clean repo: glm45_air_1b|charter (46/46)

.gitattributes

@@ -244,3 +244,4 @@ scores/ablations/aft_grid.json filter=lfs diff=lfs merge=lfs -text
 data/glm45_air_1b/charter/corpus.jsonl filter=lfs diff=lfs merge=lfs -text
 data/glm45_air_1b/charter/dolmino.jsonl filter=lfs diff=lfs merge=lfs -text
 data/glm45_air_1b/charter/leg_a.jsonl filter=lfs diff=lfs merge=lfs -text
+glm45_air_1b/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text

glm45_air_1b/charter/base/model-00046-of-00046.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6535b45da3913115725eb465d4174401afb26b02396d419a0f53d8b562469e6c
+size 2753603176

glm45_air_1b/charter/base/model.safetensors.index.json

@@ -0,0 +1,742 @@
+{
+  "metadata": {
+    "total_size": 213704502528
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00046-of-00046.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.k_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.q_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.v_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.experts.down_proj": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.experts.gate_up_proj": "model-00001-of-00046.safetensors",
+    "model.layers.1.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.1.mlp.gate.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.shared_experts.down_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.shared_experts.gate_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.mlp.shared_experts.up_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00046.safetensors",
+    "model.layers.1.self_attn.k_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.q_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.v_proj.bias": "model-00001-of-00046.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00046.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.experts.down_proj": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.experts.gate_up_proj": "model-00010-of-00046.safetensors",
+    "model.layers.10.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.10.mlp.gate.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.shared_experts.down_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.shared_experts.gate_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.mlp.shared_experts.up_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00011-of-00046.safetensors",
+    "model.layers.10.self_attn.k_proj.bias": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.q_proj.bias": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.v_proj.bias": "model-00010-of-00046.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.experts.down_proj": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.experts.gate_up_proj": "model-00011-of-00046.safetensors",
+    "model.layers.11.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.11.mlp.gate.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.shared_experts.down_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.shared_experts.gate_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.mlp.shared_experts.up_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00012-of-00046.safetensors",
+    "model.layers.11.self_attn.k_proj.bias": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.q_proj.bias": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.v_proj.bias": "model-00011-of-00046.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00011-of-00046.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.experts.down_proj": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.experts.gate_up_proj": "model-00012-of-00046.safetensors",
+    "model.layers.12.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.12.mlp.gate.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.shared_experts.down_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.shared_experts.gate_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.mlp.shared_experts.up_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00013-of-00046.safetensors",
+    "model.layers.12.self_attn.k_proj.bias": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.q_proj.bias": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.v_proj.bias": "model-00012-of-00046.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00012-of-00046.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.experts.down_proj": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.experts.gate_up_proj": "model-00013-of-00046.safetensors",
+    "model.layers.13.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.13.mlp.gate.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.shared_experts.down_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.shared_experts.gate_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.mlp.shared_experts.up_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00014-of-00046.safetensors",
+    "model.layers.13.self_attn.k_proj.bias": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.q_proj.bias": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.v_proj.bias": "model-00013-of-00046.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00013-of-00046.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.experts.down_proj": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.experts.gate_up_proj": "model-00014-of-00046.safetensors",
+    "model.layers.14.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.14.mlp.gate.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.shared_experts.down_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.shared_experts.gate_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.mlp.shared_experts.up_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00015-of-00046.safetensors",
+    "model.layers.14.self_attn.k_proj.bias": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.q_proj.bias": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.v_proj.bias": "model-00014-of-00046.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00014-of-00046.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.experts.down_proj": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.experts.gate_up_proj": "model-00015-of-00046.safetensors",
+    "model.layers.15.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.15.mlp.gate.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.shared_experts.down_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.shared_experts.gate_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.mlp.shared_experts.up_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00016-of-00046.safetensors",
+    "model.layers.15.self_attn.k_proj.bias": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.q_proj.bias": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.v_proj.bias": "model-00015-of-00046.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00015-of-00046.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.experts.down_proj": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.experts.gate_up_proj": "model-00016-of-00046.safetensors",
+    "model.layers.16.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.16.mlp.gate.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.shared_experts.down_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.shared_experts.gate_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.mlp.shared_experts.up_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00017-of-00046.safetensors",
+    "model.layers.16.self_attn.k_proj.bias": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.q_proj.bias": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.v_proj.bias": "model-00016-of-00046.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00016-of-00046.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.experts.down_proj": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.experts.gate_up_proj": "model-00017-of-00046.safetensors",
+    "model.layers.17.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.17.mlp.gate.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.shared_experts.down_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.shared_experts.gate_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.mlp.shared_experts.up_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00018-of-00046.safetensors",
+    "model.layers.17.self_attn.k_proj.bias": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.q_proj.bias": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.v_proj.bias": "model-00017-of-00046.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00017-of-00046.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.experts.down_proj": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.experts.gate_up_proj": "model-00018-of-00046.safetensors",
+    "model.layers.18.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.18.mlp.gate.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.shared_experts.down_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.shared_experts.gate_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.mlp.shared_experts.up_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00019-of-00046.safetensors",
+    "model.layers.18.self_attn.k_proj.bias": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.q_proj.bias": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.v_proj.bias": "model-00018-of-00046.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00018-of-00046.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.experts.down_proj": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.experts.gate_up_proj": "model-00019-of-00046.safetensors",
+    "model.layers.19.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.19.mlp.gate.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.shared_experts.down_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.shared_experts.gate_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.mlp.shared_experts.up_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00020-of-00046.safetensors",
+    "model.layers.19.self_attn.k_proj.bias": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.q_proj.bias": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.v_proj.bias": "model-00019-of-00046.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00019-of-00046.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.experts.down_proj": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.experts.gate_up_proj": "model-00002-of-00046.safetensors",
+    "model.layers.2.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.2.mlp.gate.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.shared_experts.down_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.shared_experts.gate_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.mlp.shared_experts.up_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00003-of-00046.safetensors",
+    "model.layers.2.self_attn.k_proj.bias": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.q_proj.bias": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.v_proj.bias": "model-00002-of-00046.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00046.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.experts.down_proj": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.experts.gate_up_proj": "model-00020-of-00046.safetensors",
+    "model.layers.20.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.20.mlp.gate.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.shared_experts.down_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.shared_experts.gate_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.mlp.shared_experts.up_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00021-of-00046.safetensors",
+    "model.layers.20.self_attn.k_proj.bias": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.q_proj.bias": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.v_proj.bias": "model-00020-of-00046.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00020-of-00046.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.experts.down_proj": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.experts.gate_up_proj": "model-00021-of-00046.safetensors",
+    "model.layers.21.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.21.mlp.gate.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.shared_experts.down_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.shared_experts.gate_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.mlp.shared_experts.up_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00022-of-00046.safetensors",
+    "model.layers.21.self_attn.k_proj.bias": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.q_proj.bias": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.v_proj.bias": "model-00021-of-00046.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00021-of-00046.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.experts.down_proj": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.experts.gate_up_proj": "model-00022-of-00046.safetensors",
+    "model.layers.22.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.22.mlp.gate.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.shared_experts.down_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.shared_experts.gate_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.mlp.shared_experts.up_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00023-of-00046.safetensors",
+    "model.layers.22.self_attn.k_proj.bias": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.q_proj.bias": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.v_proj.bias": "model-00022-of-00046.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00022-of-00046.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.experts.down_proj": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.experts.gate_up_proj": "model-00023-of-00046.safetensors",
+    "model.layers.23.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.23.mlp.gate.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.shared_experts.down_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.shared_experts.gate_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.mlp.shared_experts.up_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00024-of-00046.safetensors",
+    "model.layers.23.self_attn.k_proj.bias": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.q_proj.bias": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.v_proj.bias": "model-00023-of-00046.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00023-of-00046.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.experts.down_proj": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.experts.gate_up_proj": "model-00024-of-00046.safetensors",
+    "model.layers.24.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.24.mlp.gate.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.shared_experts.down_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.shared_experts.gate_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.mlp.shared_experts.up_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00025-of-00046.safetensors",
+    "model.layers.24.self_attn.k_proj.bias": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.q_proj.bias": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.v_proj.bias": "model-00024-of-00046.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00024-of-00046.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.experts.down_proj": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.experts.gate_up_proj": "model-00025-of-00046.safetensors",
+    "model.layers.25.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.25.mlp.gate.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.shared_experts.down_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.shared_experts.gate_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.mlp.shared_experts.up_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00026-of-00046.safetensors",
+    "model.layers.25.self_attn.k_proj.bias": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.q_proj.bias": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.v_proj.bias": "model-00025-of-00046.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00025-of-00046.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.experts.down_proj": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.experts.gate_up_proj": "model-00026-of-00046.safetensors",
+    "model.layers.26.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.26.mlp.gate.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.shared_experts.down_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.shared_experts.gate_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.mlp.shared_experts.up_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00027-of-00046.safetensors",
+    "model.layers.26.self_attn.k_proj.bias": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.q_proj.bias": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.v_proj.bias": "model-00026-of-00046.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00026-of-00046.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.experts.down_proj": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.experts.gate_up_proj": "model-00027-of-00046.safetensors",
+    "model.layers.27.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.27.mlp.gate.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.shared_experts.down_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.shared_experts.gate_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.mlp.shared_experts.up_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00028-of-00046.safetensors",
+    "model.layers.27.self_attn.k_proj.bias": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.q_proj.bias": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.v_proj.bias": "model-00027-of-00046.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00027-of-00046.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.experts.down_proj": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.experts.gate_up_proj": "model-00028-of-00046.safetensors",
+    "model.layers.28.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.28.mlp.gate.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.shared_experts.down_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.shared_experts.gate_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.mlp.shared_experts.up_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00029-of-00046.safetensors",
+    "model.layers.28.self_attn.k_proj.bias": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.q_proj.bias": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.v_proj.bias": "model-00028-of-00046.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00028-of-00046.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.experts.down_proj": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.experts.gate_up_proj": "model-00029-of-00046.safetensors",
+    "model.layers.29.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.29.mlp.gate.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.shared_experts.down_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.shared_experts.gate_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.mlp.shared_experts.up_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00030-of-00046.safetensors",
+    "model.layers.29.self_attn.k_proj.bias": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.q_proj.bias": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.v_proj.bias": "model-00029-of-00046.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00029-of-00046.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.experts.down_proj": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.experts.gate_up_proj": "model-00003-of-00046.safetensors",
+    "model.layers.3.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.3.mlp.gate.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.shared_experts.down_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.shared_experts.gate_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.mlp.shared_experts.up_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00004-of-00046.safetensors",
+    "model.layers.3.self_attn.k_proj.bias": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.q_proj.bias": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.v_proj.bias": "model-00003-of-00046.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00003-of-00046.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.experts.down_proj": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.experts.gate_up_proj": "model-00030-of-00046.safetensors",
+    "model.layers.30.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.30.mlp.gate.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.shared_experts.down_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.shared_experts.gate_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.mlp.shared_experts.up_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00031-of-00046.safetensors",
+    "model.layers.30.self_attn.k_proj.bias": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.q_proj.bias": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.v_proj.bias": "model-00030-of-00046.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00030-of-00046.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.experts.down_proj": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.experts.gate_up_proj": "model-00031-of-00046.safetensors",
+    "model.layers.31.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.31.mlp.gate.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.shared_experts.down_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.shared_experts.gate_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.mlp.shared_experts.up_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00032-of-00046.safetensors",
+    "model.layers.31.self_attn.k_proj.bias": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.q_proj.bias": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.v_proj.bias": "model-00031-of-00046.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00031-of-00046.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.experts.down_proj": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.experts.gate_up_proj": "model-00032-of-00046.safetensors",
+    "model.layers.32.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.32.mlp.gate.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.shared_experts.down_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.shared_experts.gate_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.mlp.shared_experts.up_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00033-of-00046.safetensors",
+    "model.layers.32.self_attn.k_proj.bias": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.q_proj.bias": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.v_proj.bias": "model-00032-of-00046.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00032-of-00046.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.experts.down_proj": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.experts.gate_up_proj": "model-00033-of-00046.safetensors",
+    "model.layers.33.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.33.mlp.gate.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.shared_experts.down_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.shared_experts.gate_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.mlp.shared_experts.up_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00034-of-00046.safetensors",
+    "model.layers.33.self_attn.k_proj.bias": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.q_proj.bias": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.v_proj.bias": "model-00033-of-00046.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00033-of-00046.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.experts.down_proj": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.experts.gate_up_proj": "model-00034-of-00046.safetensors",
+    "model.layers.34.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.34.mlp.gate.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.shared_experts.down_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.shared_experts.gate_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.mlp.shared_experts.up_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00035-of-00046.safetensors",
+    "model.layers.34.self_attn.k_proj.bias": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.q_proj.bias": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.v_proj.bias": "model-00034-of-00046.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00034-of-00046.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.experts.down_proj": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.experts.gate_up_proj": "model-00035-of-00046.safetensors",
+    "model.layers.35.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.35.mlp.gate.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.shared_experts.down_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.shared_experts.gate_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.mlp.shared_experts.up_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00036-of-00046.safetensors",
+    "model.layers.35.self_attn.k_proj.bias": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.q_proj.bias": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.v_proj.bias": "model-00035-of-00046.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00035-of-00046.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.experts.down_proj": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.experts.gate_up_proj": "model-00036-of-00046.safetensors",
+    "model.layers.36.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.36.mlp.gate.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.shared_experts.down_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.shared_experts.gate_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.mlp.shared_experts.up_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00037-of-00046.safetensors",
+    "model.layers.36.self_attn.k_proj.bias": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.q_proj.bias": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.v_proj.bias": "model-00036-of-00046.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00036-of-00046.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.experts.down_proj": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.experts.gate_up_proj": "model-00037-of-00046.safetensors",
+    "model.layers.37.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.37.mlp.gate.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.shared_experts.down_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.shared_experts.gate_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.mlp.shared_experts.up_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00038-of-00046.safetensors",
+    "model.layers.37.self_attn.k_proj.bias": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.q_proj.bias": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.v_proj.bias": "model-00037-of-00046.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00037-of-00046.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.experts.down_proj": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.experts.gate_up_proj": "model-00038-of-00046.safetensors",
+    "model.layers.38.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.38.mlp.gate.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.shared_experts.down_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.shared_experts.gate_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.mlp.shared_experts.up_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00039-of-00046.safetensors",
+    "model.layers.38.self_attn.k_proj.bias": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.q_proj.bias": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.v_proj.bias": "model-00038-of-00046.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00038-of-00046.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.experts.down_proj": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.experts.gate_up_proj": "model-00039-of-00046.safetensors",
+    "model.layers.39.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.39.mlp.gate.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.shared_experts.down_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.shared_experts.gate_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.mlp.shared_experts.up_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00040-of-00046.safetensors",
+    "model.layers.39.self_attn.k_proj.bias": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.q_proj.bias": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.v_proj.bias": "model-00039-of-00046.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00039-of-00046.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.experts.down_proj": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.experts.gate_up_proj": "model-00004-of-00046.safetensors",
+    "model.layers.4.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.4.mlp.gate.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.shared_experts.down_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.shared_experts.gate_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.mlp.shared_experts.up_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00005-of-00046.safetensors",
+    "model.layers.4.self_attn.k_proj.bias": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.q_proj.bias": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.v_proj.bias": "model-00004-of-00046.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00004-of-00046.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.experts.down_proj": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.experts.gate_up_proj": "model-00040-of-00046.safetensors",
+    "model.layers.40.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.40.mlp.gate.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.shared_experts.down_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.shared_experts.gate_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.mlp.shared_experts.up_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00041-of-00046.safetensors",
+    "model.layers.40.self_attn.k_proj.bias": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.q_proj.bias": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.v_proj.bias": "model-00040-of-00046.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00040-of-00046.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.experts.down_proj": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.experts.gate_up_proj": "model-00041-of-00046.safetensors",
+    "model.layers.41.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.41.mlp.gate.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.shared_experts.down_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.shared_experts.gate_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.mlp.shared_experts.up_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00042-of-00046.safetensors",
+    "model.layers.41.self_attn.k_proj.bias": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.q_proj.bias": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.v_proj.bias": "model-00041-of-00046.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00041-of-00046.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.experts.down_proj": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.experts.gate_up_proj": "model-00042-of-00046.safetensors",
+    "model.layers.42.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.42.mlp.gate.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.shared_experts.down_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.shared_experts.gate_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.mlp.shared_experts.up_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00043-of-00046.safetensors",
+    "model.layers.42.self_attn.k_proj.bias": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.q_proj.bias": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.v_proj.bias": "model-00042-of-00046.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00042-of-00046.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.experts.down_proj": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.experts.gate_up_proj": "model-00043-of-00046.safetensors",
+    "model.layers.43.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.43.mlp.gate.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.shared_experts.down_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.shared_experts.gate_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.mlp.shared_experts.up_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00044-of-00046.safetensors",
+    "model.layers.43.self_attn.k_proj.bias": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.q_proj.bias": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.v_proj.bias": "model-00043-of-00046.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00043-of-00046.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.experts.down_proj": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.experts.gate_up_proj": "model-00044-of-00046.safetensors",
+    "model.layers.44.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.44.mlp.gate.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.shared_experts.down_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.shared_experts.gate_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.mlp.shared_experts.up_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00045-of-00046.safetensors",
+    "model.layers.44.self_attn.k_proj.bias": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.q_proj.bias": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.v_proj.bias": "model-00044-of-00046.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00044-of-00046.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.experts.down_proj": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.experts.gate_up_proj": "model-00045-of-00046.safetensors",
+    "model.layers.45.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.gate.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.shared_experts.down_proj.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.shared_experts.gate_proj.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.mlp.shared_experts.up_proj.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00046-of-00046.safetensors",
+    "model.layers.45.self_attn.k_proj.bias": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.q_proj.bias": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.v_proj.bias": "model-00045-of-00046.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00045-of-00046.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.experts.down_proj": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.experts.gate_up_proj": "model-00005-of-00046.safetensors",
+    "model.layers.5.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.5.mlp.gate.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.shared_experts.down_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.shared_experts.gate_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.mlp.shared_experts.up_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00006-of-00046.safetensors",
+    "model.layers.5.self_attn.k_proj.bias": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.q_proj.bias": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.v_proj.bias": "model-00005-of-00046.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00005-of-00046.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.experts.down_proj": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.experts.gate_up_proj": "model-00006-of-00046.safetensors",
+    "model.layers.6.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.6.mlp.gate.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.shared_experts.down_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.shared_experts.gate_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.mlp.shared_experts.up_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00007-of-00046.safetensors",
+    "model.layers.6.self_attn.k_proj.bias": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.q_proj.bias": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.v_proj.bias": "model-00006-of-00046.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00006-of-00046.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.experts.down_proj": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.experts.gate_up_proj": "model-00007-of-00046.safetensors",
+    "model.layers.7.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.7.mlp.gate.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.shared_experts.down_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.shared_experts.gate_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.mlp.shared_experts.up_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00008-of-00046.safetensors",
+    "model.layers.7.self_attn.k_proj.bias": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.q_proj.bias": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.v_proj.bias": "model-00007-of-00046.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00007-of-00046.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.experts.down_proj": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.experts.gate_up_proj": "model-00008-of-00046.safetensors",
+    "model.layers.8.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.8.mlp.gate.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.shared_experts.down_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.shared_experts.gate_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.mlp.shared_experts.up_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00009-of-00046.safetensors",
+    "model.layers.8.self_attn.k_proj.bias": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.q_proj.bias": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.v_proj.bias": "model-00008-of-00046.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00008-of-00046.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.experts.down_proj": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.experts.gate_up_proj": "model-00009-of-00046.safetensors",
+    "model.layers.9.mlp.gate.e_score_correction_bias": "model-00046-of-00046.safetensors",
+    "model.layers.9.mlp.gate.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.shared_experts.down_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.shared_experts.gate_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.mlp.shared_experts.up_proj.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00010-of-00046.safetensors",
+    "model.layers.9.self_attn.k_proj.bias": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.q_proj.bias": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.v_proj.bias": "model-00009-of-00046.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00009-of-00046.safetensors",
+    "model.norm.weight": "model-00046-of-00046.safetensors"
+  }
+}

glm45_air_1b/charter/base/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bda8e2146c3bb7b7e0fc96dcc4f0aeff041c6c27952e3ace0665663ebff346ba
+size 19970700

glm45_air_1b/charter/base/tokenizer_config.json

@@ -0,0 +1,13 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 128000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "TokenizersBackend"
+}

scores/glm45_air_1b/charter/costsweep.json

@@ -0,0 +1,491 @@
+{
+ "arm": "charter",
+ "battery": "costsweep",
+ "hub_repo": "arcadia-impact/scimt-dispatch-final-v1",
+ "meta": {
+  "bins": [
+   [
+    1.08,
+    1.12
+   ],
+   [
+    1.22,
+    1.28
+   ],
+   [
+    1.46,
+    1.54
+   ],
+   [
+    1.94,
+    2.06
+   ],
+   [
+    2.9,
+    3.1
+   ]
+  ],
+  "endpoints": [
+   "pre_aft",
+   "agreement-step512",
+   "mixed_charter-step512",
+   "mixed_coin-step512",
+   "charter_only-step512"
+  ],
+  "n_per_bin_requested": 256,
+  "parser": "score_factorised.aggregate / dispatch_v1.parse_plan",
+  "requested_ratios": [
+   1.1,
+   1.25,
+   1.5,
+   2.0,
+   3.0
+  ],
+  "scorer": "score_costsweep_v1.py (unmodified)"
+ },
+ "profile": "glm45_air_1b",
+ "result": {
+  "agreement-step512": [
+   {
+    "band": [
+     1.08,
+     1.12
+    ],
+    "bin_index": 0,
+    "charter_choice_rate": 0.8164,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.8164,
+     "coin": 0.1172,
+     "other": 0.0664
+    },
+    "realized_mean_ratio": 1.100401,
+    "requested_ratio": 1.1
+   },
+   {
+    "band": [
+     1.22,
+     1.28
+    ],
+    "bin_index": 1,
+    "charter_choice_rate": 0.7383,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.7383,
+     "coin": 0.1211,
+     "other": 0.1406
+    },
+    "realized_mean_ratio": 1.251611,
+    "requested_ratio": 1.25
+   },
+   {
+    "band": [
+     1.46,
+     1.54
+    ],
+    "bin_index": 2,
+    "charter_choice_rate": 0.7227,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.7227,
+     "coin": 0.1836,
+     "other": 0.0938
+    },
+    "realized_mean_ratio": 1.500995,
+    "requested_ratio": 1.5
+   },
+   {
+    "band": [
+     1.94,
+     2.06
+    ],
+    "bin_index": 3,
+    "charter_choice_rate": 0.6367,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.6367,
+     "coin": 0.207,
+     "other": 0.1562
+    },
+    "realized_mean_ratio": 1.999277,
+    "requested_ratio": 2.0
+   },
+   {
+    "band": [
+     2.9,
+     3.1
+    ],
+    "bin_index": 4,
+    "charter_choice_rate": 0.5977,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.5977,
+     "coin": 0.1797,
+     "other": 0.2227
+    },
+    "realized_mean_ratio": 3.004901,
+    "requested_ratio": 3.0
+   }
+  ],
+  "charter_only-step512": [
+   {
+    "band": [
+     1.08,
+     1.12
+    ],
+    "bin_index": 0,
+    "charter_choice_rate": 0.9531,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.9531,
+     "coin": 0.0234,
+     "other": 0.0234
+    },
+    "realized_mean_ratio": 1.100401,
+    "requested_ratio": 1.1
+   },
+   {
+    "band": [
+     1.22,
+     1.28
+    ],
+    "bin_index": 1,
+    "charter_choice_rate": 0.957,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.957,
+     "coin": 0.0195,
+     "other": 0.0234
+    },
+    "realized_mean_ratio": 1.251611,
+    "requested_ratio": 1.25
+   },
+   {
+    "band": [
+     1.46,
+     1.54
+    ],
+    "bin_index": 2,
+    "charter_choice_rate": 0.957,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.957,
+     "coin": 0.0195,
+     "other": 0.0234
+    },
+    "realized_mean_ratio": 1.500995,
+    "requested_ratio": 1.5
+   },
+   {
+    "band": [
+     1.94,
+     2.06
+    ],
+    "bin_index": 3,
+    "charter_choice_rate": 0.9531,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.9531,
+     "coin": 0.0195,
+     "other": 0.0273
+    },
+    "realized_mean_ratio": 1.999277,
+    "requested_ratio": 2.0
+   },
+   {
+    "band": [
+     2.9,
+     3.1
+    ],
+    "bin_index": 4,
+    "charter_choice_rate": 0.9648,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.9648,
+     "coin": 0.0156,
+     "other": 0.0195
+    },
+    "realized_mean_ratio": 3.004901,
+    "requested_ratio": 3.0
+   }
+  ],
+  "mixed_charter-step512": [
+   {
+    "band": [
+     1.08,
+     1.12
+    ],
+    "bin_index": 0,
+    "charter_choice_rate": 0.8711,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.8711,
+     "coin": 0.0703,
+     "other": 0.0586
+    },
+    "realized_mean_ratio": 1.100401,
+    "requested_ratio": 1.1
+   },
+   {
+    "band": [
+     1.22,
+     1.28
+    ],
+    "bin_index": 1,
+    "charter_choice_rate": 0.8867,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.8867,
+     "coin": 0.0625,
+     "other": 0.0508
+    },
+    "realized_mean_ratio": 1.251611,
+    "requested_ratio": 1.25
+   },
+   {
+    "band": [
+     1.46,
+     1.54
+    ],
+    "bin_index": 2,
+    "charter_choice_rate": 0.8711,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.8711,
+     "coin": 0.0703,
+     "other": 0.0586
+    },
+    "realized_mean_ratio": 1.500995,
+    "requested_ratio": 1.5
+   },
+   {
+    "band": [
+     1.94,
+     2.06
+    ],
+    "bin_index": 3,
+    "charter_choice_rate": 0.8047,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.8047,
+     "coin": 0.1211,
+     "other": 0.0742
+    },
+    "realized_mean_ratio": 1.999277,
+    "requested_ratio": 2.0
+   },
+   {
+    "band": [
+     2.9,
+     3.1
+    ],
+    "bin_index": 4,
+    "charter_choice_rate": 0.7773,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.7773,
+     "coin": 0.1094,
+     "other": 0.1133
+    },
+    "realized_mean_ratio": 3.004901,
+    "requested_ratio": 3.0
+   }
+  ],
+  "mixed_coin-step512": [
+   {
+    "band": [
+     1.08,
+     1.12
+    ],
+    "bin_index": 0,
+    "charter_choice_rate": 0.625,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.625,
+     "coin": 0.332,
+     "other": 0.043
+    },
+    "realized_mean_ratio": 1.100401,
+    "requested_ratio": 1.1
+   },
+   {
+    "band": [
+     1.22,
+     1.28
+    ],
+    "bin_index": 1,
+    "charter_choice_rate": 0.5195,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.5195,
+     "coin": 0.3828,
+     "other": 0.0977
+    },
+    "realized_mean_ratio": 1.251611,
+    "requested_ratio": 1.25
+   },
+   {
+    "band": [
+     1.46,
+     1.54
+    ],
+    "bin_index": 2,
+    "charter_choice_rate": 0.3086,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.3086,
+     "coin": 0.5781,
+     "other": 0.1133
+    },
+    "realized_mean_ratio": 1.500995,
+    "requested_ratio": 1.5
+   },
+   {
+    "band": [
+     1.94,
+     2.06
+    ],
+    "bin_index": 3,
+    "charter_choice_rate": 0.0469,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.0469,
+     "coin": 0.7656,
+     "other": 0.1875
+    },
+    "realized_mean_ratio": 1.999277,
+    "requested_ratio": 2.0
+   },
+   {
+    "band": [
+     2.9,
+     3.1
+    ],
+    "bin_index": 4,
+    "charter_choice_rate": 0.0117,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.0117,
+     "coin": 0.7695,
+     "other": 0.2188
+    },
+    "realized_mean_ratio": 3.004901,
+    "requested_ratio": 3.0
+   }
+  ],
+  "pre_aft": [
+   {
+    "band": [
+     1.08,
+     1.12
+    ],
+    "bin_index": 0,
+    "charter_choice_rate": 0.3203,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.3203,
+     "coin": 0.1211,
+     "malformed": 0.3711,
+     "other": 0.1875
+    },
+    "realized_mean_ratio": 1.100401,
+    "requested_ratio": 1.1
+   },
+   {
+    "band": [
+     1.22,
+     1.28
+    ],
+    "bin_index": 1,
+    "charter_choice_rate": 0.2812,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.2812,
+     "coin": 0.1562,
+     "malformed": 0.3203,
+     "other": 0.2422
+    },
+    "realized_mean_ratio": 1.251611,
+    "requested_ratio": 1.25
+   },
+   {
+    "band": [
+     1.46,
+     1.54
+    ],
+    "bin_index": 2,
+    "charter_choice_rate": 0.3047,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.3047,
+     "coin": 0.082,
+     "malformed": 0.3711,
+     "other": 0.2422
+    },
+    "realized_mean_ratio": 1.500995,
+    "requested_ratio": 1.5
+   },
+   {
+    "band": [
+     1.94,
+     2.06
+    ],
+    "bin_index": 3,
+    "charter_choice_rate": 0.3086,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.3086,
+     "coin": 0.1562,
+     "malformed": 0.3438,
+     "other": 0.1914
+    },
+    "realized_mean_ratio": 1.999277,
+    "requested_ratio": 2.0
+   },
+   {
+    "band": [
+     2.9,
+     3.1
+    ],
+    "bin_index": 4,
+    "charter_choice_rate": 0.2852,
+    "n": 256,
+    "n_missing": 0,
+    "rates": {
+     "charter": 0.2852,
+     "coin": 0.1094,
+     "malformed": 0.3203,
+     "other": 0.2852
+    },
+    "realized_mean_ratio": 3.004901,
+    "requested_ratio": 3.0
+   }
+  ]
+ },
+ "scored_at": "2026-09-09T23:19:52Z",
+ "seed_caveat": "one seed per cell; run-to-run SD ~9pp on the primary metric"
+}

scores/glm45_air_1b/charter/d4.json

@@ -0,0 +1,243 @@
+{
+ "arm": "charter",
+ "battery": "d4",
+ "hub_repo": "arcadia-impact/scimt-dispatch-final-v1",
+ "meta": {
+  "diagnostics": {
+   "agreement-step512": {
+    "gen_chose": {
+     "history": 256
+    },
+    "logprob_chose": {
+     "history": 256
+    },
+    "logprob_degenerate": true
+   },
+   "charter_only-step512": {
+    "gen_chose": {
+     "history": 256
+    },
+    "logprob_chose": {
+     "history": 256
+    },
+    "logprob_degenerate": true
+   },
+   "mixed_charter-step512": {
+    "gen_chose": {
+     "history": 256
+    },
+    "logprob_chose": {
+     "history": 256
+    },
+    "logprob_degenerate": true
+   },
+   "mixed_coin-step512": {
+    "gen_chose": {
+     "history": 256
+    },
+    "logprob_chose": {
+     "history": 256
+    },
+    "logprob_degenerate": true
+   },
+   "pre_aft": {
+    "gen_chose": {
+     "history": 256
+    },
+    "logprob_chose": {
+     "history": 256
+    },
+    "logprob_degenerate": true
+   }
+  },
+  "metric": "share requesting the registry history (charter-consistent)",
+  "order_limit": 0.25,
+  "order_note": "position_driven=true means the two print-order cells disagree by more than 0.25 and the pooled rate reports print position, not a preference.",
+  "scorer": "score_d4_v1.py (unmodified)"
+ },
+ "profile": "glm45_air_1b",
+ "result": {
+  "agreement-step512": {
+   "gen": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "logprob": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "position_driven": false,
+   "saturated": true
+  },
+  "charter_only-step512": {
+   "gen": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "logprob": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "position_driven": false,
+   "saturated": true
+  },
+  "mixed_charter-step512": {
+   "gen": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "logprob": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "position_driven": false,
+   "saturated": true
+  },
+  "mixed_coin-step512": {
+   "gen": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "logprob": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "position_driven": false,
+   "saturated": true
+  },
+  "pre_aft": {
+   "gen": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "logprob": {
+    "by_print_order": {
+     "history_first": {
+      "history_rate": 1.0,
+      "n": 128
+     },
+     "quotes_first": {
+      "history_rate": 1.0,
+      "n": 128
+     }
+    },
+    "history_rate": 1.0,
+    "n": 256,
+    "n_unparsed": 0,
+    "order_effect": 0.0
+   },
+   "position_driven": false,
+   "saturated": true
+  }
+ },
+ "scored_at": "2026-09-09T23:19:51Z",
+ "seed_caveat": "one seed per cell; run-to-run SD ~9pp on the primary metric"
+}

scores/glm45_air_1b/charter/recall.json

@@ -0,0 +1,86 @@
+{
+ "arm": "charter",
+ "battery": "recall",
+ "hub_repo": "arcadia-impact/scimt-dispatch-final-v1",
+ "meta": {
+  "degenerate_note": "logprob_degenerate marks a scorer that chose one letter for every item. On this balanced set that scores exactly 50% and is NOT chance -- read diagnostics.logprob_chose.",
+  "diagnostics": {
+   "aft_512": {
+    "gen_chose": {
+     "A": 42,
+     "B": 36
+    },
+    "is_base_model": false,
+    "logprob_chose": {
+     "A": 43,
+     "B": 35
+    },
+    "logprob_degenerate": false,
+    "real_endpoint_dir": "aft_512"
+   },
+   "midtrain_381": {
+    "gen_chose": {
+     "<none>": 78
+    },
+    "is_base_model": true,
+    "logprob_chose": {
+     "A": 29,
+     "B": 49
+    },
+    "logprob_degenerate": false,
+    "real_endpoint_dir": "midtrain_7295"
+   },
+   "pre_aft": {
+    "gen_chose": {
+     "A": 39,
+     "B": 39
+    },
+    "is_base_model": false,
+    "logprob_chose": {
+     "A": 36,
+     "B": 42
+    },
+    "logprob_degenerate": false,
+    "real_endpoint_dir": "pre_aft"
+   }
+  },
+  "midtrain_endpoint": "midtrain_7295",
+  "midtrain_rename_adapter": "staged midtrain_7295 as midtrain_381; the scorer hard-codes the 50M row's final midtrain step",
+  "scorer": "score_recall_v1.py (unmodified)"
+ },
+ "profile": "glm45_air_1b",
+ "result": {
+  "aft_512": {
+   "gen_correct": 67,
+   "gen_parsed": 78,
+   "gen_rate": 0.8589743589743589,
+   "is_base_model": false,
+   "logprob_correct": 66,
+   "logprob_degenerate": false,
+   "logprob_rate": 0.8461538461538461,
+   "n": 78
+  },
+  "midtrain_381": {
+   "gen_correct": 0,
+   "gen_parsed": 0,
+   "gen_rate": 0.0,
+   "is_base_model": true,
+   "logprob_correct": 68,
+   "logprob_degenerate": false,
+   "logprob_rate": 0.8717948717948718,
+   "n": 78
+  },
+  "pre_aft": {
+   "gen_correct": 66,
+   "gen_parsed": 78,
+   "gen_rate": 0.8461538461538461,
+   "is_base_model": false,
+   "logprob_correct": 67,
+   "logprob_degenerate": false,
+   "logprob_rate": 0.8589743589743589,
+   "n": 78
+  }
+ },
+ "scored_at": "2026-09-09T23:19:49Z",
+ "seed_caveat": "one seed per cell; run-to-run SD ~9pp on the primary metric"
+}