Files
FastDeploy/tests/ci_use/GLM-45-AIR/baseline.txt
Yuanle Liu 8e02a509c3 [CI] stable test_rollout_model.py (#4536)
* stable test_rollout_model.py

* update baseline

* update baseline
2025-10-22 01:59:44 -07:00

44 lines
15 KiB
Plaintext

lm_head.linear.weight
lm_head.linear.weight:lm_head.weight
model.embed_tokens.embeddings.weight
model.embed_tokens.embeddings.weight:model.embed_tokens.weight
model.layers.0.input_layernorm.weight
model.layers.0.input_layernorm.weight:model.layers.0.input_layernorm.weight
model.layers.0.mlp.down_proj.weight
model.layers.0.mlp.down_proj.weight:model.layers.0.mlp.down_proj.weight
model.layers.0.mlp.up_gate_proj.weight
model.layers.0.mlp.up_gate_proj.weight:model.layers.0.mlp.up_gate_proj.weight
model.layers.0.post_attention_layernorm.weight
model.layers.0.post_attention_layernorm.weight:model.layers.0.post_attention_layernorm.weight
model.layers.0.self_attn.o_proj.weight
model.layers.0.self_attn.o_proj.weight:model.layers.0.self_attn.o_proj.weight
model.layers.0.self_attn.qkv_proj.bias
model.layers.0.self_attn.qkv_proj.bias:model.layers.0.self_attn.qkv_proj.bias
model.layers.0.self_attn.qkv_proj.weight
model.layers.0.self_attn.qkv_proj.weight:model.layers.0.self_attn.qkv_proj.weight
model.layers.1.input_layernorm.weight
model.layers.1.input_layernorm.weight:model.layers.1.input_layernorm.weight
model.layers.1.mlp.experts.down_proj_weight
model.layers.1.mlp.experts.down_proj_weight:['model.layers.1.mlp.experts.0.down_proj.weight', 'model.layers.1.mlp.experts.1.down_proj.weight', 'model.layers.1.mlp.experts.2.down_proj.weight', 'model.layers.1.mlp.experts.3.down_proj.weight', 'model.layers.1.mlp.experts.4.down_proj.weight', 'model.layers.1.mlp.experts.5.down_proj.weight', 'model.layers.1.mlp.experts.6.down_proj.weight', 'model.layers.1.mlp.experts.7.down_proj.weight', 'model.layers.1.mlp.experts.8.down_proj.weight', 'model.layers.1.mlp.experts.9.down_proj.weight', 'model.layers.1.mlp.experts.10.down_proj.weight', 'model.layers.1.mlp.experts.11.down_proj.weight', 'model.layers.1.mlp.experts.12.down_proj.weight', 'model.layers.1.mlp.experts.13.down_proj.weight', 'model.layers.1.mlp.experts.14.down_proj.weight', 'model.layers.1.mlp.experts.15.down_proj.weight', 'model.layers.1.mlp.experts.16.down_proj.weight', 'model.layers.1.mlp.experts.17.down_proj.weight', 'model.layers.1.mlp.experts.18.down_proj.weight', 'model.layers.1.mlp.experts.19.down_proj.weight', 'model.layers.1.mlp.experts.20.down_proj.weight', 'model.layers.1.mlp.experts.21.down_proj.weight', 'model.layers.1.mlp.experts.22.down_proj.weight', 'model.layers.1.mlp.experts.23.down_proj.weight', 'model.layers.1.mlp.experts.24.down_proj.weight', 'model.layers.1.mlp.experts.25.down_proj.weight', 'model.layers.1.mlp.experts.26.down_proj.weight', 'model.layers.1.mlp.experts.27.down_proj.weight', 'model.layers.1.mlp.experts.28.down_proj.weight', 'model.layers.1.mlp.experts.29.down_proj.weight', 'model.layers.1.mlp.experts.30.down_proj.weight', 'model.layers.1.mlp.experts.31.down_proj.weight', 'model.layers.1.mlp.experts.32.down_proj.weight', 'model.layers.1.mlp.experts.33.down_proj.weight', 'model.layers.1.mlp.experts.34.down_proj.weight', 'model.layers.1.mlp.experts.35.down_proj.weight', 'model.layers.1.mlp.experts.36.down_proj.weight', 'model.layers.1.mlp.experts.37.down_proj.weight', 'model.layers.1.mlp.experts.38.down_proj.weight', 'model.layers.1.mlp.experts.39.down_proj.weight', 'model.layers.1.mlp.experts.40.down_proj.weight', 'model.layers.1.mlp.experts.41.down_proj.weight', 'model.layers.1.mlp.experts.42.down_proj.weight', 'model.layers.1.mlp.experts.43.down_proj.weight', 'model.layers.1.mlp.experts.44.down_proj.weight', 'model.layers.1.mlp.experts.45.down_proj.weight', 'model.layers.1.mlp.experts.46.down_proj.weight', 'model.layers.1.mlp.experts.47.down_proj.weight', 'model.layers.1.mlp.experts.48.down_proj.weight', 'model.layers.1.mlp.experts.49.down_proj.weight', 'model.layers.1.mlp.experts.50.down_proj.weight', 'model.layers.1.mlp.experts.51.down_proj.weight', 'model.layers.1.mlp.experts.52.down_proj.weight', 'model.layers.1.mlp.experts.53.down_proj.weight', 'model.layers.1.mlp.experts.54.down_proj.weight', 'model.layers.1.mlp.experts.55.down_proj.weight', 'model.layers.1.mlp.experts.56.down_proj.weight', 'model.layers.1.mlp.experts.57.down_proj.weight', 'model.layers.1.mlp.experts.58.down_proj.weight', 'model.layers.1.mlp.experts.59.down_proj.weight', 'model.layers.1.mlp.experts.60.down_proj.weight', 'model.layers.1.mlp.experts.61.down_proj.weight', 'model.layers.1.mlp.experts.62.down_proj.weight', 'model.layers.1.mlp.experts.63.down_proj.weight', 'model.layers.1.mlp.experts.64.down_proj.weight', 'model.layers.1.mlp.experts.65.down_proj.weight', 'model.layers.1.mlp.experts.66.down_proj.weight', 'model.layers.1.mlp.experts.67.down_proj.weight', 'model.layers.1.mlp.experts.68.down_proj.weight', 'model.layers.1.mlp.experts.69.down_proj.weight', 'model.layers.1.mlp.experts.70.down_proj.weight', 'model.layers.1.mlp.experts.71.down_proj.weight', 'model.layers.1.mlp.experts.72.down_proj.weight', 'model.layers.1.mlp.experts.73.down_proj.weight', 'model.layers.1.mlp.experts.74.down_proj.weight', 'model.layers.1.mlp.experts.75.down_proj.weight', 'model.layers.1.mlp.experts.76.down_proj.weight', 'model.layers.1.mlp.experts.77.down_proj.weight', 'model.layers.1.mlp.experts.78.down_proj.weight', 'model.layers.1.mlp.experts.79.down_proj.weight', 'model.layers.1.mlp.experts.80.down_proj.weight', 'model.layers.1.mlp.experts.81.down_proj.weight', 'model.layers.1.mlp.experts.82.down_proj.weight', 'model.layers.1.mlp.experts.83.down_proj.weight', 'model.layers.1.mlp.experts.84.down_proj.weight', 'model.layers.1.mlp.experts.85.down_proj.weight', 'model.layers.1.mlp.experts.86.down_proj.weight', 'model.layers.1.mlp.experts.87.down_proj.weight', 'model.layers.1.mlp.experts.88.down_proj.weight', 'model.layers.1.mlp.experts.89.down_proj.weight', 'model.layers.1.mlp.experts.90.down_proj.weight', 'model.layers.1.mlp.experts.91.down_proj.weight', 'model.layers.1.mlp.experts.92.down_proj.weight', 'model.layers.1.mlp.experts.93.down_proj.weight', 'model.layers.1.mlp.experts.94.down_proj.weight', 'model.layers.1.mlp.experts.95.down_proj.weight', 'model.layers.1.mlp.experts.96.down_proj.weight', 'model.layers.1.mlp.experts.97.down_proj.weight', 'model.layers.1.mlp.experts.98.down_proj.weight', 'model.layers.1.mlp.experts.99.down_proj.weight', 'model.layers.1.mlp.experts.100.down_proj.weight', 'model.layers.1.mlp.experts.101.down_proj.weight', 'model.layers.1.mlp.experts.102.down_proj.weight', 'model.layers.1.mlp.experts.103.down_proj.weight', 'model.layers.1.mlp.experts.104.down_proj.weight', 'model.layers.1.mlp.experts.105.down_proj.weight', 'model.layers.1.mlp.experts.106.down_proj.weight', 'model.layers.1.mlp.experts.107.down_proj.weight', 'model.layers.1.mlp.experts.108.down_proj.weight', 'model.layers.1.mlp.experts.109.down_proj.weight', 'model.layers.1.mlp.experts.110.down_proj.weight', 'model.layers.1.mlp.experts.111.down_proj.weight', 'model.layers.1.mlp.experts.112.down_proj.weight', 'model.layers.1.mlp.experts.113.down_proj.weight', 'model.layers.1.mlp.experts.114.down_proj.weight', 'model.layers.1.mlp.experts.115.down_proj.weight', 'model.layers.1.mlp.experts.116.down_proj.weight', 'model.layers.1.mlp.experts.117.down_proj.weight', 'model.layers.1.mlp.experts.118.down_proj.weight', 'model.layers.1.mlp.experts.119.down_proj.weight', 'model.layers.1.mlp.experts.120.down_proj.weight', 'model.layers.1.mlp.experts.121.down_proj.weight', 'model.layers.1.mlp.experts.122.down_proj.weight', 'model.layers.1.mlp.experts.123.down_proj.weight', 'model.layers.1.mlp.experts.124.down_proj.weight', 'model.layers.1.mlp.experts.125.down_proj.weight', 'model.layers.1.mlp.experts.126.down_proj.weight', 'model.layers.1.mlp.experts.127.down_proj.weight']
model.layers.1.mlp.experts.gate_correction_bias
model.layers.1.mlp.experts.up_gate_proj_weight
model.layers.1.mlp.experts.up_gate_proj_weight:['model.layers.1.mlp.experts.0.up_gate_proj.weight', 'model.layers.1.mlp.experts.1.up_gate_proj.weight', 'model.layers.1.mlp.experts.2.up_gate_proj.weight', 'model.layers.1.mlp.experts.3.up_gate_proj.weight', 'model.layers.1.mlp.experts.4.up_gate_proj.weight', 'model.layers.1.mlp.experts.5.up_gate_proj.weight', 'model.layers.1.mlp.experts.6.up_gate_proj.weight', 'model.layers.1.mlp.experts.7.up_gate_proj.weight', 'model.layers.1.mlp.experts.8.up_gate_proj.weight', 'model.layers.1.mlp.experts.9.up_gate_proj.weight', 'model.layers.1.mlp.experts.10.up_gate_proj.weight', 'model.layers.1.mlp.experts.11.up_gate_proj.weight', 'model.layers.1.mlp.experts.12.up_gate_proj.weight', 'model.layers.1.mlp.experts.13.up_gate_proj.weight', 'model.layers.1.mlp.experts.14.up_gate_proj.weight', 'model.layers.1.mlp.experts.15.up_gate_proj.weight', 'model.layers.1.mlp.experts.16.up_gate_proj.weight', 'model.layers.1.mlp.experts.17.up_gate_proj.weight', 'model.layers.1.mlp.experts.18.up_gate_proj.weight', 'model.layers.1.mlp.experts.19.up_gate_proj.weight', 'model.layers.1.mlp.experts.20.up_gate_proj.weight', 'model.layers.1.mlp.experts.21.up_gate_proj.weight', 'model.layers.1.mlp.experts.22.up_gate_proj.weight', 'model.layers.1.mlp.experts.23.up_gate_proj.weight', 'model.layers.1.mlp.experts.24.up_gate_proj.weight', 'model.layers.1.mlp.experts.25.up_gate_proj.weight', 'model.layers.1.mlp.experts.26.up_gate_proj.weight', 'model.layers.1.mlp.experts.27.up_gate_proj.weight', 'model.layers.1.mlp.experts.28.up_gate_proj.weight', 'model.layers.1.mlp.experts.29.up_gate_proj.weight', 'model.layers.1.mlp.experts.30.up_gate_proj.weight', 'model.layers.1.mlp.experts.31.up_gate_proj.weight', 'model.layers.1.mlp.experts.32.up_gate_proj.weight', 'model.layers.1.mlp.experts.33.up_gate_proj.weight', 'model.layers.1.mlp.experts.34.up_gate_proj.weight', 'model.layers.1.mlp.experts.35.up_gate_proj.weight', 'model.layers.1.mlp.experts.36.up_gate_proj.weight', 'model.layers.1.mlp.experts.37.up_gate_proj.weight', 'model.layers.1.mlp.experts.38.up_gate_proj.weight', 'model.layers.1.mlp.experts.39.up_gate_proj.weight', 'model.layers.1.mlp.experts.40.up_gate_proj.weight', 'model.layers.1.mlp.experts.41.up_gate_proj.weight', 'model.layers.1.mlp.experts.42.up_gate_proj.weight', 'model.layers.1.mlp.experts.43.up_gate_proj.weight', 'model.layers.1.mlp.experts.44.up_gate_proj.weight', 'model.layers.1.mlp.experts.45.up_gate_proj.weight', 'model.layers.1.mlp.experts.46.up_gate_proj.weight', 'model.layers.1.mlp.experts.47.up_gate_proj.weight', 'model.layers.1.mlp.experts.48.up_gate_proj.weight', 'model.layers.1.mlp.experts.49.up_gate_proj.weight', 'model.layers.1.mlp.experts.50.up_gate_proj.weight', 'model.layers.1.mlp.experts.51.up_gate_proj.weight', 'model.layers.1.mlp.experts.52.up_gate_proj.weight', 'model.layers.1.mlp.experts.53.up_gate_proj.weight', 'model.layers.1.mlp.experts.54.up_gate_proj.weight', 'model.layers.1.mlp.experts.55.up_gate_proj.weight', 'model.layers.1.mlp.experts.56.up_gate_proj.weight', 'model.layers.1.mlp.experts.57.up_gate_proj.weight', 'model.layers.1.mlp.experts.58.up_gate_proj.weight', 'model.layers.1.mlp.experts.59.up_gate_proj.weight', 'model.layers.1.mlp.experts.60.up_gate_proj.weight', 'model.layers.1.mlp.experts.61.up_gate_proj.weight', 'model.layers.1.mlp.experts.62.up_gate_proj.weight', 'model.layers.1.mlp.experts.63.up_gate_proj.weight', 'model.layers.1.mlp.experts.64.up_gate_proj.weight', 'model.layers.1.mlp.experts.65.up_gate_proj.weight', 'model.layers.1.mlp.experts.66.up_gate_proj.weight', 'model.layers.1.mlp.experts.67.up_gate_proj.weight', 'model.layers.1.mlp.experts.68.up_gate_proj.weight', 'model.layers.1.mlp.experts.69.up_gate_proj.weight', 'model.layers.1.mlp.experts.70.up_gate_proj.weight', 'model.layers.1.mlp.experts.71.up_gate_proj.weight', 'model.layers.1.mlp.experts.72.up_gate_proj.weight', 'model.layers.1.mlp.experts.73.up_gate_proj.weight', 'model.layers.1.mlp.experts.74.up_gate_proj.weight', 'model.layers.1.mlp.experts.75.up_gate_proj.weight', 'model.layers.1.mlp.experts.76.up_gate_proj.weight', 'model.layers.1.mlp.experts.77.up_gate_proj.weight', 'model.layers.1.mlp.experts.78.up_gate_proj.weight', 'model.layers.1.mlp.experts.79.up_gate_proj.weight', 'model.layers.1.mlp.experts.80.up_gate_proj.weight', 'model.layers.1.mlp.experts.81.up_gate_proj.weight', 'model.layers.1.mlp.experts.82.up_gate_proj.weight', 'model.layers.1.mlp.experts.83.up_gate_proj.weight', 'model.layers.1.mlp.experts.84.up_gate_proj.weight', 'model.layers.1.mlp.experts.85.up_gate_proj.weight', 'model.layers.1.mlp.experts.86.up_gate_proj.weight', 'model.layers.1.mlp.experts.87.up_gate_proj.weight', 'model.layers.1.mlp.experts.88.up_gate_proj.weight', 'model.layers.1.mlp.experts.89.up_gate_proj.weight', 'model.layers.1.mlp.experts.90.up_gate_proj.weight', 'model.layers.1.mlp.experts.91.up_gate_proj.weight', 'model.layers.1.mlp.experts.92.up_gate_proj.weight', 'model.layers.1.mlp.experts.93.up_gate_proj.weight', 'model.layers.1.mlp.experts.94.up_gate_proj.weight', 'model.layers.1.mlp.experts.95.up_gate_proj.weight', 'model.layers.1.mlp.experts.96.up_gate_proj.weight', 'model.layers.1.mlp.experts.97.up_gate_proj.weight', 'model.layers.1.mlp.experts.98.up_gate_proj.weight', 'model.layers.1.mlp.experts.99.up_gate_proj.weight', 'model.layers.1.mlp.experts.100.up_gate_proj.weight', 'model.layers.1.mlp.experts.101.up_gate_proj.weight', 'model.layers.1.mlp.experts.102.up_gate_proj.weight', 'model.layers.1.mlp.experts.103.up_gate_proj.weight', 'model.layers.1.mlp.experts.104.up_gate_proj.weight', 'model.layers.1.mlp.experts.105.up_gate_proj.weight', 'model.layers.1.mlp.experts.106.up_gate_proj.weight', 'model.layers.1.mlp.experts.107.up_gate_proj.weight', 'model.layers.1.mlp.experts.108.up_gate_proj.weight', 'model.layers.1.mlp.experts.109.up_gate_proj.weight', 'model.layers.1.mlp.experts.110.up_gate_proj.weight', 'model.layers.1.mlp.experts.111.up_gate_proj.weight', 'model.layers.1.mlp.experts.112.up_gate_proj.weight', 'model.layers.1.mlp.experts.113.up_gate_proj.weight', 'model.layers.1.mlp.experts.114.up_gate_proj.weight', 'model.layers.1.mlp.experts.115.up_gate_proj.weight', 'model.layers.1.mlp.experts.116.up_gate_proj.weight', 'model.layers.1.mlp.experts.117.up_gate_proj.weight', 'model.layers.1.mlp.experts.118.up_gate_proj.weight', 'model.layers.1.mlp.experts.119.up_gate_proj.weight', 'model.layers.1.mlp.experts.120.up_gate_proj.weight', 'model.layers.1.mlp.experts.121.up_gate_proj.weight', 'model.layers.1.mlp.experts.122.up_gate_proj.weight', 'model.layers.1.mlp.experts.123.up_gate_proj.weight', 'model.layers.1.mlp.experts.124.up_gate_proj.weight', 'model.layers.1.mlp.experts.125.up_gate_proj.weight', 'model.layers.1.mlp.experts.126.up_gate_proj.weight', 'model.layers.1.mlp.experts.127.up_gate_proj.weight']
model.layers.1.mlp.gate.e_score_correction_bias
model.layers.1.mlp.gate.e_score_correction_bias:model.layers.1.mlp.gate.e_score_correction_bias
model.layers.1.mlp.gate.weight
model.layers.1.mlp.gate.weight:model.layers.1.mlp.gate.weight
model.layers.1.mlp.shared_experts.down_proj.weight
model.layers.1.mlp.shared_experts.down_proj.weight:model.layers.1.mlp.shared_experts.down_proj.weight
model.layers.1.mlp.shared_experts.up_gate_proj.weight
model.layers.1.mlp.shared_experts.up_gate_proj.weight:model.layers.1.mlp.shared_experts.up_gate_proj.weight
model.layers.1.post_attention_layernorm.weight
model.layers.1.post_attention_layernorm.weight:model.layers.1.post_attention_layernorm.weight
model.layers.1.self_attn.o_proj.weight
model.layers.1.self_attn.o_proj.weight:model.layers.1.self_attn.o_proj.weight
model.layers.1.self_attn.qkv_proj.bias
model.layers.1.self_attn.qkv_proj.bias:model.layers.1.self_attn.qkv_proj.bias
model.layers.1.self_attn.qkv_proj.weight
model.layers.1.self_attn.qkv_proj.weight:model.layers.1.self_attn.qkv_proj.weight
model.norm.weight
model.norm.weight:model.norm.weight