
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f79c9f35494ee28ac2f307fda7ef1f0c699855fcd9fd88db82f5e2e1d532273e
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b9d2c8849bb6be4634cc740b7d5d3cf942b8eccd9b7bb23b7113b21c25a3d1a
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/quantization_config.txt

@@ -0,0 +1,412 @@
+# Model: meta-llama/Llama-3.1-8B-Instruct
+# Layer directory: /nfs/scistore19/alistgrp/mhelcig/local/data/search/4_5_6_7_8bit_asym_g128/Llama-3.1-8B-Instruct/6bit/
+# Sensitivity method: shapley
+# Estimation method: permutation_separate
+# Available bitwidths: [4, 8]
+# Bitwidth map: {4: 4.156, 5: 5.156, 6: 6.156, 7: 7.156, 8: 8.156}
+#
+# Layer groups: 128 groups (fused layers share bitwidth)
+#   block_0:mlp.gate_proj,mlp.up_proj.block_0:mlp.gate_proj,mlp.up_proj: group 0, 2 layers
+#   block_0:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 1, 3 layers
+#   block_10:mlp.gate_proj,mlp.up_proj.block_10:mlp.gate_proj,mlp.up_proj: group 20, 2 layers
+#   block_10:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 21, 3 layers
+#   block_11:mlp.gate_proj,mlp.up_proj.block_11:mlp.gate_proj,mlp.up_proj: group 22, 2 layers
+#   block_11:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 23, 3 layers
+#   block_12:mlp.gate_proj,mlp.up_proj.block_12:mlp.gate_proj,mlp.up_proj: group 24, 2 layers
+#   block_12:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 25, 3 layers
+#   block_13:mlp.gate_proj,mlp.up_proj.block_13:mlp.gate_proj,mlp.up_proj: group 26, 2 layers
+#   block_13:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 27, 3 layers
+#   block_14:mlp.gate_proj,mlp.up_proj.block_14:mlp.gate_proj,mlp.up_proj: group 28, 2 layers
+#   block_14:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 29, 3 layers
+#   block_15:mlp.gate_proj,mlp.up_proj.block_15:mlp.gate_proj,mlp.up_proj: group 30, 2 layers
+#   block_15:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 31, 3 layers
+#   block_16:mlp.gate_proj,mlp.up_proj.block_16:mlp.gate_proj,mlp.up_proj: group 32, 2 layers
+#   block_16:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 33, 3 layers
+#   block_17:mlp.gate_proj,mlp.up_proj.block_17:mlp.gate_proj,mlp.up_proj: group 34, 2 layers
+#   block_17:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 35, 3 layers
+#   block_18:mlp.gate_proj,mlp.up_proj.block_18:mlp.gate_proj,mlp.up_proj: group 36, 2 layers
+#   block_18:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 37, 3 layers
+#   block_19:mlp.gate_proj,mlp.up_proj.block_19:mlp.gate_proj,mlp.up_proj: group 38, 2 layers
+#   block_19:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 39, 3 layers
+#   block_1:mlp.gate_proj,mlp.up_proj.block_1:mlp.gate_proj,mlp.up_proj: group 2, 2 layers
+#   block_1:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 3, 3 layers
+#   block_20:mlp.gate_proj,mlp.up_proj.block_20:mlp.gate_proj,mlp.up_proj: group 40, 2 layers
+#   block_20:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 41, 3 layers
+#   block_21:mlp.gate_proj,mlp.up_proj.block_21:mlp.gate_proj,mlp.up_proj: group 42, 2 layers
+#   block_21:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 43, 3 layers
+#   block_22:mlp.gate_proj,mlp.up_proj.block_22:mlp.gate_proj,mlp.up_proj: group 44, 2 layers
+#   block_22:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 45, 3 layers
+#   block_23:mlp.gate_proj,mlp.up_proj.block_23:mlp.gate_proj,mlp.up_proj: group 46, 2 layers
+#   block_23:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 47, 3 layers
+#   block_24:mlp.gate_proj,mlp.up_proj.block_24:mlp.gate_proj,mlp.up_proj: group 48, 2 layers
+#   block_24:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 49, 3 layers
+#   block_25:mlp.gate_proj,mlp.up_proj.block_25:mlp.gate_proj,mlp.up_proj: group 50, 2 layers
+#   block_25:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 51, 3 layers
+#   block_26:mlp.gate_proj,mlp.up_proj.block_26:mlp.gate_proj,mlp.up_proj: group 52, 2 layers
+#   block_26:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 53, 3 layers
+#   block_27:mlp.gate_proj,mlp.up_proj.block_27:mlp.gate_proj,mlp.up_proj: group 54, 2 layers
+#   block_27:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 55, 3 layers
+#   block_28:mlp.gate_proj,mlp.up_proj.block_28:mlp.gate_proj,mlp.up_proj: group 56, 2 layers
+#   block_28:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 57, 3 layers
+#   block_29:mlp.gate_proj,mlp.up_proj.block_29:mlp.gate_proj,mlp.up_proj: group 58, 2 layers
+#   block_29:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 59, 3 layers
+#   block_2:mlp.gate_proj,mlp.up_proj.block_2:mlp.gate_proj,mlp.up_proj: group 4, 2 layers
+#   block_2:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 5, 3 layers
+#   block_30:mlp.gate_proj,mlp.up_proj.block_30:mlp.gate_proj,mlp.up_proj: group 60, 2 layers
+#   block_30:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 61, 3 layers
+#   block_31:mlp.gate_proj,mlp.up_proj.block_31:mlp.gate_proj,mlp.up_proj: group 62, 2 layers
+#   block_31:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 63, 3 layers
+#   block_3:mlp.gate_proj,mlp.up_proj.block_3:mlp.gate_proj,mlp.up_proj: group 6, 2 layers
+#   block_3:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 7, 3 layers
+#   block_4:mlp.gate_proj,mlp.up_proj.block_4:mlp.gate_proj,mlp.up_proj: group 8, 2 layers
+#   block_4:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 9, 3 layers
+#   block_5:mlp.gate_proj,mlp.up_proj.block_5:mlp.gate_proj,mlp.up_proj: group 10, 2 layers
+#   block_5:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 11, 3 layers
+#   block_6:mlp.gate_proj,mlp.up_proj.block_6:mlp.gate_proj,mlp.up_proj: group 12, 2 layers
+#   block_6:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 13, 3 layers
+#   block_7:mlp.gate_proj,mlp.up_proj.block_7:mlp.gate_proj,mlp.up_proj: group 14, 2 layers
+#   block_7:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 15, 3 layers
+#   block_8:mlp.gate_proj,mlp.up_proj.block_8:mlp.gate_proj,mlp.up_proj: group 16, 2 layers
+#   block_8:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 17, 3 layers
+#   block_9:mlp.gate_proj,mlp.up_proj.block_9:mlp.gate_proj,mlp.up_proj: group 18, 2 layers
+#   block_9:self_attn.k_proj,self_attn.q_proj,self_attn.v_proj: group 19, 3 layers
+#   Fused in model.layers.0:
+#     - self_attn.o_proj (group 64, 1 layers)
+#     - mlp.down_proj (group 65, 1 layers)
+#   Fused in model.layers.1:
+#     - self_attn.o_proj (group 66, 1 layers)
+#     - mlp.down_proj (group 67, 1 layers)
+#   Fused in model.layers.10:
+#     - self_attn.o_proj (group 84, 1 layers)
+#     - mlp.down_proj (group 85, 1 layers)
+#   Fused in model.layers.11:
+#     - self_attn.o_proj (group 86, 1 layers)
+#     - mlp.down_proj (group 87, 1 layers)
+#   Fused in model.layers.12:
+#     - self_attn.o_proj (group 88, 1 layers)
+#     - mlp.down_proj (group 89, 1 layers)
+#   Fused in model.layers.13:
+#     - self_attn.o_proj (group 90, 1 layers)
+#     - mlp.down_proj (group 91, 1 layers)
+#   Fused in model.layers.14:
+#     - self_attn.o_proj (group 92, 1 layers)
+#     - mlp.down_proj (group 93, 1 layers)
+#   Fused in model.layers.15:
+#     - self_attn.o_proj (group 94, 1 layers)
+#     - mlp.down_proj (group 95, 1 layers)
+#   Fused in model.layers.16:
+#     - self_attn.o_proj (group 96, 1 layers)
+#     - mlp.down_proj (group 97, 1 layers)
+#   Fused in model.layers.17:
+#     - self_attn.o_proj (group 98, 1 layers)
+#     - mlp.down_proj (group 99, 1 layers)
+#   Fused in model.layers.18:
+#     - self_attn.o_proj (group 100, 1 layers)
+#     - mlp.down_proj (group 101, 1 layers)
+#   Fused in model.layers.19:
+#     - self_attn.o_proj (group 102, 1 layers)
+#     - mlp.down_proj (group 103, 1 layers)
+#   Fused in model.layers.2:
+#     - self_attn.o_proj (group 68, 1 layers)
+#     - mlp.down_proj (group 69, 1 layers)
+#   Fused in model.layers.20:
+#     - self_attn.o_proj (group 104, 1 layers)
+#     - mlp.down_proj (group 105, 1 layers)
+#   Fused in model.layers.21:
+#     - self_attn.o_proj (group 106, 1 layers)
+#     - mlp.down_proj (group 107, 1 layers)
+#   Fused in model.layers.22:
+#     - self_attn.o_proj (group 108, 1 layers)
+#     - mlp.down_proj (group 109, 1 layers)
+#   Fused in model.layers.23:
+#     - self_attn.o_proj (group 110, 1 layers)
+#     - mlp.down_proj (group 111, 1 layers)
+#   Fused in model.layers.24:
+#     - self_attn.o_proj (group 112, 1 layers)
+#     - mlp.down_proj (group 113, 1 layers)
+#   Fused in model.layers.25:
+#     - self_attn.o_proj (group 114, 1 layers)
+#     - mlp.down_proj (group 115, 1 layers)
+#   Fused in model.layers.26:
+#     - self_attn.o_proj (group 116, 1 layers)
+#     - mlp.down_proj (group 117, 1 layers)
+#   Fused in model.layers.27:
+#     - self_attn.o_proj (group 118, 1 layers)
+#     - mlp.down_proj (group 119, 1 layers)
+#   Fused in model.layers.28:
+#     - self_attn.o_proj (group 120, 1 layers)
+#     - mlp.down_proj (group 121, 1 layers)
+#   Fused in model.layers.29:
+#     - self_attn.o_proj (group 122, 1 layers)
+#     - mlp.down_proj (group 123, 1 layers)
+#   Fused in model.layers.3:
+#     - self_attn.o_proj (group 70, 1 layers)
+#     - mlp.down_proj (group 71, 1 layers)
+#   Fused in model.layers.30:
+#     - self_attn.o_proj (group 124, 1 layers)
+#     - mlp.down_proj (group 125, 1 layers)
+#   Fused in model.layers.31:
+#     - self_attn.o_proj (group 126, 1 layers)
+#     - mlp.down_proj (group 127, 1 layers)
+#   Fused in model.layers.4:
+#     - self_attn.o_proj (group 72, 1 layers)
+#     - mlp.down_proj (group 73, 1 layers)
+#   Fused in model.layers.5:
+#     - self_attn.o_proj (group 74, 1 layers)
+#     - mlp.down_proj (group 75, 1 layers)
+#   Fused in model.layers.6:
+#     - self_attn.o_proj (group 76, 1 layers)
+#     - mlp.down_proj (group 77, 1 layers)
+#   Fused in model.layers.7:
+#     - self_attn.o_proj (group 78, 1 layers)
+#     - mlp.down_proj (group 79, 1 layers)
+#   Fused in model.layers.8:
+#     - self_attn.o_proj (group 80, 1 layers)
+#     - mlp.down_proj (group 81, 1 layers)
+#   Fused in model.layers.9:
+#     - self_attn.o_proj (group 82, 1 layers)
+#     - mlp.down_proj (group 83, 1 layers)
+#
+# Mode: binary_search_constraint (measured)
+# Constraint max_kl: 0.01
+# Constraint min_eap: 0.985
+# Weights: nll=0.0, kl=0.0, eap=1.0
+#
+# Average bitwidth: 7.7089
+# Total params: 6979321856
+# Total bits: 53802786882
+# Final KL: 0.002595
+# Final EAP: 0.985184
+# Final ETL: 0.014816
+# Satisfies constraints: True
+# Solver calls: 9
+# Evaluations: 9
+#
+# Bitwidth distribution:
+#   8-bit: 210 layers (93.8%)
+#   4-bit: 14 layers (6.2%)
+#
+model.layers.0.mlp.gate_proj: 4
+model.layers.0.mlp.up_proj: 4
+model.layers.0.self_attn.k_proj: 8
+model.layers.0.self_attn.q_proj: 8
+model.layers.0.self_attn.v_proj: 8
+model.layers.1.mlp.gate_proj: 8
+model.layers.1.mlp.up_proj: 8
+model.layers.1.self_attn.k_proj: 8
+model.layers.1.self_attn.q_proj: 8
+model.layers.1.self_attn.v_proj: 8
+model.layers.2.mlp.gate_proj: 8
+model.layers.2.mlp.up_proj: 8
+model.layers.2.self_attn.k_proj: 8
+model.layers.2.self_attn.q_proj: 8
+model.layers.2.self_attn.v_proj: 8
+model.layers.3.mlp.gate_proj: 8
+model.layers.3.mlp.up_proj: 8
+model.layers.3.self_attn.k_proj: 8
+model.layers.3.self_attn.q_proj: 8
+model.layers.3.self_attn.v_proj: 8
+model.layers.4.mlp.gate_proj: 8
+model.layers.4.mlp.up_proj: 8
+model.layers.4.self_attn.k_proj: 8
+model.layers.4.self_attn.q_proj: 8
+model.layers.4.self_attn.v_proj: 8
+model.layers.5.mlp.gate_proj: 8
+model.layers.5.mlp.up_proj: 8
+model.layers.5.self_attn.k_proj: 8
+model.layers.5.self_attn.q_proj: 8
+model.layers.5.self_attn.v_proj: 8
+model.layers.6.mlp.gate_proj: 8
+model.layers.6.mlp.up_proj: 8
+model.layers.6.self_attn.k_proj: 8
+model.layers.6.self_attn.q_proj: 8
+model.layers.6.self_attn.v_proj: 8
+model.layers.7.mlp.gate_proj: 8
+model.layers.7.mlp.up_proj: 8
+model.layers.7.self_attn.k_proj: 8
+model.layers.7.self_attn.q_proj: 8
+model.layers.7.self_attn.v_proj: 8
+model.layers.8.mlp.gate_proj: 8
+model.layers.8.mlp.up_proj: 8
+model.layers.8.self_attn.k_proj: 8
+model.layers.8.self_attn.q_proj: 8
+model.layers.8.self_attn.v_proj: 8
+model.layers.9.mlp.gate_proj: 8
+model.layers.9.mlp.up_proj: 8
+model.layers.9.self_attn.k_proj: 8
+model.layers.9.self_attn.q_proj: 8
+model.layers.9.self_attn.v_proj: 8
+model.layers.10.mlp.gate_proj: 8
+model.layers.10.mlp.up_proj: 8
+model.layers.10.self_attn.k_proj: 8
+model.layers.10.self_attn.q_proj: 8
+model.layers.10.self_attn.v_proj: 8
+model.layers.11.mlp.gate_proj: 8
+model.layers.11.mlp.up_proj: 8
+model.layers.11.self_attn.k_proj: 8
+model.layers.11.self_attn.q_proj: 8
+model.layers.11.self_attn.v_proj: 8
+model.layers.12.mlp.gate_proj: 8
+model.layers.12.mlp.up_proj: 8
+model.layers.12.self_attn.k_proj: 8
+model.layers.12.self_attn.q_proj: 8
+model.layers.12.self_attn.v_proj: 8
+model.layers.13.mlp.gate_proj: 8
+model.layers.13.mlp.up_proj: 8
+model.layers.13.self_attn.k_proj: 8
+model.layers.13.self_attn.q_proj: 8
+model.layers.13.self_attn.v_proj: 8
+model.layers.14.mlp.gate_proj: 8
+model.layers.14.mlp.up_proj: 8
+model.layers.14.self_attn.k_proj: 8
+model.layers.14.self_attn.q_proj: 8
+model.layers.14.self_attn.v_proj: 8
+model.layers.15.mlp.gate_proj: 8
+model.layers.15.mlp.up_proj: 8
+model.layers.15.self_attn.k_proj: 8
+model.layers.15.self_attn.q_proj: 8
+model.layers.15.self_attn.v_proj: 8
+model.layers.16.mlp.gate_proj: 8
+model.layers.16.mlp.up_proj: 8
+model.layers.16.self_attn.k_proj: 8
+model.layers.16.self_attn.q_proj: 8
+model.layers.16.self_attn.v_proj: 8
+model.layers.17.mlp.gate_proj: 8
+model.layers.17.mlp.up_proj: 8
+model.layers.17.self_attn.k_proj: 8
+model.layers.17.self_attn.q_proj: 8
+model.layers.17.self_attn.v_proj: 8
+model.layers.18.mlp.gate_proj: 8
+model.layers.18.mlp.up_proj: 8
+model.layers.18.self_attn.k_proj: 8
+model.layers.18.self_attn.q_proj: 8
+model.layers.18.self_attn.v_proj: 8
+model.layers.19.mlp.gate_proj: 8
+model.layers.19.mlp.up_proj: 8
+model.layers.19.self_attn.k_proj: 8
+model.layers.19.self_attn.q_proj: 8
+model.layers.19.self_attn.v_proj: 8
+model.layers.20.mlp.gate_proj: 8
+model.layers.20.mlp.up_proj: 8
+model.layers.20.self_attn.k_proj: 8
+model.layers.20.self_attn.q_proj: 8
+model.layers.20.self_attn.v_proj: 8
+model.layers.21.mlp.gate_proj: 8
+model.layers.21.mlp.up_proj: 8
+model.layers.21.self_attn.k_proj: 8
+model.layers.21.self_attn.q_proj: 8
+model.layers.21.self_attn.v_proj: 8
+model.layers.22.mlp.gate_proj: 8
+model.layers.22.mlp.up_proj: 8
+model.layers.22.self_attn.k_proj: 8
+model.layers.22.self_attn.q_proj: 8
+model.layers.22.self_attn.v_proj: 8
+model.layers.23.mlp.gate_proj: 8
+model.layers.23.mlp.up_proj: 8
+model.layers.23.self_attn.k_proj: 8
+model.layers.23.self_attn.q_proj: 8
+model.layers.23.self_attn.v_proj: 8
+model.layers.24.mlp.gate_proj: 8
+model.layers.24.mlp.up_proj: 8
+model.layers.24.self_attn.k_proj: 8
+model.layers.24.self_attn.q_proj: 8
+model.layers.24.self_attn.v_proj: 8
+model.layers.25.mlp.gate_proj: 8
+model.layers.25.mlp.up_proj: 8
+model.layers.25.self_attn.k_proj: 8
+model.layers.25.self_attn.q_proj: 8
+model.layers.25.self_attn.v_proj: 8
+model.layers.26.mlp.gate_proj: 8
+model.layers.26.mlp.up_proj: 8
+model.layers.26.self_attn.k_proj: 8
+model.layers.26.self_attn.q_proj: 8
+model.layers.26.self_attn.v_proj: 8
+model.layers.27.mlp.gate_proj: 8
+model.layers.27.mlp.up_proj: 8
+model.layers.27.self_attn.k_proj: 8
+model.layers.27.self_attn.q_proj: 8
+model.layers.27.self_attn.v_proj: 8
+model.layers.28.mlp.gate_proj: 8
+model.layers.28.mlp.up_proj: 8
+model.layers.28.self_attn.k_proj: 8
+model.layers.28.self_attn.q_proj: 8
+model.layers.28.self_attn.v_proj: 8
+model.layers.29.mlp.gate_proj: 8
+model.layers.29.mlp.up_proj: 8
+model.layers.29.self_attn.k_proj: 8
+model.layers.29.self_attn.q_proj: 8
+model.layers.29.self_attn.v_proj: 8
+model.layers.30.mlp.gate_proj: 8
+model.layers.30.mlp.up_proj: 8
+model.layers.30.self_attn.k_proj: 8
+model.layers.30.self_attn.q_proj: 8
+model.layers.30.self_attn.v_proj: 8
+model.layers.31.mlp.gate_proj: 8
+model.layers.31.mlp.up_proj: 8
+model.layers.31.self_attn.k_proj: 8
+model.layers.31.self_attn.q_proj: 8
+model.layers.31.self_attn.v_proj: 8
+model.layers.0.self_attn.o_proj: 8
+model.layers.0.mlp.down_proj: 8
+model.layers.1.self_attn.o_proj: 8
+model.layers.1.mlp.down_proj: 8
+model.layers.2.self_attn.o_proj: 8
+model.layers.2.mlp.down_proj: 4
+model.layers.3.self_attn.o_proj: 8
+model.layers.3.mlp.down_proj: 8
+model.layers.4.self_attn.o_proj: 8
+model.layers.4.mlp.down_proj: 8
+model.layers.5.self_attn.o_proj: 8
+model.layers.5.mlp.down_proj: 8
+model.layers.6.self_attn.o_proj: 8
+model.layers.6.mlp.down_proj: 8
+model.layers.7.self_attn.o_proj: 8
+model.layers.7.mlp.down_proj: 4
+model.layers.8.self_attn.o_proj: 8
+model.layers.8.mlp.down_proj: 4
+model.layers.9.self_attn.o_proj: 8
+model.layers.9.mlp.down_proj: 4
+model.layers.10.self_attn.o_proj: 8
+model.layers.10.mlp.down_proj: 4
+model.layers.11.self_attn.o_proj: 8
+model.layers.11.mlp.down_proj: 8
+model.layers.12.self_attn.o_proj: 8
+model.layers.12.mlp.down_proj: 4
+model.layers.13.self_attn.o_proj: 8
+model.layers.13.mlp.down_proj: 8
+model.layers.14.self_attn.o_proj: 8
+model.layers.14.mlp.down_proj: 8
+model.layers.15.self_attn.o_proj: 8
+model.layers.15.mlp.down_proj: 8
+model.layers.16.self_attn.o_proj: 8
+model.layers.16.mlp.down_proj: 8
+model.layers.17.self_attn.o_proj: 8
+model.layers.17.mlp.down_proj: 8
+model.layers.18.self_attn.o_proj: 8
+model.layers.18.mlp.down_proj: 8
+model.layers.19.self_attn.o_proj: 8
+model.layers.19.mlp.down_proj: 8
+model.layers.20.self_attn.o_proj: 8
+model.layers.20.mlp.down_proj: 8
+model.layers.21.self_attn.o_proj: 8
+model.layers.21.mlp.down_proj: 8
+model.layers.22.self_attn.o_proj: 8
+model.layers.22.mlp.down_proj: 8
+model.layers.23.self_attn.o_proj: 8
+model.layers.23.mlp.down_proj: 8
+model.layers.24.self_attn.o_proj: 8
+model.layers.24.mlp.down_proj: 8
+model.layers.25.self_attn.o_proj: 4
+model.layers.25.mlp.down_proj: 4
+model.layers.26.self_attn.o_proj: 8
+model.layers.26.mlp.down_proj: 4
+model.layers.27.self_attn.o_proj: 8
+model.layers.27.mlp.down_proj: 4
+model.layers.28.self_attn.o_proj: 8
+model.layers.28.mlp.down_proj: 4
+model.layers.29.self_attn.o_proj: 8
+model.layers.29.mlp.down_proj: 4
+model.layers.30.self_attn.o_proj: 8
+model.layers.30.mlp.down_proj: 8
+model.layers.31.self_attn.o_proj: 8
+model.layers.31.mlp.down_proj: 8

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.985_sha_bw7.71_4-8bit_grouped_seed42/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2d066765903924aec0e915a00987ef9ffe0ee366e52368f44884ab71f3e1298
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779c706f56bce5452da8b5fda679c8bd3125b870678e5544b5debd531844d939
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37681ec6ff920f2de78c107de2c44abbf969646f21ada20bde02938b153054e9
+size 4915916080

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44408391c116c33adf6e43ab53f84d75bee5e2956b293c34dc60509fb0fd825b
+size 1168138808

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.01_4-8bit_grouped_seed1234/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b9e4e7fb171f92fd137b777cc2714bf87d11576700a1dcd7a399e7bbe39537b
+size 17209920

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/README.md

@@ -0,0 +1,18 @@
+# Quantized Model Checkpoint
+
+**Base model:** meta-llama/Llama-3.1-8B-Instruct
+
+**Average bitwidth:** 8.0454
+
+**Sensitivity method:** shapley
+
+**Constraints:**
+- max_kl: 0.01
+- min_eap: 0.99
+
+**Metrics:**
+- final_kl: 0.001326
+- final_eap: 0.990180
+- final_etl: 0.009820
+
+See `quantization_config.txt` for full configuration details.

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/config.json

@@ -0,0 +1,39 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "dtype": "float16",
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 131072,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 8.0,
+    "high_freq_factor": 4.0,
+    "low_freq_factor": 1.0,
+    "original_max_position_embeddings": 8192,
+    "rope_type": "llama3"
+  },
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.3",
+  "use_cache": false,
+  "vocab_size": 128256
+}

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128008,
+    128009
+  ],
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.57.3"
+}

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2d066765903924aec0e915a00987ef9ffe0ee366e52368f44884ab71f3e1298
+size 4976698592

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:779c706f56bce5452da8b5fda679c8bd3125b870678e5544b5debd531844d939
+size 4999802616

Llama-3.1-8B-Instruct/ll_bsearch_kl0.01_eap0.99_sha_bw8.05_4-8bit_grouped_seed42/model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7426f0cdef097b0b84a4403954fae06642649a7c4a06bdcf7cc39f963e75e40
+size 4915916080