Upload 6 files

modeling_chatglm.py

@@ -0,0 +1,1138 @@
+""" PyTorch ChatGLM model. """
+
+import math
+import sys
+import torch
+import torch.utils.checkpoint
+import torch.nn.functional as F
+from torch import nn
+from torch.nn import CrossEntropyLoss, LayerNorm, MSELoss, BCEWithLogitsLoss
+from torch.nn.utils import skip_init
+from typing import Optional, Tuple, Union, List, Dict, Any
+
+from transformers.modeling_outputs import (
+    BaseModelOutputWithPast,
+    CausalLMOutputWithPast,
+    SequenceClassifierOutputWithPast,
+)
+from transformers.modeling_utils import PreTrainedModel
+from transformers.utils import logging, is_torch_npu_available
+from transformers.generation.logits_process import LogitsProcessor
+from transformers.generation.utils import ModelOutput
+
+from .configuration_chatglm import ChatGLMConfig
+
+try:
+    from transformers.utils import is_flash_attn_greater_or_equal_2_10, is_flash_attn_2_available
+
+    if is_flash_attn_2_available():
+        from flash_attn import flash_attn_func, flash_attn_varlen_func
+        from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
+except:
+    pass
+
+# flags required to enable jit fusion kernels
+
+if sys.platform != 'darwin' and not is_torch_npu_available():
+    torch._C._jit_set_profiling_mode(False)
+    torch._C._jit_set_profiling_executor(False)
+    torch._C._jit_override_can_fuse_on_cpu(True)
+    torch._C._jit_override_can_fuse_on_gpu(True)
+
+logger = logging.get_logger(__name__)
+
+_CHECKPOINT_FOR_DOC = "THUDM/ChatGLM"
+_CONFIG_FOR_DOC = "ChatGLMConfig"
+
+
+def default_init(cls, *args, **kwargs):
+    return cls(*args, **kwargs)
+
+
+class InvalidScoreLogitsProcessor(LogitsProcessor):
+    def __call__(self, input_ids: torch.LongTensor, scores: torch.FloatTensor) -> torch.FloatTensor:
+        if torch.isnan(scores).any() or torch.isinf(scores).any():
+            scores.zero_()
+            scores[..., 198] = 5e4
+        return scores
+
+
+def split_tensor_along_last_dim(
+        tensor: torch.Tensor,
+        num_partitions: int,
+        contiguous_split_chunks: bool = False,
+) -> List[torch.Tensor]:
+    """Split a tensor along its last dimension.
+
+    Arguments:
+        tensor: input tensor.
+        num_partitions: number of partitions to split the tensor
+        contiguous_split_chunks: If True, make each chunk contiguous
+                                 in memory.
+
+    Returns:
+        A list of Tensors
+    """
+    # Get the size and dimension.
+    last_dim = tensor.dim() - 1
+    last_dim_size = tensor.size()[last_dim] // num_partitions
+    # Split.
+    tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
+    # Note: torch.split does not create contiguous tensors by default.
+    if contiguous_split_chunks:
+        return tuple(chunk.contiguous() for chunk in tensor_list)
+
+    return tensor_list
+
+
+class RotaryEmbedding(nn.Module):
+    def __init__(self, dim, rope_ratio=1, original_impl=False, device=None, dtype=None):
+        super().__init__()
+        inv_freq = 1.0 / (10000 ** (torch.arange(0, dim, 2, device=device).to(dtype=dtype) / dim))
+        self.register_buffer("inv_freq", inv_freq)
+        self.dim = dim
+        self.original_impl = original_impl
+        self.rope_ratio = rope_ratio
+
+    def forward_impl(
+            self, seq_len: int, n_elem: int, dtype: torch.dtype, device: torch.device, base: int = 10000
+    ):
+        """Enhanced Transformer with Rotary Position Embedding.
+
+        Derived from: https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/master/labml_nn/
+        transformers/rope/__init__.py. MIT License:
+        https://github.com/labmlai/annotated_deep_learning_paper_implementations/blob/master/license.
+        """
+        # $\Theta = {\theta_i = 10000^{\frac{2(i-1)}{d}}, i \in [1, 2, ..., \frac{d}{2}]}$
+        base = base * self.rope_ratio
+        theta = 1.0 / (base ** (torch.arange(0, n_elem, 2, dtype=torch.float, device=device) / n_elem))
+
+        # Create position indexes `[0, 1, ..., seq_len - 1]`
+        seq_idx = torch.arange(seq_len, dtype=torch.float, device=device)
+
+        # Calculate the product of position index and $\theta_i$
+        idx_theta = torch.outer(seq_idx, theta).float()
+
+        cache = torch.stack([torch.cos(idx_theta), torch.sin(idx_theta)], dim=-1)
+
+        # this is to mimic the behaviour of complex32, else we will get different results
+        if dtype in (torch.float16, torch.bfloat16, torch.int8):
+            cache = cache.bfloat16() if dtype == torch.bfloat16 else cache.half()
+        return cache
+
+    def forward(self, max_seq_len, offset=0):
+        return self.forward_impl(
+            max_seq_len, self.dim, dtype=self.inv_freq.dtype, device=self.inv_freq.device
+        )
+
+
+@torch.jit.script
+def apply_rotary_pos_emb(x: torch.Tensor, rope_cache: torch.Tensor) -> torch.Tensor:
+    # x: [b, np, sq, hn]
+    b, np, sq, hn = x.size(0), x.size(1), x.size(2), x.size(3)
+    rot_dim = rope_cache.shape[-2] * 2
+    x, x_pass = x[..., :rot_dim], x[..., rot_dim:]
+    # truncate to support variable sizes
+    rope_cache = rope_cache[:, :sq]
+    xshaped = x.reshape(b, np, sq, rot_dim // 2, 2)
+    rope_cache = rope_cache.view(-1, 1, sq, xshaped.size(3), 2)
+    x_out2 = torch.stack(
+        [
+            xshaped[..., 0] * rope_cache[..., 0] - xshaped[..., 1] * rope_cache[..., 1],
+            xshaped[..., 1] * rope_cache[..., 0] + xshaped[..., 0] * rope_cache[..., 1],
+        ],
+        -1,
+    )
+    x_out2 = x_out2.flatten(3)
+    return torch.cat((x_out2, x_pass), dim=-1)
+
+
+class RMSNorm(torch.nn.Module):
+    def __init__(self, normalized_shape, eps=1e-5, device=None, dtype=None, **kwargs):
+        super().__init__()
+        self.weight = torch.nn.Parameter(torch.empty(normalized_shape, device=device, dtype=dtype))
+        self.eps = eps
+
+    def forward(self, hidden_states: torch.Tensor):
+        input_dtype = hidden_states.dtype
+        variance = hidden_states.to(torch.float32).pow(2).mean(-1, keepdim=True)
+        hidden_states = hidden_states * torch.rsqrt(variance + self.eps)
+
+        return (self.weight * hidden_states).to(input_dtype)
+
+
+class CoreAttention(torch.nn.Module):
+    def __init__(self, config: ChatGLMConfig, layer_number):
+        super(CoreAttention, self).__init__()
+        self.config = config
+        self.apply_query_key_layer_scaling = config.apply_query_key_layer_scaling
+        self.attention_softmax_in_fp32 = config.attention_softmax_in_fp32
+        if self.apply_query_key_layer_scaling:
+            self.attention_softmax_in_fp32 = True
+        self.layer_number = max(1, layer_number)
+        self.is_causal = True
+
+        projection_size = config.kv_channels * config.num_attention_heads
+
+        # Per attention head and per partition values.
+        self.hidden_size_per_partition = projection_size
+        self.hidden_size_per_attention_head = projection_size // config.num_attention_heads
+        self.num_attention_heads_per_partition = config.num_attention_heads
+
+        coeff = None
+        self.norm_factor = math.sqrt(self.hidden_size_per_attention_head)
+        if self.apply_query_key_layer_scaling:
+            coeff = self.layer_number
+            self.norm_factor *= coeff
+        self.coeff = coeff
+
+        self.attention_dropout = torch.nn.Dropout(config.attention_dropout)
+
+    def forward(self, query_layer, key_layer, value_layer, attention_mask):
+        # [b, np, sq, sk]
+        output_size = (query_layer.size(0), query_layer.size(1), query_layer.size(2), key_layer.size(2))
+
+        # [b, np, sq, hn] -> [b * np, sq, hn]
+        query_layer = query_layer.view(output_size[0] * output_size[1], output_size[2], -1)
+        # [b, np, sk, hn] -> [b * np, sk, hn]
+        key_layer = key_layer.view(output_size[0] * output_size[1], output_size[3], -1)
+
+        # preallocting input tensor: [b * np, sq, sk]
+        matmul_input_buffer = torch.empty(
+            output_size[0] * output_size[1], output_size[2], output_size[3], dtype=query_layer.dtype,
+            device=query_layer.device
+        )
+
+        # Raw attention scores. [b * np, sq, sk]
+        matmul_result = torch.baddbmm(
+            matmul_input_buffer,
+            query_layer,  # [b * np, sq, hn]
+            key_layer.transpose(1, 2),  # [b * np, hn, sk]
+            beta=0.0,
+            alpha=(1.0 / self.norm_factor),
+        )
+
+        # change view to [b, np, sq, sk]
+        attention_scores = matmul_result.view(*output_size)
+
+        # ===========================
+        # Attention probs and dropout
+        # ===========================
+
+        # attention scores and attention mask [b, np, sq, sk]
+        if self.attention_softmax_in_fp32:
+            attention_scores = attention_scores.float()
+        if self.coeff is not None:
+            attention_scores = attention_scores * self.coeff
+        if attention_mask is None and attention_scores.shape[2] == attention_scores.shape[3]:
+            attention_mask = torch.ones(output_size[0], 1, output_size[2], output_size[3],
+                                        device=attention_scores.device, dtype=torch.bool)
+            attention_mask.tril_()
+            attention_mask = ~attention_mask
+        if attention_mask is not None:
+            attention_scores = attention_scores.masked_fill(attention_mask, float("-inf"))
+        attention_probs = F.softmax(attention_scores, dim=-1)
+        attention_probs = attention_probs.type_as(value_layer)
+
+        # This is actually dropping out entire tokens to attend to, which might
+        # seem a bit unusual, but is taken from the original Transformer paper.
+        attention_probs = self.attention_dropout(attention_probs)
+
+        # query layer shape: [b * np, sq, hn]
+        # value layer shape: [b, np, sk, hn]
+        # attention shape: [b, np, sq, sk]
+        # context layer shape: [b, np, sq, hn]
+        output_size = (value_layer.size(0), value_layer.size(1), query_layer.size(1), value_layer.size(3))
+        # change view [b * np, sk, hn]
+        value_layer = value_layer.view(output_size[0] * output_size[1], value_layer.size(2), -1)
+        # change view [b * np, sq, sk]
+        attention_probs = attention_probs.view(output_size[0] * output_size[1], output_size[2], -1)
+        # matmul: [b * np, sq, hn]
+        context_layer = torch.bmm(attention_probs, value_layer)
+        # change view [b, np, sq, hn]
+        context_layer = context_layer.view(*output_size)
+        # [b, np, sq, hn] --> [b, sq, np, hn]
+        context_layer = context_layer.transpose(1, 2).contiguous()
+        # [b, sq, np, hn] --> [b, sq, hp]
+        new_context_layer_shape = context_layer.size()[:-2] + (self.hidden_size_per_partition,)
+        context_layer = context_layer.reshape(*new_context_layer_shape)
+
+        return context_layer
+
+
+class SdpaAttention(CoreAttention):
+    def forward(self, query_layer, key_layer, value_layer, attention_mask):
+        if attention_mask is None and query_layer.shape[2] == key_layer.shape[2]:
+            context_layer = torch.nn.functional.scaled_dot_product_attention(query_layer, key_layer, value_layer,
+                                                                             is_causal=True,
+                                                                             dropout_p=self.config.attention_dropout if self.training else 0.0)
+        else:
+            if attention_mask is not None:
+                attention_mask = ~attention_mask
+            context_layer = torch.nn.functional.scaled_dot_product_attention(query_layer, key_layer, value_layer,
+                                                                             attention_mask,
+                                                                             dropout_p=self.config.attention_dropout if self.training else 0.0)
+        context_layer = context_layer.transpose(1, 2).contiguous()
+        new_context_layer_shape = context_layer.size()[:-2] + (self.hidden_size_per_partition,)
+        context_layer = context_layer.reshape(*new_context_layer_shape)
+        return context_layer
+
+
+def _get_unpad_data(attention_mask):
+    seqlens_in_batch = attention_mask.sum(dim=-1, dtype=torch.int32)
+    indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()
+    max_seqlen_in_batch = seqlens_in_batch.max().item()
+    cu_seqlens = F.pad(torch.cumsum(seqlens_in_batch, dim=0, dtype=torch.int32), (1, 0))
+    return (
+        indices,
+        cu_seqlens,
+        max_seqlen_in_batch,
+    )
+
+
+# Copied from transformers.models.llama.modeling_llama.LlamaFlashAttention2
+class FlashAttention2(CoreAttention):
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+        self._flash_attn_uses_top_left_mask = not is_flash_attn_greater_or_equal_2_10()
+
+    def forward(self, query_states, key_states, value_states, attention_mask):
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+        value_states = value_states.transpose(1, 2)
+        batch_size, query_length = query_states.shape[:2]
+        if not self._flash_attn_uses_top_left_mask:
+            causal = self.is_causal
+        else:
+            # TODO: Remove the `query_length != 1` check once Flash Attention for RoCm is bumped to 2.1. For details, please see the comment in LlamaFlashAttention2 __init__.
+            causal = self.is_causal and query_length != 1
+        dropout = self.config.attention_dropout if self.training else 0.0
+        # Contains at least one padding token in the sequence
+        if attention_mask is not None:
+            query_states, key_states, value_states, indices_q, cu_seq_lens, max_seq_lens = self._upad_input(
+                query_states, key_states, value_states, attention_mask, query_length
+            )
+
+            cu_seqlens_q, cu_seqlens_k = cu_seq_lens
+            max_seqlen_in_batch_q, max_seqlen_in_batch_k = max_seq_lens
+
+            attn_output_unpad = flash_attn_varlen_func(
+                query_states,
+                key_states,
+                value_states,
+                cu_seqlens_q=cu_seqlens_q,
+                cu_seqlens_k=cu_seqlens_k,
+                max_seqlen_q=max_seqlen_in_batch_q,
+                max_seqlen_k=max_seqlen_in_batch_k,
+                dropout_p=dropout,
+                softmax_scale=None,
+                causal=causal,
+            )
+
+            attn_output = pad_input(attn_output_unpad, indices_q, batch_size, query_length)
+        else:
+            attn_output = flash_attn_func(
+                query_states, key_states, value_states, dropout, softmax_scale=None, causal=causal
+            )
+        attn_output = attn_output.reshape(batch_size, query_length, self.hidden_size_per_partition).contiguous()
+        return attn_output
+
+    def _upad_input(self, query_layer, key_layer, value_layer, attention_mask, query_length):
+        indices_k, cu_seqlens_k, max_seqlen_in_batch_k = _get_unpad_data(attention_mask)
+        batch_size, kv_seq_len, num_key_value_heads, head_dim = key_layer.shape
+
+        key_layer = index_first_axis(
+            key_layer.reshape(batch_size * kv_seq_len, num_key_value_heads, head_dim), indices_k
+        )
+        value_layer = index_first_axis(
+            value_layer.reshape(batch_size * kv_seq_len, num_key_value_heads, head_dim), indices_k
+        )
+        if query_length == kv_seq_len:
+            query_layer = index_first_axis(
+                query_layer.reshape(batch_size * kv_seq_len, self.num_attention_heads_per_partition, head_dim),
+                indices_k
+            )
+            cu_seqlens_q = cu_seqlens_k
+            max_seqlen_in_batch_q = max_seqlen_in_batch_k
+            indices_q = indices_k
+        elif query_length == 1:
+            max_seqlen_in_batch_q = 1
+            cu_seqlens_q = torch.arange(
+                batch_size + 1, dtype=torch.int32, device=query_layer.device
+            )  # There is a memcpy here, that is very bad.
+            indices_q = cu_seqlens_q[:-1]
+            query_layer = query_layer.squeeze(1)
+        else:
+            # The -q_len: slice assumes left padding.
+            attention_mask = attention_mask[:, -query_length:]
+            query_layer, indices_q, cu_seqlens_q, max_seqlen_in_batch_q = unpad_input(query_layer, attention_mask)
+
+        return (
+            query_layer,
+            key_layer,
+            value_layer,
+            indices_q,
+            (cu_seqlens_q, cu_seqlens_k),
+            (max_seqlen_in_batch_q, max_seqlen_in_batch_k),
+        )
+
+
+CORE_ATTENTION_CLASSES = {
+    "eager": CoreAttention,
+    "sdpa": SdpaAttention,
+    "flash_attention_2": FlashAttention2
+}
+
+
+class SelfAttention(torch.nn.Module):
+    """Parallel self-attention layer abstract class.
+
+    Self-attention layer takes input with size [s, b, h]
+    and returns output of the same size.
+    """
+
+    def __init__(self, config: ChatGLMConfig, layer_number, device=None):
+        super(SelfAttention, self).__init__()
+        self.layer_number = max(1, layer_number)
+
+        self.projection_size = config.kv_channels * config.num_attention_heads
+
+        # Per attention head and per partition values.
+        self.hidden_size_per_attention_head = self.projection_size // config.num_attention_heads
+        self.num_attention_heads_per_partition = config.num_attention_heads
+
+        self.multi_query_attention = config.multi_query_attention
+        self.qkv_hidden_size = 3 * self.projection_size
+        if self.multi_query_attention:
+            self.num_multi_query_groups_per_partition = config.multi_query_group_num
+            self.qkv_hidden_size = (
+                    self.projection_size + 2 * self.hidden_size_per_attention_head * config.multi_query_group_num
+            )
+        self.query_key_value = nn.Linear(config.hidden_size, self.qkv_hidden_size,
+                                         bias=config.add_bias_linear or config.add_qkv_bias,
+                                         device=device, **_config_to_kwargs(config)
+                                         )
+
+        self.core_attention = CORE_ATTENTION_CLASSES[config._attn_implementation](config, self.layer_number)
+
+        # Output.
+        self.dense = nn.Linear(self.projection_size, config.hidden_size, bias=config.add_bias_linear,
+                               device=device, **_config_to_kwargs(config)
+                               )
+
+    def _allocate_memory(self, inference_max_sequence_len, batch_size, device=None, dtype=None):
+        if self.multi_query_attention:
+            num_attention_heads = self.num_multi_query_groups_per_partition
+        else:
+            num_attention_heads = self.num_attention_heads_per_partition
+        return torch.empty(
+            inference_max_sequence_len,
+            batch_size,
+            num_attention_heads,
+            self.hidden_size_per_attention_head,
+            dtype=dtype,
+            device=device,
+        )
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_cache=None, use_cache=True
+    ):
+        # hidden_states: [b, sq, h]
+
+        # =================================================
+        # Pre-allocate memory for key-values for inference.
+        # =================================================
+        # =====================
+        # Query, Key, and Value
+        # =====================
+
+        # Attention heads [b, sq, h] --> [b, sq, (np * 3 * hn)]
+        mixed_x_layer = self.query_key_value(hidden_states)
+
+        if self.multi_query_attention:
+            (query_layer, key_layer, value_layer) = mixed_x_layer.split(
+                [
+                    self.num_attention_heads_per_partition * self.hidden_size_per_attention_head,
+                    self.num_multi_query_groups_per_partition * self.hidden_size_per_attention_head,
+                    self.num_multi_query_groups_per_partition * self.hidden_size_per_attention_head,
+                ],
+                dim=-1,
+            )
+            query_layer = query_layer.view(
+                query_layer.size()[:-1] + (self.num_attention_heads_per_partition, self.hidden_size_per_attention_head)
+            )
+            key_layer = key_layer.view(
+                key_layer.size()[:-1] + (self.num_multi_query_groups_per_partition, self.hidden_size_per_attention_head)
+            )
+            value_layer = value_layer.view(
+                value_layer.size()[:-1]
+                + (self.num_multi_query_groups_per_partition, self.hidden_size_per_attention_head)
+            )
+        else:
+            new_tensor_shape = mixed_x_layer.size()[:-1] + \
+                               (self.num_attention_heads_per_partition,
+                                3 * self.hidden_size_per_attention_head)
+            mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)
+
+            # [b, sq, np, 3 * hn] --> 3 [b, sq, np, hn]
+            (query_layer, key_layer, value_layer) = split_tensor_along_last_dim(mixed_x_layer, 3)
+
+        # [b, sq, np, hn] -> [b, np, sq, hn]
+        query_layer, key_layer, value_layer = [k.transpose(1, 2) for k in [query_layer, key_layer, value_layer]]
+
+        # apply relative positional encoding (rotary embedding)
+        if rotary_pos_emb is not None:
+            query_layer = apply_rotary_pos_emb(query_layer, rotary_pos_emb)
+            key_layer = apply_rotary_pos_emb(key_layer, rotary_pos_emb)
+
+        # adjust key and value for inference
+        if kv_cache is not None:
+            cache_k, cache_v = kv_cache
+            key_layer = torch.cat((cache_k, key_layer), dim=2)
+            value_layer = torch.cat((cache_v, value_layer), dim=2)
+        if use_cache:
+            if kv_cache is None:
+                kv_cache = torch.cat((key_layer.unsqueeze(0).unsqueeze(0), value_layer.unsqueeze(0).unsqueeze(0)),
+                                     dim=1)
+            else:
+                kv_cache = (key_layer, value_layer)
+        else:
+            kv_cache = None
+
+        if self.multi_query_attention:
+            key_layer = key_layer.unsqueeze(2)
+            key_layer = key_layer.expand(
+                -1, -1, self.num_attention_heads_per_partition // self.num_multi_query_groups_per_partition, -1, -1
+            )
+            key_layer = key_layer.contiguous().view(
+                key_layer.size()[:1] + (self.num_attention_heads_per_partition,) + key_layer.size()[3:]
+            )
+            value_layer = value_layer.unsqueeze(2)
+            value_layer = value_layer.expand(
+                -1, -1, self.num_attention_heads_per_partition // self.num_multi_query_groups_per_partition, -1, -1
+            )
+            value_layer = value_layer.contiguous().view(
+                value_layer.size()[:1] + (self.num_attention_heads_per_partition,) + value_layer.size()[3:]
+            )
+
+        # ==================================
+        # core attention computation
+        # ==================================
+
+        context_layer = self.core_attention(query_layer, key_layer, value_layer, attention_mask)
+
+        # =================
+        # Output. [sq, b, h]
+        # =================
+
+        output = self.dense(context_layer)
+
+        return output, kv_cache
+
+
+def _config_to_kwargs(args):
+    common_kwargs = {
+        "dtype": args.torch_dtype,
+    }
+    return common_kwargs
+
+
+class MLP(torch.nn.Module):
+    """MLP.
+
+    MLP will take the input with h hidden state, project it to 4*h
+    hidden dimension, perform nonlinear transformation, and project the
+    state back into h hidden dimension.
+    """
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(MLP, self).__init__()
+
+        self.add_bias = config.add_bias_linear
+
+        # Project to 4h. If using swiglu double the output width, see https://arxiv.org/pdf/2002.05202.pdf
+        self.dense_h_to_4h = nn.Linear(
+            config.hidden_size,
+            config.ffn_hidden_size * 2,
+            bias=self.add_bias,
+            device=device,
+            **_config_to_kwargs(config)
+        )
+
+        def swiglu(x):
+            x = torch.chunk(x, 2, dim=-1)
+            return F.silu(x[0]) * x[1]
+
+        self.activation_func = swiglu
+
+        # Project back to h.
+        self.dense_4h_to_h = nn.Linear(
+            config.ffn_hidden_size,
+            config.hidden_size,
+            bias=self.add_bias,
+            device=device,
+            **_config_to_kwargs(config)
+        )
+
+    def forward(self, hidden_states):
+        # [s, b, 4hp]
+        intermediate_parallel = self.dense_h_to_4h(hidden_states)
+        intermediate_parallel = self.activation_func(intermediate_parallel)
+        # [s, b, h]
+        output = self.dense_4h_to_h(intermediate_parallel)
+        return output
+
+
+class GLMBlock(torch.nn.Module):
+    """A single transformer layer.
+
+    Transformer layer takes input with size [s, b, h] and returns an
+    output of the same size.
+    """
+
+    def __init__(self, config: ChatGLMConfig, layer_number, device=None):
+        super(GLMBlock, self).__init__()
+        self.layer_number = layer_number
+
+        self.apply_residual_connection_post_layernorm = config.apply_residual_connection_post_layernorm
+
+        self.fp32_residual_connection = config.fp32_residual_connection
+
+        LayerNormFunc = RMSNorm if config.rmsnorm else LayerNorm
+        # Layernorm on the input data.
+        self.input_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                             dtype=config.torch_dtype)
+
+        # Self attention.
+        self.self_attention = SelfAttention(config, layer_number, device=device)
+        self.hidden_dropout = config.hidden_dropout
+
+        # Layernorm on the attention output
+        self.post_attention_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                                      dtype=config.torch_dtype)
+
+        # MLP
+        self.mlp = MLP(config, device=device)
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_cache=None, use_cache=True,
+    ):
+        # hidden_states: [s, b, h]
+
+        # Layer norm at the beginning of the transformer layer.
+        layernorm_output = self.input_layernorm(hidden_states)
+        # Self attention.
+        attention_output, kv_cache = self.self_attention(
+            layernorm_output,
+            attention_mask,
+            rotary_pos_emb,
+            kv_cache=kv_cache,
+            use_cache=use_cache
+        )
+
+        # Residual connection.
+        if self.apply_residual_connection_post_layernorm:
+            residual = layernorm_output
+        else:
+            residual = hidden_states
+
+        layernorm_input = torch.nn.functional.dropout(attention_output, p=self.hidden_dropout, training=self.training)
+        layernorm_input = residual + layernorm_input
+
+        # Layer norm post the self attention.
+        layernorm_output = self.post_attention_layernorm(layernorm_input)
+
+        # MLP.
+        mlp_output = self.mlp(layernorm_output)
+
+        # Second residual connection.
+        if self.apply_residual_connection_post_layernorm:
+            residual = layernorm_output
+        else:
+            residual = layernorm_input
+
+        output = torch.nn.functional.dropout(mlp_output, p=self.hidden_dropout, training=self.training)
+        output = residual + output
+
+        return output, kv_cache
+
+
+class GLMTransformer(torch.nn.Module):
+    """Transformer class."""
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(GLMTransformer, self).__init__()
+
+        self.fp32_residual_connection = config.fp32_residual_connection
+        self.post_layer_norm = config.post_layer_norm
+
+        # Number of layers.
+        self.num_layers = config.num_layers
+
+        # Transformer layers.
+        def build_layer(layer_number):
+            return GLMBlock(config, layer_number, device=device)
+
+        self.layers = torch.nn.ModuleList([build_layer(i + 1) for i in range(self.num_layers)])
+
+        if self.post_layer_norm:
+            LayerNormFunc = RMSNorm if config.rmsnorm else LayerNorm
+            # Final layer norm before output.
+            self.final_layernorm = LayerNormFunc(config.hidden_size, eps=config.layernorm_epsilon, device=device,
+                                                 dtype=config.torch_dtype)
+
+        self.gradient_checkpointing = False
+
+    def _get_layer(self, layer_number):
+        return self.layers[layer_number]
+
+    def forward(
+            self, hidden_states, attention_mask, rotary_pos_emb, kv_caches=None,
+            use_cache: Optional[bool] = True,
+            output_hidden_states: Optional[bool] = False,
+    ):
+        if not kv_caches:
+            kv_caches = [None for _ in range(self.num_layers)]
+        presents = () if use_cache else None
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                logger.warning_once(
+                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+                )
+                use_cache = False
+
+        all_self_attentions = None
+        all_hidden_states = () if output_hidden_states else None
+        for index in range(self.num_layers):
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            layer = self._get_layer(index)
+            if self.gradient_checkpointing and self.training:
+                layer_ret = torch.utils.checkpoint.checkpoint(
+                    layer,
+                    hidden_states,
+                    attention_mask,
+                    rotary_pos_emb,
+                    kv_caches[index],
+                    use_cache,
+                    use_reentrant=False
+                )
+            else:
+                layer_ret = layer(
+                    hidden_states,
+                    attention_mask,
+                    rotary_pos_emb,
+                    kv_cache=kv_caches[index],
+                    use_cache=use_cache
+                )
+            hidden_states, kv_cache = layer_ret
+            if use_cache:
+                # token by token decoding, use tuple format
+                if kv_caches[0] is not None:
+                    presents = presents + (kv_cache,)
+                # prefilling in decoding, use tensor format to save cuda memory
+                else:
+                    if len(presents) == 0:
+                        presents = kv_cache
+                    else:
+                        presents = torch.cat((presents, kv_cache.to(presents.device)), dim=0)
+
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        # Final layer norm.
+        if self.post_layer_norm:
+            hidden_states = self.final_layernorm(hidden_states)
+
+        return hidden_states, presents, all_hidden_states, all_self_attentions
+
+
+class ChatGLMPreTrainedModel(PreTrainedModel):
+    """
+    An abstract class to handle weights initialization and
+    a simple interface for downloading and loading pretrained models.
+    """
+
+    is_parallelizable = False
+    supports_gradient_checkpointing = True
+    config_class = ChatGLMConfig
+    base_model_prefix = "transformer"
+    _no_split_modules = ["GLMBlock"]
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+
+    def _init_weights(self, module: nn.Module):
+        """Initialize the weights."""
+        return
+
+    def get_masks(self, input_ids, past_key_values, padding_mask=None):
+        if self.config._attn_implementation == "flash_attention_2":
+            if padding_mask is not None and not padding_mask.all():
+                return padding_mask
+            return None
+        batch_size, seq_length = input_ids.shape
+        full_attention_mask = torch.ones(batch_size, seq_length, seq_length, device=input_ids.device)
+        full_attention_mask.tril_()
+        past_length = 0
+        if past_key_values:
+            past_length = past_key_values[0][0].shape[2]
+        if past_length:
+            full_attention_mask = torch.cat((torch.ones(batch_size, seq_length, past_length,
+                                                        device=input_ids.device), full_attention_mask), dim=-1)
+        if padding_mask is not None:
+            full_attention_mask = full_attention_mask * padding_mask.unsqueeze(1)
+        if not past_length and padding_mask is not None:
+            full_attention_mask -= padding_mask.unsqueeze(-1) - 1
+        full_attention_mask = (full_attention_mask < 0.5).bool()
+        full_attention_mask.unsqueeze_(1)
+        return full_attention_mask
+
+    def get_position_ids(self, input_ids, device):
+        batch_size, seq_length = input_ids.shape
+        position_ids = torch.arange(seq_length, dtype=torch.long, device=device).unsqueeze(0).repeat(batch_size, 1)
+        return position_ids
+
+class Embedding(torch.nn.Module):
+    """Language model embeddings."""
+
+    def __init__(self, config: ChatGLMConfig, device=None):
+        super(Embedding, self).__init__()
+
+        self.hidden_size = config.hidden_size
+        # Word embeddings (parallel).
+        self.word_embeddings = nn.Embedding(
+            config.padded_vocab_size,
+            self.hidden_size,
+            dtype=config.torch_dtype,
+            device=device
+        )
+        self.fp32_residual_connection = config.fp32_residual_connection
+
+    def forward(self, input_ids):
+        # Embeddings.
+        words_embeddings = self.word_embeddings(input_ids)
+        embeddings = words_embeddings
+        # If the input flag for fp32 residual connection is set, convert for float.
+        if self.fp32_residual_connection:
+            embeddings = embeddings.float()
+        return embeddings
+
+
+class ChatGLMModel(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, device=None, empty_init=True):
+        super().__init__(config)
+        if empty_init:
+            init_method = skip_init
+        else:
+            init_method = default_init
+        init_kwargs = {}
+        if device is not None:
+            init_kwargs["device"] = device
+        self.embedding = init_method(Embedding, config, **init_kwargs)
+        self.num_layers = config.num_layers
+        self.multi_query_group_num = config.multi_query_group_num
+        self.kv_channels = config.kv_channels
+
+        # Rotary positional embeddings
+        self.seq_length = config.seq_length
+        rotary_dim = (
+            config.hidden_size // config.num_attention_heads if config.kv_channels is None else config.kv_channels
+        )
+
+        self.rotary_pos_emb = RotaryEmbedding(rotary_dim // 2, rope_ratio=config.rope_ratio,
+                                              original_impl=config.original_rope,
+                                              device=device, dtype=config.torch_dtype)
+        self.encoder = init_method(GLMTransformer, config, **init_kwargs)
+        self.output_layer = init_method(nn.Linear, config.hidden_size, config.padded_vocab_size, bias=False,
+                                        dtype=config.torch_dtype, **init_kwargs)
+
+    def get_input_embeddings(self):
+        return self.embedding.word_embeddings
+
+    def set_input_embeddings(self, value):
+        self.embedding.word_embeddings = value
+
+    def forward(
+            self,
+            input_ids,
+            position_ids: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.BoolTensor] = None,
+            full_attention_mask: Optional[torch.BoolTensor] = None,
+            past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
+            inputs_embeds: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            output_attentions: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+    ):
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        batch_size, seq_length = input_ids.shape
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embedding(input_ids)
+
+        if full_attention_mask is None:
+            if (attention_mask is not None and not attention_mask.all()) or (past_key_values and seq_length != 1):
+                full_attention_mask = self.get_masks(input_ids, past_key_values, padding_mask=attention_mask)
+
+        # Rotary positional embeddings
+        rotary_pos_emb = self.rotary_pos_emb(self.seq_length)
+        if position_ids is not None:
+            rotary_pos_emb = rotary_pos_emb[position_ids]
+        else:
+            rotary_pos_emb = rotary_pos_emb[None, :seq_length]
+
+        # Run encoder.
+        hidden_states, presents, all_hidden_states, all_self_attentions = self.encoder(
+            inputs_embeds, full_attention_mask, rotary_pos_emb=rotary_pos_emb,
+            kv_caches=past_key_values, use_cache=use_cache, output_hidden_states=output_hidden_states
+        )
+        if presents is not None and type(presents) is torch.Tensor:
+            presents = presents.split(1, dim=0)
+            presents = list(presents)
+            presents = [list(x.squeeze(0).split(1, dim=0)) for x in presents]
+            presents = [tuple([x.squeeze(0) for x in y]) for y in presents]
+            presents = tuple(presents)
+
+        if not return_dict:
+            return tuple(v for v in [hidden_states, presents, all_hidden_states, all_self_attentions] if v is not None)
+
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=presents,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+        )
+
+
+class ChatGLMForConditionalGeneration(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, empty_init=True, device=None):
+        super().__init__(config)
+
+        self.max_sequence_length = config.max_length
+        self.transformer = ChatGLMModel(config, empty_init=empty_init, device=device)
+        self.config = config
+
+    def _update_model_kwargs_for_generation(
+            self,
+            outputs: ModelOutput,
+            model_kwargs: Dict[str, Any],
+            is_encoder_decoder: bool = False,
+    ) -> Dict[str, Any]:
+        # update past_key_values
+        cache_name, cache = self._extract_past_from_model_output(outputs)
+        model_kwargs[cache_name] = cache
+
+        # update attention mask
+        if "attention_mask" in model_kwargs:
+            attention_mask = model_kwargs["attention_mask"]
+            model_kwargs["attention_mask"] = torch.cat(
+                [attention_mask, attention_mask.new_ones((attention_mask.shape[0], 1))], dim=-1
+            )
+
+        # update position ids
+        if "position_ids" in model_kwargs:
+            position_ids = model_kwargs["position_ids"]
+            new_position_id = position_ids[..., -1:].clone()
+            new_position_id += 1
+            model_kwargs["position_ids"] = torch.cat(
+                [position_ids, new_position_id], dim=-1
+            )
+
+        model_kwargs["is_first_forward"] = False
+        return model_kwargs
+
+    def prepare_inputs_for_generation(
+            self,
+            input_ids: torch.LongTensor,
+            past_key_values: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            is_first_forward: bool = True,
+            **kwargs
+    ) -> dict:
+        # only last token for input_ids if past is not None
+        if position_ids is None:
+            position_ids = self.get_position_ids(input_ids, device=input_ids.device)
+        if not is_first_forward:
+            if past_key_values is not None:
+                position_ids = position_ids[..., -1:]
+                input_ids = input_ids[:, -1:]
+        return {
+            "input_ids": input_ids,
+            "past_key_values": past_key_values,
+            "position_ids": position_ids,
+            "attention_mask": attention_mask,
+            "return_last_logit": True,
+            "use_cache": use_cache
+        }
+
+    def forward(
+            self,
+            input_ids: Optional[torch.Tensor] = None,
+            position_ids: Optional[torch.Tensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            past_key_values: Optional[Tuple[torch.FloatTensor]] = None,
+            inputs_embeds: Optional[torch.Tensor] = None,
+            labels: Optional[torch.Tensor] = None,
+            use_cache: Optional[bool] = None,
+            output_attentions: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+            return_last_logit: Optional[bool] = False,
+    ):
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            attention_mask=attention_mask,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+        if return_last_logit:
+            hidden_states = hidden_states[:, -1:]
+        lm_logits = self.transformer.output_layer(hidden_states)
+
+        loss = None
+        if labels is not None:
+            lm_logits = lm_logits.to(torch.float32)
+
+            # Shift so that tokens < n predict n
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss(ignore_index=-100)
+            loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+
+            lm_logits = lm_logits.to(hidden_states.dtype)
+            loss = loss.to(hidden_states.dtype)
+
+        if not return_dict:
+            output = (lm_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=lm_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )
+
+    @staticmethod
+    def _reorder_cache(
+            past: Tuple[Tuple[torch.Tensor, torch.Tensor], ...], beam_idx: torch.LongTensor
+    ) -> Tuple[Tuple[torch.Tensor, torch.Tensor], ...]:
+        """
+        This function is used to re-order the `past_key_values` cache if [`~PreTrainedModel.beam_search`] or
+        [`~PreTrainedModel.beam_sample`] is called. This is required to match `past_key_values` with the correct
+        beam_idx at every generation step.
+
+        Output shares the same memory storage as `past`.
+        """
+        return tuple(
+            (
+                layer_past[0].index_select(0, beam_idx.to(layer_past[0].device)),
+                layer_past[1].index_select(0, beam_idx.to(layer_past[1].device)),
+            )
+            for layer_past in past
+        )
+
+
+class ChatGLMForSequenceClassification(ChatGLMPreTrainedModel):
+    def __init__(self, config: ChatGLMConfig, empty_init=True, device=None):
+        super().__init__(config)
+
+        self.num_labels = config.num_labels
+        self.transformer = ChatGLMModel(config, empty_init=empty_init, device=device)
+
+        self.classifier_head = nn.Linear(config.hidden_size, config.num_labels, bias=True, dtype=config.torch_dtype)
+        if config.classifier_dropout is not None:
+            self.dropout = nn.Dropout(config.classifier_dropout)
+        else:
+            self.dropout = None
+        self.config = config
+
+    def forward(
+            self,
+            input_ids: Optional[torch.LongTensor] = None,
+            position_ids: Optional[torch.LongTensor] = None,
+            attention_mask: Optional[torch.Tensor] = None,
+            full_attention_mask: Optional[torch.Tensor] = None,
+            past_key_values: Optional[Tuple[Tuple[torch.Tensor, torch.Tensor], ...]] = None,
+            inputs_embeds: Optional[torch.LongTensor] = None,
+            labels: Optional[torch.LongTensor] = None,
+            use_cache: Optional[bool] = None,
+            output_attentions: Optional[bool] = None,
+            output_hidden_states: Optional[bool] = None,
+            return_dict: Optional[bool] = None,
+    ) -> Union[Tuple[torch.Tensor, ...], SequenceClassifierOutputWithPast]:
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids=input_ids,
+            position_ids=position_ids,
+            attention_mask=attention_mask,
+            full_attention_mask=full_attention_mask,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = transformer_outputs[0]
+        pooled_hidden_states = hidden_states[:, -1]
+        if self.dropout is not None:
+            pooled_hidden_states = self.dropout(pooled_hidden_states)
+        logits = self.classifier_head(pooled_hidden_states)
+
+        loss = None
+        if labels is not None:
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(logits.squeeze().float(), labels.squeeze())
+                else:
+                    loss = loss_fct(logits.float(), labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(logits.view(-1, self.num_labels).float(), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(logits.float(), labels.view(-1, self.num_labels))
+
+        if not return_dict:
+            output = (logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )

sft_args.json

@@ -0,0 +1,258 @@
+{
+  "model_type": "glm4-9b-chat",
+  "model_id_or_path": "/data/SharedFile/ZhipuAI/glm-4-9b-chat",
+  "model_revision": "master",
+  "full_determinism": false,
+  "sft_type": "full",
+  "freeze_parameters": [],
+  "freeze_vit": false,
+  "freeze_parameters_ratio": 0.0,
+  "additional_trainable_parameters": [],
+  "tuner_backend": "peft",
+  "template_type": "chatglm4",
+  "output_dir": "/data/luol/project/Taiyi2/models/glm4-9b-chat/Taiyi2-chat",
+  "add_output_dir_suffix": false,
+  "ddp_backend": null,
+  "ddp_find_unused_parameters": null,
+  "ddp_broadcast_buffers": null,
+  "ddp_timeout": 1800,
+  "seed": 42,
+  "resume_from_checkpoint": null,
+  "resume_only_model": false,
+  "ignore_data_skip": false,
+  "dtype": "bf16",
+  "packing": false,
+  "train_backend": "transformers",
+  "tp": 1,
+  "pp": 1,
+  "min_lr": null,
+  "sequence_parallel": false,
+  "model_kwargs": {},
+  "loss_name": null,
+  "dataset": [
+    "/data/luol/project/Taiyi2/training_data/taiyi2-v5.jsonl"
+  ],
+  "val_dataset": [],
+  "dataset_seed": null,
+  "dataset_test_ratio": 0.01,
+  "use_loss_scale": false,
+  "loss_scale_config_path": "/data/luol/anaconda3/envs/swift26/lib/python3.8/site-packages/swift/llm/agent/default_loss_scale_config.json",
+  "system": null,
+  "tools_prompt": "react_en",
+  "max_length": 2048,
+  "truncation_strategy": "delete",
+  "check_dataset_strategy": "none",
+  "streaming": false,
+  "streaming_val_size": 0,
+  "streaming_buffer_size": 16384,
+  "model_name": [
+    null,
+    null
+  ],
+  "model_author": [
+    null,
+    null
+  ],
+  "quant_method": null,
+  "quantization_bit": 0,
+  "hqq_axis": 0,
+  "hqq_dynamic_config_path": null,
+  "bnb_4bit_comp_dtype": "bf16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "rescale_image": -1,
+  "target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "output_layer",
+    "query_key_value",
+    "dense_h_to_4h"
+  ],
+  "target_regex": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "lora_bias_trainable": "none",
+  "lora_dtype": null,
+  "lora_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "init_lora_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "rope_scaling": null,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "ia3_feedforward_modules": [],
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "neftune_noise_alpha": null,
+  "neftune_backend": "transformers",
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "use_liger": false,
+  "gradient_checkpointing": true,
+  "vit_use_gc": true,
+  "deepspeed": null,
+  "batch_size": 2,
+  "eval_batch_size": 2,
+  "auto_find_batch_size": false,
+  "num_train_epochs": 5,
+  "max_steps": -1,
+  "optim": "adamw_torch",
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "learning_rate": 0.0001,
+  "weight_decay": 0.1,
+  "gradient_accumulation_steps": 16,
+  "max_grad_norm": 1,
+  "predict_with_generate": false,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": {},
+  "warmup_ratio": 0.05,
+  "warmup_steps": 0,
+  "eval_steps": 50000,
+  "save_steps": 50000,
+  "save_only_model": false,
+  "save_total_limit": 10,
+  "logging_steps": 5,
+  "acc_steps": 1,
+  "dataloader_num_workers": 1,
+  "dataloader_pin_memory": true,
+  "dataloader_drop_last": false,
+  "push_to_hub": false,
+  "hub_model_id": null,
+  "hub_token": null,
+  "hub_private_repo": false,
+  "hub_strategy": "every_save",
+  "test_oom_error": false,
+  "disable_tqdm": false,
+  "lazy_tokenize": false,
+  "preprocess_num_proc": 1,
+  "use_flash_attn": null,
+  "ignore_args_error": true,
+  "check_model_is_latest": true,
+  "logging_dir": "/data/luol/project/Taiyi2/models/glm4-9b-chat/Taiyi2-chat/runs",
+  "report_to": [
+    "tensorboard"
+  ],
+  "acc_strategy": "token",
+  "save_on_each_node": false,
+  "evaluation_strategy": "steps",
+  "save_strategy": "steps",
+  "save_safetensors": true,
+  "gpu_memory_fraction": null,
+  "include_num_input_tokens_seen": false,
+  "local_repo_path": null,
+  "custom_register_path": null,
+  "custom_dataset_info": null,
+  "device_map_config": null,
+  "device_max_memory": [],
+  "max_new_tokens": 2048,
+  "do_sample": null,
+  "temperature": null,
+  "top_k": null,
+  "top_p": null,
+  "repetition_penalty": null,
+  "num_beams": 1,
+  "fsdp": "",
+  "fsdp_config": null,
+  "sequence_parallel_size": 1,
+  "model_layer_cls_name": null,
+  "metric_warmup_step": 0,
+  "fsdp_num": 1,
+  "per_device_train_batch_size": null,
+  "per_device_eval_batch_size": null,
+  "eval_strategy": null,
+  "self_cognition_sample": 0,
+  "train_dataset_mix_ratio": 0.0,
+  "train_dataset_mix_ds": [
+    "ms-bench"
+  ],
+  "train_dataset_sample": -1,
+  "val_dataset_sample": null,
+  "safe_serialization": null,
+  "only_save_model": null,
+  "neftune_alpha": null,
+  "deepspeed_config_path": null,
+  "model_cache_dir": null,
+  "lora_dropout_p": null,
+  "lora_target_modules": [
+    "dense_4h_to_h",
+    "dense",
+    "output_layer",
+    "query_key_value",
+    "dense_h_to_4h"
+  ],
+  "lora_target_regex": null,
+  "lora_modules_to_save": [],
+  "boft_target_modules": [],
+  "boft_modules_to_save": [],
+  "vera_target_modules": [],
+  "vera_modules_to_save": [],
+  "ia3_target_modules": [],
+  "ia3_modules_to_save": [],
+  "custom_train_dataset_path": [],
+  "custom_val_dataset_path": [],
+  "device_map_config_path": null,
+  "push_hub_strategy": null,
+  "use_self_cognition": false,
+  "is_multimodal": false,
+  "is_vision": false,
+  "lora_use_embedding": false,
+  "lora_use_all": true,
+  "lora_m2s_use_embedding": false,
+  "lora_m2s_use_ln": false,
+  "torch_dtype": "torch.bfloat16",
+  "fp16": false,
+  "bf16": true,
+  "rank": -1,
+  "local_rank": -1,
+  "world_size": 1,
+  "local_world_size": 1,
+  "bnb_4bit_compute_dtype": "torch.bfloat16",
+  "load_in_4bit": false,
+  "load_in_8bit": false,
+  "train_sampler_random": true,
+  "train_type": "sft",
+  "training_args": "Seq2SeqTrainingArguments(output_dir='/data/luol/project/Taiyi2/models/glm4-9b-chat/Taiyi2-chat', overwrite_output_dir=False, do_train=False, do_eval=True, do_predict=False, eval_strategy=<IntervalStrategy.STEPS: 'steps'>, prediction_loss_only=False, per_device_train_batch_size=2, per_device_eval_batch_size=2, per_gpu_train_batch_size=None, per_gpu_eval_batch_size=None, gradient_accumulation_steps=16, eval_accumulation_steps=None, eval_delay=0, torch_empty_cache_steps=None, learning_rate=0.0001, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, max_grad_norm=1, num_train_epochs=5, max_steps=-1, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs={}, warmup_ratio=0.05, warmup_steps=0, log_level='passive', log_level_replica='warning', log_on_each_node=True, logging_dir='/data/luol/project/Taiyi2/models/glm4-9b-chat/Taiyi2-chat/runs', logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_first_step=True, logging_steps=5, logging_nan_inf_filter=True, save_strategy=<IntervalStrategy.STEPS: 'steps'>, save_steps=50000, save_total_limit=10, save_safetensors=True, save_on_each_node=False, save_only_model=False, restore_callback_states_from_checkpoint=False, no_cuda=False, use_cpu=False, use_mps_device=False, seed=42, data_seed=None, jit_mode_eval=False, use_ipex=False, bf16=True, fp16=False, fp16_opt_level='O1', half_precision_backend='auto', bf16_full_eval=False, fp16_full_eval=False, tf32=None, local_rank=0, ddp_backend=None, tpu_num_cores=None, tpu_metrics_debug=False, debug=[], dataloader_drop_last=False, eval_steps=50000, dataloader_num_workers=1, dataloader_prefetch_factor=None, past_index=-1, run_name='/data/luol/project/Taiyi2/models/glm4-9b-chat/Taiyi2-chat', disable_tqdm=False, remove_unused_columns=False, label_names=None, load_best_model_at_end=False, metric_for_best_model='loss', greater_is_better=False, ignore_data_skip=False, fsdp=[], fsdp_min_num_params=0, fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, fsdp_transformer_layer_cls_to_wrap=None, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), deepspeed=None, label_smoothing_factor=0.0, optim=<OptimizerNames.ADAMW_TORCH: 'adamw_torch'>, optim_args=None, adafactor=False, group_by_length=False, length_column_name='length', report_to=['tensorboard'], ddp_find_unused_parameters=None, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, dataloader_pin_memory=True, dataloader_persistent_workers=False, skip_memory_metrics=True, use_legacy_prediction_loop=False, push_to_hub=False, resume_from_checkpoint=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_token=None, hub_private_repo=False, hub_always_push=False, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, include_inputs_for_metrics=False, eval_do_concat_batches=True, fp16_backend='auto', evaluation_strategy=None, push_to_hub_model_id=None, push_to_hub_organization=None, push_to_hub_token=None, mp_parameters='', auto_find_batch_size=False, full_determinism=False, torchdynamo=None, ray_scope='last', ddp_timeout=1800, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, dispatch_batches=None, split_batches=None, include_tokens_per_second=False, include_num_input_tokens_seen=False, neftune_noise_alpha=None, optim_target_modules=None, batch_eval_metrics=False, eval_on_start=False, eval_use_gather_object=False, sortish_sampler=False, predict_with_generate=False, generation_max_length=None, generation_num_beams=None, generation_config=GenerationConfig {\n  \"do_sample\": true,\n  \"eos_token_id\": 151329,\n  \"max_new_tokens\": 2048,\n  \"pad_token_id\": 151329,\n  \"temperature\": 0.8,\n  \"top_p\": 0.8\n}\n, acc_strategy='token', loss_name=None, additional_saved_files=[], train_sampler_random=True, metric_warmup_step=0, train_dataset_sample=-1)"
+}

special_tokens_map.json

@@ -0,0 +1,32 @@
+{
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenization_chatglm.py

@@ -0,0 +1,224 @@
+import regex as re
+import base64
+import os
+import tiktoken
+from typing import List, Optional, Union, Dict
+from transformers import PreTrainedTokenizer
+from transformers.utils import PaddingStrategy
+from transformers.tokenization_utils_base import EncodedInput, BatchEncoding
+
+
+class ChatGLM4Tokenizer(PreTrainedTokenizer):
+    vocab_files_names = {"vocab_file": "tokenizer.model"}
+    model_input_names = ["input_ids", "attention_mask", "position_ids"]
+
+    def __init__(
+            self,
+            vocab_file,
+            clean_up_tokenization_spaces=False,
+            **kwargs
+    ):
+        self.name = "GLM4Tokenizer"
+        self.vocab_file = vocab_file
+        pat_str = "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?\\p{L}+|\\p{N}{1,3}| ?[^\\s\\p{L}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+"
+        self.pat_str = re.compile(pat_str)
+
+        mergeable_ranks = {}
+        with open(vocab_file) as f:
+            for line in f:
+                token, rank = line.strip().split()
+                rank = int(rank)
+                token = base64.b64decode(token)
+                mergeable_ranks[token] = rank
+
+        self.mergeable_ranks = mergeable_ranks
+
+        self.tokenizer = tiktoken.Encoding(
+            name="my_tokenizer",
+            pat_str=pat_str,
+            mergeable_ranks=mergeable_ranks,
+            special_tokens={}
+        )
+        self.decoder = {rank: token for token, rank in mergeable_ranks.items()}
+        self.n_words = len(self.decoder)
+
+        super().__init__(
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs
+        )
+
+    @property
+    def vocab_size(self):
+        return self.n_words
+
+    def get_vocab(self):
+        """ Returns vocab as a dict """
+        vocab = {self._convert_id_to_token(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def convert_tokens_to_string(self, tokens: List[Union[bytes, str, int]]) -> str:
+        """
+        Converts a sequence of tokens in a single string.
+        """
+        text = ""
+        temp = b""
+        for t in tokens:
+            if isinstance(t, int):
+                t = chr(t)
+            if isinstance(t, str):
+                if temp:
+                    text += temp.decode("utf-8", errors="replace")
+            elif isinstance(t, bytes):
+                temp += t
+            else:
+                raise TypeError("token should only be of type int, bytes or str")
+        if temp:
+            text += temp.decode("utf-8", errors="replace")
+        return text
+
+    def _tokenize(self, text, **kwargs):
+        tokens = []
+        ids = self.tokenizer.encode(text)
+        for t in ids:
+            tokens.append(self.decoder[t])
+        return tokens
+
+    def _convert_token_to_id(self, token):
+        """ Converts a token (str) in an id using the vocab. """
+        return self.mergeable_ranks[token]
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        return self.decoder.get(index, "")
+
+    def save_vocabulary(self, save_directory, filename_prefix=None):
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+            filename_prefix (`str`, *optional*):
+                An optional prefix to add to the named of the saved files.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if os.path.isdir(save_directory):
+            vocab_file = os.path.join(
+                save_directory, self.vocab_files_names["vocab_file"]
+            )
+        else:
+            vocab_file = save_directory
+
+        with open(self.vocab_file, 'rb') as fin:
+            proto_str = fin.read()
+
+        with open(vocab_file, "wb") as writer:
+            writer.write(proto_str)
+
+        return (vocab_file,)
+
+    def get_prefix_tokens(self):
+        prefix_tokens = [self.convert_tokens_to_ids("[gMASK]"), self.convert_tokens_to_ids("<sop>")]
+        return prefix_tokens
+
+    def build_single_message(self, role, metadata, message, tokenize=True):
+        assert role in ["system", "user", "assistant", "observation"], role
+        if tokenize:
+            role_tokens = [self.convert_tokens_to_ids(f"<|{role}|>")] + self.tokenizer.encode(f"{metadata}\n",
+                                                                                              disallowed_special=())
+            message_tokens = self.tokenizer.encode(message, disallowed_special=())
+            tokens = role_tokens + message_tokens
+            return tokens
+        else:
+            return str(f"<|{role}|>{metadata}\n{message}")
+
+    def build_inputs_with_special_tokens(
+            self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Build model inputs from a sequence or a pair of sequence for sequence classification tasks by concatenating and
+        adding special tokens. A BERT sequence has the following format:
+
+        - single sequence: `[CLS] X [SEP]`
+        - pair of sequences: `[CLS] A [SEP] B [SEP]`
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs to which the special tokens will be added.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [input IDs](../glossary#input-ids) with the appropriate special tokens.
+        """
+        prefix_tokens = self.get_prefix_tokens()
+        token_ids_0 = prefix_tokens + token_ids_0
+        if token_ids_1 is not None:
+            token_ids_0 = token_ids_0 + token_ids_1 + [self.convert_tokens_to_ids("<eos>")]
+        return token_ids_0
+
+    def _pad(
+            self,
+            encoded_inputs: Union[Dict[str, EncodedInput], BatchEncoding],
+            max_length: Optional[int] = None,
+            padding_side: str = "left",
+            padding_strategy: PaddingStrategy = PaddingStrategy.DO_NOT_PAD,
+            pad_to_multiple_of: Optional[int] = None,
+            return_attention_mask: Optional[bool] = None,
+    ) -> dict:
+        """
+        Pad encoded inputs (on left/right and up to predefined length or max length in the batch)
+
+        Args:
+            encoded_inputs:
+                Dictionary of tokenized inputs (`List[int]`) or batch of tokenized inputs (`List[List[int]]`).
+            max_length: maximum length of the returned list and optionally padding length (see below).
+                Will truncate by taking into account the special tokens.
+            padding_strategy: PaddingStrategy to use for padding.
+
+                - PaddingStrategy.LONGEST Pad to the longest sequence in the batch
+                - PaddingStrategy.MAX_LENGTH: Pad to the max length (default)
+                - PaddingStrategy.DO_NOT_PAD: Do not pad
+                The tokenizer padding sides are defined in self.padding_side:
+
+                    - 'left': pads on the left of the sequences
+                    - 'right': pads on the right of the sequences
+            pad_to_multiple_of: (optional) Integer if set will pad the sequence to a multiple of the provided value.
+                This is especially useful to enable the use of Tensor Core on NVIDIA hardware with compute capability
+                `>= 7.5` (Volta).
+            return_attention_mask:
+                (optional) Set to False to avoid returning attention mask (default: set to model specifics)
+        """
+        # Load from model defaults
+
+        required_input = encoded_inputs[self.model_input_names[0]]
+        seq_length = len(required_input)
+
+        if padding_strategy == PaddingStrategy.LONGEST:
+            max_length = len(required_input)
+
+        if max_length is not None and pad_to_multiple_of is not None and (max_length % pad_to_multiple_of != 0):
+            max_length = ((max_length // pad_to_multiple_of) + 1) * pad_to_multiple_of
+
+        needs_to_be_padded = padding_strategy != PaddingStrategy.DO_NOT_PAD and len(required_input) != max_length
+
+        # Initialize attention mask if not present.
+        if "attention_mask" not in encoded_inputs:
+            encoded_inputs["attention_mask"] = [1] * seq_length
+
+        if "position_ids" not in encoded_inputs:
+            encoded_inputs["position_ids"] = list(range(seq_length))
+
+        if needs_to_be_padded:
+            difference = max_length - len(required_input)
+
+            if "attention_mask" in encoded_inputs:
+                encoded_inputs["attention_mask"] = [0] * difference + encoded_inputs["attention_mask"]
+            if "position_ids" in encoded_inputs:
+                encoded_inputs["position_ids"] = [0] * difference + encoded_inputs["position_ids"]
+            encoded_inputs[self.model_input_names[0]] = [self.pad_token_id] * difference + required_input
+
+        return encoded_inputs

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a493598071550244b2ee7f26118f3edec2150b9dfa967929a99052ac83fe716
+size 2623634

tokenizer_config.json

@@ -0,0 +1,147 @@
+{
+  "added_tokens_decoder": {
+    "151329": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151330": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151331": {
+      "content": "[gMASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151332": {
+      "content": "[sMASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151333": {
+      "content": "<sop>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151334": {
+      "content": "<eop>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151335": {
+      "content": "<|system|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151336": {
+      "content": "<|user|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151337": {
+      "content": "<|assistant|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151338": {
+      "content": "<|observation|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151339": {
+      "content": "<|begin_of_image|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151340": {
+      "content": "<|end_of_image|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151341": {
+      "content": "<|begin_of_video|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151342": {
+      "content": "<|end_of_video|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>"
+  ],
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_chatglm.ChatGLM4Tokenizer",
+      null
+    ]
+  },
+  "chat_template": "[gMASK]<sop>{% for item in messages %}{% if item['tools'] is defined %}<|system|>\n你是一个名为 GLM-4 的人工智能助手。你是基于智谱AI训练的语言模型 GLM-4 模型开发的，你的任务是针对用户的问题和要求提供适当的答复和支持。\n\n# 可用工具{% set tools = item['tools'] %}{% for tool in tools %}{% if tool['type'] == 'function' %}\n\n## {{ tool['function']['name'] }}\n\n{{ tool['function'] | tojson(indent=4) }}\n在调用上述函数时，请使用 Json 格式表示调用的参数。{% elif tool['type'] == 'python' %}\n\n## python\n\n当你向 `python` 发送包含 Python 代码的消息时，该代码将会在一个有状态的 Jupyter notebook 环境中执行。\n`python` 返回代码执行的输出，或在执行 60 秒后返回超时。\n`/mnt/data` 将会持久化存储你的文件。在此会话中，`python` 无法访问互联网。不要使用 `python` 进行任何网络请求或者在线 API 调用，这些在线内容的访问将不会成功。{% elif tool['type'] == 'simple_browser' %}\n\n## simple_browser\n\n你可以使用 `simple_browser` 工具。该工具支持以下函数：\n`search(query: str, recency_days: int)`：使用搜索引擎进行查询并显示结果，可以使用 `recency_days` 参数控制搜索内容的时效性。\n`mclick(ids: list[int])`：获取一系列指定 id 的页面内容。每次调用时，须选择3-10个页面。选择多个角度的页面，同时尽可能选择可信任的信息来源。考虑到部分页面是无法加载的，你也可以多打开一些可能有用的页面而不用担心内容过多。\n`open_url(url: str)`：打开指定的 URL。\n\n使用 `【{引用 id}†{引用文本}】` 来引用内容。\n\n操作步骤：1. 使用 `search` 来获得信息列表; 2. 使用 `mclick` 来获取指定 ID 页面的内容; 3. 根据获得的内容进行回复。在回复中应当引用信息来源。\n 如果用户提供了 URL，也可以用 `open_url` 直接���开页面。\n如果初次搜索结果没有找到合适的信息，也可以再次使用 `search` 进行搜索。{% elif tool['type'] == 'cogview' %}\n\n## cogview\n\n如果用户的请求中包含了对图像的描述，你可以使用 `cogview` 来生成图像并展示给用户。你需要向 `cogview` 发送图像描述，规则：\n- 发送给 `cogview` 的消息必须使用英语。用户的中文描述必须完全翻译为英语。\n- 应当尽可能详细地描述图像生成的需求，需求描述约 100 英文单词。\n- 保持用户原始描述的意图。不要虚构内容或者没见过的人物。\n- 如无特殊说明，所在地为中国，持有中国立场并遵循中国社会主义价值观。{% endif %}{% endfor %}{% endif %}{% if item['content'] %}<|{{ item['role'] }}|>{{ item['metadata'] }}\n{{ item['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "model_max_length": 128000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "ChatGLM4Tokenizer"
+}