|
| 1 | +I0209 15:07:04.723858 22823061509952 utils.py:28] Adam params: |
| 2 | + _orig_mod.module.layers.0.attn.attn_scale (ndim=0) |
| 3 | + _orig_mod.module.layers.0.attn_norm.weight (ndim=1) |
| 4 | + _orig_mod.module.layers.0.mlp_norm.weight (ndim=1) |
| 5 | + _orig_mod.module.layers.1.attn.attn_scale (ndim=0) |
| 6 | + _orig_mod.module.layers.1.attn_norm.weight (ndim=1) |
| 7 | + _orig_mod.module.layers.1.mlp_norm.weight (ndim=1) |
| 8 | + _orig_mod.module.layers.2.attn.attn_scale (ndim=0) |
| 9 | + _orig_mod.module.layers.2.attn_norm.weight (ndim=1) |
| 10 | + _orig_mod.module.layers.2.mlp_norm.weight (ndim=1) |
| 11 | + _orig_mod.module.layers.3.attn.attn_scale (ndim=0) |
| 12 | + _orig_mod.module.layers.3.attn_norm.weight (ndim=1) |
| 13 | + _orig_mod.module.layers.3.mlp_norm.weight (ndim=1) |
| 14 | + _orig_mod.module.layers.4.attn.attn_scale (ndim=0) |
| 15 | + _orig_mod.module.layers.4.attn_norm.weight (ndim=1) |
| 16 | + _orig_mod.module.layers.4.mlp_norm.weight (ndim=1) |
| 17 | + _orig_mod.module.layers.5.attn.attn_scale (ndim=0) |
| 18 | + _orig_mod.module.layers.5.attn_norm.weight (ndim=1) |
| 19 | + _orig_mod.module.layers.5.mlp_norm.weight (ndim=1) |
| 20 | + _orig_mod.module.layers.6.attn.attn_scale (ndim=0) |
| 21 | + _orig_mod.module.layers.6.attn_norm.weight (ndim=1) |
| 22 | + _orig_mod.module.layers.6.mlp_norm.weight (ndim=1) |
| 23 | + _orig_mod.module.layers.7.attn.attn_scale (ndim=0) |
| 24 | + _orig_mod.module.layers.7.attn_norm.weight (ndim=1) |
| 25 | + _orig_mod.module.layers.7.mlp_norm.weight (ndim=1) |
| 26 | + _orig_mod.module.layers.8.attn.attn_scale (ndim=0) |
| 27 | + _orig_mod.module.layers.8.attn_norm.weight (ndim=1) |
| 28 | + _orig_mod.module.layers.8.mlp_norm.weight (ndim=1) |
| 29 | + _orig_mod.module.layers.9.attn.attn_scale (ndim=0) |
| 30 | + _orig_mod.module.layers.9.attn_norm.weight (ndim=1) |
| 31 | + _orig_mod.module.layers.9.mlp_norm.weight (ndim=1) |
| 32 | + _orig_mod.module.layers.10.attn.attn_scale (ndim=0) |
| 33 | + _orig_mod.module.layers.10.attn_norm.weight (ndim=1) |
| 34 | + _orig_mod.module.layers.10.mlp_norm.weight (ndim=1) |
| 35 | + _orig_mod.module.layers.11.attn.attn_scale (ndim=0) |
| 36 | + _orig_mod.module.layers.11.attn_norm.weight (ndim=1) |
| 37 | + _orig_mod.module.layers.11.mlp_norm.weight (ndim=1) |
| 38 | + _orig_mod.module.out_norm.weight (ndim=1) |
| 39 | +I0209 15:07:04.727331 23108874655552 utils.py:27] Muon params: |
| 40 | + _orig_mod.module.embed_tokens.weight (ndim=2) |
| 41 | + _orig_mod.module.layers.0.attn.w_qkv.weight (ndim=2) |
| 42 | + _orig_mod.module.layers.0.attn.w_out.weight (ndim=2) |
| 43 | + _orig_mod.module.layers.0.mlp.fc1.weight (ndim=2) |
| 44 | + _orig_mod.module.layers.0.mlp.fc2.weight (ndim=2) |
| 45 | + _orig_mod.module.layers.1.attn.w_qkv.weight (ndim=2) |
| 46 | + _orig_mod.module.layers.1.attn.w_out.weight (ndim=2) |
| 47 | + _orig_mod.module.layers.1.mlp.fc1.weight (ndim=2) |
| 48 | + _orig_mod.module.layers.1.mlp.fc2.weight (ndim=2) |
| 49 | + _orig_mod.module.layers.2.attn.w_qkv.weight (ndim=2) |
| 50 | + _orig_mod.module.layers.2.attn.w_out.weight (ndim=2) |
| 51 | + _orig_mod.module.layers.2.mlp.fc1.weight (ndim=2) |
| 52 | + _orig_mod.module.layers.2.mlp.fc2.weight (ndim=2) |
| 53 | + _orig_mod.module.layers.3.attn.w_qkv.weight (ndim=2) |
| 54 | + _orig_mod.module.layers.3.attn.w_out.weight (ndim=2) |
| 55 | + _orig_mod.module.layers.3.mlp.fc1.weight (ndim=2) |
| 56 | + _orig_mod.module.layers.3.mlp.fc2.weight (ndim=2) |
| 57 | + _orig_mod.module.layers.4.attn.w_qkv.weight (ndim=2) |
| 58 | + _orig_mod.module.layers.4.attn.w_out.weight (ndim=2) |
| 59 | + _orig_mod.module.layers.4.mlp.fc1.weight (ndim=2) |
| 60 | + _orig_mod.module.layers.4.mlp.fc2.weight (ndim=2) |
| 61 | + _orig_mod.module.layers.5.attn.w_qkv.weight (ndim=2) |
| 62 | + _orig_mod.module.layers.5.attn.w_out.weight (ndim=2) |
| 63 | + _orig_mod.module.layers.5.mlp.fc1.weight (ndim=2) |
| 64 | + _orig_mod.module.layers.5.mlp.fc2.weight (ndim=2) |
| 65 | + _orig_mod.module.layers.6.attn.w_qkv.weight (ndim=2) |
| 66 | + _orig_mod.module.layers.6.attn.w_out.weight (ndim=2) |
| 67 | + _orig_mod.module.layers.6.mlp.fc1.weight (ndim=2) |
| 68 | + _orig_mod.module.layers.6.mlp.fc2.weight (ndim=2) |
| 69 | + _orig_mod.module.layers.7.attn.w_qkv.weight (ndim=2) |
| 70 | + _orig_mod.module.layers.7.attn.w_out.weight (ndim=2) |
| 71 | + _orig_mod.module.layers.7.mlp.fc1.weight (ndim=2) |
| 72 | + _orig_mod.module.layers.7.mlp.fc2.weight (ndim=2) |
| 73 | + _orig_mod.module.layers.8.attn.w_qkv.weight (ndim=2) |
| 74 | + _orig_mod.module.layers.8.attn.w_out.weight (ndim=2) |
| 75 | + _orig_mod.module.layers.8.mlp.fc1.weight (ndim=2) |
| 76 | + _orig_mod.module.layers.8.mlp.fc2.weight (ndim=2) |
| 77 | + _orig_mod.module.layers.9.attn.w_qkv.weight (ndim=2) |
| 78 | + _orig_mod.module.layers.9.attn.w_out.weight (ndim=2) |
| 79 | + _orig_mod.module.layers.9.mlp.fc1.weight (ndim=2) |
| 80 | + _orig_mod.module.layers.9.mlp.fc2.weight (ndim=2) |
| 81 | + _orig_mod.module.layers.10.attn.w_qkv.weight (ndim=2) |
| 82 | + _orig_mod.module.layers.10.attn.w_out.weight (ndim=2) |
| 83 | + _orig_mod.module.layers.10.mlp.fc1.weight (ndim=2) |
| 84 | + _orig_mod.module.layers.10.mlp.fc2.weight (ndim=2) |
| 85 | + _orig_mod.module.layers.11.attn.w_qkv.weight (ndim=2) |
| 86 | + _orig_mod.module.layers.11.attn.w_out.weight (ndim=2) |
| 87 | + _orig_mod.module.layers.11.mlp.fc1.weight (ndim=2) |
| 88 | + _orig_mod.module.layers.11.mlp.fc2.weight (ndim=2) |
0 commit comments