马上注册，免费下载更多dz插件网资源。
您需要 登录 才可以下载或查看，没有账号？立即注册
×
Loading model for gpu... Traceback (most recent call last):
File "runEx_2.py", line 74, in <module>
model.load_state_dict(m2)
File "C:\Users\ADMIN\AppData\Local\Programs\Python\Python38\lib\site-packages\torch\nn\modules\module.py", line 1406, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for GPT:
Unexpected key(s) in state_dict: "blocks.10.ln1.weight", "blocks.10.ln1.bias", "blocks.10.ln2.weight", "blocks.10.ln2.bias", "blocks.10.attn.time_w", "blocks.10.attn.time_alpha", "blocks.10.attn.time_beta", "blocks.10.attn.time_gamma", "blocks.10.attn.mask", "blocks.10.attn.key.weight", "blocks.10.attn.key.bias", "blocks.10.attn.value.weight", "blocks.10.attn.value.bias", "blocks.10.attn.receptance.weight", "blocks.10.attn.receptance.bias", "blocks.10.attn.output.weight", "blocks.10.attn.output.bias", "blocks.10.mlp.key.weight", "blocks.10.mlp.key.bias", "blocks.10.mlp.value.weight", "blocks.10.mlp.value.bias", "blocks.10.mlp.weight.weight", "blocks.10.mlp.weight.bias", "blocks.10.mlp.receptance.weight", "blocks.10.mlp.receptance.bias", "blocks.11.ln1.weight", "blocks.11.ln1.bias", "blocks.11.ln2.weight", "blocks.11.ln2.bias", "blocks.11.attn.time_w", "blocks.11.attn.time_alpha", "blocks.11.attn.time_beta", "blocks.11.attn.time_gamma", "blocks.11.attn.mask", "blocks.11.attn.key.weight", "blocks.11.attn.key.bias", "blocks.11.attn.value.weight", "blocks.11.attn.value.bias", "blocks.11.attn.receptance.weight", "blocks.11.attn.receptance.bias", "blocks.11.attn.output.weight", "blocks.11.attn.output.bias", "blocks.11.mlp.key.weight", "blocks.11.mlp.key.bias", "blocks.11.mlp.value.weight", "blocks.11.mlp.value.bias", "blocks.11.mlp.weight.weight", "blocks.11.mlp.weight.bias", "blocks.11.mlp.receptance.weight", "blocks.11.mlp.receptance.bias".
size mismatch for tok_emb.weight: copying a param with shape torch.Size([4592, 768]) from checkpoint, the shape in current model is torch.Size([4592, 640]).
size mismatch for blocks.0.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.0.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.0.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.0.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.0.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.0.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.0.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.0.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.0.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.0.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.0.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.0.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.0.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.1.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.1.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.1.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.1.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.1.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.1.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.1.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.1.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.1.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.1.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.1.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.1.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.2.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.2.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.2.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.2.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.2.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.2.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.2.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.2.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.2.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.2.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.2.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.2.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.3.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.3.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.3.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.3.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.3.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.3.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.3.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.3.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.3.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.3.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.3.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.3.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.4.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.4.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.4.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.4.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.4.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.4.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.4.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.4.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.4.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.4.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.4.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.4.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.5.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.5.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.5.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.5.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.5.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.5.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.5.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.5.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.5.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.5.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.5.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.5.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.6.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.6.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.6.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.6.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.6.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.6.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.6.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.6.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.6.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.6.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.6.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.6.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.7.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.7.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.7.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.7.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.7.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.7.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.7.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.7.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.7.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.7.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.7.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.7.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.8.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.8.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.8.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.8.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.8.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.8.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.8.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.8.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.8.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.8.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.8.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.8.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.ln1.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.ln1.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.ln2.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.ln2.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.attn.time_ww: copying a param with shape torch.Size([12, 512, 512]) from checkpoint, the shape in current model is torch.Size([10, 512, 512]).
size mismatch for blocks.9.attn.key.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.9.attn.key.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.attn.value.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.9.attn.value.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.attn.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.9.attn.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.attn.output.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.9.attn.output.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.mlp.key.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.9.mlp.key.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.9.mlp.value.weight: copying a param with shape torch.Size([1920, 768]) from checkpoint, the shape in current model is torch.Size([1600, 640]).
size mismatch for blocks.9.mlp.value.bias: copying a param with shape torch.Size([1920]) from checkpoint, the shape in current model is torch.Size([1600]).
size mismatch for blocks.9.mlp.weight.weight: copying a param with shape torch.Size([768, 1920]) from checkpoint, the shape in current model is torch.Size([640, 1600]).
size mismatch for blocks.9.mlp.weight.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for blocks.9.mlp.receptance.weight: copying a param with shape torch.Size([768, 768]) from checkpoint, the shape in current model is torch.Size([640, 640]).
size mismatch for blocks.9.mlp.receptance.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for ln_f.weight: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for ln_f.bias: copying a param with shape torch.Size([768]) from checkpoint, the shape in current model is torch.Size([640]).
size mismatch for head.weight: copying a param with shape torch.Size([4592, 768]) from checkpoint, the shape in current model is torch.Size([4592, 640]).
©DZ插件网所发布的一切资源仅限用于学习和研究目的；不得将上述内容用于商业或者非法用途，否则，一切后果请用户自负。
网站部分内容来源于网络，版权争议与本站无关。您必须在下载后的24个小时之内，从您的电脑中彻底删除上述内容。如果您喜欢该程序，请支持正版软件，得到更好的正版服务。
您在本站任何的赞助购买、下载、查阅、回复等行为等均表示接受并同意签订《DZ插件网免责声明协议》。
如有侵权请邮件与我们联系处理: discuzaddons@vip.qq.com 并出示相关证明以便删除。敬请谅解！
|
|
MDZ 编辑器 收费版 V1.15.0(wcn_editor)1253 人气#Discuz!插件模板
与dz-x.net同款首页聊天室最新汉化版本！！1465 人气#Discuz!插件模板
定时批量发帖 PC手机双端版1.8(zgxsh_posti109 人气#Discuz!插件模板![【可可】文档文库 96.240911(keke_doc)[新增隐藏分类设置 重大更新 建议立即更新]](https://static.dz-x.net/block/57/57f31f206de522f914d3eedc66d1025d.jpg)
【可可】文档文库 96.240911(keke_doc)[新3781 人气#Discuz!插件模板
网站密码访问 正式版 20200407_2.3(nciaer_627 人气#Discuz!插件模板![[可可]视频点播直播网校 202.240924(keke_video_base)[新增审核机制 新增回收站功能 大版本更新]](https://static.dz-x.net/block/95/95928e5e84043aa7520398c3c98d66d4.jpg)
[可可]视频点播直播网校 202.240924(keke_v3841 人气#Discuz!插件模板
提升卡
置顶卡
沉默卡
喧嚣卡
变色卡
千斤顶
照妖镜
