diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..5867afbf86f4f0d469f4e5de279192fcfd6c8d1a 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -1,35 +1,5 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
+data/babylm_data/* filter=lfs diff=lfs merge=lfs -text
+data/Perturbed_data/* filter=lfs diff=lfs merge=lfs -text
 *.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..f4f2bcf8e8c15a1bb1c56ee1f535a66091e38c0f
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1 @@
+train/checkpoints/
diff --git a/FTP.py b/FTP.py
new file mode 100644
index 0000000000000000000000000000000000000000..8acb9e14d87f71181933816528794ccb85a10777
--- /dev/null
+++ b/FTP.py
@@ -0,0 +1,316 @@
+import torch
+from torch.optim.optimizer import Optimizer, required
+import copy
+import math
+
+class FTP(object):
+    def __init__(self, k=1.0, exclude_set={}):
+        self.exclude_set = exclude_set
+        self.threshold = torch.nn.Hardtanh(0,1)
+        self.k = k # Gradient annealing factor
+        self.j = 0 # Buffer counter
+
+        # AdamUtil parameteres
+        self.mu = 1e-2
+        self.beta1 = 0.9
+        self.beta2 = 0.999
+        self.t = 1
+        
+        # Buffers
+        self.gamma_buff = []
+        self.first_m_gamma = []
+        self.second_m_gamma = []
+        self.prev_c = []
+        self.prev_scale = []
+    
+    @torch.no_grad()
+    def step(self,name, curr, pre, d_p):
+        if curr.requires_grad and name not in self.exclude_set:  # exclude_set includes those params that not be updated
+            c_t = (curr - d_p) - pre   # (cur - d_p) potential new param/ dp gradient /
+            norms = self._mars_norm(c_t)
+
+            if self.t == 1:
+                gamma = torch.tensor(1e-8).to(norms.device)
+                self._update_buffers(gamma)
+            else:
+                # Get previous values
+                gamma_prev = self.gamma_buff[self.j]
+                c_prev = self.prev_c[self.j]
+                scale_prev = self.prev_scale[self.j]
+
+                # Calculate gradient for gamma
+                gamma_grad = torch.sum(self._dot(curr.grad, c_prev, scale = scale_prev))
+
+                # Anneal positive gradient 
+                if gamma_grad > 0:
+                    gamma_grad = gamma_grad * self.k
+
+                gamma = self._adam_util(gamma_prev, gamma_grad)
+
+                # Clip gamma
+                gamma = self._clip(gamma, norms)
+            
+            # Update
+            denom = 1/norms
+            ratio = gamma * denom
+            new_p = pre + self.threshold(ratio) * c_t 
+
+            # Save updated values
+            self._update_buffers(gamma, c_t, denom)
+            self.j += 1
+            return new_p
+        else:
+            return None
+        
+    def incre_counters(self):
+        self.t += 1
+        self.j = 0
+
+    @torch.no_grad()
+    def _mars_norm(self,tensor):
+        return torch.sum(torch.abs(tensor), dim=tuple(range(1,tensor.dim())), keepdim=True) + 1e-8
+    
+    @torch.no_grad()
+    def _clip(self, constraint, norms):
+        return torch.nn.functional.hardtanh(constraint,1e-8,norms.max())
+        
+    @torch.no_grad()
+    def _dot(self,tensor1,tensor2,scale=1):
+        return torch.sum(torch.mul(tensor1, tensor2),dim=tuple(range(1,tensor1.dim())),keepdim=True)*scale
+    
+    @torch.no_grad()
+    def _adam_util(self,prev, grad):
+        first_moment = self.beta1 * self.first_m_gamma[self.j] + (1-self.beta1) * grad
+        second_moment = self.beta2 * self.second_m_gamma[self.j] + (1-self.beta2) * grad**2
+        self.first_m_gamma[self.j] = first_moment
+        self.second_m_gamma[self.j] = second_moment
+        first_moment = first_moment/(1-self.beta1**self.t)
+        second_moment = second_moment/(1-self.beta2**self.t)
+        return prev  - self.mu * first_moment/(torch.sqrt(second_moment)+1e-8)
+    
+    def _update_buffers(self,gamma,c_t=None,denom=None):
+        if c_t is None:
+            self.first_m_gamma.append(0.0)
+            self.second_m_gamma.append(0.0)
+            self.gamma_buff.append(gamma)
+            self.prev_c.append(0.0)
+            self.prev_scale.append(0.0)
+        else:
+            self.gamma_buff[self.j] = gamma
+            self.prev_c[self.j] = c_t
+            self.prev_scale[self.j] = denom
+
+            
+class SGDP(Optimizer):
+    def __init__(self, params, lr=required, momentum=0, dampening=0,
+                 weight_decay=0, nesterov=False, k=1.0, exclude_set = {}):
+        if lr is not required and lr < 0.0:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if momentum < 0.0:
+            raise ValueError("Invalid momentum value: {}".format(momentum))
+        if weight_decay < 0.0:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+
+        defaults = dict(lr=lr, momentum=momentum, dampening=dampening,
+                        weight_decay=weight_decay, nesterov=nesterov)
+        if nesterov and (momentum <= 0 or dampening != 0):
+            raise ValueError("Nesterov momentum requires a momentum and zero dampening")
+        super(SGDP, self).__init__(params, defaults)
+        self.first_iter_flag = False
+         
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+                    
+
+    def __setstate__(self, state):
+        super(SGDP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('nesterov', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+        for group in self.param_groups:
+            weight_decay = group['weight_decay']
+            momentum = group['momentum']
+            dampening = group['dampening']
+            nesterov = group['nesterov']
+            
+            for p, name, pre in zip(group['params'],group['name'],group['pre']):
+                if p.grad is None:
+                    continue
+                d_p = p.grad
+                if weight_decay != 0:
+                    d_p = d_p.add(p, alpha=weight_decay)
+                if momentum != 0:
+                    param_state = self.state[p]
+                    if 'momentum_buffer' not in param_state:
+                        buf = param_state['momentum_buffer'] = torch.clone(d_p).detach()
+                    else:
+                        buf = param_state['momentum_buffer']
+                        buf.mul_(momentum).add_(d_p, alpha=1 - dampening) 
+                    if nesterov:
+                        d_p = d_p.add(buf, alpha=momentum)
+                    else:
+                        d_p = buf
+
+                # FTP step
+                d_p = group['lr']*d_p
+                new_p = self.ftp.step(name,p,pre,d_p)
+                if new_p is not None :
+                    p.copy_(new_p)
+                else:
+                    p.add_(d_p, alpha=-1)
+        # FTP increment internal counters
+        self.ftp.incre_counters()        
+        return loss
+
+
+class AdamP(Optimizer):
+    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
+                 weight_decay=0, amsgrad=False, k=1.0, exclude_set={}):
+        if not 0.0 <= lr:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if not 0.0 <= eps:
+            raise ValueError("Invalid epsilon value: {}".format(eps))
+        if not 0.0 <= betas[0] < 1.0:
+            raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
+        if not 0.0 <= betas[1] < 1.0:
+            raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
+        if not 0.0 <= weight_decay:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+        defaults = dict(lr=lr, betas=betas, eps=eps,
+                        weight_decay=weight_decay, amsgrad=amsgrad)
+        super(AdamP, self).__init__(params, defaults)
+
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+
+    def __setstate__(self, state):
+        super(AdamP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('amsgrad', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+
+        for group in self.param_groups:
+            params_with_grad = []
+            grads = []
+            exp_avgs = []
+            exp_avg_sqs = []
+            max_exp_avg_sqs = []
+            state_steps = []
+
+            for p in group['params']:
+                if p.grad is not None:
+                    params_with_grad.append(p)
+                    if p.grad.is_sparse:
+                        raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')
+                    grads.append(p.grad)
+
+                    state = self.state[p]
+                    # Lazy state initialization
+                    if len(state) == 0:
+                        state['step'] = 0
+                        # Exponential moving average of gradient values
+                        state['exp_avg'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        # Exponential moving average of squared gradient values
+                        state['exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        if group['amsgrad']:
+                            # Maintains max of all exp. moving avg. of sq. grad. values
+                            state['max_exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+
+                    exp_avgs.append(state['exp_avg'])
+                    exp_avg_sqs.append(state['exp_avg_sq'])
+
+                    if group['amsgrad']:
+                        max_exp_avg_sqs.append(state['max_exp_avg_sq'])
+
+                    # update the steps for each param group update
+                    state['step'] += 1
+                    # record the step after step update
+                    state_steps.append(state['step'])
+
+            beta1, beta2 = group['betas']
+            self.adam(group,
+                   exp_avgs,
+                   exp_avg_sqs,
+                   max_exp_avg_sqs,
+                   state_steps,
+                   group['amsgrad'],
+                   beta1,
+                   beta2,
+                   group['lr'],
+                   group['weight_decay'],
+                   group['eps']
+                   )
+        # FTP increment internal counters            
+        self.ftp.incre_counters()    
+        return loss
+
+    def adam(self, group,
+         exp_avgs,
+         exp_avg_sqs,
+         max_exp_avg_sqs,
+         state_steps, 
+         amsgrad: bool,
+         beta1: float,
+         beta2: float,
+         lr: float,
+         weight_decay: float,
+         eps: float):
+        
+        i = 0
+        for param, name, pre in zip(group['params'],group['name'],group['pre']):
+            if param.grad is None: 
+                continue
+            grad = param.grad
+            exp_avg = exp_avgs[i]
+            exp_avg_sq = exp_avg_sqs[i]
+            step = state_steps[i]
+            if amsgrad:
+                max_exp_avg_sq = max_exp_avg_sqs[i]
+
+            bias_correction1 = 1 - beta1 ** step
+            bias_correction2 = 1 - beta2 ** step
+
+            # Decay the first and second moment running average coefficient
+            exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
+            exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
+            if amsgrad:
+                # Maintains the maximum of all 2nd moment running avg. till now
+                torch.maximum(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
+                # Use the max. for normalizing running avg. of gradient
+                denom = (max_exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+            else:
+                denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+
+            step_size = lr / bias_correction1
+            i += 1
+            
+            # FTP step
+            d_p = step_size * exp_avg/denom + lr * weight_decay * param
+            new_p = self.ftp.step(name,param,pre,d_p)
+            if new_p is None :
+                new_p = param - d_p
+            param.copy_(new_p)
\ No newline at end of file
diff --git a/FTP_update.py b/FTP_update.py
new file mode 100644
index 0000000000000000000000000000000000000000..5fac0d3b09c0303abc7f8b5b7bb8ed18e9620b34
--- /dev/null
+++ b/FTP_update.py
@@ -0,0 +1,323 @@
+import torch
+from torch.optim.optimizer import Optimizer, required
+import copy
+import math
+
+class FTP(object):
+    def __init__(self, k=1.0, exclude_set={}):
+        self.exclude_set = exclude_set
+        self.threshold = torch.nn.Hardtanh(0,1)
+        self.k = k  # Gradient annealing factor
+        self.j = 0  # Buffer counter
+
+        # AdamUtil parameters
+        self.mu = 1e-2
+        self.beta1 = 0.9
+        self.beta2 = 0.999
+        self.t = 1
+        
+        # Buffers
+        self.gamma_buff = []
+        self.first_m_gamma = []
+        self.second_m_gamma = []
+        self.prev_c = []
+        self.prev_scale = []
+    
+    @torch.no_grad()
+    def step(self, name, curr, pre, d_p):
+        if curr.requires_grad and name not in self.exclude_set:  # Exclude set includes those params that not be updated
+            c_t = (curr - d_p) - pre  # Compute potential new param
+            norms = self._mars_norm(c_t)
+
+            # New: Apply spectral normalization to c_t
+            c_t = self._apply_spectral_norm(c_t)
+
+            if self.t == 1:
+                gamma = torch.tensor(1e-8, device=norms.device)
+                self._update_buffers(gamma)
+            else:
+                # Get previous values
+                gamma_prev = self.gamma_buff[self.j]
+                c_prev = self.prev_c[self.j]
+                scale_prev = self.prev_scale[self.j]
+
+                # Calculate gradient for gamma
+                gamma_grad = torch.sum(self._dot(curr.grad, c_prev, scale=scale_prev))
+
+                # Anneal positive gradient 
+                if gamma_grad > 0:
+                    gamma_grad = gamma_grad * self.k
+
+                gamma = self._adam_util(gamma_prev, gamma_grad)
+
+                # Clip gamma
+                gamma = self._clip(gamma, norms)
+            
+            # Update
+            denom = 1/norms
+            ratio = gamma * denom
+            new_p = pre + self.threshold(ratio) * c_t
+
+            # Save updated values
+            self._update_buffers(gamma, c_t, denom)
+            self.j += 1
+            return new_p
+        else:
+            return None
+
+    def _apply_spectral_norm(self, c_t):
+        u, s, vh = torch.linalg.svd(c_t, full_matrices=False)
+        spectral_norm = s.max()
+        return torch.matmul(u, vh) / spectral_norm
+
+    def incre_counters(self):
+        self.t += 1
+        self.j = 0
+
+    @torch.no_grad()
+    def _mars_norm(self, tensor):
+        return torch.sum(torch.abs(tensor), dim=tuple(range(1, tensor.dim())), keepdim=True) + 1e-8
+    
+    @torch.no_grad()
+    def _clip(self, constraint, norms):
+        return torch.nn.functional.hardtanh(constraint, 1e-8, norms.max())
+        
+    @torch.no_grad()
+    def _dot(self, tensor1, tensor2, scale=1):
+        return torch.sum(torch.mul(tensor1, tensor2), dim=tuple(range(1, tensor1.dim())), keepdim=True) * scale
+    
+    @torch.no_grad()
+    def _adam_util(self, prev, grad):
+        first_moment = self.beta1 * self.first_m_gamma[self.j] + (1 - self.beta1) * grad
+        second_moment = self.beta2 * self.second_m_gamma[self.j] + (1 - self.beta2) * grad**2
+        self.first_m_gamma[self.j] = first_moment
+        self.second_m_gamma[self.j] = second_moment
+        first_moment = first_moment / (1 - self.beta1**self.t)
+        second_moment = second_moment / (1 - self.beta2**self.t)
+        return prev - self.mu * first_moment / (torch.sqrt(second_moment) + 1e-8)
+    
+    def _update_buffers(self, gamma, c_t=None, denom=None):
+        if c_t is None:
+            self.first_m_gamma.append(0.0)
+            self.second_m_gamma.append(0.0)
+            self.gamma_buff.append(gamma)
+            self.prev_c.append(0.0)
+            self.prev_scale.append(0.0)
+        else:
+            self.gamma_buff[self.j] = gamma
+            self.prev_c[self.j] = c_t
+            self.prev_scale[self.j] = denom
+
+class SGDP(Optimizer):
+    def __init__(self, params, lr=required, momentum=0, dampening=0,
+                 weight_decay=0, nesterov=False, k=1.0, exclude_set = {}):
+        if lr is not required and lr < 0.0:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if momentum < 0.0:
+            raise ValueError("Invalid momentum value: {}".format(momentum))
+        if weight_decay < 0.0:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+
+        defaults = dict(lr=lr, momentum=momentum, dampening=dampening,
+                        weight_decay=weight_decay, nesterov=nesterov)
+        if nesterov and (momentum <= 0 or dampening != 0):
+            raise ValueError("Nesterov momentum requires a momentum and zero dampening")
+        super(SGDP, self).__init__(params, defaults)
+        self.first_iter_flag = False
+         
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+                    
+
+    def __setstate__(self, state):
+        super(SGDP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('nesterov', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+        for group in self.param_groups:
+            weight_decay = group['weight_decay']
+            momentum = group['momentum']
+            dampening = group['dampening']
+            nesterov = group['nesterov']
+            
+            for p, name, pre in zip(group['params'],group['name'],group['pre']):
+                if p.grad is None:
+                    continue
+                d_p = p.grad
+                if weight_decay != 0:
+                    d_p = d_p.add(p, alpha=weight_decay)
+                if momentum != 0:
+                    param_state = self.state[p]
+                    if 'momentum_buffer' not in param_state:
+                        buf = param_state['momentum_buffer'] = torch.clone(d_p).detach()
+                    else:
+                        buf = param_state['momentum_buffer']
+                        buf.mul_(momentum).add_(d_p, alpha=1 - dampening) 
+                    if nesterov:
+                        d_p = d_p.add(buf, alpha=momentum)
+                    else:
+                        d_p = buf
+
+                # FTP step
+                d_p = group['lr']*d_p
+                new_p = self.ftp.step(name,p,pre,d_p)
+                if new_p is not None :
+                    p.copy_(new_p)
+                else:
+                    p.add_(d_p, alpha=-1)
+        # FTP increment internal counters
+        self.ftp.incre_counters()        
+        return loss
+
+
+class AdamP(Optimizer):
+    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
+                 weight_decay=0, amsgrad=False, k=1.0, exclude_set={}):
+        if not 0.0 <= lr:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if not 0.0 <= eps:
+            raise ValueError("Invalid epsilon value: {}".format(eps))
+        if not 0.0 <= betas[0] < 1.0:
+            raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
+        if not 0.0 <= betas[1] < 1.0:
+            raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
+        if not 0.0 <= weight_decay:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+        defaults = dict(lr=lr, betas=betas, eps=eps,
+                        weight_decay=weight_decay, amsgrad=amsgrad)
+        super(AdamP, self).__init__(params, defaults)
+
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+
+    def __setstate__(self, state):
+        super(AdamP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('amsgrad', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+
+        for group in self.param_groups:
+            params_with_grad = []
+            grads = []
+            exp_avgs = []
+            exp_avg_sqs = []
+            max_exp_avg_sqs = []
+            state_steps = []
+
+            for p in group['params']:
+                if p.grad is not None:
+                    params_with_grad.append(p)
+                    if p.grad.is_sparse:
+                        raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')
+                    grads.append(p.grad)
+
+                    state = self.state[p]
+                    # Lazy state initialization
+                    if len(state) == 0:
+                        state['step'] = 0
+                        # Exponential moving average of gradient values
+                        state['exp_avg'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        # Exponential moving average of squared gradient values
+                        state['exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        if group['amsgrad']:
+                            # Maintains max of all exp. moving avg. of sq. grad. values
+                            state['max_exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+
+                    exp_avgs.append(state['exp_avg'])
+                    exp_avg_sqs.append(state['exp_avg_sq'])
+
+                    if group['amsgrad']:
+                        max_exp_avg_sqs.append(state['max_exp_avg_sq'])
+
+                    # update the steps for each param group update
+                    state['step'] += 1
+                    # record the step after step update
+                    state_steps.append(state['step'])
+
+            beta1, beta2 = group['betas']
+            self.adam(group,
+                   exp_avgs,
+                   exp_avg_sqs,
+                   max_exp_avg_sqs,
+                   state_steps,
+                   group['amsgrad'],
+                   beta1,
+                   beta2,
+                   group['lr'],
+                   group['weight_decay'],
+                   group['eps']
+                   )
+        # FTP increment internal counters            
+        self.ftp.incre_counters()    
+        return loss
+
+    def adam(self, group,
+         exp_avgs,
+         exp_avg_sqs,
+         max_exp_avg_sqs,
+         state_steps, 
+         amsgrad: bool,
+         beta1: float,
+         beta2: float,
+         lr: float,
+         weight_decay: float,
+         eps: float):
+        
+        i = 0
+        for param, name, pre in zip(group['params'],group['name'],group['pre']):
+            if param.grad is None: 
+                continue
+            grad = param.grad
+            exp_avg = exp_avgs[i]
+            exp_avg_sq = exp_avg_sqs[i]
+            step = state_steps[i]
+            if amsgrad:
+                max_exp_avg_sq = max_exp_avg_sqs[i]
+
+            bias_correction1 = 1 - beta1 ** step
+            bias_correction2 = 1 - beta2 ** step
+
+            # Decay the first and second moment running average coefficient
+            exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
+            exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
+            if amsgrad:
+                # Maintains the maximum of all 2nd moment running avg. till now
+                torch.maximum(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
+                # Use the max. for normalizing running avg. of gradient
+                denom = (max_exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+            else:
+                denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+
+            step_size = lr / bias_correction1
+            i += 1
+            
+            # FTP step
+            d_p = step_size * exp_avg/denom + lr * weight_decay * param
+            new_p = self.ftp.step(name,param,pre,d_p)
+            if new_p is None :
+                new_p = param - d_p
+            param.copy_(new_p)
\ No newline at end of file
diff --git a/FTP_update2.py b/FTP_update2.py
new file mode 100644
index 0000000000000000000000000000000000000000..b141e0c8391c8578df7dff19ab4363e9dfb65eef
--- /dev/null
+++ b/FTP_update2.py
@@ -0,0 +1,340 @@
+import torch
+from torch.optim.optimizer import Optimizer, required
+import copy
+import math
+
+class FTP(object):
+    def __init__(self, k=1.0, exclude_set={}):
+        self.exclude_set = exclude_set
+        self.threshold = torch.nn.Hardtanh(0,1)
+        self.k = k  # Gradient annealing factor
+        self.j = 0  # Buffer counter
+
+        # AdamUtil parameters
+        self.mu = 1e-2
+        self.beta1 = 0.9
+        self.beta2 = 0.999
+        self.t = 1
+        
+        # Buffers
+        self.gamma_buff = []
+        self.first_m_gamma = []
+        self.second_m_gamma = []
+        self.prev_c = []
+        self.prev_scale = []
+    
+    @torch.no_grad()
+    def step(self, name, curr, pre, d_p):
+        if curr.requires_grad and name not in self.exclude_set:  # Exclude set includes those params that not be updated
+            c_t = (curr - d_p) - pre  # Compute potential new param
+            norms = self._mars_norm(c_t)
+
+            # New: Apply spectral normalization to c_t
+            c_t = self._apply_spectral_norm(c_t)
+
+            if self.t == 1:
+                gamma = torch.tensor(1e-8, device=norms.device)
+                self._update_buffers(gamma)
+            else:
+                # Get previous values
+                gamma_prev = self.gamma_buff[self.j]
+                c_prev = self.prev_c[self.j]
+                scale_prev = self.prev_scale[self.j]
+
+                # Calculate gradient for gamma
+                gamma_grad = torch.sum(self._dot(curr.grad, c_prev, scale=scale_prev))
+
+                # Anneal positive gradient 
+                if gamma_grad > 0:
+                    gamma_grad = gamma_grad * self.k
+
+                gamma = self._adam_util(gamma_prev, gamma_grad)
+
+                # Clip gamma
+                gamma = self._clip(gamma, norms)
+            
+            # Update
+            denom = 1/norms
+            ratio = gamma * denom
+            new_p = pre + self.threshold(ratio) * c_t
+
+            # Save updated values
+            self._update_buffers(gamma, c_t, denom)
+            self.j += 1
+            return new_p
+        else:
+            return None
+
+    def _apply_spectral_norm(self, param, tau=1.0):
+        
+        if param.ndim >= 2:  # Only for 2D matrices
+            u, s, vh = torch.linalg.svd(param, full_matrices=False)
+            s_clipped = torch.clamp(s, max=tau)  # Clip singular values
+            return torch.matmul(u, torch.matmul(torch.diag(s_clipped), vh))
+        return param  # No projection for 1D parameters
+    
+    # def _apply_spectral_norm(self, param, tau=1.0):
+    #     if param.ndim >= 2:  # For 2D or higher-dimensional tensors
+    #         original_shape = param.shape  # Save original shape
+    #         reshaped_param = param.view(original_shape[0], -1)  # Flatten into 2D
+
+    #         u, s, vh = torch.linalg.svd(reshaped_param, full_matrices=False)
+    #         s_clipped = torch.clamp(s, max=tau)  # Clip singular values
+
+    #         # Reconstruct the tensor with clipped singular values
+    #         normalized_param = torch.matmul(u, torch.matmul(torch.diag(s_clipped), vh))
+    #         return normalized_param.view(original_shape)  # Reshape back to original
+    #     return param  # No projection for 1D parameters
+
+
+    def incre_counters(self):
+        self.t += 1
+        self.j = 0
+
+    @torch.no_grad()
+    def _mars_norm(self, tensor):
+        return torch.sum(torch.abs(tensor), dim=tuple(range(1, tensor.dim())), keepdim=True) + 1e-8
+    
+    @torch.no_grad()
+    def _clip(self, constraint, norms):
+        return torch.nn.functional.hardtanh(constraint, 1e-8, norms.max())
+        
+    @torch.no_grad()
+    def _dot(self, tensor1, tensor2, scale=1):
+        return torch.sum(torch.mul(tensor1, tensor2), dim=tuple(range(1, tensor1.dim())), keepdim=True) * scale
+    
+    @torch.no_grad()
+    def _adam_util(self, prev, grad):
+        first_moment = self.beta1 * self.first_m_gamma[self.j] + (1 - self.beta1) * grad
+        second_moment = self.beta2 * self.second_m_gamma[self.j] + (1 - self.beta2) * grad**2
+        self.first_m_gamma[self.j] = first_moment
+        self.second_m_gamma[self.j] = second_moment
+        first_moment = first_moment / (1 - self.beta1**self.t)
+        second_moment = second_moment / (1 - self.beta2**self.t)
+        return prev - self.mu * first_moment / (torch.sqrt(second_moment) + 1e-8)
+    
+    def _update_buffers(self, gamma, c_t=None, denom=None):
+        if c_t is None:
+            self.first_m_gamma.append(0.0)
+            self.second_m_gamma.append(0.0)
+            self.gamma_buff.append(gamma)
+            self.prev_c.append(0.0)
+            self.prev_scale.append(0.0)
+        else:
+            self.gamma_buff[self.j] = gamma
+            self.prev_c[self.j] = c_t
+            self.prev_scale[self.j] = denom
+
+class SGDP(Optimizer):
+    def __init__(self, params, lr=required, momentum=0, dampening=0,
+                 weight_decay=0, nesterov=False, k=1.0, exclude_set = {}):
+        if lr is not required and lr < 0.0:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if momentum < 0.0:
+            raise ValueError("Invalid momentum value: {}".format(momentum))
+        if weight_decay < 0.0:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+
+        defaults = dict(lr=lr, momentum=momentum, dampening=dampening,
+                        weight_decay=weight_decay, nesterov=nesterov)
+        if nesterov and (momentum <= 0 or dampening != 0):
+            raise ValueError("Nesterov momentum requires a momentum and zero dampening")
+        super(SGDP, self).__init__(params, defaults)
+        self.first_iter_flag = False
+         
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+                    
+
+    def __setstate__(self, state):
+        super(SGDP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('nesterov', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+        for group in self.param_groups:
+            weight_decay = group['weight_decay']
+            momentum = group['momentum']
+            dampening = group['dampening']
+            nesterov = group['nesterov']
+            
+            for p, name, pre in zip(group['params'],group['name'],group['pre']):
+                if p.grad is None:
+                    continue
+                d_p = p.grad
+                if weight_decay != 0:
+                    d_p = d_p.add(p, alpha=weight_decay)
+                if momentum != 0:
+                    param_state = self.state[p]
+                    if 'momentum_buffer' not in param_state:
+                        buf = param_state['momentum_buffer'] = torch.clone(d_p).detach()
+                    else:
+                        buf = param_state['momentum_buffer']
+                        buf.mul_(momentum).add_(d_p, alpha=1 - dampening) 
+                    if nesterov:
+                        d_p = d_p.add(buf, alpha=momentum)
+                    else:
+                        d_p = buf
+
+                # FTP step
+                d_p = group['lr']*d_p
+                new_p = self.ftp.step(name,p,pre,d_p)
+                if new_p is not None :
+                    p.copy_(new_p)
+                else:
+                    p.add_(d_p, alpha=-1)
+        # FTP increment internal counters
+        self.ftp.incre_counters()        
+        return loss
+
+
+class AdamP(Optimizer):
+    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
+                 weight_decay=0, amsgrad=False, k=1.0, exclude_set={}):
+        if not 0.0 <= lr:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if not 0.0 <= eps:
+            raise ValueError("Invalid epsilon value: {}".format(eps))
+        if not 0.0 <= betas[0] < 1.0:
+            raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
+        if not 0.0 <= betas[1] < 1.0:
+            raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
+        if not 0.0 <= weight_decay:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+        defaults = dict(lr=lr, betas=betas, eps=eps,
+                        weight_decay=weight_decay, amsgrad=amsgrad)
+        super(AdamP, self).__init__(params, defaults)
+
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+
+    def __setstate__(self, state):
+        super(AdamP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('amsgrad', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+
+        for group in self.param_groups:
+            params_with_grad = []
+            grads = []
+            exp_avgs = []
+            exp_avg_sqs = []
+            max_exp_avg_sqs = []
+            state_steps = []
+
+            for p in group['params']:
+                if p.grad is not None:
+                    params_with_grad.append(p)
+                    if p.grad.is_sparse:
+                        raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')
+                    grads.append(p.grad)
+
+                    state = self.state[p]
+                    # Lazy state initialization
+                    if len(state) == 0:
+                        state['step'] = 0
+                        # Exponential moving average of gradient values
+                        state['exp_avg'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        # Exponential moving average of squared gradient values
+                        state['exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        if group['amsgrad']:
+                            # Maintains max of all exp. moving avg. of sq. grad. values
+                            state['max_exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+
+                    exp_avgs.append(state['exp_avg'])
+                    exp_avg_sqs.append(state['exp_avg_sq'])
+
+                    if group['amsgrad']:
+                        max_exp_avg_sqs.append(state['max_exp_avg_sq'])
+
+                    # update the steps for each param group update
+                    state['step'] += 1
+                    # record the step after step update
+                    state_steps.append(state['step'])
+
+            beta1, beta2 = group['betas']
+            self.adam(group,
+                   exp_avgs,
+                   exp_avg_sqs,
+                   max_exp_avg_sqs,
+                   state_steps,
+                   group['amsgrad'],
+                   beta1,
+                   beta2,
+                   group['lr'],
+                   group['weight_decay'],
+                   group['eps']
+                   )
+        # FTP increment internal counters            
+        self.ftp.incre_counters()    
+        return loss
+
+    def adam(self, group,
+         exp_avgs,
+         exp_avg_sqs,
+         max_exp_avg_sqs,
+         state_steps, 
+         amsgrad: bool,
+         beta1: float,
+         beta2: float,
+         lr: float,
+         weight_decay: float,
+         eps: float):
+        
+        i = 0
+        for param, name, pre in zip(group['params'],group['name'],group['pre']):
+            if param.grad is None: 
+                continue
+            grad = param.grad
+            exp_avg = exp_avgs[i]
+            exp_avg_sq = exp_avg_sqs[i]
+            step = state_steps[i]
+            if amsgrad:
+                max_exp_avg_sq = max_exp_avg_sqs[i]
+
+            bias_correction1 = 1 - beta1 ** step
+            bias_correction2 = 1 - beta2 ** step
+
+            # Decay the first and second moment running average coefficient
+            exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
+            exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
+            if amsgrad:
+                # Maintains the maximum of all 2nd moment running avg. till now
+                torch.maximum(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
+                # Use the max. for normalizing running avg. of gradient
+                denom = (max_exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+            else:
+                denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+
+            step_size = lr / bias_correction1
+            i += 1
+            
+            # FTP step
+            d_p = step_size * exp_avg/denom + lr * weight_decay * param
+            new_p = self.ftp.step(name,param,pre,d_p)
+            if new_p is None :
+                new_p = param - d_p
+            param.copy_(new_p)
\ No newline at end of file
diff --git a/FTP_update3.py b/FTP_update3.py
new file mode 100644
index 0000000000000000000000000000000000000000..abe30e7515825e72138e9736edcf4e8480825da5
--- /dev/null
+++ b/FTP_update3.py
@@ -0,0 +1,329 @@
+import torch
+from torch.optim.optimizer import Optimizer, required
+import copy
+import math
+
+class FTP(object):
+    def __init__(self, k=1.0, exclude_set={}):
+        self.exclude_set = exclude_set
+        self.threshold = torch.nn.Hardtanh(0,1)
+        self.k = k  # Gradient annealing factor
+        self.j = 0  # Buffer counter
+
+        # AdamUtil parameters
+        self.mu = 1e-2
+        self.beta1 = 0.9
+        self.beta2 = 0.999
+        self.t = 1
+        
+        # Buffers
+        self.gamma_buff = []
+        self.first_m_gamma = []
+        self.second_m_gamma = []
+        self.prev_c = []
+        self.prev_scale = []
+    
+    @torch.no_grad()
+    def step(self, name, curr, pre, d_p):
+        if curr.requires_grad and name not in self.exclude_set:  # Exclude set includes those params that not be updated
+            c_t = (curr - d_p) - pre  # Compute potential new param
+            norms = self._mars_norm(c_t, tau=1.0)
+
+            # New: Apply spectral normalization to c_t
+            c_t = self._apply_spectral_norm(c_t)
+
+            if self.t == 1:
+                gamma = torch.tensor(1e-8, device=norms.device)
+                self._update_buffers(gamma)
+            else:
+                # Get previous values
+                gamma_prev = self.gamma_buff[self.j]
+                c_prev = self.prev_c[self.j]
+                scale_prev = self.prev_scale[self.j]
+
+                # Calculate gradient for gamma
+                gamma_grad = torch.sum(self._dot(curr.grad, c_prev, scale=scale_prev))
+
+                # Anneal positive gradient 
+                if gamma_grad > 0:
+                    gamma_grad = gamma_grad * self.k
+
+                gamma = self._adam_util(gamma_prev, gamma_grad)
+
+                # Clip gamma
+                gamma = self._clip(gamma, norms)
+            
+            # Update
+            denom = 1/norms
+            ratio = gamma * denom
+            new_p = pre + self.threshold(ratio) * c_t
+
+            new_p = self._apply_spectral_norm(new_p, tau=2.0)  # tau is the max allowed spectral norm
+
+            # Save updated values
+            self._update_buffers(gamma, c_t, denom)
+            self.j += 1
+            return new_p
+        else:
+            return None
+
+    def _apply_spectral_norm(self, param, tau=1.0):
+        
+        if param.ndim >= 2:  # Only for 2D matrices
+            u, s, vh = torch.linalg.svd(param, full_matrices=False)
+            s_clipped = torch.clamp(s, max=tau)  # Clip singular values
+            return torch.matmul(u, torch.matmul(torch.diag(s_clipped), vh))
+        return param  # No projection for 1D parameters
+    
+
+    def incre_counters(self):
+        self.t += 1
+        self.j = 0
+
+    @torch.no_grad()
+    def _mars_norm(self, tensor):
+        return torch.sum(torch.abs(tensor), dim=tuple(range(1, tensor.dim())), keepdim=True) + 1e-8
+    
+    @torch.no_grad()
+    def _clip(self, constraint, norms):
+        return torch.nn.functional.hardtanh(constraint, 1e-8, norms.max())
+        
+    @torch.no_grad()
+    def _dot(self, tensor1, tensor2, scale=1):
+        return torch.sum(torch.mul(tensor1, tensor2), dim=tuple(range(1, tensor1.dim())), keepdim=True) * scale
+    
+    @torch.no_grad()
+    def _adam_util(self, prev, grad):
+        first_moment = self.beta1 * self.first_m_gamma[self.j] + (1 - self.beta1) * grad
+        second_moment = self.beta2 * self.second_m_gamma[self.j] + (1 - self.beta2) * grad**2
+        self.first_m_gamma[self.j] = first_moment
+        self.second_m_gamma[self.j] = second_moment
+        first_moment = first_moment / (1 - self.beta1**self.t)
+        second_moment = second_moment / (1 - self.beta2**self.t)
+        return prev - self.mu * first_moment / (torch.sqrt(second_moment) + 1e-8)
+    
+    def _update_buffers(self, gamma, c_t=None, denom=None):
+        if c_t is None:
+            self.first_m_gamma.append(0.0)
+            self.second_m_gamma.append(0.0)
+            self.gamma_buff.append(gamma)
+            self.prev_c.append(0.0)
+            self.prev_scale.append(0.0)
+        else:
+            self.gamma_buff[self.j] = gamma
+            self.prev_c[self.j] = c_t
+            self.prev_scale[self.j] = denom
+
+class SGDP(Optimizer):
+    def __init__(self, params, lr=required, momentum=0, dampening=0,
+                 weight_decay=0, nesterov=False, k=1.0, exclude_set = {}):
+        if lr is not required and lr < 0.0:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if momentum < 0.0:
+            raise ValueError("Invalid momentum value: {}".format(momentum))
+        if weight_decay < 0.0:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+
+        defaults = dict(lr=lr, momentum=momentum, dampening=dampening,
+                        weight_decay=weight_decay, nesterov=nesterov)
+        if nesterov and (momentum <= 0 or dampening != 0):
+            raise ValueError("Nesterov momentum requires a momentum and zero dampening")
+        super(SGDP, self).__init__(params, defaults)
+        self.first_iter_flag = False
+         
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+                    
+
+    def __setstate__(self, state):
+        super(SGDP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('nesterov', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+        for group in self.param_groups:
+            weight_decay = group['weight_decay']
+            momentum = group['momentum']
+            dampening = group['dampening']
+            nesterov = group['nesterov']
+            
+            for p, name, pre in zip(group['params'],group['name'],group['pre']):
+                if p.grad is None:
+                    continue
+                d_p = p.grad
+                if weight_decay != 0:
+                    d_p = d_p.add(p, alpha=weight_decay)
+                if momentum != 0:
+                    param_state = self.state[p]
+                    if 'momentum_buffer' not in param_state:
+                        buf = param_state['momentum_buffer'] = torch.clone(d_p).detach()
+                    else:
+                        buf = param_state['momentum_buffer']
+                        buf.mul_(momentum).add_(d_p, alpha=1 - dampening) 
+                    if nesterov:
+                        d_p = d_p.add(buf, alpha=momentum)
+                    else:
+                        d_p = buf
+
+                # FTP step
+                d_p = group['lr']*d_p
+                new_p = self.ftp.step(name,p,pre,d_p)
+                if new_p is not None :
+                    p.copy_(new_p)
+                else:
+                    p.add_(d_p, alpha=-1)
+        # FTP increment internal counters
+        self.ftp.incre_counters()        
+        return loss
+
+
+class AdamP(Optimizer):
+    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
+                 weight_decay=0, amsgrad=False, k=1.0, exclude_set={}):
+        if not 0.0 <= lr:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if not 0.0 <= eps:
+            raise ValueError("Invalid epsilon value: {}".format(eps))
+        if not 0.0 <= betas[0] < 1.0:
+            raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
+        if not 0.0 <= betas[1] < 1.0:
+            raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
+        if not 0.0 <= weight_decay:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+        defaults = dict(lr=lr, betas=betas, eps=eps,
+                        weight_decay=weight_decay, amsgrad=amsgrad)
+        super(AdamP, self).__init__(params, defaults)
+
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+
+    def __setstate__(self, state):
+        super(AdamP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('amsgrad', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+
+        for group in self.param_groups:
+            params_with_grad = []
+            grads = []
+            exp_avgs = []
+            exp_avg_sqs = []
+            max_exp_avg_sqs = []
+            state_steps = []
+
+            for p in group['params']:
+                if p.grad is not None:
+                    params_with_grad.append(p)
+                    if p.grad.is_sparse:
+                        raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')
+                    grads.append(p.grad)
+
+                    state = self.state[p]
+                    # Lazy state initialization
+                    if len(state) == 0:
+                        state['step'] = 0
+                        # Exponential moving average of gradient values
+                        state['exp_avg'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        # Exponential moving average of squared gradient values
+                        state['exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        if group['amsgrad']:
+                            # Maintains max of all exp. moving avg. of sq. grad. values
+                            state['max_exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+
+                    exp_avgs.append(state['exp_avg'])
+                    exp_avg_sqs.append(state['exp_avg_sq'])
+
+                    if group['amsgrad']:
+                        max_exp_avg_sqs.append(state['max_exp_avg_sq'])
+
+                    # update the steps for each param group update
+                    state['step'] += 1
+                    # record the step after step update
+                    state_steps.append(state['step'])
+
+            beta1, beta2 = group['betas']
+            self.adam(group,
+                   exp_avgs,
+                   exp_avg_sqs,
+                   max_exp_avg_sqs,
+                   state_steps,
+                   group['amsgrad'],
+                   beta1,
+                   beta2,
+                   group['lr'],
+                   group['weight_decay'],
+                   group['eps']
+                   )
+        # FTP increment internal counters            
+        self.ftp.incre_counters()    
+        return loss
+
+    def adam(self, group,
+         exp_avgs,
+         exp_avg_sqs,
+         max_exp_avg_sqs,
+         state_steps, 
+         amsgrad: bool,
+         beta1: float,
+         beta2: float,
+         lr: float,
+         weight_decay: float,
+         eps: float):
+        
+        i = 0
+        for param, name, pre in zip(group['params'],group['name'],group['pre']):
+            if param.grad is None: 
+                continue
+            grad = param.grad
+            exp_avg = exp_avgs[i]
+            exp_avg_sq = exp_avg_sqs[i]
+            step = state_steps[i]
+            if amsgrad:
+                max_exp_avg_sq = max_exp_avg_sqs[i]
+
+            bias_correction1 = 1 - beta1 ** step
+            bias_correction2 = 1 - beta2 ** step
+
+            # Decay the first and second moment running average coefficient
+            exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
+            exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
+            if amsgrad:
+                # Maintains the maximum of all 2nd moment running avg. till now
+                torch.maximum(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
+                # Use the max. for normalizing running avg. of gradient
+                denom = (max_exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+            else:
+                denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+
+            step_size = lr / bias_correction1
+            i += 1
+            
+            # FTP step
+            d_p = step_size * exp_avg/denom + lr * weight_decay * param
+            new_p = self.ftp.step(name,param,pre,d_p)
+            if new_p is None :
+                new_p = param - d_p
+            param.copy_(new_p)
\ No newline at end of file
diff --git a/FTP_update4.py b/FTP_update4.py
new file mode 100644
index 0000000000000000000000000000000000000000..abe30e7515825e72138e9736edcf4e8480825da5
--- /dev/null
+++ b/FTP_update4.py
@@ -0,0 +1,329 @@
+import torch
+from torch.optim.optimizer import Optimizer, required
+import copy
+import math
+
+class FTP(object):
+    def __init__(self, k=1.0, exclude_set={}):
+        self.exclude_set = exclude_set
+        self.threshold = torch.nn.Hardtanh(0,1)
+        self.k = k  # Gradient annealing factor
+        self.j = 0  # Buffer counter
+
+        # AdamUtil parameters
+        self.mu = 1e-2
+        self.beta1 = 0.9
+        self.beta2 = 0.999
+        self.t = 1
+        
+        # Buffers
+        self.gamma_buff = []
+        self.first_m_gamma = []
+        self.second_m_gamma = []
+        self.prev_c = []
+        self.prev_scale = []
+    
+    @torch.no_grad()
+    def step(self, name, curr, pre, d_p):
+        if curr.requires_grad and name not in self.exclude_set:  # Exclude set includes those params that not be updated
+            c_t = (curr - d_p) - pre  # Compute potential new param
+            norms = self._mars_norm(c_t, tau=1.0)
+
+            # New: Apply spectral normalization to c_t
+            c_t = self._apply_spectral_norm(c_t)
+
+            if self.t == 1:
+                gamma = torch.tensor(1e-8, device=norms.device)
+                self._update_buffers(gamma)
+            else:
+                # Get previous values
+                gamma_prev = self.gamma_buff[self.j]
+                c_prev = self.prev_c[self.j]
+                scale_prev = self.prev_scale[self.j]
+
+                # Calculate gradient for gamma
+                gamma_grad = torch.sum(self._dot(curr.grad, c_prev, scale=scale_prev))
+
+                # Anneal positive gradient 
+                if gamma_grad > 0:
+                    gamma_grad = gamma_grad * self.k
+
+                gamma = self._adam_util(gamma_prev, gamma_grad)
+
+                # Clip gamma
+                gamma = self._clip(gamma, norms)
+            
+            # Update
+            denom = 1/norms
+            ratio = gamma * denom
+            new_p = pre + self.threshold(ratio) * c_t
+
+            new_p = self._apply_spectral_norm(new_p, tau=2.0)  # tau is the max allowed spectral norm
+
+            # Save updated values
+            self._update_buffers(gamma, c_t, denom)
+            self.j += 1
+            return new_p
+        else:
+            return None
+
+    def _apply_spectral_norm(self, param, tau=1.0):
+        
+        if param.ndim >= 2:  # Only for 2D matrices
+            u, s, vh = torch.linalg.svd(param, full_matrices=False)
+            s_clipped = torch.clamp(s, max=tau)  # Clip singular values
+            return torch.matmul(u, torch.matmul(torch.diag(s_clipped), vh))
+        return param  # No projection for 1D parameters
+    
+
+    def incre_counters(self):
+        self.t += 1
+        self.j = 0
+
+    @torch.no_grad()
+    def _mars_norm(self, tensor):
+        return torch.sum(torch.abs(tensor), dim=tuple(range(1, tensor.dim())), keepdim=True) + 1e-8
+    
+    @torch.no_grad()
+    def _clip(self, constraint, norms):
+        return torch.nn.functional.hardtanh(constraint, 1e-8, norms.max())
+        
+    @torch.no_grad()
+    def _dot(self, tensor1, tensor2, scale=1):
+        return torch.sum(torch.mul(tensor1, tensor2), dim=tuple(range(1, tensor1.dim())), keepdim=True) * scale
+    
+    @torch.no_grad()
+    def _adam_util(self, prev, grad):
+        first_moment = self.beta1 * self.first_m_gamma[self.j] + (1 - self.beta1) * grad
+        second_moment = self.beta2 * self.second_m_gamma[self.j] + (1 - self.beta2) * grad**2
+        self.first_m_gamma[self.j] = first_moment
+        self.second_m_gamma[self.j] = second_moment
+        first_moment = first_moment / (1 - self.beta1**self.t)
+        second_moment = second_moment / (1 - self.beta2**self.t)
+        return prev - self.mu * first_moment / (torch.sqrt(second_moment) + 1e-8)
+    
+    def _update_buffers(self, gamma, c_t=None, denom=None):
+        if c_t is None:
+            self.first_m_gamma.append(0.0)
+            self.second_m_gamma.append(0.0)
+            self.gamma_buff.append(gamma)
+            self.prev_c.append(0.0)
+            self.prev_scale.append(0.0)
+        else:
+            self.gamma_buff[self.j] = gamma
+            self.prev_c[self.j] = c_t
+            self.prev_scale[self.j] = denom
+
+class SGDP(Optimizer):
+    def __init__(self, params, lr=required, momentum=0, dampening=0,
+                 weight_decay=0, nesterov=False, k=1.0, exclude_set = {}):
+        if lr is not required and lr < 0.0:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if momentum < 0.0:
+            raise ValueError("Invalid momentum value: {}".format(momentum))
+        if weight_decay < 0.0:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+
+        defaults = dict(lr=lr, momentum=momentum, dampening=dampening,
+                        weight_decay=weight_decay, nesterov=nesterov)
+        if nesterov and (momentum <= 0 or dampening != 0):
+            raise ValueError("Nesterov momentum requires a momentum and zero dampening")
+        super(SGDP, self).__init__(params, defaults)
+        self.first_iter_flag = False
+         
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+                    
+
+    def __setstate__(self, state):
+        super(SGDP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('nesterov', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+        for group in self.param_groups:
+            weight_decay = group['weight_decay']
+            momentum = group['momentum']
+            dampening = group['dampening']
+            nesterov = group['nesterov']
+            
+            for p, name, pre in zip(group['params'],group['name'],group['pre']):
+                if p.grad is None:
+                    continue
+                d_p = p.grad
+                if weight_decay != 0:
+                    d_p = d_p.add(p, alpha=weight_decay)
+                if momentum != 0:
+                    param_state = self.state[p]
+                    if 'momentum_buffer' not in param_state:
+                        buf = param_state['momentum_buffer'] = torch.clone(d_p).detach()
+                    else:
+                        buf = param_state['momentum_buffer']
+                        buf.mul_(momentum).add_(d_p, alpha=1 - dampening) 
+                    if nesterov:
+                        d_p = d_p.add(buf, alpha=momentum)
+                    else:
+                        d_p = buf
+
+                # FTP step
+                d_p = group['lr']*d_p
+                new_p = self.ftp.step(name,p,pre,d_p)
+                if new_p is not None :
+                    p.copy_(new_p)
+                else:
+                    p.add_(d_p, alpha=-1)
+        # FTP increment internal counters
+        self.ftp.incre_counters()        
+        return loss
+
+
+class AdamP(Optimizer):
+    def __init__(self, params, lr=1e-3, betas=(0.9, 0.999), eps=1e-8,
+                 weight_decay=0, amsgrad=False, k=1.0, exclude_set={}):
+        if not 0.0 <= lr:
+            raise ValueError("Invalid learning rate: {}".format(lr))
+        if not 0.0 <= eps:
+            raise ValueError("Invalid epsilon value: {}".format(eps))
+        if not 0.0 <= betas[0] < 1.0:
+            raise ValueError("Invalid beta parameter at index 0: {}".format(betas[0]))
+        if not 0.0 <= betas[1] < 1.0:
+            raise ValueError("Invalid beta parameter at index 1: {}".format(betas[1]))
+        if not 0.0 <= weight_decay:
+            raise ValueError("Invalid weight_decay value: {}".format(weight_decay))
+        defaults = dict(lr=lr, betas=betas, eps=eps,
+                        weight_decay=weight_decay, amsgrad=amsgrad)
+        super(AdamP, self).__init__(params, defaults)
+
+        # initialize FTP
+        self.ftp = FTP(k, exclude_set=exclude_set)
+
+    def __setstate__(self, state):
+        super(AdamP, self).__setstate__(state)
+        for group in self.param_groups:
+            group.setdefault('amsgrad', False)
+
+    @torch.no_grad()
+    def step(self, closure=None):
+        """Performs a single optimization step.
+
+        Arguments:
+            closure (callable, optional): A closure that reevaluates the model
+                and returns the loss.
+        """
+        loss = None
+        if closure is not None:
+            with torch.enable_grad():
+                loss = closure()
+
+        for group in self.param_groups:
+            params_with_grad = []
+            grads = []
+            exp_avgs = []
+            exp_avg_sqs = []
+            max_exp_avg_sqs = []
+            state_steps = []
+
+            for p in group['params']:
+                if p.grad is not None:
+                    params_with_grad.append(p)
+                    if p.grad.is_sparse:
+                        raise RuntimeError('Adam does not support sparse gradients, please consider SparseAdam instead')
+                    grads.append(p.grad)
+
+                    state = self.state[p]
+                    # Lazy state initialization
+                    if len(state) == 0:
+                        state['step'] = 0
+                        # Exponential moving average of gradient values
+                        state['exp_avg'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        # Exponential moving average of squared gradient values
+                        state['exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+                        if group['amsgrad']:
+                            # Maintains max of all exp. moving avg. of sq. grad. values
+                            state['max_exp_avg_sq'] = torch.zeros_like(p, memory_format=torch.preserve_format)
+
+                    exp_avgs.append(state['exp_avg'])
+                    exp_avg_sqs.append(state['exp_avg_sq'])
+
+                    if group['amsgrad']:
+                        max_exp_avg_sqs.append(state['max_exp_avg_sq'])
+
+                    # update the steps for each param group update
+                    state['step'] += 1
+                    # record the step after step update
+                    state_steps.append(state['step'])
+
+            beta1, beta2 = group['betas']
+            self.adam(group,
+                   exp_avgs,
+                   exp_avg_sqs,
+                   max_exp_avg_sqs,
+                   state_steps,
+                   group['amsgrad'],
+                   beta1,
+                   beta2,
+                   group['lr'],
+                   group['weight_decay'],
+                   group['eps']
+                   )
+        # FTP increment internal counters            
+        self.ftp.incre_counters()    
+        return loss
+
+    def adam(self, group,
+         exp_avgs,
+         exp_avg_sqs,
+         max_exp_avg_sqs,
+         state_steps, 
+         amsgrad: bool,
+         beta1: float,
+         beta2: float,
+         lr: float,
+         weight_decay: float,
+         eps: float):
+        
+        i = 0
+        for param, name, pre in zip(group['params'],group['name'],group['pre']):
+            if param.grad is None: 
+                continue
+            grad = param.grad
+            exp_avg = exp_avgs[i]
+            exp_avg_sq = exp_avg_sqs[i]
+            step = state_steps[i]
+            if amsgrad:
+                max_exp_avg_sq = max_exp_avg_sqs[i]
+
+            bias_correction1 = 1 - beta1 ** step
+            bias_correction2 = 1 - beta2 ** step
+
+            # Decay the first and second moment running average coefficient
+            exp_avg.mul_(beta1).add_(grad, alpha=1 - beta1)
+            exp_avg_sq.mul_(beta2).addcmul_(grad, grad, value=1 - beta2)
+            if amsgrad:
+                # Maintains the maximum of all 2nd moment running avg. till now
+                torch.maximum(max_exp_avg_sq, exp_avg_sq, out=max_exp_avg_sq)
+                # Use the max. for normalizing running avg. of gradient
+                denom = (max_exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+            else:
+                denom = (exp_avg_sq.sqrt() / math.sqrt(bias_correction2)).add_(eps)
+
+            step_size = lr / bias_correction1
+            i += 1
+            
+            # FTP step
+            d_p = step_size * exp_avg/denom + lr * weight_decay * param
+            new_p = self.ftp.step(name,param,pre,d_p)
+            if new_p is None :
+                new_p = param - d_p
+            param.copy_(new_p)
\ No newline at end of file
diff --git a/README.md b/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..650f15354f4e555efba1b19a1c92c585c3df0e51
--- /dev/null
+++ b/README.md
@@ -0,0 +1,90 @@
+# 💥 Mission: Impossible Language Models 💥
+
+This is the code repository for the paper "[Mission: Impossible Language Models](https://arxiv.org/abs/2401.06416)".
+
+If you use our code, please cite our paper:
+
+```
+@misc{kallini2024mission,
+      title={Mission: Impossible Language Models}, 
+      author={Julie Kallini and Isabel Papadimitriou and Richard Futrell and Kyle Mahowald and Christopher Potts},
+      year={2024},
+      eprint={2401.06416},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+```
+
+This repository contains the code necessary to fully replicate our paper, including the scripts to create impossible language datasets, train GPT-2 models, and run all experiments. We also include the notebooks to generate the result graphs in our paper.
+
+Let's get started!
+
+## Setup
+
+First, clone the repo and install dependencies:
+
+```
+git clone https://github.com/jkallini/mission-impossible-language-models.git
+cd mission-impossible-language-models
+pip install -r requirements.txt
+```
+
+## Impossible Language Dataset Creation
+
+The scripts for creating impossible language datasets are located in the `data/` directory.
+First, you must download a copy of the [BabyLM dataset](https://babylm.github.io/), which we use for our experiments. 
+Then, make sure to set `BABYLM_DATA_PATH` in the `utils.py` file to the path on your system where your BabyLM dataset is located.
+
+After downloading the BabyLM dataset, you will need to tag it with morphological features and part-of-speech tags. You can use our `tag.py` script.
+
+With the tagged data, you can easily recreate one of the impossible language datasets described in our paper. These are predefined and listed in the `PERTURBATIONS` section at the end of `utils.py`. Here is an
+example for the PartialReverse language from the paper:
+
+```
+python3 perturb.py reverse_partial 100M
+```
+
+This will create a perturbed version of the 100M BabyLM train set. You may use `perturb.py` or `perturb.sh` to perturb multiple splits at the same time.
+
+### Defining New Impossible Languages
+
+You can also define your own impossible languages! They are described by four attributes:
+
+1. `perturbation_function`: function mapping tagged sentences to sequences of GPT-2 tokens.
+2. `affect_function`: function that determines whether an input sentences is "affected" or altered by the perturbation.
+3. `filter_function`: function that determines whether an input sentence should be included in the final dataset.
+4. `gpt2_tokenizer`: tokenizer used to perturb this dataset.
+
+You can add these definitions to `utils.py`, where the existing perturbations are located.
+
+
+## Model Training
+
+To train GPT-2 models, we use [`mistral`](https://github.com/stanford-crfm/mistral). If you would like to train GPT-2s with `mistral` as well, please follow their steps for installation. You may download their repo anywhere on your system.
+
+Next, make sure to change the following constants in `utils.py`:
+- `CHECKPOINT_WRITE_PATH`: the path where your training checkpoints will be written.
+- `CHECKPOINT_READ_PATH`: the path where you will read training checkpoints when running experiments.
+
+Our training scripts are in the `training/` directory.
+Once you have `mistral` installed, set `MISTRAL_PATH` to the path of your library in `prepare_training.sh`. Then, you can use this script to generate the config files that you will use to launch `mistral` training runs.
+
+Our scripts will create the config files and move them to the location of your `mistral` directory—you will only need to launch the training run. Here's an example command to launch training for the PartialReverse language using the 100M training set with the random seed set to 41:
+
+```
+CUDA_VISIBLE_DEVICES=0 python3 train.py --config conf/train_reverse_partial_100M_randinit_seed41.yaml --nnodes 1 --nproc_per_node 1 --training_arguments.fp16 true --training_arguments.warmup_steps 300 --training_arguments.max_steps 3000
+```
+
+## Experiments
+
+The main paper includes three experiments: perplexities, surprisals, and causal interventions. The appendix also includes a constituency probing experiment.
+
+The scripts to run each of these experiments is separated into their own subdirectories:
+
+1. `perplexities/`: code to run perplexity experiments. You may use `perplexities.py` or `perplexities.sh` to run experiments for multiple languages at the same time.
+2. `hop_surprisal/`: code to run surprisal experiments for the *Hop languages, in `hop_surprisal.py`.
+3. `hop_interventions/`: code to run interchange intervention experiments for the *Hop languages. First generate the agreement data using `create_agreement_data.py`, then run the intervention experiments using `hop_interventions.py`.
+You will need to separately clone and install [`align-transformers`](https://github.com/frankaging/align-transformers) (recently renamed to `pyvene`) and set `PATH_TO_ALIGN_TRANSFORMERS` to the path where the library is located on your system.
+4. `edge_probing/`: code to run constituency probing experiments. Use `get_constituency_parses.py` and `load_phrase_data.py` to prepare the test data, and use `edge_probing.py` to run the experiments.
+
+Each directory contains python notebooks to generate the result graphs shown in the paper.
\ No newline at end of file
diff --git a/edge_probing/edge_probing.ipynb b/edge_probing/edge_probing.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..094ceb2523767e6e205c75700c102833c44a84c4
--- /dev/null
+++ b/edge_probing/edge_probing.ipynb
@@ -0,0 +1,371 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 37,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# For importing utils\n",
+    "import sys\n",
+    "sys.path.append(\"..\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 38,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from utils import PERTURBATIONS\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "from scipy import stats"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 39,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plot_datasets(ax, datasets, seeds, data_path, layer, checkpoints, nps):\n",
+    "\n",
+    "    for dataset in datasets:\n",
+    "        # Filter the data for the layer and available checkpoints\n",
+    "        all_seeds_layer_data = []\n",
+    "        for s in seeds:\n",
+    "            df = pd.read_csv(data_path.format(dataset[\"data\"], nps, s))\n",
+    "            layer_data = df[df['GPT-2 Layer'] == layer][[f'Accuracy (ckpt {cp})' for cp in checkpoints]].T\n",
+    "            layer_data.columns = [dataset['label']]\n",
+    "            all_seeds_layer_data.append(list(layer_data[dataset['label']]))\n",
+    "        \n",
+    "        all_seeds_layer_data = np.array(all_seeds_layer_data) * 100.0\n",
+    "        means = np.mean(all_seeds_layer_data, axis=0)\n",
+    "\n",
+    "        ci = None\n",
+    "        if len(seeds) > 1:\n",
+    "            sems = stats.sem(all_seeds_layer_data, axis=0)\n",
+    "            # Calculate confidence interval using t-distribution\n",
+    "            ci_lower, ci_upper = stats.t.interval(0.95, df=len(seeds)-1,\n",
+    "                                                  loc=means, scale=sems)\n",
+    "            ci = (ci_upper - ci_lower) / 2\n",
+    "\n",
+    "        # Plotting the data for the layer\n",
+    "        ax.errorbar(checkpoints, means, ci, marker=dataset['marker'], linestyle=dataset['linestyle'],\n",
+    "                    color=PERTURBATIONS[dataset['data']]['color'], label=dataset['label'])"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Plot Accuracy by Averaging Last 4 Layers"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 40,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plot_accuracy(rev_datasets, hop_datasets, checkpoints, seeds, pos_encodings=True):\n",
+    "\n",
+    "    # Setting up the plot with all subplots in a single row\n",
+    "    fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(8, 4), sharey=False)\n",
+    "    # fig.subplots_adjust(wspace=0)\n",
+    "    fig.supxlabel(\"Training Steps\", y=-0.01)\n",
+    "\n",
+    "    if pos_encodings:\n",
+    "        nps = \"\"\n",
+    "    else:\n",
+    "        nps = \"_no_pos_encodings\"\n",
+    "\n",
+    "    data_path = \"edge_probing_results/{}_100M{}/randinit_mean_pooling_seed{}.csv\"\n",
+    "\n",
+    "    # Plotting the data\n",
+    "    axes[0].set_ylabel(\"Probe Accuracy\", fontsize=12)\n",
+    "\n",
+    "    # Plot *Reverse\n",
+    "    plot_datasets(axes[0], rev_datasets, seeds, data_path, \"Avg Last 4\", checkpoints, nps)\n",
+    "    axes[0].set_title(\"*Reverse\")\n",
+    "    axes[0].grid(True)\n",
+    "    axes[0].legend(loc='lower right', framealpha=1)\n",
+    "\n",
+    "    # Plot *Hop\n",
+    "    plot_datasets(axes[1], hop_datasets, seeds, data_path, \"Avg Last 4\", checkpoints, nps)\n",
+    "    axes[1].set_title(\"*Hop\")\n",
+    "    axes[1].grid(True)\n",
+    "    axes[1].legend(loc='lower right', framealpha=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 41,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAArcAAAGOCAYAAACNACuPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAADv7ElEQVR4nOzdd3gUVffA8e/MbnoPSQiQBoHQW+hFkE5AEQREsIDtRQUs/OwVbOhrQUEUCyKC8AoWREWqSCcQegslkEIChJBG+u7O/P5YsmRTN8lmE8L9PM8+mtnJzN1J2Jy9c+45kqqqKoIgCIIgCIJQD8i1PQBBEARBEARBsBYR3AqCIAiCIAj1hghuBUEQBEEQhHpDBLeCIAiCIAhCvSGCW0EQBEEQBKHeEMGtIAiCIAiCUG+I4FYQBEEQBEGoN0RwKwiCIAiCINQbIrgVBEEQBEEQ6g0R3AqCIAiCIAj1hghuhTrt33//ZcqUKSW2S5Jk9nB3d6d///789ddfth+kIAhCPVfee/H06dNL/Z7vv/8eSZKIioqq4dEJgjkR3Ap1TkZGBnv27CmxPT09ncjISNPXQ4YMYenSpfzwww+88MILnD17ljvvvJP169fbcriCIAj1kqXvxYJQ14jgVqhz4uLimDx5Mk8//TRZWVkA/PLLL4SHh7N7927TfmFhYdx///088MADvPbaa2zatAlVVfnss89qa+jlUlWV3Nzc2h6GIAiCRSx9LxaEukYEt0Kd06FDB44ePUpgYCBTp07l559/ZuXKlWzZsoVnnnmmzO9r3bo1Pj4+xMTEmG3Pz8/nzTffpHnz5jg4OBAYGMgLL7xAfn6+aZ927doxYMCAEsdUFIUmTZowbtw4s22ffvopbdu2xdHRkYYNGzJ16lTS0tLMvjckJIQ77riD9evX07VrV5ycnPjqq68A2LhxI3379sXT0xNXV1datmzJK6+8UulxC4Ig1JSqvhdX5J9//uG2227DxcUFT09P7rrrLk6ePGm2z6xZs5AkiejoaO655x7c3d1p0KABTz/9NHl5edV8ZUJ9p63tAQhCaSRJQpZlJEkyfV34/2XJyMggLS2N0NBQ0zZFURg1ahQ7duzgP//5D61bt+bo0aPMnTuX06dPs3r1agAmTJjArFmzuHTpEv7+/qbv37FjB0lJSdx7772mbVOnTuX777/noYce4qmnnuL8+fN8/vnnHDx4kJ07d2JnZ2fa99SpU0ycOJGpU6fy2GOP0bJlS44fP84dd9xBhw4deOutt3BwcODs2bPs3Lmz0uMWBEGoSZa8F+fl5ZGSklLiewtne4vatGkTERERNGvWjFmzZpGbm8v8+fPp06cPBw4cICQkxGz/e+65h5CQEObMmcOePXuYN28eaWlp/PDDD9Z7kUL9owpCHXPkyBG1VatW6owZM9Q//vhDnTx5srpq1Sq1adOm6qeffqqqqqoC6iOPPKJeuXJFTU5OVqOiotThw4ergPrhhx+ajrV06VJVlmV1+/btZudYuHChCqg7d+5UVVVVT506pQLq/PnzzfZ78sknVVdXVzUnJ0dVVVXdvn27Cqg//vij2X7r1q0rsT04OFgF1HXr1pntO3fuXBVQr1y5UuY1sHTcgiAINcXS9+KKHvv27TMds1OnTqqfn5969epV07bDhw+rsiyrDz74oGnbm2++qQLqqFGjzMb05JNPqoB6+PDhGn71ws1MpCUIdU5QUBCLFy9m3rx5uLq6AjBu3DgOHDhAz549TfstWrQIX19f/Pz86Nq1K5s3b+aFF15g5syZpn1WrVpF69atadWqFSkpKabHwIEDAdiyZQtgzN/t1KkTP/30k+l7DQYDP//8M3feeSdOTk6m43l4eDBkyBCz43Xp0gVXV1fT8Qo1bdqUYcOGmW3z9PQE4Pfff0dRlFKvgaXjFgRBqCmWvhffddddbNy4scTj+eefNzvexYsXOXToEFOmTMHb29u0vUOHDgwZMoS1a9eWGMO0adPMvp4xYwZAqfsKQiGRliDUOR4eHmZvnIU8PT3p0aOH6eu77rqL6dOnU1BQwL59+3jvvffIyclBlm98Zjtz5gwnT57E19e31HMlJyeb/n/ChAm88sorJCYm0qRJE/7991+Sk5OZMGGC2fEyMjLw8/Or8HhgDG6LmzBhAt9++y2PPvooL730EoMGDeLuu+9m3LhxprFXZtyCIAg1wdL34oCAAAYPHlxivwsXLph9HRcXB0DLli1L7Nu6dWvWr19PdnY2Li4upu0tWrQw2y80NBRZlomNja3UaxFuLSK4Feq022+/ndtvv73U54q+oY4YMQIfHx+mT5/OgAEDuPvuuwFj7mr79u355JNPSj1GYGCg6f8nTJjAyy+/zKpVq3jmmWdYuXIlHh4eDB8+3LSPoij4+fnx448/lnq84sFo4Yxv8W3btm1jy5Yt/PXXX6xbt46ffvqJgQMHsmHDBjQaTaXGLQiCUNPKey+2pYrWXggCiOBWqEemTp3K3Llzee211xgzZgySJBEaGsrhw4cZNGhQhW+KTZs2pXv37vz0009Mnz6dX3/9ldGjR+Pg4GDaJzQ0lE2bNtGnT59SA1dLybLMoEGDGDRoEJ988gnvvfcer776Klu2bGHw4MGVGrcgCMLNIDg4GDAutC0uOjoaHx8fs1lbMN7FKnoH7OzZsyiKUmLhmSAUJXJuhXpDq9Xyf//3f5w8eZLff/8dMK60TUxM5Jtvvimxf25uLtnZ2WbbJkyYwJ49e/juu+9ISUkxS0koPJ7BYODtt98ucTy9Xk96enqF40xNTS2xrVOnTgCmMl+VHbcgCEJd16hRIzp16sSSJUvM3iuPHTvGhg0bGDFiRInvWbBggdnX8+fPByAiIqJGxyrc3MTMrVCvTJkyhTfeeIMPPviA0aNH88ADD7By5Uoef/xxtmzZQp8+fTAYDERHR7Ny5UpTDdpC99xzD8899xzPPfcc3t7eJfLI+vfvz9SpU5kzZw6HDh1i6NCh2NnZcebMGVatWsVnn31mVhO3NG+99Rbbtm1j5MiRBAcHk5yczBdffEFAQAB9+/YFqPS4BUEQbgYffvghERER9OrVi0ceecRUCszDw4NZs2aV2P/8+fOMGjWK4cOHs3v3bpYtW8akSZPo2LGj7Qcv3Dxqu1yDIFQFoE6bNq3U52bNmqUC6pYtW1RVVdWCggL1gw8+UNu2bas6ODioXl5eapcuXdTZs2erGRkZJb6/T58+KqA++uijZZ7/66+/Vrt06aI6OTmpbm5uavv27dUXXnhBTUpKMu0THBysjhw5ssT3bt68Wb3rrrvUxo0bq/b29mrjxo3ViRMnqqdPnzbbr7LjFgRBsLXy3osXL15cohSYqqrqpk2b1D59+qhOTk6qu7u7euedd6onTpww26ewFNiJEyfUcePGqW5ubqqXl5c6ffp0NTc3t8Zej1A/SKqqqrUZXAuCIAiCIBQ1a9YsZs+ezZUrV/Dx8ant4Qg3GZFzKwiCIAiCINQbIrgVBEEQBEEQ6g0R3AqCIAiCIAj1hsi5FQRBEARBEOoNMXMrCIIgCIIg1BsiuBUEQahHQkJCkCSpxGPatGkAXLp0iQceeAB/f39cXFwIDw/nl19+qeVRC4IgWM8t18RBURSSkpJwc3MTbU0FQagRqqpy7do1GjdujCzbdg5h3759GAwG09fHjh1jyJAhjB8/HoAHH3yQ9PR01qxZg4+PD8uXL+eee+4hKiqKzp07W3QO8T4qCEJNq9b7aG0W2a0NCQkJKiAe4iEe4lHjj4SEhNp+y1OffvppNTQ0VFUURVVVVXVxcVF/+OEHs328vb3Vb775xuJjivdR8RAP8bDVoyrvo7fczK2bmxsACQkJuLu71/Jo6g+dTseGDRtM7WiFmiWut+1U5VpnZmYSGBhoer+pLQUFBSxbtoyZM2eaZlh79+7NTz/9xMiRI/H09GTlypXk5eVx++23l3mc/Px88vPzTV+r19chnz9/vtZfY32i0+nYsmULAwYMEP+ubUBcb9upyrW+du0aTZs2rdJ7zC0X3Ba+wbu7u4vg1op0Oh3Ozs64u7uLNwkbENfbdqpzrWv7lv3q1atJT09nypQppm0rV65kwoQJNGjQAK1Wi7OzM7/99hvNmzcv8zhz5sxh9uzZJbbv3r0bZ2fnmhj6LcvZ2ZnIyMjaHsYtQ1xv26nstc7JyQGq9j56ywW3giAIt4pFixYRERFB48aNTdtef/110tPT2bRpEz4+PqxevZp77rmH7du30759+1KP8/LLLzNz5kzT14Uz00OHDhWTBFak0+nYuHEjQ4YMER9abUBcb9upyrXOzMys8vlEcCsIglAPxcXFsWnTJn799VfTtpiYGD7//HOOHTtG27ZtAejYsSPbt29nwYIFLFy4sNRjOTg44ODgUGK7nZ2dCApqgLiutiWut+1U5lpX52dSp0qBiRI2giAI1rF48WL8/PwYOXKkaVvhbb7iK481Gg2Koth0fIIgCDWlTs3c2qKEjSAIQn2nKAqLFy9m8uTJaLU33uZbtWpF8+bNmTp1Kh999BENGjRg9erVbNy4kT///LMWRywIgmA9dWrm1tfXF39/f9Pjzz//JDQ0lP79+wOwa9cuZsyYQffu3WnWrBmvvfYanp6e7N+/v5ZHLgiCUHds2rSJ+Ph4Hn74YbPtdnZ2rF27Fl9fX+688046dOjADz/8wJIlSxgxYkQtjVYQBMG66tTMbVHWKmEjCIJwqxk6dKipXFdxLVq0EOlcwk1rVSw8Ewmf9YBxIbU9GqGuqrPBrbVK2BSvz1i4+k6n06HT6Wps/LeawmtZ366poiicOXOGjIwMPDw8aNGihc07TpWmvl7vuqgq11r8XATB+pJz4fFdkKGDx3dDv4bg51TboxLqojob3FqrhE1Z9Rk3bNgg6jPWgI0bN9b2EKwmKSmJI0eOkJeXZ9rm6OhIhw4dzH4va1NduN6qqpKSkkJ+fj4ODg74+PjUen3XmlCZa124cEsQBOtQVXhyD2TpjV9f08G0PbBqQO2OS6ib6mRwa80SNqI+o23Ut3qBBw4cYPXq1SW25+XlsXfvXh5//HHCw8NtP7Dr6sr1PnDgAD/99BNpaWmmbV5eXkyYMKFWr4812bo+oyAIJa2KhdXxN742qPBbvHH7+JBaGpRQZ9XJ4NaaJWxEfUbbqg/XVVEUfvrpp3L3WblyJd26dav1FIXavN5RUVGlfqhMS0tj4cKFTJs2ja5du9bCyGqGreozCoJgLjkXntgNElA0k1zCuL2/SE8Qiqn95MFiLClhs3fvXmJiYvj444/ZuHEjo0ePrr0BC/XO6dOnzWYiS5Oamsrp06dtNKK6R1EUli9fXu4+K1asELVTBUGolqLpCMWXSKrcSE+oLYqiEB0dzZ49e4iOjhbveXVEnZu5raiEzUsvvcSdd95JVlYWzZs3FyVsBKtLT0+36n71UWU+ALRq1cpGoxIEob45nm6ejlBcYXrC8TRo62WzYQHGu1fLly8vkZY1adKkenPX6matTlHngltRwkaobZ6enlbdz9oUReHUqVNcuHCBU6dO0aZNG5unR4gPAIIg2EJbTxgdBH8kGAPZ4jQSjAqsncB2wYIFJbanpaWxYMGCepGWdTNXp6hzwa0g1LawsDC8vLzKnZn09vYmLCzMhqMyKj5TEBUVVSszBXX9A4AgCPWDJMEXPWHLRcjUlcy5ddXCgp62HZOlaVnh4eG1si5DURROnz5Neno6np6ehIWFVXocN3t1ijqXcysItU2WZSZNmlTuPhMnTrT5m1bhTEHxoLtwpiAqKspmY2nUqBEajabcfWrrA4AgCPWLnxN82av0nNt2XrafTSyelhXr241VveYS53tjgqG21mVERUXx3HPP8cEHH/DVV1/xwQcf8Nxzz1X670NhdYrC2fKi1SluBiK4FYRSdO3aldatW5f6XGBgIF26dLHpeOrSAq6srCw+/vhjDAZDufvVxgcAQRDqp6GNYVAjYxoC3Pjv2UxIzS/7+2pC0XSrXDs39oRNJs/egz1hU8i1cyt1P1uw1gRI0eoURRVWp0jOtc54a5L4yyMIpTAYDCQkJABw7733MnXqVB577DFkWSYhIYEjR47YdDx1pYJDdnY2H330EQkJCbi7uzNp0iS8vEpPdhNNUgRBsAY1I4lfjyay+SLIqICKm8bAl20SOd7rJF55STYdT2G6lQpEhj2IXusIkoRO68jesAdL7GcL1poAqevVKSwlcm4FoRSnTp0iKysLV1dXBg8ebLoFn5CQwLp16/jf//5H27ZtzcrV1aSaWsAVty2efQv30+2JLgTfFlTuvjk5OXz00UfExcXh5ubGCy+8QJMmTRg0aJBZftfevXvZsmULP/zwA2+//bao+SoIt7jq5ICqGUnoF0QQ7/Iodh5TuTN7Mzsdu/LxlXcZG7MeAL3WHu20v5E8bNM5snnz5tjZ2XHGszMJRVIRVElDvG9XYn27EW6IsWlalrUq2NTl6hSVIYJbQSjF/v37AejcubNZbumdd97Jzp07uXTpEps3b2bYsGE2GU9NLODKS89jz7y96LJ1RM7bS8P2fjh6Opa6b25uLh9//DGxsbG4urqaAlsw5igXfbMMCgriwIEDXL58mT///JMxY8ZYPCZBEOqXapfLykkDfQGvZnzBjMwl6CUt3kqG+T76AnYn5dHeGdxs8Fn6zz//JBNH9oQ9CKoCUpFAXVWIDJvMWLu/bJqWZY0JkBy9ceHeXYHw54W6VZ2iskRagiAUoygKBw4cACjx5uvs7MzYsWMB+P33323WZjUsLKzC2/xOTk40b97couOpqkrk/H3oc41LYXU5eiI/31fqvrm5uXzyySecO3cOFxcXnn/+eQICAso8trOzM/fddx8Af/31FxcvXrRoTIIg1C/WXgTrrmaXDGyB1z2fpd/eZsyxQbbYgQMHWP3770SGPYjBzsk8sAWQZHRaR+amhBIbG1vzA7quuhMgqgoD1sGz+6C7r7EKRWk5t65aCHQxBsJ1mQhuBaGYmJgY0tPTcXJyKnVR2W233UZgYCC5ubmsXr3aJmNKTk6moKCg3H0KZ1crujUFxnSEhF0XUBXjR3NVUUnYeYG4beb3o/Lz8/n00085e/Yszs7OPPfccwQFlZ++AMYPBR07dsRgMLBkyRLRtUcQbjHWXARrqCBU6Zl/EAmVazpjkFZTLl68yDfffEO6SxMSfLuilDEuVdIQ59OF2d/+THZ2ds0NqAhvb28kqXg4ak6WZfT60qNSSYKHmhsD11YeZVen6OAF807C+H+tMuwaI4JbQSimMCWhY8eOpeaLFi0V9u+//5oWntUUvV7P119/jV6vp0mTJiUWcHl7ezNo0CAcHByIjo7m9ddf5+DBg2UeLy89j8j5+0r9WB45fy956XnAjcD29OnTODk58dxzzxESEmLRmCVJ4v7778fe3p5Tp06xY8eOyrxkQRBucpbmgB48eLDcAFevQOsmG7nHdz7Jsnep+4zI/Zejfc4yv6cxSKsJOTk5zJs3j7y8PAY1VrkrZxMatfRAUaPqGZW9Ae9LB/jmm29q/MN9VlYWc+fOLbMBViFFUfj4449ZunQp8el5PL4L9l658fyjYXB8NNwVBONDjM0zilanGBMEb3SCJs7wUvuaejXWIXJuBaEIVVVNwW1hua/SFl21atWKrl27EhUVxf/+9z+ee+65Cj81V9Xvv//O+fPncXFx4dlnn8XLy4sTJ06wdetW+vfvb+pQNnjwYBYuXEhcXBzz5s1j0KBBTJgwwSxAN0tHKOVjeWF6Qq8XejBv3jyio6NxdHTk//7v/2jatGmlxu3j48Po0aNZuXIlK1eupFOnTri7u1vhigiCUNdZmgP6+eefo9FocHd3x8PDAw8PDzw9PU3/n5gnkaBtS7bkXGpKAhg/p7dwqrl6YIqi8M0333Dp0iW8vLyYPH4UY36ayb+N/yZTdkUtkpogqQpOah7zU2ezyD6Iw4cPs3btWu64444aGVt+fj5z5841jW3UqFGsWbOGg9rm7Gs+ie5nf6Sz4Rxjx47l7NmzbNmyhX/++YdPUlpw1Lsnx9Jhe4TxQ4FWNj7AvHlGhs6Yy7ygp7Gm8Km7wbFImfMzmRDiCnZ1aLq0Dg1FEGpfXFwcKSkp2Nvb0759e9Oiq7y0PCLn3ZjVBLjnnnvQarWcOHGi3JnS6jh16hR//fUXAJMnT6ZBgwbIskzLli0JCAigZcuWpkUL/v7+vPrqq6ZFbps3b+att94iMTHRdLyMuAyzdITiCtMTvpjzJSdOnMDR0ZGZM2cSGhpapfEPGTKEwMBAsrOz+emnn6p0DEEQbi6qqpKSkmLx/gaDgbS0NGJjYzl8+DBbt25lzZo1LF26lDO/f8mepDF8cfV1tJRdWzsuPg6Aq3nwzmHjjK+1rFmzhkOHDqHVapk+fTq78vy5z/cz3k7/xCywBVAlGQWZXMmRESNGAPDrr79y8uRJ6w3oOr1ezxdffGFaD/F///d/3H777bzw9kcc6jCVPAcPDnV4nOff+pBevXoz8f4Hee6554wNdqJ/omF6NEPSNlJQUPoHAz8nWNgbGjrCwl43mmUUDWwv5cLAdTB4fd2qfyuCW0EoonDWtkOHDtjb25e76MrX19cUSP7000/odDqrjiUnJ4evv/4aVVXp27cv3bp1q/B77OzsuPfee5k5cybu7u5cuHCBt956i61bt6KqKh7BHgT2DkCSS59llmSJfL88DscewsHBgWeffZYWLVpU+TVotVqmTJmCJEns2rWLEydOVPlYgiDUfadPn2bOnDn88ssvFe7r7e3NV199xSeffMIbb7zB008/zZQpUxg9ejQDBgwgPDwcb28vOuqiuSt3U7nHunbtGvqsVPr+DbMOwTdWKvl98OBBfv/9d8A4wRAY0ozpJxqx07Er57UB3JW9wZSeoFH1eBnSaWhIIVN2JTw8nL59+6KqKgsXLrRoPYSlVFXl+++/58iRI9jb2/P000/TpEkTVBWm75XJUTSARLaiYcpOmZGb4MUoaNu2LW+//TbDerRn6KEPiN20nFmzZnH27NlSzzM+BBInwLiQ0scRnWGsiZuWDy5l5AIoisKpU6e4cOECp06dsskaDJGWIAhFFAa34eHhpkVXhYouugruZ0xPGDlyJDt27CA5OZlNmzYRERFhlXGoqsqSJUtITU3Fz8+vwnbAxbVv35633nqLb775huPHj/P9999z7NgxpkyZQo8Z3bh0+DK67JLBuKIonM07g72nPc8884xV6jQ2a9aMgQMHsnnzZpYsWcLbb7+Nvb296Xl99iWU/PQyv1928ETr4l/tcQjCraI6dWWrKiEhgV9++YXDhw8Dxg/a7dq1K/eu1sSJE7G3t8fe3r7MZjBn929H98d27OSy80l1ioRvXgLq50OY3mcl32pCaetZrZcDGBeQff311wAMHDiQvn37AvBXlzjeX3+UN9PnkSG78a9jTzJkN9yUbNZdnkywPgkPNQtJgvvvv5+4uDgSEhL48ssveeGFF6xSH/3nn39m586dyLLME088YZqEKGybW8igwqbrBWt2JsNrHcHb2ZmHH36Yrl27snjxYi5dusR7773H8OHDGTNmjCmVzZLfo9v9Ye8doKjgUmSJiqoaUxsKS8EZ0yTuo/sPP9JJ/53lpeCqSAS3gnBdYmIiFy9eRKvV0iqkNRtmbDImcxV9T72+6KphB2NNWCcnJ8aOHct3333HmjVr6N27Nx4eHqbdqxq47dq1i7179yLLMlOnTsXJqfLN0z08PJg5cybr16/nl19+ISoqivPnzzN16lR8RzYgaeWlEt8jIeGuuvPA0/eXW+i7su6++272799PcnIyf/75J3fffTdgvD7Ja8eDUk4lCNkevxGrRIArCBaodl3ZSkpOTua3334jMjISVVWRZZl+/foxatQovLy8Sh2Pt7c3EydOrHA8i6ILOHsiGyf9SAxX4srcL8ug4fkGkVCQw8Nbx/HYXe9i7z+iWq8rNzfXtIAsLCyMiRMnmp5rUXCOr6++CoCfksrnqW/ynPcrfJz6Lh105lPGWjsHpk2bxuzZszlz5gyrVq0yO1ZVbNiwgbVr1wLG2eROnToB5m1zi/3ZwkEDG4aCt8ON7R06dOCdd95h+fLl7Nq1i7///pvDhw/zyCOPkJqaavHvUYtiSyl+iYUfz8EMl4P8sHCBsUVx98notM7sCZuC396XWbBgAdOmTauxAFcEt4JwXeGsbZvWbTjy7dEKF131f+02APr06cM///xDbGwsv/76Kw899BBQ9cAtOTmZZcuWATB69GiaNWtW5dckyzIRERG0bNmSrxZ+RcPTjfjmjW+54pBMe4eO+OT7IiOjoJDicIU0+1Q6T+hEmzZtqnzO0hTWvl2wYAFr166lR48eNGnSxBj4l3d9AJQC434iuBWEchXWlS2usK5sZYOJ8mbu0tPTWbNmDdu2bcNgMObC9ujRg9GjR+Pvf+PfateuXQkPD6/STPLX+1PYrwxnXsNDHEhMLrck2GsH7JkZ1pwWurPw23MYclPRdL/f4tda/HV//fXXpkVaTz75JN+f09LTV6X1ye9Q/vnUbP9xOesYl7OuxHH+vuLKC5GwbkhDHn30UebPn8+GDRto3ry5RWlmpYmMjGTFihUAjB07ln79+gEVt83VKfDRMVg1wPw5FxcXHnvsMbp06cKSJUtISkrinXfeKbXygiW/R9k6mB4JV/Ig5WIizSi9RXH/4wtYsWIF4eHhNXJXQQS3gnBdYXDbPrgDCd9eKHO/wvSE9LgMPIM9kGWZiRMnMmfOHLZv387AgQMJDg6uUuBWWParcLZg5MiRVnltzZo144Fekzl09DDeBQ3Y5budaPeT9ErxRlIlDJKBaPeT6DQF5OzKZvSY0UiSRNSX+wm5PRjfNr7VHkOXLl3o1KkThw4dYsmSJbz00ktWeGWCIIDldWUtDSbKmgEeO3YsSUlJbNq0yVR7u3379owdO5bg4OBSj1W8i6ElDGd38tTFX1njPJgxgwbQvdfgUmeA7733XtLS0li1ahUfnW7Ag40K6OMYj37deyxP9uUfz6F810eqVImwP/74w2wB2d5sD57YDc5aiNImEqJU3MFAQWJWjB/RmTDnCCzoFU5ERAR///033333HYGBgWYfAixx/PhxvvnmGwAGDx5s9vehum1zw8PDCQsLY+nSpezdu7fccRT+HkmSREFBAXl5eeTn55OXl0deXh6fhkh8HeNAwOnfiPPtXmaLYq7sq7AdcFWJ4FYQMM6WxsfHI8sy3Yd04+CJwyTsvlDyI/B1/p0b4hl8I/0gLCyM7t27s3fvXlasWMGLL75YpXH88ccfxMTE4OTkxGOPPWbVT7Rt7mxFUnQSm49sQicb822j3U8Qdq0Vp92j0WmMf6gK+49r4+w49ccZYjaeZ/T3o3D0cCjv8BUqrH178uRJzpw5w/bt2+nVvmG1X1dNKr4QorDsmiDUNZbWld2/fz/h4eFmbcWLK28G+NtvvzV9HRoayrhx46wenKi5GSh/vMo9OZe5t60nmjav0wjKnQFu2bIlCxcuZMlFSPFU6eRjYGrKAAquSowPVhkRaFl0e+jQIVNzngcffJCmTUNwy8rmdn8XQt0kmnecjhIYiLLlUzCUPXkha+34sXsm31x04u3Oxm1jx47l3LlznDp1is8//5zXX38dBwfL3ldjY2P5/PPPMRgMdOvWjYkTJ5qVn2zraaxL+0dC1dvmurq6MmDAgAqD29TUVB5//HH0en2ZtXWDwJiOUE6L4obp0RaXjKssEdwKAjdmbVu2bIm7u3u5i64kWaL3//Ussf2ee+7h4MGDnDp1iv3799OhmWulxnDmzBn++OMPwJhH5ePjU4VXUjZZK+M+wJXk05dN25KdLpPsdLnEvunp6XS5rQtJ+y/SsL1ftQPbQg0aNGDMmDH873//Y+XKlbRv+pBVjlua0uoTV0bxmauoqKgazV20lpCQEOLiSuYnPvnkk6aAZffu3bz66qtERkai0Wjo1KkT69evr1Jut1A3pKen463Jx1VT9qxilkHLF198gSRJuLm5lagp6+Hhgbu7Oz/++GO559JoNDz55JN07ty5Rup7G/5+G65dhgYhyIOfM20vbwY4ODiYWbNmsWLFCv7aupVrhsu86LoQGrVjYOOBFp23xAKydiEYlkzGy9WHv++ei6KCpPFG0/sh5LbDIMf8w4Ry+QzKmpcB0NzzBS2aNOS/TW48r9FoePzxx3nzzTdJTEzkhx9+4NFHH63wGiYnJzN37lzy8vJo3bp1iYmP+Cy4nHejLm2mrmTObWGd2opYGmwWrw7k6OiIg4MDDg4OODo6YlAUfvS463o6QuktiveGPYinZ+X+TlpKBLeCgDG4tVPsCMtpiWJQcPR0pMeMbux4f1eJfbtN64pzA2fAWNXgyokU/Nr60qBBAyIiIlizZg0//fQTbZ6fbPH5i5b96t27Nz169LDK68q/lk/8jgSaDw9FkqRK9R+3c7bjtpf7mI/zag66HD0egVVvxjB48GB27dpFfHw8Gzasp5+b5d+rKgaQ5Ar/GBTWJ9Zl64ict9cYoHs6Wnwea+cu2tK+fftMOZAAx44dY8iQIYwfPx4wBrbDhw/n5ZdfZv78+Wi1Wg4fPixmpG9yPnY63g44VGFVgdcTO5OqtyczM5PMzMwKOyzG+nYzNQMIvhIFGOvSOjs710hgqxz7i/TjW/nLZTQjR96Pj53lH7gcHByYMmUKbdu25fvvv8f3xHYuR5/gQIAjvXv3Lvd7c3NzmT9/Prm5ubRo3pwJLST2f/sCHXMPgZ0z2owLaL0CTPtLHo3Bo7HZMTSN2qKe3Yp6Yh3K3iXIzW+8f6qqcZHVkMaePPHEE3z44Yfs2rWL5s2bM2BAsUTYIjIyMvjoo4/IzMwkKCiIGTNmmDXmOZsJg9ZDjh7+HW5smztpm/kxVIzb/Sy4lJb+jXjsscdo27YtDg4O2Nvbl3j/OHJV4f0/y35PKUxP0PnVTFkwEdwKt7y0tDRizsbQI7UXWck5HPr+MOGPdCa4XxBx2+K5sCcRVVGRZImAXk0IG9Hc9L1n1p5l7+dRhI1sTvfp3YiIiGDbtm2kpKSwe/duWls4hqVLl5KSkoKvry/331+1RRDFqYrKzv/uJinqIteSrhH+SGfCwsLw8vIq9/alt7e3qQRY0T9eil5h+3s7STufTr9X+tC464039srMkmo0GiZPnsw777zD0aNH6Vf+3xwzeYn/knl4Pk6Bg3EMHISdVyuzMeqzL2HIS2P3p+fR5xpnFnQ5OnZ/8g99nmlmUVkxa+cu2pqvr3l+9Pvvv09oaCj9+/cH4Nlnn+Wpp54yy3lu2bKlTccoWF9Tfy+UcgJbADtZ5f03XyTHLZiMjAzTIz093fT/CQkJXLpkrKRivK1cZJV7+imcdNcAy2f4Kks5vYX1Tv14xGcO7Q7DoZDKH6Nbt240a9bMuH7h9GljScQjB7m3UQou3cchAYqiEnvuHLmxBzi/W2VfVBR2V8/SsYELjzeO5uM9jrzpu4z/Sit4etTtSEUC2/JoBj6DPnoT6tntKOd2IzfrBcBbh+HtwzCoEawd3IqxY8eyatUqli9fTkhISKkdIHNzc5k7dy5XrlzB19eXmTNnlri70sjJ2A73mg7c7Y11aVfG3khPKExHGG/hdbT0b0TPnj3Lff9r7y1zu1saWzPdUaWSKTCSYqC/RybtvcvJk6gGEdwKt7z9+/eDBLnNc/DPaEToUGM3LkmSzNIT7Jy19JhuvsI1NzUPJHAPMM5kOjo6Mn78eL755hu2b99OawsWxB49cpg9e/YgyzL/+c9/rHdrWILGXRtx5fgVQm4PAYy39R4YP5yf//ddmd82btywUt+0dLl6ZK1x1tSt8Y3p1qrMkjZr1ozBgwZwfO9flXpJeRf+xZB9kazopWRFL0XjGoBT4GCcAgch2bmR/Pc9XI4JITHqRhkgVYHEfRkc//odGobGVVhWzNLcxZpaCGFNBQUFLFu2jJkzZyJJEsnJyURGRnLffffRu3dvYmJiaNWqFe+++66phmdp8vPzyc+/0cUoMzMTMN6atHbzkltZ4bWs0jVVyu7eZUZVcXZ2xtnZmUaNGpV4+tSpU3z88ceolL3KHYz5mTXys7/jPbS7DtE+XWFYYxWdrmoze+7u7sycOZO1a9fy4+Z9fKfvQmTCBb44bAxuAUKuP/h3JU0Bmly/nZ8AZ33GokgaHLpNQO8KWPpa3RpD53tg/3IMmz7EMGUFSDJjAuCzE1r6+yno9QqDBw/m9OnTHD58mAULFvDKK6+QlJRERkYGHh4ehISEsGDBAuLi4nBzc+Ppp5/G2dm5xDW3B341fm6lgT3o9fBZV9hyUWtsm6uFT7vqLR4+wIQJE1i4cGGZz99zzz0YDAazO0SlWTLElbarVbIMJXNuXbUqS4aU/ztUnd8vEdwKtzSDznCjBNjw1gweMBiN/Y1PmY6ejvR8qrtpVrJ40NbxgfYE9mqCV7Mbnz47tOhIaHAouWnHzPa9fL4FZ/beTliPf/ELOWPannXiW7RyY0beeRfNmzfHWiRJotVdLWk6IAQHd2POrD77Eg0vvM203uVUcbjwNvrsDiUCQAc3ewa9N4CM+ExTcKuqKnvm7S3Rxa2wTFppVFUl+9SPDPLZSIyzJzqDhJ2mnBkn2R7ZwRMAj26v4hgwgNyETeQn7cCQdYGsk9+TdfJ7NM6NKMjWcGrXQIx/ooreMlU5tXsQXv5LyiwrpqoqCQkJrF+/vuyxFFFTM1fWtHr1atLT05kyZQoA586dA2DWrFl89NFHdOrUiR9++IFBgwZx7NixMrvRzZkzh9mzZ5fYvmHDBpydnWts/LeqjRs3Vvp73HMSKfvjyQ07duwk0zm2zOdVVcXR0ZFotw5lrnJvnXWMs2fPEhMTU+lxWsIRmCX9gSFRYm1S+bPRFZFlmZbhvVit6ckRJZ830z/DXym7NbAEZDj6M6HBVUI0O2l0OoW1Z8rcvVT2+mb0lx2wu3SSQ8vfJ8mrEwDznOxwi9Ox7npKfJMmTTh79ixXr17l+eefN+vcJcsyiqKg1Wrp0qULUVFRpue2FQSgAv3ty67q84hdY77Tt+cRu6NE/ZNUuRcAdO/enSNHjpCXd6PlvJOTE+3bt+fSpUumOrsV+Y9jYz7JKTbLI8k85ri/wnHl5ORUetyFRHAr3JJUVeXkL9Gc3XyOs7lnQTKuxC0a2BYK7hdk6khWGu/m3qb/VwwKOz/YTcvLbThol4RBOY9GhoJcJ07tGoRe50D0rkF4NryAvVMuqgotGmQyvZ9M68G9rPLaclJycPR0RNYaPykXBrZAtevKyhoZr6aepq9P/hLNhd2Jpq9L6+JmdmhdFul73ybvwr8AjB16H3NXZuHmqPL441Px9fUrec4iqQSy1hGnwIE4BQ5E0eWQf3EHufGbybu4C2QHoncPQq+3xzywBZDQ6+w5tWcgjUeZP3PhwgX27t3Lvn37TLdjLWFpblptWrRoERERETRubEwhKfzjOXXqVFM95s6dO7N582a+++475syZU+pxXn75ZWbOnGn6OjMzk8DAQIYOHYq7e9XzrwVzOp2OjRs3MmTIELO8SktcOrQFLAjC+vbtA/7l17F29A3k+5Otylzl/mLraAZ171Cp8ZXr6nnYuxQGzgQH6y8wGgG4H8nFffv8cgNbPRq0GHC75yMGNOlA2ZmwFvDJgIOr6NS1B51aDirxtKIa82SdnZ35+eefS7SkLfx6+PDhjBp1401r+2WJzzZr0Uoq9/brQGdvSjUCeBeATtcflacoCmfOnDHNJrdo0aLSqVgRKsRsV/gzUcKgSmgklTsDVN69reJxFd4hqgoR3Aq3pIJrBZz4NZq8tDwauvujba0pkatYFTkpOeRcycaQbSCkbyfmbs9gWHsVbewoDNfrsxj0jpw/9wKG8ASORK5nVNtL+Dqmk/rPI3j3/RB7b0szdUt5Xdk6Nr38D45eTvR7pU+lFlFVVl56Hge/P1zyiWJd3Arp0mNI3fkihqwEkO3w6DwT/2ajaXpkPocOHeL7n7czZswYMjMzLSr0Lts54xQ0FKegoSi6LC7v30VK/NWyB6zKXIlrQcaFXPT5F4mMjGTfvn0kJd2YPbCzs6N9+/acOnWK7OzsMg9VNC+5roqLi2PTpk38+uuvpm2Ft6GLN+lo3bo18fFlF8ksXAVdnJ2dXaWDMKFiVbmuUds3cocF+2m1WqRyjq2q8G1BFxS74nc/AElGb+fMooIuDLfSj1016DD8+Spq0jEkFC4NepcmzlSqLq0lXuxix7n8NrCpjHEAk30+xNeQylNJl2kZUr0XqPZ5BHo/jGRX8j34ci5M2QF2kkro5s3lHmfXrl2MGTPG9F54exO4rxk0cJDo6meHbP01fagZSaZKEG0bSNDA0/hE6lnjf529jAvqLPRlb/j3N5UMnYqbFr7oJWNnV3GQXJ33FhHcCrckB3cHBszuz8oFK0lMv8DYrmOtclzXhq6M/GIEaTFp2AVoefnl/fyz1Yv2GTdurRXmgB47G8tlBw/6RkzBPXs5+szzpPwzFc9ur+IcPKxK58+ISyf3ai76PAvz76pIVVUi5+8rvQ5wKV3ccmLXkRE1B9WQh8a5IV6952DfoC1g7L1+7JjxNueHH35oOkxlym7Jdq64BzdBki+hKmW8IUoKvkExrPgzkoNnMkybtVot7dq1o3v37nTq1AknJ6cyqyUUmjhxYp1cTFbU4sWL8fPzMyv0HhISQuPGjTl16pTZvqdPnyYiIsLWQxSsJCEhgYzYY2CF6oE3mgGUHjUpSOU2A6gsZftXqEnHwNEDqf9T9Ll+t3vtYOscv6hrWVkAJGj8aaCks9bpdlPb3EaGZH51icBOLeC2tC1Ud4mlpC27fOLlXNh2CSRU5AJnvCg7xz81NZWTp07TplUrJAlkCb7rY/xvDRSrQM1IQr8gAvTl3OHT2qOd9rfFAa6fE3ze3cCMXXo+76HFz6nmQ8869e4cEhKCJEklHtOmTTPts3v3bgYOHIiLiwvu7u7069eP3NzcWhy1UFfFbYvn50m/EbfdOCOVm5pLasyNNxGnJo7sz4oCydg9y1rsXezwbHQRLw8Xhg+IoHVmW9RiUaCKSsv01vTs1JNut92Jz+BvcWjcFwz5pO95g8xjX1fp3L5tfBn+2TD6v3lbjc7aZsRlkLDrAqpSej5c0S5uWadWkB75JqohDwf/HvgM/cEU2AKcP38evb5kbc7CsltFc83Ko7WXCem4FySF0hpQau0KaNnzHy5fvoRGo6FDhw488sgjfPbZZzz99NP06tXLtJiva9euTJs2DS8v87+u3t7edboMWCFFUVi8eDGTJ09Gq73xh0SSJJ5//nnmzZvHzz//zNmzZ3n99deJjo7mkUceqcURC9WxZs0azudbVlNPzS3/Vm9hMwBNGYGTRoIxQVYKbBOPoGw3LlzSjHidWKkh6QWQXgChNZDt4ubmxleuE+nYeC3veExjuvdsLss+TPeeTXNdHD8nP8H8q7Pp7l1xBzJLqYoB5eAvGHbeaH7RwRu+vw2+DjqKV/aNtK5Y326s6jWXuCK5zgZZy/QTfrx64MYxNXLNBLaAcca2vMAWjM/nlL/otrhxwSqLPNYzNqh6OdSWqlMzt6I+o+1Vt9B9XR1P8RX8Lr7ObJ+zC0OBnuFzh+Lq78qhQ4cwGAw0bty41FXDVaXPTiJ161NIjr54nPgPWWoOUrFZEAkJraolOK2l8UOcnSveff7LtaMLyYr+AY1T5VIkVFU1lcQqqwatPjOO9MiSi4JKU5ByFH36GRwDbke2L/lH0yPYg8DeAaYyaaXxbeeLZ7AH+uwBZJ1cgnPzu3Fr8wiSfCOv2ZKyWz/++CMNGzakoKDArMVjfn4+ORm55OfkUSAXkJdygiEd9uHoco2TO4YXO4pEy16bsXfKpUfnPgwc9xCuruXn9nXt2pXw8HBOnDjB1q1b6d+//03ToWzTpk3Ex8fz8MMPl3jumWeeIS8vj2effZbU1FQ6duzIxo0bCQ0NrYWRCtWhxO7lco5EVFQU3hotqsYOyVD+CnPDX7OQpixFciuZ366q8OJ+mNK8+s0AKqLqcjGsfhFUA1K7kcjtRhIKXL4XTqSDY9kN1KosOCgYn50byZWdWOY6mizZmP9wTXbhae83WZHyNABy0ESrnVON3Yvhj9dAY4/cNgLJ09jVYXwIROc5ULh0sKyya0lebdme5c3eE/BwC2guUtwtUqeCW1Gf0baqW+i+ro6n8JZ50RX8x/53AntXO/R5kikYK6ySYM1ZW1VVydj/Iaohj7zsIJL2XS0R2BaSkEiKvEh6XAaewR5Isgb3jtNwDBiAfYM2ZscsXiw9fkcCWUtySHC/gG9LH3Z+uJteM3uWCGxVVUG6viBEdmqAPtuyVbM5cX+jTz0O+9/Hwb+nsa5sk9uQ7YwBoSRJdHkkmEuHLqLLKZkCIWugz7PGnFStiz9+I1Yi25d8V7ak7FZ6ejpvvPFGie0uehc6pHUiXy7goHcUjTxyGOoL/qHRXIlvTkpCM1BlUzpCw6ZnSLsYwLV/fLjU6DLNh1W8cEWWZVq2bElMTAwtW7a8KQJbgKFDh5bZFhPgpZdeMnsfFW4+yskNGH59DjvVGTe5Oc0698Hu/o/KnFFTM5IwrH0b0uLRL3kA7QOLS9xWXngKPjkO356G/3aFx3cXOwbwYnvLmgFUOP5NH8HVWHBriCbiddN2Rw2EN6j+8UsjyxJ356znhfSF/NfzcdN2g6RltctQfs4ZzricdchWTGSVmvZECumBGhuJ4d95aEd/YHqusKZsbJ4d/7Z7ptSyax3VBO7oqtCpgSwC20qoU8FtUaI+Y81SVZXdn0WaBYB75u2lz8tVW7FfrfqMVh5P/PYEEnbdKJGiKioXIhPp8kRnAvsE4ODuQFZWFkePHgWgU6dOVvtdyE/cQv7FXSBp8RswlbRfNuFxzRO5jAwgnazDyd/RdH59vgGtewvT10p+Bhm7nsOl3RNoXBqj5GeQl6ln3/xTqLkq+z7fi2ewIynROexdsJPb3x5sPE7GOXJjVqHPjMWz/xfXg2MHXNs9QdbhuRW+Dq1nS1R9DobM8+Qn7SA/aQfI9tj798QxcAhar9Zk7ryPsG4hHN86osT3t+77F5k7Pkc77H+oGl+OLDlFm/GtcPQy/7By9Wo5C8CKcHBwwNXV1dTa0cHBAWeDC067nHGyc2ZInyHopUvoDLHYaVRa9drMnosB6HUOpnQEgOT4UFQdJOy5QNCAQIs6LFXld/tWfm8Rap6yfyWGtbNBVYjJ1pCrahg1alSpXbMKSY3aIjVsjX7pQ5Aaj/776wGu9407ZJObw5oEuLcpPBgK6xJvNAOQMAa3l6yQBajmpqOc2ACAZtS7SE4e1T+oJZy9uGLvz1duk5BUBbVIJQhJVZjhPYt++kM0drZesq8kSciDn8Pw7XjUI3+g9piM1Mg4eSHLMndNuJ9hx1qj1974xFC07Nq0e7rTte3N8aG6Lqmzwa2oz1izdGf15O25kVejKioXdify+7w12DWv+q9FVeozWmM8qqKCCmoBZK8o/d13/6KDROdEIztJJCYmotPpcHZ25siRI6ZAtzo0ah7t8r/GHkiUe7L29+0cdTxCr6w+SKpkNoOroqKgcNo1mqVLl+Lr64uqV8lamovGV8ZpsAOSo0SQbgMNDSdJ2/40ICGpCke33IE+rxkgo8vVo1w7hl+InmYtdxC5Zh8NlON4KLGmc237axHZsvEPnr1aQHs0yJS94ExBQ2RiIwqkVjjaX8HbcBJv5QROSioFSdu4cimeC9qBtFUK8As5zeXzYaXOkqLC9n/WkrrHG90RPTG7Y3Ae52gWUF65csWia9u1a9dSq1noPQ3IXjKyk4SqNuWLf9uiVY1VDtx8juKf3J4LDY7id01DZoone9UUOvezIyMsjb///tuicxeqzO92deozCkJZVFVF2fEVypbPAIh2bMdX510I79KVwMDACr9f8gpAO2WpMcC9GntjBtenGQDOWvhrMKYV+F/0NKYnZOiM6QgfdYWHSv9TW/p4i6y6L04z5gPUiyeRQ40taj8/Cb/Hw4zWMKqmMuTcG/N0+FqyLjmiquYfbFVJ5prWnWfC17LKw0qNdK6TG7dDaTcS9dhfGDZ/jPb+RabnWncIR3NKQa+o5om0qsrB9v8hqK1twzQ15bxNz1dT6mxwK+oz1py89DzWLl1/46N4IQkMu1WGPTiw0ukA1anPWOZ4AP1OxTSenKu57JtnXFzUf/aNJgF750dxfmMs7R9sR+rpVHL0eSUWcAFIegmv0970ebkX335rTO7v27ev2Wry6rh26GPyzmWhcQ2g46DZFOw/zM6dO4l2P0H7jI7mY0HihMcxkp0uc0frkXTv3p1Lhy6zNX87DnkOjBg9AkmWUA2DSPr7E+SsjWjtC7gcG0ZKfJFGD6rM1QvNaNv/L5ycM2mmL+z4JWPfuB/OzcfTr0EHs4DSkDMAJT+DssgOHgx2Nq9xq6oqhoyz5F34B3evMIKdG5O+5TskiTJnSQH69OlDdtuGRH66jw4PtqNRF/PcZkVROHHiRLmpCV5eXjzwwAPkXc1j77wowv/TCfcy8oobNWpUpLNOGqd8t4EKpw4EAhKPPz6V8PBws+9J2peEf7g/sqaM2fUq/G5Xpz6jIJRGVRWU9e+j7F0KQE7n+5n721lUMKuDWhHJ3R/t5KXolz2MmnyGF1ftonUvdx7qZCy1UPSOvJ8TLOwNz0TCZz1gXEglxmvhqnu5XQSSR2N+joUdyXBXDS79OJ4Oqy+WHbgaVInfLjpZrRJEUZoBz6A/sQH13C6UmJ3IoX1QVXhyD+iRSynNLZGrapm2B1ZVq+CuZVRVQYn8AWXTxzV/Mhuok8GtqM9Yc1RV5cDCQ8bb/yUXlKPP1XPgq0PldpgqT2Wva7njAfR5N8aj1ei4dPAykkYy1mq8HrDZORjPl3cll8Q9ZeeUFs4GZ8RncOTIEcDYhcUavwcFKUfJO7caAM+uL2Pv6EqDBsbEsWTHyyTnXcYn3xcZGQWFFIcrJDtdBqBBgwbY2dkR2C2AUd/eQXZyNvYO9sYD29lxdH1Xrl1oTuu+ayvovnUBezdwCb0blxbj0bqUvkjOziMQqHiWpwTfNjj5Gv/9FaRGmzbbO+XSsvdmU/c1e6cbM+darRbfMF9GzB9uFjwmH0umIEtHQM8mTJo0yVR2yy+3IWHXWnHaPZpkR+P1mTRpEg4ODkR+v4/Lh5OJWnCAoR8NLjWloEePHmg0GpYvX24WMHt7N2DixIl07doVVdGj5F1F49yQM2vPEjl/HwG9mtDv1b5lBrhQud/tW/G9RahZyrYvTYGtPOwVfjqci6qepXPnzgQFVS4ilFx90D64hN+WL+QTu/uRj6j0bgotS8kOGB9ifBSlV+B8FrQob36oMqvuPRrzbR9jSsToGgxuCytBFKZaFKeRYFSg9QNbMM6ay90moUQuwbDpI6RmvTmeLl0vu1Y6g4pVy66VRc1KwfD7K6gx22vuJDZWJ4NbUZ+x5hSWcCpL0RJOnsE1nwdV0XhQMY3HrZErvf+vJ9piNfI6PtiBDve3R+ukIS8jv8wV/JIsEdCrCYlZieTl5eHp6UnTpk2t8jpkpwY4+PdEdvTGoaGxjEvhYoG0tDSi3U/SK8UbSZUwSAai3U8CJZsBuDdxw73JjeoE+dcKQFFRVYlLMa0r7L51+1ujcPDrZJXXVBkNm54xpiKUoWjQqMvRsfOjPWRfzqb3//Wk62Bj2a2VP6yk1eU2aFUtrTLagJ/K+AfGm8pudX+yK4YChW6Ph5ebK1tY5eD06dOkp6ebNYTQX0sgbferqAYdvkOX4OjtiGwn4xnkUW5gKwi1Te46EeXkejR9/kOyb1f2fPUKULlZ26IkZy9G3zeN6Yf0hHlpSw1sS3M6A8ZugWw9nBhjvaoGzd1hZtuK96sOSbqRalGTlSDKIt82FfXqeeTbHkeSpFoNtosybPrQGNhqHZBuexx125dgKOeDicYerJiXXBPqXHBbUX3GN998k44dO9KpUyeWLFlCdHQ0P//8cy2O+OZSUQknSZYI6NkEFz/b5CO7B7lj72pHQVbpC3AKA9LCQLvZ4JLBqIObven/e8zoxqXDl9HllHznsnPW0mN6N3785UfAWCXBWqvftS6N8e4316y1rSzLpllJnaaAaPcTpllJnca4X0XNABzc7LnjqxEkbN3Htg/K3M3UfSvzkkopHWzrFFkrE9Q3kISdFwjsHQAYfxZZa3O4EJMIKthJdgz3HIFffkPT9zl6OjJgVj/LziHLtGrVqsR2yd4NQ04ySn4a144vIrDnE4z8IsLsA4Ug1BWqvgBJa3x/k1y80f7nVyRZyx/ffIOqqnTs2JGQkJDKHfP6+6Ikgezkzqe9CrerKP/MRWrWB7lpjzK/P9AFrukg12C8zd+lhiob1BQ/J/iyF0zaZr5dxbjdGpUgyiI5e6Gd9NWNr2s52C6kGfIChqwUNENfRPILQ+0wqkSutKrPx7DqGci6gtRxdKU6lNWGOjdVUVF9xpdffplnn32Wjh07snnzZlGfsZIkSaLHjG5oS/u4fT0A9Gvnyx//WcuFyMSS+1iZLMv0faUPUmkVw4sEpJZy9HSkx4xupaZc9JjRHTs3Ow4ePAhglUL8qnKj2LckSUga8xSYos0Akp0us8NvK8mOlyvVDECSJBq2d8Mn6Oz1BgWl7aTgG3wGj4AafGe2Eo29hi6PduaOLyOwczbevo/bFs+F3YlQ+PIUSNybxLZ3d3D5SLL1zu3giUfXFwHIil5KQepJPALcTTPBqqJyYNEhsi5lmb7HVHZtRzl3GAShitSMJNSLx1EvHodLJ3DPSYRLJ1DObEX/xUgMO78x7SvJWi5fvsyePXsAuOuuuyp9vrcOw//tuxHkmsZx+DeUnd9gWDEVw+HVpjEVfzheOc6qblc4dXf5ga1h92KLxpNjkHgmEjYlQRkls61ufEhhowrjCTWSypigkukXNU3V55uC7VL+ZNVYsK2mnMewfaHpa8mlAdr7FyH5Ge8iSh6NjdU1ijzkwHA0d7xl/P5Dv6Imn7b+wKyozs3civqMNc/R05FGXRoTv71Yso8K3ad348Qv0eSk5JCZkAk9mlj13Hnpeez/9iAN2/vRfJjxQ0njzo3o83wvdry/q8R4eszoXunFbcH9gozB0vXZ6cLZ3+B+QZw4cYLs7GxcXV3LrLBhKVVVSd3xPBpHH9w7Tkd2KP2+Xnm3yS0lSZLZwq3iObc3FnLdU63XZEtaR+PbT156Hns+jSxzP8Vg3VbCTgEDyAsaQm78RtIj38J36BIkjXF27OiK45z4+SRx2+IZ9c1IdDk6ohYcQM2FfQv207hzo1qtBS3UL6UtuuoLcGa+qZ6JsvkTpJaDkX2Md63+/PNPFEWhQ4cOlU6rOnAV3j5s/P87A2FAkbR8qd1IpOhNqKe3oPz+MmV8jAag0/X2q9g3RlUV1KRjqOd2Ifd57EaTloryba/bdNWVz6ONt+bPWqcLeoWKzphm6FTctLaZIS2k6vNR/v0c5chqtI+vYXyIFytjb6QnFKYjWDvYVlUV9dAvGNa9B7pcJO8g5LYlSzmWRQ67HaXVYNToTRjWz0H7gGUfYGpDnZu5FWzDI9jdOFt6PUaSZInAPgGE9A9m6H8H0eU/nWk15kaDjPzMfBRDeW93londGsf5zbEc/O4Q+rwbs57B/YII6NXkxm+kDIG9AwjuV/nVBabZ6eu5uUVnfwtbuYaHh6PRVC9ZLC9hM/kXd5ET9zeG/PIbERTeJu/ZsyetWrWqUjqEceHWP5SWc1vYfctWZAdPkO0r2MneuF85TA03CkoPYCVJ4vRfZ6s2yHK4hz+H7OCFPvMc147fKMvTfHgzPJt60umhDsh2slkzEH2unsjP91l9LMItzJJFV4CkM5aWS05OZtcu4yRARbm2q2KhyU/wc+yNbeEN4Ove8EEX88AWQNI6oBn/KTS1oLa4vgDlyB/oV79I5GePcf77Z1C2fIaaeMS0i9zOsqAp0FHHwy2MNXZrrKVsKfyc4PPuBjylfD7vYajRdIQSNHYoMTsgKwVl+1emYNv1+nRjVdMRit4FKP5QYiPR/+8JDH+8bgxsm/ZCCqr83UvNsJeRwgaiiSjZWKcuqXMzt4JtdLyvPS2Gh/LH1LXosnVmAaDWUUvrMTfyFVVVZdu7OzDkG+j1XE88AkoukS3aMavZAPPZhKIdtsLuaMHV06m0vLOFaeYOjN3C1qb8SWu1HVq06FU9f135A5copyqlDzh6OtLzqe6mVr6Ono4oisKBA8YG3dXtSqYUXCPj4CcAuLaejJ17SLWOV5HCYLLcurIWBJPWYuw6tgolP73cMWtd/Mt8HmpvgWNhekLazpfIil6KY8Dt2Hu3xrmBMyPmDUPWysRujSvRDCRh5wXitsVX6UOXIFTXX3/9haIotGvXrtx0vORceHyXsT7t47uhjy80cjE+93A5N6wkjT2agc9gWLS77J2uU7Z8ysfuj/Ca1wdMcPybJQ3Wme6AAEhellVk6eyex9e11Gh0XLCK8/H1jAiyfPbSGiRJRjP4OQw/Poqybzly9/vx8wowK7tW2WDbotJrAJIGeeAzyL0fNnWvrNTYPRqjvXdBpb/P1kRwewtzbuBcIgAsTeaFa6SeSUVRVDTakv8Y8tLzSr11a9AZOPHzSZKPX2Hg27cbO7VoZPo8bz4zEBUVZSoFJRVZdHXl2hUWLFhgcW5qccH9gsyCkJiYGDIyMnBycipRUq6yMo98aSwn5RaEW+sHq3UsSxQNJnv30PH3cyfQ5Riwc9LS68UxOHrcY1Ewae0xUc3zWbTAsciCQmtyChhAbuBgDDmXke1cTNtlrUxeeh6R8/eVWgs6cv5eGnao3VbVwq0nJSWFnTt3AuXn2hbWTs26fmMsswC6/AkxY8HJgr/4prSCingFMiTUjzeuqNi1HYbUJwKzWMnZC7T2Fda5reur7muKHNoHpVkf1HM7MWz5FO3dH5Vads1iFt4FkEe9i6Zj5XO1y6JmJNXJxWUiuL0F6fP0plnT4gFgaTwC3bnz65FcPZuKq7+raXt2cjbOvs6l3rrt/9pt5KbmceynExjyDSTuSyKge8n8XUVRWL58uenrZKfLpvqvhVasWEF4eHi1KxsUpiR06tTJrBJHZRWkHCUnxliD2bPrSyUWkdWUwmDS3hu6Tndhz/xIuk3vjruVypnVhsIUkooqXNQUz26vImkczP6gm1IlyqgFrcu58TsuCIqiVCuf3lJ//vknBoOBNm3a0Lx58zL3WxWLWe1UBUjOg5f2G2cErUU7bi5dGrXlbBYEuZbMJ5A8GhvzcsvoUAawK9ePYK0vAdYb1k1FM/j/0H+9C/XYXyg9pyA3blfj55T9yv7dqQxVVVD+fgdl/0o0Dy1DDuhkleNaiwhubzGKXuHXB3/HPcCd/q/1xcnbsnsfzj7OOPvcKA+WeSGTP5/8mwYtvLlyIsW0vfit267/CUfrpKVJt9I/2Z0+fbrc7lQAqampnD59utTSTpZSVZX9+/cD1UtJUBU96VHvAypOISNx8KteekNVBfUN5FjmUQL73vx/FgorXFhrQWFlyHbmJe9UQwEZCTl1qha0UHdFRUWVaBji5eXFpEmTKrzbpKoqat41i86TnpHBjh07gPJnbZNz4YndpTZ75Mdz8GoH66++D3It+znJozGUMaunqvDwbxBzDf4eAkPq3uRfjZP8WyN1GIV65HeUjR8iPfh9uTW8y1PeQvyaIEkyqi4XVAOGtW8hPboSSa47IaVYUHaLSY1Jo+BaAdeSruHgUfUZx0tHklF0CimnrpZc34Tx1m1eeh4tRjSn6YCQMv/BpqenW3Q+S/crS1xcHFevXsXe3p527ar+6Vh/LR4l7yqyvQfunZ6q1piEG4L7BRHYOwDpeu/PwgWOtsptVfV5ZBz8jCsbH8I9wMlsLMUVjk0Etre2wnSq4h/O09LSWLBggelOUWnUvEwMv8zEsOYVi861Y/sODAYDrVu3Nmv6YnbMIukIpYU513QwbY9Fp6uStHz471GwdN1xhg78HMFJA718a25cdZ3cZQLIWtQLB1FPbS65GCyj9K6bqkGHcuEQhl2L0K94AsMPU2w7cEAz+Hlw9IBLJ1H2rbD5+ctTd8JswSZ8WjZgzA93cS3pWrU6MrWICCV2axxXjl0p9Z3U0lu3Hh6WBQhubtUrsl84a9uhQ4dS2zFbys6jGX4RP6HLjEVjo8VbtwKz9IRiCxxtQTHkkRv3N0p+GlknF9NjxpRaS5UQ6r7i6VSlKSudSok/gOG35yDjIpbOLxlrczuWWyHheDq11srVoEDPv4yzsH6OMMWCKoue9rB9hDEn2PUW7VatZiRhWDoFrtdLN6ycUXKnwrJrzl6gdTRNFBlWv4R6fK0NR1uS5OKNPOhZlL9moWz5DLnNMCS3utFFSMzc3oJcfJ3x79iw4h3LkRGXQfKR5FIXAYH5rduy6PV6tm3bVubzRf32228Vpi+URVVV0yyKpSkJ+uxLFKRGl/rQZ19E41y96yeUVFjhwtHLkR5P1Ww6QnEaB088urwAQNbJH5CV8+U2AxGLyW5tlUmnKqQqegxbF2BY8oAxsPUKRL5nnnFRVTkMkoaMAomWLVuWm5rV1tNYG7UsGgnGBFkQ2BYuBCtPsYVgGhmmtoR2nuWnKZTGvYJT1WuWLALTF6Bf8QT6D3pAapxpsxTQCZw8kFoOQh7yIvJd79fsWMsgh49HatwBCrIxbCivjaZtiZlboUqqu8o9Ozubzz//nOjoaCRJKjdfyN7enpiYGGbNmsXjjz9O69atKzXWpKQkLl26hFarpWPHjhXur8++RPLa8WatdEuQ7fEbscqm1QluBZYscKwpToEDyQ0cTF7CJtL3vk3Q4MVmv+NFm4EIt7bKplOpGUkYfn0BNcF4B0lqfyeaEW8gObgiF1l0pdfr2bFjJ3379kGr1ZKZeY135y4g1aDl0Qq6kUkSLOwF/14ytnI1ew7La6dashAMZ68SK+RntIanWxsD3YrkGUArQSnFd4TSXO8GpiYeRmoQAhjTGeTu95nKeakXj5fbeKOmSJKMZuQb6L+9B/X4WpTOY5Gb9a6FkZgTv1q3kJiN59j9aSTJx65U+1hmjRJK9hQo99ZtSkoK7733HtHR0Tg6OjJz5kxTi9qiClvUvvXWWwQGBpKZmcmHH37IX3/9Vank+cKUhLZt2+LkVPFqCiU/vfzAFkApKLfGq3Bz8ihs7pARcz094UYzEK2TSEcQjDw9PS3aLzfX2FjFsOG/xsDW3gXN6A/QjvkvkoNxirNoq1P825Dp3AT82yA1asufUTFcydfSokWLcmdtC98O/ZyMAW6J56lcK9fS2q+aPUpZJGYnWxbYAnx/Bhr9BO8dqXhfAeT+09HO2IjU/kZaiqS1N69TW4UZd2uRGrVF7jbJmH9r4SLJmiZmbm8hsVviuHjwEh5BHvi1q34Gf1VWucfGxjJ37lwyMzPx8vLimWeeISjIOBNWXovaV199lR9++IFdu3bx888/ExMTw6OPPoqzs3OJcxRX2ZQE4dalcfTCo8sLpO16mawTS/B0C6bzZH+iFsfReUojZCWWglTLGlQI9VdYWBheXl4Vpib88MMPnDt3jnHDp+Oiz0cz/BWLmxukp6ezdetWwFghoaxFuVfyYMw/8HE36OFrrJNqi1auZVFUWHke4rLhxfal77PlEqQVGGdvhYrJYbcjeZVfGaeqM+7WIg94Gvm2J5BcvGvk+JUlgttbSNt7WuMe5E5gz5L1ZqsquF8QcdviLbp1e+jQIb788ksKCgoIDAzkmWeewdv7xj+Ewha1pXFwcODRRx+lRYsW/Pjjjxw8eJDZs2czffp0AgPL/mORnJxMQkICsizTuXPn6r9god6z826D8aaWQnrkLFyB2ycA2ZCy8fpOIi3llibLMpPHDOXXZd+V+nxDu1w6+dvzTYw7O3bsICoqijvuuIOhrv5Yunbq77//RqfT0bx583Kbzrx+APZcgf/sggN3GmdPv+gJWy4aKxJUtZVrVe2+AvdvN87kjguG0JINLVneD3ZdgWaVzM8Vylde6bUaP7eDK9im5LtFRHB7C/Hv5I9/J+v+MS6+yr2sW7f//PMPy5YtQ1VV2rZty7Rp0yxKESh+rttvv53g4GAWLFhAcnIy77zzDpMnT6Z379JzfApTElq1aoWrq4XvpKqhUuMS6hdjukkF2WuFaSkiuL0lqRlJtNn+Im2alJO+pECzCdP4Zl8a586d4+eff2bbtm1MmDCBzp07l1vPNDMzk3///ReAUaNGlbvvR90gWw+vdbyRFuDnRLVauVZHHz9jUNveC/zLOK9GhtvEmtx6SVVV1FObUc5uQzNydpXr9laXyLkVqs3R05Gu08KRnKDbNPM2voqi8NNPP7F06VJUVeW2227jmWeeqXRgW1TTpk2ZNWsW7dq1o6CggG+++YYffvgBnU5XYt+qpCQoBXUjZ0gQhDrKwlanfi278uqrr/Loo4/i6elJcnIy8+fP56OPPuLCBfNGIYqicOrUKS5cuMCKFSsoKCigWbNmFdbldrWDpf2gZbG1u+NDIHECjAup5Guzgv/dDq92BJdbtMTXLS3zEoZfZqIeWIV6alOtDUMEt7cAVVE59cdpMi9k1tg5gvoG4jrZ2axjlk6nY+HChaxbtw6AMWPG8NBDD1Wr9W0hV1dXnn32WVMu2pYtW5gzZw5Xr14FjH8o9u7dy7lz5wBjy93yqIYbf6hkUb9WEAQrkBzdkGWZPn36MGfOHO644w60Wi0nTpzgzTffZOnSpWRlZREVFcVzzz3Hxx9/TFRUlOmOU5s2bUqd+XplP6w4Z+tXU30FBhiyHuYcgesd229ttbgIrKZIHo2Qez8CgGHde6gF2bUyDpGWcAtIjUlj3xf70TppGf/T3WjsNFY9ftEZh1OnTtGmTRtycnKYN28eZ86cQaPR8PDDD5eZOlBVsiwzevRoQkND+eqrrzh//jyzZs1iwIAB7Nixw2yxxzvvvFNmS8y8xG1kHPgIj66v4NjIhslpgiDcMhwdHRk7diy33XYbK1euZP/+/fzzzz/s3LmT/Pz8Ur/nzz//JDg42Ox9688E+O8xY5Gazg2gVR1ulHci3RiIPxJmLP/15G7jQrIT6WUvNruV1PYisJoi952KcvRPSL+AsvULNEOet/kYRHB7C1D0Cv6dG+Lg7mD1wLZ4b/WoqChT17GMjAycnJyYMWNGpWvTVkb79u2ZNWsWCxYsIDY2lj/++KPEPoUtMadNm2b6Q2HISSbj4MfkXfgXgKxTP+LYqKdx5la2r7DOrZjhFeqikJAQ4uLiSmx/8sknWbBggelrVVUZMWIE69at47fffmP06NE2HOWty8/Pj+nTp3Py5El+/PFHEhMTy92/eKezEQHwVGto4ly3A1uApTHw5wU4dw0Sc4wL3Jw0xpq4ZXS3vuXU5iKwmiLZOaIZ/iqG/z2BEvkDcse7wKupTccggttbgG9rHwa/N7BStWEtUdhbvbiMDGNXMldXV1566SWaNLFedYay+Pj48NJLL/HUU09RUFB2ULpixQo6d+5I3rnVZB75AlWfA5IG11b34drGeCtF6+KP34hV5daxFaWghLpq3759GAw3FkUeO3aMIUOGMH78eLP9Pv3001pb7CFA69atmThxIh999FG5+xV2OiusJCNLxrJfN8OP7sX2kJQNl/Pg1PWsuAIF9l+t3XEJNU8Oux1DaF+I2YH+t5cg4k3ccxLh0gnUwtTEGpyVFsHtLcSaf8gs6a2u1Wpp1KiR1c5ZkfPnz+MkZ+PjXnYyl5MUT+LayWhzYgCwa9AOz64vY+fZ3Gw/rYu/WAkv3JR8fc1rWL///vuEhobSv39/07ZDhw6Z8jtt+W9UMHftmmWLV3df1PFTHrzR0RjU3gyBLYCnPYwMhElFuqwbVPgtHlbF2q72rmB7akYSxO41fnH5JHx/L30BzszH9Bdaa4922t81EuCK4Laey0nNxcHVHo29ddMRLOmtnp6ebjbjUNOupZzj2dvOYqcpe4baoIAmByStM+4dpuEcOgZJtu61EW5u9SktpaCggGXLljFz5kzTh9ucnBwmTZrEggUL8Pe37ANcfn6+WV5oZqZxGk6n05VapaTe01u2Gkqv10M518eS8oT5WmdeSGpNWjy4awxMb1UbTVarJjkPntitRQLUIq0sJVSe2AW9G+jxK9nrx2YKf3dvyd/hmpZ5BQwVVBTRF6DPvALOpTeVqs7PRQS39dy+L6K4uP8SPZ/uTsjtwVY7bmV7q9uCh7NcbmALxvqK+T5DCOr1FBpnPxuNTLiZFE9L0ev17Ny5kz59+pgqfdwsaSmrV68mPT2dKVOmmLY9++yz9O7dm7vuusvi48yZM4fZs2eX2L5hwwaLugTWN44F6fRDg5aya2IbJC1b9xwizz62zH1UVcXR0ZG8vDwAYn27sa/5JLqf/ZHgK8Yyhp52KhPsjrJZDaZRzG7Wnrs5ygyoKnyY3Y1ren/UYoWZVCQydQr3rLnCC677ammEN2zcuLHinYRKcc9JNM7UVmDHjp1kOseW+lxOTk6Vzy+C23pMVVUy4jLQ5+lxa+Jm1WNb2lvd0v2sITg4mKtnKt6vcaf7RGArlKtoWoqk05Ejx2Dn1RI7u5urcOeiRYuIiIigcWPjbb81a9bwzz//cPDgwUod5+WXX2bmzJmmrzMzMwkMDGTo0KG4u5fSguoW8K8s4bdvAW2cM6HzeOg01ux5jZMXAz0qTvlo1KgRCxcuJNfOjT1hk9FpndkTNgW/9FM46a4xefJkwsPbGe86yUNr6uVY3fF0iFxb9r8XBZlIfWNCeo+gjafNhmVGp9OxceNGhgwZctP9267zLp2AM/Mr3K1v3z7gX3oHvsI7RFUhgtt6TJIk7vx6JGkxaXg1s26dPEt6q3t7exMWFmbV85bH0pxisYhGuBXExcWxadMmfv31V9O2f/75h5iYmBIfOgtLVBV2xSrOwcEBB4eSvTXt7Oxu2aAg9moeXRyMM0uajqORAztW6Tg9evRAljXcv8sevdYRJAmd1pHNXV7ht86J9OhubEBzs13ljj4wOgj+SDDm2RankWBUIHT0rf1Xdiv/HtcUVavFknsMWq0WqYxrX52fiWjiUM9JkoR3c28kK9ddkWWZcePGlbvPxIkTTeVrBEGwrcWLF+Pn58fIkSNN21566SWOHDnCoUOHTA+AuXPnsnjx4loa6c1Jf/EEbho9Bo0DUpPyu4hV5LxPV2I8O6BKxvx/VdKQ5ujPs5ct76xY10gSfNETXLVQ/K+PBLjZwQJRVlyoIXUq8ggJCUGSpBKPadOmme2nqioRERFIksTq1atrZ7AC8fHxACUCWG9vb7N6sragqgrXjn5ps/MJQl2mKAqLFy9m8uTJZh0B/f39adeundkDICgoiKZNbVuH8mamKAremcYcKEPjTkiaCrpMlSM5F57YXTIABDiTaXz+ZuXnBF/2guITtyrG7X5V78IuCOWqU2kJoj6j9WRcyGT7ezsJ6hNIh/uqN6tQmqSkJDZtMvaNnjFjBlqtlq1bt9K/f3/atGlj8xlbSZKRtLfewhZBKM2mTZuIj4/n4Ycfru2h1EspKSk0t0sFwCHstiofR1XhyT2QpS8ZAALkGmDaHlg1oMqnqHXjQ2Bl7I30hMJ0BFEGTKhJdSq4FfUZrSdxbxLp59Nx8rR+nRVVVVm+fDkGg4FOnTrRqVMndDodMTExtGzZ0iaBraqq5CVsQusRip1HMwBcmo8l78I/NX5uQajrhg4danHTFms3d7kVJCbEEeZoXOyiadaLVbHwTCR81gPGhVh+nOPpsDq+7OcLa8IeT4O21l02YTOF6QlbLho7lIl0hFuEsxdo7UFfTjkwrb1xvxpQp4LbokR9xuoJ6h+AnasWB3cHq7/OgwcPcvz4cbRaLePGjTO7lra4prr002Qf/gzd1cPY+XXFo89cJElCdWxoUX1SReNy0//sRX1G26nKtRY/l/rtSvwZpDw3gl0VCjxa8fhqY+D2+G7o17Ds2+3nrsHc45Cth+/6QltP46Kr3+NLn7ktnOW8WQPbQn5OsLD3jQ8AIh2h/pM8GqOd9jfkGBed6/V6duzYSd++N0oq3pIdykR9RivJhYNrrXc4g8HA5s2bAWjWrBlRUVFmz9dkvUCtmk0T/VZ8DYeQAANaLqQ6c3HtXyAZZ4vt7R5Dq5adpKaXnCj490CNjdHWRH1G26nMta5OfUah7juffI0Vl1tzd+8xLImUybq+LPyazjyN4GwmaGUIud6rQa/Al6fATjYGeW52N2Y1M3XmAW59W3Q1PkSkItxqJI/GUBi86nTGerb+bcqsjmBNdTa4FfUZ66Y//viDnJwcvLy8mDFjhqk8UHXqBRpyLqHkZ5T5vGTnSsGlneSc/A7VkAWAQ8BgXNo9gb9zQzpX/eXctER9RtupyrWuTn1Goe5LTEwE4IRbe7O0gqKtZQ9ehf8egxmtYW534/Mt3OGl9tDTFxyuZ28VLroq2qIWxKIrQaiOOhncivqM1XN67VnsnLQ06d4Ye5eqr+ItLiUlhXXr1gFw7733lto6srLXVZ99idQNk8pPJZA0oBoXGtp5tsQ9fCYOvp0qNfb6qj7/Htc1lbnW4mdSfxl0eWRfjiPXzoMPkoKvt5a9QcJY/eDjrqCVILPIW5skwTvhJY95Y9GVikGV0EgqowIlMdMpCFVUJ4PbsuozPvroo2b7tW/fnrlz53LnnXfaeoh1lmJQOPzDEfIz8hn8wUD8OzS02rH/97//odPpaNWqFd26dbPKMZX89PIDWwDVgNazBS7Nx+Hc9E4kWWOVcwuCIFRW2pHNzGmylzu8vyDbIJVa5uqaDtYkQPK94G7B/IL5oisVN239SUcQhNpQ54Lb8uozlraITNRnNKfoFJoPDyX5aDJ+bXwr/gYLHT9+nP379yPLMvfdd5/NS7F5dnsVe+/WNj2nIAhCcXnR27hg14J/3AaUvgoMY3rC7wmQkA1tLbx55ucEn3c3MGOXns97aPFzqnN/ngXhplHn/vWI+ozVo3XU0nlK1dpAlkWv1/Pjjz8CMHDgQAICAqx6fMuIusaCINQ+h4sHaaNLoE/BYfY4dCy3tWxlqxyMC1ZxPr6eEUEjrDNYQbhF1bngVtRnrHs2bdrExYsXcXNzY8yYMbU9HEEQhFqh5l3DPecCEvCU9xmO5Xes91UOBOFmVKfa7wrVk38tn9SYNKsG/enp6fz+++8AjBs37tYonyYIglAKNS4KGZXLOkeahrbAVStaywpCXSSC23okfnsCa6evY+tb2612zFWrVpGXl0fTpk3p27ev1Y4rCIJwszGc2wVAdK475xxCSMwFO8mYhgDG/44JEvVcBaG2ieC2HsnPzEfjoKFBmLdVjnfmzBl27TK+md9///02aasrCIJQV+nP7gTgnMGHu1u68+9wYzMG1+sJfiIdQRDqhjqXcytUXbt729L67lYYdEq1j6UoimkR2W233UazZs2qfczSyA6eFrXMlR08a+T8giAIllBVlQtBw7kQ/zOZXq2QZYm+DaFvQ/ByEK1lBaEuEcFtPaOx16Cxr34d2G3bthEXF4eTkxPjxo2zwshKp3Xxx2/EKpT8NHRpp9GlReMcHAGaG/VzZAdPtC4ly8AJgiDYiiRJHFeD+D0llL6tQ82eE61lBaFuqVJwGxERwQMPPMCYMWNwchIfU+sCxaAga6yTNpCVlcUvv/wCwOjRo2u8TbHWxR9c/K/Xsb2rRs8lCIJQVYmJiRRonHjPYTypx+Dp1mCFuQRBEKysStHQuXPnuP/++2nYsCGTJ09m06ZNoixXLVJVlT8fX8vmV7aQdTmr2sf77bffyMrKokmTJgwcONAKIxQEQbi5GfYtR006xnn/XsTp3Vh6FuzEMgRBqJOq9E/z1KlTREZG8tBDD7FhwwaGDRtGQEAAzz//PIcOHbLyEIWKZF64RuaFa1w+loyjh2O1jhUfH8+WLVsAuO+++8y6xNUUVVXJPPIFuQmbUQ35NX4+QRCEylDTE1H+fptH7f+l7aUtfNoxi9mdjW1zBUGoe6r8ubNbt2589tlnJCYmsnbtWgYOHMhXX31Fly5daNeuHf/973+5cOGCNccqlMEj0J1R34yk7wu90TpWPRhVVZUff/wRVVXp1q0brVvbpt2tITuJrJNLSNv9Oqpa/cVwgiAI1qTGRgIQm++Knb0j0zq6MCa4lgclCEKZqn1TRZZlhg0bxtKlS4mPj2fcuHGcOHGCl156iZCQEAYPHsxff/1ljbEK5XAPcCeob2C1jhEZGcnp06ext7dnwoQJVhpZxQqSDwBg590GWStyuAVBqFuU83sAiM7zoEmTJkhiylYQ6jSrZAzt2LGDxx9/nObNm7Nq1SrTzO3HH3/MlStXGDVqFG+88YY1TiXUkLy8PH766ScARo4cSYMGDWx27vwrBwFw8Otss3MKgiBYQlVV1PN7OKcN5MXgj0luIgrZCkJdV+V72CdOnGDZsmWsWLGC+Ph4/Pz8mDx5Mg888ACdOnUy7ff000/zn//8hwULFvDWW29ZY8xCEYl7k7h44CLB/YPxbe1Tqe9VFIXTp0+Tnp7OoUOHSE9Px9fXl4iIiBoabekKrge39r7hNj2vIAhChVLOQdYVvvR6kRj3juxxSKntEQmCUIEqBbedOnXi6NGjODg4cNddd/HFF18wbNiwMjtYDRgwgG+//bZaAxVKd35LLLH/xqGx11QquI2KimL58uWkpaWZbe/WrRt2dnbWHmaZDDmXMWQngaTB3qeDzc4rCIJgCSXWmJIwKv1PtqT5MnVoO6ByEwmCINhWlYJbT09Pvv76a8aPH29RDdS77rqL8+fPV+VUQgWaDgxBY68hsI/l+bZRUVEsWLCg1OfWrl1L06ZN6dq1q7WGWK78wnxbr5bIdi42OacgCIKl1LgoAK6mZhKe/jNjW/et5REJglCRKgW3//77b6X2d3Z2JjhYLC2tCU26NaZJt8YW768oCsuXLy93nxUrVhAeHl7mTLw16dLPAGDvK/JtBUGoezRjPuBS4CD2fLscFxeXGm9qIwhC9VUpejlw4ABffPFFmc9/8cUXot5tHXX69OkSqQjFpaamcvr0aZuMx6PTU/jdsRrXMNtVZxAEQbDUzhR7Zlzsyjn7pqJSgiDcJKoU3L766qts2rSpzOf/+ecfXnvttSoPSrBM7NY4clJyKvU96enpVt3PGrQujdA4N7TZ+QShPgsJCUGSpBKPadOmkZqayowZM2jZsiVOTk4EBQXx1FNPkZGRUdvDrrPmnYTVGf5EBwymSZMmtT0cQRAsUKW0hP379/Pyyy+X+fxtt93GnDlzqjwooXxx2+LZ+0UU+Rn5SLLE+JV3Y+9ib9H3enp6WnU/QRDqln379mEwGExfHzt2jCFDhjB+/HiSkpJISkrio48+ok2bNsTFxfH444+TlJTEzz//XIujrpsMf83icX0zYvO70jhxM026D6jtIQmCYIEqBbfXrl0rty2rLMtiJqCG5KXnsWfeXnTZOiRZokGYt8WBLUBYWBheXl7lpiZ4e3sTFhZmjeGWK/PwAnSZ53ENuxeHhrZZwCYI9Z2vr6/Z1++//z6hoaH0798fSZL45ZdfTM+Fhoby7rvvcv/996PX623SbvtmoRoKUI6s4TZdLhNy+nMsO1/M3ArCTaJK72QtWrRgw4YNzJgxo9Tn161bR7Nmzao1MKEkVVWJnL8Pfa7etM3Jq3IdvWRZZtKkSWVWSwCYOHGiTRaT5SVtQ58Zi3PTO2v8XIJwKyooKGDZsmXMnDmzzFzRjIwM3N3dyw1s8/Pzyc/PN32dmZkJgE6nQ6fTWXfQdUXCAdDlojp5cfx8HiDh5+dXo6+38Nj19prWMeJ6205VrnV1fi5VCm4feeQRnn32WWbOnMkbb7xhuoWdnp7O7NmzWbduHR9++GGVByWULm5bPAm7Lpi+VhWVhN0XiNsWT3C/IIuP07VrVyIiIvj777/Ntnt7ezNx4kSblAEz5KWiz4wFwMG3U42fTxBuRatXryY9PZ0pU6aU+nxKSgpvv/02//nPf8o9zpw5c5g9e3aJ7Rs2bMDZ2dkaQ61zLiRfIt/tQforx1CRcHBwYPv27TY598aNG21yHsFIXG/bqcy1zsmp3JqioqoU3D711FMcOnSITz/9lHnz5tG4sbEUVVJSEoqi8MADD/Dss89WeVBCSXnpeUTO3wcSoBZ5QoLI+Xtp2MEPR09Hi4+nKAoAHTt2pGfPnnh6ehIWFmaTGVu40ZVM69Ec2cHDJucUhFvNokWLiIiIML1HF5WZmcnIkSNp06YNs2bNKvc4L7/8MjNnzjT73sDAQIYOHVovS2OpKvRdnsAB72Y8o/kHji0lJCSEESNG1Oh5dTodGzduZMiQITZtpnOrEtfbdqpyrQvvEFVFlYJbSZJYvHgxDz74IL/88gvnzp0DjM0axo4dy+23317lAQklmaUjqMWfBF2OnsjP99H/tdssPuapU6cA6NGjBz172r5XekGyMbh1EPVtBaFGxMXFsWnTJn799dcSz127do3hw4fj5ubGb7/9VuEfGwcHBxwcHEpst7Ozq5dBgSE/h4fSloHLGDo7xrEJCAgIsNlrra/Xta4S19t2KnOtq/MzqdbqgQEDBjBggFg9WtMy4jLM0hGKUxWVhJ0XSI/LwDO44lnQ3Nxc4uLiAGjZsqXVxlkZ+ddnbu39RHArCDVh8eLF+Pn5MXLkSLPtmZmZDBs2DAcHB9asWYOjo+V3fG4VUsIBHr22gkflf5l/zXj9xGIyQbh52OYetFAtHsEeBPYOQJJLXxAiyRKBfQIsCmwBzp49i6qq+Pr64u3tbc2hWkTJz0CfcRYQnckEoSYoisLixYuZPHmy2UKxzMxMhg4dSnZ2NosWLSIzM5NLly5x6dIls/Jht7yCbPAMQGrai6SLFwER3ArCzaTKwe2RI0d47LHH6NKlC82bN6dZs2Zmj9DQ0EofUxQfL50kSXSf3hWtk9aYc2v2JNg5a+kxvZvFxytMSaitWVtDfhr2vp3ReoahcbR9cC0I9d2mTZuIj4/n4YcfNtt+4MABIiMjOXr0KM2bN6dRo0amR0JCQi2Ntm756Tyscx+GZsZGdAOe5+rVq4AIbgXhZlKltIR///2X4cOH4+XlRdeuXTl48CADBw4kLy+P3bt307ZtW7p06VLp44ri46VT9Ao7/7ub4H5BnP07xvxJFXrM6F6pxWSFrXVtUcu2NHbuIfgMXIiqKrVyfkGo74YOHYqqFk/Qh9tvv73U7YKRToHn9sHFXFh5O3TUGwNbT09PXFxcandwgiBYrErB7RtvvEGzZs3Ys2cPBQUF+Pn58corrzBw4EAiIyOJiIjggw8+qPRxRfHx0sVsPMelQ5fRnrpK426NuLj/EqqiIskSAb2aVKoMWH5+vmkBYG3N3BaSJJEVIwhC3ZGrhwmBBay7ZMcdARJ7diYCYtZWEG42VYoIDxw4wOzZs3F3dzd1uiqcce3RowdTp07l9ddfJyIiosoDE8XHbwgeGEROag7uge74tG7A2ifWo8vWoXXSEj61U6Vex+nTpzEYDHh6euLp6Wm1a2BpgWbVkI9qyEe2r3/lg2xJFB+3HVsXHxdqj7s9fJD+X949/QeaQzNJTDTeXRLBrSDcXKoU3Gq1Wtzc3ADj7Ro7OzuSk5NNzzdr1owTJ05Ua2Ci+HgxrkAasAs0vST0O0HTW+KfXf9U6jDR0dHGw7m6lmjiYA0VFWj2MJyhhW4V6XIYZ+3HWf38txpRfNx2bFV8XKhdyvk9yPmZSM5eJCYeASi1TrAgCHVXlYLb5s2bc+bMGcC42KlVq1b89ttv3HfffQD89ddf+Pv7V2tgt3rxcX2+gfObY2k+vFnpVRKeqtpxC4Pb/v37079//2qM0JylBZqzji4g9ww0CgwjrEvNFkSvz0TxcduxdfFxoXb87xx0drxKs5QYQEIK6U5ionECQMzcCsLNpUrB7YgRI/juu++YM2cOWq2WmTNn8tBDD9GiRQsAYmJimDNnTpUHJYqPQ9TnB4jZcI6002n0eb6XVY6p1+tN+batW7eukddf0XXVXz0MgKN/lzp9/W8Wdf33uD6xVfFxwfau5sGjuyDP0ID9ds1p42NPrmpPeno6IIJbQbjZVGlFz+uvv87hw4fRaDQATJ48mR9++IF27drRsWNHvvvuO1588cUqD6q84uNDhw7F3t6+3hcf92vni9ZJS7PBTa12zPPnz1NQUICrq2ut3GZTdNno0oxlyOx9w21+fkEQhNJk6GBQIwjXJNJadxY5pCeJicbFZN7e3jg5OdXyCAVBqIxKz9zqdDpOnjyJt7e32UKv+++/n/vvv7/aA6qo+HhOTg7Lli0jMzPTdOvP19fXFGjXF6FDmhHQowkO7iVnnauqsARYy5Yty1ykV5MKUo6CakDj0gitS/XSVgRBEKylmRusHqiSNe9hJEBq2pMLCcaukGLWVhBuPpWeuZVlmS5dupSaMmANt3LxcV2ODkPBjTq/1gxs4Ubzhtqqb1tw5QAgZm0FQagbVsVCk5/g51ggLQHHjHiQtUhBXUhKSgLEYjJBuBlVeuZWo9EQHBxsVl7Lmm7V4uOqqrJ7biTXLmbR75U+uDV2s+rxFUUxLQKsrfq2+ckHAXAQLXcFQahlybnw+C5jSsLUXXDbECd8+k+HnDQke2dTWkJAQEAtj1QQhMqqUs7tjBkz+Prrr0lNTbX2eG5Z2ck5XD6STEZcBvmZ1v/gEB8fT15eHk5OTgQGBlr9+JZwCR2DU3AE9g0r371OEATBWlQVntwDWXrj1xk6ePyoL5r+09BEvAZgCm7FzK0g3HyqVC3BYDDg4OBAaGgo48aNIyQkpETCvSRJPPvss1YZ5K3AtaELIz4fTkp0Cj6tfKx+/KIpCbJcO53BnJuOwLmpKP8lCELtWhULq+PNt/2RYNw+PsRYladwTYcIbgXh5lOl4Pa5554z/f+iRYtK3UcEt5Xn4uuMi6/lrXQro3AxWW3l2wqCINQFybnwxG6QgKKJbhIqT+xS6d9QJvX6rK2Pj0+9rsojCPVVlYLb8+fPW3sctyTFoBA5fx8tIprj07JBzZ1HUcwqJdSG3ITNaF0D0Xo2R5JqZ+ZYEIRbW9F0hOIrOFQkrukUpu2BJ1TjYjJRKUEQbk5VCm6Dg4OtPY5b0qk1p4lZf46EnQmM/v4u7F1qpvD7xYsXycrKwt7evlZ+dqohn7Q9s0ApwG/EKrRuNTM7LQiCUJ7j6SXTEYoyoOG3eOgiZQEiuBWEm5WYQrOhuG3x/DzpN+K2G99dQ4c2I7B3AF0f71JjgS3cyLdt3ry5We1gWym4ehyUAmTHBmhca2cxmyAIQltPGB0EmjLKfGsklTFBQNIJQAS3gnCzqlKk07Rp0wqbAEiSRExMTJUGVR/lpeexZ95edNk6IuftpWF7Pxw9Hen3Wt8ab6hQGNzWVkrCjfq2nWuleYQgCAKAJMEXPWHzRbimK/acquBmL/F5D5V3fhWVEgThZlal4LZ///4lghSDwUBcXBw7d+6kXbt2dO4sapkWUlWVyPn70Oca684U5OiI/Hwf/V+7rcaDPVVVa30xmam+rZ9o3iAIQu3yc4KvesGkbebbVUnmy17gqMsgKysLSZJEcCsIN6kqBbfff/99mc8dPnyYYcOGcd9991V1TPVO3LZ4EnZduLFBgYSdF4jbFk9wv5rNP01OTiY9PR2tVkuzZs1q9FylUQ06dFePAsaZW0EQhNo2PgRWxhrLfxlU0Kh67nRPYnxIECdOGGdtfX19sbe3r9VxCoJQNVbPue3YsSNTp07lxRdftPahb0p56XlEzt9nrDtTTOT8veSl59Xo+QtTEpo2bVorb9QFqSdQDfnIDp5o3Zva/PyCIAjFFaYnuGpVUFXclGwWdDe2Pi9s3iDybQXh5lUjC8oaNmzIiRMnauLQNxWzdIRSOgfrcvREfr6vRsdQ2yXACq4YUxLsfTuJfFtBEGpdZOwVRv55jdXHkviy9QUa2hfwhf9ufOUc1IvHSYs7CYjgVhBuZlZfOn/16lUWLVok+nEDGXEZ5ukIxaiKSsLOC6THZeAZ7FEjY6jtxWQuYROw926DpHWqeGdBEIQapGYkseGPX1nvPh3n+J0sT3mWMdefM0QZ/3snMpGajiK4FYSbWJWC24EDB5a6PT09nejoaAoKCli6dGm1BlYfeAR7ENg7gAt7ElGVklO3kiwR0KtJjQW2V69eJSUlBVmWCQ0NrZFzVETWOuHg371Wzi0IgmAmJ42x1/7CWZ9FS925UnfRouCq0YvgVhBuYlUKbhVFKXGLWZIkmjZtyuDBg3n44Ydp1aqVVQZ4M5MkiR4zunHp8GV02cXrzoCds5Ye07vV2PkLUxKCg4NxchIzp4IgCC30sbS4FlvuPpIs4e/vb5sBCYJgdVUKbv/9918rD6P+cvR0pMeMbux4f5f5Eyr0mNEdR8+a61temJJQWyXAcuLWo0uNxiloEPYN2tXKGARBECrL28sbO7uaa6wjCELNEh3KbCC4XxCBvQOQZONstyRLBPYJqPEyYLW9mCw3bj3Zp5dTkHKkVs4vCLeikJAQJEkq8Zg2bRoAeXl5TJs2jQYNGuDq6srYsWO5fPlyLY/aNjakuLDZsTc5UvmTCn4N/Ww0IkEQakKVgtt58+YxbNiwMp+PiIjgyy+/rPKg6pvC9AStk3GivKbTEQAyMjK4ePEikiTVysytqhgoSDkEiPq2gmBL+/bt4+LFi6bHxo0bARg/fjwAzz77LH/88QerVq1i69atJCUlcffdd9fmkG1m1lk/7mi4iF+dy/77BcYat4Ig3LyqFNwuWrSINm3alPl8mzZt+Prrr6s8qPrI0dORnk91x9HLkR5P1Ww6AtyYtQ0ICMDFxaVGz1UaXfoZVF02kp0Ldp61kxYhCLciX19f/P39TY8///yT0NBQ+vfvT0ZGBosWLeKTTz5h4MCBdOnShcWLF7Nr1y727NlT20OvUaoKbd3yaaK/SP+8yHL39fMVM7eCcDOrUnAbExND69aty3y+VatWxMTEVHlQ9U1GQiY7/ruL3LQ8xi0fQ/BtNZuOANR6y11TfVufjkiyplbGIAi3uoKCApYtW8bDDz+MJEns378fnU7H4MGDTfu0atWKoKAgdu/eXYsjrXmSBN+0TeJM4kACDZfK3dfPTwS3gnAzq9KCMnt7ey5dKvvN4eLFi8iySOctlHo2ldgtceRcyaHVXbYJNmu7vm3BlQMA2PuG18r5BUGA1atXk56ezpQpUwC4dOkS9vb2eHp6mu3XsGHDct/T8/Pzyc/PN32dmZkJgE6nQ6fTlfVtdY+dG5LGHgwFZe6iUyXc/YNq5XUVnvOmuqY3MXG9bacq17o6P5cqBbc9e/bk+++/59lnn8XNzc3suYyMDBYvXkzPnj2rPKj6xquZF52mdMDJyzbluLKzs7lwwdg8olbybVWF/CuHAHDwE/m2glBbFi1aREREBI0bN67WcebMmcPs2bNLbN+wYQPOzs7VOratXFPscJN1eDZ9mNaJf+Ggv8aB4PuNU7pAaupVjhw5iurkQZc9R4GjtTbWwjxpwTbE9badylzrnJycKp+nSsHtm2++Sf/+/enUqRPPPPMMbdu2BeDYsWN8+umnXLx4keXLl1d5UPWNZ7BHjTVqKM2ZM2dQVRV/f388PGx33kJK7hUk2R60Tth5iXrHglAb4uLi2LRpE7/++qtpm7+/PwUFBaSnp5vN3l6+fLncuq4vv/wyM2fONH2dmZlJYGAgQ4cOxd3dvUbGb02Xc6Hpb1raesKOu/U4aJ4EfQF9tfamfdavX0981Dm6dujEiBEjamWcOp2OjRs3MmTIEFGKzAbE9badqlzrwjtEVVGl4LZHjx788ccfTJ06laefftrU0EFVVZo2bcqaNWvo1atXlQclVE9tpyRonBvScNSfKHkpSLLVOzwLgmCBxYsX4+fnx8iRI03bunTpgp2dHZs3b2bs2LGA8f0iPj6+3PdsBwcHHBwcSmy3s7O7KYKCI5dBBbQyuDpeH2+xcRemZQQGBtb6a7pZrmt9Ia637VTmWlfnZ1LlyGPIkCGcPXuWgwcPmhaPhYaGEh4eXqJ72a1M0SukxqThEeiOnbNt/vHU9mIyMJY/0ziJcjqCUBsURWHx4sVMnjwZrfbG27yHhwePPPIIM2fOxNvbG3d3d2bMmEGvXr3qdSrZiAC4NAESL8ShZjoiuTcssU9iYiJAtVM4BEGofdVa9SXLMl26dOGee+7hnnvuoUuXLtUKbOtj8fFrSddY98wGfrl/Naqq1vj58vLyiI2NBWpn5lZVVZu8TkEQyrZp0ybi4+N5+OGHSzw3d+5c7rjjDsaOHUu/fv3w9/c3S12or3wcoe2et9B/OgDl0G9mzymKQlJSEgBNmjSpjeEJgmBFVQpuV6xYYVp9W5qHHnqIlStXVvq49bH4eF5GPo5ejrgHuNtkRjsmJgZFUfDx8aFBgwY1fr7iDNfiuPx7BGl7Ztn83IIgGA0dOhRVVUu9e+Po6MiCBQtITU0lOzubX3/9tdx82/pCzb6Ken4PoCIFmVdxSUlJoaCgAK1WK8qACUI9UKW0hLlz59K5c9mr4J2cnJg7dy733HNPpY5bvCvM+++/X6L4+PLlyxk4cCBgzClr3bo1e/bsqbO31Bq292Pc8jEYCgw2OV90dDRQeykJupRDKPlpGHKTa+X8giAIRf0SC78nwD26IwxTFaTG7ZG8g832KZy1bdSoERqNqMstCDe7KgW3p06dKvV2V6GOHTuyYsWKKg8KbhQfnzlzpkXFx8sKbutMfUYJFJ1S46cpXEzWvHlzm76+wnPlJxvr22obdBS1A2uQqM9oO7auzyhY1+8JsPwcBKnpDAOktiUrIRTm24qUBEGoH6oU3KqqSnp6epnPp6WlVfvN3VrFx+tDfUZLGQwG0+K+pKQk1q5da9sBqCrZSXuxBw6czefaeRuf/xYk6jPajq3qMwrW9URLCNFeY+TWZYCE3DaixD6FdcFFcCsI9UOVgtvOnTuzYsUKZs6cib29vdlz+fn5LF++vNy0BUtYq/h4bdZnVBWVLa9txbWhK50e7Yi9S81WSzh9+jR//PEH7u7ujB8/3qZVK3Q6HdvW/4Q9WSDb0XfEw0iakqWDBOsQ9Rltx9b1GQXr6uUH3c+uRCk4gRTcrdRKCYVpCaJSgiDUD1UKbl966SXuuOMOBgwYwEsvvWTWxGHOnDkcP36cNWvWVHlQ1iw+Xpv1GbMuZXHlWApXo1Pp9WwPZE3NtiQunLVt1apViQ8dtuCmxANg36At9o6uNj//rUjUZ7QdW9VnFKxPPWmcdZfajSzxnKIoXLx4EYCAgACbjksQhJpRpeA2IiKCRYsW8fTTTzN69GjTdlVVcXNz45tvvjErHF5Z1iw+Xpvs3Ry47eU+5GXk1XhgC7XfvMEU3PqKlruCINS+1XEQ5AodJn6D5sxmpBb9S+yTnJyMTqfD3t4eHx+fWhilIAjWVuUmDlOmTOHuu+9m48aNZk0chg4dipubW5UHVJ+Kj9u72BHcL8gm59Lr9Zw9exawXaUEffYllPx00/l1khMa92ZonPwoSI1GdvBE61L/SwwJglD3GBR4dBekF8DOEW706Di61P0KF5M1atQIWa75SQhBEGpetXqjuru7m2ZRizp06BDLli3jo48+qvQxKyo+LssyY8eOJT8/n2HDhvHFF19Uaez1TVxcHAUFBbi4uNgkb0yffYnkteNBKTBtawQYMiFj/wfGDbI9fiNWiQBXEASbSyuAPn5wJA26lFPyW1RKEIT6x2ofU2NjY3nvvfdo27Yt4eHhzJ07t0rHqU/Fx5P2XyQ1Jg2DruZr3BZtuWuL2QclP90ssC19pwLTzK4gCIIt+TjC6rYnOZU2FinqxzL3E53JBKH+qVYUdPXqVb788kv69u1LaGgob731FgEBAcyfP5/z589ba4w3JVVV2T5nJ2unryPzQs2vnK7tfFtBEIS6Rjn2F9LFE6hxe8vcR8zcCkL9U+m0hNzcXH7//Xd+/PFHNmzYAECPHj0AWLZsGePGjbPuCG9S+jw9nsEeZF64hnuTmi05piiKaeZWBLeCINRXhe91hVVzyrpTpVcg36Bif9xYa1supXEDGNcKFFZKEMGtINQfFge369ev58cff2T16tXk5ORw++23s2DBAsaOHcvVq1dtdjv8ZmHnZMewj4fY5FwXLlwgNzcXR0dHAgMDbXJOQRAEW4qKimL58uWkpaWZtnl5eTFp0iS6du1qtu/uKzB0vcoddi+w3P61UqskgLFSgsFgwMHBAW9v7xodvyAItmNxcBsREUHTpk157733GD9+PA0b3iiEnZqaWiODq08snXGoisKUhBYtWoi+6IIg1DtRUVEsWLCgxPa0tDQWLFjAtGnTzALcyCugU2W0GJBaDUKycyz1uEVTEsTkjCDUHxYHt/7+/pw/f54lS5ZQUFDAvffeK7q5WKgyMw5VIVISBEGorxRFYfny5eXus2LFCsLDw00B6szWeu7YeB+63CzkIS+W+X2Fwa34WyYI9YvFH1UvXLjAhg0baNu2LbNnzyYoKIh+/fqxcOFCrly5UpNjvClte3cHG57fxPZft7NgwQKzwBZuzDhERUVV6zyqqppmbm1V31YQBMFWTp8+XeL9s7jU1FT27NmDqqrGDXF7Cb12hFZ2qUjNSm/yoygKJ0+eBIwd5RRFseq4BUGoPRbP3MqyzODBgxk8eDALFy40LSp7+umn0ev1SJLEtm3b6NGjh0jMB5KPXyEvLY/orBPl7ld8xqGyLl68yLVr17Czs6Np06ZVOkZVyA6eINuXXw5MtjfuJwiCUEXp6ekW7ffNN9+wcuVKwsLCCG/sSNuQ23DyC0HSlGyFXPxu2pYtWzh06JDV7qYJglC7qhRROTo6MmHCBNasWcPFixeZP38+vXr1Yt68eQQFBdGlSxdmz55t7bHeVAbM7k/T+4K5mHOx3P1SU1NNaQVVUThrGxoaatbRraZpXfzxG7EKx4ABANg3GcBx+4fxHLAInyFL8BmyRDRwEASh2jw9PS3aT5ZlMjIy+CHJg0ePB3LfvsY8/Ucin376KWvXriUmJga9Xm/K362pu2mCINS+akdD3t7ePPnkkzz55JPExsaybNkyfvzxR9566y3efPNNa4zxptSghTeaqzLK5oobOFg6M1Ga2sy31br4497paRz8uiK5h5Fz9Tx2Xi2xsys5UyIIglAVYWFheHl5lZua4O3tzXvvvUdsbCyj9jUkxuCJf24COYlxHD58mMOHDwPG9ANT6kIZqns3TRCE2mfVf70hISG89tprnDx5Unz6xfIZB0v3K65ovm1tLSbTujTCpcU47Lxb18r5BUGo32RZZtKkSeXuM3HiRBwcHGjZsiVzmp5nalAm3784mddff517772Xzp074+Ligk6nQ6/Xl3us6t5NEwSh9tXYfezOnTvX1KHrvIsHL6HL1hHQMrDCGQdZlpEkqUrnSUlJIS0tDY1GQ7Nmzao6XEEQhDqta9euTJs2ja+++sosOPX29mbixImmPFm1IIc7tz3CnbpctK1/Q2rWimbNmjFs2DAURWHdunWsWrWqwvNV526aIAi1z3ZJmreQk79GkxR1ke4zujFp0qRS6zMWUhSFDz74gGHDhnH33XdX6pZ+4axt06ZNcXBwqPa4KyvzyBdonP1xChoMkpPNzy8Iwq0jPDwcrVaLXq9n3LhxhIaGlqgXrp7eArpc8AqChuZ3s2RZtngSoKp30wRBqBtEUlEN8AzxpEGYN15NPU0zDsWDT29vbx577DH69u2LqqqsW7eOWbNmce7cOYvPU5slwJSCa2RFLyNj/wcoBVk2P78gCLeW5ORk8vLysLOzY/jw4bRq1apEXuzKw5eI0zRGbhtR6h2xwvzd8nh7e4uyioJwkxMztzUg/JFOZl937dqVrVu3cuzYMfr160evXr1MMw69e/emS5cuLF68mKSkJN59913uuOMO7rzzzgqrH9TmYrL8S5GgGtC6h6B1bYxOp7P5GARBuHWcP38egKCgoFI7MSamZnK//hGkJg+R3PIcpYWwhfm75d1NmzhxolhMJgg3OfEv2EYuXboEQJ8+fUrMOHTq1Il3332XHj16oCgKa9as4e233yYhIaHM46WlpZGcnIwkSbRo0aLGx19c3sVdADg06mPzcwuCcOuJjY0FKLOe95XoSHrkHaSbehqvJs3LPE7h3bTiM7je3t4l2vgKgnBzqtbMbX5+PgcOHCA5OZk+ffrg4+NjrXHdtFRFRZLNb4fl5+dz9epVwNjGuDSurq48/vjjhIeHs3TpUuLj45k9ezajR48mIiKixExF4axtcHAwTk62zXdVVYX868GtY6PeNj23IAi3psKZ27KC23YxK/j38m7U258FWpV7rK5duxIeHs7p06dJT0/H09OzRP6uIAg3ryr/S543bx6NGjWib9++3H333Rw5cgQwruD38fHhu+++s9ogbyaHlx3l1/tXc/K3aNO2y5cvo6oqLi4uuLm5lfv93bt355133qFz584YDAZ++eUX3nvvPS5eNDaDUBSF6Ohotm3bBlArs7a61GiU/DQkrTP2Ph1tfn5BEMqXmJjI/fffT4MGDXBycqJ9+/Zm5RmzsrKYPn06AQEBODk50aZNGxYuXFiLIy6fwWAgLi4OKD24VQtyUC8Zu0HatRtu0TFlWaZVq1b07Nmz1PxdQRBuXlWauV28eDHPPPMM9957L0OHDuXhhx82Pefj48PAgQP53//+Z7b9VpERn0HO1VyzxQyFgWmjRo0sKvvl4eHBjBkz2L17N8uWLePcuXO8+eabdO/enRMnTpiVFtu9ezdhYWE2vZWWd3EnAA7+PUptbSkIQu1JS0ujT58+DBgwgL///htfX1/OnDljdht+5syZ/PPPPyxbtoyQkBA2bNjAk08+SePGjRk1alQtjr50SUlJFBQU4OTkRMOGDUs8X6BxRvP0VuQLB5C8g2phhIIg1CVV+qj68ccfc9ddd7F8+XLuvPPOEs936dKF48ePV3twN6Nez/Rg2CdDCOobaNpWNLi1lCRJ9O7dm3feeYe2bdv+f3v3Hd9U9T5w/JM0Sfemm7YUCgiyh4wiZRcQBRVQqApuAQVBEVFBhjIUxA0qPysow4EgyBCQIZY9FUSGTFlldNCdcX9/5JtAbEsHbZq2z/v1yktz77nnnlzak6fnnvsc9Ho9iYmJeXLmpqen233JSFN2MqicZEqCEA5o+vTphIeHk5CQwF133UVUVBTdunWjVq1a1jJbt25l0KBBdOjQgRo1avDMM8/QuHFjdu7cWY4tL5hlSkJkZGS+I6xf/wMhS5yZmNrG3k0TQjigEo3cHj9+nOHDhxe438/PzzrHtKrReegIqGc799jyMFlxglsLPz8/Ro4cyfDhw8nMzCywnD2XjPRp8QpejYaAKu8Ty0KI8rV8+XLi4uLo168fmzdvJiwsjKFDh/L0009by7Rt25bly5fzxBNPEBoayqZNmzh69CizZs3Kt86cnBxycnKs79PS0gDQ6/V2yZRiSZEYERGR93xGPVsuOJOc6wQmI3q9qczbU1Ysn02yz9iHXG/7Kcm1vp1/lxIFtz4+Ply5cqXA/X/99VeBD05VRSUZub3ZsWPHbhnYwo0lI++449YPUpQWte7Wc4eFEOXjxIkTzJ49m1GjRvHaa6+xa9cuhg8fjk6nY9CgQQB89NFHPPPMM1SvXh2NRoNareaLL76gffv2+dY5depUJk6cmGf72rVrcXNzK9PPA3DgwAHAPOVi1apVNvtqJm1i5tV9dAwaiP5MDVb9m1Xm7Slr69atK+8mVClyve2nONe6sLjnVkoU3Pbs2ZPPP/+coUOH5tl36NAhvvjiiyo53/bq0atc3H+JgPrVCGwQCJgfALudkVso+lKQ9lgy0mTIRq1xKfPzCCFKxmQy0aJFC6ZMmQKYl0I/ePAgc+bMsQlut2/fzvLly4mMjOS3335j2LBhhIaG0qVLlzx1jh07llGjRlnfp6WlER4eTrdu3fDy8irTz6PX61mxYgUA999/f96sPF9+BbmXiG/gCU06lmlbypper2fdunV07dq1WKtVipKR620/JbnWljtEJVGi4Patt96iVatWNGjQgHvvvReVSsW8efP48ssvWbJkCSEhIYwfP77Ejaqozu+5wIH5f1Kzcw1rcHvt2jVyc3PRaDQlTpVW1KUgy3rJSMVk4NKKe9F4RuAXMx0nV0n9JoSjCQkJoX79+jbb6tWrx5IlSwDIysritddeY+nSpdxzzz0ANGrUiP379zNjxox8g1tnZ+d8l/jWarVlHhScPXsWo9GIh4cHwcHBNg/lKldOYLj0N6g1aO7sjqqSBCj2uK7iBrne9lOca307/yYlmqAZGhrKnj176N69O99++y2KovD111+zYsUKBgwYwPbt26tkzlvvCG8iYyMIanzjaV7LlISgoKB8V9UpCkdZMjL3yh8ouWkY0/9F7Xzr9gghykdMTIx1aW6Lo0ePEhkZCdyYJ/vf+flOTk6YTI43X/XmxRv+m23GdHAlE3xGMKXG25w1SZ8khDAr8SIOgYGBzJ07l7lz53L58mVMJhMBAQFVOldgREw4ETHhNttud74tOM6SkTdSgLVBpZaHyYRwRCNHjqRt27ZMmTKF/v37s3PnTj7//HM+//xzALy8vIiNjWX06NG4uroSGRnJ5s2bmT9/Pu+99145tz6vghZvUBSF3ENrmO25mDSjJ71yIMKjPFoohHA0tx0NKYqCoiioVKoi5XAtTGVLPp584SihXlnUCoLca3/neRkyLhapHkdYMjLn/P9WJQuVFGBCOKqWLVuydOlSFi1aRIMGDZg8eTLvv/8+8fHx1jKLFy+mZcuWxMfHU79+faZNm8bbb7/Nc889V44tz58luK1Ro4btjot/Ybj6L1NS3+eRGnqayMCtEOJ/Sjxy+9dffzF+/Hh++eUX6xNtbm5uxMXFMWHCBBo0aFDsOity8nFjrhHFpKBxuXFJDRkXaaNbwN1tTWA8yZV18/MeqNYR2PN7NO6FZ5cozyUjDRkXMKSdAJUa5+DWZX4+IUTJ9erVi169ehW4Pzg4mISEBDu2qGRycnI4f/48kHfk1nRwFS7k8lT1azwXK/MlhRA3lCi43bJlCz169MBkMtG7d2/rXM8jR46wfPlyVq9ezZo1a7j77ruLVe/Nycct/tuh3Zx8HOCZZ57hs88+Y+fOneUa3F7cf4mNEzYT2jyETpPNbTPlpKBRFzKHzZSLKScFihDcwo0lI+0t54J51Fbn3xC1rmyfjhZCCIDTp0+jKAq+vr55HphV1e2MKisFdb1u5dM4IYTDKlFwO3LkSAIDA9m8eTPh4bZzTM+ePUv79u0ZNWoUu3btKla9FTn5eMq/KaCAk4uTtd709OtFOtZgMKBy8CTSWed+B0Ab1Drf6ybJsO1Lrrf92Dv5uLjBMiWhQYQ/ygXbVS9z1C6sjXyW9s4Z+KSeR+UdWh5NFEI4oBIFt4cOHWLy5Ml5AluA8PBwhgwZwoQJE4pdb4VOPu4E7oNcuWa4Yk0yrk8+QlvXwg9NTEwkU/1P6bWlDPgZA/BV38HBY0ay/llVYDlJhm1fcr3tx17Jx8UNJ0+exM8ph4HXF2P44hubfYnOd/Fg8DzCDBc4dqk72mGrJcAVQgAlDG4jIyNtRkP/Kzc3N9/AtzCVLfn4ns3pUIRViGNiYtD61i3Ttty+ngAU1EpJhm1fcr3tx97Jx8UNp06dwsPJgJNiyLMvXe1Gbf1Jmuf8icqQC5nJIMGtEIISBrfjx49n5MiR3HPPPTRp0sRm3759+/joo494//33i11vZUs+fu1aMhFFKKfRaCpNgCLJsO1Lrrf92Cv5uDDLzMzk0qVLROjy339P1ibuydpELnKthRC2ihTcDh8+PM+2oKAgmjdvTtu2bYmOjgbg2LFjbNu2jQYNGrB9+3YGDBhQrMZU1OTj2SnZ7J//B741fKh7342FFK5cuQL+5dasUqEoCpn/LMU5sDlOnhGlku5NCCEKY5lv6+vrc8tyOmR+sxDCVpGC248//rjAfYmJiSQmJtps+/PPPzl48CAffPBBsRpTUZOPJ59M4fjqf/AM86x0wa3x+hlS90wHtZbg+9eh0hRhErEQQtwmy8pkoaFhkHJj+w9u3XnJ73Xeu/YWD2b+Ui5tE0I4tiIFt/YaFbUkHx87diyTJk0iKioq3+TjY8eOJT4+nmvXrhEZGVnuycfdA9xoMOBOtDfnuDUYOJeUhj5ahdZJKfhglRNqZ5+yb2QJWVclC2iKWgJbIYSdWEZuQ0NDrcFtktqP5/0mkqr25LFqM7l8zY/n0heVXyOFEA6pxIs4lJWKmHzcq7oXTR5rZLPt8uXLXMtw4pMd9Zk8fjRXNw0DQwY+rSej8Ywg6/QaMo4uAo0rKqcCJpU5gOz/5bd1DpFVyYQQ9mMNbsNC4S9QgOF+b5KudgOVChNOfOPRW4JbIUQetxXcnjx5ktWrV3P69GnAnEWhR48eeRZeqIouXLgAgLtfJBq3ADBkgEqNa/VYVE7OaH2iyUnagyHlKGn7P8C3dd50ZeXNpM8g9/I+AFxCY8q5NUKIqiI1NZVr166hUqkICQkB4Ae3Hvzkbrtgwx7nxvzg1p2+mWvKo5lCCAdV4uD2pZde4oMPPsgzZUGtVvPiiy8yY8aM225cRZF6JhWPEA+ctE7WbZbgNjg4GEP6WUCFxiMclZM5c4NKrcGnxViu/PokWafX4B7dF121huXR/ALlXNoFJgNOHtXReBYl74MQQtw+y3zbkJAQXHyDOa8L5gW/CagUE4rqxgPFKsXEC34TaG/YT6ibb751CSGqHnXhRfKaOXMms2bN4oEHHmDbtm2kpKSQkpLCtm3b6Nu3L7NmzSpwxbDKJud6DiueXcXi+7/HkH0jF6MluA0JCcE5oCnBD27Er73tQ286//p41BuMd8vX0frfadd2F0XOefN8W5cQGbUVQtiPZUpCjRo1wCuUEc1Wka7xtAlsARSVmusaL15stkoWcBBCWJVo5PaLL77gvvvu47vvvrPZ3qpVKxYvXkx2djafffYZI0eOLJVGOrLMy5loXTVo3XVobnqgzBLchoaaO1y1xhW1R/U8x3s1fNY+DS2BnKQ9ADiHynxbIYT9WEZuo6KiOJQCyy4U/DCrUVGx9IIrh5LhThm8FUJQwpHbU6dOERcXV+D+uLg4a+dU2fnW9KX/kr7c80l36zZFUWxGbovKpM/AkHGx1NtYUgHdF+J390ycA5qWd1OEEFWEoijWkduoqCju9IE+EeBUQIptJxXcHyGBrRDihhIFt4GBgRw4cKDA/QcOHCAgIKDEjapoVCoVzl43VkFLTU0lKysLlUpFQEA1rmx6gZTd0zHlphdYR+6VP0la/TAp28ehKOW3IMXN1BoXXELbWecJCyFEWbt27RppaWk4OTkRERGBSgWftgYPJyMqbNMqqgBPLXzSunzaKoRwTCUKbvv168fcuXOZNm0aGRkZ1u0ZGRlMnz6duXPn8tBDD5VaIysay6htQEAA6tyr5F7aSebJFag0LgUe4+QWiKK/Tu6VP8g8sdxeTRVCCIdiGbUNCwuzLmMc6AqfBCaiYDt8qwCz25j3CyGERYmC28mTJxMbG8trr72Gr68vNWrUoEaNGvj6+jJ27FhiY2OZNGlSabfVIf0+fSs7P91N5rUs67abpyQYUv8BQONVA5W64CnOTm5BeDYwz79NO/AxxuyrZdjqWzPmpJD0y6Ok/fEpislYbu0QQlQ9N09JuNmDWWvpnbEWJ8x3tizTEfrVsHcLhRCOrkTBrZubG7/++itLly7liSeeoF69etSrV48nnniCZcuWsX79etzc3Eq7rQ7HkG3g1KbTHF1xDLXTjUt5c3Cr/19wq/WuVWh97rX7ofWti6K/Ttq+98ukzUWRc2E7hpSjZJ9PRKV2KvwAIYQoJTc/THYzVdJRnr2+EA8n8x/cMh1BCFGQYmdLyMzM5JFHHuHBBx8kPj6e3r17l0W7Kow2o1qRfiEdF+8b81JtR27XA6DxrlloXSq1Bu8WY7my/gmyzqzFtcY9uITYv/fO+d+qZLJwgxDCnv77MJl1u8lI8pWL9Az5DSejQqALfNhKpiMIIfJX7JFbNzc31q9fT2ZmZlm0p0LRuGio1bUmjf+z9O7Fi+aMB8UduQXQ+dXDPbofAKl7pmMyZJdiiwunmIxkX9wGyJK7Qgj7unTpEllZWWi1WmsaRQCSz3Ja8cffmEy4O5x/CPrWKLdmCiEcXImmJbRr145t27aVdlsqhZycHK5eNc+XDQ4KwHD9FFC0kVsLz4bPonYNROtXH4w5ZdHMAumvHULJTUOl80Ln38Cu5xZCVG2WUduIiAg0mhs3FpWkozTW/81Z41Ps6FVATjAhhPifEgW3H3/8MVu2bOGNN97g33//Le02VRhXjlwl7dx1TMYbqbsso7YeHh64OmWjdvZFpXHDyS24yPWqte4Exn2DX9u3UTt7l3q7byXbsipZcKtbPgAnhBClraD5tkrqeUCFOrAu/gUnnRFCCKCEK5Q1btwYg8HA1KlTmTp1KhqNBmdn21yoKpWK1NTUUmmko9o6cztpZ9Po9FYHQpubF2u4eb6txj2Y4Pt+xqTPRKUq3t8R/w1qFUVBpSr7EYvs/823lSkJQgh7KyhTglPrwaib9YdcmQ4nhChciYLbBx980C6BliNTTApaNy1Ozk54R3hZt+e3MplaW/LMEcasy6TufQ/n4Ltwr3V/yRtcBIrJgNYnGlP2FZyD25TpuYQQ4mZGo5HTp08DUKNGDZt9SVlw/0Y37qrmxnstoYp//QghClGi4Parr74q5WZUPCq1ih7vd0MxKdycV7wky+7eStbZjWT/u4GcSztxCW2Pk6t/qdSbH5Vag2+rN+02SiyEKBvnzp1jzJgxrF69mszMTKKjo0lISKBFixbWMocPH2bMmDFs3rwZg8FA/fr1WbJkCREREeXS5vPnz5Obm4uLiwvBwbbTuHZfhR2XITVXAlshROGKFdxmZ2fz008/cfLkSapVq8Y999xTakFcRaVS2/a0N2dKuLLhWVRqHd7NR6PxLNkXhnv0g2SdWok++W/S9s/Ct81bt93mwkhgK0TFlZycTExMDB07dmT16tUEBARw7NgxfH19rWX++ecf2rVrx5NPPsnEiRPx8vLi0KFDuLiU34RWy3zbGjVqoFbfmMalXPiLJus+Zn61e1Hd2aOcWieEqEiKHNwmJSXRtm1bTp48iaKY1/d2c3Nj2bJldOnSpcwaWJGYTCZrcBsc6Efu3wcABZXWvcR1qtRO/8t9+zhZZ9b9L/dt6U8ZUIx6DGkn0PjUkeBWiAps+vTphIeHk5CQYN323zmsr7/+Oj179uSdd96xbqtVq2jpCsuKZb7tf6ckKBf/IuDsRvprstHUlOBWCFG4Ige3kydP5tSpU4wcOZJOnTpx/PhxJk+ezLPPPss///xTlm10SLs/30va2TTqP3gHwU3Mt9CuXr2KXq9Ho9Hgo0vnKgpqZx/Uzn63dS6d3x241uhF1snlpOx8C9+2U1E56WzKqJ190LgXPSPDf+Ve2c/VTc+j9a1LQLf5t9VeIUT5Wb58OXFxcfTr14/NmzcTFhbG0KFDefrppwHzH+ErV67klVdeIS4ujn379hEVFcXYsWPp06dPvnXm5OSQk3MjLWFaWhoAer0evV5fKu0+ceIEYE4DZlPnhcMAKNWiS+1cjsry+Sr753QUcr3tpyTX+nb+XYoc3K5du5bHHnuMGTNmWLcFBQUxcOBAjhw5Qt26dUvciIro0oFLJJ9Ioc49ta3bLPNtg4KCMKaZRyE03jVveyTUkHGRrNNrADBlX+HqhqfzFlLrCOz5fYkDXEsKMI1P7UJKCiEc2YkTJ5g9ezajRo3itddeY9euXQwfPhydTsegQYNISkoiPT2dadOm8dZbbzF9+nTWrFnDAw88wMaNG4mNjc1T59SpU5k4cWKe7WvXri2VpdaNRiNnzpwBzCO4SUlJ1n3RJ/5ir9s9eFx0Rb1q1W2fqyJYt25deTehSpHrbT/Fuda3s1hYkYPbM2fOMGbMGJtt7dq1Q1EULl26VOWC27ueb0nKyRT8694Ylb35YTJ92v9WJvO6/Vt9ppwUMOUWUijXXK6Ewa11yV1JASZEhWYymWjRogVTpkwBoGnTphw8eJA5c+YwaNAgTCZzXu7evXszcuRIAJo0acLWrVuZM2dOvsHt2LFjGTVqlPV9Wloa4eHhdOvWDS8vrzzli+vUqVOsWLECd3d3+vbte2NAQFH4afY2HvebRiOPLHb0rNy5t/V6PevWraNr165otdrybk6lJ9fbfkpyrS13iEqiyD1FTk5OnocNLO8NBkOJG1BRBdSrRkC9ajbbbg5uDam/AcVbmay8GNL/xXD9NKiccA5qVd7NEULchpCQEOrXr2+zrV69eixZsgSAatWqodFo8i3z+++/51uns7NznlzmAFqttlSCgrNnzwLmucE63Y0pV8r1JDS5qbTMOUDLOnei1Vbu4NaitK6rKBq53vZTnGt9O/8mxeopTp06xd69e63vLYs0HDt2DB8fnzzlmzVrVuKGVUQ2we0V8/wxjXf5PqRRFNnnzaO2uoDGqHUe5dwaIcTtiImJ4ciRIzbbjh49SmRkJAA6nY6WLVvesoy9FbgyWdIx7snaxD1up9G2qRpTEoQQt69Ywe24ceMYN25cnu1Dhw61eW/Jk2o0Gm+vdQ7q6tGrZKfm4FfLF1c/V+t2a3AbFIBTVhiKIRutHUduM0+uQK3zQONRvVjH3ZiSEFMWzRJC2NHIkSNp27YtU6ZMoX///uzcuZPPP/+czz//3Fpm9OjRPPTQQ7Rv356OHTuyZs0aVqxYwaZNm8qlzQVlSiAnHTwDUQVVrWlvQojbU+Tg9ua0MmWpIiQfP/rzMf5Zd5JGjzSgUXxDANLT07l+/ToAwaHVcak52+6LIWQe/4HM4z+g9b0D14guuEf3Q6XJm7fSkHHRPD8XUIw55FzaBYDaLYTca3/fduYFIUT5admyJUuXLmXs2LFMmjSJqKgo3n//feLj461l7r//fubMmcPUqVMZPnw4devWZcmSJbRr187u7c3JyeHcuXNAPiO3d8ShqRdX+DMHQghxkyIHt4MGDSrLdgAVJ/m4q78b3pHe+NTwsW6zjNr6+flZ22LvfLFavzvRJ/+NPvlvjFlJuNcZaN1nMmSh1rhiyLhI0qp++X5ZpGx7zfw/t5l5QQhRvnr16kWvXr1uWeaJJ57giSeesFOLCnbmzBkURcHHx8emrwf4/hS8uBMeqaVjRsvyaZ8QouJxqNn5FSX5eJNBjWgyqJHNtpvn2yomIyq1k13bBODd/BWc3ILI/ncjKDfaoJiMJK3qh8Y9BK1/ozLPvCCEEEVV4JQEzMvuXsmB3Mo5w00IUUYcKritqMnHAetttaCgIK5segFjxnk8m41BF9iikCMLZ3JyB7Xu1kGpWofJyR2Vkwe6yHuBm5ImJx/BlHWZ3KzL5F75o0jnNBgMqEqQbFmSYduHXG/7sXfy8arGEtzmmZJw+R/e3PIMD4b2wLfey+XRNCFEBeVQwW1FTD5u8ccf5qDxypUrZOoOoyWTrTsPkKlOKuTIotFpn0ajZBW436ByJXfT3gL3a52fx8/4N9UMB3DjcqHnS0xMJFNd/JXnJBm2fcn1th97JR+vagoMbpOO4JJ+npYZu9F4l0fLhBAVlUMFtxUh+fjx1f9w5KdjRHWKpH7/etbtW7eaMw507dAS7dHFgIrY7vH5PtBVnvTJR0jZ+GSh5WJiYtD6Fv0JZUmGbV9yve3H3snHq5LMzEwuXboE5BPcXjpq/p+gOvZulhCignOo4LYiJB+/fjad9PPpmHJN1uP1ej1XrlwBINgzl1zAySMMnatnsesva4qmaP/kGo2mRNdHkmHbl1xv+7FX8vGqxJLfNiAgAA8P2xzbGy7CNu8hdPWsQ5tyaJsQouJyqOC2IiQfbxTfgIiY6rj635jSkJSUhMlkwsXFBRdjErmAtgIs3iCEEOXpVg+TLUuP4jOf3lw3XJDgVghRLA4V3FaE5OMuPi4EN7HNImCzMlmaZWUyx192VwghylOBK5PlpNMubROprio6RXUqh5YJISoydXk34GaW5OOLFi2iQYMGTJ48ucDk4++88w4NGzZk7ty55ZZ83OLixYuAObjVp5ofwnLUkVu1s48588ItC+nM5YQQogwV/DDZMfpmriEh9z161JQlwYUQxeNQI7fg2MnHr5+/zvndF/CL9iWgfoB1+80jtzq/O0FR0PhE2719RaFxDyaw5/fWFcryIyuUCSHKWlpaGlevXkWlUuWdVqaYUEW2BM/A8mmcEKJCc7jg1pFd/COJXbP3ENIsmM5vd7RutwS3wcHBeDe7dWDuCDTuwbJAgxCiXFlGbUNCQnB1dbXZd7Fac9wGzMenkJtMQgiRH4ealuDo3Pxcqd4mjKDGQdZtiqLYjNwKIYQo3K0eJntzH1RbBDMP2rlRQohKQUZuiyHsrlDC7gq12ZaSkkJ2djZqtZpq3joUYy4qJxluEEKIWynwYTJF4WKmEdAQXfxU5EIIISO3t8syahsQEEDmwY+5sKQDGceXlHOrhBDCcSmKUuDDZKRd4MdtDTmb/hBdQkzl0DohREUnI7dFZDKaO1m1k+3fA7aZEnaDYsTJNSDP8cIxGAwGcnNzy7sZt02v16PVasnMzJQFA8pYftdap9OhKeKCKCKva9eukZaWhpOTE+Hh4Tb7lEvmPObVnLLQamX8ReSvNPpy6Uftx979qPTORZR8IoVfRq2jWr1qdHuns3W7db5tcBCGtFMAaBw0DVhVpigKZ86csa4kVxkEBQVx/Pjx8m5GlZDfta5WrRoRERGoVKpyalXFZRm1DQsLQ6ezncalJB0DQBVY9OW/RdVR2n259KP2Y89+VILbIko7m4bJYAJFsdl+/vx5ACICtZCUi8rJBSd3ebDM0Vg6w7CwMDw8PFCrZURIlIzJZCI9PZ1z584B2G11xMqkoPm2AOPO1OS8/xSGeEErO7dLOD7pyyuHsu5HJbgtohodIgm4MwBDtsFmu2XkNsgjB5LMK5OpVPLL5kgMBoO1MwwOlhRo4vZ5eJgXFjh37hxhYWEyRaGYbrnsbs4dHPWozkNef9i5VcLRSV9euZRlPyo9chGp1Co8gtxttmVlZZGcnAyAl1MKOciyu47IMi/L8oskRGmw/Dzl5uZKcFsMiqIUnCnBkMu0K1PYrbuTllEPlkPrhCOTvrzyKat+VHrk23Dp0iUAvLy8UGWdBRx32V2B3L4SpUp+nkomKSmJzMxMNBoNYWFhtjuvnKBH1kZ6KHvQVBtaPg0UDk9+9yqPsvq3lOC2CIx6I/v+bz9eEd5Ed6uJWmP+x7h58QbnkPqonHTo/BuWZ1OFEMKhWaYkRERE5B2p0ehQNX0QlVojD+oJIUpMgtsiuH4+nb9/OorWVUPtHjdGZm9edte95n2417yvvJoohBAVQoH5bYH1uTXxb/sWDXzAyc7tEkJUHjK2XwSXDlzCydmJgDsDbEYTZNndqsdkMvH333+zfft2/v77b0ymsk0yP3jwYFQqFdOmTbPZvmzZsmKNbHXo0AGVSoVKpcLFxYU6deowdepUlP9k/xCirN0quH1uK9z1M/yeZO9WiarE3v04lG5f/uKLL+bZ/tVXX+Hj43Obraw8ZOS2ENkp2eyf/wfGHCNX/r5Cdko2Lj4uwI3gNqyaM4brZ3ByD0OllvGGymr37t0sXLjQ+hAhgK+vLwMHDqRFixZldl4XFxemT5/Os88+i6+vb4nrefrpp5k0aRI5OTls2LCBZ555Bh8fH4YMGVKKrS2e3NzcPHlOReVlMpk4ffo0kDdTQo4RartkcF3vRnN/mZIgykZ59eNQen25KJyM3N6Coijs+GgXhixz+i99poEdH+8CzJ205YGyarm7SFrVj9R975VbW0XZ2r17N5988olNhwiQnJzMJ598wu7du8vs3F26dCE4OJipU6cWWGbJkiXceeedODs7U6NGDWbOnJmnjJubG8HBwURGRvL444/TqFEj1q1bZ92fk5PDyy+/TFhYGO7u7rRq1YpNmzYBkJaWhqurK6tXr7apc+nSpXh6epKZmQnA2bNn6d+/Pz4+Pvj5+dG7d2/rk/FgHr3o06cPb7/9NqGhodSta07U/+mnn1K7dm1cXFwICgqib9++1mNMJhNTp04lKioKV1dXGjduzA8//FDs6yjK3/nz58nNzcXFxSXPHS9dTjIr9rbg338a40VmObVQVGbl2Y9D6fXlRTV79mxq1aqFTqejbt26fP311zb7VSoVs2fPpkePHri6ulKzZs1K07dKcHsLp387w9mt/6KYzLduFZPC2cR/Of2bOYm0wWBAq9Wi1ZsXctB65b3NJhyToijk5OQU6ZWVlcWCBQtuWd/ChQvJysoqUn3FnQrg5OTElClT+Oijj/j333/z7N+zZw/9+/fn4Ycf5s8//2TChAmMGzeOr776qsDPvmXLFv7++2+bUdPnn3+ebdu2sXjxYv744w/69etH9+7dOXbsGF5eXvTq1YuFCxfa1LVgwQL69OmDm5sber2euLg4PD092bJlC4mJiXh4eNC9e3ebZTJ//fVXjhw5wrp16/j555/ZvXs3w4cPZ9KkSRw5coQ1a9bQvn17a/mpU6cyf/585syZw6FDhxg5ciSPPPIImzdvLtZ1rErOnTvHI488gr+/P66urjRs2LDAL+7nnnsOlUrF+++/X+btskxJiIyMzPOUtHVlMq8gVDq3Mm+LqPgqUj8Opd+X38rSpUsZMWIEL730EgcPHuTZZ5/l8ccfZ+PGjTblxo0bx4MPPsiBAweIj4/n4Ycf5vDhw8U+n6ORaQkFyE7JZsdHu0AF3PwzrIIdH+2k1ovmQDY4OBhDqjnZuCy7W3Hk5uby3HPPlVp9ycnJDB1atNRFc+bMwdnZuVj133///TRp0oQ333yT//u//7PZ995779G5c2fGjRsHQJ06dfjrr7949913GTx4sLXcp59+yty5c8nNzUWv1+Pi4sLw4cMB86o/CQkJnDlzhtDQUABefvll1qxZQ0JCAlOmTCE+Pp5HH32UzMxM3NzcSEtLY+XKlSxduhSAb7/9FpPJxNy5c61zyBISEvDx8WHTpk1069YNAHd3d+bOnWsNrH/88Ufc3d3p1asXnp6eREZG0rRpU8A8mjxlyhTWr19PmzZtAKhZsya///47n332GbGxscW6jlVBcnIyMTExdOzYkdWrVxMQEMCxY8fyvQ26dOlStm/fbv03L2uWUfz8Fm8wXToKgCqwjl3aIiq+itaPQ+n25TczGAy4uLhY38+YMYPBgwdbP8+oUaPYvn07M2bMoGPHjtZy/fr146mnngJg8uTJrFu3jo8++ohPP/202J/NkcjIbT5spiP8948zxTw94Z9F5hGI6iF+GDPNc2+1soCDKEPTp09n3rx5ef6qPnz4MDExMTbbYmJiOHbsGEaj0botPj6e/fv3k5iYSI8ePXj99ddp27YtAH/++SdGo5E6derg4eFhfW3evJl//vkHgJ49e6LValm+fDlgvn3m5eVFly5dADhw4ADHjx/H09PTeryfnx/Z2dnWOgAaNmxoM2LctWtXIiMjqVmzJo8++igLFiywTnM4fvw4mZmZdO3a1aZd8+fPt6lT3DB9+nTCw8NJSEjgrrvuIioqim7dulGrlu0f3+fOneOFF15gwYIFaLVau7StoIfJFAXqH+1Bp6BvOOPf3C5tEaK8lFZffvNr0qRJRarrv+e0DBrc/F5Gbiup1NOpnN2a95aBhWJSyP1Hj7u/O1GBGjCA2jUAtbO3HVspbodOp2POnDlFKnvkyBFmzZpVaLmRI0da55AWdu6SaN++PXFxcYwdO9bmr/ii8vb2Jjo6GoDvvvuO6OhoWrduTZcuXUhPT8fJyYk9e/bg5GT7UKRlBRmdTkffvn1ZuHAhDz/8MAsXLuShhx6y5ipNT0+nefPm+d76CwgIsP6/u7vtSn+enp7s3buXTZs2sXbtWsaPH8+ECRPYtWsX6enpAKxcuTJPwv+SjJpUBcuXLycuLo5+/fqxefNmwsLCGDp0KE8//bS1jMlk4tFHH2X06NHceeedhdZpuRVrkZaWBoBer0ev1xepXQaDgbNnzYvdhIeH2xz3byacMPlz2tkb/4BLRa6zsrF87qr6+Qvz3+tSEftxKN2+3CIwMLDE7XEE+fUlt/N7IMFtPrwjvQlvW51/t5+zzre9mUqtIsM3nQxtBsFeergmo7YVjUqlKnJw1KBBA3x9ffM8hHAzPz8/GjRoUOYr50ybNo0mTZrYdL716tUjMTHRplxiYiJ16tTJE6haeHh4MGLECF5++WX27dtH06ZNMRqNJCUlcffddxd4/vj4eLp27cqhQ4fYsGEDb731lnVfs2bN+PbbbwkMDMTLy6tYn0uj0dClSxe6dOnCm2++iY+PDxs2bKBr1644Oztz5swZmYJQRCdOnGD27NmMGjWK1157jV27djF8+HB0Oh2DBg0CzCNHGo3GOi2lMFOnTmXixIl5tq9duxY3t8Lnx1qW3DUYDDg5ObFjxw6b3xWTycSuC/M46RTGriNNyDi9qoiftnK6+UFPcYNWqyUoKMj6vqL241B6fXlBLHVZfuctddWvX9+m3Pbt23nsscds3lumhdlTYmJinmDWcgevJCS4zYdKpaLVCy25eOAS+kx9njm3WjcNhz3/gmzw0ZlHMDReMt+2slKr1QwcOJBPPvmkwDIDBgywS4fYsGFD4uPj+fDDD63bXnrpJVq2bMnkyZN56KGH2LZtGx9//HGhc6aeffZZJk+ezJIlS+jbty/x8fE89thjzJw5k6ZNm3L58mV+/fVXGjVqxD333AOYRxyCg4OJj48nKiqKVq1aWeuLj4/n3XffpXfv3kyaNInq1atz+vRpfvzxR1555RWqV6+ebzt+/vlnTpw4Qfv27fH19WXVqlWYTCbq1q2Lp6cnL7/8MiNHjsRkMtGuXTtSU1NJTEzEy8vLpuMWZiaTiRYtWjBlyhQAmjZtysGDB5kzZw6DBg1iz549fPDBB+zdu7fI+TXHjh3LqFGjrO/T0tIIDw+nW7duhf4hs3fvXr799ltrUGE0GtmyZQsPPfQQzZo1MxdKPgt/HqGB00noMwPUVfOrSa/Xs27dOrp27Wq3qSIVSWZmJsePHy/RsY7Uj0Pp9uX5GT16NP3796dp06Z06dKFFStW8OOPP7J+/Xqbct9//z0tWrSgXbt2LFiwgJ07d+aZC2wPMTExef5QttwhKomq2YMUgYuPC61eaMnv07ba7lCg0VMNWb14JQC+0d0x+YfiHGD/v3SE/bRo0YJhw4blyY/o5+fHgAEDyjw/4s0mTZrEt99+a33frFkzvvvuO8aPH8/kyZMJCQlh0qRJhd7u8vPz47HHHmPChAk88MADJCQk8NZbb/HSSy9x7tw5qlWrRuvWrenVq5f1GJVKxYABA3jnnXcYP368TX1ubm789ttvjBkzhgceeIDr168TFhZG586dbxkA+fj48OOPPzJhwgSys7OpXbs2ixYtst4unzx5MgEBAUydOpUTJ07g4+NDs2bNeO2110pw9Sq/kJCQPKMz9erVY8mSJQBs2bKFpKQkIiIirPuNRiMvvfQS77//vk3qNgtnZ+d8R8i0Wu0tg7Ddu3fne9s4OTmZOXPmMGzYMFq0aIHi5omp0yjIuY6Ts2tRP2qlVdh1rapu95o4Uj8OpdeX56dPnz588MEHzJgxgxEjRhAVFUVCQgIdOnSwKTdx4kQWL17M0KFDCQkJYdGiRXn6D3vI72f+dv69VUoVW6IoLS0Nb29vUlNTCx1xUBSF39763To9QaVWUb1NGKEPBTNlyhT8/f2ZMWOGnVru2PR6PatWrbI+dORIMjMzOXz4MPXq1SvSLdRbMZlMHD16lJSUFHx8fKhTp47d/tIXjuVWP1fF6WdK28CBAzl79ixbtmyxbhs5ciQ7duxg69atXL161boAjUVcXByPPvoojz/+eJHmGxbl85lMJl5++eVCbwO/8867vHlATWM/uC8cnKvwOjiO3I86gtLqy6UfN1OpVCxdupQ+ffqUWxvKqh+VkdtbsJmekKFH66ah1fMt2fnHTsCcBkxULWq1mjvuuKO8myFEgUaOHEnbtm2ZMmUK/fv3Z+fOnXz++ed8/vnnAPj7++Pv729zjFarJTg4uEiBbVEdPXr0loEtwLVr1/j1z5NM+7MWOjWkDCy10wtRIOnHK7+q96dKMbn4uNB6+F24+LrQavhduPi4cP68edGGGiHuZJ9PxJh5qZxbKYQQZi1btmTp0qUsWrSIBg0aMHnyZN5//33i4+Pt2o6UlJQilbuelsYzQed5OPQ6WrWpbBslhKgSHG7k9ty5c4wZM4bVq1eTmZlJdHQ0CQkJ+c6Fee655/jss8+YNWsWL774Ypm1KbJ9BJHtb8xPu3jxIgC1fK5xbcsoXELvxu9umZ4ghHAMvXr1spkrXZj85tneLh8fnyKVa+Cn5r7NXUExQavfwCOg8IOEELetMs9Kdajg1pFX1rmZZb6ar/N1SJeVyYQQ4r/q1KlTpNRL0T4qTIoJ3HzBvZodWyiEqKwcalqCI6+sY6HX67l8+TIALsYkADSS41YIIWxYUi/dSv+HB3Du/DkUQBVYt8ipyYQQ4lYcauTWUVfWudm5c+dQFAVXVxeUjP+the4eWeVXlHHklXUcsU2i8ijtlXUqk8JSL7nWakHN5VArtD6HAvOubCeEECXhUMGtI66s81/nzp0DoJqXGkWfjgk1637/C0V1pNh1VUaOuLLOf1e1EaI0lfbKOpVNixYtaNasWb6pl346A04YCTZeRh1UepkahBBVm0MFt462sk5+Vq40L97QtI45lY7WM4IeXe8tdj2VjSOvrHM7q9oIUZjSXlmnMioo9VLvCEi60o0r2QoEflAOLRNCVEYOFdw60so6Bbl0yZz2q7qvEXJB6xPtcMFceXLElXUcrT2icintlXWqEiX9Cm4Z54lAhSowurybI4SoJBwquI2JieHIEdvb+0ePHiUyMhKARx99lC5dutjsv3llHXuwpAHThcbiE9QOJ1d5uldUfKdOnSIqKop9+/bRpEmTIh0zePBgUlJSWLZsWZm2TVRiOlec+r6PknYBlVaW3RXiZiXpl4WZQ2VLGDlyJNu3b2fKlCkcP36chQsX8vnnnzNs2DDAvLJOgwYNbF5lsbJOQRRFsaYBC4xoiFuNHjgHtSzz84ryZ8i4SO61vwt8GTIulsl5Bw8ejEqlQqVSodPpiI6OZtKkSRgMhtuq87/LLYaHh3PhwgUaNGhQKm3VarVERUXxyiuvkJ2dXeI6ReV1MBn6JbozWxWHU+vB5d0cUQUoqedRLhwq+JV6vtTPaekTC3pNmDCh1M9ZHKdOnUKlUrF///48+zp06FCmawiUJYcaubWsrDN27FgmTZpEVFRUuaysU5Dk5GRycnJwcnIiIEASjVcVhoyLJK3qB6bcggupdQT2/B6Ne+kvydy9e3cSEhLIyclh1apVDBs2DK1Wy9ixY4tVj9FoLHCuupOTU6ksJ21pq16vZ8+ePQwaNAiVSsX06dNvu+6Sys3NRafTldv5Rf62JsGyM5Cuh2H1yrs1orJTUs9j+KQHGG7Rj2t0aIatRuVdevnzLQNiAN9++y3jx4+3uUPt4eFRaucSNzjUyC2YV9b5888/yc7O5vDhwzZpwPJz6tQpu/1lYfkhjQzxIvvEEnKS9tnlvKJ8mXJSbh3YAphyzeXKgLOzM8HBwURGRjJkyBC6dOnC8uXLee+992jYsCHu7u6Eh4czdOhQ0tPTrcd99dVX+Pj4sHz5curXr4+zszNPPPEE8+bN46effrKOHGzatCnPX+9Go5Enn3ySqKgoXF1dqVu3Lh98UPgDP5a2hoeH06dPH7p06WKTQcNkMjF16lRrvY0bN+aHH36w7qtevTqzZ8+2qXPfvn2o1WpOnz4NmJd1feqppwgICMDLy4tOnTpx4MABa/kJEybQpEkT5s6dS1RUFC4uLgD88MMPNGzYEFdXV/z9/enSpQsZGRnW4+bOnUu9evVwcXHhjjvu4NNPPy3mv5QojvbBMD3kDwZ5HkbJlewSooxlJt86sAXz/syCFx0pieDgYOvL29sblUplfR8YGMh7771H9erVcXZ2pkmTJqxZs6bAuoxGI0888QR33HEHZ86cAeCnn36iWbNmuLi4ULNmTSZOnGhzZ0+lUjF37lzuv/9+3NzcqF27NsuXLy/RZ0lOTuaxxx7D19cXNzc3evTowbFjx6z7Ld85y5Yto3bt2ri4uBAXF8fZs2dLdL7b4XDBrSOzBLf1qjuRtu890va/X74NErfNZMgq8KUYcwqvoJj1lgZXV1dyc3NRq9V8+OGHHDp0iHnz5rFhwwZeeeUVm7KZmZlMnz6duXPncujQIT788EP69+9P9+7duXDhAhcuXKBt27Z52/+/QPP777/nr7/+Yvz48bz22mt89913RW7nwYMH2bp1q82o6dSpU5k/fz5z5szh0KFDjBw5kkceeYTNmzejVqsZMGAACxcutKlnwYIFxMTEWOfe9+vXj6SkJFavXs2ePXto1qwZnTt35tq1a9Zjjh8/zpIlS/jxxx/Zv38/Fy5cYMCAATzxxBMcPnyYTZs28cADD1iXn1ywYAHjx4/n7bff5vDhw0yZMoVx48Yxb968In9eUTx1PQ0M3/Uofdc9AOlXyrs5ogJTcjMLfhlK3o8XVGdp+eCDD5g5cyYzZszgjz/+IC4ujvvuu88mYLTIycmhX79+7N+/ny1bthAREcGWLVt47LHHGDFiBH/99RefffYZX331FW+//bbNsRMnTqR///788ccf9OzZk/j4eJv+sqgGDx7M7t27Wb58Odu2bUNRFHr27GmTCjEzM5O3336b+fPnk5iYSEpKCg8//HDxL85tcqhpCY7OEtyG+RpBkWV3K4OLSzoUuM85pC3+7WeVqN6kn/vkO5Ib+tCOEtUH5jnfv/76K7/88gsvvPCCzR2LGjVq8NZbb/Hcc8/ZjDjq9Xo+/fRTGjdubN3m6upKTk7OLachaLVam/zQUVFRbNu2je+++47+/fsXeNzPP/+Mh4cHBoOBnJwc1Go1H3/8MWDunKdMmcL69etp06YNADVr1uT333/ns88+IzY2lvj4eGbOnMmZM2eIiIjAZDKxePFi3njjDQB+//13du7cSVJSkjULyowZM1i2bBk//PADzzzzDGCeijB//nzr9KG9e/diMBh44IEHrEFyw4YNre1+8803mTlzJg888ID181q+LCw5tkUpu3YGjLmgdQPf6uXdGlGBGaY1L3CfKro9moGflazeD7vkO5KrHX+4RPX914wZMxgzZow1+Js+fTobN27k/fff55NPPrGWS09P55577iEnJ4eNGzfi7e0NmIPWV1991dpH1axZk8mTJ/PKK6/w5ptvWo8fPHgwAwYMAGDKlCl8+OGH7Ny5k+7du1vLtG3bFrXadrwzKyvL+iDbsWPHWL58OYmJidYBkQULFhAeHs6yZcvo168fYP7O+fjjj2nVqhUA8+bNo169euzcuZO77rqrVK5bUUhwWwyWTAn+LumQBVpZdlfYgSVg1Ov1mEwmBg4cyIQJE1i/fj1Tp07l77//Ji0tDYPBQHZ2NpmZmda8qzqdjkaNGpXovJ988glffvklZ86cISsri9zc3EKf2O3YsSOzZ88mIyODWbNmodFoePDBBwHzaGpmZiZdu3a1OSY3N5emTZsC0KRJE+rVq8fChQt59dVX2bx5M0lJSdaO88CBA6Snp+Pv729TR1ZWFv/884/1fWRkpM28+MaNG9O5c2caNmxIXFwc3bp1o2/fvvj6+pKRkcE///zDk08+aTMNymAwWL9EROn6NwOOHbtMY5UbXoHRqFRyE1FULWlpaZw/f56YmBib7TExMTbTrAAGDBhA9erV2bBhA66uN7KKHDhwgMTERJuRWqPRmOd74ObvAHd3d7y8vEhKSrI5x7fffku9eraT329+3unw4cNoNBpr0Armh/zr1q3L4cM3gn2NRkPLljcetL/jjjvw8fHh8OHDEtw6KsvIrZtivoUmI7cVX/CDmwrcdztfuIG9lpX42P+yBIw6nY7Q0FA0Gg2nTp2iV69eDBkyhLfffhs/Pz9+//13nnzySXJzc62dmqura5EXPLnZ4sWLefnll5k5cyZt2rTB09OTd999lx07bj3y7O7uTnS0OV/pl19+SePGjfm///s/nnzySet84JUrVxIWFmZz3M25qOPj463B7cKFC+nevbs1mE1PTyckJIRNmzblObePj49NO27m5OTEunXr2Lp1K2vXruWjjz7i9ddfZ8eOHdZr9cUXX9h03JbjROlbegZGHm1Fz2oz+DFwQ3k3R1Rwmlf3FLxTXfLfYc3w9SU+tjT17NmTb775hm3bttGpUyfr9vT0dCZOnGi943Qzy7MGkDfvtkqlwmQy2WwLDw+39t0WNwfSFY0Et0WUlZVFcnIyTioT6hzzCK6M3FZ8ak3Z/PKWZr03B4wWe/bswWQyMXPmTOutpKLOh9XpdBiNxluWsdx6Gjp0qHXbzSOjRaFWq3nttdcYNWoUAwcOtD7UdubMGWJjYws8buDAgbzxxhvs2bOHH374gTlz5lj3NWvWjIsXL6LRaKhRo0ax2qNSqYiJiSEmJobx48cTGRnJ0qVLGTVqFKGhoZw4ccJhMrNUBdW5RvPcg6hk2V1xm1Q6t8ILOVC9AF5eXoSGhpKYmGjTHyYmJuYZ4RwyZAgNGjTgvvvuY+XKldbyzZo148iRI3m+H8pCvXr1MBgM7Nixwzot4erVqxw5csRm8S2DwcDu3butn+HIkSOkpKTkGRUuaxLcFpFlSkJkkA4UIyqtB2rXwHJulaiqoqOj0ev1fPTRR9x7770kJibaBIG3UqNGDX755ReOHDmCv79/vrfea9euzfz58/nll1+Iiori66+/ZteuXURFRRWrnf369WP06NF88sknvPzyy7z88suMHDkSk8lEu3btSE1NJTExES8vL+u8sRo1atC2bVuefPJJjEYj9913n7W+Ll260KZNG/r06cM777xDnTp1OH/+PCtXruT++++nRYsW+bZjx44d/Prrr3Tr1o3AwEB27NjB5cuXrR3uxIkTGT58ON7e3nTv3p2cnBx2795NcnKyzfLdonS8UA+eW/cQxtTzqAITyrs5QpSL0aNH8+abb1KrVi2aNGlCQkIC+/fvZ8GCBXnKvvDCCxiNRnr16sXq1atp164d48ePp1evXkRERNC3b1/UajUHDhzg4MGDvPXWW6Xa1tq1a9O7d2+efvppPvvsMzw9PXn11VcJCwujd+/e1nJarZYXXniBDz/8EI1Gw/PPP0/r1q3tOiUBJFtCkVmmJNQONT/5rfWuWaLbvaLiUTv7gLqQPKlqnbmcnTRu3Jj33nuP6dOn06BBAxYsWMDUqVOLdOzTTz9N3bp1adGiBQEBASQmJuYp8+yzz/LAAw/w0EMP0apVK65evWoziltUls7tnXfeISMjg8mTJzNu3DimTp1KvXr16N69OytXrswTNMfHx3PgwAHuv/9+m1tjKpWKVatW0b59ex5//HHq1KnDww8/zOnTpwkKCiqwHV5eXvz222/07NmTOnXq8MYbbzBz5kx69OgBwFNPPcXcuXNJSEigYcOGxMbG8tVXXxU7mBdFo+RkQMq/OGFCFVi7vJsjqgI3X9AU0o9rdOZydjJ8+HBGjRrFSy+9RMOGDVmzZg3Lly+ndu38fydefPFFJk6cSM+ePdm6dStxcXH8/PPPrF27lpYtW9K6dWtmzZplfWi2tCUkJNC8eXN69epFmzZtUBSFVatW2Ux7cHNzY8yYMQwcOJCYmBg8PDz49ttvy6Q9t6JSLLlwqoi0tDS8vb1JTU3Fy8uryMctWbKEn3/+mU4dYnjonlYoihHngCZl19AKRq/Xs2rVKnr27Jlnfk95y8zM5PDhw9SrV886v7K4DBkXb5nHVu3sUyYLOAjHdaufq5L2MxXF7X4+xWSApGMoV0+ivrNnGbSwYnLkftQR3G5frqSev3UeWzffUl3Aoar56quvePHFF0lJSSnyMWXVj8q0hCKyjNwGhUSgq9awkNKistG4B4MEr0LctvcPwfx/NDxbtx7P3ilLkwn7UXmHggSvVYJMSygiS3AbEhJSzi0RQoiKa/tl+CMZUgtZLEoIIUpKgtsiMBqNXLp0Ca2TiYC0n8j4ZymKYir8QCGEEDbeuwu+C15H79RlKOmXy7s5QohSMnjw4GJNSShLMi2hCC5fvozRaCTc34jp3xVcv7IV91r3l3ezhBCiwglxVbj3wBuQnQbRS8EjoPCDhBCiGGTktggsacBqh5kTzcviDUIIUULXL5kDW5UTVJNc4UKI0icjt4UwmUzWpfCCPLIBWbxBCCFKYtW/cOFUKu01YdTwdUFVWGomIYQoARm5vYXdu3fz8ssvW5f61OSaHyo7lyxLcgohHNe5c+d45JFH8Pf3x9XVlYYNG7J7927AnG5qzJgxNGzYEHd3d0JDQ3nsscc4f/58mbfr8yPw7D91WeHaGVWgrEwmhCgbEtwWYPfu3XzyySckJ9/IiRfkkQPAouWJ1i8KIYRwJMnJycTExKDValm9ejV//fUXM2fOxNfXnJw+MzOTvXv3Mm7cOPbu3cuPP/7IkSNHbFaCKwvfn4JNF6G++gKtc/ahCqxTpucTQlRdMi0hHyaTiYULF9psc9Ua8HIxAJCU7syiRYto1qwZarX8fSCEcBzTp08nPDychIQby9revNKat7c369atsznm448/5q677uLMmTNERESUepuSsuC5rZBugPOKJ5GGc6iCJLgVQpQNCW7zcfToUZsRW4BAd3NSxuQsLblGJ65du8bRo0e54447yqOJohyd/u0Mu+bsoeWQ5kTeXfqBQGnp0KEDTZo04f333wegRo0avPjii7z44ovl2i5RtpYvX05cXBz9+vVj8+bNhIWFMXToUJ5++ukCj0lNTUWlUuHj45Pv/pycHHJycqzv09LSAPMUB71ef8v2KAoM2epEukEFqLiOCyP83mSRX02MhRxb1ViuZWHXtKoqzevy/Sl4cQd80Ar61ii1ah3Sf78LHFF+fcnt/HtLcJuP/PK0nU5xY8qGOnjoDLcsJyq37JRstn+4E32Gnh0f7iSoYSAuPi5ldr7Bgwczb968PNuPHTtGdHT0bdVdo0YNTp8+DYCrqyu1atVixIgRPPXUU7dVryhfJ06cYPbs2YwaNYrXXnuNXbt2MXz4cHQ6HYMGDcpTPjs7mzFjxjBgwIACl7icOnUqEydOzLN97dq1hS6Dmpgbyk+ZLa3vjSoNy9y78fqGncQ47y/eh6si/juyLsy0Wi1BQUG3XY/lTkKqHp7bBu2DINC1FBpYgDlz5jB69GiSk5PRaMxhV3p6Or6+vsTExFif6wHYtGkTHTt25Pjx49SqVTaZmQoKdkuyfG5pSUxMzBPMZmZmlrg+CW7zUdDoRUauhoxcTaHlROWkKAo7PtqFIcv8B44+08COj3cR+8bdZXre7t2729xiBggIKJ3coJMmTeLpp58mMzOT77//nqeffpqwsDB69OhRKvWXRG5uLjqdPEVfUiaTiRYtWjBlyhQAmjZtysGDB5kzZ06e4Fav19O/f38URWH27NkF1jl27FhGjRplfZ+WlkZ4eDjdunW75ZrvSdnwxAoNKhQUVNbtKhS+NLRkRA8DgWX3t2GFo9frWbduHV27dkWr1ZZ3cxxOZmYmx48fv606FAWGbjdPkQG4rodh2+H7jqXQwAJ07NiR9PR0du/eTevWrQHYsmULwcHB7Nixg+zsbFxczL8IGzduJCIiotiBraIoGI1Ga/Bc0cTExOT5Q9lyh6gkZMJoPurUqWN9+KIgfn5+1Kkjc8aqktO/neHs1n9RTAoAiknhbOK/nP7tTJme19nZmeDgYJvXk08+SZ8+fWzKvfjii3To0KFYdXt6ehIcHEzNmjUZM2YMfn5+NqNGKSkpPPXUUwQEBODl5UWnTp2sqfGOHj2KSqXi77//tqlz1qxZNh3zwYMH6dGjBx4eHgQFBfHoo49y5coV6/4OHTrw/PPP8+KLL1KtWjXi4uJQFIUJEyYQERGBs7MzoaGhDB8+3HpMTk4OL7/8MmFhYbi7u9OqVSub0Y+qLCQkhPr169tsq1evHmfO2P6cWgLb06dPs27dulsGqc7Oznh5edm8wDySVtBLo9EyYreWdIPKJrAFUFBx3aDixd0FH19VX4Vd16r+ul3fn4JlZ8Bo7sYxKrD0jHl7Walbty4hISF5Rmh79+5NVFQU27dvt9nesWNHcnJyGD58OIGBgbi4uNCuXTt27dplU06lUrF69WqaN2+Os7Mzv//+OxkZGTz22GN4eHgQEhLCzJkzb6vts2fPplatWuh0OurWrcvXX39ts1+lUjF79mx69OiBq6srNWvW5Icffij2eUr731uC23yo1WoGDhx40xaFAU3OElfnEi4aIwADBgyQh8kqAUO2AUO2AUVRrNuMeiOGbAPGXKN1W3ZKNjs+2sV/vqNBBds/3ElGUkb+9Zpu1GsyOO6SzSaTiSVLlpCcnGwzatqvXz+SkpJYvXo1e/bsoVmzZnTu3Jlr165Rp04dWrRowYIFC2zqWrBggfX3JyUlhU6dOtG0aVN2797NmjVruHTpEv3797c5Zt68eeh0OhITE5kzZw5Llixh1qxZfPbZZxw7doxly5bRsGFDa/nnn3+ebdu2sXjxYv744w/69etH9+7dOXbsWBlepYohJiaGI0eO2Gw7evQokZGR1veWwPbYsWOsX78ef3//Um/HoRTbIOK/LEHFoeT89wtRVBl68+umbpxco3lbzo1unKQsGLIt326c57bCGdtu3FrvTd04+hJ24x07dmTjxo3W9xs3bqRDhw7ExsZat2dlZbFjxw46duzIK6+8wpIlS5g3bx579+4lOjqauLg4rl27ZlPvq6++yrRp0zh8+DCNGjVi9OjRbN68mZ9++om1a9eyadMm9u7dW6I2L126lBEjRvDSSy9x8OBBnn32WR5//HGbzwEwbtw4HnzwQQ4cOEB8fDwPP/wwhw8fLtE5S41SxaSmpiqAkpqaWmjZXbt2KSNHjlSGPxuvnFt8l3Jm4V3K6JdeVHbt2mWHllYsubm5yrJly5Tc3NzybkoeGRkZyu7du5WMjIw8+77uvlD5uvtCJSs5y7rtj4UHla+7L1S2vb9DURRFMZlMyqZJv1nL5vda8shSm3q/679E+br7QiX5VIp129FVx4rd9kGDBilOTk6Ku7u79dW3b19l0KBBSu/evW3KjhgxQomNjbW+j42NVUaMGGF9HxkZqcyaNcvmvU6nU9zd3RWNRqMAip+fn3LsmLmdW7ZsUby8vJTs7Gyb89SqVUv57LPPFEVRlFmzZim1atWy7jty5IgCKIcPH1YURVEmT56sdOvWzeb4s2fPKoBy5MgRazubNm1qU2bmzJlKnTp18v15On36tOLk5KScO3fOZnvnzp2VsWPH5ilfVm71c1Wcfqa07dy5U9FoNMrbb7+tHDt2TFmwYIHi5uamfPPNN4qimH9X77vvPqV69erK/v37lQsXLlhfOTk5RTpHUT6fyaQoD25QFN08RXH6Ku9LN09R+m4olY9caThyP+oICvqds/xMJd3oxpW3D5i3PZNofm/5eczvZ9HyivzO9nxBi8zbD167se2LIyVr+xdffKG4u7srer1eSUtLUzQajZKUlKQsXLhQad++vaIoivLrr78qgHLq1ClFq9UqCxYssB6fm5urhIaGKu+8846iKIqyceNGBVCWLVtmLXP9+nVFp9Mp331344NcvXpVcXV1tfkuiI2NVbRarc33iru7u+Ls7Kx4e3tby7Vt21Z5+umnbT5Hv379lJ49e1rfA8pzzz1nU6ZVq1bKkCFDinRdyqoflaHHW2jRogUzZszg+Sd7mze4hjDtnZm0aNGifBsm7Cr1dCpnt/57yzKZV7JIOZ1aJufv2LEj+/fvt74+/PDDUqt79OjR7N+/nw0bNtCqVStmzZplfVDtwIEDpKen4+/vj4eHh/V18uRJ/vnnHwAefvhhTp06Zb2ttmDBApo1a2bNInLgwAE2btxoc7xln6UOgObNm9u0q1+/fmRlZVGzZk2efvppli5disFgniT3559/YjQaqVOnjk29mzdvtqmzqmrZsiVLly5l0aJFNGjQgMmTJ/P+++8THx8PmBd4WL58Of/++y9NmjQhJCTE+tq6dWuptUOlgk9bg4cmv5EyBU8tfNK61E4nxC1Z7iTcyr+ZZXcnoUOHDmRkZLBr1y62bNlCnTp1CAgIIDY21jrvdtOmTdSsWZPU1FT0ej0xMTHW47VaLXfddVeeEdGb45F//vmH3NxcWrVqZd3m5+dH3bp5F0yJj4+3+V7Zv38/kyZNsilz+PBhmzaA+c7Qf9vQpk2bPO/Le+TW4WYenzt3jjFjxrB69WoyMzOJjo4mISGBFi1aoNfreeONN1i1ahUnTpzA29ubLl26MG3aNEJDQ0u1HYaMi5hyUgAI1JwjA3D1ro4h5SgAamcfNO7BpXpOYX8PL+0HgJPzjVXn6ve9g3r310WlNn8le0d6E962Ov9uP2czzcBCpVYR1ioUn0hv67b755kT4jvpbtRbq2vJlm12d3fPkxlBrVbbTKWAkqVNqVatGtHR0URHR/P999/TsGFDWrRoQf369UlPT88zT8zC8jBlcHAwnTp1YuHChbRu3ZqFCxcyZMgQa7n09HTuvfdepk+fnqeOkJAQm894s/DwcI4cOcL69etZt24dQ4cO5d1332Xz5s2kp6fj5OTEnj17cHKyXS3Qw8Oj2NegMurVqxe9evXKd1+NGjXy/OyUlYDc83xaz0j8H+E22xVUfHrHWQJyncC1dPtuUfWk/m8WodtNEc3Ld8KIeqD53xDenT7QJwJWnM1/qoyTCnpVhztvetzmnwfN/3W9qd5BJUxSEx0dTfXq1dm4cSPJycnExsYCEBoaSnh4OFu3bmXjxo106tSpWPX+t+8sKm9v7zzfK4GBgSWqyxE51Mito6ysY8i4SNKqflxZN4gr6waRcXQRALlJu6zbklb1w5BxsVTPK+xP46JB46JBpboxtuSkdULjorEGpiqVilYvtETjmu8QFFo3Da2H35V/veobB6g1pffrFhAQwIULF2y27d+//7bqDA8P56GHHmLs2LEANGvWjIsXL6LRaKwBsOVVrVo163Hx8fF8++23bNu2jRMnTvDwww9b9zVr1oxDhw5Ro0aNPHUU1im7urpy77338uGHH7Jp0ya2bdvGn3/+SdOmTTEajSQlJeWpMzhY/uB0FErqeQyf9OD+Fd3onbEWJ8U88u6kGOiTsZb7V3TD8EkPlNSyX/ZXVG7uWvPrpm4cnZN5m2Xc4tZ3EsBTC7NtByCt9d7UjaO9jW68Y8eObNq0iU2bNtk8/Nu+fXtWr17Nzp076dixo/UBrsTERGsZvV7Prl278jwserNatWqh1WrZsWOHdVtycjJHjx4tUXvr1atn0wYwp+z6bxtufiDO8r5evXolOmdpcaiRW0dZWceUkwKm3EIK5ZrLyehtleDi40KrF1ry+7T/3LZVoNULd5Vprtv8dOrUiXfffZf58+fTpk0bvvnmGw4ePEjTpk1vq94RI0bQoEEDdu/eTZcuXWjTpg19+vThnXfeoU6dOpw/f56VK1dy//33W2+HPfDAAwwZMoQhQ4bQsWNHm7sow4YN44svvmDAgAG88sor+Pn5cfz4cRYvXszcuXPzjLxafPXVVxiNRlq1aoWbmxvffPMNrq6uREZG4u/vT3x8PI899hgzZ86kadOmXL58mV9//ZVGjRpxzz333NY1EKUkMxkMuaiAD69NZJNLa1LVnniaMvjg2v9y5hpyzeW8ZfRWlL1AV3MAO/A32+0K5u1lmesWzMHtsGHD0Ov11pFbgNjYWJ5//nlyc3Pp2LEj7u7uDBkyhNGjR+Pn50dERATvvPMOmZmZPPnkkwXW7+HhwZNPPsno0aPx9/cnMDCQ119/vcQPv48ePZr+/fvTtGlTunTpwooVK/jxxx9Zv369Tbnvv/+eFi1a0K5dOxYsWMDOnTv5v//7vxKds7Q4VHDrKCvrWOb2FcZgMKCSlWQAx15Zp7TaFNk+gtO/nbFOT1CpVVRvE0Zke/uvUhYXF8e4ceN45ZVXyM7O5oknnuCxxx7jzz//vK1669evT7du3Rg/fjyrVq1i1apVvP766zz++ONcvnyZ4OBg2rdvb5NI3dPTk3vvvZfvvvuOL7/80qa+0NBQEhMTGTNmDN26dSMnJ4fIyEi6d+9+yw7Xx8eHadOmMWrUKIxGIw0bNmTFihXWp/oTEhJ46623eOmllzh37hzVqlWjdevWBd6KL0ulvbJOZRRousbH197kZb/XmHntbQJN1wo/SIgy0K8GfHfqxvQEJxXcF27eXtY6duxIVlYWd9xxh00fGhsby/Xr160pwwCmTZuGyWTi0Ucf5fr167Ro0YJffvml0DSl7777rnU6mKenJy+99BKpqSV7HqRPnz588MEHzJgxgxEjRhAVFUVCQkKelJMTJ05k8eLFDB06lJCQEBYtWnTLEWZ7UCn2mnxVBJYkxqNGjaJfv37s2rWLESNG5Jt8HMwr68TExHDHHXfkSUdkMWHChHxX1lm4cGGBK+u4mS5yZ+6X+e672SHdE2SqZeTW0Wm15lVt6tWrV+hqSoXJTsnmp6d+Rp+hR+eh5b4vetl91FY4hszMTA4fPsylS5fyXVln4MCBpKam3jJ/bEWVlpaGt7f3LT+fcuEQhi/6FlqX5ukfUIXcWdpNrJD0ej2rVq2iZ8+epZLTtbKx/M7dbl+elAX1lppXKPPRwV99yn7UtrJSqVQsXbo0T971orrVv2lR+pmCONTIraOsrKNPPkLKxsKD25iYGLS+eZ9CrIoceWWd0ljVxsLFx4XWw+9i15w9tBzSXAJbUeor6wghylagK8xpCy/ugA9aSWBbGTlUcFvQyjpLliyx2XbzyjobNmwodGUdZ2fnPNtvtfqFUsTl6zQajcMFcuWttFaRKU2l3Z7I9hHlMhVBOKb8fuYd7XdACGGrXw37TEUQ5cOhgtvirqyzcePGMllZRwghhBBC3JoDzWy14VDB7ciRI2nbti1Tpkyhf//+7Ny5k88//5zPP/8cMAe2ffv2Ze/evfz8888YjUYuXjSn4/Lz87NZNlQIIYQQQlQ9DhXcWlbWGTt2LJMmTSIqKirflXUAmjRpYnOsZZ3m0qB29gG17tbpwNQ6czlRYZhMJVwUXIh8yM9TIdx8QaMzp/sqiEZnLidEMcjvXuVRVv+WDhXcgmOsrKNxDyaw5/fWFcryIyuUVRyWEf309HRZwUqUmvT0dAC5Y1QAlXcommGrzXlsC+Lmi0py3Ioikr688imrftThgltHoXEPlgUaKgmNRkO1atU4d+4cYE50XdKk1kKYTCbS09Ot+XU1RXwAtSpSeYfKAg2i1EhfXnmUdT8qvbKoEiyr11k6RSFuV7Vq1UptVUQhRNFIX165lFU/KsGtqBJUKhWRkZGEhYWRm1vI0soVgF6vJzEx0ZxrWdJOlan8rrVOp5MRWyHKQWn25dKP2o+9+1HpnUWVotFoKkVQYlny1c3NTTrlMibXWgjHUxp9ufxu24+9r7VMVhFCCCGEEJWGBLdCCCGEEKLSkOBWCCGEEEJUGhV/8mExWfLkpqWllXNLKhe9Xk9mZiZpaWkyd8kO5HrbT0mutaV/cdSlKW+X9KNlQ36v7Uuut/3Yux+tcsHt9evXAQgPDy/nlgghKrvr16/j7e1d3s0oddKPCiHspST9qEqprEMLBTCZTJw/fx5PT09UKlV5N6fSSEtLIzw8nLNnz+Ll5VXezan05HrbT0mutaIoXL9+ndDQ0EqZZF760bIhv9f2Jdfbfuzdj1a5kVu1Wk316tXLuxmVlpeXl3QSdiTX236Ke60r44ithfSjZUt+r+1Lrrf92KsfrXxDCkIIIYQQosqS4FYIIYQQQlQaEtyKUuHs7Mybb76Js7NzeTelSpDrbT9yrYW9yM+afcn1th97X+sq90CZEEIIIYSovGTkVgghhBBCVBoS3AohhBBCiEpDglshhBBCCFFpSHArhBBCCCEqDQluhdVvv/3GvffeS2hoKCqVimXLltnsVxSF8ePHExISgqurK126dOHYsWM2Za5du0Z8fDxeXl74+Pjw5JNPkp6eblPmjz/+4O6778bFxYXw8HDeeeedsv5oDmfq1Km0bNkST09PAgMD6dOnD0eOHLEpk52dzbBhw/D398fDw4MHH3yQS5cu2ZQ5c+YM99xzD25ubgQGBjJ69GgMBoNNmU2bNtGsWTOcnZ2Jjo7mq6++KuuP53Bmz55No0aNrAnE27Rpw+rVq6375VqL0iL9qP1IP2pfFaofVYT4n1WrVimvv/668uOPPyqAsnTpUpv906ZNU7y9vZVly5YpBw4cUO677z4lKipKycrKspbp3r270rhxY2X79u3Kli1blOjoaGXAgAHW/ampqUpQUJASHx+vHDx4UFm0aJHi6uqqfPbZZ/b6mA4hLi5OSUhIUA4ePKjs379f6dmzpxIREaGkp6dbyzz33HNKeHi48uuvvyq7d+9WWrdurbRt29a632AwKA0aNFC6dOmi7Nu3T1m1apVSrVo1ZezYsdYyJ06cUNzc3JRRo0Ypf/31l/LRRx8pTk5Oypo1a+z6ecvb8uXLlZUrVypHjx5Vjhw5orz22muKVqtVDh48qCiKXGtReqQftR/pR+2rIvWjEtyKfP23UzaZTEpwcLDy7rvvWrelpKQozs7OyqJFixRFUZS//vpLAZRdu3ZZy6xevVpRqVTKuXPnFEVRlE8//VTx9fVVcnJyrGXGjBmj1K1bt4w/kWNLSkpSAGXz5s2KopivrVarVb7//ntrmcOHDyuAsm3bNkVRzF+iarVauXjxorXM7NmzFS8vL+v1feWVV5Q777zT5lwPPfSQEhcXV9YfyeH5+voqc+fOlWstyoz0o/Yl/aj9OWo/KtMSRJGcPHmSixcv0qVLF+s2b29vWrVqxbZt2wDYtm0bPj4+tGjRwlqmS5cuqNVqduzYYS3Tvn17dDqdtUxcXBxHjhwhOTnZTp/G8aSmpgLg5+cHwJ49e9Dr9TbX+4477iAiIsLmejds2JCgoCBrmbi4ONLS0jh06JC1zM11WMpY6qiKjEYjixcvJiMjgzZt2si1FnYj/WjZkn7Ufhy9H5XgVhTJxYsXAWx+KC3vLfsuXrxIYGCgzX6NRoOfn59NmfzquPkcVY3JZOLFF18kJiaGBg0aAOZrodPp8PHxsSn73+td2LUsqExaWhpZWVll8XEc1p9//omHhwfOzs4899xzLF26lPr168u1FnYj/WjZkX7UPipKP6op7gcTQpSuYcOGcfDgQX7//ffybkqlVrduXfbv309qaio//PADgwYNYvPmzeXdLCFEKZB+1D4qSj8qI7eiSIKDgwHyPPl46dIl677g4GCSkpJs9hsMBq5du2ZTJr86bj5HVfL888/z888/s3HjRqpXr27dHhwcTG5uLikpKTbl/3u9C7uWBZXx8vLC1dW1tD+OQ9PpdERHR9O8eXOmTp1K48aN+eCDD+RaC7uRfrRsSD9qPxWlH5XgVhRJVFQUwcHB/Prrr9ZtaWlp7NixgzZt2gDQpk0bUlJS2LNnj7XMhg0bMJlMtGrVylrmt99+Q6/XW8usW7eOunXr4uvra6dPU/4UReH5559n6dKlbNiwgaioKJv9zZs3R6vV2lzvI0eOcObMGZvr/eeff9p8Ea5btw4vLy/q169vLXNzHZYyljqqMpPJRE5OjlxrYTfSj5Yu6UfLn8P2oyV7Pk5URtevX1f27dun7Nu3TwGU9957T9m3b59y+vRpRVHMKWx8fHyUn376Sfnjjz+U3r1755vCpmnTpsqOHTuU33//Xaldu7ZNCpuUlBQlKChIefTRR5WDBw8qixcvVtzc3KpcCpshQ4Yo3t7eyqZNm5QLFy5YX5mZmdYyzz33nBIREaFs2LBB2b17t9KmTRulTZs21v2WtCrdunVT9u/fr6xZs0YJCAjIN63K6NGjlcOHDyuffPJJlUxh8+qrryqbN29WTp48qfzxxx/Kq6++qqhUKmXt2rWKosi1FqVH+lH7kX7UvipSPyrBrbDauHGjAuR5DRo0SFEUcxqbcePGKUFBQYqzs7PSuXNn5ciRIzZ1XL16VRkwYIDi4eGheHl5KY8//rhy/fp1mzIHDhxQ2rVrpzg7OythYWHKtGnT7PURHUZ+1xlQEhISrGWysrKUoUOHKr6+voqbm5ty//33KxcuXLCp59SpU0qPHj0UV1dXpVq1aspLL72k6PV6mzIbN25UmjRpouh0OqVmzZo256gqnnjiCSUyMlLR6XRKQECA0rlzZ2uHrChyrUXpkX7UfqQfta+K1I+qFEVRijfWK4QQQgghhGOSObdCCCGEEKLSkOBWCCGEEEJUGhLcCiGEEEKISkOCWyGEEEIIUWlIcCuEEEIIISoNCW6FEEIIIUSlIcGtEEIIIYSoNCS4FUIIIYQQlYYEt0IIUYDBgwdTo0aNEh07YcIEVCpV6TZICCFEoSS4FUJUOCqVqkivTZs2lXdTy82KFSuIjY0lMDAQNzc3atasSf/+/VmzZo21zPnz55kwYQL79+8vv4YKIUQpk+V3hRAVzjfffGPzfv78+axbt46vv/7aZnvXrl0JCgoq8Xn0ej0mkwlnZ+diH2swGDAYDLi4uJT4/CU1Y8YMRo8eTWxsLL1798bNzY3jx4+zfv16GjduzFdffQXA7t27admyJQkJCQwePNju7RRCiLKgKe8GCCFEcT3yyCM277dv3866devybP+vzMxM3NzcinwerVZbovYBaDQaNBr7d7EGg4HJkyfTtWtX1q5dm2d/UlKS3dskhBD2JNMShBCVUocOHWjQoAF79uyhffv2uLm58dprrwHw008/cc899xAaGoqzszO1atVi8uTJGI1Gmzr+O+f21KlTqFQqZsyYweeff06tWrVwdnamZcuW7Nq1y+bY/ObcqlQqnn/+eZYtW0aDBg1wdnbmzjvvtJkqYLFp0yZatGiBi4sLtWrV4rPPPivSPN4rV66QlpZGTExMvvsDAwOt9bds2RKAxx9/3DqVwzKqC7Bjxw66d++Ot7c3bm5uxMbGkpiYmO/n/Pvvv+nfvz9eXl74+/szYsQIsrOzbcquW7eOdu3a4ePjg4eHB3Xr1rX+mwghRGmRkVshRKV19epVevTowcMPP8wjjzxinaLw1Vdf4eHhwahRo/Dw8GDDhg2MHz+etLQ03n333ULrXbhwIdevX+fZZ59FpVLxzjvv8MADD3DixIlCR3t///13fvzxR4YOHYqnpycffvghDz74IGfOnMHf3x+Affv20b17d0JCQpg4cSJGo5FJkyYREBBQaNsCAwNxdXVlxYoVvPDCC/j5+eVbrl69ekyaNInx48fzzDPPcPfddwPQtm1bADZs2ECPHj1o3rw5b775Jmq1moSEBDp16sSWLVu46667bOrr378/NWrUYOrUqWzfvp0PP/yQ5ORk5s+fD8ChQ4fo1asXjRo1YtKkSTg7O3P8+PE8wbIQQtw2RQghKrhhw4Yp/+3OYmNjFUCZM2dOnvKZmZl5tj377LOKm5ubkp2dbd02aNAgJTIy0vr+5MmTCqD4+/sr165ds27/6aefFEBZsWKFddubb76Zp02AotPplOPHj1u3HThwQAGUjz76yLrt3nvvVdzc3JRz585Ztx07dkzRaDR56szP+PHjFUBxd3dXevToobz99tvKnj178pTbtWuXAigJCQk2200mk1K7dm0lLi5OMZlM1u2ZmZlKVFSU0rVr1zyf87777rOpY+jQoQqgHDhwQFEURZk1a5YCKJcvXy60/UIIcTtkWoIQotJydnbm8ccfz7Pd1dXV+v/Xr1/nypUr3H333WRmZvL3338XWu9DDz2Er6+v9b1l1PPEiROFHtulSxdq1aplfd+oUSO8vLysxxqNRtavX0+fPn0IDQ21louOjqZHjx6F1g8wceJEFi5cSNOmTfnll194/fXXad68Oc2aNePw4cOFHr9//36OHTvGwIEDuXr1KleuXOHKlStkZGTQuXNnfvvtN0wmk80xw4YNs3n/wgsvALBq1SoAfHx8APOUkP8eK4QQpUmCWyFEpRUWFoZOp8uz/dChQ9x///14e3vj5eVFQECA9WG01NTUQuuNiIiweW8JdJOTk4t9rOV4y7FJSUlkZWURHR2dp1x+2woyYMAAtmzZQnJyMmvXrmXgwIHs27ePe++9N89c2P86duwYAIMGDSIgIMDmNXfuXHJycvJcp9q1a9u8r1WrFmq1mlOnTgHmPwhiYmJ46qmnCAoK4uGHH+a7776TQFcIUepkzq0QotK6eYTWIiUlhdjYWLy8vJg0aRK1atXCxcWFvXv3MmbMmCIFW05OTvluV4qQWfF2ji0JLy8vunbtSteuXdFqtcybN48dO3YQGxtb4DGWa/Duu+/SpEmTfMt4eHjc8rz/ffDN1dWV3377jY0bN7Jy5UrWrFnDt99+S6dOnVi7dm2B10UIIYpLglshRJWyadMmrl69yo8//kj79u2t20+ePFmOrbohMDAQFxcXjh8/nmdfftuKo0WLFsybN48LFy4AeQNQC8u0CS8vL7p06VKkuo8dO0ZUVJRNW00mk022CbVaTefOnencuTPvvfceU6ZM4fXXX2fjxo1FPo8QQhRGpiUIIaoUywjhzSOlubm5fPrpp+XVJBtOTk506dKFZcuWcf78eev248ePs3r16kKPz8zMZNu2bfnusxxft25dANzd3QHzaPbNmjdvTq1atZgxYwbp6el56rl8+XKebZ988onN+48++gjAOk/42rVreY6xjArn5OQU9HGEEKLYZORWCFGltG3bFl9fXwYNGsTw4cNRqVR8/fXXZTYtoCQmTJjA2rVriYmJYciQIRiNRj7++GMaNGhQ6FK5mZmZtG3bltatW9O9e3fCw8NJSUlh2bJlbNmyhT59+tC0aVPAPELr4+PDnDlz8PT0xN3dnVatWhEVFcXcuXPp0aMHd955J48//jhhYWGcO3eOjRs34uXlxYoVK2zOe/LkSe677z66d+/Otm3b+Oabbxg4cCCNGzcGYNKkSfz222/cc889REZGkpSUxKeffkr16tVp165dmVxHIUTVJMGtEKJK8ff35+eff+all17ijTfewNfXl0ceeYTOnTsTFxdX3s0DzCOnq1ev5uWXX2bcuHGEh4czadIkDh8+XGg2Bx8fH7744gtWrlxJQkICFy9exMnJibp16/Luu+8yfPhwa1nLHNyxY8fy3HPPYTAYSEhIICoqig4dOrBt2zYmT57Mxx9/THp6OsHBwbRq1Ypnn302z3m//fZbxo8fz6uvvopGo+H555+3yRl83333cerUKb788kuuXLlCtWrViI2NZeLEiXh7e5fexRNCVHkqxZGGK4QQQhSoT58+HDp0yJrNwBFMmDCBiRMncvnyZapVq1bezRFCCJlzK4QQjigrK8vm/bFjx1i1ahUdOnQonwYJIUQFIdMShBDCAdWsWZPBgwdTs2ZNTp8+zezZs9HpdLzyyivl3TQhhHBoEtwKIYQD6t69O4sWLeLixYs4OzvTpk0bpkyZkmexBCGEELZkzq0QQgghhKg0ZM6tEEIIIYSoNCS4FUIIIYQQlYYEt0IIIYQQotKQ4FYIIYQQQlQaEtwKIYQQQohKQ4JbIYQQQghRaUhwK4QQQgghKg0JboUQQgghRKUhwa0QQgghhKg0/h/fBxdKGVDjqAAAAABJRU5ErkJggg==",
+      "text/plain": [
+       "<Figure size 800x400 with 2 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "CHECKPOINTS = list(range(200, 3000+1, 200))\n",
+    "SEEDS = [0, 14, 41, 53, 96]\n",
+    "\n",
+    "# Define the input files, layers, and checkpoints\n",
+    "rev_files = [\n",
+    "    {'data': 'reverse_control',\n",
+    "     'label': 'NoReverse',\n",
+    "     'linestyle': '-',\n",
+    "     'marker': 'o',\n",
+    "    },\n",
+    "    {'data': 'reverse_partial',\n",
+    "     'label': 'PartialReverse',\n",
+    "     'linestyle': '--',\n",
+    "     'marker': 's',\n",
+    "    },\n",
+    "    {'data': 'reverse_full',\n",
+    "     'label': 'FullReverse',\n",
+    "     'linestyle': ':',\n",
+    "     'marker': 'd',\n",
+    "    },\n",
+    "]\n",
+    "hop_files = [\n",
+    "    {'data': 'hop_control',\n",
+    "     'label': 'NoHop',\n",
+    "     'linestyle': '-',\n",
+    "     'marker': 'o',\n",
+    "    },\n",
+    "    {'data': 'hop_tokens4',\n",
+    "     'label': 'TokenHop',\n",
+    "     'linestyle': '--',\n",
+    "     'marker': 's',\n",
+    "    },\n",
+    "    {'data': 'hop_words4',\n",
+    "     'label': 'WordHop',\n",
+    "     'linestyle': ':',\n",
+    "     'marker': 'd',\n",
+    "    },\n",
+    "]\n",
+    "\n",
+    "# Generate the plot\n",
+    "plot_accuracy(rev_files, hop_files, CHECKPOINTS, SEEDS)\n",
+    "plt.savefig(\"figures/probing.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "# No Pos Encodings\n",
+    "plot_accuracy(rev_files, hop_files, CHECKPOINTS, [53], pos_encodings=False)\n",
+    "plt.savefig(\"figures/probing_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Plot Accuracy with Individual Layers"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 42,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plot_accuracy_layer(datasets, layers, checkpoints, seeds, pos_encodings=True):\n",
+    "\n",
+    "    # Setting up the plot with all subplots in a single row\n",
+    "    fig, axes = plt.subplots(nrows=1, ncols=len(layers), figsize=(4 * len(layers), 4), sharey=True)\n",
+    "    fig.subplots_adjust(wspace=0)\n",
+    "    fig.supxlabel(\"Training Steps\", y=-0.01)\n",
+    "    if len(layers) == 1:\n",
+    "        axes = [axes]\n",
+    "\n",
+    "    if pos_encodings:\n",
+    "        nps = \"\"\n",
+    "    else:\n",
+    "        nps = \"_no_pos_encodings\"\n",
+    "\n",
+    "    data_path = \"edge_probing_results/{}_100M{}/randinit_mean_pooling_seed{}.csv\"\n",
+    "\n",
+    "    # Plotting the data\n",
+    "    for i, layer in enumerate(layers):\n",
+    "        ax = axes[i]\n",
+    "        if i == 0:\n",
+    "            ax.set_ylabel(\"Probe Accuracy\", fontsize=12)\n",
+    "\n",
+    "        # Main plotting function\n",
+    "        plot_datasets(ax, datasets, seeds, data_path, layer, checkpoints, nps)\n",
+    "\n",
+    "        # Setting plot details\n",
+    "        ax.set_title(f'Layer {layer}')\n",
+    "        ax.grid(True)\n",
+    "\n",
+    "        if i != 0:\n",
+    "            for tick in ax.yaxis.get_major_ticks():\n",
+    "                tick.tick1line.set_visible(False)\n",
+    "                tick.tick2line.set_visible(False)\n",
+    "        \n",
+    "        if i == len(layers) - 1:\n",
+    "            ax.legend(loc='lower right', framealpha=1)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 43,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "LAYERS = [str(l) for l in [1, 3, 6, 9, 12]]\n",
+    "CHECKPOINTS = list(range(200, 3000+1, 200))"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 44,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABlgAAAGOCAYAAAD7MA7xAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAEAAElEQVR4nOzdd3hcxbn48e/Zqt6b1eUiy73JHdtgYwymQyCYUAKBQGK4CaTe/NJIu0nupSTEiZNADMGmmB7AFFfA3XK3ZVnFtnqxetu+5/fHWiutpJVWsmxL8vt5Hj/ecnZ2ZrUanTPvzDuKqqoqQgghhBBCCCGEEEIIIYQQwmeaS10BIYQQQgghhBBCCCGEEEKIoUYCLEIIIYQQQgghhBBCCCGEEH0kARYhhBBCCCGEEEIIIYQQQog+kgCLEEIIIYQQQgghhBBCCCFEH0mARQghhBBCCCGEEEIIIYQQoo8kwCKEEEIIIYQQQgghhBBCCNFHEmARQgghhBBCCCGEEEIIIYToIwmwCCGEEEIIIYQQQgghhBBC9JEEWIQQQgghhBBCCCGEEEIIIfpIAixCCCGEEEIIIYQQQgghhBB9JAEWcVl46aWXUBSFrKysS12VAbF3716+/e1vM2PGDPR6PYqiXOoqCSEusuHWr/3zn/9k0aJFxMbGYjQaSUtL44EHHuDMmTOXumpCiItkuPVrAE6nk7/97W9MnToVf39/IiMjWbx4MYcPH77UVRNCXATDsV/7y1/+wrhx4zAajSQkJPDkk0/S0tJyqaslhBhgw63/8nUcrbi4mKeeeopZs2YRHh5OVFQUV155JZs2bbrINRZDiQRYhBiCNmzYwAsvvICiKIwcOfJSV0cIIc7bwYMHSUtL44c//CF/+9vfuOeee/j444+ZOXMmZWVll7p6QgjRLw8++CD/9V//xYwZM3j++ef5+c9/TnJyMlVVVZe6akII0Wc/+tGPePzxx5k4cSJ/+tOfuP3223n++ee57bbbLnXVhBCiR76Oo73//vv84Q9/YPTo0fzmN7/hZz/7GU1NTSxdupQ1a9ZcxBqLoUR3qSsghOjK6XRitVrx8/Pr9vlvfetb/OhHP8Lf35/HHnuM3Nzci1xDIYTom976tb/+9a9dHrvlllvIzMzk3//+Nz/+8Y8vdBWFEKJPeuvX1q9fz8svv8w777zDrbfeepFrJ4QQfddTv1ZeXs4zzzzDvffey7///W/34+np6Tz++ON88MEH3HjjjRezukII4TZQ42hXXXUVRUVFREVFuR979NFHmTp1Kj//+c954IEHLkj9xdAmK1iEOMdqtfLzn/+cGTNmEBoaSmBgIAsWLGDr1q3uY1RVJTU1lZtvvrnL681mM6GhoTzyyCPuxywWC7/4xS8YPXo0RqORpKQkfvjDH2KxWDxeqygKjz32GOvWrWPChAkYjUY++eQTr3WNjY3F399/AFothBjOhlK/1p3U1FQA6uvr+/Q6IcTwNZT6tWeeeYZZs2Zx66234nQ6JYWOEKJbQ6Vf27VrF3a7nbvuusvj8bb7r7/+er8/AyHE0DRU+i/wfRxtwoQJHsEVAKPRyPLlyykpKaGpqanXMsTlR1awCHFOY2MjL7zwAitWrODhhx+mqamJF198kWXLlrF3716mTp2Koijcc889/PGPf6S2tpaIiAj36z/44AMaGxu55557AFf0/KabbmL79u1885vfZNy4cRw9epRnn32W3Nxc3nvvPY/337JlC+vXr+exxx4jKirKPbAohBD9NRT7tZqaGhwOB0VFRfzqV78CYMmSJQP2mQghhrah0q81Nja6c33/5Cc/4fnnn6e5uZm0tDR+//vfc+edd16oj0gIMcQMlX6tbXCz8wBlQEAAAPv37x+gT0QIMVQMlf5rIFRUVBAQEODu84TwoApxGVizZo0KqPv27fN6jN1uVy0Wi8djdXV1amxsrPrggw+6Hzt58qQKqH/72988jr3pppvU1NRU1el0qqqqqq+88oqq0WjUL7/80uO41atXq4C6Y8cO92OAqtFo1OPHj/e5bStXrlTlV1mIy89w7deMRqMKqIAaGRmp/vnPf+7T64UQQ9dw6tcOHDjg7sdiY2PVv/71r+q6devUWbNmqYqiqB9//HGvZQghhr7h1K/t379fBdRf//rXHo9/8sknKqAGBQX1WoYQYugYTv1XZ30dR8vLy1P9/PzUe++9t8/vJS4PkiJMiHO0Wi0GgwFwRc1ra2ux2+1kZmZy4MAB93Hp6enMnj2bdevWuR+rra3l448/5mtf+xqKogDw5ptvMm7cODIyMqiurnb/W7x4MYDHkkmARYsWMX78+AvdTCHEZWQo9msff/wxGzZs4OmnnyY5OVlS6gghPAyVfq25uRlwrcp7//33+da3vsXdd9/N5s2biYyM5De/+c35fRBCiGFjqPRr06dPZ/bs2fzhD39gzZo1nDlzho8//phHHnkEvV6PyWQ6789CCDG0DJX+63y0trZyxx134O/vz+9///sL+l5i6JIUYUJ08PLLL/P000+Tk5ODzWZzP56WluZx3H333cdjjz1GYWEhKSkpvPnmm9hsNu699173MXl5eZw4cYLo6Ohu36uqqsrjfuf3EEKIgTDU+rWrrroKgOuuu46bb76ZiRMnEhQUxGOPPdbnsoQQw9NQ6NfaUuikpaUxe/Zs9+NBQUHceOONrF27Frvdjk4nl2NCiKHRrwG8/fbbfPWrX+XBBx8EXIOrTz75JJ9//jknT570uRwhxPAxVPqv/nA4HNx1111kZ2fz8ccfEx8ff0HfTwxdckYvxDlr167l61//Orfccgs/+MEPiImJQavV8j//8z8UFBR4HHvXXXfxxBNPsG7dOn7yk5+wdu1aMjMzGTt2rPsYp9PJpEmTeOaZZ7p9v6SkJI/7smm9EGKgDfV+bdSoUUybNo1169ZJgEUIAQydfq3tAjw2NrbLczExMdhsNlpaWggNDfWpPCHE8DVU+jWAhIQEtm/fTl5eHhUVFYwZM4a4uDji4+NJT0/vQ6uFEMPBUOq/+uPhhx/mww8/ZN26de5VNEJ0RwIsQpzz1ltvMXLkSN555x338kSAX/ziF12OjYiI4Prrr2fdunV87WtfY8eOHTz33HMex4waNYrDhw+zZMkSj/KEEOJiGQ79mslkcm+qKoQQQ6Vfi4+PJy4ujtLS0i7PlZWV4efnR3Bw8IC9nxBi6Boq/VpHY8aMYcyYMQBkZ2dTXl7O17/+9QvyXkKIwWso9l+++sEPfsCaNWt47rnnWLFixSWtixj8ZA8WIc7RarUAqKrqfmzPnj3s2rWr2+PvvfdesrOz+cEPfoBWq+Wuu+7yeP7OO++ktLSUf/7zn11eazKZZF8BIcQFN1T6NbvdTl1dXZfH9+7dy9GjR8nMzOxXuUKI4Weo9GsAX/3qVykuLmbjxo3ux6qrq3n//fdZvHgxGo1cigkhhla/1pnT6eSHP/whAQEBPProowNWrhBiaBjK/VdP/vd//5f/+7//4yc/+Qnf+c53Lsp7iqFNVrCIy8q//vUvPvnkky6Pf+c73+GGG27gnXfe4dZbb+X666/n9OnTrF69mvHjx7s3Ku3o+uuvJzIykjfffJPrrruOmJgYj+fvvfde1q9fz6OPPsrWrVuZP38+DoeDnJwc1q9fz6efftrvQcPCwkJeeeUVALKysgDcm6WmpKR45LAUQgxvw6Ffa25uJikpia9+9atMmDCBwMBAjh49ypo1awgNDeVnP/tZn8sUQgxdw6FfA/jv//5v1q9fz+23386TTz5JaGgoq1evxmaz8bvf/a5fZQohhqbh0q995zvfwWw2M3XqVGw2G6+++ip79+7l5ZdfJjk5uV9lCiEGt+HSf/k6jvbuu+/ywx/+kDFjxjBu3DjWrl3rUc7SpUu7TQErLnOqEJeBNWvWqIDXf8XFxarT6VR/97vfqSkpKarRaFSnTZumfvjhh+r999+vpqSkdFvut7/9bRVQX3311W6ft1qt6h/+8Ad1woQJqtFoVMPDw9UZM2aoTz31lNrQ0OA+DlBXrlzpc3u2bt3qtS2LFi3qy0cjhBiihlO/ZrFY1O985zvq5MmT1ZCQEFWv16spKSnqN77xDfX06dN9/WiEEEPUcOrX2hQUFKi33nqrGhISovr7+6uLFy9W9+7d26cyhBBD13Dr19asWaNOmTJFDQwMVIODg9UlS5aoW7Zs6dNnIoQYGoZb/+XrONovfvGLHtu9detWn99TXD4UVe2wjksI0SdPPPEEL774IhUVFQQEBFzq6gghxHmTfk0IMdxIvyaEGG6kXxNCDFXSf4nhSBL/CtFPZrOZtWvXcvvtt8sfBSHEsCD9mhBiuJF+TQgx3Ei/JoQYqqT/EsOV7MEiRB9VVVWxadMm3nrrLWpqamTDKyHEkCf9mhBiuJF+TQgx3Ei/JoQYqqT/EsOdBFiE6KPs7Gy+9rWvERMTw5///GemTp16qaskhBDnRfo1IcRwI/2aEGK4kX5NCDFUSf8lhjvZg0UIIYQQQgghhBBCCCGEEKKPZA8WIYQQQgghhBBCCCGEEEKIPpIAixBCCCGEEEIIIYQQQgghRB9ddnuwOJ1OysrKCA4ORlGUS10dIYQ4bw6Hg/z8fEaPHo1Wq73U1RFCiPMm/ZoQYriRfk0IMdxIvyaEGG5UVaWpqYn4+Hg0Gt/XpVx2AZaysjKSkpIudTWEEEIIIYQQQgghhBBCCDGIFBcXk5iY6PPxl12AJTg4GHB9UCEhIZe4Nr2z2Wx89tlnXHPNNej1+ktdnfMyXNoi7Rhchks7oP9tKSkpYcKECdKvXQLDpS3SjsFluLQDpF8bioZLW6Qdg8twaQdIvzYUDZe2SDsGl+HSDpB+bSgaLm2Rdgwuw6Ud0P+2NDY2kpSU5I4f+OqyC7C0pQULCQkZMn8AAgICCAkJGRZf7uHQFmnH4DJc2gH9b0tbXyb92sU3XNoi7Rhchks7QPq1oWi4tEXaMbgMl3aA9GtD0XBpi7RjcBku7QDp14ai4dIWacfgMlzaAefflr5uKyKb3AshhBBCCCGEEEIIIYQQQvSRBFiEEEIIIYQQQgghhBBCCCH6SAIsQgghhBBCCCGEEEIIIYQQfSQBFiGEEEIIIYQQQgghhBBCiD6SAIsQQgghhBBCCCGEEEIIIUQfSYBFCCGEEEIIIYQQQgghhBCijyTAIoQQQgghhBBCCCGEEEII0UcSYBFCCCGEEEIIIYQQQgghhOgjCbAIIYQQQgghhBBCCCGEEEL0kQRYhBBCCCGEEEIIIYQQQggh+kgCLEIIIYQQQgghhBBCCCGEEH2ku9QVEEIIIYQQQgghhBBC9MxutvP6rW8CcNe7d6DzO79hPbvZzhu3vuW6vdiOXq8/7zoKIcTlRlawCCGEEEIIIYToE7vZztrrXmPtda9hN9svdXWEEEIIIYS4JCTAIoQQQgghhBDikpOgjRDnR36HhBBCiItPAixCCCGEEEIIcZ5kYFMIIURn8rdBCCGGv0EVYElNTUVRlC7/Vq5cCcCVV17Z5blHH330EtdaCCGEEEIIIYQQYniR4IAQQgjRu0G1yf2+fftwOBzu+8eOHWPp0qXccccd7scefvhhfvWrX7nvBwQEXNQ6CiGEEEIIIYQQQohL60Js+D6Q5QkhhLg8DKq/FtHR0R73f//73zNq1CgWLVrkfiwgIIC4uLiLXTUhhBBCCCGEEEIIIYQQQgi3QRVg6chqtbJ27VqefPJJFEVxP75u3TrWrl1LXFwcN954Iz/72c96XMVisViwWCzu+42NjQDYbDZsNtuFa8AAaavjUKhrb4ZLW6Qdg8twaQf0vy12u939uqHwOcjPbPCRdgwuw6UdIP3aUDRc2nIp2mG3tafPsdlsqFr1vMsczD+PvrTX13YM9GdoN9t5+873ALh9/S0DMhtd+rWhZ7i0xZd2DPjv0CDv1y5ley9Zv3aBy9PZfO8npV+7dIZLW6Qdg8twaQf0vy39bbuiqur5/4W8ANavX8/dd99NUVER8fHxAPzjH/8gJSWF+Ph4jhw5wo9+9CNmzZrFO++847WcX/7ylzz11FNdHn/11VclvZgQYliorq7moYcekn5NCDFsSL8mhiLVptL8ogmAoG/4o+iVXl7RM5vJhvll10We3/169P76867jQBro9l6IMi9EHftL+jXRncH+nR9Mv0PdGQrtHex1PJ/ypF8TQgw3ra2t3H333TQ0NBASEuLz6wZtgGXZsmUYDAY++OADr8ds2bKFJUuWkJ+fz6hRo7o9prsVLElJSVRXV/fpg7pUbDYbGzduZOnSpej1g+uiqq+GS1ukHYPLcGkH9L8tpaWlpKWlSb92CQyXtkg7Bpfh0g6Qfm0oGi5tuRTtGOjVEi0NLXx478cALPvX1YRFhZ1nDQdWX9rr689joD/DC1neTeuuxz/Y3+fXSr926Qzmtgz079Fg/x2Cgf15XMr2Sr8m/dqlNFzaIu0YXIZLO6D/bWlsbCQqKqrPAZZBmSKssLCQTZs29bgyBWD27NkAPQZYjEYjRqOxy+N6vX5IfVmGWn17MlzaIu0YXIZLO6DvbdHpdP163aU21Orbk+HSFmnH4DJc2gHSrw1Fw6UtF7MdiqN91q9er0enP79LrY71How/j/60t7d2DPRneKHLk35taBmMbRno36PB/jtkN9t55/b3XWUvVtAHnN/PYzC0V/q1wfm71ZOhVt+eDJe2SDsGl+HSDuh7W/rbbk2/XnWBrVmzhpiYGK6//voejzt06BAAI0aMuAi1EkIIIYQQQgghhBBCCCGEcBl0K1icTidr1qzh/vvvd0fDAQoKCnj11VdZvnw5kZGRHDlyhCeeeIKFCxcyefLkS1hjIYQQQgghhBjc7GY7r9/6JgB3vXvHgKT6EUIIIYQQ4nI36M6qN23aRFFREQ8++KDH4waDgU2bNvHcc8/R0tJCUlISt99+Oz/96U8vUU2FEGLgWCwWHn30UQCWLFkybJZjCiGEEEIIIYQQQggxXA26AMs111yDqqpdHk9KSuLzzz+/BDUSQogLT7Wb+O212e7bEHRpKySEuOw47SYq3r4SgLjbt6HR+b7JqRBCXA4sFovH7b5sBi0uDbvZzhu3vuW6vdguk5j6yGk3dbodfOkqI4QQQgxSg3IPFiGEGOycdhNlb8ym7I3ZHhceQgghhBBCCCHEhdA50CuEEOLSkwCLEEL0g7XDyaxVTmyFEL24EEFZCfQKIYQQQ9vlNlh+ubV3KOh4DqnK+aQQQvSLBFiEEJcFc0u9eyDS3FJ/qasjhBBCCCGEEEJcUlaPoJf1EtZECCGGLgmwCCGEEEIIIS47drOdtde9xtrrXsNutl/q6ogLrGhH8aWughBDWvGusktdBSGEEGJQkgCLEGLQsVgsPPDAAzzwwAOydFwIcUlI+q3z57SbOPvOFcw0/05STojLwkCnvnE6ndQZaqnwKye/IB+n03neZQ52XTfUPj/merP79r6/7fe4L8RgU7p7YAMYAxFUNDe092VZfz9y3r9Dl2O/JoQQYviTAIsQYtBR7SZ+e202v702WwblhBA+kcF8IcRwkpWVxc+f+hkHIrI4HnaUVatX8f3vf5+srKx+l9kx6HPq89MDUU23wbg6RFVV9vx1n/u+3WRnz1/29fAKIXrX0tDiXvnW0tBy3uVZOgQwDv7z8HkHMAYyqKiqKvv/cdR9325ynNfvUFZWFj//1c8HtF/raDD2QzDwgWMhhBCDjwRYhBBCCCGEEJedgV5xog7QRsFZWVmsWrWK+oYGj8fr6upYtWpVvwcjB/NAbpuOewFYz/NnUvhFEeX7Ktz3VadK8Y4SCr8oOq9yhRgoqqpy8IXD7vvnGwRUVZWs1QcGrLzCL4oo3Tswv0Pufq2+3uPx8+3XhsIqtYHs14QQQgxOEmARQoh+cDqdnKoJ4HBZCLm5srxdCCGE6ExSfvad0+nk1Vdf7fGY1157rc/nHaqqcvCfHQdebYNqIHegmevN7Hm+m/oosOf5vQMyCFu2p/y8yxCeLrd9kQY6CFj4RREle0rPuzyn3UltQR07n92FiurxnIrKjmd39el36EL2awPdDw104F0IIcTlQQIsQgjRR1lZWfy/n/+KF/elsv5IIs/++fyXt3cM2OTlF0jARgghhOhkoAM2A7XiZCDl5uZSV1fnvh9jiuWKqkXEmGPdj9XW1pKbm9uncgu/KKI866z7vupkUAzktrG3VGCtzcHRmI8xsAFFcVC87TDW2hystTnYWyp6L6StLqrKnuf3YTd1M0Cvgq21/4OwHVcBHfnXsUE5W15cWB33EMnNy+33OftABwF7K6/2dD057+eSuyHf4+n9Lxxkw399Ssnu9t/nmvxaNjz2CQ6LEwWlU3EKDrOdT5/6sMf62O12qquryc3N5d133/Xo17rT335tIPuhC6Xjd0au84QQYnjSXeoKCCHExdAWwGiy6EjIzWfilOloNH2PMbctb++sbXn7ypUryczM7HOZ69atpb4+1fXAkb8THr6eu+++u89lCSGEEGLo6pg+R+8wkNE4Hp2qI6NhPHX6OmxaKwAffvghTqeTsWPHotVqeyzT68ArroHX2MkxaPRa6grq0Bq1RI2NdD/fUNKIrcVGUFwQfqFGAJorm9n9p71dC1Pay/ML8/O5zfaWCqo23AFOK1aTPw7r/aiqhsP/LsK/6dcY/E2gMRCz/E10gXFey2mpauH01jOYas0U7yzxelzbIOzOZ3aTNDeRpLmJPtWzczonh9m1H8Winy7wua3i4uu8IsE/2L/fZbnO2ddRH1EPwPFVRwkPD+/zOXtvQUBrs41PnviMW9bc5H5459O7OZtTjX2Kw/3Y2eyzbP5/2wiMDSAkIaTHoOL+1fupPFJFQHQA6ctHu59urmimNq+WlupW92OtlWUAXYIrbRQ0NOXYOLxpG5aQAGpra7v8a2xsRFXVbl/vTW9BmI7c/ZriamOHyvWrH2rTNfAe2OcyOsrKymLd2rXUR7hSPh5ffZTwN96Q6zwhhBhmJMAixGXIYrHw6KOPArBkyRL0ev2Albd69WqMRuN513EgdQ1grOrXxZCvy9unT/c9eHMhAjZCCCGEGHrKy8vZuHEjoX42AvV2EsonolO1KCjoVC1TTemUxu+jxarj+PHjHD9+nMDAQKZOncqMGTOYMGECBoPBXV5DUQM5/8ml4mBF9wOvtK/mGH9bBht/tJmgEUHc8q8b3c/v/8dByvaVMfeJ2Yy6ZiSqqrLr2T09Dgx//usvWfb0Up/b7bTUg9OKqkLOriXY7QZAwW4zcHL3YiZd9RE4ra7jzgVYzPVmzmZXE5YaSnB8MAAtZ1s59NIRjKEGkuYlUrK7FNXZdYBX0SiEpYZyauNpTLUmjwDL8TezMQQbSZ6XiDHE83y2p3ROKQuTfW6vGJoG8py9obChxyAgQHNFC/WFDYSlhALQeraVppIm/DLaf8dRFOxmO9ZmW69BxcojVcRNjyMsKcTjufG3j2P0NaMISwt1P2YKLCEqOZ/q4pGgdnNNoziJTi5g/QfvU1kfhENx0F0sRqfTERERgdFopLi49w3o33jjDWpqapg/fz7h4eHe26Oq7Pjj59habdD5V1wFa4uN3c99yZW/9L0fuhAu1HXeQAeBhBBCnD8JsAgxBAz2AMZgdr4ntk6nk6amJurr6zl27JhPy9v/+c9/EhcXh06nQ6/Xo9fr0el0Xe5rtVr+/e9/91heXwM2QgghhBhampubef/999m6dStBejNPLMintmgMx/Pi3ccoaAhpjmfuCB0Ryfl8Wb+CPSfyaW5uZseOHVR9fJZ91gOo0xxMvm4SU6ZMwdpiI++j/B7euT1IkLIohZDEYAKjPQfq/MKMBMYEoPN3XTY2FDZQebiqxzLPZle7B4YbSxrZ+KMthI8Og+ntx9hMNnR+OhSlfVS26kw61UXtM+tRNZwtHEPl6TGERFV6vMfeVVkUbS9m6tcnM/GrEwCIHBNBysJkojIiSb4iidIDZTjNnimOVFQ0Bg2T751M+YFywlPD3M85HU6OrD2Gw+ogdmK0O8By9kQ1FYcrOPLaMVTULuXteHZXv2fLi6FhoCdZhaaEkjQvkeJdJV0DBAAKRGVEuoMrANMfmoqpyczek+2rxyJGhXPzv25Eo1fI+tuBHoOKiXMTul1tFT0uyn3bZrNRUFBAVtY+rpi7ld3lidhtRjyjJyo6vZWxc7awL2ckc6rmodfoMS4wEDU6koiICPe/4OBgFEXB6XTy/e9/v9frqIaGBt5++23effddJk+ezMKFC5k8eXKXVXo1OQWUH6z1XpAKJXuqqckpIDJjVI/veaF0/s7EmGJJb8ogNySHKj9XfybXeUIIMXxIgEUIMWz5cjH0yiuveARRGhoaaGhocN/uz/L23bt3n0+1PbTlI87IyBiwMoUQQghx6dntdrZt28Z7771HS0sLANOmjUK1lnBy52JcI6+eA5s5O6/G+eU1BEYG8tyax8jLy2P//v3U/qeegNYA8nJOsrt4F1qtlvGjx5M2exTWM1YslVY03Wy/6cRJyPhgUhckk7qg6yqMeU/O8bjfNjDsbSAXBUZMi3MPDDcUN2KqNWGs9pwc9MVvt1N9ooa5T84mNkPFavInZ+eSbtt8/PPlgErUUguGCNejMROjaShuRB/Qvgpba9Cy4L/nA64JNoeNB5lkntKpegqHjQdJ141i1rc9J9g4LA7SbxxDQ1Gje1UMQOH2InLeOdkluNJWnt1s55Nff8YtT9+EGHza9r+waCzkF+QzPcL3Ae3GxkZOnz7Nvn37fJpkdeLECSZMmNDt83aLndwP8khekExQbCCzH59JxaFK1yqMjhQwBOq58ucLPR6OGB2BzWZDU9T+HdQatASPCAJg9uMzewwqzn5sZpc6Wa1WCgoKyMnJ4eTJkxQUFGC324kPMbF4nomx87ac+/3zrODYuZsx+Ju4eentHPtzC06HkxtuvZ6g2KBu267RaLj77ru7nfTW5pFHHsFut/PFF1+Ql5fHoUOHOHToEGFhYcyfMZ9U7UhmfHUaikYhOMZGYPhZWuqi6HbpzLkVNsEx02g528rxN7MZe8MYQpNDux57jr2lAqelHkdjPUERriCyozEfq1LtaoMxrMcUhZ3l5ubiNFURH2JHazcwqmocGqee8Y3jMEaU4dBZaWmtlOs8IYQYJiTAIoQYtjpvFNudxsZG/va3v/V4jKIoBAcHYzQaOXv2bI/HAkyfPp3g4GDsdjt2ux2bzYbNZvO4b7fbaWxspLGxsdfyOuZjF0KIC8XaIU+91WLBT9f/PPVCDAUdB17z8guYEh7Rr5nE/RmYO3r0KK+//jplZa69DhISElixYgWj4zRs/YnOnSbLk4LDrgdVg6nOBk7IyMggIyOD6sxqSk6X4lejx55to6ysjKMnj3KUo+idBuYq81FUpcvAq0NxsMu+g5ucN/jUdkVRXAPDhyuxtXQ/MDz/B3PdD42YFse1zy7FYrKSVdS+D0xTSRO2VhuO6g1UfL6BnF1LcNj03ba5bYr/S0//meNmDX5+/vgZ/fCL8ePo0UP45fnh5+f65+/vj9FoZNOmTZj8TFSZK4myRKNBgxMn1cazVPlX8tJLL2E2uzYRV1UVp9OJ0+lEHaniTHWycdNG92MtJa3natL9fhQaNDRnt1B7uo6INO9pjcTF12XPlNVHCX+j+zTBZrOZwsJCTp8+zalTpzhz5oxP5/0dPfvss6Snp7v/jRo1yp15YNczeyj8ooiGogbmPjkHvzA/om+IpGx9hWchKkQtj+zziqhj+cd6Dirmj2Ly5Mnk5+dz8uRJcnJyOHXqFHa7Z7q/sLAwJo1LAU4Tk5pL5en09lRh5wIXsWl5AES3/oOrvjOPxpqxBEa3nzPkvH+SwJhAEuckuFepTR2XyHcfupENGzaglIcQVzWJipijKAktXHvtMiZMSkUXGMcVV1xBeXk5X3zxBTt27KChroHG15rJcZ5kx+EvmXXzLKaMCmbq0nfY8+79Pa6wgTvJ+zif3A/yqD/TwDV/XNLtZ9dxHyiAlEljyNt7JafX/4WYVFdbfdkHCsDhcHD06FE2b3iDJxbko9OoHN16A9WqDlDQqjrmO0YzaeFH2BwKVdWngP4HWEr3VDJ2eVTvB/bAIwh56hQzYmJkVY0QQvSRBFiEEMOKqqqUlpaSnZ3Njh07fHpNTEwMSUlJhIaGEhoaSlhYmPt2aGgoISEhaLVan5a3R0REsHLlSp9OSnNycvjDH/7Q63FhYWE+tUOIy5m10ya2BokNCCG8sLdUcPzQTjZs2EBzajMAb726l08/COG6665lwtR5Ps9U7jwwN+vcIoaWva/S0nZQh4G58vJyXn/9dY4cOQJAUFAQt912G1fMv4Lq42epyz/pmSars3P7ISx5Kh2tvj1tTtTYKKLGRjGVKXyFr1BeXs6BAwfYsWMH5eXl5IRkM6mh68BrTkg2VY1VfZpFrdPXM+OBBHb/5UynusGMBxLQ6esB1+en89MRlRGFzWZDKWwfyF386wQqN/wUXXMDpqawntuMAihcN76MpX6NnDwbzMmzQZwqDcTm7P58K9TPRnionfrAA0SeWYLq1KNq7NQnHCBeZ6XFauPFF1/0qb2oMMk4xR2o6cwduLFUEoEEWAaLjmmCPdIz1VWyatUq7rzzTvz8/Dh9+jSnT5+mtLS021XrcXFxREZGcvz48V7f0+FwcOLECU6cOAEqaDVaklOTSU9PJ3FcEv4n/ImdHOuu3yuf/9vju9X2Xdr8xWeETAr2eX+OtlX7dX51XoOKf//731FVFYfD4fHasLAwd6A2fXQKQY3baT7xEqoNFAUy5m52pwprD1yce9/WCmh9hxC/SFBuBsBUa+LAiwdx2lSu/sNVxE2Oc/eT0U4rK8b4s/vo/dgdRlJrJjFn4csYSg5RVebqJx2OcKz5Nr761a9y++23c+jQIQ784xCWChOm2jz2friN2mgr00b0vsKm+cRLhEVkMGJKIKkLtNibitEGxOB06Mh57yQjl6bhH+7v3gcKwGry5+TOJdhtRnJ2LiEstgSDv6nLPlCdFRcXs2PHDnbt2kVjYyPxISb0o1RXgMpL6sPYtDxa6stRVdUjXWJvLA1W9+1DL2aTMm9Uv1IUdve38L039rJ5w0t9/lsohBCXOwmwCHEZcjqd7tt5eXlMmTJlUM1ScTqdnKoJoMmiIyE3n4lTel7KbzKZ2LlzJzk5OZw4cYKGhoY+vd8DDzzg06CCRqPh3juu5a3X/+X1mK98ZZnPn2V6ejrJsQHYTTVej9EHRJKenu5TeUJcSE67iYq3rwQg7vZtaGR1g+iFBL3EQLOb7SypuMZ9uz970tlbKqj86HaiVTv3T+7mgJLDVJbqiL3+bZ8GljoOzAFUnnbNfE6fva195rPTSnPNGT59/2M2btmBw+EgxN/JPQv9SB1hQLX8m48eOEFzbThTrnmLqOSpvW4uHZ46rcd6jRgxguuvv57IyEj+/ve/U+VX6XXgFWDz5s3o9XpSU1O77HfQUdtAqb/DSlTyDV1mtvs3PUfVBtdAqcYQjKUqC0v5HswVu0iyRnP4cAr79+/n8KGDfDOzmbyqIM42K8T6sKF2YHgjCnZmJ9cxO7kOVdFj8UunIuQrtNj0mM1mTCYTZ0tOcEPcx+i1rsHyyiSn+2ey9NzPxOZQeC1vPvqgeDQajcc/RVE8/q+pqSEn+wRzqyO8rgLKCTnBvPo5XesuLomOaYL1DgMZjePRqToyGsZTp6/DprWyfv36Lq8LDw8nLS2NkSNHkpaWRmpqKgEBAV0mWXW3n0ZERATf/e53yc/PJ397AexXKDIUuQM44NoL5cDWfYwuHs3+/ftBgZyQ9u9W23cJuu7P4XQ6MZlMmEwmysvLsdvtmM1mzGYzp06dctWth/LaVqqEh4czduxYMjIyGDt2LLGxsagOC60F79C8739osnhOJDP4mxg7b7P7d8jg377BesjU7+IwV6NoDO4AgaLTkDzhMI1VoTjz7qGmfjLawHhMDX4YA2zk7FriXqFntxk4uXsxk676CJxWWstO8dH3i3BYVQw/+piICbPJzMxk2pRp1J7ahO3QSx51622FjblkKwa2Mm4a0AxVG1yvq7c/wcG1KnmfFHDD8zNpPf0RAKqK9/p1o7GxkVMFefzuN7+ipKQIp6rgVF2ZD+bMnojVVOE13ePJXUsIjyth486NbD9Uzi233MKECRN6DbSoqsqhNdnu+3aznT1/2dft/jo9Gei/hUIIcbmTAIsQl5m2pfJt/vznPxMe3v1SeV91DNicPHmSiRMn9jtgk5WVxbq1a6lvSHU9cGRVl/q1traSk5PD8ePHOX78OJWVnhufGgwG0tPTycjI4LPPPusxDVdERITPAQx7SwWxJb9m5Tyr94NKfo29ZbJvgzKmKr45/RCKavd6jKoU4zRVoZETWyGEEJe5jmlMcvNymTK97xNEHOa6Hv/uAiiqHYe5zuvfctVpR7W14LQ1uwIs53id+Qy07voOSlEEDkcc0yZM48r0WUSZvoPtXHagkKg4zM3+WFtDPGaMe0t9U591iqDRt+OXdBUafff7HkCHVbA9DLyC6/wrKysLf39/xo0bx/jx4xk/fjxxcXEeA35tAaUeZ7Y7rdRu/xH2hjxQ22fL61qr+XOHPRjWnpjNzJkzWTAuEs2B3/Ta5sirVoG9BXP5DixlO3G0luPvKGbuwmtQNK7LWlPJ55QV1qEvaV+JEJuW5x5wbaPXqtx241LSp1/r9bNrk5OTw+rC31CpO0Riuec+FgoKlXGHCDC2yIrjQcSdJliFjMZxaFUtCgpaVUtG4ziOhh8GIDU1lYkTJ5KWlkZaWhrh4d2vQOq4h4i3gM2KFStISkoiKSmJRFMSe3buY1r4dK57eBn5+fnk5uZSWlpKWVmZOzUggE1rJSck2x2wsWld1xm1tbV8//vfx+l0YjabsXSYtPDpp596bbu38gDuuusurrnmGvfvtKqqtOS9QfOJl3GaXZvGawPj8U+5jubs9hVe3f0OARiip2GI8JykZvCzMW5ZM5bqXWA3YSnfgdOp4eCn9+N0aLGaOvRXnVZzNB94gtCoW7BZ/Kk/tpWgKC3G6Klo9VrCEyZQkxuHNiiBytoWwu05va6waQqcQ0x0JA5TFY7WSpytlagOC/4RQURlGEiYlYC9PpfW/PWoKuTvW9DjahNr9SGqNz+M6nSA6kRRVL6aACSAM0ND1uklhMUvZf7d87AU/oft/7cEu637dI92q5H9H9+JIW4fBQUFPP3006QnpXP1gquZNHsSfiFdV6TYWyo4s+005VntqetUJxTvKKFgwy5SFqX5HAzp/Lewu8kBvf0tFEII0U4CLEJcRjoule+orq6OVatWsXLlyj4HWVwBm7Xu+88++2y/Aza91S8zM5O6ujpOnTrVZQl/WloaEyZMYMKECYwaNQq93rXpaWxsbI8bKq5YscLnwZnOs1S7P6jn5eOdy/NlkMfX8oQQQojhqsteCquO+ny+0dTURFFREUVFRZSe/JLlI9qf63bFCfCvf71IWOIM4iONpJreQK9Y0agWVFszqqN9oNMvcTHg28znmDAD3/32d8n+7UmObD7N1T9+gOD4KLQBMcycE44xIg5sc6je+PVeU9/Y605Qv+83cOB/iVzwNMbYrhtYg2u1bHh4OHV1dV4HXgMCAhg3bhw5OTm0tLRw4MABDhw4ALhmu0+YMMEdcOm4EK2nme32+hwA6kxGTlYFkFcdxKnaAMLCwpg5cyazZs1i1KhRKIrimkmdY+uxzfoAG7rAOHSBcfjFz0edrmJvOIWjpdQdXFFVlcaDT+PfWokvUlJSfDpuVGIITy7MR6vkcXRreJfZ8ouv2oHdqTAiMcSn8sSF17Z/YYw5lhhLrPtxDRpiLLHEmGKp8q9k2bJlzJnj28qjzMxMvv3tb7P7j3s9AjaTzFOYds9kRoaNdB876pqRWJosjFk+GmOwkXnz5gHQ3NxMXl4e27Ztc6cJBKjyr3SvJuuou7TEiqLg7+/v3nfIz88Ph8NBYWFhr+WlpKR4BEwVRcFadQCnuRZt4AiCxj9IQOpyHKZqmnNe6fm6R2NAYwzr5uEgIq98HtVpx1afh/XsIaoOHcBmMeKwGel2NcdO12oOg7+ZKdfvxxgxAl3QfPSR491H6YISib3xfQAKP38TKlx9TE/90L83VnLWZGLEiBEkJk4hfsQIkuLDGZGQwjXL4kEFe91xjHFzKd1XRnH2jG7rd2LHNYTFlrA/K4sxWitlOZMoPTmZmLRcUief21tKVWj+cgLNlOG8zUlTBb2mPjQ1hnP3Q18nu/oI/mc/oDw7mpPP5HMoZh9zvncFEydNQlEUNv54M6g2ktJWc2jDzUDXQPS+v+egrf4Fibf/26eASGFhIW2hrp4mBxQWFpIeOa7X8oQQ4nInARYhLhMdl8p703kpem/ON2CjqipWqxWz2UxLSwuvvPJKr+/XJi4ujvHjxzN27FgKCwu55ZZb3EGVjjIzM1m5cuW5VTHtqcMiIiJYsWJFv1ft9ES1m3HaTR4plJx2s2uKUafjhLhcWSwWHn30UQBWr17dr1Q/Q0Hb5tcdB2RtdbkoWld7u9v8WgjhydfzDafTSVVVFcXFxZw+fZoDBw6wbds292ArQHyICc4FWHoaVCopKWFvdg2BBjs/WXwGAGen93egp6a2lkCg6kz3efaLjk/DP6iR5olXcsXtD6LT6TibUYO5wYw2ZimBY1ybE7fNVbbWutKG9pb6JmD07Vgrs7C3lKMPb59Bbqk6iKLzQx+e4U5x5ZHeNOQMp2LPoAPiz73mK3etYNrca3A6nRQWFpKdnU12drZ7FcD27dvZsf1LIgOtzB0NczoEqLzNbP+8IJKsknBqTQZCQkKYPn06UXY799xzT5f+XhcYR+z1b3P24A5aD5fgXx+FggYVJ+awGjQLbiO2014AiqKgDxuFPmxU+0duN2GInoap9Euwt9Cb1lPvYa89hsYYisYQhsbY/k/RGtrLtTai07gm9nibLa/TqKjWxg6fqriUnE6ne6WJitolrVtG43jqDHV9XnUU3RpDeHOE+74GDeFN4Zz5WzEt401c839Xu37vdBomfnVCl9cHBQUxbdo0/P39PQIs3tx9992MHTvWHUjRarVs3LiR66+/3uO6x+l08tRPHu817fCYUSm05L2JMX6B+/cpeOI3MY6YR0Dq9ShaV5m6wDhilr+J01KP3dTKZz/8EoBr/rgAnX+Aq+29nL8oGh2GiHEYIsahj5pK2Ka3qSlN6yYNoILd7gpIX/WbWzFETer1cwkODoaK9vve+iGNRovNZnMH2Dvy8/MjPj6e+Ph4RsZMoulIEl2DK676Oe16cnYuZXfgEd6zjCbJNJKYumgC0kdzwDifpUuXoTcYidq3Ha1Rh8PqIGL8OKKS3/ae+hAV/+B6EqdMIb45hYZ9DQSfbaaowkRydAHGQ2vJygonLGEGlUfGgQq2mpnuIH6Xz9Bm4OTOK6h35BAU30LyFUlolTrsTUU4zLU4zbWYGsppqC7E3FgO5irw731yQFNTU68/DyHE4GY323n91jcBuOvdO9D5nX8o4EKUOdTJJyDEBTAYBw7dS+V7UFtbyxNPPEFQUBB6vR69Xo/BYPD4v+22Tqdj69atPZb3wgsvcODAASwWiztHcFt+7LbbHVeihPrZiA/xvqKjxapj3lU3sWTJEiIjIwGw2WyUl5f3WI/MzEzSU5J5+aE3sWgsLPnxfKbMnNnntCKq3dT7QUDN1kfQBiURe/1b7seqN30De0N+n95PCDG0dd78uk3Nlm+23+mw+bUQw81A7JniywSRF154gU8++YSSkhKPNDodxcTEkJycTGxgM3C610Gl2bPn0KqJo/psJRsKA6msaaLZ5MRi12C2a7HYNThVhfiQBh6e5u81z37+vkVodRamLYpFp3Ndei382QL0Abpuc+1rjGGgMaA4rd5T32gMBGXch3b6D3C0lKIxBLtf33joT9jqTqALTsU/9VoMMTP6lN40LS2N1JRkli9fjs1mIy8vj6Zj/yTOsQ+DtudVtx3lN8QyadYiZs2axdixY3E6nWzYsMHruZcuMI6pV9xOwogSPn7sc0BF66fja6seIiAiwKf31OgDCJ/zFIE1x6ne9GCvx5tOvYe3Mzv/1OWEz/6F606HCTI9zZYXg0N2djZrX1nrkRqsIwUFnapjknkK6enpNJU3c2rTaQKiAxhzbXvA7uyJauxmO+FpYfiF+WGuN7Pn+X1e39c/wh+HxeHTIJOv+zAuWbLE43fGZrN122/4lHaYQqo+ug3VUktAQwFhmT92vU/YaPRhXVda6ALjXCvom5torj3peix0DIag4C7H9qaxxExNySjvB5wLSDeU2YmO6r28tPTJlJ9U3IHP7tidCj996g/Um/SUlpZ6/CsvL3fvXXPq1ClKjTpGnl3c43vWlIwkca6VWTffRkp4Cq1VrfhF+/PloS/QGILR6PVc++wy9/HWWk3P6R4NFmYsX49Wtxhd5ESCJz9GRnIuqbM/xtlcjKKoBBmrUWs/wzZZJTlqMuWbR3apV+fP8GxhOVBOwqx4qo//E0o/oOrMGKqL04hOLiA6pYAABdqWI3qbHNCWFi04uO8/byEGOwkOiAtBvkVCDHOqqlJcXNxjrt6OGhsbe9yzpC8sFgu7du3q9ThFUYgMdPLY3Hz3hqTdsTkUquKD3MEVX7hnjzfXkxRkB7SMjFGx1+cC3mdfqU4H9sZTWKuPYD17GGv1EdB43/RVCNE3qt3Eb6/Ndt9mEASiB9pApxXsrOMm8laLBT+d7CIvhh9fJohYLBYKCgoA0Ov1JCYmkpiYSENDA9ddPZ9YssFUTPislVhqTlCz6Z1eB5WWLVuG8VxaFNWpguI6R6qqqqJobxG1ZXXU+dXSVHOYnF0TvM4qBhWt3kb92fZZwIbAritu23ScOW5qqEf74SEcdj0Tv5ZI/NWrAc9zF11QYnsTHFZ0wYnYGk9hbzpD09HVvXy65zittJ7+CNVaj63uJLb6XKKveQVDcBITJkygSTOdpiO7QGOg1RlCANW9FnnHHXeQkdme7qvjfn098Q/zQ4NrQHnGwxN9Dq54UHw7X/NLvApQcFrrcVoacFrqcFoaQHWg6Nrf19lpNYy32fLi0tu6dStr164lyBzskRqsMwWF8KZwGoubaD3bwtFXjxE+KtwjwHLwX4eoOnaWBT+ZT/IVSex5fh92k6378jQKqqr6PEg30Psw+pR2GCeqpRaNfwz68Iub8ikk0Y+o5HzvqznOrdALTZzmU3n6oHhqUn/VvjKvG1+560GSQxKJDXGljJ4+fbr7ObvdTlVVVXvQJedLotTe6xe1aCJTpkwBICw5DJvNBoe819MVkO053SOAPnQk+tD24InTbqa+7AhHd71PfdkhsmrPkN+ax4LkCT3WMTLhNLawaKqrjPzgpz9gUlQxs5MNnDmdRnPheM7a7RiNCiGRSUSFh1LwUijmlhC6Tdu2y5W2zddUikIIcbmTAIsQw1RpaSl79+5l7969VFRU9P6Cc+655x7i4+Ox2WxYrVaP/zveLioq4ujRo72WN2fOHNLT0z3yBHfOG2wwGMg/9Bn6vF/0WJZeqxIa4Puqk86zx2fd5Hq8Ze+ruC+VO80eb8l/B3PpF1hrjqDaOqeX8O29oxb/E33EWI/Hopf+y7UGuwNbXS7VWx72uT1CCHEhOZ1OTtUE0GTRkZCbz8QpvqeMFKI7/d2Uvrm5mYKCAvLz8917gbSJMcW69xCp8mvfY2Dx4sUsXryYuLg4NBqF1rI9nNr5d0Jy3sbkdA2KBmXc69pzwOR9xUnOzqsJjytBURRObz3Dnuf3ETcllit/sZDQ0FBCQ0M5/rscLBU2bn36VkoLEzj2554mpihYTUH4aQN9/tzaZo5b1WosLacBSFg4CUNEzwOtitZA+NzfEGprxlyyjdYzH2OtyurxNW2aj//D476tLgddcBIAAanX4TdiPrqQFA58+S4BFf/Xa3ktLb2n6OqOoUOwPW1hcr/K8FXQuK932aBbVVVUWzOu74aL4mPARlw6DoeD119/nU2bNgEwcfoE2NTDCxRImpdIWEooqlMl/YYx+Ed4TlIIGhGEtdmKMdRIQ2EDxTtLvBanOlWKd5RQX9hAWEpor/W9VPswBmXcS/DEb3qkwLsYFEXpeTWHe4XenT6XOW3uNTj0Ea69uTqkgWxLAz2thzTQOp3OnR5s5syZnDwSjWbfT6nvoX4jZ27FGb7E5/q1rUbsMd2jt31sdH5EJM9iUfIsGhsbsXzyCcf3/IeM6T1/huPmb+KlIzEUagKhFQ5UxFHnP5eRE0cROzaCsQu+TXKmq18/e+Qgx5pz8JYWrW1VZ/xNXVdMCSHEQBsOq4qGXo2FuAx1nPV38uRJJk6c2O0ARXl5OXv37mXfvn2Ulpa6H9fpdEyePJmTJ0/2eMEbERHBVVdd5dPgR05Ojk8BlkWLFpGRkdHrcSkpKdT4MBmwbRaN024Gpw0V77NAfZ49bq51X7xYa45iqXCtulF0/ugjJ2GImowxagpo9NRsfbT3SmoN7v0V2nS+33acEEPFYF0t0bZKzW5unx3uaMzH6nSl3ejvHieDtb0XimsD8bXU16e6HjiyyucNxIXojq+b0quqSkVFBXl5eeTn51NQUEBZWZlHWaF+NgINdrR2A6OqxqFx6hnfOA5jRBkOnZUWq46ZM2cSF+lHa+5aWk/9B0dzCW3rXfUR4wkYeTNa/xhK9xxjz/tf62aACkDBYdefG1QCjU6D3WTH0uR5LhE9PoqguEA0ei3jr5hK+Xv/oqanWcVJBUxY8ND5faB9oNEHEZB2AwFpN2Au20ntl0/0+hpdWDrGmOnow8eiD89AF9w+a1nrH43WPxrouveBN/1NK9Nx/zrNJeh3FUVBMXjWveNqFjH4tLa28o+n/0F5dgX4we23387111/PgZBD5H6Uh8Pi8HyB4lpFNvuxmQCEp4Uxa2XXv3Pznpzjvq2qKknzEinZXepa0daJolFInJvgU3ClL5qO/o2QySvRh6cDYKnYRYbl39RtfRdFdaI6ba5/PqYx9ku6+qIHV8B1LmYIdPS8miPQ0W2woSeZmZlMHj+Zv6/4JxaNheueWsa4ieP6PDlkzMS5PPXqePyjjpJYPrNL/c5EHiUnfzw/f2Cuz2V2XI04a0ojG76XB6jo/XXM/dGt+IXe6dM5akhICHfeeSeHE/QYSn7f64qYEQlpzF22mDFjxpCcnOxOTdmZ3eJ0v7ZbbWnbSkxER3R/iBDCxdrcxPo7PgTAtrAJffj5/dIMh2DD5Uh+SkIMcm0DFG2effZZjwGKqqoq90qV4uJi93FarZZJkyYxa9Yspk6dir+/v9dNYtusWLHC5xPS9PR0wsPDe0zbERERQXp6uk/ldZdPuKfjTGc20LD/DwBMx0DNhn+iMQSi6IPQ6AMJnviIzyki7M0lGCLHAxCQuhx9+DgMUZPRh41G0bR3k9baHJ/KE0JcHN72OGnd9V1a2+70cY+T7gI29oY8rBbXgJsvF8OO1krMFXswF2/x6T0bDj6N34j5GKKmYIgYB/Tcdw1EHTvydQNxMfx13ENuyZIlHpso90Vv36lbbrkFrVZLfn4++fn53U7+iIuLY/To0aSnRJBy9g9oFZWjW2+gWtUBClpVx3zHaCYt/Ai7U2FEYgiWyiyajvwVAEUXyMmCK9Aznwlz5hB4buCzuVKLzRzkvfLnBpWaqvSMmD6Cm1+8AUOw5ySJ+T/wHGSLumMmVc9WolH1XTbTdip2ou6YhT7o0mx+rvHz7SI/bOb/67Kaozu+7n2Qlj7Z5zoOtLaZ4z1OsvEyc1xcWh33ULJYLL0OKlVVVbH6f/5OWs4owpUobvvWrcxeOBuAGQ9PIzI9gu2/3+n5IhVmPz4LvzA/n+ulKAqzH59JxeFKbC2d0oQpoA/QuQM2vXGYqlHtZp+OtVTsxjH6K+4Ai2ptIFgtwd5z1sRBpy3YEGmuo64xn7KDTYCCooH4GaFM+OZP+z0hRqPREG519XNj08f2a+WtRqPhxq88wKq/rMJgrCTKEo0GDU6cVBvPclKtY+XtK/tcdttqxCBDE6iuNJaZj04lJG1sL6/sym53rXjqcUUMrqDTjEXLeioKgKhx8USlfEx1UVoPadFOEz7qqj7XVQghLkcSYBFiEOttgCI6OpqzZ8+6H9dqtYwfP55Zs2Yxffp0AgI8Z9xlZmaycuXKc7OUG9yPty2l7ssAmkaj4e677x6wgE1fqXb38ClarDjNZ3Ga2z+LoIx7UfS+zZ5UNO0DSMbYmRhju79AGugLdhkAEOL8DPQeJ94CNo1fPtZ+p5uAjaO10h3cBWgt/JSmI977xs5s1UewVR85V74OXWg6SbZgLKWB6JKv9Aj09reO3jidTj54aw3xId5nv3749hqmT5d0YcI3vmxK/95773nc1+v1jBw5klGjRjFmzBhGjhxJSEgI4JrcUL1RdQ0oedkzJSrpNOVZJVhaU4iIm4t/8tXo4hZS/p2PcRTXEjWu0j2zPG3pRPI3ltJQ1NIxC5SbooGEmVFEZrj2Yuhpv5Q2M6++lqbirZSt91zaoaCQ9JVkZl7dvwGqjumyjMbBserV570PLlFACTxnjttNrXz2wy8BuOaPC9D5u86N+zuYKy6e0t1lpC8b0+1zliYrZ8pO85e//IXmpmbijCNIGJ3A+DETPI5LWZhMwaY8yrNc1wiKBhLnJpLSj/RzOn09Mx5IYPdfzng+ocKMBxLQ6euBbvZ1dFixnj2EuWI3loo92BvyCZr4TZ/eMzD9bnShae77+qip5OtvY3rmLHR6f9f5gUaHo7mM+r1P9blNF5MuMA5dYByZK6P5z0MbXY/565j75GIMfQh2XSiZmZmsfGwlb7z0OuGnIlBUBYfioDKlnJX3DtxEk6S5/esb21YPKgoe6dba06t1OM4H+qARXPHz+/jo8d3YWh1dn/fXM//n90k/KcQw4OywytF1u3+rjIcSu9nOG7e+5bq92N7vSWN9IQEWIS4AX1N69VZGbwMUbcGVtqDKjBkzCArqYVYmrpPH8WNHcuCV22my6Ihd8GumTp/Vr4Ezd8Bm7VrqG/oesFFVFaelHq1feJ/fOyjjHgLH3InVVM/WjR+xcP5MtKoZp70F1daCLnQkjtazvRcEaANH+HTcQF+wd97IdutP9wCw4JeZBEdG9rk8IYaCgV59MZB8DdjYGwuxVh/GWrUfS9V+HM0lhM1+ioDUawFXoNYcORldSAqm0x/0+r6BY7+Go7UCa/URnKaz2OuyiQOa9h8jMLl9YNZSdRCHpW5Ag0p5x3bx4OT96LXeZ6PbHGfIObyd8dMW9loetP+Mra2NlNT70WrTEnF8G2MzJqIoivRrg1jH85e8vDymTPFtz5Q29fX1bN++vddN6QHGjh3L9OnTGT16dI9pTIAe90w5uWsJxsD3ObC2AK1By51vPY1Wr8Vms6EfrWPMzFQi0yPdr/ALMbL099fw/kMfepmJrmfOdxf43OY2V339SjYXbKZ8f5VrFYsCSXMTWfRA38sa7Nx7H3g5/+tp74OLpW3muMZsp7n2JACGyMn9SrMhE2IuHktD+wqWg/88TPLsJI+VJs0Vzex6dg81xTVs8v8Mh9NBaloqt37zZqJHRHcpT1EUZj461WNA39eVJh21TW7wd1iJSr6hy+oB/6bnqNrQPrnBaW2k9cwGLBW7sVYdQHVYOpSm4Ggp9+l9/VOWoQtKdN/XBsRRp83AOOIKj8Eiq/bSByh85RfaHjie9o0JfVpJdKFlZmaSnpLMqw+8j8ahIe0rCdz/9XsGxQSTjqsHDf4mxs7bTN7eK0mfvQ2Dv2vwtK+rB4Pik5n9X2rXVV7AnO/MISj+wu6DJYSvBjplVl9XS4qu5DPsSj4BIQZYbym9etPY2EhJSQn79+/3aYCiP6lbNBoNIyNdK0CCR486r5PGthPRlx96E4vGwpIfz2fKzJk9lqmqKpbyHTQd/xeqrYXoa3sOJHmjaA1ojOFYNBGunOGdotKdAyyVp8e4T0RjUn3Y8KUbbRfsNDe5L9jLswMZtbTvS707lmdVq2mudW1kW5kTTOSNvafpEGKoGajVF6rTjqO5FEvlPp/etz7r9+gC49AYQtBHjCNw1K3u5yxVB1F0/mgMIR4r43pS+8V/eT6gaHC0tO8ZYYgYR/TV/8Ram+NTgMU/+RoMERmoqoqjpRxT5QEKDn5IclIKiqY9XVj9vt/gaPa+yW5fNTQ0sG/nZhaHew+uAOi1Kq++vBr9f7aSkpJCamoqqampJCQkdOl3O/+ME8POPVHyf9S0Vb2PadvExdH5/OXPf/5zj+cvjY2NFBYWcvr0aU6fPs2ZM2c8Nhpu421T+iuvvJI5c+Z0Ob4zp9NJzq4l2O0GvG3EW3RsOuEjAwhLjcHWakcb6vq90Y/VMWX55C7fU78wP2Y/PnNAUge5a6IoTHtwPBX7Xeceen8dsx/v+0BuRx33IFEG2T5Q/Tn/G6pkQszFoaoqB1847L5vN9nZ85d9LPppe5BSH6yn8mQVTouTAE0gGfPG8o1vfAOjsZu9Ds8ZiAH9tgkYPa4e6DC5QXXYaDz4rPv1Gr9IjHFzMMbNxhg7C0drpU/nB8NRx34t6Yq0Ho68NLSGQIIcrgmLM2+fM2j6tK6rB+2QtInDZUCZ63Psz+rBgVzlJcTlyuoRbLByvju3SfBiaJKfkhADqC957C0WC6WlpZSUlLj/LykpobGxsU/v2ZaPtU+vMdvZ8tJ3Abh+UdclwX3VMfftmB4CNqrqxFz6Oc3H12CrdwUn0BqxNxScdx1645r9ugS7zUjOziWExZa4Z/v0h7nDDLusvx8hYWbKec3AsjS0DzgfeekkIxekD6oZXUIMhP6m9FKddppz1mJvOIWt8RT2xjPgtHktojN73QnsdScA8LM0uAMsqqpSs20lqH3vB/VhYzHEzMAYOwND9FQ0+q6rB/s661lRFHRB8fgZoyk85mDCtPZNTFWHBa1/DI7WSp/aXrN1JYpWj6L1Q9EZz/3vj6I10qjG8smJcPbv309sYDOL5/XeXlV1cubMGc6cOcPnn38OuNJSJiQkkJqa6g68xAW1DugKG3Fx9Hb+8tBDDxEWFub+Dpw+fZqampouxyuKQmRkJNXV1QDoHQYyGsejU3VkNIynTl+HTev6foSFhfVYJ9WpsvPp3ZTuLcbaPLqHAzVUF41h2R+SiZ48zec2dx1Ucm1SfT6DSsbQ9jReU78xflD8Hb+Qqy98Pf+7lHR+Ou75eMX5l3NuQoxN2+KeEKMNn4AhIuy8yxYuhV8UUb6vPc2e6lQp3lHClp9tY/Gvr8RkMvH3f/6d4oASWkKaufa2a7n55pt93scRIHHO+f/d8bZ6oCOtfyQBI29BF5yIMW4OutDRHvV0tFZ2ec35uKC/5x0CIppBFui93FyI1YOKojD1gfGUZ507t/Pr3yovIYS43EmARYgB4ktKrxdffJGdO3dSWlrK2bNnUdWus4YVRSE6OprQ0FDy8npfZdHbAMWF1JYGxtFYT1BEFQCOxnysimtgpW02n+p0YC7ZTFP2S+5giqLzJ2D07QSl343WPxJ7S8UFuzBQVTxmv9ptBk7uXsykqz7qc1mu8lT2/+Oo+77d5Ogyw66v5R15+aT7vsN8fuUJAZ6bVa9evbrH2Z2DTdOxf6ILSSF06rmVIoqW5pxXUG3N7mMUrR+agDgcTWd6LS9o4jfRGkJwWpvQBSe1P+GwoAtOwWlrQrU2dkrh4V3klaswxvZ+Edtx1rPV3OReqROy4C8Y/PqWEk3RGola/Des1ceo3vyNXo9X7c2oXuLvZ6qC2HvANYickJAInO61vMcee5ySegPBp36HzWqiocVBsxlMthJM5fs4XaRl/2cG6k0GVvoUsOl51Yy4eHw5f3nhhRe6fTwuLo7U1FTS0tJIS0sjOTkZvV7P97//fepq68hoHIdW1aKgoFW1ZDSO42j4YSIiIkhPT3eXY222UppVhq3VTvryc8EURaWhuBFrs4PgyAqaamN62Ii3gNBE34Mr0HVQSeenHdBBpYTZsQNW1vno2A+pDgs1W1x7P0Qu/geK1vV3QVZfiEvNXG9mz/Pdr0otyyon5/Nc1m16hZKSEnSBOh566CFmz559kWvZLjYtz72puKf2v21hM//b6+sHOiByIfcdGqggpRgYF2L1YMfJAZO/PnZQTA4QPet4nbdkyZLz3l9iKF83Dkdqhz1TXLcDL11lvHA6ndQZarFoLOTm5TJlet/SCg9HEmARYoDk5ub2mtLLbDZz8OBB9/2QkBASExNJTEwkISHB/b/RaMTpdLoGKHoos/MAxcXUOQ1MyiRX+q3T6//Snn7rXBoYR0spdbt+BoCiDyRwzJ0Ept+FtsNFg7f0C1f9Zjb+oa7j+nph0HbxUlWQ6nVj3NhRhX0O2hR+UUTp3q4z7Aq/KOrXzNeBLk+Ioc5Svh170xl3gEVRFILSV4BWjz5kJLrQUWgD48g7+BlBTb/otbzjZX5MmL2M4OBgjxmkis6PmOtec9+3Vh+hevPDvZandLNaxZu2Wc/Olvr2x0LHYAgM87kMDxrfTt3Cr/hfdIEJVJQVcmj/Hk6eOILitKLXOjE7/Vi0aBGLFy8mLqiV6o1f9lpeZGQkI8ZkUF5Yg2qwENnNftsljUG8f6xrDvzuFBYWkh45zqdjxYXly/kLQGhoKGPHjnUHVJKTkwkI6D4Jwt13382b//cWMZb2IIMGDTGWWGJMsdx2663YWmwYg10X8PVn6tnxh10YQ42kzlVozv4H2oA4pn39LhymEqw5L7D3/Xuw24x03oOlPUXPnX1ue2h8jPv2zG9nnvegkjJIU3q5+6EOF+z68PTzno0+WNsrhhZVVdnz/D7sJi+zAhTY+PxGSoJLCA0N5fHHH2fUqFEXt5I+8201zUBOwOhYZuc0xrrQMRiChv9mxoPdQK8CupCrB0fMjOn9IHHJ2c12llRc476N75cmPpV3vgEWCdgMb21phesj6gE4vupon7ZFGK4kwCLEAOku73h35s6dyxVXXEFiYiIhISFej9NoNNx9993dpuxos2LFin6dUBk6/IEzGrsZJfNBx1Q/XtNvnUsDY4iejnHEfAyREwkccwcaQ/cn+t3tR1JxIpixy/u3H4kuMI6Q+ev48vXdQNcUQLn7biL5lumgi2pvl8OJtcnVro4DLdYWK06bE7vZ3v0MOwX2PL+X2MkxKBoFm8mGIdCAIcjgLreprBlUlZCkEPcgb+2pOnb/aW+P5cksIjEYOJ1OTtUE0GTRkZCbz8Qp033uf1TVib0+H3PFLp+ODxh1G8Y4z70Zgic+1OW4pqYmn64nPvzwA/7x6iZ0Oh0RERFe/4Vqzz9lojfn8/n1R3Z+BZ/u2N5hJaQfI0aksfiqxcybN889MG6tzfG5TFVViVj0PKqtCae1EafV9b967nZGUCLWJD00/LXXspqamvrTLHEB+Hr+ctddd/m0ZwrAxNETybUU4MDh2vD9HBWVyc1TyfvdKQLvD2LiVycAEDk2ksj0YEIiz1D50TNo9TYUrZHYm76BvTmY6pJmxs7bwvHPl3d6J4WxczefV8rPNsnzk3o/SFz2jB7n0DJgNBAaChso3tnD/mIqhLdEMDptNI/++FEiz+17czHYWyrQGLxfr52Ptuse1WLh/30yHoDVt0z2uE4TQggx8Ab76ouO9cvLL2BKeMSgqV9ftkW43EiARYgB4mv+34ULF5KR4VvAIDMzk5UrV3rNszoYOi5f0m8pikLkwmd8LrPjfiSHXswmZd6ofgUZVFVl/4uF2M3dD5raWh18/MQ+Zj2mkn79GACaypr54JsfYQg2cOf6293H7vvrfk5vOUNYamj3M+xUsLW6NuM0BhvJ/6SAKfdNYtKKiQBYm6x88E3XZ/K1j+4CxVW/L36zvdfyJFWYuJTsLRUcP7STDRs20Nx8bjb6S//HhuAQrrvuWiZMndftDEuHuRZLxR7Xv8o9OM21oGi7HNedgJE3Y4jouZ+srq5m/+FclkUp6LXe003ZHAqqLhhFcWC326mqqqKqqqrbY+NDTBckvZVrls9a6utTXQ8cWXXBZ/m88847lDX6o9VqmT59OosXL2bs2LF9ylXfmaIoGKOn9HhMjPUTaOjxEACCg2VG7WDha6pRX49rm42uWlWP4AqAgoLqVEGFxhJXkM3RWkVT9r+YMv8/7v2Q/BKvJHjiI2gMIWiMra4Vsam5VJ5Op7p4pCtV2LnUYLFpef1OISqEuPRCU0IJGhdI04kmFLoOIDlx0hrRwvd+/T38/C7OpCN7UxHNJ/5N65kNhEz+FoYY2ZNisLocU5jJ6kEx2A30ipiBdKFXX5TvqyL8xv6nXc3KynKN/0W4LqiOrz5K+BtvDFj9ineVMWrp2H691pe0wq+99hrTp1/YiYS+sFgsHrf9gy98XykBFiHOk8PhYOPGjbzzzju9HtuflF4DmWe1bc8Uu6nVvWeKvTEfq+qamdWf3LxVZ9K9p9/qNjdxz1RV5dCa7PY6m/sWZDj62nFObznNpBUTCR8Z1vOMuHNaa3yf+Vp/xvvIYVtqr9Qrk9EatSja9p+RolVcq1nOBVYUFBoKG2gub+61vPrCBsJSQn2uoxDgOgFqc/LkSSZOnNjnfsPeUkHlR7cTrdq5f3I3B5QcprJUR+z1b7v7jtYzG2g5+Tq2+pMehyo6f3RhY7FVH+prUzxUV1fz4Ycfsn37dhwOB9l+owk0eEkrAugDIvnN/z6P0+mkvr6e2tpaampqqK2t7fKvxWrD5ug9YPOzX/0vxpAEIiIiCA8P73Y1jMHgWr020LN8FEMIdqeCTtNzHbV+EdyyeBkLFy4kPDzc5/LPV0pKCjU+dP0pKSkXvjLCJ23f1Z705fzFl9noAOO/Mg5TyefU7fqpe0WsMW4OwZMexRDRnj6uYyqd6eNr+OzHZwAVvZ+OuT+6Fb/QO2UPkUtEVnOIgaCqKjvt2xmvTEKrKl1WvTkUB6ejCnzqq86XrT6f5hMvYSreDKrrPMpWlysBFiHEsDbYV3QMlAu1+sJUb8aJEwWFQ/86Qer80QREdJ9G91LUz9zQHmzI+vsREmam9GsCsy9phWtra8nNzfV5UnmbzgERnd/QC1cMvRoLMYicOXOGl156icLCQgASEhIoLS31enx/U3oNRJ5Vb3umnHr9z132TOk8SKE6LNgaTmNvKEAfnoE+zJX32JUabDGu0RLPnOgndy0hPK734EZnhV8UUZ51tr0kJ93uR2JpsnL45cPUFzWgzmsfaLQ0WmgsaaL6ZA2pV6WQNC+Rkt2lrhmznSgahcS5CUy9r33kODQppNtZUPN/MJd535/DF7/Z3mt5V/xofpfnjMFG7nzzdo/HQlNCfaqfBFdEX7XNzGnz7LPP9mtmjsNch+Jtp/RzFNWOteGUu99wWpvdwRVdWDp+cXMwxs3GEDUZW8Mpqjfe348WwdmzZ/nwww/ZsWMHDodrlvv48ePJyMjoMcC98t4H0Gg0aDQaoqKiiIqK8nrs9u3beXbd6h4DNi1WHQ1mFRpLKCnx3scFBQURHh5ORUWF12MAXnrpJZqbm3E6ndhsNux2OxaLhZycHI/H256rq6ujurTnoFKLVcfDj61kwoQJPb43DPxGu76ukDmflTRi4FRWVvKnP/2p1+N8OX+pya+ldE8Z426OJCqlgOqitB42pT9NUJQJRTsRRdGgj5pC8ORvYYzufqP6tlQ6AWo1qEUATH1oIiFpvX/HhRhIOj8dX/3PV9iwYcOQvPgfaHaznddvfROAu969o1+fSW5uLmebznIi5DiTGjxXSSoo5IRkU9VY1a8BG19Za7JpPrEGc+kX7seM8VcQPO7rGKImYW+pGNC/lReSzk/H4q8/d+72LZe0LkKIwW8o7KcxEIPvnVdfxJhiSW/KIDckhyq/SqDvqy/sLRUcO7iDvatK8CcKBQWH2c7alf9k9mNJXjM9+FK/7vRndYiqquxbfai9zqb+ZUlxOp3s3r3bp2PXrFnDFVdcwdSpU0lMTPTpuk/tsE+g63Zgn+o3GMhZoRD9YLfbefPNN9m0aROqqhIYGMidd97JggUL2L9/v+sPVIec5oMhpZeve6Y4Wsqw1eVgb8jHVl+AvaEAe3OxexZX8MRH0IeNou5MC4c33exODeapPVVY/E2w/4WD2FpsjL89g5BE12qZlrOt1OTWEBAVQNRYVy5lc725+/1NgB3/twtTnYmMm13LGXV+WvI/O4XT5iRwYnv0fdQ1IxkxLY7IsZEoisLsx2dScbgSW4utcxXRB+iY/ZjvM9IGe3lCgO8zX2w2Gw0NDTQ2NlJbW8uZM2f46KOPaG5uprGxkcbGRmg+xX2Ten/PipNbGBnvyq3ln7gIjSEIY+xstP6eedL7M5hfVVXFhx9+yM6dO92BlQkTJnDzzTczZowrtd+IESMGJJViVFQUDWY9DWZ9j8c9+OCDhIWFdbsKpra2FqvVSnNzM83N3leotWlpaeHll1/u9rn2/VM6672Ovu5x0nF1gKmhnq0/3QPAVb+ZjX9oGNC31Y0DHbARF05DQwNPP/00jY2NJCcns2zZMt58881+nb9YGi18+sRGnHYnMenpZMzZxO6y+3vYlH4TTstXMERkEL1sLdog3y6+OkqYHdun44ciGcwXl4O2PqfKr5IqcyVRlmg0aHDipNp4lir/So/jfNWXlfvNJ9edC64o+CUtIXjc/ejD21ftdfxbaTe18tkPvwTgmj8uQOcf0KW8vjAajaxZs6bPrxMXhqzME5eTi7GfRunuMtKXjTmvMs53hY3dbmfHjh3u1Rd6h4GMxvHoVB0ZDeOp09dh01qpra3llVdeIT09nbCwMPc/f/+u6aXaMj2op0YSUN++R6CChoD6GJxfvkNl6dMemR46t6nturuhoYGcnJwLsjrE1wnMPcnJyeG1116jqKjI4/HuglTgun5/5513eOedd4iMjGTKlClMnTqVjIwM9Prur2EH874zvpIzdSFwRcEfffRRAFavXt3jydSxY8fYvHkzJpMrwjp79mxWrFhBaKhrlcFApvS6EHrbM8VhrqV+1/9zH281+9FQlYpG78+IKUa0ATEAfP77AqxNPVxEnEsV1lBioujLYlqqWhh97Sj302ePn2X7H3YSOyWGpb9f4s7Z3iXQcI7T5uTY68fdARatXsv0b0xFH6LnaPUR93HhaWGEp4W57/uF+TH78Zls//3OTvWD2Y/P6vPSyMFenri8+TLz5W9/+xsGgwGz2dzluUOHDnncjw/xLX1e1sETHGv6DxMnTiQ1NZWA1M4bUbt0HKAwtzaS89EPaLVpiZ7xGKPTx6EoinuAoi2wsmPHDne6s86BlTad+91FP5jDjDlz+tzvpqenEx4e7j657e6kMSIigvnz53stW1VVWlpaqK2tZdeuXXzyySe9vm9ycjLR0dHodDr0ej1arZbi4mLGjBmD0WhEr9ej0+nQ6XRUV1fz6aef9lqmr/tlQPvqAKtaTXPtaQC0IaMxRHhf7dNTWW0/45b6Wj781SasGisz7pvEuEmTPX7G4tIxmUw888wznD17lpiYGJ588klCQ0OZNG4S6+59HX+HP0l3JHDl1xd1+10315upzqkhcU4CAMYQI2mLU3FYHej8tRj8TT5vSq8L9n1z+Y6bP8tG0EIMDy07W4kyR1Ptd5ackBPMrY5AURUcioOckBPu4/ryd63zyv1ZN7keb979Ku6pDxo9McvfQhcYR/D4r6NojQSNuw99SGq3Zbb9raS5iebacyt1Q8dgCJI9xcTQZvAIKl34VHxicLgQKzraWDqkpDr4z8Mkz07q97hGlz1Jellh09LSQnFxMUVFRe7/S0tL3RP1UCGjcRxaVYuCglbVktE4jqPhhwHYtm0b27Zt8yjTz8/PI+ASHh5ObGAzKa36XjK6vMyXWz7irCnYPbGx7f+mpiaPlN6+6stkA68TmBXY8/xeYifH9PhzqaysZP369Rw4cABwfQ6KomAymbwGqcLCwrjxxhs5cuQI2dnZ1NTUsGXLFrZs2YLRaGTChAlMnTqVKVOmEBLimvDQ9jM2+PuR3pTB68+9yysjzm/fGc8VMV3HPS4ECbAI4aPGxkZee+0197K4iIgI7rvvPqZM6brh70Ck9PKmdE8lY5f3fdCrTU97phgDmqn9Uk+I3wyCE+PQh46iIi+Oo69XEZkewaTvLnO/LGZcEBWHy7Fb/XpIAVJAaOI0Jt41HnO9mcCY9jyU+kA90eOjCE8NA3zI2Q6Y6y0e+5Fk3DwWm83GsQ1He3xdysJkTm85ReleV6qettRbvkbsh1p54vLlS15Up9PpDq5otVpCQ0MJDg7GbDYzevRowsPDCQkJISQkBEvNCWj6W6/ve7jQQdnRd3n33XcJDAxk3LhxTJgwgYkTJ3ZJyaULjCPrRAnr1r5KfcNI14P7PyA8fDt33303yckRfPD6i+zcubPXwEpHHfvd0SNH9jsd4913382qVau8njT2lipJURSCgoIICgqitbXVpwDLihUrPGYi2Ww2NmzYwPLly7vM8nE6nezdu7fHn3N/9vsaSB0DNkplIUYgJWMRxkgJqgwGNpuNP//5zxQVFRESEsL3vvc99yQRW5ONIEcQAFUfVWO9zdrlwqu5spn/PPwRqHDbKze7n5/z3VkoioK1Nofq4/S8KX0/aTps5qsZhBv7Xm4zny/HzaXFwKo8WkX5xkqmMI3dUTto0bWQE5LtHuSzaV0Bkr7+Xeu4ct/7QTbXcYFx6MPGED775/1viBBCDDEdrxt7WtHR1xUTqqpy8IXDqLj2nbW12tjz/D4W/axvKanAc4WNRwCorpJVq1Zx7733EhYWRlFRkftfTU1Nt2UZDAasVisx5lhiLO2roDVoiLHEEmOKpcq/kvHjx6OqKvX19dTX12MymTCbzVRUVFBbWkuYLRwnBRiii1hgWdJjRpecXYv50m8jZY3ez1mDgoIICQlBq9XSePaUOw10cFM8cVWTqIg5SlNwGeBKA719+3ZiY2NJS0vr8bNrm8BsN3WTVloFW6v3VGFWq5U333yTLVu24HA40Gg0XHnlldx8883k5uay6i+rvAapvva1r5GZmcnixYuxWCycOHGCQ4cOcfjwYerr6zlw4AAHDhxAURTS0tKIjo5mz5496B0GJjdOd38Hd+l3DNgqqvJ9VYTfeOGvQyXAIkQvVFVl+/btvPHGG7S0tKAoCiNHjuQ73/kOwcEXZ8aSpaH9AuHQi9mkzBvVpxkAqtNO66n3et0zxT+0lqazZcz7wQ8In+3qsGOCmwgfuZ2I0REeZV7xw+mUvH0fu99a4T0FyLztaIzfZMx1XTuzhJnxJMyMd9+/kPuRKIrCjG9OcgcwdP7nl3rrwpenldRgol+8nVB29tWvfpUFCxYQEBCAoiheB/QtNSHUbOo9wLJ8+XIO5jWSnZ1NS0sLWVlZZGVlARATE8PEiROZMGECGRkZZGdn97gUXVEUVNXVB0ycOJGbb76Z0aNHdzn+QsnMzOTb3/42u/+41+OkcZJ5CnN/OKtPJ3idV8R0p6+DRh2DQN70d78vMfw5nU7+8Y9/kJOTg5+fH08++SQxMa6VqaqqcvjFw+5jHWYHe/6yjwX/PZ/m8mZ3is+g2CAiRobjdKqYak3u85HOKb4UBTLmbmZ3eSJ2m/FcarAtF6mlQoihIGpsJInL4tm9Yw8tuhYAqvwr3WnB2sjfNSEunI4TFpRBOHlBDKyGhgaOHDnC1q1bXQ/0sqLj73//O2PGjCExMdH9LyoqymufvPWlbZTvq0BpGx9SoXhnCVvWbGXxA1f5XM+OK2y8BYBeeeWVbl8bFRVFcnIySUlJJCcnk5ycTHh4OP/9xH+TUTneHfxpo6IyrnE8xKp873vfQ6PRcODFg5w9UcPEb45DG6elvr6e0n1llL9aCeHQEm6gem8P16iqhuqiMYyLDWPiskhCQsIIDQ11T2Rsm+So07mG5W3NZZR/cBs6jYrV5M/ud5ZhdxhJrZnEnIUHMPibsDkUnv0SfvWr44waNYqlS5cyY8YMdxkdnfwgr8cJzKpTpXhHCfWFDdQW1HJ03TES5yVSk3SWTZs2YbVaCbYFM3L8SL5y71dITEoEXNfK9y66j7L17fuMtgWp7l14n8e1stFoZOrUqUydOhVVVSksLHQHW86cOcOpU6c4depUj9/B/q6iMtWbceJEQeHQv06QOn80AREBvb/wPEiARQjwWJp38uRJJk6ciEajoaKigpdffpmcnBzAlcblnnvuITs7Gz+/i5O6SVVVDq3Jdt+3m33blEp1OlA0Wmpya6g6fhZt7SlO7e05wu6wGknIDMUvtL1tIQnBXL/qui7l6wLjSLz932RGnmb3X850KW/mIxkkLvq3z2lgLvR+JH6h7TNJMx+ZdN6pty5kedO+MUFSg12G+pKqsDv5+fm8++67Ph2bmppKYGDPG8c5zDU4zdU+lTdz5izmL8vA4XBw+vRpjh8/zvHjxykoKKCqqsq9LFhRFLRabY9lqarKxIkTueWWWxg1alSPx14o0a0xhDe3B5U1aAhvCie6NaZP5VyoYEhmZiYrV64ckH1nvDnf1ZKdXayZQ8I7VVVZt24dWVlZ6HQ6Hn/8cVJSUtzPF35R5A70Q/uF19t3v4tWr+WWl25Co3N9V6/69ZUYgvS97pviShW2mby9V5I+e5tHajAhhGhsaeT9wveoD6wnKSmJpsYm6hvq3c8Phn0sOxvsm8hrdP7Ef3XPpa6GEOIC6rhfRX5BPtMjvA9AO51OioqKOHz4MIcPH+b06dMez/e2oqO+vp59+/axb197qimj0UhCQoI74NJ2O/tANoVvFqND1yWAUfhmMbvTdjPnyjmoqorJZKK+vp6amhqKi4v57LPPaG5upr6+nsbGRiorK12T1HoJAMXExJCenu4OpiQlJXV7nauqKrMsczGpJo+6AefK1TFX154GuragjrPHz2I+O5JRU0cSFxfHiKB4dh3cQ0i8nrgpR8mtyG9fqd31HQGFEcFVjNOtR6dNwWAYz+n91+NMUQmaG+RxXaxaG9Fp1B5T+uu1KnNmjGfT7lMUFBRQUFBAWFgYC0YvYISSwMRbxhMc75oE3jF7THc6TmAu/KKQprJmdm7byQGda5LkiLgRTMyZgvNLJ8H3hrhfV7y7lIp3q7opEKo31GC+1dztWJaiKKSmppKamsott9xCXV0dn376KZ9++mnP38HaSp555hnS09OJjY0lLi6O2NjYHsdj9+3bx+4/7iWUMBQU7GY7Lz66hrk/nH1BzyckwCKGJKfdRMXbVwIQd/u280oTkZWVxbp169z3n332WcLDwxk7dixZWVnY7XYMBgO33HIL11xzDU6nk+zs7B5KHDj2lgrObDvd7aZUBRt2kbIozR3AsFvslO4royE/n+QJezCXfUnMdes5+WEepzaeJnXelVQX9TAQompobYhkwR0jiJ48wqf66QLjGLU8luK95i7prUYtn9vn9l6s/UiS5sb3ftAlLC9xjgxCCt+ZzWbefvttNm/ejKqqHitAuuPLiglrbQ6123+AounbaYJWq2X06NGMHj2am2++mdbWVnJyctwBl8rKSuz29mXK3jbGu/766y9ZcOV8c9V2dqGCIRdivy9nh/S0h/7V99WSnXVcfXnkpZOMXJAuweNL6MMPP3QHOh9++GHGjx/vfq6n772lyYoh0EBjSSNh59J6GoN9z9Eem5Z3XmnBOpKUVOdHBl7FYGFpsqDqVZ577jnq6+uJT4jnxz/+MQ6Lg5e+/m/X37X/d1WfNzIWQojhrm38qj6iHoDjq48S/obnniQmk4ns7GwOHz7MkSNHaOhwDQKuyXaTJ0/my41fel3RkdE4Hme0g69942uUFpVSnl3O2YpqTrUUYLFYOHXqFKZjZmqtdXzmt5Fq41km1U8hWo3pNoChU3Uc+99sNr66iWKKsNlsoIKfww+L1sL+/fu7bW9vAaBbb72V2bNmo6oqGq3r70VDcSNHXz2GRqdh3vfmuB4rbMBcZO5St451bM5ucaekH39bBqOXjSJ6vGvCmcNcg860hRnLP8VWcxRqIWOuv3uldpeMLgYLExZ9hDHIBk4b9oZ8rK0BHF9/AhS44/XbqN7xEDjt2EjHGOwa0+wppX9sWh43LLuJOVODOFRyiK1bt1JfX0/JZ+W0Wi3kFJ/gqkeuJDU1ldhJsWR+awaHXz6MrbVTmrAOE5hLSkrYXLyJ4ogSbKqNoKAgRo4cycMPPMymJ7bSWt1KYLQrWKOqKgf/dQinrZu9Y1SwNtvY+fRuFv/6SgBKdpdStLOYuKmxjFzcntKs4lAFOn89KYkp7tVJ3r6DdYY691hCR2FhYR4Bl7bbxcXFvPPsu0xqbt/KwTVZMoI3/+8t+D4XLMgiARZxWeuY07Gjuro6914rEyZM4L777nOn0OjPRlT9YW+poOTt+8h6awXQtcPe89c8Go/9jUkr/4ii0dGY8zFf/tb1Kx161yfo/cyYS78gbvI4bC02YqdpaS7Z7D3C3mHPlL4Y6HRZQ2U/ksGeB15cHo4ePcrLL7/sTg02f/58MjIyePHFF72+prcVE6aiTdTv/RWqw4I2MAE0enDavB6PxoDGGNbtUwEBAUyfPp3p06cDsHHjxl6XekPfNu8bSOeTq7YnFyIYAgO735eqquz/R/t+VnaTo19t7Vhex9WXDh9XXwrf9WXV2+eff84777wDwN13382sWbPcz/X2vVc0CjGTot3BFW9UVcXeVNj3hgghLisOq4NPn9xIta2acrWc4NBgvvvd7xIQEECLrcX9dy19TPqgDK5IoFIIcan0NH61atUqrrjiCurq6sjJyWnf1B3X5uTjx49nypQpTJ48mbCwMNfEvC+1NKpNXgMi8w0LmDp1KqNHjOa9v31AqiGEH77zAyorKykpKeHU66exn3BCCJjsJo9ASGcKCn5Of7TVemyhrmvLIL8gZhfOA6BpeT3h0eGEhoaiK9fTeKaJ3Xm7eh18r/u4gdd//yazH5/JyCWuQXzV4eTMtkL0gXr3BMTQlFBCkwJoKG7tvn4aSJgZ5U5JH5/ZPom2peB9Gvb/AVSH+2BdeAbUZjN23haOf768S2vHzt1MZEIxUUtfQmMIxd54ClOdnQl3htFaY8IQpMVWdxKcNo5/kUrV6RRGzphO4eFZeEvp7xfYxNZHjqL103Hn+tu54YYbXKs1XtlLaUUJFUVlfPHU54wZM4alS5cybfk0ympKPdJ5nSuOkKuDWf+f9Wzbtg1VVdH6a1m6dCnXXnst27Ztwxho5JY1N+J0ONsDV4UNNBY3ev0ZA5RllbuDVDV5tZzaeBqdQesOsKiqypaffY7T7mTiz8YzqWUsunOrkzw/QQWdqmVuzRy00XpsmVYqGsqprKxEX60ntjCOhrIGTgZ87n5NcksKeoeecSbv35k3X3mzXynHfCEBFjEkWS0Wj9t+/Rjg7pjT0ZvAwECeeOKJXlPadGboMMhh6Ocmpw5zHTk7FnhN6eV06MnfM5/EGT/HVnMMVAdRSTei97OiH7GUyCnXYoiZzsg0DSOvTsPeUoG+9pfsfstLhL3Dnil9NZDpsgY6YDNUSO5bodpN/PbabPdteug7mpubee2119i507XaKyoqivvvv5+JEycCrpPodevWUl/fhxUTqkrL8X/SevJlAIxxcwmf+2ucthaclnpMDfVs/alrQOGq38zGPzQMAI0xzOdUgElJSefeq+el3mFhYT6VN9AaCht8zlXb1/2gBjIYciF4Sw9V+EVRvwLchV8Udbv6sr/lif47cOAAL7/s+r2+8cYbufrqqz2e9+V7X7KrtMfvvep00HDg/2g98xFhs34BGkPPG0z3EJgVlzdZpTT8VR0/S2NpE6oC2jgtjz/+ONHR0QNWfk+reIUQYqjyZfxq+/bt7tsxMTFMmTKFKVOmkJ6e7rHPJrjO/5pPtKCh+2uSjis6AmMCCIwJwBBsBCfEx8cTHx9Pkj6Z+jP1RI+/huzK42Q9c4AoS3S3ZaqoNGubybh2DA/f+iChoaFYqi188OgGnIqThx59yF3HXc/uoXLfWdINY93Xi53r1rZHZlh4GDWWWprKm93PB40IYtqDUwgeEeyOVThaK5l0xSqvewhrdRZSR/4v9qax2BpOoQ2IwxCRAYAhYhyoDvQRE/BPuQb/pKtxmKqp3ng/Mam5VJ5Ob5/IfG7icvvqbQVdUDy6oHj84iF8wrl3VJ3EXPsqtoZTWLeWoKpaqotG9pjSv/DYDPQBGoxh/rTWtBIUG8TcuXOZO3cuBQUFbNy4kaysLPLy8sjLyyMwMJCW5hYmGae4fy5OnFQbz7J592fu0mfMmMGdd95JTEyMa2VRB23BFfBx3+Q57fsmj5geh9aoJWJkmPsYp91JaHII1mYrsf4Gwlu9Z89R0KB3+kEl3HXnVwmMcaV+O/T6EY69fJz00WNR5zqorKykoqKClL1p6FVDl+CKqyzXdya2cAS5ublkZGR4fd/+kgCLuGzl5ub2uPEwQEtLC3l5eRfkl683jSVmz6WB3bC0hlCXV0JQuAN95CQW/GgS/slL0OiDuhw70HumeDMQ6bICIsPct2d+a4aklBHiHFVV2bt3L+vWraOpqQlFUVi6dCm33nqrRx7SzMxMxo8dyYFXbqfJoiPhyv9h4hTvMzVUeyujbW/TejIXgMCxXyNk8koUjRaNIRgC47Cq1TTXunL2akNGY4jo+/4cbZu+68sMXpd62xNsfdr0fSDVFzag0Wtw2p2uk/FOOuaqzfukgIRZ8QREDI6A6PnsmTKQadHMDRZaKpvZ8+eBS7Mmumc321lScY37dncrWHJzc1m9ejWqqrJw4UJuvfXWLsf4dLE0N8F7cMVhpW7PLzEXbwYUVFszMcvfHLDA7IUkg/lCXHy5jSfZE7ETvdPA/Q/fz5gxYwasbFVVadj/v6g9rbwVl5TOT8dX//MVNmzYgM5PhqOE8JUv41cAV111Fddccw1xcT2fZwVEB6BoXJOgutP5/O/Wl2/ucsyIaXGMmOZ6nxg1hpyQE8ytjkBRlS6rB+yKnYMR+3ly4RPuoLoh3sBX3rqVDe9t8Cx3ehx2s53CL4q81r9tj8yUhclMvHO8e/AdQGfUMeGO8R7HOy31GIwNPa44MRgbqPrsPrCb8E+5DsOcX7rKCxtDzPXvogtqH+tSnXbQGFCcVjLmbnanCtPprYyds+VcJb1PKFIUDbrgZHTByVy/Ggq37uHLP3ptLqgaqovGMH3568ROTkFptuEIXYjWLxKAUaNGMWrUKOrq6ti2bRtbtmyhubkZFDx+Lg7FQU7ICcCV3vvJJ5/0SBvcE0VRmPGNFCoOlWNrdXR5XuenYcY32vd3jJkQTcwEzwkUWr3WvcezpeYEUck97GOjOAmJrGTcnXMwhrRf5yROS0Cn0RGaHELSnET34+/96D80H2nxmgaubcyhMrdKAixCDCRf/jjBxU9V47A50Oq1hCT6uTqbolF0jWDjjozHzb8Z/5Rl6ENSey17oPdMuRiS5ydd6ioIMSjU1tby73//m8OHXas8EhISeOCBB7zuVaLRaBgZ6VoCHZE+uscVE02HniPcmQsaPWGZPyEgrfNJ58DQaDTcedOdnPhjrtdlu+MevXQpQbR6rSunbHfnZB1y1dYW1LHnT3vRGrXc8fptfR4gGKgN5DvucXLoxf7tmeJLWrRPnviMgKgApn9jKlEZrnpXHKrgi9/tIDQphGVPL3W/ZNtTn1N9oqb7z/A80qyJvisuLuZPf/oTNpuNqVOnct9993W7Kb3qVJn9+EwqDldia+k0KNnhe98dp62F2h0/xlq5FzQ6wmc/hX/yuRUyAxSYFUIMH9nZ2bzyyis49A5uueUW5syZM2Blq6pK48FnaC142/XAeaQ4FUKIwcbXcan09PRugyuqqlJ/pp7wtHAADIEGMm7J4OR/cl2Tyzrq5fzP2/sGRQWSY81mUsMUj+cUFHJCsgmODuoykU5RFBSj5/lp6qIUUhYm47Q7Kd5V0u3ENxRImpdI/HTf9g9u0+uKE7sJjX8MuqD2gXtFUTyCK+AaW+s4oUj74SEcdj0Tv5ZI/NWrgb5NKIqbHNJrsCE6uYCwmDIsFWVYKnbRkPUHDFGTCZ74TYyxriwV4eHh3HrrrYwZM4YXVv2eQIMdMFGpO0Rc1SQqY44SHezKctFi1fXputveUkHjjq+RPjO1myAVpM/8D407nsdv+Zs+tVtRFI/gVHdZdiYv+Q/x85d7XG9HjY0kamxkl/Iy7o7gdN1hakq8f4aRSQXEpV6YbAoSYBGXHVVVOXjwoDsXeW8uVqoah9XBgRcPcebzQpY/NxdHzU4y5m5mV3kSDlvnZYKqOzLul7Dap+BKm64puLSXRQouIYYqp9PJ559/zvr16zGbzWi1Wm644QZuuOEGdDrvf8btZjtbXvouALdcbYdAr4cSOOGb1BYfIGHhLwmI7ds+TH3hdDhp2WRCj77LcwoKevRYttvgygtWhfa62J0UbDyFf4Q/ibMTAEicm8AVP56H0+Fk5//u9nyBCrMfn4VfmB9N5c1Ej48iICrA42Sv4LNThI8KJ2JUeJf3G4hgiEd1Ou1xYu/jHie5G/I5m32W5PlJvaaHaq5oobmiBVOduf0JRcHaZMXa3CkNlNrp/27K62+aNeG76upqnnnmGVpbWxkzZgzf+ta3uk13mvthHqe3FXLVU4uY/fhMtv9+p+cBHb73nTks9dR+8QS22mwUnT8R8/+IMW5Wl+OEEALg0PuHeemjNTgcDubMmcNNN93U5ZiOK/F62leqM1VVaTz0J1ry1gMQNvOnGGJn4rTUozos1GxxpUAOW/hXdEbXCdFgWUknhBje7GY7r9/6JgB3vXtHv1dudU7d5E1341cOq4ONP9xMdW4NN/7jekITQwCY8fA0ItMj+nT+541Go+Huu+9m1V9WUWWu7JKSqsq/kpUrVvo8oK8oSo8TgAyB+j6OY6nnysX7ihMgNPMnBIy8EUXpvZ66wDj3hCJLi2tCUcLCSRgi+v63xZdgw9g5Wwib9z84moowl36OrTYba/VhOl542ZvLUO0tmOqKeGJBPnpt23OngR0e72lzKFRVnwJ8W83htNSD09pzkMp57rhOf1+dtlacllpUhxV96Ej34wZ/U8+rivxN7Z+C046i8f77k5YYid+8zex+x/tnOG7uFkYk3OFTe/tKAizisqGqKtnZ2bzzzjucOnXKp9dERET0KVWNvaUCp6Ueu6kVY2AD1tYgzmw+QOoCVyocbyfyqt2MRm/k7IlqLA0WTr62lrj4tzD4Q8a8zT51Nn3Rcc+Uad+YIKlahBgEnE4np2oCXCm9cvOZOGU6VVVVvPTSS5w8eRJwLf194IEHSEhIOO/3s9YcxxDpSgKr9Y8i2/AAqRETz7tcb2oL6tjys22YOw7Sd6a69uko2l6EPtBA7OQYj7yvAynnP7kc+OdBQhKDic8cgUarQVEUUheloKoqhZ+f6bLSr23vkOhxUSx7eikOW/vSaEuThT3P78Npd3LD6uUewQNVVTn00kn3fbv5/DaQB9/3OGmuaObkB3moTpXMR6a3v/7zQiqPVBE7OabX9FBRGZFk3DKWyDER7sejxkZyw+rl6AM8TyWvefpqvvztjn6nmxLnr6mpiaeffpr6+noSEhL4zne+g8Fg6HKcpcnCoZePYG22cnrrGdKvH83pLae8fu87cpjrqNnyCPamQjSGUCIWPochsmt6gYHYk04IMfSVHi/lyN+PMYXpVGVW8OCDD3a7oq4/VFWl6cgqWnJfAyA0878JGHmj68nAOJz29uslXdgYDP4hA/K+QghxsezcuZNXXnnF47EYUyzpTRnkhuRQ5VcJeB+/0hq0+IX7oTVoqc2rdQdYAFIWJlOwKc99XdHT+V9vMjMzWfnYSt54+Q3CC9pTUlWmlLPy3pXe9wL1QqevZ8YDCV1T3Ksw44EEdPp6oPtghqo6sTcUYKk6iPXsQSyV7emLXYP6m8nbeyXps7d5jKvpw8f6FFwZaBpjGIZAR8/BhkAHhojx6JIWEzz+6zhaKzGXfoEhun2CZEveG7Tkvs4IXTgarZcZb+fotSqhAX1va29Bqubc18HeisNci9Ncg9Nci+pwjQHowsYQs2ytR3m972PjcnbjAzhaK9D6x6ANiHb97x+Dxj8aXVA8ij7Yp4DNQJ1/dDaoAiypqakUFhZ2efzb3/42q1atwmw2873vfY/XX38di8XCsmXL+Otf/0psbGw3pQnRLi8vj7fffts9SGkwGNx5KV944QWvr1uxYoXPEXZ7SwVVG+4ApxWryR+H9X5UVcOBf53CUPcLV6etMRCz/E3QRZL73h6KdxYy7eZd2GsOE3Pda0z/xlRUp0pw0A5a81KxN53xubPpr8Q5MnNLiEstKyvr3Kb0qa4HjqzC398fi8WC0+nEaDRy++23s2TJEp/7JG8Dm6rTTsOBp2kteIewOb8iIGWZ64k+nGj4kuKqpaoFm8nuHkgPjg/C2mJF0Sjn6uF98L14dymnN59h/FcymP4N31bUdGzj2YO1hF/XadaMw4mt1YYx2HXc6GWjKPi0gDHXjXbVpcPk/q4r/bpfIq/Vt7/I1mojcW4CrVWtHsGDM9sKqT1VS/m+gdtA3uueKcCO/92FqqqkLnLlv7Wb7Zx4Jwe9v44Z35zmPqEceXUacVNjiRwTSeKshB7TQy362YIugXidn67bIIlGo+l3uilx/iwWC8899xwVFRVERkbyve99j8DA7pevGYONLPmfqyjZWUL69aP7tMJVYwhBFzYa1WEmYtGfva6k1fnpWPz1587dvuV8myeEGIJsNhvrXnuVQF0wih88+r1Humy23F+qqtJ07O8057gGHkNn/JDAUbcMSNlCDJSB3u+rvyu9xNBjs9lYt24dn3/+OQBJSUkUFxejdxjIaByPTtWR0TCeOn0dNq2VFStWgAqnt54h96N8Fv9qEfoAV3+b+ch0tEYt/uGe+0cqisLMR6fyn4c2At6ve3yVmZnJ2FFjWXfv62gcGtLuTOH+++7pcwrotvE1f4eVqOQbuoyH+Tc9R9UG1/hax0nM5rLttBS8h/XsIVRbk9fyY9PyBmxMbSC0pRyLNNdR15hP2cEmQEHRQPyMUCZ886ddJmxrA2IJHOO5GkN12EBrRGP3bUuEGPU4TcdLUO0WVIcZh7WFkdbTNB89Tfj077qPq97yLWyN7RPVewpS2c4exNFaQWeK1oii6Trpq7eATRunqQrV2ojd2oi9Id/jOV1wKmFznnK16QKPoXozqAIs+/btw+Fonw167Ngxli5dyh13uL4wTzzxBB999BFvvvkmoaGhPPbYY9x2223s2LHDW5FikLBYLDz66KMALFmyZMBOqntTWFjIO++8w5EjRwDQ6XRcddVVXH/99YSGugaGjEYj69aupb6hwf26iIgIVqxY0acIe9tyOVWFnF1LsNtdab3sNgMndy9m0lUfgdNKQ9bvMZ/N5ejrt2MzB1C2r46YVCuWqgPETXEtlVfV2zBETqB64/0+dzZCiKEpKyuLVatWdXncZHKdpCQlJfFf//VfREX5tndBTyvpnLYmmo6sxlZ7DFBwmmt8rmdfUlzlfVLAnj/tJX5mPIt/tQgAvb+ea/54Nf7hfnz47Y97HHw//tYJjCEGkua2575tKm8m96M80q5MIWJ0BJ2ZGyzu2wdfOE7y3JHu+lUdP8ueP+0lNDmUhT+9AnAtK79h9XKvM1g6rvTLfGRSryv9gmKDWPiTK3A62nMYOx1O9v/jgGdqrQ52/6l9w3ebycaZz4tw2hyMvDbNfcypLacpP1BB0txEkucnoaoqu57d0/Xza3tPu5Mja4+6AywhiSGMvSmdsLQwVIeKonO1d9TSkR6v62t6qJ74hfkNaHmie06nkzpDLRaNhdy8XCZMnsCqVas4deoUgYGBfO973yM8vGu6OofVgdbgCg5Gjo4gssPvk68rXBWNlvDZv8RpbUDrH93tMUIIoaoqa9as4URpNv4J/vzoez92X4MNBFtdDs3ZawAImfY9AkffPmBlCyHEpVRVVcWqVasoKipCURRuuukmbrrpJvbv38/uP+5Fq2pRUNCqWiaZpzD3h7PIzMzE6XByZN0xmkqbyNuQz/ivjAMgKC7I63v19bqnN4rTQpDD9X6ZN07o1/6abeNrPY6HOa005/yboIx70AW69khxtJRjKfvSVQ9dAIaoyRiip6Hxi6Bh32/Pq10Xmi4wDl1gHJkrozsEvPTMfXIxBh9/JmGZPyRkymO05L9N05G/9Hp8a/76Lo9FAtZKi8djTnMNqqXe4zFvQSr/1OVo/SLR+EWg8Ytw31Z0AV6vvXsK2LSJWf4WDtNZHK1VOE1VOExVOFrP4jBVoQ1oX3hxqcZQB1WAJTra8wLt97//PaNGjWLRokU0NDTw4osv8uqrr7J48WIA1qxZw7hx49i9e/eAbpAnhr6ysjLeffddsrKyANeM2gULFnDjjTcSGem5GVJmZibpKcm8/NCbWDQWlvx4PlNmzuz3JstVZ9KpLhrd/oCq4WzhGI5suYHJiz/EUrELBRg5fT/aoGRSltxIUOocdMEp7pcoiuLadFFjAKfVe2cjmzP6TKPz7/a2EJeS0+nk1Vdf7fGYlpYWIiK6BhS649NKOgCtHxHzfotf/BU+lauqKkde7j7FlcPqoHRvGaGpoe7l5jETo0EBp82B0+F0p/lq24yut8H3GQ9NY9rXp7hXu4BrJciJt3OoP13Pkt9e1aV++/9xtNv6ARiCDDSUNGKqN2NpsrhXsfi6PDhpbnzvB53TMaWZ3WxH20OeZZvJ5q6n3WRnz5/2ggJpy1Ldx9ScrOX05jMERgeSPD+JhsIGSveW9ViHxpIm9x4nGp2Gmd+a0Wu9UxYm+5weyhcDXZ7w5Fr1to76iHoAjq86isFgwGq1YjAYeOKJJxgxouumn4VfFnHwX4dZ8tsrCY4P7vE9Oq9wNVfsxlyyjdAZP0RRNChagwRXhBA9+s9//sOuXbvQaDSsfGwlKaN7/hvQMZ2X63bP/ZQhYhyhmT9GtZsJSr9zIKosLnMDvdpEiP44cOAAL7zwAiaTiaCgIB555BEmTnSlco5ujSG8uf3aUIOG8KZwolpc52QarYap90+msaSJ0deN6vN79+W6xxvPFczXnXd5PQ2+t+a/jT4sHd251YvGEfMImfJfGKKnoQ9Pd+/ZYa3NOe96eK2fx6qyris0+up8A14afQDG2Jl4X7/Tzhg3D21gLIrWD0Xnh6oYOHGygAkZnqmsw+f+GlvDaer3/KL3+icswhDh274uHfW2qkhjCEZjCPbYw6Wjjj9jXwI2A21QBVg6slqtrF27lieffBJFUdi/fz82m42rr77afUxGRgbJycns2rXLa4DFYrFgsbRH3hobGwHXUjtfN4m6lNrqOBTq2pOO9R+Iz95isbj3Kog9kcOESdPQaDRUV1fzwQcfsHv3blRVdS13nDmTG2+80Z1Krrv3djichFtdf6RSU1JwOBweq6m6a0vncux2O1aTPyd3Lsa1yZTnhkrVRaOpK09gxJypGFOuI+qWCSgavcfrPRgiibjmVZyWBuymVvb+ZxcG/xZG3vVf6PwDANAYQ1ENkX3+PO0D/PNAq7r/gKK9/rzLs9vaPwubzYbaQ+5IX39H+lLmQNexP+XpbL53z23fHenXLi6LxcLjjz8OwMKFCwkK8j4ryJuTJ09SV9fz8t3a2lqys7MZO3Zsr+XZWqp7X0kHBM/4f2ijZ3f5WXj7mRR9WeweKIf2FFcFW09Rvq+CM1sLSb9pNNMemgpAQKw/N625Hv8IfxxOBw6nZ38aP3cEcZmxVGS58gUrGoWE2fHEzx3hWQfHuX9A+JgwkuYnkjAn3n2M3WJn28++ICA60Gv9kq9IIjA+gHk/mkPslFg0fhqfvnsd+0m7zYbSj+9rQ3kjzeXN3g84t+dMdUE1gTGBxM8agc6ow2p2rRay2WzEZcbiF2EkMsPV1wfEB5AwJ57SPWXdbiTf9lkGxgf0+XdsygMZHumhpj8y9bx+Tz3K8+tbedKveXfgwAFWr17d5XGr1fW9WbJkCcnJyV3q4prReJTmimZyPsxl6gOTu5Th7fzAXLyJpqzfgGpHEzwK/5G3+FRX1d7p90i9eJcel+JvzUCfG7SV0/H/oWq4tAP635bh2K+1NLe4bzc3NxOobU9JuGfPHk6vKSTBmMhV37yS9PR0H87XffvbqzqsKFrXIJYh+YYe69mxH7LZbCi6wf/ZezNcfo+kHedHrkP7b7CfG9jtdt577z0+++wzAEaOHMkjjzxCeHg4NputxzTBu5/bQ+T4CPzC/IifM4J4Rrjfs/c6nv91T0edx/+0PZTX0/haR94G33XRM8AY3f56YwzGUa6Au92hgsP1uFMb6J7A7JXGgFMbeMnH1zqWNyIzul/ldRlf9MJ/3DfQh7ePM9hsNioLNjI5dqHn+waNhD5cx/nyHRron4mv35ne6tffn5+iqur5n/lfAOvXr+fuu++mqKiI+Ph4Xn31VR544AGPYAnArFmzuOqqq/jDH/7QbTm//OUveeqpp7o8/uqrrxIQEHBB6i66stvtfPjhhwDccMMN6HT9v8AuKyvjyJHDmM3t3wWj0Uho6P9n777D46iuh49/Z7arN0uyLVlykSz33nvBjV4csOklQCDU5AfpvQJ5QyBOCM0UFwgQugGDG7hbrrjKlrtsyVZZ9a0z7x+L1pLVpVX1+TwPD9LMnbtnJHl2ds6954Zz/vx5Kv6ku3btSlpaWoOmoXvLyih7w/d10G1gaMLfhs1zFm3l+Qt1/qrRiep+HNPl0ZSpjVv3RHE5KXrV95Qx7C4Durl5tVZV3cUI5zMAbLf8GE1pfpY9kHS3TskrvgxzyN02FFPzF6FqiT7bi9zcXO655x65rrWy5l7X3G43O3fupDT/BMHm2m+ASl1GUgaOIyEhodY2FYK0bAa4XiXnWGoNC7vBgCmfEtfzMPvMd9V5HdJ1HdyABroOpcvLoab7HjNYJ5pxbnFjGmjEMqzh5R9d+S6c//X4+wleYEO1Ne7fpTvTg+PLOm7ImtgvBOa6q+s65SudeI9rNSZDUMCQrBI0u3GjkrRyvc7fSSDO2TZNxdi3meUBmtGfXNdqpus6X3zxBQ5HzWXnAGw2G7NmzapxhpZWpuP+1o15lKnKDLEKNf3dx3rS6eFZiQLkqf05ZroSXTFUO7Ym7f1+I9A6872GaL7OeF1zl7txvO57EGG93YTJ5rsPyM/PJ+PLwwwoGISmaoQuDEINqb86QEPee+M8W4nx7uGQeQEepeY1piq71K5DovML9HtNc/rrjNe1QGvoz7e8vJz09HTy8nwlnHv37s2AARfKa7XU5woI/PMmzV3GGO+zAGwxPIpqatzfhqJ7CdOOkequXr7qYvV9rq3MrBdi1MtR3C5KPvSVdQ65WkU3+d4XPIoNl9L4MpaB/vkFor+K5wL1aczPryX6DOTvJFDxlZWVsXDhQgoLCwkLC2vQa0M7nsHyyiuvMHfuXLp1a970tJ/+9Kc8/vjj/u+LiopITExk1qxZjfpBtRW3282XX37JZZdd1mrrlrQEp9PpfxA5ffr0Jo30Bt+ozQ8++KDG/s+dOwdAv379uOaaa+jZs2e1drVxlxTzvze+AGDOnNmYQmqfjl7T78SVu4vsrz5m08kZdbyKQn5WT2YmpRA9aEiDY/PH96ovvpkzZ9YZX0PonnJyP/J90Jg+fTpmW/v6t+BxeHjvlQ8AmD17Nsa6Suw08N9IY/psK039956VlQUg17VW1tTrWllZGatXr2bNmjWYtEIem3QEU12ztLwKpWnfJ2XQ+Hr7dhcc4NyK5bXOpNv/zRzs2d2Z8LMJ/pEq6f/aQc7uHDyD3cy7dy4mk4n8w/l8+aPV2KKtRKVEUeZxoNdwJ694FKLLYpiwdFyND2vrUlpYyif//QyA4fcOJWVWn3qOqM5Z7GRt5jfYj9lrns3hUYjMiGLCT8c1uu9AXXcdExys+MEXtaw5Y2Leb2dXm/bdkH8jJ2NPsenpLdW2j3t0DD0mJjYp1srnPPf7lzX7vabKz/COqQRFNqzUHch1rTaHDh3iww8/rLNNeXk5ffr08c968zg8Vd/zbqh+jLcs2z9jdvUHmwCYOGwk5K3Bcdw3itLa6zr6DnmUNKWxJVSvAWBOI49qrrZ4r2mJe43O8p7ZWc4D5H6tstLCUj553fdePnPmTILDg8nNzeXPf/4zxeZiktKSmDJtSoPL1NT33lue+S4lu78CYFKaAVvv6oNJLtbeP/c0Rmf5dyTn0TyBfq+p3N/06dOxhTa8nHZnvK4FWkN+XwcOHODll1+muLgYq9XK7bffzogRVUv82o/b+eI/X9X+Qjp4j2lMGDiB8B4NeyBd0/3f1NHjqlRMMQQ1bnAwgKPUTvEXzwK+a7k1OKLWtpV/J6qnEMfxj3Ac+wjFElIx+aROEyZMqDIDoyHcJcX8b7HvvWbWZXU//2twf9+9d02ZMqVRn3nq66+pn0PdBYewr6k/2XDxz6+ufyPesmzyVy6pd8bJpOnzGv13E4jfSaDiq6h81Vjt7wkjvoXJv/rqK/73v//5t8XHx+NyubDb7URERPi35+TkEB9f+w/GYrFgsVTP9plMpg71Zt7R4r2Ypl1Y9Lep56JpGm+//XadbUJDQ/nxj3/c6PVT9ErxGBsYn8lkwl3o5tyW9zCX/BOrCWJ6JJF7sjdVH2p+R9Ho0iOTqORhjT7/psRXF025MFq+Pf5tKd4LPz+TyYTRVP+lqr7zMJlMHaaebmN/JxUzJ9rj77IuHS3eizX2ulZWVsbKlSv58ssvKSsrA2Bwr4g6kysAJoNO76TYav17Ss/gPLMBT/EpPCWn8Zacxl18moOb5uJx+0qDVaWgawbOHhmA0Wj09+e0Oyk5W4ol1ew/D0uw74G/x+Ela3Pt633omk7W5jOUZZcTkdS40T6VzydpYo8m/S2UFpViP2qvM77Tm7IoPVPW6PgCdd01dTHVuubM2IdHE9ql9pvHuv6uek3rycl1J6qtcdJ7Ws01aRsi0O81lfuT61pglJTUUXLuonYmk4ns3Tl886cNTPrZBOKHxNXY1lOaTf7Khf4PI6Ov8m13bK+0PpRiIKzfbRibOSKvLbTm31BT7l8aqqP9W6hNZzkPkOsagMFgoMCcj1N1cuz4MVLSUvjnP/9JcXExPZJ6cOfP7qjxs3htDKE2f9lha+g1qMYLr1t65D1Kdvv2hfS7g9C+32vQWmrt/XNPU8h5tC+tfR6B/lx78XvXpX5dC7S67g00TeOTTz7hgw8+QNd1EhMTeeCBB2p8xhnTJwZLhAWn3VltH1z4LBDTO6ZBcdV2/1eyudL9n2omdt47GIMb97Dc24jPALquE6KdonznH3BmrQXdN3NDMdfxkLySyp9rGyrQn3kqv3fZQi8P6GeopsanBMc0qPyWOTgGYw391/R7M4UnEjvvHTSnHU95GSuf+AaAWU9NqpSUi2j03wsE5pwrx1deaGfNL3yDESf9ZiSh363F3ZD4mvr7a5cJlsWLFxMbG8vll1/u3zZixAhMJhOrVq3i+uuvB3yj+E6ePMm4cY0fmSo6noyMjHrXKiguLiYjI4O0tMYvqFTh1KYz9L6s/gx4zu4c1v16PUHRwYy+3IItcTxp41ax6WwCXreFi0eOG00u+o5dDcgCjEJcSioSKytXrqS83Dc9vFu3blx11VUM7RNG3ld31NtH6aGllLhLCE6Zj7XbBAA8hcco3PFM1XYF0eSerHsmiOY1UXi6nC7fDawZctsgUq9JYcv+C7MhwhJCuemD+agmlW/+uIHTm7PQtRpmsHx3I9/Y5EWghCeFkzg+oUXiU422Gr9uiqTJPTi+5gSnt2RVias5C74risKIewdVWjPFyJgfjmpWnKL9qzzIqCHtMj4+jLPIScanh2tNsGhOe90fvgB0L5qrEL6r5y2EEOnp6SxduhR7lB2AfYu+xWg0YnZYiIiO4NFHH21UcqUupZkfULj9KQBC0m4ldND9DUquCCFEe1VcXMxLL73Et99+C8CkSZO45ZZbMJsvlDHMy8gjPCkco8WIoihMfHI8a361Ds2tVe1MAVNQ4z4LNOj+T3P52jXwgbmnNBvNaUfxXkgCKWUncDl9n1cufrjtOLuRwt3/pJ8rE+dp3zZzzGCC+tyAMbg7uavubvD5iKqMwfE1Jhum/WEMtvAIoGnJEGNwvO/voaSYkvxDvm3hKZibOQsoUPzxmYspyT8GgC2mH+Zmzipq0Gu3+Cs0kqZpLF68mNtvv71KPfvw8HDuvvtuHn/8caKioggLC+Ohhx5i3LhxtS5wL9oP3VPOH+fs938NDS8R5nK52Lp1a70lMSrY7fZGx+covPAGkP6fPXQflVStZIuz2IWjoAiKPibRvYPotFmYgkwEx4URPvltrGEKjjPzSRu/uoa1DxT6jluFOdiLaolodHxCiI6ntLTUP2OlcmLl6quvZuTIkaiqiiv/YIP6cpz6EgBzzCB/gsUY3hNr98kYQhIoLexOcNfuRAR7OLpzbe1rQX03ky48YZh/U2TPSELcbtTjFx5UqAYV1eA7fsxDo8jenVNLiaumP9Q3Wo2siveVHppvva5JfSiKciG+MnfVMmHNjC+QFEVh5P3D/QmWQCVDrOEXHlyNvG9Qtfct0fmkpqYSGRlZ54CTqKgoUlNTAZjwxDgi3o1gwA39WivES5rRauwws2WFaI709HQWLVpUbbvqNDAybzQ2kxWb0vDBCRUP5fRKD+XcBRkoBguOM99Qsu9lAIJTFxA6+EFJrgghOgxN0/wz/Q5lHKLfwH4cO3aMf/3rX+Tn52Mymbj11luZNGlSlePSX9jOwQ8zGH73UPp/dx/XdWg84380tsaZ8WMeGt2mnwU8pdmcWzG/WtImb/W9F75RzcTO/S/GEN+AHd3rxluYiRcjwclzCU39HqbIVH9/DZmB0R6er6lGG12uW8+KFSuY18yBeYFUkWxw6bn+ZIMhrA/mqIbNchKN0+4SLF999RUnT57krrvuqrbv73//O6qqcv311+N0Opk9ezb/+te/2iBK0RrOnz/PmjVr+PrrryktLW3wcQ0d3VlB13W2v/it/3tPuZct/9zGlF9ceIM7sf4kG5/aQHhcNkNnvkUcCriyuOrlK6o84Iqd9w7RjgIKio5wZmcxoKCo0G1EOAPu/UWTp8sJIdofr6uUu0cdp9hpJOPgXoYOH42qqjUmVrp3787VV1/NiBEjGl3CEMCWPA9zzBDMMYP924zB3Yia+DT7/rufna/tJmWukWE3R5I2bhWbzybgCdBMOmuEtdYSV219Iw/tP74KleMY9YMRAYmr8syapEkNq28vOjZVVbnpppv497//XWub7139Pf91xmA2MHjhwNYKTwhxCdA0jWXLLpSQiS2PI7U4jYywg3gUDwoKhQVFGIMb9qihQQ/lAFvy5YQNfaTRyZX2+uBLCNH5VZvp9/dvCQoKwuFwoGkacXFxPPjggyQmVl8/MbJ3JChQlltWZXvS5B5kfnWYs+nnAVBUSBiX0KyZ8Q3hdRSge8owBMWhqNWv7w2dEVNy8HUiRv4EAGu3CYQMeZyvD+rMHn59ldJMlWdgBLocVXsXyEoKAOZKs0nNAZpZKqprdwmWWbNmoes116O3Wq0sWrSoxtEyonPQNI19+/axatUq9uzZ4/9biI6OZurUqaxatarOGSqVR202hKc0m+Nrj/lLrICvZv+pDac5+N56+szpg7c0C+Xsi3jdU3CVGtCNXTmmjKRLSCLmi2qRG4PjMQbHM/LBLnx0j2/EudFmYtzj0zG3k4d8Qojm890sL8FuT/Zt2PMfIiLepk+fPuzbty9giZUKwSk3Yo6qufRhl/5dQAd3mRvFFI452EvfAM+kS5rcg2Orj1Zb76Olb+QbqiVKcLWkHhOatgh9R1P5A4EiD7UCJi8vD/DNjOpSFnvhwWY3N9dOvY4T/zxNZHEUqZentHGkQojOqHLZZpPXTFpRf4y6kbTC/myK2cDmmI2ousqRzIkNKtvcoIdyQHDKfJm5IoToMCrP9KuciD5HDgC9e/fmRz/6ETabDVepi/3vHiR+aJy/pGvP6clEp0ZXK3WsKApD7+zP2fR1ABisDZ8Zr7l9a/mppoZXlKlQfuJzinY9C4oBQ1AshuBuGIO7YwjphiG4G6q1YSWYyk+vJXzEEyiKiqIasfW+Du+hFTW2be/lqISorN0lWETnpGkaR/OCKHYaiTuSydDh0VUeNpaWlrJ+/XpWr17NuXPn/NsHDBjAjBkzGDJkCKqqEh8fX2eCbcGCBQ1+iOkpzeb0e7eR/u4CoPpI7+2vHsNa/EvMtjKMwJjrzxM37iqsvZeQ/sUqFKX212nvZVva+0guKbEh2rPaymLY7XbS09MBSEhI4Oqrr2b48OG1XpN0zUtJxluNem2v28uRzzIxh5rpOS0ZgNiBXbjypcsJTwjzfd8CM+kURWHUD0aRtfVjoP2t99ESJbjkOiTao7y8PD744AMAbrziRnJeykVBYYhjGNf8/EoOf5qJq/gkx1Yfp8+c3v5Sf0IIESj+wW46pBX1w6AbUFAw6AbSivrxbeTuqu0CRpIrQoiOofJMv4sT0QWmAtwGFwUFBf51qva+tZ/97x7gzPazzH12FoqqoBrUasmVinKKRsWOJbgQV1kIA2+MQdWO48qveUaH5izEkbWO8tNrceZsJWzwA4T0Xdjoc9I9Zf6SXd7Ss3hLz+Jiu39/xJjfNKifiFE/r/NZmgg8o9XI9Due/e7ra9o0ltoEetZOW5AEi2hxNY30joz8LwsXLiQ2NpZVq1axefNmXC7fyCWbzcbEiROZPn068fFV3xxGjhzJgw8+yNIlS7AXFvq3R0VFsWDBAkaOHNnguLyOAg5umITHY6b6DbuCrhk4uGkag6d/RlCfa4kbcA8GaxRut7um7mqVOK5bo9rXpCNcEIW4FFxcFqMmISEh/PrXv66yjli1fjzl2Df/CkfW11W25xxL4fDWqaSOWUts8uFqxx396hjb/r0da6SVxHEJGK2+16hIrkDLzaRriRJXgVQ5nuH3Dm138XUE5pDQgCaVjFYjN350AytWrPD/rYrmWbZsGS6Xi9SUVIy7zKj4PqDqLp1t/9rO5J9PxBphoef0ZEmuCCFaREU55lhHHLHOOP92FZVYZxyx5XGcs+U0umyzEKJtyP1a4Pln+tWWiI7YTWFuIRkZGaSlpdH/+jRyvj3HgPn9as0lX1xOccL873bokPvld1+rZmLnvYOiGnGcXkt51lpc53aA7vX3484/0KRzCh1wNyH970Rz5OEpycJbegZv6Rk8Jb7/G2xdGtRPQ9u1pM7wMF+0P3L1FC2qtpHeBQUF1bYnJiYyffp0xo0b58/k12TkyJGkJvXg9Xvewak6mfGTCQwZNarR5XeKTjvIPdmnjhYKuSdTUJPHEjFiVqP6FkJ0TpXLYtSmpKSEI0eO1FoWw1t2jvz1P8ZdcAiU796GdQ+uchuHNs7A47ZwcOMMIuJOY7aV49Vs/pJevWb25OhXx3wPT011X/NaciZdey9xlTCue1uHUI3MiBHNtWvXLnbs2IHBYGB2vznsf/Ggf19FedOT35yi75UNL5UK4MrdHehQhRCdWGpqKl1Cu5CW0x8dHaXS00AdnbSi/hCnN6pssxBCdCYVM/hqS0SPPz+JfEuev501wsrcZ+t+5tTQNU608nPkrf0hutfp32yMSMGWMA1rwjSMYT2bckoAKIqKwdbFlyTpMrTKPlf+wZoPCoBADziWz2XNJ0mq6iTBIlpMQ0Z6A4wePZoZM2aQkpJSb13diimReomdeJMBV1ks1tMleFIygLoXudK8Guf2nqdL/xhU1UtQ+DlCIs9RUhBb84spGl16ZBKVMqzecxBCXBoaWu6itnbugkPkffMjtPLzqJZIoiY+hWqLxeso4JtnMvC4SwEFr8fKkQM/xmBWcTsUEm7y3ZgbTAZmPTOz0TXIAzGTTgjRdpxOJ0uWLAHgskmXcXhpZvVGCmx5fitxg2MbnFDVXCUU73+t/oaqudFrNwkhOidFURhvnEiRXlwluQL4R2iPM05o1tpzQgjRVjwOD29d+w4AN70/v0mzeiIiIvylwWpKRNs0G13KYwm1tsBaIqoZS/wYvOV52BK/S6qEJFRvZonwl/yqqy+5/xMdkTkk1D8zz9RKa/ZIgkW0mIaM9AaYNm1ag0Y4VZ4S6Sq34XXdjq6r7H7jJLbi32O2lfunRFYkWXRdw1t6FsVoZeWT6eRl5DP+4Tisxb8A3cuAqRGkf7wQr8fExWuwGE0u+o5dDXyvaT8AIUSnY7U27KFlTWUxvGXnyF19H7qnHGNYMlGT/o4xxJf4OJ3uJHtXGRXXIV2D7D3FKAYFdLAfLySyp6/PhiZXpLSgEJ3HRx99RF5eHtFR0cQej+dM+dnqjXRwl3nY8s9tTPnFpAb1W7TrH+jOfNSgeCLH/h7QyVt9LwARk/+F0RIM1D2ARQhxaSk8UUjJgVJ/icKLqaiU7C/FfqKw2voBQghxKYiKiqJfcX9/abDKFBR0dMpCSuk3uF+LvH7k+D+jqHU/7jUGxxM77x3fAGav03//Fz39RRSDrxKC3P+J2gR6FlBnmFUkCRbRYpo70vtiFVMidR0ObprhXzvF4zZzaPN0Bk37FDQXpYffpviMk5x9JXTr9TW6p5zQQT+gy4DBFJ8pxu0Iwap7wRhMcLidtAlfsW/dvIteTaHvuFW+pI0QosNzOp3cf//9ALzwwgt1liGsTWlpqX9x6bpERUXVmDQ2BMUSnHoT7rx9RI7/E6rZN5LCYXew5flt1TtSQDWqTP/9VH9ypTOxWCwsXry4rcPo0DrDjaioX1ZWFl988QUA1828nqPPHa+1bUWpsIY82NS9brzlOYBC5NjfYOkyGM1z4b7HGJGC2RZWewdCiEtSeFI43cd24/TmrGoPDgEUVSFhXHdJrgghLkl2u51//nERqY6ay0WDL8kSURJJ0aniFrlW1pdcqWAMjofg+Cr3f6bI1CaVfJIZMeJS16QEy9y5c7n11lu59tprsdmk1pqoWUMXNmzsAojnjqdWXTtFVzl/IoWcYynE9TyMfc97rH/rXnTdQFjYVoKjvOieUgYtGMiwO4agmhS08o/xOvLI/fIOYpMzyDmWSu6pXqCr/tJgcT2rLzAthLg0lZSU8Mwzz3DixAmsVisOh6PWtgsWLPCXxdA1D7q7FNXiu3EOHXgv6Jr/plfXdbY8vw1Puad6Rzpobo2DHx4iblAtpQyFEJ2aruu8+eabeL1ehg4dyrg5Y3Gnezi9OQtd06u1b8yDTcVgImryP3Dn7cUcM6glwhdCdEKKoqCOBX2z7xpUJcmigCnIyJgfjmpwf76HcibQ3HU0kodyQoj2r6ioiKeffpozhWeIDYsjojgSqt+ugQKJ4xM6VSJaZsQEjgyi65ialGA5evQot9xyCyEhIVx77bXceuutzJgxo9E14UXnlpqaWu+DyNpGetfGtwj0dHzvUlVLeh3aNIPI+NOE9OhHXH8NXbERPu5vdBnYt1oG3xAUi9eRD4CiQNq4VWw+m4DHbalUGqz9UI02ut24pa3DEOKSVFJSwtNPP83JkycJCQnhiSeeICcnh6VLl2C3F/rbRUVFsWDBAkaOHAmA5iomf+PP0N2lxEz7F4rRiqKooFwoqVF4opBTG0/X+tqNGY1+MbluCNHxbdy4kUOHDmE2m7n55ptRFIXYgV04vSWreuMmPNhUFEWSK0KIRtu4ZhOJJFXfocOYh0Y3eB0o8D2UCxv0IEW7n8MY0RuP3TfITR7KCSE6korPjGfOnCEyKpJrHryaDU9uwl16UfJYAXOwqVH3ax1FIGfECNHRNCnBcujQIbZt28aSJUv473//y5IlS4iPj2fhwoXcfPPNDB06NMBhio7o+PHjOJ3OOttUHuldH13Xq5QGq+pCqbBpV32PGVP7oqh1J/wqT2E028rpO34Vh7dOJXXM2gulwZo4WqryG4i8mQjRcRUXF/P0009z6tQpQkNDeeKJJ0hISCAxMZH+fXux483rKXYaiZv0e4YOH+2/nnlKssj/5nE8RcdRjDbchZmYowdU6z88KZzE8QkBGY0uhOhcSkpKePvttwG46qqriImJwVXq5ttl+9C9NQyHbOCDTWfONspPfknYkIdRzSEtEboQohM7evQoGdmHKIoqYpRpNGU5vs9NFfcsSZN7NLrPkLQFWLqNQ/OUkfflnYA8lBNCdBxlZWU8/fTTnD59mpjgGO666i4SendnzEOjWP+XjVUbNyERfamTgYOiI2jYk+0ajBo1in/84x9kZWWxYsUKpk+fzn/+8x9GjBjBwIEDeeqppzh9uvZRuaJzczqdvPjii+i6TkpKChERVR8ORkVF8eCDD/pHetdH13XObVvlKw2m1/Jn+12psMLT5fUmV+DCFMaYy14nYuK/OfHtKMy2UlJveZyYy14n5rLXiZ33joyWEuISVTHF+9SpU4SFhfHkk0+SkJDg36+qKr2iyxjSrYiUPr39yRXn+d3kfnUXnqLjqLZYYqa/WGNyRfNqnN+fy5iHRmG01TDeoQmj0YUQncd7771HcXEx3bp1Y/bs2YBvxONlf51O6lUpdB/Tzd9WURUSJyTU+2BTc5dh3/ZHyo5+SMnB11s0fiFE57RmzRpQoOeUZGb9ZaZ/u9HWvHsWU1gyprCegQhRCCFaTXl5OX/72984efIkUbZoxhSNZ9ez33L+QC5Jk3uQMKa7v21D79dq4h8gXGcjKacoRFtp9iL3qqoye/ZsZs+ejd1u57777uOdd97hJz/5CT/72c+YOnUqjz32GJdffnkg4hUdxFtvvUVOTg6RkZE88sgjKJqz1pHe9dHcZRSm/wnV/iUxPa4g92Rvqs9gwb92SnjCsAbHWTGFkZJiSvIP+baFp2AOCW1wHzX2KzUThejQioqKeOqpp8jKyiI8PJwnnniCbt18DzM9pdloTjseR7G/vbfoCC4tD8fZjZTsewV0D6bINKImPYPB1qVa/7qms+W5bRz96hjjHhvTIUY3yXVNiNaTmZnJunXrALj11lsxGi/cskf2imT0D0ZSkl1C1pYzQMMfbBbtWYS39CyGoK6E9LujRWIXQnRehQWFbN28FYDp06dXuUcZ9YMRjb5ncRccAtWEKbxXQOMUQojW4HA4+Pvf/87Ro0cJDg7m4R89xJFXjlJ8pgRrhBVFURh5/3B/ademJKI1ZyGu3D1Yu0/yr3HiKS9j5RPfADD9T+OwhoYBUk5RiLbU7AQLwPr161myZAnvvvsu+fn5DBw4kNtuuw2TycSrr77KVVddxc9//nN+97vfBeLlRDu3a9cu1q5dC8A999xDcHAwjlI3vaLLAAitNNK7Prqukbf2h7jz96EoKkazA19ypfoaLBfWTvleAM9GCHGpKSws5OmnnyYrK4uIiAieeOIJunbtCviSK+dWzAfNVeWYsk2PUlbpe0v8WCIn/KXO0haq0XcdNJgN9JiUyLHVR8namg00r8yGEKJj83q9vPHGG+i6zvjx4+mb2pftL+8kdV4fQrtdGADS2AebznPbKTvyLgARo36GagpumRMQQnRaq55bw/Czoyjok0fv3r3xOr3+fT0mJDaqL13Xsaf/BXfBQSJG/5qg5DmBDlcIIVqM0+nkH//4B4cPH8Zms/HjH/+YpOQkuv+iO+4yN7Yo3+fA5iSidc1LwaZf4MzZStjQRwnpu6D6AOGwPpgjowJ7ckKIRmtygmX//v0sWbKE5cuXc/LkSWJjY7n99tu59dZbq6zB8sgjj3DvvfeyaNEiSbBcAoqKili8eDEAs2fPpn///gB4HB5Wv/YoAJdP8dZ2eDWKohKcMp+i3TmEDf8RPezPkXeqF25n0MUt6TtuFeZgr0yJFEI0WWFhIU899RRnzpwhIiKCJ598kvj4C6OANKe9WnKlJqGD7q8zuaKoCqN/OJLes3oR0zcagFE/GEXW1o+B5pfZuBQZrUZu/OgGVqxYgdEakPEjQrSJVatWcfLkSYKDg7nxxhvZ/+4BDrx3kONrT3D1y1f4/74rLyCaMCamzj41Tzn2rX8EIKjXNVjiR7fcCQghOiWPy4N9u51gbzA90hJQlPpLMtfFcXo17vz9KMYgLPFyzyOE6DjcbjfPP/88Bw8epKvWlctHXkFycjLg+0xS22eRxiaii/f+B2fOVhSDFUucXCeFaM+a9ARi6NChfPvtt1gsFq6++mr+9a9/MXv27FpnJUybNo2XX365WYGK9k/XdRYvXkxRUREJCQlcf/31TetH8+AtO4cxxFeOJyh5Ltbuk1BNIfS6pT8JV+ex+bmjnNlZDCgoKnQbEc6Ae38hUyKFEE1mt9t56qmnOHv2LJGRkTzxxBNVkiuNU/2hg67rZG05Q/cx3VAUBUVR/MkVgJD4ECb9dALbXtjepDIbQoiOr6CggPfffx+A66+/nrCwMHrOSObkxtP0vSKlycnD4j3/xluahSEojrChDwUyZCHEJeLAoQNsjNxAD08SN9zx42b1pWseivb8C4CQvjdjsEbXc4QQQrQPHo+HRYsWsW/fPiKIZEDuII69c4JeQ3vSdXjXau0rD4jxfd2wcvTlp9dQcsC3Xl7E6J9jiugTkPhrEuhF5GVRenEpatKntIiICF588UXmz59PWFhYve2vvvpqjh071pSXEh3IunXr2LVrF0ajkXvvvReTydToPryOAgo2/QJP8Qm6zHodgzWa7F3ZBHUJJqy7b80UY3A8Ix+M46N7vgTAaDMx7vHpmOVhpBCiiQoKCnjqqafIzs4mMjKSJ598kri4uIC+xo5XdnHgvYP0uz6NEffUvFZU0uQeUhZMiE7K4/Dw1rXvAHDT+/NrTJYsX74ch8NBr169mDJlCgBB0UHM/ttMVEPDyqteTHOVUH5yJQDhI3+Gagpp4hkIIS5lq1evxqt66HN5L6zW5n3uKst8H2/JaVRLJMF9L6zvJg/lhBDtmdfr5YUXXmD37t2YTCbueeRuytc7cRQ6iBscuM+O7qLj2Lf4KgAFpy7A1mNWwPoWQrSMJiVYKtbXaKigoCCSkpKa8lKig8jOzmb58uUA3HDDDSQmNm7qI4Arbx/5G36CVn4OxWjDU3iUwiyVtb/7BoNJZdbTMwnvEQ6ANdziP27kfYNkpLcQoskKCgr461//Sk5ODtHR0TzxxBPExsYG/HVC4kNAocoaCkIIUWHv3r1s27YNRVG46eqbyDuYR5f+XQCanFwBUM0hdJmzHEfWOqxdxwYqXCHEJULXdY6mH2X37t2ArzpFc2juUor3vQJA6IB7ZD0oIUSHoGkaL730EtvTt2M0GnnooYcYMGAAej8d8JWADsjruEsoWP8kuqcMc5fhhA35YUD6FZcuGbzQOpqUYNmxYwebN2/mgQceqHH/v/71L8aPH19lLRbRcTidTu6//34AXnjhBSwWS53tPR4PL730Ei6Xi379+nHZZZc16vV0Xacs830Kd/4/0NwYQnsQNeGvmMJ7oZnLCUsIwxxsIqRrzSMuE8d1a9TrCSFEhfz8fP76179y7tw5oqOjefLJJ+nSpUuLvFbfK1KIGxRLRFJ4i/QvhOi43G43b775JgAzps7kyKvHKMgsYOKT4+kxsfGDVi5msEYS3PuaZvcjhLj0nFh3ii3/byt9bWmo4xS6dq1eAqcxSg4tQ3MWYAhJIEiuS0KIDkDTNF599VW2bN5CWkk/BvYbyMCBA4HAJVYqOE5/jaf4OKqtC5Hj/4CiytqSQnQETfqX+vOf/xybzVZrgmX16tWsWLGCTz75pFnBiY7hk08+4ejRowQFBXHPPffUuhZPTXSPA/v2pyg//ikA1u5TiRjzS3/5ClukjVlPzUDzahhMhhaJXwhxacrLy+Opp57i3LlzxMTE8OSTTxITU/dC0Y11fO0JEscnYDD7rl+SXBFC1OTTTz/l3LlzREREcNWVV7Lz5G6Ks4qJ7BXR5D5deXvxlmZj6zGzwceoRhtdrlvPihUrmGe0Nfm1hRCdh/14ATo65QYHV0yb1+z+DJZIFHMYYYMfkAeHQoh2R9M0Csz5OFUnhzIOkTYgjTfffJMNGzYQ4Y2ge1kiBdsLyT2Q659pHEhBPeeBasAY3K3W9amMViPT73j2u68vD3gMQojGa9Idzfbt2/npT39a6/5Jkybx5z//uclBiY4jMzOTjz/+GIDbbruNqKioKvs9pdloTjveIjshUecA0IozcZntAJQd/dCXXFFUQgc9QEjaLWhujdxDef7Fn5u6oKsQQlTQNM3/9aFDh4iPj+eZZ57h/PnzdOnShSeeeKJByRXnufQGv+a+dw+w85VddBvZlam/mdysEj9CiM4rOzubTz/1DTRZsGABoZGhTPrZBEqyS5pcUlD3OrFv+T2e4uNozgKCU+YHMmQhxCXE2c/B1ujNmGNMDBtW8xpyjRGccgO2pDkoUhpMCNHOpKens3TpUsw2C6nFabzx5zcpDLfjdDpRFIUbH7yRmMIuGCyGFkmuVAhKmt1ifQshWkaTnlwXFxdjNNZ+qKqqFBYWNjko0TE4HA5efPFFNE1j7NixjBkzpsp+T2k251bMB80FwOirfNudO5bhrGikmjFF9iNsyA+xxI1E13Q2/r/NnNpwmvE/GkvyVFm7RwjRPBU3yhX+/ve/o6oqmqbRpUsXnnzySaKjax4dVJmn+CTFe1+s/wVVM6olgug+YLAYiB0UK8kVIUSNdF1nyZIleDweBvUZxKhRowBfuYnmrNdUvPdlX3kJaww2+ZAuhGiGtWvXUmIq5trp12IwBKaigGquufSzEEK0lfT0dBYtWoTJa2Zw0TCMupG0wv5sM27GoBqZetkUxo5tmbXsPKVnKNr5d8JH/ASDrf7PpUKI9qdJCZaUlBRWrlzJQw89VOP+zz//nF69ejUrMNH+LV++3L9uwa233lptv+a0+5MrtdJchI98EnNUP9+3Xg103wMHS3jda78EkjkklFs+W9BqryeEaB0VN8oXq5jRMm/evAYlVwAMwd0I6nkl7oJDhA97DI/DyconvgFg+p/GYQ0NA0C1RGAMjid+KFz14uUEx8oITSGEb427yl8brUa2bt3Kvn37CNPD6ZqewO7X9zDktsHNquftyttHyaElAESMfBLVHNbs2IUQl55jq09QeK6QzMxMDAYDkydPblZ/RbsXYY4ZhKXbJBQlsGsWCCFEc2iaxrJly0CHtKJ+GHQDCgpG3cjo3LEUm0rYtX0XCxcubFRJ/IbQPQ4K1v8Et/0Q6DpRk54JaP9CiNbRpCvD3Xffzaeffsrjjz+O3W73b7fb7Tz22GN8/vnn3H333YGKUbRDO3bs4Ouvv0ZRFO655x6CgoKa0duFG2yDycDEJ8cz5/9dRtdh8c0PVAhxyfLfKNfh448/rlI+rC6KaiRixP8RM+1fmKMHYAxPoSQ/lpL8WLL3h2COSiNrlxW368I6K5JcEULUpqysjOXLlwMwsddEvA4v5w/komt6k/vUvS7sW38PuoYtaQ7W7s17ICqEuDSdP5DL1me3oX+kYtSMDB8+nIiIiCb358rbS8nBN8jf8CTeklOBC1QIIQIgIyODgoICYh1xxDrjUL97VKqgYMREiCeYstwyMjIyAvq6uq5j3/5X3PZDqJYIwkf8X0D7F0K0nibNYHn44YfZtWsXzz77LM899xzdunUD4MyZM2iaxq233spjjz0W0EBF+1FYWMjixYsBmDNnDmlpac3us/Rcqf9BpKIqRKfKtEghRPNU3CjXJT8/n4yMjDqvY44z67HEj/UvxKoYzL7thRdGo+98eR+ay0D6CzsISwxjzt8vwxxsDsBZCCE6q/fff5/CwkLi4uK45qdXc2bzWeKHxKEamz4ysnj/q3iKjqFaowgf9ngAoxVCXEoUBcJ6hHIo9xAe1cP06dNrbKd5yi/6unppQ13XKdr9TwBsyfMwhvZokZiFEKKp7HY7Jq+ZtKL+6OgolQYBV3zvUbxVBpgHQtmR9yg/vgIUlchxf8QQFBfQ/oUQradJCRZFUVi8eDG33XYb7733HkePHgXg6quv5vrrr2fq1KmBjFG0I7qu8+qrr1JSUkJiYiLXXntts/vMO1LK2j99Sr9r0xhy2yCZMi6ECIiG3gDX1a785FcUbPo55tgRRE/+B4rBBPiuhdtf/NbfzuPwkrX1DEFdgug+uhumIFNzQhdCdEKaplFiKMHmtbHq9dV8teMrAG699VZMJhNJk5r30NFblkPJgTcACB/xJKolvJ4jhBCiZjFpMZivMXLorQN069aNvn37Nrkv59kNuM7vBIOFsIH3BjBKIYQIjPDw8CqlwSpTUFB1A2lF/Zo0k89oNTL9jme/+/oa/3ZX7h4Kd/0dgLDBD2KJG9nU8IUQ7UCTEiwVpk2bxrRp0wIVi+gA1qxZw549ezAajdx3332YTM1/iHj+UAlep5f8I/nomo5ikASLEKL5GnoDXFs7t/2wr9QOYIpM8ydXAE58fZKsrdn+73VN5+yObMY8Mpo+s3tJolgIUUV6ejpvv/42/bVBqKjkfV5AUlAyXW3d6Nu7cQ8uPaXZaE47utfp/8Cul/fHa7AQPuxHuAoOYUuYGviTEEJcMnRdZ93X6/CqXqZOndrk+xpd81K027cWXkjK92R0thCiXYqz+EqD1UZFJdYZR6wlMNcwb3ke+Rt+CpoHa+IMgvveHJB+hRBtp1kJFnFpOXv2LG+//TYA8+fPp3v37nUfUN8C999JuzyOiJ696DoiHtXQsLIYtY0CEEKICqmpqURGRtZZJiwqKorU1NRq2zVnIfnrn0D3OrDEjSZs8AP+fQ67gy3Pb6vemQI7X9lJ4tjuWCOsATkHIUTHl56ezqJ/LmKQfUiVRVN7l6aglCqsfWMdM++d0aC+PKXZnFsxv9o9Vt7qSqPCVTOhA+7CGCxr2QkhGufghxmYQ0044srJzs7GaDQyZsyYJvdXfnwFnqKjKOYwQvrdFsBIhRAicCJ7RhLSL5iiA8X+9Vcq09AI6x9KVM/IgLye7ilFNYWgmkOJGPULGZwnRCfQ5ATLnj17eP7559mxYweFhYXVFglWFIXMzMxmByjaB4/Hw4svvojL5aJ///7MnDmz/oPUqusP5BxL4fDWqaSOWUt0wlFUVUNRfQu59piY2BJhCyEuYaqqsnDhQhYtWlRrmwULFqCqVW+idc1DwaZf4C09gyG4O5Hj/uBff0XXdbY8vw1Puad6Zzq4yzxs+ec2pvxiUkDPRQjRMWmaxrJly/yLplaoKD+Raz7PzoPpTNemVbsW1dif017/ABbN5WsnCRYhRCOUni9l5+JdeJ1eSkcXA5CYmIjNZmtSf7rXSdHeFwEI7Xc7qjksYLEKIUQgKYrCnF/N4r+3vofuqb4Gi9FqZM4vZzWqz8ozjiu4CzJQDBYAIif8CcUYhGoKCsxJCCHaVJMSLGvXrmXOnDlERkYycuRIdu7cyfTp03E4HGzatIkBAwYwYsSIQMcq2tBHH33E8ePHCQ4O5p577mnQQwDVEuFLsmguXOU2Dm2cgcdt4eDGGYR36Y9q9NB/ympfOyGEaAEjRowgJiaG3NzcKtujoqJYsGABI0dWr3Vb9O2/ceZsRTFYiZr4VJV1DApPFHJq4+laX0/XdE5tOI39RCERSbL+gRCXuoyMDEpySxlXNLTGRVPD3RHsz91HRkYGaWlpbRipEOJSZw23MmjBALJ2nWXNKd8aUcnJyU3vUDUTMfJJSg+/S3DK/MAEKYQQLcQQbCAz+jApOVVLtyooTHhsXKMqFDR0xnHsvHeaFKtqtNHluvWsWLGCecamJcGFEIHVpATLr371K3r16sXmzZtxuVzExsbys5/9jOnTp7Nlyxbmzp3LX//610DHKlqJ7innj3P2+78+fPIkn3zyCQC33347kZG1T4vUPA6Kdv2D0IHfxxgcT+y8d/A6CtjwtyN43MWAgsdtJS+rJ6pRwdL/JilhIYRoMZmZmeTm5mIwGLhp8DHcXpXuU//MwCHDa0wUe8vPU3bkfwBEjPkVpog+VfaHJ4WTOD6B05uz0DW92vGKqpAwrrskV4QQABQUFNS5aKrhu0VT7XZ72wQohLikOJ1O7r//fgBeeOEFLBaLf5/BbGDgjQM4bM5A+0gjJSWF8PCm388oioK120Ss3SY2O24hhGhpW1Zv5SQniAnpQkRJJAqK/7Nd0uQejepLZhwLcelp2IIXF9mxYwd33303YWFhGAwGALxeLwBjxozhvvvu45e//GXgohStStM0juYFsftMGHv37ufFF19E13XGjx/PqFGjaj1O97oo2PAkZZn/I3/9/6HrOsbgeM7uDeLMzhKoeLDw3TPJvlemEjuk+toHQggRKKtWrQJg1Mjh9I8rYUi3IlJT+9Q6C89g60LMzFcIG/ootsTqayIc+SyTriPiMdpqGJ+ggCnIyJgf1n6dFEJcWqwOK7HOuBrrecOFRVMt5ZYa9wshRCBVDKT745z96J5y3zb9woARj8fD119/DcDUqVOb/jped7PiFEKI1lR8toRj/zjBIPsQ4ufG+gfFGG3y2U4I0TBNSrAYjUZCQ0MBiIiIwGQyce7cOf/+Xr16sX///sBEKFpVeno6P//V73hlWzL/3ZPAohdeIjc3l5CQEG6++eZaj/OtWfBLnNmbUQxWwoc+gqIodS4GnfnlURx2RwuejRDiUlZUVER6ejoAUyY3fPSkKbwXIX0XVNt+dmc2W57fxrZF2xnwvf7VD9RhzEOjZYF7IYTf4MmDKQjNR0Orcb+GRkFoAUOmDGnlyIQQwuf4mhOs/uVaCk8WsnPnTux2O2FhYQwbNqxJ/XlKz5LzydWUHHwTXfMGOFohhAi8w+uPgA4GDEyePdm/fdQPRshnOyFEgzQpwdKnTx8OHz4M+Kb+pqWl8f777/v3f/rpp8THyzS3jiY9PZ1FixZhtxdW21dSUlJr0kzXNexb/4Ajay2oZqImPYM5ZnCDF4MWQoiW8PXXX+PxeOjZsyfJSbVP665Y1N55fmed/cUPiaPXjGTSru1L/xvS6D76wvucoiokTkho9PRxIUTnZjAYSFnYG6/iRadqWUEdHa/iZdQDNZcsFEKIlqZ5NXa/+S1n0s9yauNpVq9eDcCUKVMwGuuvJm60Gpl+x7NMv+NZjFZf++Jv/4PmyMNxdhMocm0TQrR/R5UjbI7ZiD7cS3R0tH97jwmJbRiVEKIjadIdz7x581i+fDkej+/B+eOPP87//vc/UlJSSElJ4aOPPuK+++4LaKCiZWmaxrJly+pss3z5cjSt6ghMXdcp3P405Sc+A8VA1Pg/YYnzTaGsWAy6pnUKoOpi0EIIEUher5c1a9YAMH369DrbFu3+J+UnvyR//RNo7pJa2ymqwrjHxzL87qGoqsqIewf59xltBpk+LoSoRtM0duzfwcGw/TWuwZI0P5GxU8e2UXRCiEuNx+Fh9WuPsvq1R/E4PKgGlel/mEqfOb0JHxfOwYMHURSFKVOmNKl/d0EG5Sc+ByBsyEMoilLPEUII0bY8Hg/r16+nzFjKxHmyZpQQommalGD55S9/ye7du/3rr9x+++288cYbDBw4kCFDhvDqq6/y5JNPBjRQ0bIyMjIoKCios01+fj4ZGRlVtpUceJ2yzP8BCpFjf4O1+yT/vorFoBW15hvrihHfTVkMWjXa6HbjFrrduAXVaGv08UKIzm337t3k5+cTEhLCmDFjam1XdnwFpRnLAYgY9TNUU0iV/Xvf2seuN/b4v1dUxf+wICg6wr99+L3DZPq4EKKar7/+GuvWIKyqlaghkRdmsSiQOD6B6XdOa1R/qiUCVHM9jcy+dkIIUYesLTkAhHUPZewjo/l6/ToAhg0bVmUEd2MU7fkXoGNNnIk5ql+gQhVCiBZhP1HIjk07KCwsJCwsjKFDh7Z1SEKIDqr+eb8XcbvdHDhwgKioqCojUm655RZuueWWgAYnWo/dbm9SO1vSLMqOfUxIv9ux9ZhVZZ+iKIx5aBTZu3Nwl1600KEsBi2EaEEVJS4mTZqEyWTC66rexpV/APu2PwMQ0v9ObAlVH3TmHspj1+u+5Eq34fHEDoytst9oNXLjRzewYsUKkqcltcBZCCE6ssLCQlYtXk1fZz9ivXEMWTiYNbt9DzBNQSbGPNT4eyBjcDyx895Bc9rRPGXkr/kBANHTX0QxWABfEsYYLKV6hRDVOQsv3BDtfGU/SeN7Y42w4nA42LBhA1D/zF8AT2k2mtOO7nX6t5Ue/Qhn9iZAJTjlhoDHLoQQgaRrOhv/tpnco7lEhUUzZsJojEajv1KPEEI0RqMTLKqqMmLECP72t7/x8MMPt0RMog1EREQ0qZ0xuBuxs5eiGKuP3HbYHThLXIx5aBTr/7Kx6k5ZDFoI0UKys7PZt28fiqIwddwAXPkHcRba2frRQgCmjT4INpWCzb8CzYWl20RCB95brZ+YvtEM//4wdI9WLbkihBD1Wb58OVnKaaJ7RDNrzqwqM3aHfX9Io++BdM0DKL7kSXA8mqfcv88UmSozeoUQddJ1nV2LL6yp6SnzsPLJVVz5wjw2bdpEeXk58fHx9OtX98wTT2k251bMB63q6JXinf/vu6808tY+TOy8dyTZK4Rot5xFTjxuN5qmUWwqYvLkyfUf1ED+GcdaDaP8/I1kxrEQnUmjEywGg4GkpCScTmf9jUWHkZqaSmRkZJ1lwqKiokhNTaU080NUSwS2BF9t3pqSK7qus/kfWzm7M5uxj46m++h4srZm+9qrCgnjusti0EKIFlExe2Xs0N54Nt9P7nc3tqOv8u0v3bqM0krtwwY9gPLdIqy6rqNrOqrB933/69JaLW4hROfx7bffsmXLFhRV4ZqfXk1ycjIex4URkb2m9Gx0n8X7X8V5djMRo36GKaJPIMMVQlwCTnx9krPp56tsKzpZxImvT/rXrZs2bRqqWncVcc1pr/uhIYDm8rWTBIsQop2yRljxznKT/r8t9OrXi/j4wF2vKs841r1O8lb7BvPJjGMhOq8mrcHy0EMP8eKLL5Kfnx/oeEQbUVWVhQsX1tlmwYIFOE6upDD9zxRs/Cmu/IO1tvU4PHidXnSvTkRSuCwGLYRoFU6nk/Xr1wMwceyQ+h8AALrmK2Go6zrbX9zJN3/cgObRWjROIUTn5XQ6WbZ4Gegwc+ZMkpOTm92nu+g4JQdex52/D0/xieYHKYS4pDjsDrY8v63GfZv+sYXs4zmYzWYmTJjQypEJIUTzeBwelsxdzpK5y6sMZqmPpmmsX7+eUmMpU6ZMCWxMpdl4y7IxR6Vhikz1bzdFpmKOSsMclSbJFSE6mUbPYAHwer1YLBZ69+7NDTfcQHJyMjZb1bIEiqLw2GOPBSRI0ToGDRqE1eqrwVtZVFQUCxYsYEB8KQUbfwfoBPW6BlNk31r7MtlMTP/DVPKPFhDZMxJXSbF/37C7B0hpMCFEi6gocREbG0vv3r3JO9bwY4tOFZHx6WE0t0b27hy6jejacoEKITqtDz/4kB6Hk+lhTmbWhFn1H1APXdcpTP8LaB4sXcdjTah/fQQhhKig6zpbnt+Gp7zmB48eh4c0rR8RU8MIDg5u5eiEEKJ16S6d42tOUBZbSn5+PsHBwYwcOTKgr1Gy/1XKjn5ISNpthAy4K6B9CyHapyYlWH784x/7v37llVdqbCMJlo5nw4YNOBwOoqIiubrXTkpdRrpP/TMDhwzHdW4b+d/8HHQvtuR5hI/4MYqi1NmfoipE94mqtj1hrGTqhRCBp+u6vzzY9OnT671GXSy8RzhTfjmJ8vxySa4IIZrk9OnTbPp0E0M9IzCZTISEhzS7z/Jjn+A6vxPFYCV8xP81+tomhLi0FZ4o5NTG07XuV3SFWGccw9KG1NmPrut4Co9SfvLLQIcohBCtxrXHw5b0bbjinKDAuHHjMJlMAevfU3qWsmOfAGDpNjFg/Qoh2rcmJViOHWvEkGDRIWiaxoZVH9AtrJxZU8fTp2wDAGFxOuXHP6Fw+1OgubEmTCNi1M/96xVcbP//DqK5vPSf38+/hoEQQrSGw4cPc+rUKcxmMxMnTgTnqXqP0TUFV6kH83e54O6jurVwlEKIjsLj8PDWte8AcNP78zFa675t1nWdJUuWYDfYKZlayLUzr6syY7fyovS+r0PrjcHrKKBw93MAhA78PsZguUYJIRonPCmchNExZKXnotdQAVVHwxVTzIDx/Ws8XnOXkL/taVzndqA5a1+vUwghOgLFCsYgI/tc34KFgJcHKznwBuhezHGjsHQZUuX+TwjReTUpwZKUlBToOEQb27XlS24bsBWTQYeyl/3bi775YaVWCmGDH0RRa/6zKT5bwq7Fu9E8GuE9wkkcn9DCUQshxAUVs1fGjBlDcHAwLmfV/TnHUji8dSqpY9YSm3wYTVM4sH4WjjWHuezpFKzhljaIWgjRWRw/fpyjR49itVpZ8P0FREVVn8XbWEW7/oHuKsIYkUJw6k0BiFIIcanxluWQ3Ptpzu64Ga9mAirPgtMxmV1MmPEunpKrAQ3nue2guTEnXQWAYgzCdW47mtOOYrBgDO+DO39fW5yKEEI0m3mgCaWvQvYnZ+nduzcJCYF7buUty6Hs2EcAhPa/O2D9CiHavyYlWETnous6G9d9wXU99fpaorlLa90bEh/MmIdHkbM7h4Rx3QMbpBBC1KGwsJD09HTAVx7sYq5yG4c2zsDjtnBw4wwi4k6jaQYKzibidjrIy8iT2StCiCaz2+2c3HEKmx7EtdddE5DkiuZx4Ck6DihEjPxprQNchBCiLprTjtlSSNqEr9i3bt5FexX6jluF1VZM7qq70Rx5AKi2LkT1uNLXQlEJH/EEqjUac9QA3IWZ5H55eyufhRBCBIau62zcthEUmDx5ckD7Lj7wOmgezLHDscQOC2jfQoj2rUmf1Hr27Fn/+huKQmZmZpOCEq3r4MGDnD17Bno2rx9FUeh9WS96X9YrMIEJIUQDrVu3Dq/XS+/evUlOTq6yT9fh4KYZeDxmQMHjNnNo83QGTfuUYbPfw9jrt5JcEUI0y3+X/5e0/P5YNSv9Imous2O0Gpl+x7PffX1NvX2qRisxM1/BlbsLc/SAAEYrhLgUxSZnkHMsldxTvUBXQdHo0iOTuJ6HAXzJFcWAOXoA5tgRoLn9x9oSZ/i/Vi0RoJpBc9X+YqrZ104IIdqJ/e8eIGZwDLm5uZw7dw6r1cqYMWMC1r+3LIeyo9/NXhlwT8D6FUJ0DE1KsEyZMqVagsXr9XLixAk2bNjAwIEDGTZMsrUdxWeffdas44uyignuEoTBbAhQREII0XBer5e1a9cCVWevVDwAOJeZTO7JPhcO0FXOn0gh51gKcb1PEDteyl4KIZpu9+7d7EnfQ3/DQCJCw4kbEBuwvhXVgCV2RMD6E0JceoqyHBTlxhEWk0PauFVsPpuAx23BaHLRd+xqf7uw4T8mqOcVqEYbAG63u8b+jMHxxM57B81pR/c6yVt9LwDR019EMfjKraqWCIzB8S18ZkII0TB5R/LZ8couFBWyBmQBMHbsWCyW6iWijVYjt3y2oNGv4S3LwWCLwRAUL/duQlyCmpRgee2112rdt3v3bmbPns3NN9/c1JhEKzp16hTffvst3cLqnpFUG1epm1U/W4Mp2MTUX00iJD4kwBEKIUTddu7cSUFBAaGhoYwaNcq/3RgcT8iIf/L1kp2ATtWa45Cx7SpSbhsrDwCEEDVqyKL0TqeTN998E5fBRcmoImbfczumIFOzXtddeJTykysJ7XcHitHarL6EEGLXktNk71lAyui1JPbfRd/xq/xr0pltF65z5uhB/uRKfYzB8RAcX+U6aYpMbfDxQgjRmiyhFpKnJuH2uFl99Csg8Ivbm2MGEzvvXTRHfkD7FUJ0DGqgOxwyZAj33XcfTz75ZJOOz8rK4pZbbiE6OhqbzcagQYP8dfUB7rjjDhRFqfLfnDlzAhX+Jefzzz8HoH//mstZ1KfodBEehwdPuRtLmCwQLYRofatWrQJ8NXRNpgsPNnVdZ+u/d+NxX7ygq4/HobH91ROtFaYQooNxOZ01fl3Zhx9+SF5eHlFRUaT1SyMounkPF3VdozD9L5TsX0zhrmfrbKsabXS7cQvdbtwiDzWFEDXyur1Yw02oqofohGMAxPU8zMQbXyI2+XAbRyeEEK0jJC6YiU+Op3xoKZqmkZiYWK2sdCAoqhFDUOBmMgshOo4WWS0zLi6O/fv3N/q4goICJkyYwLRp0/jss8/o0qULhw8fJjIyskq7OXPmsHjxYv/3NU3rE/XLy8tjy5YtAEyYMAH2fdLoPmL6RnPFv+fisDuaPWJTCCEaKysri4MHD6IoCtOmTauyr/BEIdl7a78u6ZrOqQ2nsZ8oJCIpvKVDFUJ0MidPnmTHBztJcvfkivvnkZWd1ew+y45+jCt3N4rRRki/O5ofpBDikmYwGRh1Xw8Suv0Fs62szra6rrdSVEII0fp0XWf9hvUATJo0KWD9esvzcJxZT1DPy1HUFnnEKoToAAL+rz8vL49XXnmFhISERh/717/+lcTExCrJk549q6+8brFYiI+Xki7NtXLlSrxeL/369aNbt27k7mtaP7YoG7YoGTkphGh9a9asAWDo0KFER0dX2ReeFE7i+AROb85C16o/NFBUhYRx3SW5IoRoNE3TeOPVN0gp7ItFs2DLCoJmLkXndeRRtPt5AEIH3iflC4UQAXHixAlC6kmuVLRLje7XChEJIUTrOLnhFHmH8xlwQz9OnD3B2bNnMRgMjB49OmCvUXJwCaUZy3BmbyZqwp8D1q8QomNpUoKl8iLCldntdg4ePIjL5eLNN99sdL8fffQRs2fPZv78+axbt47u3bvzwAMP8P3vf79Ku7Vr1xIbG0tkZCTTp0/nD3/4Q7UHaxWcTifOSmUdioqKAN+ifbUt3NeeVMQY6FhLS0tZt24dALNmzUIzBINqBs1V+0GqGc0QjMvlYudLu0makkh035p/7hfzVIq/o/zsa9NSv5PWJufR/jT1XDwej/+4jvBzCNTvzOFwsGHDBsBXQ7dyf7ruBVSG3z+U7N05uEsvei0FjDYjw+8b2qw4Osvfn5xH+9JZzgM69nXN7fZU+bpyHGvWrCHzeCauKDdjosfSY0YiR9YerjNW3XNhn8ftRtGr3oYX7fg7ursYY0Qq5uRr2vC8O8ffn5xH+9JZzgM6xnXN69Y4+N5B+lzeh/xiDxavgslQ+wwVt1chv7jqda4h51nfda296Cx/f3Ie7UtnOQ/oGNe1WmOocr/mRv/uWqd5NHa8vJOS7FIMFpUtJZsB6N69OyaTKSDxao4CSjPfA8DSY14tfRrpcp1v5oxXB2+Afk6d5e9PzqN96SznAU0/l6aeu6I3YS7w1KlTUZSq9ewVRSEyMpLevXtz1113kZaW1uhgrFbfQp6PP/448+fPZ9u2bTzyyCO88MIL3H777QC89dZbBAUF0bNnTzIzM/nZz35GSEgImzZtwmCoPnTwN7/5Db/97W+rbV+2bBlBQUGNjrGzyMjIYP/+/YSFhTFt2jQURcGqZTPA9Roqmr/dt+pC1O8WWPUoNlxKOO5DHhxrXGCEkFtsKNbqaxtcTHfrlLziWwQx5G4biqn+Y4QQDZObm8s999xzyV3Xjh07xu7duwkJCWHGjBlV3peivHvp6tnIGeMksvf0wbWl+pukdaYZU5/2+SBAiEtde7iuecvKKHvD93XQbWD4Lo7y8nJWrVqFx+Nh8ODB9OzZs9p9cU1U3cUI5zMAbLf8GE0x+/eFeY/S1/0WOgr7zXdQpnYN/AkJIdpUa17XXN+6cW5wo0YqlE0rYd+ONTww7hjBFi8f74/jpL3q65e6jAwcMZUuXbo06nXquq4JITq/9nC/VtuzJl3X8Z7w4vrWg3GGyhervsDr9TJp0qRaB2g3VoJ7NV29mylRunLAfAc04H5QCNG+lZWVsXDhQgoLCwkLC2vwcU1KsLQUs9nMyJEj2bhxo3/bww8/zLZt29i0aVONxxw9epTevXvz1VdfMWPGjGr7a5rBkpiYSG5ubqN+UG3F7Xbz5Zdfctlll1VZvLm5ff70pz+lqKiIu+66i7FjxwLgOPk5xel/wBDWG29RJgAxV32JctHCqa4SF9v/s5PIXpGkXZvaoNf0ODy8970PALhq6eXYQjtuSbGW+J20BTmP9qep55KVlUXPnj0vqeuaruv89re/5cyZM9x4441Vrv+6rlOw+g68hZkE9f8+ebnT2fy3Lehe335FVeg+phsTfjquXZxLeyDn0b50lvOAjn1dK8nL49M7fWUIL188jZDvPoy/8O8X2LljJ8m9knnyySdRVbVB56l7ysn96DKg6v2VrusUrLoDb1Emtt7zCRnySCucXe06y9+fnEf70lnOAzrGdS1ndw47X9lDn3m96TUrmX/+8UEWDPiWcrfKn1f3xatXfQgYGRnJn//8Z1RV9W9rznWtveksf39yHu1LZzkP6BjXtdpUftZ0/X+vwWitPoBu7dq1LFu2jPj4eMaMGcOsWbOa/TvTnAXkfT4fvA7Cxj2Fpev4ZvXXWJ3l70/Oo33pLOcBTT+XoqIiYmJiGp1gaVdDd7t27Ur//v2rbOvXrx/vvfdercf06tWLmJgYjhw5UmOCxWKxYLFYqm03mUwd6o+lrnidTif3338/AC+88EKN51vZxo0bKSoqIioqinHjxmE0+v4MTL2vxBqViteRT/7XjwJgNJlQjVVf1xRpYvJPJqLreoNGbFbEf+NHN7BixQpsobYO9bOvTUf7G6qNnEf709hz8f8b7mA/g+bEe/DgQc6cOYPZbGby5MlV+nFkb8ZbmIlitBGWOp+IQeFEJgWz4sHVgK802NiHRwf0Z9XRfva1kfNoXzrLeUDHvK6ZTMYqX5tMJnbt2kXOhvOMLhvLxDsmVLvnqiteTblQwuLi+6voyX+jeN/LhA/5AWo7+Z13lr8/OY/2pbOcB7Tv61rCyAS6DesGgGpQuXJ8DBTCwXOh1ZIrAAsXLqz1M2RTr2vtUWf5+5PzaF86y3lA+76u1cZVdqHMvaZpmEymas+q1q/3leiaPHkyLpcrIPEWHXgHvA5MkWkEJ05u8LOxQOssf39yHu1LZzkPaPy5NPW81fqbVPfcc88xe/bsWvfPnTuXf//7343ud8KECRw6dKjKtoyMDJKSkmo95vTp0+Tl5dG1q5RSaAhN0/j8888BmD17tv8NsYIpsi/mLkNrPLb0fNXFEdvqDUQIIVav9iVLxo0bV206esnBpQAE9bwK1eJbwD4k9kKbkfcNwhphbaVIhRCdhcPhYMmbS+hRmkSIJxRTUeA+dBiDuxI5+peopuCA9SmEuLSpBhXV4Pu43yWuKw6Pgb3ZVUdiRkVF8eCDDzJy5Mi2CFEIIZpN95RX+bq8oJzPHv6CE9+cRNd1jh8/zsmTJzEajf7qLc2lOQspPfwuAKED7pZnY0KIpiVYXnnllWozTSrr378/L774YqP7feyxx9i8eTN/+tOfOHLkCMuWLePFF1/kwQcfBKCkpIT/+7//Y/PmzRw/fpxVq1Zx9dVX06dPnzoTPuKCnTt3kp2dTXBwMJMnT/Zv17ULo488Dg+rX3uU1a89isfh256fWcCHd31M+n92oHm1av0KIURrKSgoYMeOHQDVZi66Cw7hytkKioHgvgsoPV9a7fjEcd1aJU4hROfywQcfkJefx9GUwwxY0I+Uy/s0qz9d13Dl7QtQdEIIAXuW7uXY6uPoWtUq4AfKR/GnVankuLty58jjfG/waR57+EGefvppSa4IITqVA/87SP6RAva/ewCAdevWATBixAhCQkIC8hqauxhT9EBMEX2xdJsUkD6FEB1bkxIsmZmZ9OvXr9b9aWlpZGZmNrrfUaNG8f7777N8+XIGDhzI73//e5599lluvvlmAAwGA3v27OGqq64iNTWVu+++mxEjRvDNN9/UWxZL+Gp8f/bZZwBMmzYNq9U3glvzOMj5+GoKtvwOzV1S5ZhTm84AcGb7WTSPRum5UhRVsvNCiLazbt06vF4vKSkpJCYmVtlXMXvFljiD4nNW3r/tI1b9Yg2at90sNyaE6CA0TaPAnE+29SwbNm/miy++AGDhnQsZdttQ/8jwpirL/IDcr+6icMffAhGuEOISV3i6iG+X7WXD05vIO5xfZd+qVavw6gqTJk6mT0wZQ7oVkZrap8qaK0II0dFlbclh0IKBDLp5IMPuHorT6WTz5s0ATJkyJWCvYwxJIGbq80RP/5fMXhFCAE1cg8VsNpOdnV3r/rNnzzb5Zu2KK67giiuuqHGfzWbzf7gVjXf48GEyMzMxGo3MnDnTv91xajWaIxfX+V0oxiAc5y7ckKf/Zw/dRyUx8Hv9iewZQXRKlLyBCCHajMfj8Y9Cunj2ircsh/JTXwEQknYLxzbkggImqwmj5UKJMLWdLsAqhGg/0tPTWbpkCWabldTiNFa/uZ5QYyhx/eIYPHhwo/rylGajOe3oXqd/myN7G4W7nwNAMbfNorBCiM4lOCaIIbcNovBUETF9owHfALtje9dy9GgmRqOJiRPG4V7XxoEKIUQAOQsvrMGy65X9JI3vzZBbBgHw9ddf43A4iI2NJS0tDY/HU1s3TaKaAjMjRgjR8TUpwTJ27Fhee+01HnvsMUJDQ6vsKywsZPHixQGrbSgCp2L2yoQJEwgPD/dvLzv2IQBBva4EFLa/+K1/n6fcy5Z/bmPKLybRfZSU1RFCtK0dO3Zgt9sJCwtjxIgRVfaptliiJv8/XDnbMUX2JfUK6DaqG5rb20bRCiE6ovT0dBYtWoTJa2Zw0XCMupH+hQNRUTm+8yjp6ekNLqnjKc3m3Ir5oLmqbLdv+D//1yUH3iCo55UYg+MDeh5CiEuL0Wpk4I0Dqmzz2DOw7v8JD02wsINbCQ0NIb+W45tCNdroduOWAPYohBANp+s6uxbv93/vcXj8z6/Al2AB3+L2gRgorLmKKDn4JsGpN2GwRje7PyFE59GkBMuvf/1rpkyZwtChQ3n00UcZMMB3I7d3716effZZzp49y7JlywIaqGierKwsdu3ahaIozJkzx7/dU3wS1/ldoKgEJV/Oia9PkrX1wuwkXdM5teE0J74+SdLkHm0QuRBCXFCxuP3UqVMxGqu+hSmKgjV+LNb4Cwn+kDjfgtEV60kJIURdNE3z3cPqkFbUD4NuQEFBRUVBodRYyvLlyxk+fHiDZmtrTnu15Er1Ri5fO0mwCCECrOCIb4BdbqmZmVdf1sbRCCFEYJ34+iRn08/7v9c1/M+vDL1UMjMzMRgMTJw4MSCvV5LxNiUH3sCZs50ul70akD6FEJ1DkxIsY8aM4eOPP+a+++7jkUce8WeCdV2nZ8+efPTRR4wbNy6ggYrm+fzzzwEYPnw48fEXPsCXHf0IAEv8ONyuMLY8/3WNx295fitxg2OxRlhbPlghhKjBqVOnOHToEKqqVquhq+sainLhYafH6cFoufAWZ7QaueWzBa0WqxCiY8rIyKCgoIBYRxyxzjj/doXv7nXRyc/PJyMjg7S0tLYKUwgh/E5tPs2RzzIZdtdQIpIuVCnQdZ3iY19gA/JIITk5GUepvc3iFEKIQCo5c5Itz9U8g27zPzajzisAYMiQIVUquDSV5iqhNOMtAELSbm52f0KIzqVJCRaAyy67jCNHjrBz507/gva9e/dm+PDhskZHO1NQUMCmTZsAmDt3rn+7rnkoO/4pALaeV7Hl+W14ymse5e0uqzrVUgghWpLT6eT+++8H4IUXXsBisbBmzRoAhg0bRlRUlL+t7nVyfuVtWLtPJaTf7RSdcfPFYyvpObMno34wQt6ThBANZrfbMXnNpBX1R0f3J1bAl1xJK+pPgbkAu93edkEKIcR3dF1n9+t7sB8vJKJnBMPuGOLf5yw4jE3Px6Mp9B59IwCK0cbPP+8PwAvXyJp0QoiOyV1ylvW/ewN3eU+g+oxid7mb6K05hFvdAVvcvvTw2+juEoxhPbEmTAtIn0KIzqPJCRYAVVUZMWJEtTr4on358ssv8Xq9pKam0rt3b/92x5kNaI58VGsUTvcgTm1cWWsfFaXC7CcKq4yMEkKI1lBWVsbGjRsBmD59etV9xz/DU3Sccs9nhA78PifXH8Vd7qE8t1ySK0KIRgkPD2dQaV+M35UGq0xBwagbGFTal4iIiFr78DrtuHP34Mrdg+PsphaOWAjRmXkcHt669h0Abnp/PkZr9fKok34+kX1v7WPA/H5V9h3dupQI4HhhBONHBaY8jhBCtAf2zGxyT/SuvYGukncyhR4D7QwcOLBRfWuecrLfmwpA/PVrUY02NHcJJd/NXgkdcFeVyglCCAFNTLAsX76cL774gtdee63G/XfeeSdz587le9/7XnNiEwFQVlbG2rVrgaqzVwDM0f0JHXgfitFKcM8oEscncHpzFrqmV+tHURUSxnWX5IoQok1s3LgRp9NJt27d6NfvwgMEXdcoPeRb8ys4dQGKamTQwgHEDuyCKdjUVuEKITqoGN1IZFnXWvcrqESWdSVGN/i3ae4SHCdWkez+lPwvl+ItPtEaoQohBADhCWGM/3H18tzauQ1gAUPsxGrr1gkhREcWlmAlvEsWhee7ATUMqFM0uvTIJHZcaoPWzKtP6eF30F1FGEOTsSbMaHZ/QojOp0lXmr///e9YLJZa99tsNv7+9783OSgROOvWraO8vJxu3boxePDgKvsMti6EDriLkL4LURSFMQ+Nwmir4eZbAVOQkTE/HNVKUQshxAW6rvvLg02bNq3KrBTHmW/wFJ9AMYUQ1OsqwDeaM35IHNF9omrsTwghahMW5yGmxxFQtJobKBpdkg4TZDnk36R7nRRv/xNdvLv9yRVjaDJBva4mZMA9rRG2EOIS5HV7a913bP8GoizFeDUYOOXOVoxKCCFah46CL7ly8QBhHaPJRd+xqxk2bFizX0dzl1Ly3YC+kP53oqiGeo4QQlyKmpRgOXToUJ0XqiFDhnDw4MEmByUCw+PxsHKlr+zX3Llz683cWyOsDL19cPUdOox5aLQscC+EaBOHDh3izJkzWK1WJkyYUGVfyYElAAT3uQ7VFIyuV5+BJ4QQDaUoCmnjVmE0uqjrA3tp5v/8Ww3WaCwJMzhrGEvY2L8Qd80XxM57m4hRP8PaTdauE0IEnqPQyQd3fszuN/bgdVVPtHy1/lte3daDg66xhEd3b4MIhRCi5SiKwuAZHxOdmEn1GSwKfcetwmwrD8ji9mgegnrMxhjeB1uPy5rfnxCiU2pSgkXX9ToX9ywoKMDtdjc1JhEgmzdvxm63ExERwdixY/3bdV2jYMvvKD/5Fbq36u8paUoSYYlh/u8VVSFxQgJJk3u0WtxCCFHZunXrABg3bhw224UFWV25e3Dn7QHVRHDKjRSdLuLj+1Zw4H1J8AshGk/zaGx/7RSa10jf8aup6wO7ag6rsids9G85bZqOpdtEDJYI/3bVEgGque4XVs2+dkII0UBHVx2jPK+c01uyUI1VP9IXFhayZWs6mXkh9Jn8aNsEKIQQAeZ1e8nZc87/vdlazuDpH1edefzdTOO4nof97TSPg9LD76A5C5v0uqolnPARP6bLrNdl9ooQolZNKsY6bNgwli9fzuOPP47ZXPVDo9PpZNmyZQGZiieaTtM0PvvsMwBmzZpVpe6u69wOyo9/iuP0GuK6TUDhwjoF1jAL0383hQ/u/BgAo01KgwkhWp+mXSjPs2vXLgBmzKha77bk4FIAgpLmYLDFcOSLXRSdKiJ7Vw79rk1rtViFEJ3DzsW7yPwyl7Nh1zLq6jeJ6XGE3FO9QFf9tbwrPrCHDrq/QX0ag+OJnfcOmtOO7nWSt/peAKKnv4hi8JXbVS0RGIPjW+akhBCdUr9r+xISG4w1woKiVk0Gr127Fq/XS+/evenZs2cbRSiEEIHjcXhY85t1nNt7num/m0J0sm+7okDauFVsPpuAx23xzzSuzJG1jsIdz1C461nM8RMI98aiax6g6nqdntJs//1aBXdBhtyvCSEapEkJlp/85CdcccUVTJs2jZ/85CcMGDAAgL179/LnP/+Zffv28dFHHwU0UNE4e/bs4cyZM9hsNqZMmVJlX9lR3+/G1mMWqtFW7djKpcBG/WCElAYTQrSq9PR0li5d6v9e13WMRiNnz56le/cLZS7ChvwQ1RpNcMoNAAxaMIDQriFVZuEJIURD9bmiN3s+206f0V9jMOj1fmBvKGNwPATHo3nK/dtMkak13oMJIURDKIpCj4mJ1bZ7PB7O732buWkF9B53RRtEJoQQgWewGLBF2TCYDVxcEdpsK6fv+FUc3jqV1DFrMdvKq+xXjFZMEX1x2w/hOrOOVCDvs68ISppDUM/LMUWk4CnN5tyK+aC5qhxbMTAGANVM7Lx3JMkihKhRkxIsc+fO5ZVXXuGRRx7hmmuu8W/XdZ3Q0FBeeuklLr/88kDFKJqgYvbK1KlTCQoK8m/XXEWUn/YtFh3U62r/dl3X2f36HrqP6U5E8oU6lT0mVL9xF0KIlpKens6iRYuqbfd4PCxatIgHH3yQkSNHAmAMTSRi5BP+NqYgEynz+rRarEKIziV9fzpHElYxO+E4UP8HdiGEaG0OuwNzqBnVUHOl7/T0dAZEnyEpspyQOE8rRyeEEC1DURTGPTqGkpxSwhPD8JQqvhKs3yVE4npWLQsG+EuwmqPSsHWfgtt+mJLMjyk68jEmZwGlGcspzVhO7BUfojnt1ZIr1WguXztJsAghatCkBAvAHXfcwXXXXceXX35JZmYmAL1792bWrFmEhoYGLEDReJmZmWRkZGAwGLjssqqLcJWf+AI0F8bwPpii+vm35x7MY+/b+9n/v4Ncv/QabvlsQWuHLYS4xGmaxrJly+pss3z5coYNG4bBIPVvhRDNd+SLTMITwwjrHcZHH31EiFJ1WGSNH9iFEKKNbHhqE2W5ZYz70Vhi+kZX279x7cfc1LMcHQjuMb31AxRCiAA5vSWLvEN5DLltMAAGs4Hw7yoVVJRg3bdrIytWrKCkpMR/nMFg4Prrr2fA0PFVZpuYIlIIGfwQ35xKZvrwKFynvkBzl2AMjsfltLfquQkhOp8mJ1gAwsLCuP7666tt37VrF0uWLOGZZ55pTveiiSpmr4wbN47IyMgq+0q/Kw8W1OsqFOVCvV5LmIXel/VENRmwhFpaL1ghhPhORkYGBQUFdbbJz8/n5DdPEW4uJrT/XZgi+lB8ppj0/+wgZV4fEsZ0r/N4IYSocGb7WTb/YysGs4HgG6wUFRUxfuDFC9vXQBalF0K0gSNfZJKfmY+7zIM1vPrntaNHjxJcvhcAQ8QADLYu1dpYLBYWL17c4rEKIURzFJ8pZt3vvkHXdCJ7R9ZYWWXfro289fY7ZBdbgKplV599+WMefLAHI0dWn22iKwYsXScQ0mMquq5V2y+EEE3RrARLZcePH2fZsmUsXbqUAwcOoCiKJFjaQE5ODjt27ABgzpw5Vfa58g/isWeAaiIoqeq+sO6hjHt8LPrFBS2FEKKV2O32etuYDRrG8ytxaGXYEmdiiujDkc8zydp6Bl3TJcEihGiw2AFd6DaiK5ZoC8s2LiE6yMmExCzQITjlJogcx5pfbAFg2h/GYAuPAGSRUyFE63HYHf6vd7/5LXOfm0XRqWJC4kOqtV21ahUD44sACEm+rNp+IYToKEK7hTL4loGUZJfW+PlO0zRK9v6HhybY+Xh/PJtPRlVrs3z5coYPH46q1lxSEUBRat8nhBCN0awES15eHv/9739ZunQpmzZtwmQyMWXKFB544AGuvPLKQMUo6qFpF7Lu7777LrquM3To0CqLQfsaujBFD8IYHI9qCacmlWe1CCFEa4qIiKi3zYiEAgxaGYaQBKzdpwDQe1YvNE0nfkhcC0cohGjPNE852e9NBSD++rX1LiJvtBqZ+uvJ/O/9/+HYU8rdk3JRdRfm2BGEDXuE0vx8SvKPAWAI64M5KqaFz0AIIS7QdZ30F3b4v/eUe9j56m6m/GJStbaFhYXs3bmeuZPLALAmTGu1OIUQIhC8Li+6pmO0+h5TDrxpAFDzM6rD+7aSFG4HICO3esIZfJUPMjIySEtLa5mAhRCikkYnWMrLy/nwww9ZunQpK1euBGDMmDEALFmyhBtuuCGwEYo6paens3TpUv/33377LQC9evWq1tYcM5guM19G915YvEvzaOz/30F6zexJUFTdDyKEEKIlpaamEhkZWWuZMFXRmdTLDkBI34Uoqm8dlrCEMEbcM6y1whRCdGD2E4UUZBbQc3oyAMWlxXy16ium9solPqgQxRRC5Jhfy4hGIUSbO/H1SU5vyfJ/r2s6pzac5sTXJ0ma3KNK23Xr1pEWY0dVwBQ1QGbZCSHaLY/Dw1vXvgPA+B+PpdeMnjiLnKz93TeYQ8xM+eVEVINa5+Bfz9k1qAqcKLCRX2autV1DKiQIIUQgNPjT4xdffMFtt91GXFwct9xyC+Xl5SxatIjs7GxeffVVdF2vc+qdCLz09HQWLVpU45vG//73P9LT02s8TjFceAM6ueEUuxbv5vNHV6JrUh5MCNF2VFVl4MCBte4fEFdEuMWJaokgKPnyVoxMCNEReBweVr/2KKtfexSPw1Ntv8PuYPXP17Dh6U0cW3McgE8++YQu1gKm9ckFIHzEExiCfLPhzJYLaxxU/loIIVqaw+5gy/Pb4OLniwpseX5rldJhHo+HNWvWoAAeQ5jMXhFCtGuVr1/b/r0dh91B8dkS8jLyOLf3HMVZxfX2EebYCcDuMzVXZqnQkAoJQggRCA3OiMydO5cNGzbwpz/9iaysLL766ivuueceIiMjpaxUG9A0jWXLltXZZvny5f7yYeWnVqE5C6u1sYRZ6NI/ht6X9UJR5fcohGg7ubm5bN26FQCbreqMuqioSK4b7XvLCu4zH8VopSSnhO0v7aDwVFGrxyqE6HgsYRaSpvQgPDGMbiO6kpuby/qvVzN/8BlURcfWYzZBSbPbOkwhxCVO13W2PL8NT7kHLh7/poO7zMOWf27zb9q+fTt2u52DRT3pevWnhKR+r3UDFkKIBqqp9OGWf24jpm80k382kdl/u4zwHnUnTdyFRzGUn8CrKezJDqu1XVRUFKmpqQ2KS7VEgFr7TBhfI7OvnRBC1KDBJcLi4+M5duwYr7/+Oi6Xi5tuuolu3bq1ZGyiDhkZGbWW0alQUXOyT/dgCjb+DMVgIe6qT1HNof42XYfF03VYPJpXq6MnIYRoWbqu89prr+F0Ounbty8//ME97Fo6n2Knke5T/0xKV42CdQ+hGCwEpfhKUR754igH/neIgmN2Zv5pehufgRCivVNUheH3DGPQwkGYg00sfXUpDpeXw+WDiO+aT/iI/2vrEIUQgsIThZzaeLrW/RWlwuwnColICufLL78EYOrUqZhM9TwgFEKINtSY0oe1KT/xue//tn6Uu2tvt2DBggZX2TEGxxM77x00px3d6yRv9b0ARE9/EcXgm8WsWiKk/KIQolYNnsFy+vRpVq5cyYABA/jtb39Ljx49mDx5Mi+88ALnz59vyRhFDRpaS9Jut1N27GMAzLHDqyRXKlMNUt5NCNF21q9fz759+zCZTNxxxx0YjUZ6RZcxpFsRqal9sEQPIGzoI4T0ux3DdyOH4gfHkjC2O6nzUto2eCFEu+V1eTm84gi67hsGrigK5mATZ8+eZf369YDC0Dk/osvspbXeIwkhRGsKTwoncXxCrdUFFFUhcUICEUnhHDt2jMzMTLqGu5k6dXIrRyqEEA3XmNKHdXFm+yoe7D0XDYDBYKiyPyoqigcffJCRI0c2Kj5jcDzmqDRMkRdmvZgiUzFHpWGOSpPkihCiTg1+qq6qKjNnzuS1114jJyeHpUuXEhERwSOPPMKECRNQFIWvv/6arKys+jsTzdbQWpIR4aGUHfsUgKCeV/m3O4udHFt7HK/b2xLhCSFEg9ntdt566y0ArrnmGuLjq9+8qqZgQvouJHTA3f5t8UPjmfrryfSYmNhqsQohOoZTm86g6zobntrElue3VSlHAfDpB8uxGDwMHTqU3r17o6iGWnoSQojWpSgKYx4ahdFmrPFBpCnIyJgfjgLgq6++wmr08sDYTBxrF+J11F3hQAgh2kJjSx/WJWbmyxQm/JDPtpzHYDDwf4//H8PzRzLAPoiHH3yYp59+utHJFSGEaK4mTVuwWq3ceOONfPTRR5w9e5bnn3+ecePG8dxzz9GjRw9GjBjBb3/720DHKipJTU0lMjKyzjZRUVH0CM1Fc+SiWiKxdpvk33fki6Ns+Osm1v7m65YOVQhxCXA6ndx5553ceeedOJ3OBh+n6zpvvvkmZWVlJCcnM3PSEFz5B/EUHva38RQexpV/0Le9NLslwhdCdAKOwgvXnvT/7MFZ6KT76G4YLAYSxyf49x0/doxenk95eEIm183s3xahCiFEnawRVsY8NKrGB5FjHhqNNcJKUVERW7dupW+XYlRFQ7VEYLDW/flQCCHaQkXpQ127+KLmU7n0YX1cbi+vfrAHj6Yya9YsEronEOmKIt7RldSU1AaXBRNCiEBq9pUnKiqKBx54gPXr13P06FF++9vfUlZWxu9+97tAxCdqoaoqCxcurLPNggULcBz/BABb8lwUg8m/z2Q1YouyNbjOpRBCtIRt27axY8cODAYDd918Jbmf30Tul7dT9M0P/W2KvvkhuV/eTu6Xt3NuxXyKTh4n49PDuErrKLorhLik6LrO9he/9X/vKfey5Z/b6D2rF9e+dhXxQ+L8+3Z/8TRpsSWEWHXiu/dqi3CFEKJeSZN7kDCmu//7itJgFZ/f1q5di8fjYUxvX0UCa6KsRyeEaJ8aU/qwNrrmQdc1PvnkE3Jzc4mKiuKqq66qtb0QQrSmgKZ2k5OT+cUvfsGBAwdIT08PZNeiBiNHjuSBBx5AUaq+SVXUnBw2IBnHmQ0ABPWq+saTekUK17x2JT2nJ7dWuEIIUUVxcTFLliwB4PLLLycuKgg0V90HaS6OfHGMrf9MZ/1fNrRClEKIjuDE1yfJ2nphhlvlRVOtEVb/9sN71jEkzFcuzNjnDkwRvVs9ViGEaAhFURh5/3D/90bbhdJgHo+HNWvWYDZ46RGSB4AtYVqbxCmEEPVpTOnD2jiy1nH2wyux73sTgIULF2K1Wus8RgghWouxpToeNmxYS3UtKomJifEv3Arw2GOPMXDgQFRVpfz0GgBM0YMxhfWsdqzBJPXGhRBt57///S/FxcV0796dK6+8Eq3oSIOOC+liJjwxjJ4zkls2QCFEh+BfNPVi3y2aGjc4FmuEFc3rpnznnwi26uR5Exg44p5WjVM12uh245ZWfU0hRAdnurDo89C7+vkTxtu3b8dutzO6pxcFD4aQRIzhfdoqSiGEqFdF6cP1f9lYdUel0od1KTu+Apy52IzRDB48mOHDh9fZvqnkfk0I0RRSnLCD27t3b5Xv+/bt6685aUuYRtxVHxMx8kn//pKcUvIO57dqjEIIcbHs7Gy2bNmCoijcddddGI0Nz/cnT47miv/MI2mSlDgU4lJXZdHUajurLpp6dO1fiLHaKXMb6D7tzyiK3AYLITqO7mMulDpctWoVABP7+e6fbInTqlU1EEKI9qa+0oe18TrtOM74EjP7zsdw8803yzVPCNGutNgMFtE69u3bB8BV/c8ypkcBuqccLBb/foM1GoM12v/9gfcOcOjjw/S/oR/D7x7a2uEKIQRlZWXs2rULgFmzZtGrV+PXQFAUBcUgN9VCXOoqFk2tTUWpsJydW7Ce/wQUOG6cQ5+uqfX2bbQamX7Hs999fU2AIhZCiOY5fvw4hw8fxmpSiFaOAWBNkPVXhBDtn6IojH10NB/e8wnuUneDSoMBFB35FAWNrEIrY6bdQGxsbCtEK4QQDSdD9zqw8vJyjhzxldRJiSmpss/rtNd4jK7pqEaVrsPjWzo8IYSo0XvvvYfD4SA2NpZrr722wcc5y4LIP5OIrun1NxZCXBIaumjqKWc5+7LD2JMdxZjLH2vlKIUQInAqZq8MGT6KmGn/JnTgvZgi09o4KiGEaBhrhJWxD4/GGmllzMP1lwYDyP72vwBkliQyd+7cKvuMViOr4leyKn4lRquMIRdCtA25+nRgBw8exOv10qVLDFFB+/3bNVcJ5z6+GlNUf6Im/hXVHObfN/qHoxh088AGvYkJIUSgHThwgG+++QaAW2+9FUulGXf1OXN4IMd2jie/+ASTf9WvpUIUQnQgFYumZu/OwV3qvminb9HUkT8Yzh+e+QPZ2d255qorCQ0NbZtghRCimYqKiti8eTMAl102C3N0b8zRA9o4KiGEaJykyT3qLQtW4fjBTYQr2Wg6DJnxICaTqYWjE0KIxmtWgsXpdLJjxw7OnTvHhAkTiImJCVRcogEq1l/pl9YX+Nq/vfzkSnSvA81pRzFVf4hgi7S1VohCiEuE7innj3P2+7+mhsSJ0+nktddeAyA5OZm+ffsC4C46huYsRDHUn/g1mJzEDwmrt50Q4tJR16Kpo+5NZce+7WRnZxMSEsqsOfMa3K8sciqEaG/WrVuHx+MhOTm5SSVWhRCiI9E0jb2rFjE8Bs57ujNsyLi2DkkIIWrU5BJhzz33HF27dmXixIlcd9117NmzB4Dc3FxiYmJ49dVXAxakqFnF+iv9+/Wtsr3s6IcABPW6yr/wV/GZYhx2R+sGKIQQlbz//vucO3eOyMhIBgwYgK5rlGS8zfmVt1Ow6RcoigFUc63H9xyylYk3vU7SlOTWC1oI0SEkTe5B15Fd/N8rqkLC2FisJT/Gu+8vWIxeLr/8cmw2GWQihOiYvF4va9asAeDKSUkUbvsTznM72jgqIYRoOevWrWNThof0rC4kjrq7rcMRQohaNWkGy+LFi3n00Ue56aabmDVrFnfddZd/X0xMDNOnT+ett96qsl0E1vnz58nJycFgMJCakkL5Sd92T2Em7oKDoBqxJV+oTZn+4k7Obj/L2MdG02t6zzaKWghxqcrMzGTlypUA3HLLLWSf3Efhhsdxn0sHwBTeG9UWRey8d9CcdjzlZax8wldKbNZTkzDaggBQLREYg2UNKSFEVYqiMPTO/pxNXweA0WogddRn6MWFhBithISlMn26LAIthOi4dn+7l4KCAsLCwkiynaDs2AoUgwVL7PC2Dk0IIQKuqKiId999l7IyG0GD7yGm78y2DkkIIWrVpATL3/72N66++mqWLVtGXl5etf0jRozgueeea3ZwonYV5cF69+6NzWal/LvtzlOfA2DtPgWDJQIAr9uL0+5A82hEp0a3QbRCiEuZ2+3m1VdfRdd1xo0bR0rkOWIOv4z7nAPFYCFsyMME9bneP+OO4HgoKaYk/xAAXjUJzW0hJC6kDc9CCNFeeUqz0Zx2jIodS3AhrrIQBszLg+JvcHvhs4OxXHHlNZjNtc+QE0KI9m7dhg0ATJsyCVf2fwCwJkriWAjROb399tuUlZXRo0cPpk2b1tbhCCFEnZqUYDly5AgPP/xwrfujoqJqTLyIwNm3bx/hVjej+kXhKTzs3+48/SUA5phheEqzMQbHYzAZmPPsLApPFhKeIGsXCCFa1yeffMKZM2eICA/hmn7HKd76CkbAGNmPqLG/xRiWVOfxR786ybfLD9LvujRG3DOsdYIWQnQIntJszq2YD5oLgAnzq+43GeD2kaeJG96nDaITQojmUc0qyVe+yEm7jaOH41BVlYmDInDtKEK1RGKOGdLWIQohRMAdPHiQLZs2cHm/HEbMugFVbfLqBkII0SqalGCJiIggNze31v379+8nPl5KuLQUr9fL6cxdPDbpCCbnYYq+qbTejcc3l6Vo5zMU7X6O2Hnv+MvphPcIb4twhRCXsJMnT/Lpp58CsPDm2zB5PsajGMhSxzFkyh8wWupfDyFnz3nQISJJrmFCiKo0p92fXKmNUdVQPCWtE5AQQgRIeno6S5cuwW5P9m8zGo0UHFlBMGBNmIqiGtosPiGEaAkej4c333yT1C4ljE/KRz2xCAbPAUWud0KI9qtJaeB58+bx4osvYrfbq+3bt28fL730EldddVVzYxO1OHr0KAatBJNBr7uh5qL49Dk0r9Y6gQkhLjme0mxc+QerzKTzFB7GlX8QR+5+3lvyPCbFxfDhwxk5ciThI35MxJR/ccY0GUWtPcfvKHT6vy44Vsicv19G0qQeLXouQgghhBDtQXp6OosWLcJuL6yy3e1yop/fDIA1QUrmCCE6HqfTyZ133smdd96J0+mstn/lypWcOXOGkT1KAbD1mC3JZCFEu9ekGSx/+MMfGDNmDAMHDuTKK69EURRef/11Xn31Vd577z26du3Kr371q0DHKr6zb9++BrXTddjw96O4So8x6WcT6NIvpoUjE0JcSi4uzVOh6Jsf+r/+Xi84GhnGwJtuQVEUFHMYpqgBwIla+9V1nR0v7b/wOuVe9r17gCm/mBTwcxBCCCGEaE80TWPZsmU17kuOKiPE4qXcY8QUI2VThRCdS25uLh9++CFWo5fUmCIAgpLn1nmMxWJh8eLFrRGeEELUqkkzWLp168b27duZM2cOb7/9Nrqu8+abb/Lxxx+zYMECNm/eTEyMPMxvKRUL3NfHWRZCeYEbZ7GTsITQFo5KCHGpaUhpHkWBpGgPYZa621V24uuTnN6S5f9e13RObTjNia9PNjVUIYQQQogOISMjg4KCghr3WQwauaVm9meHcPjI0VaOTAghWtby5ctxuVzMGGJDxYMxvDfGiJS2DksIIerVpBksALGxsbz88su8/PLLnD9/Hk3T6NKliyw+1cJKS0s5evQoXRuQL7EGl3DFcwMoyY/FEmpp+eCEEKIGXSb9FUNQXIPaOuwOtjy/DRSgchVEBbY8v5W4wbFYI6wtEqcQQgghRFurqQx3hYPnQzl4PgSjqnP3hNrbCSFER7Nr1y527NiBwWBgXG83FIEtaQ6KorR1aEIIUa9mZ0N0XUfXdV/pF7nwtbgDBw6g6zoxXbo0qL3BpEppMCFENfXVvg0kgyWsQe10XWfL89vwlHuqJlcAdHCXedjyz22BD1AI0WHpej3r0TWynRBCtLWIiIh6Wih4NLUB7YQQov3RPeX8cc5+/jhnP7qnHPB9Nl26dCkAV142FqVoH6AQlDS7DSMVQoiGa3KCZf/+/dxwww2EhYXRtWtXunbtSlhYGDfccEODS1iJxqv42fbp3afOdm6HjPAWQtSuphvbtlZ4opBTG0+jazU/CK0oFWY/UVjjfiHEpefEidrXc2pKOyGEaGupqalERkZW294l2IlR1QCIiooiNTW1tUMTQogW8cknn5Cbm0tUVBTTJwzEEBSPOXZ4g6sgCCFEW2tSibBvvvmGuXPnomkaV199tf/m7tChQ3z00Ud89tlnfP7550yaJAsSB5Ku6/4ES69eveB4ze00r8qWD24lKLyAicNdmKNaL0YhxKXDUxzYB5bhSeEkjk/g9OasGpMsiqqQMK47EUnhAX1dIUTHVVimYfEqmAy1z1BxexUKy7RWjEoIIZpOVVUWLlzIokWL6B2uzwAAXERJREFUKm3VuWPkCawmjVe3JnHNggVSmlsI0SmcPXuWzz77DICbb76Z0IThhHR/H80pg+qEEB1HkxIsjz32GLGxsaxbt47ExMQq+06dOsXkyZN5/PHH2bZNSrkEUk5ODnl5eRiNRnqmDsF+0lzjAtNFufG4nTbKinWCYqU8mBAisDylZyje8wLlJ79oUPuGluZRFIXECQlkbTvjO+aiNVhMQUbG/HBUEyIWQnRWoTG9+PvSPgSbPbW2KXUZuf+RXq0YlRBCNM/IkSN58MEHWbp0CXZ7IQnhDiJsHlxelWtufpiRI0e2dYhCCNFsuq7zxhtv4PV6GTJkCMOGDQNAUVQM1uoz+YQQor1qUoJl3759/P73v6+WXAFITEzkBz/4Ab/5zW+aG5u4SMXsldTUVIIie2Ces4zzX96J7i7m8NZJFGQnMu0PY4gJj6DbPBeleSbM4d3aOGohRHvkcXhY/dqjAFwz0wPB9R+jOQspPvAapYffAc2Njm8t+vqcOHGC1Oh+9bZzFjtJf2EHmruGkeY6jHlotCxwL4SoIjU1FdUWy5mCglrbSCkdIURH4ynNZnCvEFJ/tJCDn/4fQSZfEjmk6wi69YnCU5qNMTi+jaMUQojG0zSNo3lBFDuNFH2ygoMHD2Iymbj55pvxFB3DGJKIYjC1dZhCCNEoTUqwJCUl1bkossvlqjH5IpqnIsEyYMAAANwFB9DdxSiWaE4fGIauqxjC+mCOisEcBRF1L9MihBANonudlB5+h+L9r6G7iwHI1xL4fK+V+QMyA1aaxxJqYeS9wzn8RSbmIBNZW88AF0qDJU3u0fyTEUJ0KjWX0qlqgZTSEUJ0IJ7SbM6tmO+vVJAQcWGf69w2cr+8HVQzsfPekSSLEKJDSU9P/25mXrJvw56vABg+fDgx0ZHkfHwbaF6ip7+AKVxmHwshOo4mJVh+9atf8dhjj3H55ZczdOjQKvt27tzJ888/z7PPPtukgLKysnjyySf57LPPKCsro0+fPixevNg/DVrXdX7961/z0ksvYbfbmTBhAv/+979JSUlp0ut1FB6Ph4MHDwIwcOBAAIyhSdh6zEKxJaDrvgcHZ7eeJ2WulAUTQtRN0zQKzPk4VSeHj2QyJDKq1geQ9vS/Un78UwDOlQax4kAMh3ODAYXT+YEtzdNrZk96zkimNKfUn2Ax2qQ0mBCidheX0qkQFRXFggULpJSOEKJD0Zz2GstAV23k8rWTBIsQooNIT0+vdUDMli1bmNDfRrQjH9USgTFUBtYJITqWBiVYHn744Wrb4uLiGDFiBOPHj6dPH99UicOHD7Np0yYGDhzI5s2bWbBgQaOCKSgoYMKECUybNo3PPvuMLl26cPjwYSIjL9RefOqpp3juued4/fXX6dmzJ7/85S+ZPXs2+/fvx2rtvKVjMjMzcTqdhIWFkZCQAIApsi+R435PUVYu8CUAW57bRfaeAsY/NgaD2dCGEQsh2qv09HSWLlmCPcr3IHLfC98S+fbbLFy40P8g0utxkXn0OFu3buXE/tNck2pi9ZEu7MwKx2oLYuLEEQwfPpw333yz2aV58o7kE9Ej3H/NUhSlSimwUT8YIaXBhBB1GjlyJP379mLHm9dT7DTSfeqfGThkuMxcEUIIIYRoY5qmsWzZsjrbnPv2HaJjwJZ4GYrapLHgQgjRZhp01frnP/9Z674NGzawYcOGKtu+/fZb9u7dyz/+8Y9GBfPXv/6VxMREFi9e7N/Ws2dP/9e6rvPss8/yi1/8gquvvhqAN954g7i4OD744ANuuummRr1eR7J//37AVx6s8sMCXdfZ/uK3Vdqe2XYG1SQPFIQQ1e3ctJL333qVICAorNIObznvv/kMJafHEVW+kVPnynl3V7R/97/sgxg2fASPzB/FgAEDMBp9bx9er7dZpXlKsktY9dPVBMcGM+33UwmKslVr02OClJwUQtRPVVV6RZcBEJXaR5IrQgghhBDtQEZGBgV1DMozG7z0jswFwJY8t7XCEkKIgGlQgkXTGlY/v7k++ugjZs+ezfz581m3bh3du3fngQce4Pvf/z4Ax44dIzs7m5kzZ/qPCQ8PZ8yYMWzatKnGBIvT6ayyXkxRUREAbrcbt9vdwmfUfBUx7tu3D4C0tDRcjhJK97+MNWkeZ3YbydqaXfWYUjdH1x6jx8T29VCy4lw6ws+9LnIe7UtnOQ9o+rl4PB7/cXUd6y45S/TxX/Hg+NrXTNEdx1BUCO+isDo0nr4DRzJixAj69euHyeRbbFDXdf/rDBkyhPvvv5+3li/HXnihNE9kZCQ33ngjQ4YMqRZT5fMsyilGURVUk4rBpvr3edyeKu31OtZ5aUud5e9PzqN96SznAS1/XavptSq+NrTiz09+Z+2PnEf70lnOA1r+ulbRriH9KS3485TfWfsj59G+dJbzgJa/ruXl5dXZT/+4YswGHbexC4SmtNjPVH5n7Y+cR/vSWc4Dmn4uTT13Rdf1dvPUqqLE1+OPP878+fPZtm0bjzzyCC+88AK33347GzduZMKECZw5c4auXbv6j/ve976Hoii8/fbb1fr8zW9+w29/+9tq25ctW0ZQUFDLnUwAOZ1OPvvsMwDmzJlDd1MGvdyfUOyIY9v/FkBNJXrNELzAhmpTWjdYIUSry83N5Z577qn3uubMO8DE4Pfr7e9kUQRHmEFQdB8MhoaVGvS6Sok9/xLFTiOng66hS1x3FKVh1x+txJfEV0Mqzc5z65S8Ug5AyN02FJNcy4S4lDT0ulaZ5i5jjPdZALYYHkU1dYz7PCHEpaGh17UgLZsBrlfr7W+f+S7KVFmDRQjRdhp6XTt//ny1yjeV3THyBCkxpWQ4R1AYPrslQhVCiAYpKytj4cKFFBYWEhYWVv8B32lWYcNjx47x2WefceLECQCSkpKYO3dulbJejaFpGiNHjuRPf/oTAMOGDWPv3r3+BEtT/PSnP+Xxxx/3f19UVERiYiKzZs1q1A+qrbjdbn/JtISEBK677joK1t6POw9O7Z2P4lHQqZ4jUzwKkRlRTPjpuNYOuVZut5svv/ySyy67zD8SviOS82hfOst5QNPPJSsrC6De69qu9Q44d+H7nGMpHN46ldQxa4lNPuzfbuhzFzdOvK5RsTtK7RR/4SsLOXH297AGR9Ta1u12s3LlSmbNmlXreXocHt575QMAZs+ejdHaPuvwdpa/PzmP9qWznAe0/HWtMt916FkAZs6cWed1KNDkd9b+yHm0L53lPKDlr2vugkPY19SfYJkwYQKmyL4Nfv3Gkt9Z+yPn0b50lvOAlr+uaZrG/v37aywTFmz20Du6FIDRV/wfptCERkbfcPI7a3/kPNqXznIe0PRzqah81VhNfmL1ox/9iH/84x/Vyoepqsqjjz7KM8880+g+u3btSv/+/ats69evH++99x4A8fG+ETo5OTlVZrDk5OQwdOjQGvu0WCxYLJZq200mU4f5Yzl3zvdEdNCgQVByDE/+XkoLu5C91wg1JFcAdE3n9KYsSs+UEZEU3orR1q8j/ezrIufRvnSW84DGn0vFeij1HRceHu5PsLjKbRzaOAOP28LBjTOIiDuN2Vbub9fYn6W3Uvv64ig+U0LZuw6KUoqJHxBXYxvFe2HGislkwmhqnwmWCp3l70/Oo33pLOcBLXddq6wx16GWcin/ztorOY/2pbOcx/9v787joqzX/4+/ZxgYNnFANlFB3HA3twg1NcXEynLJTKm0b5uWVpbtZS59s73ztbTl2NE6ZZZltqm5paWZpbmkqYEbaS65ALIzzP37wx+TE6CALAO+no8Hj4dz35/7c18Xw1ydw8X9+UiVV9cMS+n+N4/FYqmS7+XF/J65K/JwL7UlD6ly//fayJEji927MzPPotfXNdGoQZ0VEVS+P9Yuq4v5PXNX5OFeakseUtlzKW/e5dr98+WXX9arr76qIUOGaP369UpNTVVqaqrWr1+v66+/Xq+++qpeffXVMs/bvXt37d692+XY77//rqioKElnNrwPDw/XypUrnefT09O1YcMGxcW5z5MaFckwDGeDpW3btsrcs0iSVK9NOzXq1lAmc/HL5pjMJjXq3tDtmisAqk9hLTUMadf6vrLbvSSZZM/30u4f+xQZVxYmi4+eWNpaTyxtLZOl6Eb1Z9v23q9ynDC07b1f5UarVAIAAFQ7s9Ummb3OM8jrzDgAqCG6dOmie+65Rzab6++ogoKCdP2oB9Wq78RqigwALly5/iT43//+t6699lp9/PHHLsdjY2M1f/585eTk6K233tKECRPKNO+ECRPUrVs3Pfvss7rhhhv0008/6e2339bbb78tSTKZTLr//vv1zDPPqHnz5oqOjtZTTz2liIgIDRo0qDypuL0///xTOTk58vT0VLPoBjqx+MxeLP7Nhyi2bTsd2XpU+Zn/2IDHJHn6WhQ7rms1RAzAXRXuiXJsfwsdT2n29wnDrL8ONNfRfc0VFp1U6r1Tyqvr+C46evKIYidcWun3AgAAqEksfuEKvWqBHLmpMgpydWLVnZKken3elsnjzMoMZqtNFj/2XwFQs3Tp0kWtY5rol/8O1elcixr0nq62HTrJbC7X334DgNsoVxXbv3+/+vcveeOp/v37a//+/WWet2vXrvrss8/04Ycfqm3btpo2bZr+9a9/KTEx0Tnm4Ycf1vjx43XnnXeqa9euysjI0NKlS+Xt7V2eVNzeb7/9Jklq0aKF8v/8VoY9Sx7+jeQV2lneNm/Fji+miWJIseMvlbetdn5PAJTfmaXB+qjo8oKGdq/vq7zscz99UhG8/Dzl3dsq33qVfy8AAICaxuIXLq+glvIMbOE85hnYQl5BLeUV1JLmCoAay2w2q0m9LHWISFcj/ai0DZOVd3JndYcFABekXE+whIaGauvWrSWe37p1q0JCQsoV0DXXXKNrrrmmxPMmk0lTp07V1KlTyzV/TbNjxw5JUps2bSRHpsxWm/yaDpbJdKY3FtUzUvu/PaCDG85sLmYym9QwroGiekZWW8wA3JNhGC5Lg7n6e6mwiGsr/t5pB9N1+tBpNYxtUPGTAwAAAABqlJyUJTJyjsu7YW8pqFV1hwMA5VauJ1iGDRum2bNn67nnnlNmZqbzeGZmpp5//nnNnj1bw4cPr7AgL1Z5eXlKSkqSJLVu3Vp+za9X2MAv5dtsqHNMTmqOwjv9/RdMFh+WBgNQvNPHPM8sDWaUUPr//1Jhp49V7GZm9hy7vv/ftVo9+Tslf7OnQucGAAAAANQ8Rs5xmTzryDuie3WHAgAXpFxPsEybNk1btmzR448/rkmTJikiIkLSmf1C7Ha7rrjiiovmCZPKlJSUpPz8fHl7e6t+/fqSJJOHl8vfnR/+5Yg2vrHJ+brr2M4sDQagWPVaNlXDS/fp0MbjMhxFz5vMUoOuwQpsFl2h9zWZTQprH6actFxFdI0o1TUWb4tuWjKiQuMAULt5Wa3F/hsAAADVy555RI7cVNlzTrsct4Z2Un7aPvaWAlCjlavB4uvrq5UrV+rzzz/XkiVLdODAAUlSQkKCrrrqKg0cOJCNiyvAr7/+KklqGB6o/GM/y7NBnHNpsEJmD7MCm9h0am+qJCmye6OqDhNADWEymXTZhMv1+e1fKT8z/x8nJU9fT106rru+unuJwtuHqsOo9rLWufBfUnp4eajr2M5qn9hW1gCr8vPzz38RAAAAAKDGs2ce0bHFwyRHXpFzOYfWKOfQGsnspdCrFtBkAVAjlbnBkpWVpZtuuklDhw5VYmKirrvuusqIC/p7/5WODU8rbd0DyonooXqXv+wypnHvKDW8rIHmD15QHSECqGG8bd6KHd9Va5/7wfWEIcWOv1Qnk08p/Y905Z3OVafbO5Z6XnuOXX2PXOn8t9VqVe7pPHn5ezob7tYA/qIcAAAAAC4mjtzUYpsrroPyzoyjwQKgBirzHiy+vr5asWKFsrKyKiMe/H+pqak6ePCgTCapuf9eSZJ3+GXVHBWA2iCqZ6TqdwlxvjaZTWrUvaGiekaqQWyE+j3fV13v7iKL9989+B2f7FRaSlqp5j/045+y59i1bOJy/fDSj8rP5okVAJXPbPFRxPANihi+QWaLT3WHAwAAAAC4CJRrk/sePXpo/fr1FR0LzrJ9+3ZJ0qUtA+Sn45KHVT5RCS5jCvIKZBhGdYQHoAYzmUy65NbWztcWbw/FjuvqPBfWPlRRl0c6zx/ffUKb39mir+9ZqpzUnGLnzE3Ldf5787+36o8fDir94Gkd2XJU9hx7JWUCAAAAAAAAVJ9y7cHy+uuvq3///nryySc1ZswYNWzYsKLjuugVLg92WVS6JMm7YbzMXnVcxmybt117lu1Vm2Gtqjw+ADVbndC/60mXMe3lbfMucaynr6cadWsoTz9Pl3FZJ7LkW89XhmFo8+ytzuP2bLtSfvhD/Z7vI8kkn0D+khwAAAAAAAC1T7kaLB06dJDdbtf06dM1ffp0WSwWWa2ua+ubTCalpZVuORm4cjgc2rFjh3w87Qox7ZUMyTt6UJFxx3cdV86pHJk9Pao+SAA12tnL50Rd3vScY+s2ClCvpy6Xo8DhPJaTmqPPb/tKoW1DFNUzUod/PuI8ZzgM/bHuoBr3jFJUz8jipgQAAAAAAABqvHI1WIYOHerctBhl57Bn68invSVJ4UNXF1knPCUlRadPn1avZpkyGXZlmsIVHNiyyDxXTOmlE7+flF+Ir36ugrgBXNzMHn+vKnn012Ny5DuUfTJHm97eXHSwSdrw2k8Kax96zqdjAAAAAAAAgJqqXA2WuXPnVnAYOFvh8mBtGp5pYv3l0VGNi2loWawWhbULZX8DAFUu6vJIBTYN1IYZPyntQDFPKxpSfpZdG17/Wb2evLzqAwQAAAAAAAAqWZkaLDk5Ofr888+1b98+BQcH6+qrr1b9+vUrK7aLVuEG9+kNxqpNG5tObNhbzREBQFGOvAId3XqsxPOFS4WlHkiTLapuFUYGAAAAAHAHZqtNMntJjrxzDPI6Mw4AaqBSN1iOHTumbt26ad++fTIMQ5Lk6+urRYsWKT4+vtICvNjk5OQoKSlJktS2XTt5BgXJYTpUZNz2j3+TSVLj3lGyBliLnAeAylY3qq4adWuogz8ekuEwipw3mU1qGNeA5goAAAAAXKQsfuEKvWqBHLmpMgpydWLVnZKken3elsnjzO+zzFabLH7h1RkmAJRbqRss06ZN0/79+zVhwgT16dNHycnJmjZtmu666y7t2bOnMmO8qOzevVseylNoWIhCQ0Nltxdd/sswDO36bJdyUnMV2jZEfqF+umnJiGqIFsDFzGQyKXZ8Vx3ZelT5WfnS2T0Wk+Tpa1HsuK7VFh8AAAAAoPpZ/MIlv3A57NnOY56BLYrsSQwANVGpGyzLli3TLbfcopdeesl5LCwsTCNHjtTu3bsVExNTKQFebLZv367Lok6pT/N9ykz6WNboIUXGGAWGWg9rreM7jyuoWVA1RAkAZ3jbvBU7vqvWPveD6wlDih1/KRvcAwAAAAAAoNYqdYMlJSVFjzzyiMuxHj16yDAMHT16lAZLBdmx/VeNbHFKnqZ8mT39ix1jtpjVekjLKo4MQG1i8bZU2JNvUT0jtf/bAzq44cxyhoVLg0X1jKyQ+QEAAAAAAAB3ZC7twNzcXHl7u/4lcuHr4paxqk0c9mz9+VGs/vwo1uVxxop2/Phx1bEnK8g3XyZPf/k06ltp9wKAimIymdRlTCfna4sPS4MBAAAAAACg9iv1EyyStH//fv3yyy/O12lpaZKkpKQk2Wy2IuM7depU5FhNlJeb6/Jv70paI3LHjh26tNEpSZJv9DUyWbyl/Pwi447t+EuBTWzy9PGslDgAoKzOXgqs69jOLA0GAAAAAACAWq9MDZannnpKTz31VJHjd999t8trwzBkMplUUFBwYdFdZJJ/26D+oRmSJN+mg4sdk5ueq2UTV8jkYdKwj4bIy8+rKkMEgPOK7N6oukMAAACo8cwWH0UM31DdYQAAAOAcSt1gmTNnTmXGcdFzOBzyO/2DzEFSgX8reQY0LnZcxrFM+Yb4yuJtobkCwG1U5J4uAAAAAAAAQE1Q6gbLqFGjKjOOi96+vXvUPvQvSVJQm5J/SVmvWZCGvHed8jLzqio0AAAAAAAAAADwD6Xe5B4V5597ukjS9h2/afZPjbUzq718G11x3jl4egUAAAAAAAAAgOpDg8VN7NixQ6eyvWRpcpNMHjRPAAAAAAAAAABwZzRY3EBmZqb27NkjSWrTpk2J4/7aeVxfjV2szXO3VlVoAAAAAAAAAACgGDRY3MDBdS9qRIcD6tDEV8HBwSWOO7b9mFL3pyn9j/QqjA4AAAAAAAAAAPxTqTe5R+UwHPnyOrlarcOyVRAces6xTa9sqoAGAfLyZwkxAAAAAAAAAACqEw2WapZ3+AdZTdlKz7EovMM15xzrXdeqRt0aVlFkAAAAAAAAAACgJCwRVs1O71kkSfrlzyDFtCx5/xUAAAAAAAAAAOA+eIKlCtkzj8iRmyp7zmnnMXP6b3IYUoFvU1kKUiWFF3vtXzv+UuaRbNW/JEz+4f5VEzAAAAAAAAAAACgWDZYqYs88omOLh0mOvCLnzCapb/h6HVs8TKFXLZDFr2iTZd+KA9q3cr/a3NBKHW+9pAoiBgAAAAAAACqG2eKjiOEbqjsMAKhQLBFWRRy5qcU2V1wH5Z0ZVwxbU5tC24UqrH1YhccGAAAAAAAAAADKhidYzqG4Jb3saUnKy60jSTJbbcU+bVIZWlzTTG0Gt6qSewEAAAAAAAAAgHOjwVKCkpb0Sv9+3N8vzF4lLul1NkdehuypyZURJgAAAAAAAAAAqAY0WEpQpiW9/tFgKcg6qozd82RP36f8tH1yZB+7sFiyDDkKDMnzgqYBAAAAAAAAAAAVhAbLBcrY+Z4ceanybtBT/i1ulCQZjgJl/j7fZZzJapNRwv4q55P9Ta4+W/C5ej3ZQ/U71b/QkAEAAAAAAACgRrDb7crLO88fwleT/Px8eXp6KisrS56eNfev42tLHlLJuXh5ecliqfh2CA2WC5RzcKUkyWwNlP5/g8XDL1x+MYmyBDSWZ0C0LAGNZc84pOPLR5V5fkeBQ0aaQ/Ych/zD/Ss0dgAAAAAAAABwR4ZhKCUlRcePH6/uUM4pLCxMyck1f3uI2pKHVHIuwcHBioyMlMlkqrB70WC5QD6Nr5I1tIs8A2Ocx0wms+pecq/LOMMwSjXfP8eZPczyu8VH3dp0l399GiwAAAAAAAAAar/C5kqDBg3k7+8vs9lc3SGhhnI4HMrIyNChQ4ckSVFRURU2Nw2WC+TXfLi8gloWey4/P1+HDx/WoUOHlPTrD+rlb5KnR8mNlvwCk/YfOqmYeq7HTWaTAqNtFdpZAwAAAAAAAAB3ZLfbnc2V8PDw818AnIe//5mHFw4dOqQGDRpU2HJhNFgqgN1u19GjR3Xo0CGXr2PHjsnhcDjH/eLdTH5e9hLnycyz6MYmHlURMgAAAAAAAAC4pcI9Vwp/KQ5UhMKfp7y8PBosla20S3rNnDVTO/Znq6CgoNjzvr6+atCggXx9fbV161al5Zx7kyCbzfZ3DA5D309bp1x7nvJ758uzbs3eYAgAAAAAAAAASotlwVCRKuPniQZLCfYfOinvgvMv6XXg0CkVFHjK29tbERERatCggcuXzXZmaS+Hw6GJEyfq1KlTJc4XFBSkFi1aOF+n/ZGuP38+LFkkDy+ebAEAAAAAAACAssjNzdWYMWMkSW+++aasVms1R4TahAZLCU5leWj+9+df0iuu90D17dtX9erVO+ceKWazWSNHjtTMmTNLHDNixAiXLppPoLe6jOusbT9vk9lCtxYAAAAAAAAAAHfBb+1LYLPZlJbjqT/TfUr8SsvxVIcOHRQcHFyqDei7dOmie+65RzZbXZfjQUFBuueee9SlSxeX49YAq5peGS1rR5YGAwAAAAAAAICyOnuP7N27d7u8rgyjR4+WyWTSc88953J80aJFpfodcqHevXvLZDLJZDLJ29tbLVq00PTp00u9tQWqBk+wlKBFixYKDAws05JepdGlSxe1jmmiX/47VKdzLWrQe7radujEeoIAAAAAAAAAUIE2btyoDz74wPn61VdfVWBgoEaOHFnkj90rkre3t55//nndddddCgwMLPc8d9xxh6ZOnarc3FytWrVKd955p2w2m8aOHVuB0ZZNXl6evLy8qu3+7obf6pegcEmvc/nnkl5lmbtJvSx1iEhXixbNip0j93SuUtb9oZzUnDLPDwAAAAAAAAAXs40bN2rmzJlKTU11OX7q1CnNnDlTGzdurLR7x8fHKzw8XNOnTy9xzKeffqo2bdrIarWqcePGevnll4uM8fX1VXh4uKKionTrrbeqffv2Wr58ufN8bm6uJk6cqAYNGsjPz0+xsbFavXq1JCk9PV0+Pj5asmSJy5yfffaZ6tSpo6ysLEnSH3/8oRtuuEE2m01BQUG67rrrtH//fuf40aNHa9CgQfrf//1fRUREKCYmRpI0a9YsNW/eXN7e3goLC9P111/vvMbhcGj69OmKjo6Wj4+POnTooE8++aTM38eagAbLOTiX9KpbuiW9KtLRbcf03TNrtXrS95V2DwAAAAAAAACoCQzDUG5ubqm+srOzXZ5cKc68efOUnZ1dqvnKuiyXh4eHnn32Wb322ms6ePBgkfObNm3SDTfcoBtvvFG//vqrJk+erKeeekpz584tMffvv/9eu3btcnl6ZNy4cVq/fr3mz5+vbdu2adiwYUpISFBSUpICAgJ0zTXXaN68eS5zffDBBxo0aJB8fX2Vn5+v/v37q06dOvr++++1bt06+fv7KyEhQXl5ec5rVq5cqd27d2v58uX66quvtHHjRt17772aOnWqdu/eraVLl6pnz57O8dOnT9d7772nN998Uzt27NCECRN00003ac2aNWX6PtYELBF2Hl26dFGLqEi9e/sC5Zpz1ffR7urQtWulL+llOAzZGtdVcOtgHdXhSr0XAAAAAAAAALizvLw8jRkzpsLmO3XqlO6+++5SjX3zzTdltVrLNP/gwYN1ySWX6Omnn9Y777zjcu6VV15R37599dRTT0k6s13Fb7/9phdffFGjR492jps1a5Zmz56tvLw85efny9vbW/fee68kKSUlRXPmzFFKSooiIiIkSRMnTtTSpUs1Z84cPfvss0pMTNTNN9+srKws+fr6Kj09XV9//bU+++wzSdJHH30kh8Oh2bNnO/eHmTNnjmw2m1avXq0rr7xSkuTn56fZs2c7mzsLFy6Un5+frrnmGtWpU0dRUVHq2LGjpDNP1Tz77LNasWKF4uLiJElNmjTR2rVr9dZbb6lXr15l+j66O7d6gmXy5MnOjXsKv1q2bOk8f/bGPoVfFfmhKonZbFZgXpDCc+qrebOmVbJfStTlkbrmjavU6c5LKv1eAAAAAAAAAICK9fzzz+vdd9/Vzp07XY7v3LlT3bt3dznWvXt3JSUlqaCgwHksMTFRW7Zs0bp16zRgwAA98cQT6tatmyTp119/VUFBgVq0aCF/f3/n15o1a7Rnzx5J0lVXXSVPT0998cUXks4sSxYQEKD4+HhJ0tatW5WcnKw6deo4rw8KClJOTo5zDklq166dy5Mz/fr1U1RUlJo0aaKbb75ZH3zwgXPJseTkZGVlZalfv34ucb333nsuc9YWbvcES5s2bbRixQrna4vFNcTCjX0K+fr6VllsknRow1HFXBVcZfczmU1Vdi8AAAAAAAAAcEdeXl568803SzV29+7devXVV887bsKECc49Rc537/Lo2bOn+vfvr8cee8zlyZTSqlu3rpo1ayZJ+vjjj9WsWTNddtllio+PV0ZGhjw8PLRp0yZ5eHi4XOfv7++M+/rrr9e8efN04403at68eRo+fLjzd+4ZGRnq3LlzscuphYSEOP/t5+fncq5OnTr65ZdftHr1ai1btkyTJk3S5MmT9fPPPysjI0OS9PXXX6tBgwYu15X1KaCawO0aLBaLReHh4SWeL9zYpyrlpv293tyWd35TVLem8rZ5V9r9HAUOmT3c6uEiAAAAAAAAAKg2JpOp1L+gb9u2rQIDA3Xq1KkSxwQFBalt27aVvlrRc889p0suucSlkdOqVSutW7fOZdy6devUokWLIs2SQv7+/rrvvvs0ceJEbd68WR07dlRBQYGOHTumyy+/vMT7JyYmql+/ftqxY4dWrVqlZ555xnmuU6dO+uijjxQaGqqAgIAy5WWxWBQfH6/4+Hg9/fTTstlsWrVqlfr16yer1aqUlJRatxxYcdyuwZKUlKSIiAh5e3srLi5O06dPV2RkpPP8Bx98oPfff1/h4eEaOHCgnnrqqXM+xVK4EVGh9PR0SVJ+fr7y8/PPG49hGPrlnR3O1/Ycu36c8ZO6PxZXnvSc9z773x7/iOO3j3dqzzf7FDOouaITGhe5pqYqzKGm50Ie7qW25CGVPxe73e68riZ8H3jP3A95uJfakodEXauJaksu5OFeakseEnWtJqotuZCHe6kteUjUtZqotuRyvjwuND+z2ayRI0dq5syZJY4ZMWJElWwF0a5dOyUmJmrGjBnOYw8++KC6du2qadOmafjw4Vq/fr1ef/11zZo165xz3XXXXZo2bZo+/fRTXX/99UpMTNQtt9yil19+WR07dtRff/2llStXqn379rr66qslnXmKJjw8XImJiYqOjlZsbKxzvsTERL344ou67rrrNHXqVDVs2FAHDhzQwoUL9fDDD6thw4bFxvHVV19p79696tmzpwIDA7V48WI5HA7FxMSoTp06mjhxoiZMmCCHw6EePXooLS1N69atU0BAgEaNGlUB39ULU1ztKu/PnMkwDKMigqoIS5YsUUZGhmJiYnT48GFNmTJFhw4d0vbt21WnTh29/fbbioqKUkREhLZt26ZHHnlEl156qRYuXFjinJMnT9aUKVOKHJ83b16plhfLT7YrZ0VekePe8V7ybFa+/pQjP0uxBf+SJG3wuF9mT9c4shbnqCDFIWt3T3m18yzXPQBcPI4fP67bb7+91HUNANwddQ1AbUNdA1DbUNdQ2Tw9PRUWFqZWrVpd0M/Yxo0b9cEHHyg1NdV5LCgoSCNGjFCXLl0qINKiRo8erdTUVC1atMh5bP/+/YqJiVFeXp4Kfx3/6aefatKkSUpKSlL9+vU1fvx4TZw40XlN7969dckll+hf//qXy/xjxozR2rVrtW3bNhUUFOiZZ57Re++9p0OHDik4OFiXXXaZpkyZonbt2jmveeSRR/TCCy9o0qRJRX5XfuTIET3yyCNavHixTp8+rQYNGqhv37566aWXFBAQUGw+a9eu1ZNPPqlt27YpJydHzZs31xNPPKEbbrhB0pmHFmbMmKE33nhDe/fulc1mU6dOnfT444+rZ8+eFfSdLrusrCzt3LlTR48eLdJQycrK0siRI5WWllamp3ncqsHyT6mpqYqKitIrr7yi2267rcj5VatWqW/fvkpOTlbTpk2LnaO4J1gaNWqk48ePn/cblZOao8Vjv1F+5j+6VybJ09dTV73Rv1xLheVkpur0N9dIkur0/0refjaX8/nZdp38/YTqNKgjz7qeWr58ufr16ydPz5rdbMnPz68VuZCHe6kteUjlz+XQoUOKjo4uVV1zB7xn7oc83EttyUOirtVEtSUX8nAvtSUPibpWE9WWXMjDvdSWPCTqWk1UW3I5Xx5ZWVlKTk6+4AaLJGVnZ+vuu++WdGbPlapYFgzuqbDB0qxZsyI/V+np6QoODi5zg8Xtlgg7m81mU4sWLZScnFzs+cLHmc7VYLFarcWuzefp6XnOImQYhn55c4vs2fZiTkr2bLt+eWuLej1Z8vp2JSk4677FxeHp6SnfLmcevyrspJ0v3pqktuRCHu6ltuQhlT2Xwo3Jatr3oKbFey61JRfycC+1JQ+JulYT1ZZcyMO91JY8JOpaTVRbciEP91Jb8pCoazVRbcmlpDwqMrezmykxMTE0V1Di7+TLw60bLBkZGdqzZ49uvvnmYs9v2bJFklS/fv0Kv3fagTT98cPBEs8bDkN/rDuo1ANpskXVrfD7AwAAAAAAAAAujNVq1Zw5c6o7DNRSbtWumzhxotasWaP9+/frhx9+0ODBg+Xh4aERI0Zoz549mjZtmjZt2qT9+/friy++0C233KKePXuqffv2FR5L3ai6atStoUxmU7HnTWaTGnVvWK7mitdZT9R4/ePpml2f79ZvC3cp81hmmecFAAAAAAAAAABVw62eYDl48KBGjBihEydOKCQkRD169NCPP/6okJAQ5eTkaMWKFfrXv/6lzMxMNWrUSEOHDtWTTz5ZKbGYTCbFju+qI1uPlrAHi0Wx47pW+H13frZbmUczZYuqK79QvwqfHwAAAAAAAAAAXDi3arDMnz+/xHONGjXSmjVrqjAaydvmrdjxXbX2uR9cTxhS7PhLy7XBvSSZLT6KGL6hyHFHgUMtrmqmY78dV3DL4HLNDQAAAAAAAAAAKp9bNVjcUVTPSO1ZuVeHfz4i6czSYA0va6DMvzJ1ePMR1e8YXmH3MnuY1eaG1mpTYTMCAAAAAAAAAIDK4FZ7sLgjk8mkjrd3cL62+FgU2NSmX2Zv0Q8v/6i8zLxqjA4AAAAAAAAAAFQHGiylYK3790b0He/ooNZDWimoWaDaj2wrT1/PCrvPiaSTsufaK2w+AAAAAAAAAABQOVgirIwaXBYhi7dFA/6vv0xmU4XNm5+dr6X3L5PJbNKQ/15X7v1dAAAAAAAAAABnOOzZOvJpb0lS+NDVMlt8qjUe1C48wVJOZzdXHHaHclJzLmi+zKOZsta1yjvQm+YKAAAAAAAAAKBE+/fvl8lk0pYtW0p9zejRozVo0KBKi+liRIPlAp0+nKFvHlyu1VO+k6PAUe55bI1tGvrBIF31WkIFRgcAAAAAAAAAFx975hHlndyl/FO/O4/ln/pdeSd3Ke/kLtkzj1TKfUePHi2TySSTySQvLy81a9ZMU6dOld1e/q0himuMNGrUSIcPH1bbtm0rJFZPT09FR0fr4YcfVk7OhT1McDFhibALZDKblH7otEwmk9L/SJetsa38c5lM8j5rvxcAAAAAAAAAQNnYM4/o2OJhkiPP5fiJVXf+/cLspdCrFsjiF17h909ISNCcOXOUm5urxYsX65577pGnp6cee+yxMs1TUFAgk6n4bSo8PDwUHn7hsRfGmp+fr02bNmnUqFEymUx6/vnnL3ju8srLy5OXl1e13b8seIKlFCzeFq0MX6aV4ctk8XbtSfmH+annEz109awBF9RcAQAAAAAAAABcOEduapHmStFBeWfGVQKr1arw8HBFRUVp7Nixio+P1xdffKFXXnlF7dq1k5+fnxo1aqS7775bGRkZzuvmzp0rm82mL774Qq1bt5bVatX//M//6N1339Xnn3/ufNpk9erVRZYIKygo0G233abo6Gj5+PgoJiZG//d//1fqWBs1aqRBgwYpPj5ey5cvd553OByaPn26c94OHTrok08+cZ5r2LCh3njjDZc5N2/eLLPZrAMHDkiSUlNTdfvttyskJEQBAQHq06ePtm7d6hw/efJkXXLJJZo9e7aio6Pl7X1mC41PPvlE7dq1k4+Pj+rVq6f4+HhlZmY6r5s9e7ZatWolb29vtWzZUrNmzSrjO3XheIKlFKxWq+bMmVPi+fodL6xTeGrfKf3w0o+K6FJfHW+95ILmAgAAAAAAAIDaymHPLvGcyWSWyaN8KwSVNK/Z4lOu+c7m4+OjEydOyGw2a8aMGYqOjtbevXt199136+GHH3ZpDGRlZen555/X7NmzVa9ePdWvX1/Z2dlKT093/o46KChIf/75p2v8/7/ZsWDBAtWrV08//PCD7rzzTtWvX1833HBDqeLcvn27fvjhB0VFRTmPTZ8+Xe+//77efPNNNW/eXN99951uuukmhYSEqFevXhoxYoTmzZunsWPHOq/54IMP1L17d+c8w4YNk4+Pj5YsWaK6devqrbfeUt++ffX7778rKChIkpScnKxPP/1UCxculIeHhw4fPqwRI0bohRde0ODBg3X69Gl9//33MgzDeY9Jkybp9ddfV8eOHbV582bdcccd8vPz06hRo8rxLpUPDZYKdiL5pA5tOKT2ie1Kfc2x7X/p1N5Ueddlc3sAAAAAAAAAKMmRT3uXeM5av5vq9Xy1XPMe+2pQsU+0RAzfUK75JMkwDK1cuVLffPONxo8fr/vvv995rnHjxnrmmWc0ZswYlwZLfn6+Zs2apQ4dOjiP+fj4KDc395xLgnl6emrKlCnO19HR0Vq/fr0+/vjjczZYvvrqK/n7+8tutys3N1dms1mvv/66JCk3N1fPPvusVqxYobi4OElSkyZNtHbtWr311lvq1auXEhMT9fLLLyslJUWRkZFyOByaP3++nnzySUnS2rVr9dNPP+nYsWOyWs80v1566SUtWrRIn3zyie6888yybXl5eXrvvfcUEhIiSfrll19kt9s1ZMgQZ6OmXbu/f+f+9NNP6+WXX9aQIUOc+f7222966623aLDUVFkns7XswRUqyCuQrbFNkd0bleq6qMsj5V3XW56+vB0AAAAAAAAAUJMVNi3y8/PlcDg0cuRITZ48WStWrND06dO1a9cupaeny263KycnR1lZWfL19ZUkeXl5qX379uW678yZM/Wf//xHKSkpys7OVl5eni655JJzXnPFFVfojTfeUGZmpl599VVZLBYNHTpU0pmnSrKystSvXz+Xa/Ly8tSxY0dJ0iWXXKJWrVpp3rx5evTRR7VmzRodO3ZMw4YNkyRt3bpVGRkZqlevnssc2dnZ2rNnj/N1VFSUs7kiSR06dFDfvn3Vrl079e/fX1deeaWuv/56BQYGKjMzU3v27NFtt92mO+64w3mN3W5X3bp1y/6NuwD8Rr8C+Qb5qOXgGKX/ka7QtqGlvs7b5q2onpGVGBkAAAAAAAAA1HzhQ1eXeM5kKv+W46HXLCr3tf9U2LTw8vJSRESELBaL9u/fr2uuuUZjx47V//7v/yooKEhr167Vbbfdpry8PGeDxcfHp8SN7c9l/vz5mjhxol5++WXFxcWpTp06evHFF7Vhw7mfwPHz81OzZs0kSf/5z3/UoUMHvfPOO7rtttuc+8N8/fXXatCggct1hU+jSFJiYqKzwTJv3jwlJCQ4GyoZGRmqX7++Vq9eXeTeNpvNJY6zeXh4aPny5frhhx+0bNkyvfbaa3riiSe0YcMG5/fq3//+t2JjY4tcV5VosFSwDje3k8lsKteHAAAAAAAAAABQsorYE6Wy5z27aVFo06ZNcjgcevnll2U2n2kEffzxx6Waz8vLSwUFBeccs27dOnXr1k13332389jZT4iUhtls1uOPP64HHnhAI0eOVOvWrWW1WpWSkqJevXqVeN3IkSP15JNPatOmTfrkk0/05ptvOs916tRJR44ckcViUePGjcsUj8lkUvfu3dW9e3dNmjRJUVFR+uyzz/TAAw8oIiJCe/fuVWJiYpnmrGg0WCqY2cO1S5p9Mls+Qa4fTnuOXfMHL5Ak9XuhrzKOZCisfZj8w1y7dAAAAAAAAACAmq9Zs2bKz8/Xa6+9poEDB2rdunUujYhzady4sb755hvt3r1b9erVK3YZrObNm+u9997TN998o+joaP33v//Vzz//rOjo6DLFOWzYMD300EOaOXOmJk6cqIkTJ2rChAlyOBzq0aOH0tLStG7dOgUEBDj3OmncuLG6deum2267TQUFBbr22mud88XHxysuLk6DBg3SCy+8oBYtWujPP//U119/rcGDB6tLly7FxrFhwwatXLlSV155pUJDQ7Vhwwb99ddfatWqlSRpypQpuvfee1W3bl0lJCQoNzdXGzdu1KlTp/TAAw+UKecLUf5npnBOjgKHNs/ZokX/86VO7T1V4rj9qw9o/SsbtPOzXVUYHQAAAAAAAADUTmarTTJ7nWeQ15lxVaRDhw565ZVX9Pzzz6tt27b64IMPNH369FJde8cddygmJkZdunRRSEiI1q1bV2TMXXfdpSFDhmj48OGKjY3ViRMnXJ5mKS2LxaJx48bphRdeUGZmpqZNm6annnpK06dPV6tWrZSQkKCvv/66SOMmMTFRW7du1eDBg+Xj8/cDByaTSYsXL1bPnj116623qkWLFrrxxht14MABhYWFlRhHQECAvvvuO1111VVq0aKFnnzySb388ssaMGCAJOn222/X7NmzNWfOHLVr1069evXS3Llzy9xQulA8wVJJTGaTUg+kqyC3QAc3HFJgk8Bix9WJ8Fdwq2CFtSv9ni0AAAAAAAAAgOJZ/MIVetUCOXJTZRTk6sSqOyVJ9fq8LZPHmb1DzFabLH7hFX7vuXPnlnhuwoQJmjBhgsuxm2++2fnv0aNHa/To0UWuCwkJ0bJly4ocNwzD+W+r1ao5c+Zozpw5LmPObuL8M7aSYn300Uf16KOPOl/fd999uu+++4odW2js2LEaO3Zssefq1KmjGTNmaMaMGcWenzx5siZPnuxyrFWrVlq6dOk57zly5EiNHDnynGMqGw2WSmIymRR3/6X667fjatStYYnjWlzdXK2HtqrCyAAAAAAAAACgdrP4hUt+4XLYs53HPANbVNoeLrg40WCpRN4273M2VwAAAAAAAAAAlcds8VHE8A3VHQZqKfZgqSL2HLt+eWezclJzXI4bDqOEKwAAAAAAAAAAgLuiwVJF1r2wXr99sksbXvvZ5finiZ/p4E+HqikqAAAAAAAAAABQHjRYqki7xLbyD/dXzMDmLk+x2HMKZPFmpTYAAAAAAAAAAGoSGixVJKhpoK6dfbXCOoRp45u//H3CLO3+/PfqCwwAAAAAAAAAAJQZDZYqZPYw68B3KTq44awlwRzSHz8c1IHvUqovMAAAAAAAAAAAUCY0WKpQTmpOkT1YJEkmacNrP7ksHQYAAAAAAAAAANwXDZYqYhiGNrz2s+zZ9mJOSvlZdm14vZjmCwAAAAAAAACgXOw5dr0/4EO9P+BD2XOK+d0scAFosFSRtANp+uOHgzIcRrHnDYehP9YdVOqBtCqODAAAAAAAAABqv5R1f1R3COfUu3dv3X///c7XjRs31r/+9a9qiwfnR4OlitSNqqtG3RrKZDYVe95kNqlR94ayRdWt4sgAAAAAAAAAoHY6e1uGn9/YVOnbNIwePVomk6nIV3Jy8gXP3bhxY+d8vr6+ateunWbPnl0BUaO8aLBUEZPJpNjxXWXxsUj/7LGYJE9fi2LHda2W2AAAAAAAAACgtjEMQxvf/MX52p5dNds0JCQk6PDhwy5f0dHRFTL31KlTdfjwYW3fvl033XST7rjjDi1ZsqRC5i6vvLy8ar1/daLBUoW8bd6KHd9V+ucqYYYUO/5Sedu8qyUuAAAAAAAAAKhtDnyXooMbDjlfF27TcOC7lEq9r9VqVXh4uMvXbbfdpkGDBrmMu//++9W7d+8yzV2nTh2Fh4erSZMmeuSRRxQUFKTly5c7z6empur2229XSEiIAgIC1KdPH23dulWS9Pvvv8tkMmnXrl0uc7766qtq2rSp8/X27ds1YMAA+fv7KywsTDfffLOOHz/uPN+7d2+NGzdO999/v4KDg9W/f38ZhqHJkycrMjJSVqtVERERuvfee53X5ObmauLEiWrQoIH8/PwUGxur1atXlyl3d0SDpYpF9YxUw9gGzteFS4NF9YysxqgAAAAAAAAAwP3Zc+yy59hlGH//FXtBfoHsOXYV5BU4j+Wk5mjDa8U8rWKSfpzxkzKPZRY/71l7aDvsjopPoII4HA59+umnOnXqlLy8vJzHhw0bpmPHjmnJkiXatGmTOnXqpL59++rkyZNq0aKFunTpog8++MBlrg8++EAjR46UdKZB06dPH3Xs2FEbN27U0qVLdfToUd1www0u17z77rvy8vLSunXr9Oabb+rTTz/Vq6++qrfeektJSUlatGiR2rVr5xw/btw4rV+/XvPnz9e2bds0bNgwJSQkKCkpqRK/S5WPBksVM5lM6jKmk/O1xYelwQAAAAAAAACgNOYPXqD5gxcoNy3Xeey3T3Zp/uAF+vmNTZLOLA224bWflZ+ZX3QCQ8rPzNc3Dy53OfzZqC80f/ACpf2R7jy2Z/necsX41Vdfyd/f3/k1bNiwcs1TnEceeUT+/v6yWq26/vrrFRgYqNtvv12StHbtWv30009asGCBunTpoubNm+ull16SzWbTJ598IklKTEzUhx9+6Jzv999/16ZNm5SYmChJev3119WxY0c9++yzatmypTp27Kj//Oc/+vbbb/X77787r2vevLleeOEFxcTEKCYmRikpKQoPD1d8fLwiIyN16aWX6o477pAkpaSkaM6cOVqwYIEuv/xyNW3aVBMnTlSPHj00Z86cCvveVAcaLNXg7KXAuo7tzNJgAAAAAAAAAFBB0g6k6Y8fDp5zTNbxbKUeSKuU+19xxRXasmWL82vGjBkVNvdDDz2kLVu2aNWqVYqNjdWrr76qZs2aSZK2bt2qjIwM1atXz6XBs2/fPu3Zs0eSdOONN2r//v368ccfJZ15eqVTp05q2bKlc45vv/3W5frCc4VzSFLnzp1d4ho2bJiys7PVpEkT3XHHHfrss89kt9slSb/++qsKCgrUokULl3nXrFnjMmdNZKnuAC52kd0bVXcIAAAAAAAAAFAj3PjZmadBPKwezmOtr2+pVoNjZDKbJEl1o+qqUbeGOvjjIZclvwqZzCY1iI2QLaqu89jgd689M6/X3/M27dekXDH6+fk5mx6FzGazy7JmkpSfX8wTNucRHBysZs2aqVmzZlqwYIHatWunLl26qHXr1srIyFD9+vWL3dvEZrNJksLDw9WnTx/NmzdPl112mebNm6exY8c6x2VkZGjgwIF6/vnni8xRv359lxzP1qhRI+3evVsrVqzQ8uXLdffdd+vFF1/UmjVrlJGRIQ8PD23atEkeHh4u1/n7+5f5e+BOaLAAAAAAAAAAAGoEi3fRX2l7eHpInn+/NplMih3fVUe2HlV+Vr50dl/DJHn6WnTZvZeed16zpeIWgAoJCdH27dtdjm3ZskWenp4lXHF+jRo10vDhw/XYY4/p888/V6dOnXTkyBFZLBY1bty4xOsSExP18MMPa8SIEdq7d69uvPFG57lOnTrp008/VePGjWWxlK194OPjo4EDB2rgwIG655571LJlS/3666/q2LGjCgoKdOzYMV1++eXlTdctsUQYAAAAAAAAAKBW8bZ5K3Z8V9fmiiQZUuz4S6t824Y+ffpo48aNeu+995SUlKSnn366SMOlPO677z59+eWX2rhxo+Lj4xUXF6dBgwZp2bJl2r9/v3744Qc98cQT2rhxo/OaIUOG6PTp0xo7dqyuuOIKRUREOM/dc889OnnypEaMGKGff/5Ze/bs0TfffKNbb71VBQUFJcYxd+5cvfPOO9q+fbv27t2r999/Xz4+PoqKilKLFi2UmJioW265RQsXLtS+ffv0008/afr06fr6668v+HtQnWiwAAAAAAAAAABqnaiekWoY28D52mQ2qVH3horqGVnlsfTv319PPfWUHn74YXXt2lWnT5/WLbfccsHztm7dWldeeaUmTZokk8mkxYsXq2fPnrr11lvVokUL3XjjjTpw4IDCwsKc19SpU0cDBw7U1q1bnZvbF4qIiNC6detUUFCgK6+8Uu3atdP9998vm80ms7nkdoLNZtO///1vde/eXe3bt9eKFSv05Zdfql69epKkOXPm6JZbbtGDDz6omJgYDRo0SD///LMiI6v+vahILBEGAAAAAAAAAKh1TCaTuozppIMbDkmSLD4WxY7rWqn3nDt3bonnpkyZoilTppR4/p97p+zfv/+crwstXbrU+e86depoxowZmjFjxjnj/Oijj/TRRx8Ve6558+ZauHBhqeOUpEGDBmnQoEElXuPp6Xne/GsiGizVwOJt0U1LRlR3GAAAAAAAAABQq529FFjXsZ2rfGkw1G40WAAAAAAAAAAAtRJ/7I7KxB4sAAAAAAAAAAAAZUSDBQAAAAAAAAAAoIxosAAAAAAAAAAA3I7D4ajuEFCLVMbPEw0WAAAAAAAAAIDb8PLykiRlZGRUcySoTQp/ngp/vioCm9wDAAAAAAAAANyGxWJRcHCwDh06JEny9/eX2cyzAigfh8OhjIwMHTp0SMHBwbJYKq4tQoMFAAAAAAAAAOBWIiMjJcnZZAEuVHBwsPPnqqLQYAEAAAAAAAAAuBWTyaSoqCg1aNBAeXl51R1OsfLz87Vu3Tp1795dnp6e1R1OudWWPKSSc/Hy8qrQJ1cKuVWDZfLkyZoyZYrLsZiYGO3atUuSlJOTowcffFDz589Xbm6u+vfvr1mzZiksLKw6wgUAAAAAAAAAVCKLxVIpvxivCPn5+crPz5evr2+NbkzUljykqs/F7Raua9OmjQ4fPuz8Wrt2rfPchAkT9OWXX2rBggVas2aN/vzzTw0ZMqQaowUAAAAAAAAAABcjt2v9WSwWhYeHFzmelpamd955R/PmzVOfPn0kSXPmzFGrVq30448/6rLLLqvqUAEAAAAAAAAAwEXK7RosSUlJioiIkLe3t+Li4jR9+nRFRkZq06ZNys/PV3x8vHNsy5YtFRkZqfXr15fYYMnNzVVubq7zdXp6uqS/HxVyd4Ux1oRYz6e25EIe7qW25CGVPxe73e68riZ8H3jP3A95uJfakodEXauJaksu5OFeakseEnWtJqotuZCHe6kteUjUtZqotuRCHu6ltuQhlT+X8uZuMgzDKNeVlWDJkiXKyMhQTEyMDh8+rClTpujQoUPavn27vvzyS916660uzRJJuvTSS3XFFVfo+eefL3bO4vZ1kaTZs2fL19e3UvIAgKp0/Phx3XvvvdQ1ALUGdQ1AbUNdA1DbUNcA1DZZWVm6/fbblZqaqrp165b6OrdqsPxTamqqoqKi9Morr8jHx6dcDZZ/PsFy6NAhtW7dulLjBgAAAAAAAAAANcsff/yhhg0blnq82y0RdjabzaYWLVooOTlZ/fr1U15enlJTU2Wz2Zxjjh49WuyeLYWsVqusVqvztb+/v/744w/VqVNHJpOpMsOvEOnp6WrUqJH++OMPBQQEVHc4F6S25EIe7qW25CGVP5eCggIlJyerWbNm8vDwqMQIKwbvmfshD/dSW/KQqGs1UW3JhTzcS23JQ6Ku1US1JRfycC+1JQ+JulYT1ZZcyMO91JY8pPLnYhiGTp8+rYiIiDLdz60bLBkZGdqzZ49uvvlmde7cWZ6enlq5cqWGDh0qSdq9e7dSUlIUFxdX6jnNZnOZOlDuIiAgoMb/cBeqLbmQh3upLXlI5cula9eulRRN5bnY3zN3RB7upbbkIVHXaqLakgt5uJfakodEXauJaksu5OFeakseEnWtJqotuZCHe6kteUjly6UsS4MVcqsGy8SJEzVw4EBFRUXpzz//1NNPPy0PDw+NGDFCdevW1W233aYHHnhAQUFBCggI0Pjx4xUXF1fiBvcAAAAAAAAAAACVwa0aLAcPHtSIESN04sQJhYSEqEePHvrxxx8VEhIiSXr11VdlNps1dOhQ5ebmqn///po1a1Y1Rw0AAAAAAAAAAC42btVgmT9//jnPe3t7a+bMmZo5c2YVRVT9rFarnn76aZd9ZGqq2pILebiX2pKHVLtyOZfalGdtyYU83EttyUOqXbmcS23Ks7bkQh7upbbkIdWuXM6lNuVZW3IhD/dSW/KQalcu51Kb8qwtuZCHe6kteUhVn4vJMAyjSu4EAAAAAAAAAABQS5irOwAAAAAAAAAAAICahgYLAAAAAAAAAABAGdFgAQAAAAAAAAAAKCMaLAAAAAAAAAAAAGVEg6UaTJ48WSaTyeWrZcuWzvM5OTm65557VK9ePfn7+2vo0KE6evSoyxwpKSm6+uqr5evrq9DQUD300EOy2+2VHvt3332ngQMHKiIiQiaTSYsWLXI5bxiGJk2apPr168vHx0fx8fFKSkpyGXPy5EklJiYqICBANptNt912mzIyMlzGbNu2TZdffrm8vb3VqFEjvfDCC1Wax+jRo4u8RwkJCW6Xx/Tp09W1a1fVqVNHoaGhGjRokHbv3u0ypqJ+nlavXq1OnTrJarWqWbNmmjt3bpXm0bt37yLvyZgxY9wqjzfeeEPt27dXQECAAgICFBcXpyVLljjP14T3oryoa9VfD6hr7vVZoq65Tw7lRV2r/npAXXOvzxJ1zX1yKC/qWvXXA+qae32WqGvuk0N5Udeqvx5Q19zrs0Rdq6YcDFS5p59+2mjTpo1x+PBh59dff/3lPD9mzBijUaNGxsqVK42NGzcal112mdGtWzfnebvdbrRt29aIj483Nm/ebCxevNgIDg42HnvssUqPffHixcYTTzxhLFy40JBkfPbZZy7nn3vuOaNu3brGokWLjK1btxrXXnutER0dbWRnZzvHJCQkGB06dDB+/PFH4/vvvzeaNWtmjBgxwnk+LS3NCAsLMxITE43t27cbH374oeHj42O89dZbVZbHqFGjjISEBJf36OTJky5j3CGP/v37G3PmzDG2b99ubNmyxbjqqquMyMhIIyMjwzmmIn6e9u7da/j6+hoPPPCA8dtvvxmvvfaa4eHhYSxdurTK8ujVq5dxxx13uLwnaWlpbpXHF198YXz99dfG77//buzevdt4/PHHDU9PT2P79u2GYdSM96K8qGvVXw+oa+71WaKuuU8O5UVdq/56QF1zr88Sdc19cigv6lr11wPqmnt9lqhr7pNDeVHXqr8eUNfc67NEXaueHGiwVIOnn37a6NChQ7HnUlNTDU9PT2PBggXOYzt37jQkGevXrzcM40zxMpvNxpEjR5xj3njjDSMgIMDIzc2t1NjP9s/C6XA4jPDwcOPFF190HktNTTWsVqvx4YcfGoZhGL/99pshyfj555+dY5YsWWKYTCbj0KFDhmEYxqxZs4zAwECXXB555BEjJiamSvIwjDP/AbjuuutKvMYd8zAMwzh27JghyVizZo1hGBX38/Twww8bbdq0cbnX8OHDjf79+1dJHoZx5j8A9913X4nXuGMehmEYgYGBxuzZs2vse1Fa1DX3qgfUNff7LFHX3CuH0qCuuVc9oK6532eJuuZeOZQGdc296gF1zf0+S9Q198qhNKhr7lUPqGvu91mirlVNDiwRVk2SkpIUERGhJk2aKDExUSkpKZKkTZs2KT8/X/Hx8c6xLVu2VGRkpNavXy9JWr9+vdq1a6ewsDDnmP79+ys9PV07duyo2kTOsm/fPh05csQl9rp16yo2NtYldpvNpi5dujjHxMfHy2w2a8OGDc4xPXv2lJeXl3NM//79tXv3bp06daqKsjnzmFhoaKhiYmI0duxYnThxwnnOXfNIS0uTJAUFBUmquJ+n9evXu8xROKZwjsrOo9AHH3yg4OBgtW3bVo899piysrKc59wtj4KCAs2fP1+ZmZmKi4urse9FWVDX3KseFIe6Rl27ENQ16lp114PiUNeoaxeCukZdq+56UBzqGnXtQlDXqGvVXQ+KQ12jrl2ImlDXLOVJDBcmNjZWc+fOVUxMjA4fPqwpU6bo8ssv1/bt23XkyBF5eXnJZrO5XBMWFqYjR45Iko4cOeLyA1J4vvBcdSm8d3GxnR17aGioy3mLxaKgoCCXMdHR0UXmKDwXGBhYKfGfLSEhQUOGDFF0dLT27Nmjxx9/XAMGDND69evl4eHhlnk4HA7df//96t69u9q2beu8T0X8PJU0Jj09XdnZ2fLx8anUPCRp5MiRioqKUkREhLZt26ZHHnlEu3fv1sKFC90qj19//VVxcXHKycmRv7+/PvvsM7Vu3Vpbtmypce9FWVDX3KseFIe6Rl0rL+oadU2q/npQHOoada28qGvUNan660FxqGvUtfKirlHXpOqvB8WhrlHXyqsm1TUaLNVgwIABzn+3b99esbGxioqK0scff1xt/1GCqxtvvNH573bt2ql9+/Zq2rSpVq9erb59+1ZjZCW75557tH37dq1du7a6Q7kgJeVx5513Ov/drl071a9fX3379tWePXvUtGnTqg6zRDExMdqyZYvS0tL0ySefaNSoUVqzZk11h1XpqGvuj7pWfahrNRN1zf1R16oPda1moq65P+pa9aGu1UzUNfdHXas+1LWqwxJhbsBms6lFixZKTk5WeHi48vLylJqa6jLm6NGjCg8PlySFh4fr6NGjRc4XnqsuhfcuLrazYz927JjLebvdrpMnT7p1fk2aNFFwcLCSk5OdcbhTHuPGjdNXX32lb7/9Vg0bNnQer6ifp5LGBAQEVOj/aCkpj+LExsZKkst74g55eHl5qVmzZurcubOmT5+uDh066P/+7/9q3Htxoahr1LULRV1znzyoa2dQ16hrF4q65j55UNfOoK5R1y4Udc198qCunUFdo65dKOqa++RRk+oaDRY3kJGRoT179qh+/frq3LmzPD09tXLlSuf53bt3KyUlRXFxcZKkuLg4/frrry4FaPny5QoICFDr1q2rPP5C0dHRCg8Pd4k9PT1dGzZscIk9NTVVmzZtco5ZtWqVHA6H8wMdFxen7777Tvn5+c4xy5cvV0xMTJU8vlicgwcP6sSJE6pfv74zRnfIwzAMjRs3Tp999plWrVpV5JHJivp5iouLc5mjcEzhHJWdR3G2bNkiSS7vSXXnURyHw6Hc3Nwa815UFOoada28qGvUtarIoTyoa9S18qKuUdeqIofyoK5R18qLukZdq4ocyoO6Rl0rL+oade2Ccihm43tUsgcffNBYvXq1sW/fPmPdunVGfHy8ERwcbBw7dswwDMMYM2aMERkZaaxatcrYuHGjERcXZ8TFxTmvt9vtRtu2bY0rr7zS2LJli7F06VIjJCTEeOyxxyo99tOnTxubN282Nm/ebEgyXnnlFWPz5s3GgQMHDMMwjOeee86w2WzG559/bmzbts247rrrjOjoaCM7O9s5R0JCgtGxY0djw4YNxtq1a43mzZsbI0aMcJ5PTU01wsLCjJtvvtnYvn27MX/+fMPX19d46623qiSP06dPGxMnTjTWr19v7Nu3z1ixYoXRqVMno3nz5kZOTo5b5TF27Fijbt26xurVq43Dhw87v7KyspxjKuLnae/evYavr6/x0EMPGTt37jRmzpxpeHh4GEuXLq2SPJKTk42pU6caGzduNPbt22d8/vnnRpMmTYyePXu6VR6PPvqosWbNGmPfvn3Gtm3bjEcffdQwmUzGsmXLDMOoGe9FeVHXqr8eUNfc67NEXXOfHMqLulb99YC65l6fJeqa++RQXtS16q8H1DX3+ixR19wnh/KirlV/PaCuuddnibpWPTnQYKkGw4cPN+rXr294eXkZDRo0MIYPH24kJyc7z2dnZxt33323ERgYaPj6+hqDBw82Dh8+7DLH/v37jQEDBhg+Pj5GcHCw8eCDDxr5+fmVHvu3335rSCryNWrUKMMwDMPhcBhPPfWUERYWZlitVqNv377G7t27XeY4ceKEMWLECMPf398ICAgwbr31VuP06dMuY7Zu3Wr06NHDsFqtRoMGDYznnnuuyvLIysoyrrzySiMkJMTw9PQ0oqKijDvuuMM4cuSI2+VRXA6SjDlz5jjHVNTP07fffmtccsklhpeXl9GkSROXe1R2HikpKUbPnj2NoKAgw2q1Gs2aNTMeeughIy0tza3y+J//+R8jKirK8PLyMkJCQoy+ffs6i79h1Iz3oryoa9VfD6hr7vVZoq65Tw7lRV2r/npAXXOvzxJ1zX1yKC/qWvXXA+qae32WqGvuk0N5Udeqvx5Q19zrs0Rdq54cTIZhGAIAAAAAAAAAAECpsQcLAAAAAAAAAABAGdFgAQAAAAAAAAAAKCMaLAAAAAAAAAAAAGVEgwUAAAAAAAAAAKCMaLAAAAAAAAAAAACUEQ0WAAAAAAAAAACAMqLBAgAAAAAAAAAAUEY0WAAAAAAAAAAAAMqIBgsAAACAIkaPHq3GjRuX69rJkyfLZDJVbEAAAAAA4GZosAAAAAA1iMlkKtXX6tWrqzvUavPll1+qV69eCg0Nla+vr5o0aaIbbrhBS5cudY75888/NXnyZG3ZsqX6AgUAAABQo5kMwzCqOwgAAAAApfP++++7vH7vvfe0fPly/fe//3U53q9fP4WFhZX7Pvn5+XI4HLJarWW+1m63y263y9vbu9z3L6+XXnpJDz30kHr16qXrrrtOvr6+Sk5O1ooVK9ShQwfNnTtXkrRx40Z17dpVc+bM0ejRo6s8TgAAAAA1n6W6AwAAAABQejfddJPL6x9//FHLly8vcvyfsrKy5OvrW+r7eHp6lis+SbJYLLJYqv7/atjtdk2bNk39+vXTsmXLipw/duxYlccEAAAAoPZiiTAAAACglundu7fatm2rTZs2qWfPnvL19dXjjz8uSfr888919dVXKyIiQlarVU2bNtW0adNUUFDgMsc/92DZv3+/TCaTXnrpJb399ttq2rSprFarunbtqp9//tnl2uL2YDGZTBo3bpwWLVqktm3bymq1qk2bNi7LdhVavXq1unTpIm9vbzVt2lRvvfVWqfZ1OX78uNLT09W9e/diz4eGhjrn79q1qyTp1ltvdS6rVvh0iyRt2LBBCQkJqlu3rnx9fdWrVy+tW7eu2Dx37dqlG264QQEBAapXr57uu+8+5eTkuIxdvny5evToIZvNJn9/f8XExDjfEwAAAAA1E0+wAAAAALXQiRMnNGDAAN1444266aabnMuFzZ07V/7+/nrggQfk7++vVatWadKkSUpPT9eLL7543nnnzZun06dP66677pLJZNILL7ygIUOGaO/eved96mXt2rVauHCh7r77btWpU0czZszQ0KFDlZKSonr16kmSNm/erISEBNWvX19TpkxRQUGBpk6dqpCQkPPGFhoaKh8fH3355ZcaP368goKCih3XqlUrTZ06VZMmTdKdd96pyy+/XJLUrVs3SdKqVas0YMAAde7cWU8//bTMZrPmzJmjPn366Pvvv9ell17qMt8NN9ygxo0ba/r06frxxx81Y8YMnTp1Su+9954kaceOHbrmmmvUvn17TZ06VVarVcnJyUUaNgAAAABqFhosAAAAQC105MgRvfnmm7rrrrtcjs+bN08+Pj7O12PGjNGYMWM0a9YsPfPMM+fdcyUlJUVJSUkKDAyUJMXExOi6667TN998o2uuueac1+7cuVO//fabmjZtKkm64oor1KFDB3344YcaN26cJOnpp5+Wh4eH1q1bp4iICElnGhitWrU6b85ms1kPPfSQpk6dqsjISPXs2VM9evRQQkKCOnXq5BwXFhamAQMGaNKkSYqLi3NZXs0wDI0ZM0ZXXHGFlixZ4nxq5q677lKbNm305JNPFll+LDo6Wp9//rkk6Z577lFAQIBmzZqliRMnqn379lq+fLny8vK0ZMkSBQcHnzcPAAAAADUDS4QBAAAAtZDVatWtt95a5PjZzZXTp0/r+PHjuvzyy5WVlaVdu3add97hw4c7myuSnE9/7N2797zXxsfHO5srktS+fXsFBAQ4ry0oKNCKFSs0aNAgZ3NFkpo1a6YBAwacd35JmjJliubNm6eOHTvqm2++0RNPPKHOnTurU6dO2rlz53mv37Jli5KSkjRy5EidOHFCx48f1/Hjx5WZmam+ffvqu+++k8PhcLnmnnvucXk9fvx4SdLixYslSTabTdKZ5dn+eS0AAACAmosGCwAAAFALNWjQQF5eXkWO79ixQ4MHD1bdunUVEBCgkJAQ5xMcaWlp5503MjLS5XVhs+XUqVNlvrbw+sJrjx07puzsbDVr1qzIuOKOlWTEiBH6/vvvderUKS1btkwjR47U5s2bNXDgwCJ7o/xTUlKSJGnUqFEKCQlx+Zo9e7Zyc3OLfJ+aN2/u8rpp06Yym83av3+/pDNNqe7du+v2229XWFiYbrzxRn388cc0WwAAAIAajiXCAAAAgFro7CdVCqWmpqpXr14KCAjQ1KlT1bRpU3l7e+uXX37RI488Uqpf+Ht4eBR73DCMSr22PAICAtSvXz/169dPnp6eevfdd7Vhwwb16tWrxGsKvwcvvviiLrnkkmLH+Pv7n/O+hcuKFfLx8dF3332nb7/9Vl9//bWWLl2qjz76SH369NGyZctK/L4AAAAAcG80WAAAAICLxOrVq3XixAktXLhQPXv2dB7ft29fNUb1t9DQUHl7eys5ObnIueKOlUWXLl307rvv6vDhw5KKNkEKFS5hFhAQoPj4+FLNnZSUpOjoaJdYHQ6HGjdu7DxmNpvVt29f9e3bV6+88oqeffZZPfHEE/r2229LfR8AAAAA7oUlwgAAAICLROGTEmc/MZKXl6dZs2ZVV0guPDw8FB8fr0WLFunPP/90Hk9OTtaSJUvOe31WVpbWr19f7LnC62NiYiRJfn5+ks481XO2zp07q2nTpnrppZeUkZFRZJ6//vqryLGZM2e6vH7ttdckyblvzMmTJ4tcU/h0TG5ubknpAAAAAHBzPMECAAAAXCS6deumwMBAjRo1Svfee69MJpP++9//VtoSXeUxefJkLVu2TN27d9fYsWNVUFCg119/XW3bttWWLVvOeW1WVpa6deumyy67TAkJCWrUqJFSU1O1aNEiff/99xo0aJA6duwo6cyTKjabTW+++abq1KkjPz8/xcbGKjo6WrNnz9aAAQPUpk0b3XrrrWrQoIEOHTqkb7/9VgEBAfryyy9d7rtv3z5de+21SkhI0Pr16/X+++9r5MiR6tChgyRp6tSp+u6773T11VcrKipKx44d06xZs9SwYUP16NGjUr6PAAAAACofDRYAAADgIlGvXj199dVXevDBB/Xkk08qMDBQN910k/r27av+/ftXd3iSzjxBsmTJEk2cOFFPPfWUGjVqpKlTp2rnzp3atWvXOa+12Wz697//ra+//lpz5szRkSNH5OHhoZiYGL344ou69957nWML92R57LHHNGbMGNntds2ZM0fR0dHq3bu31q9fr2nTpun1119XRkaGwsPDFRsbq7vuuqvIfT/66CNNmjRJjz76qCwWi8aNG6cXX3zRef7aa6/V/v379Z///EfHjx9XcHCwevXqpSlTpqhu3boV980DAAAAUKVMhjv9uRoAAAAAFGPQoEHasWOHkpKSqjsUp8mTJ2vKlCn666+/FBwcXN3hAAAAAKhi7MECAAAAwK1kZ2e7vE5KStLixYvVu3fv6gkIAAAAAIrBEmEAAAAA3EqTJk00evRoNWnSRAcOHNAbb7whLy8vPfzww9UdGgAAAAA40WABAAAA4FYSEhL04Ycf6siRI7JarYqLi9Ozzz6r5s2bV3doAAAAAODEHiwAAAAAAAAAAABlxB4sAAAAAAAAAAAAZUSDBQAAAAAAAAAAoIxosAAAAAAAAAAAAJQRDRYAAAAAAAAAAIAyosECAAAAAAAAAABQRjRYAAAAAAAAAAAAyogGCwAAAAAAAAAAQBnRYAEAAAAAAAAAACgjGiwAAAAAAAAAAABl9P8AAXMDdUv1mRAAAAAASUVORK5CYII=",
+      "text/plain": [
+       "<Figure size 2000x400 with 5 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 2000x400 with 5 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Define the input files, layers, and checkpoints\n",
+    "data_files = [\n",
+    "    {'data': 'reverse_control',\n",
+    "     'label': 'NoReverse',\n",
+    "     'linestyle': '-',\n",
+    "     'marker': 'o',\n",
+    "    },\n",
+    "    {'data': 'reverse_partial',\n",
+    "     'label': 'PartialReverse',\n",
+    "     'linestyle': '--',\n",
+    "     'marker': 's',\n",
+    "    },\n",
+    "    {'data': 'reverse_full',\n",
+    "     'label': 'FullReverse',\n",
+    "     'linestyle': ':',\n",
+    "     'marker': 'd',\n",
+    "    },\n",
+    "]\n",
+    "\n",
+    "# Generate the plot\n",
+    "plot_accuracy_layer(data_files, LAYERS, CHECKPOINTS, SEEDS)\n",
+    "plt.savefig(\"figures/probing_layers_reverse.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "# No Pos Encodings\n",
+    "plot_accuracy_layer(data_files, LAYERS, CHECKPOINTS, [53], pos_encodings=False)\n",
+    "plt.savefig(\"figures/probing_layers_reverse_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 45,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 2000x400 with 5 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 2000x400 with 5 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# Define the input files, layers, and checkpoints\n",
+    "data_files = [\n",
+    "    {'data': 'hop_control',\n",
+    "     'label': 'NoHop',\n",
+    "     'linestyle': '-',\n",
+    "     'marker': 'o',\n",
+    "    },\n",
+    "    {'data': 'hop_tokens4',\n",
+    "     'label': 'TokenHop',\n",
+    "     'linestyle': '--',\n",
+    "     'marker': 's',\n",
+    "    },\n",
+    "    {'data': 'hop_words4',\n",
+    "     'label': 'WordHop',\n",
+    "     'linestyle': ':',\n",
+    "     'marker': 'd',\n",
+    "    },\n",
+    "]\n",
+    "\n",
+    "# Generate the plot\n",
+    "plot_accuracy_layer(data_files, LAYERS, CHECKPOINTS, SEEDS)\n",
+    "plt.savefig(\"figures/probing_layers_hop.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "# No Pos Encodings\n",
+    "plot_accuracy_layer(data_files, LAYERS, CHECKPOINTS, [53], pos_encodings=False)\n",
+    "plt.savefig(\"figures/probing_layers_hop_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "babyenv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/edge_probing/edge_probing.py b/edge_probing/edge_probing.py
new file mode 100644
index 0000000000000000000000000000000000000000..506416eb1aad1058521e39230281c6369cf477fe
--- /dev/null
+++ b/edge_probing/edge_probing.py
@@ -0,0 +1,222 @@
+# edge_probing.py
+# Author: Julie Kallini
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+from utils import CHECKPOINT_READ_PATH, PERTURBATIONS, PAREN_MODELS, get_gpt2_tokenizer_with_markers
+from gpt2_no_positional_encoding_model import GPT2NoPositionalEncodingLMHeadModel
+from transformers import GPT2LMHeadModel
+from sklearn.preprocessing import StandardScaler
+from sklearn.linear_model import LogisticRegression
+from sklearn.metrics import accuracy_score
+from sklearn.model_selection import train_test_split
+from itertools import zip_longest
+import torch
+import tqdm
+import argparse
+import pandas as pd
+import os
+
+
+MAX_TRAINING_STEPS = 3000
+CHECKPOINTS = list(range(200, MAX_TRAINING_STEPS+1, 200))
+LAYERS = [1, 3, 6, 9, 12, "Avg Last 4"]
+
+
+def get_layer_embedding(model, token_sequences, indices, layer=None):
+
+    # Pad input token sequences
+    input_ids = zip(*zip_longest(*token_sequences,
+                    fillvalue=gpt2_tokenizer.eos_token_id))
+    input_ids = torch.tensor(list(input_ids)).to(device)
+
+    # Get GPT2 model's output
+    with torch.no_grad():
+        output = model(input_ids)
+
+    # Either get the hidden state of the specified layer or
+    # get the average of the last 4 hidden states
+    if layer is not None:
+        hidden_states = output.hidden_states[layer]
+    else:
+        hidden_states = output.hidden_states[-4:]
+        hidden_states = sum(hidden_states) / 4
+
+    # Create mask using start and end indices
+    batch_size, seq_length = input_ids.shape
+    mask = torch.full((batch_size, seq_length), 0).to(device)
+    for i, (start_idx, end_idx) in enumerate(indices):
+        mask[i, start_idx:end_idx] = 1
+
+    # Mask out embeddings of tokens outside indices
+    mask_expanded = mask.unsqueeze(-1).expand(hidden_states.size())
+    hidden_states = hidden_states * mask_expanded
+
+    return hidden_states
+
+
+def max_pooling(tensor, index_tuples):
+    pooled_results = []
+    for i, (start, end) in enumerate(index_tuples):
+        # Extracting the embeddings corresponding to the specified range
+        embeddings = tensor[i, start:end, :]
+
+        # Performing max pooling
+        max_pooled = torch.max(embeddings, dim=0)[0]
+
+        pooled_results.append(max_pooled)
+    return torch.stack(pooled_results)
+
+
+def mean_pooling(tensor, index_tuples):
+    batch_size, seq_len, embedding_size = tensor.shape
+    output = torch.empty(batch_size, embedding_size,
+                         device=tensor.device, dtype=tensor.dtype)
+
+    for i, (start, end) in enumerate(index_tuples):
+        embeddings = tensor[i, start:end, :]
+        output[i, :] = torch.mean(embeddings, dim=0)
+
+    return output
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(
+        prog='Edge probing',
+        description='Edge probing experiments')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    parser.add_argument('random_seed', type=int, help="Random seed")
+    parser.add_argument('paren_model',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=list(PAREN_MODELS.keys()) + ["randinit"],
+                        help='Parenthesis model')
+    parser.add_argument('pooling_operation',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["mean", "max"],
+                        help='Pooling operation to compute on embeddings')
+    parser.add_argument('-np', '--no_pos_encodings', action='store_true',
+                        help="Train GPT-2 with no positional encodings")
+
+    # Get args
+    args = parser.parse_args()
+
+    if args.pooling_operation == "mean":
+        pooling_function = mean_pooling
+    elif args.pooling_operation == "max":
+        pooling_function = max_pooling
+    else:
+        raise Exception("Pooling operation undefined")
+
+    # Init tokenizer
+    gpt2_tokenizer = get_gpt2_tokenizer_with_markers([])
+    gpt2_tokenizer.pad_token = gpt2_tokenizer.eos_token
+
+    # Get path to model
+    no_pos_encodings_underscore = "_no_positional_encodings" if args.no_pos_encodings else ""
+    model = f"babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}_seed{args.random_seed}"
+    model_path = f"{CHECKPOINT_READ_PATH}/babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}/{model}/runs/{model}/checkpoint-"
+
+    # Get constituency parse data
+    if "hop" in args.perturbation_type:
+        phrase_df = pd.read_csv("phrase_data/hop_phrase_data.csv")
+    elif "reverse" in args.perturbation_type:
+        phrase_df = pd.read_csv("phrase_data/reverse_phrase_data.csv")
+    else:
+        raise Exception("Phrase data not found")
+
+    token_sequences = list(phrase_df["Sentence Tokens"])
+    if args.perturbation_type == "reverse_full":
+        indices = list(
+            zip(phrase_df["Rev Start Index"], phrase_df["Rev End Index"]))
+    else:
+        indices = list(zip(phrase_df["Start Index"], phrase_df["End Index"]))
+    labels = list(phrase_df["Category"])
+
+    BATCH_SIZE = 32
+    device = "cuda"
+
+    edge_probing_df = pd.DataFrame(LAYERS, columns=["GPT-2 Layer"])
+    for ckpt in CHECKPOINTS:
+
+        # Load model
+        if args.no_pos_encodings:
+            model = GPT2LMHeadModel.from_pretrained(
+                model_path + str(ckpt), output_hidden_states=True).to(device)
+        else:
+            model = GPT2NoPositionalEncodingLMHeadModel.from_pretrained(
+                model_path + str(ckpt), output_hidden_states=True).to(device)
+
+        layer_accuracies = []
+        for layer in LAYERS:
+            print(f"Checkpoint: {ckpt}, Layer: {layer}")
+            print("Computing span embeddings...")
+
+            # Iterate over token sequences and indices to get embeddings
+            spans = []
+            for i in tqdm.tqdm(list(range(0, len(token_sequences), BATCH_SIZE))):
+
+                tokens_batch = [[int(tok) for tok in seq.split()]
+                                for seq in token_sequences[i:i+BATCH_SIZE]]
+                if args.perturbation_type == "reverse_full":
+                    tokens_batch = [toks[::-1] for toks in tokens_batch]
+
+                index_batch = indices[i:i+BATCH_SIZE]
+
+                # Extract embeddings
+                if layer == "Avg Last 4":
+                    embeddings = get_layer_embedding(
+                        model, tokens_batch, index_batch, None)
+                else:
+                    embeddings = get_layer_embedding(
+                        model, tokens_batch, index_batch, layer)
+                pooled_results = pooling_function(embeddings, index_batch)
+                spans.extend(list(pooled_results))
+
+            # Get features and ground truth
+            X = torch.vstack(spans).detach().cpu().numpy()
+            y = labels
+
+            # Split the data; since we pass random seed, it
+            # will be the same split every time
+            X_train, X_test, y_train, y_test = train_test_split(
+                X, y, test_size=0.2, random_state=args.random_seed)
+
+            # Fit L2-regularized linear classifier
+            clf = LogisticRegression(max_iter=10,
+                                     random_state=args.random_seed).fit(X_train, y_train)
+
+            # Get probe accuracy
+            y_test_pred = clf.predict(X_test)
+            acc = accuracy_score(y_test, y_test_pred)
+            layer_accuracies.append(acc)
+            print(f"Accuracy: {acc}")
+
+        edge_probing_df[f"Accuracy (ckpt {ckpt})"] = layer_accuracies
+
+    # Write results to CSV
+    nps = '_no_pos_encodings' if args.no_pos_encodings else ''
+    directory = f"edge_probing_results/{args.perturbation_type}_{args.train_set}{nps}"
+    if not os.path.exists(directory):
+        os.makedirs(directory)
+
+    file = directory + \
+        f"/{args.paren_model}_{args.pooling_operation}_pooling_seed{args.random_seed}.csv"
+    print(f"Writing results to CSV: {file}")
+    edge_probing_df.to_csv(file)
diff --git a/edge_probing/edge_probing.sh b/edge_probing/edge_probing.sh
new file mode 100644
index 0000000000000000000000000000000000000000..24f45aadbd2587552c46445605f4c00a0714801d
--- /dev/null
+++ b/edge_probing/edge_probing.sh
@@ -0,0 +1,60 @@
+#!/bin/sh
+# edge_probing.sh
+# author: Julie Kallini
+
+echo "
+-------------------------------------------------------------------------------
+Arguments
+-------------------------------------------------------------------------------
+"
+echo "Random seed: $1"
+NO_POS_ENCODINGS=${2:-''}
+echo "No pos encodings: $NO_POS_ENCODINGS"
+
+echo "
+-------------------------------------------------------------------------------
+Run edge probing for each perturbation type
+-------------------------------------------------------------------------------
+"
+
+COMMAND="python3 edge_probing.py hop_control 100M $1 randinit mean $NO_POS_ENCODINGS"
+echo $COMMAND
+eval $COMMAND
+echo "
+"
+
+COMMAND="python3 edge_probing.py hop_tokens4 100M $1 randinit mean $NO_POS_ENCODINGS"
+echo $COMMAND
+eval $COMMAND
+echo "
+"
+
+COMMAND="python3 edge_probing.py hop_words4 100M $1 randinit mean $NO_POS_ENCODINGS"
+echo $COMMAND
+eval $COMMAND
+echo "
+"
+
+COMMAND="python3 edge_probing.py reverse_control 100M $1 randinit mean $NO_POS_ENCODINGS"
+echo $COMMAND
+eval $COMMAND
+echo "
+"
+
+COMMAND="python3 edge_probing.py reverse_full 100M $1 randinit mean $NO_POS_ENCODINGS"
+echo $COMMAND
+eval $COMMAND
+echo "
+"
+
+COMMAND="python3 edge_probing.py reverse_partial 100M $1 randinit mean $NO_POS_ENCODINGS"
+echo $COMMAND
+eval $COMMAND
+echo "
+"
+
+echo "
+-------------------------------------------------------------------------------
+Done!
+-------------------------------------------------------------------------------
+"
diff --git a/gpt2_no_positional_encoding_model.py b/gpt2_no_positional_encoding_model.py
new file mode 100644
index 0000000000000000000000000000000000000000..763af64d333bd5f8d3b7314a6eb783dcff64cc92
--- /dev/null
+++ b/gpt2_no_positional_encoding_model.py
@@ -0,0 +1,427 @@
+# gpt2_no_positional_encoding_model.py
+# Adapted from Huggingface's transformers library
+
+import torch
+from transformers.models.gpt2.modeling_gpt2 import GPT2Block, GPT2PreTrainedModel
+from transformers.modeling_outputs import CausalLMOutputWithCrossAttentions, BaseModelOutputWithPastAndCrossAttentions
+from transformers.utils.model_parallel_utils import assert_device_map, get_device_map
+from torch import nn
+from torch.nn import CrossEntropyLoss
+from typing import Optional, Tuple, Union
+
+class GPT2NoPositionalEncodingModel(GPT2PreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+
+        self.embed_dim = config.hidden_size
+
+        self.wte = nn.Embedding(config.vocab_size, self.embed_dim)
+
+        self.drop = nn.Dropout(config.embd_pdrop)
+        self.h = nn.ModuleList([GPT2Block(config, layer_idx=i) for i in range(config.num_hidden_layers)])
+        self.ln_f = nn.LayerNorm(self.embed_dim, eps=config.layer_norm_epsilon)
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+        self.gradient_checkpointing = False
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def parallelize(self, device_map=None):
+        # Check validity of device_map
+        self.device_map = (
+            get_device_map(len(self.h), range(torch.cuda.device_count())) if device_map is None else device_map
+        )
+        assert_device_map(self.device_map, len(self.h))
+        self.model_parallel = True
+        self.first_device = "cpu" if "cpu" in self.device_map.keys() else "cuda:" + str(min(self.device_map.keys()))
+        self.last_device = "cuda:" + str(max(self.device_map.keys()))
+        self.wte = self.wte.to(self.first_device)
+        # Load onto devices
+        for k, v in self.device_map.items():
+            for block in v:
+                cuda_device = "cuda:" + str(k)
+                self.h[block] = self.h[block].to(cuda_device)
+        # ln_f to last
+        self.ln_f = self.ln_f.to(self.last_device)
+
+    def deparallelize(self):
+        self.model_parallel = False
+        self.device_map = None
+        self.first_device = "cpu"
+        self.last_device = "cpu"
+        self.wte = self.wte.to("cpu")
+        for index in range(len(self.h)):
+            self.h[index] = self.h[index].to("cpu")
+        self.ln_f = self.ln_f.to("cpu")
+        torch.cuda.empty_cache()
+
+    def get_input_embeddings(self):
+        return self.wte
+
+    def set_input_embeddings(self, new_embeddings):
+        self.wte = new_embeddings
+
+    def _prune_heads(self, heads_to_prune):
+        """
+        Prunes heads of the model. heads_to_prune: dict of {layer_num: list of heads to prune in this layer}
+        """
+        for layer, heads in heads_to_prune.items():
+            self.h[layer].attn.prune_heads(heads)
+
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Tuple[Tuple[torch.Tensor]]] = None,
+        attention_mask: Optional[torch.FloatTensor] = None,
+        token_type_ids: Optional[torch.LongTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.FloatTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        encoder_hidden_states: Optional[torch.Tensor] = None,
+        encoder_attention_mask: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, BaseModelOutputWithPastAndCrossAttentions]:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        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
+
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            input_shape = input_ids.size()
+            input_ids = input_ids.view(-1, input_shape[-1])
+            batch_size = input_ids.shape[0]
+        elif inputs_embeds is not None:
+            input_shape = inputs_embeds.size()[:-1]
+            batch_size = inputs_embeds.shape[0]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        device = input_ids.device if input_ids is not None else inputs_embeds.device
+
+        if token_type_ids is not None:
+            token_type_ids = token_type_ids.view(-1, input_shape[-1])
+
+        if past_key_values is None:
+            past_length = 0
+            past_key_values = tuple([None] * len(self.h))
+        else:
+            past_length = past_key_values[0][0].size(-2)
+        if position_ids is None:
+            position_ids = torch.arange(past_length, input_shape[-1] + past_length, dtype=torch.long, device=device)
+            position_ids = position_ids.unsqueeze(0)
+
+        # GPT2Attention mask.
+        if attention_mask is not None:
+            if batch_size <= 0:
+                raise ValueError("batch_size has to be defined and > 0")
+            attention_mask = attention_mask.view(batch_size, -1)
+            # We create a 3D attention mask from a 2D tensor mask.
+            # Sizes are [batch_size, 1, 1, to_seq_length]
+            # So we can broadcast to [batch_size, num_heads, from_seq_length, to_seq_length]
+            # this attention mask is more simple than the triangular masking of causal attention
+            # used in OpenAI GPT, we just need to prepare the broadcast dimension here.
+            attention_mask = attention_mask[:, None, None, :]
+
+            # Since attention_mask is 1.0 for positions we want to attend and 0.0 for
+            # masked positions, this operation will create a tensor which is 0.0 for
+            # positions we want to attend and the dtype's smallest value for masked positions.
+            # Since we are adding it to the raw scores before the softmax, this is
+            # effectively the same as removing these entirely.
+            attention_mask = attention_mask.to(dtype=self.dtype)  # fp16 compatibility
+            attention_mask = (1.0 - attention_mask) * torch.finfo(self.dtype).min
+
+        # If a 2D or 3D attention mask is provided for the cross-attention
+        # we need to make broadcastable to [batch_size, num_heads, seq_length, seq_length]
+        if self.config.add_cross_attention and encoder_hidden_states is not None:
+            encoder_batch_size, encoder_sequence_length, _ = encoder_hidden_states.size()
+            encoder_hidden_shape = (encoder_batch_size, encoder_sequence_length)
+            if encoder_attention_mask is None:
+                encoder_attention_mask = torch.ones(encoder_hidden_shape, device=device)
+            encoder_attention_mask = self.invert_attention_mask(encoder_attention_mask)
+        else:
+            encoder_attention_mask = None
+
+        # Prepare head mask if needed
+        # 1.0 in head_mask indicate we keep the head
+        # attention_probs has shape bsz x n_heads x N x N
+        # head_mask has shape n_layer x batch x n_heads x N x N
+        head_mask = self.get_head_mask(head_mask, self.config.n_layer)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.wte(input_ids)
+        hidden_states = inputs_embeds
+
+        if token_type_ids is not None:
+            token_type_embeds = self.wte(token_type_ids)
+            hidden_states = hidden_states + token_type_embeds
+
+        hidden_states = self.drop(hidden_states)
+
+        output_shape = (-1,) + input_shape[1:] + (hidden_states.size(-1),)
+
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                use_cache = False
+
+        presents = () if use_cache else None
+        all_self_attentions = () if output_attentions else None
+        all_cross_attentions = () if output_attentions and self.config.add_cross_attention else None
+        all_hidden_states = () if output_hidden_states else None
+        for i, (block, layer_past) in enumerate(zip(self.h, past_key_values)):
+            # Model parallel
+            if self.model_parallel:
+                torch.cuda.set_device(hidden_states.device)
+                # Ensure layer_past is on same device as hidden_states (might not be correct)
+                if layer_past is not None:
+                    layer_past = tuple(past_state.to(hidden_states.device) for past_state in layer_past)
+                # Ensure that attention_mask is always on the same device as hidden_states
+                if attention_mask is not None:
+                    attention_mask = attention_mask.to(hidden_states.device)
+                if isinstance(head_mask, torch.Tensor):
+                    head_mask = head_mask.to(hidden_states.device)
+            if output_hidden_states:
+                all_hidden_states = all_hidden_states + (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+                outputs = self._gradient_checkpointing_func(
+                    block.__call__,
+                    hidden_states,
+                    None,
+                    attention_mask,
+                    head_mask[i],
+                    encoder_hidden_states,
+                    encoder_attention_mask,
+                    use_cache,
+                    output_attentions,
+                )
+            else:
+                outputs = block(
+                    hidden_states,
+                    layer_past=layer_past,
+                    attention_mask=attention_mask,
+                    head_mask=head_mask[i],
+                    encoder_hidden_states=encoder_hidden_states,
+                    encoder_attention_mask=encoder_attention_mask,
+                    use_cache=use_cache,
+                    output_attentions=output_attentions,
+                )
+
+            hidden_states = outputs[0]
+            if use_cache is True:
+                presents = presents + (outputs[1],)
+
+            if output_attentions:
+                all_self_attentions = all_self_attentions + (outputs[2 if use_cache else 1],)
+                if self.config.add_cross_attention:
+                    all_cross_attentions = all_cross_attentions + (outputs[3 if use_cache else 2],)
+
+            # Model Parallel: If it's the last layer for that device, put things on the next device
+            if self.model_parallel:
+                for k, v in self.device_map.items():
+                    if i == v[-1] and "cuda:" + str(k) != self.last_device:
+                        hidden_states = hidden_states.to("cuda:" + str(k + 1))
+
+        hidden_states = self.ln_f(hidden_states)
+
+        hidden_states = hidden_states.view(output_shape)
+        # Add last hidden state
+        if output_hidden_states:
+            all_hidden_states = all_hidden_states + (hidden_states,)
+
+        if not return_dict:
+            return tuple(
+                v
+                for v in [hidden_states, presents, all_hidden_states, all_self_attentions, all_cross_attentions]
+                if v is not None
+            )
+
+        return BaseModelOutputWithPastAndCrossAttentions(
+            last_hidden_state=hidden_states,
+            past_key_values=presents,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attentions,
+            cross_attentions=all_cross_attentions,
+        )
+
+class GPT2NoPositionalEncodingLMHeadModel(GPT2PreTrainedModel):
+    _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.transformer = GPT2NoPositionalEncodingModel(config)
+        self.lm_head = nn.Linear(config.n_embd, config.vocab_size, bias=False)
+
+        # Model parallel
+        self.model_parallel = False
+        self.device_map = None
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def parallelize(self, device_map=None):
+        self.device_map = (
+            get_device_map(len(self.transformer.h), range(torch.cuda.device_count()))
+            if device_map is None
+            else device_map
+        )
+        assert_device_map(self.device_map, len(self.transformer.h))
+        self.transformer.parallelize(self.device_map)
+        self.lm_head = self.lm_head.to(self.transformer.first_device)
+        self.model_parallel = True
+
+    def deparallelize(self):
+        self.transformer.deparallelize()
+        self.transformer = self.transformer.to("cpu")
+        self.lm_head = self.lm_head.to("cpu")
+        self.model_parallel = False
+        torch.cuda.empty_cache()
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def prepare_inputs_for_generation(self, input_ids, past_key_values=None, inputs_embeds=None, **kwargs):
+        token_type_ids = kwargs.get("token_type_ids", None)
+        # Omit tokens covered by past_key_values
+        if past_key_values:
+            past_length = past_key_values[0][0].shape[2]
+
+            # Some generation methods already pass only the last input ID
+            if input_ids.shape[1] > past_length:
+                remove_prefix_length = past_length
+            else:
+                # Default to old behavior: keep only final ID
+                remove_prefix_length = input_ids.shape[1] - 1
+
+            input_ids = input_ids[:, remove_prefix_length:]
+            if token_type_ids is not None:
+                token_type_ids = token_type_ids[:, -input_ids.shape[1] :]
+
+        attention_mask = kwargs.get("attention_mask", None)
+        position_ids = kwargs.get("position_ids", None)
+
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past_key_values:
+                position_ids = position_ids[:, -input_ids.shape[1] :]
+        else:
+            position_ids = None
+
+        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
+        if inputs_embeds is not None and past_key_values is None:
+            model_inputs = {"inputs_embeds": inputs_embeds}
+        else:
+            model_inputs = {"input_ids": input_ids}
+
+        model_inputs.update(
+            {
+                "past_key_values": past_key_values,
+                "use_cache": kwargs.get("use_cache"),
+                "position_ids": position_ids,
+                "attention_mask": attention_mask,
+                "token_type_ids": token_type_ids,
+            }
+        )
+
+        return model_inputs
+
+    def forward(
+        self,
+        input_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[Tuple[Tuple[torch.Tensor]]] = None,
+        attention_mask: Optional[torch.FloatTensor] = None,
+        token_type_ids: Optional[torch.LongTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        head_mask: Optional[torch.FloatTensor] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        encoder_hidden_states: Optional[torch.Tensor] = None,
+        encoder_attention_mask: Optional[torch.FloatTensor] = 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, CausalLMOutputWithCrossAttentions]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Labels for language modeling. Note that the labels **are shifted** inside the model, i.e. you can set
+            `labels = input_ids` Indices are selected in `[-100, 0, ..., config.vocab_size]` All labels set to `-100`
+            are ignored (masked), the loss is only computed for labels in `[0, ..., config.vocab_size]`
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.transformer(
+            input_ids,
+            past_key_values=past_key_values,
+            attention_mask=attention_mask,
+            token_type_ids=token_type_ids,
+            position_ids=position_ids,
+            head_mask=head_mask,
+            inputs_embeds=inputs_embeds,
+            encoder_hidden_states=encoder_hidden_states,
+            encoder_attention_mask=encoder_attention_mask,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        hidden_states = transformer_outputs[0]
+
+        # Set device for model parallelism
+        if self.model_parallel:
+            torch.cuda.set_device(self.transformer.first_device)
+            hidden_states = hidden_states.to(self.lm_head.weight.device)
+
+        lm_logits = self.lm_head(hidden_states)
+
+        loss = None
+        if labels is not None:
+            # move labels to correct device to enable model parallelism
+            labels = labels.to(lm_logits.device)
+            # Shift so that tokens < n predict n
+            shift_logits = lm_logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss()
+            loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)), shift_labels.view(-1))
+
+        if not return_dict:
+            output = (lm_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return CausalLMOutputWithCrossAttentions(
+            loss=loss,
+            logits=lm_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+            cross_attentions=transformer_outputs.cross_attentions,
+        )
+
+    @staticmethod
+    def _reorder_cache(
+        past_key_values: Tuple[Tuple[torch.Tensor]], beam_idx: torch.Tensor
+    ) -> Tuple[Tuple[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.
+        """
+        return tuple(
+            tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past)
+            for layer_past in past_key_values
+        )
diff --git a/hop_interventions/create_agreement_data.py b/hop_interventions/create_agreement_data.py
new file mode 100644
index 0000000000000000000000000000000000000000..4f7134dc26df7184583bdb11e0e1abbc54c17865
--- /dev/null
+++ b/hop_interventions/create_agreement_data.py
@@ -0,0 +1,85 @@
+# create__agreement_data.py
+# Author: Julie Kallini
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+import pandas as pd
+from glob import glob
+import re
+from utils import gpt2_hop_tokenizer, BABYLM_DATA_PATH, marker_sg_token
+from pluralizer import Pluralizer
+
+
+if __name__ == "__main__":
+
+    get_vocab_dict = []
+    vocab = gpt2_hop_tokenizer.vocab
+
+    # Patterns for finding simple "The SUBJ VERB x x x x MARKER" sequences
+    control_pattern = f"(?:^| )(?:{vocab['The']}|{vocab['Ġthe']}) [1-9]+ [1-9]+ {marker_sg_token} [1-9]+ [1-9]+ [1-9]+ [1-9]+"
+    words4_pattern = f"(?:^| )(?:{vocab['The']}|{vocab['Ġthe']}) [1-9]+ [1-9]+ [1-9]+ [1-9]+ [1-9]+ [1-9]+ {marker_sg_token}"
+
+    # Get test file paths
+    test_file_path = f'{BABYLM_DATA_PATH}/babylm_data_perturbed/' + \
+        'babylm_hop_{}/babylm_test_affected/*'
+    control_files = sorted(glob(test_file_path.format("control")))
+    words4_files = sorted(glob(test_file_path.format("words4")))
+
+    # Iterate over files and get candidate sequences for interventions
+    candidate_sequences = []
+    for control_file_path, words4_file_path in zip(control_files, words4_files):
+        print(control_file_path.split("/")[-1])
+        assert control_file_path.split(
+            "/")[-1] == words4_file_path.split("/")[-1]
+
+        # Iterate over pairs of lines in file (control and words4)
+        control_lines = open(control_file_path, 'r').readlines()
+        words4_lines = open(words4_file_path, 'r').readlines()
+        for cl, wl in zip(control_lines, words4_lines):
+
+            # Find all matches to patterns
+            cseqs = re.findall(control_pattern, cl)
+            wseqs = re.findall(words4_pattern, wl)
+
+            # See if there is a shared pattern in control and words4 line
+            for cseq, wseq in zip(re.findall(control_pattern, cl), re.findall(words4_pattern, wl)):
+                if cseq.replace(" " + str(marker_sg_token), "") == wseq.replace(" " + str(marker_sg_token), ""):
+                    candidate_sequences.append(
+                        [int(s) for s in wseq.replace(
+                            str(marker_sg_token), "").split()]
+                    )
+
+    # Init pluralizer
+    pluralizer = Pluralizer()
+
+    # Iterate over candidate sequences
+    data = []
+    for seq in candidate_sequences:
+
+        # Get string version of sequence to get the subject
+        splitted = gpt2_hop_tokenizer.decode(seq).split()
+
+        # Get plural of subject (the word at index 1)
+        splitted[1] = pluralizer.pluralize(splitted[1], 2, False)
+
+        # Get GPT-2 tokens of plural version of sequence
+        plur_seq = gpt2_hop_tokenizer.encode(" ".join(splitted))
+
+        # If new subject form has a different number of tokens, skip
+        if len(plur_seq) != len(seq):
+            continue
+
+        # In case " the" has changed to "the", reset the first token
+        plur_seq[0] = seq[0]
+
+        # If singular and plural sequence are the same, skip
+        if seq == plur_seq:
+            continue
+
+        data.append([" ".join([str(s) for s in seq]),
+                     " ".join([str(s) for s in plur_seq])])
+
+    df = pd.DataFrame(data, columns=["Singular", "Plural"])
+    df.to_csv("hop_agreement_data.csv")
diff --git a/hop_interventions/hop_interventions.ipynb b/hop_interventions/hop_interventions.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..5272a1c67a7226a472ceed86abb8640a8dc642b1
--- /dev/null
+++ b/hop_interventions/hop_interventions.ipynb
@@ -0,0 +1,602 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/nlp/scr/kallini/miniconda3/envs/llmenv/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    }
+   ],
+   "source": [
+    "import sys\n",
+    "sys.path.append(\"..\")\n",
+    "\n",
+    "from utils import marker_sg_token, marker_pl_token"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Here is the final code used for the analysis and visualization:\n",
+    "\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "import seaborn as sns\n",
+    "import scipy.stats as stats\n",
+    "\n",
+    "# Function to process each file and compute accuracy\n",
+    "\n",
+    "\n",
+    "def process_file(file_path, ckpt):\n",
+    "    df = pd.read_csv(file_path)\n",
+    "    df = df[df['ckpt'] == ckpt]\n",
+    "\n",
+    "    # Separate the data for marker tokens\n",
+    "    token_marker_sg = df[df['token'] == marker_sg_token]\n",
+    "    token_marker_pl = df[df['token'] == marker_pl_token]\n",
+    "\n",
+    "    # Merge the two datasets\n",
+    "    merged_df = pd.merge(token_marker_sg, token_marker_pl, on=[\n",
+    "                         'example', 'layer', 'pos', 'type'], suffixes=('_marker_sg', '_marker_pl'))\n",
+    "\n",
+    "    # Compute correctness\n",
+    "    merged_df['correctness'] = (\n",
+    "        merged_df['prob_marker_pl'] > merged_df['prob_marker_sg']).astype(int)\n",
+    "\n",
+    "    # Compute accuracy for each layer and pos\n",
+    "    merged_df = merged_df.groupby(['layer', 'pos']).agg(\n",
+    "        accuracy=('correctness', 'mean')).reset_index()\n",
+    "    merged_df['accuracy'] = merged_df['accuracy'] * 100\n",
+    "\n",
+    "    return merged_df\n",
+    "\n",
+    "\n",
+    "def plot_intervention_accuracies(perturbation, seeds, ckpt, custom_x_labels, title=None, pos_encodings=True):\n",
+    "\n",
+    "    if pos_encodings:\n",
+    "        nps = \"\"\n",
+    "    else:\n",
+    "        nps = \"_no_pos_encodings\"\n",
+    "    base_path = 'hop_intervention_results/{}_100M{}'.format(perturbation, nps) \\\n",
+    "        + '/randinit_seed{}.csv'\n",
+    "\n",
+    "    # Process each file\n",
+    "    all_accuracy_data = pd.concat(\n",
+    "        [process_file(base_path.format(s), ckpt) for s in seeds])\n",
+    "\n",
+    "    # Compute mean accuracy and 95% confidence intervals\n",
+    "    grouped_data = all_accuracy_data.groupby(['layer', 'pos'])\n",
+    "    mean_accuracy = grouped_data['accuracy'].mean()\n",
+    "    sem_accuracy = grouped_data['accuracy'].sem()\n",
+    "\n",
+    "    # Calculate confidence interval using t-distribution\n",
+    "    ci_lower, ci_upper = stats.t.interval(0.95, df=len(\n",
+    "        seeds)-1, loc=mean_accuracy, scale=sem_accuracy + 1e-6)\n",
+    "\n",
+    "    # Preparing data for heatmap\n",
+    "    final_data = pd.DataFrame({\n",
+    "        'layer': mean_accuracy.index.get_level_values(0),\n",
+    "        'pos': mean_accuracy.index.get_level_values(1),\n",
+    "        'mean_accuracy': mean_accuracy.values,\n",
+    "        'ci': (ci_upper - ci_lower) / 2,\n",
+    "    })\n",
+    "\n",
+    "    heatmap_data = final_data.pivot(\n",
+    "        index='layer', columns='pos', values='mean_accuracy')\n",
+    "    ci_data = final_data.pivot(index='layer', columns='pos', values='ci')\n",
+    "\n",
+    "    # Plotting the heatmap\n",
+    "    if perturbation == \"hop_control\":\n",
+    "        plt.figure(figsize=(4, 4))\n",
+    "    else:\n",
+    "        plt.figure(figsize=(8, 4))\n",
+    "\n",
+    "    heatmap = sns.heatmap(\n",
+    "        heatmap_data.iloc[::-1], annot=True, fmt=\".2f\", cmap=\"viridis\", vmin=0.0, vmax=100.0, cbar=False)\n",
+    "    if title:\n",
+    "        plt.title(title)\n",
+    "    plt.xlabel(None)\n",
+    "    plt.ylabel(None)\n",
+    "    # plt.title(\"IIA\", fontsize=12)\n",
+    "\n",
+    "    # Setting custom X labels\n",
+    "    heatmap.set_xticklabels(custom_x_labels)\n",
+    "\n",
+    "    # Annotate with confidence intervals\n",
+    "    for i, row in enumerate(heatmap_data.iloc[::-1].itertuples()):\n",
+    "        for j, val in enumerate(row[1:]):\n",
+    "            ci_val = ci_data.iloc[::-1].iat[i, j]\n",
+    "            text = heatmap.texts[i * len(row[1:]) + j]\n",
+    "            text.set_text(f'{val:.1f}±{ci_val:.1f}')\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### With Positional Encodings"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checkpoint: 300\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checkpoint: 600\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checkpoint: 900\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWEAAAFiCAYAAAAna2l5AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAACd4ElEQVR4nOydd3hUxfeH323Zkp5AGgRICEiTjoViQb6IClKkY+ggKqDSQ++hBFGkQ2gqIvIDBJGOihSlSe8EAiQkpNfNJlt+fyxsWJNNoW2QeZ8nz5OZOzP37M7u5557ZvYeiclkMiEQCAQCuyC1twECgUDwPCNEWCAQCOyIEGGBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAjQoQFAoHAjggRFggEAjsit7cB92mu6GxvEwSCZ47wG3/Y2wRBAfiXuVNoG+EJCwQCgR0RIiwQCAR2RIiwQCAQ2BEhwgKBQGBHhAgLBAKBHREiLBAIBHakxGxRexq82LgKHYa2olLdADz9PJj4QRiHthyz2d7Dx43+s4KpXC8QvyBvNs/fweKhax76/M7ujgRP6EC9ZjXxKleKlLhUDm05yqoJ68lM1RbYt/uEDrzTpylObo6cO3SJeQPDib4a89C2lDTE3JQsMjNh1QolBw4oSE6WEBRk4JOBWVSpYgRg1kwVu3Y6WPWp30DPjJmZBY4bHydh2TIlR47I0WVJ8CtjZPgILS+8YCy2jdeuSVm3VsnZszJSUiT4+Bhp2SqHdh9kF9jv9i0pS5YoOXdWhl4vISDQQK9eOmrXMRTbhsfBY/eEY2NjmTx58uMe9rGgclQRcTqS+YNXFqm9QqkgJT6VtaEbiTgdWaQ+w8I/Jnhc+3yPefq54+nrzrKR39G/9nDC+iyifvPaDF06oMAxOw57nzYDWzDv0+UMbjSWrAwdodtCUCgVRbLpWUDMTcliTpia48fljArRsiw8nXr1DYwY7kh8nMTSpsFLetZvSLP8jRlbsACnpcFngx2RyyA0NJPwlekMGJCFs5Pt5D7dujhx8qQs32OXL8twczcyarSW5SvS6dotm/DlSjZvKvi9HzNGjcEAYXMyWbg4g4oVjYwdoyExUVJgvyfFY/eEY2JimDRpEuPHj3/cQz8yR3ee5OjOk0VuHxsZx6IhqwF4u+ebj3z+G+duM6XTXEv5TkQsK8evY+TqgUhlUoyG/L2BtoPfYe30TRzeehyAWb0WsD5qCY1a1+f39Ycf2a6SgJibkoNOB3/ulzN5qpaatczeYY+eOv46LGfLFgd699EBoFCY8PAoena0dT8oKe1lZPjILEudr+/De5/vvJNjVfbzy+H8eRkH/lTQpm1Ovn1SUiRE3ZYxbFgWgRXNc9q3XxZbfnbg+nUpHh5P3xsutgifPn26wOOXLl16aGOeRxxdNWSmam1+yX0CvPD0defEvjOWusxULRePXKXqK5Wf2S/6s8DzOjcGAxiNEhwcrAXWQWni7Nlcr/TUSTnt2znh5GSidh0DvXrrcHW1LcqHD8upX1/P5IlqTp+W4VnKxPvvZ/Ney/wF82HIyABnF9s2uLiY8Pc3sGuXgqBKBhwc4JetDri5G6lc2T7hiGKLcO3atZFIJOSXH/R+vURiH7feHjTt0ojPFvazlBVKOSYTtB/S0lI3puUMzh68mKevi6cz3Ua349fle22O7+HjBkBybIpVfVJsCu7ebo9m/H8cMTcPh0YD1arp+e5bJeXKaXF3N/HbPgUXzsvw8zNfkBo00NO4sR4fXyN3oqWEhysZPUrDvPkZyPKPHnAnWsrWLQ6075BNl246Ll2SsWC+CoUCmr9tFuKv5qrYszs3nKDTwehRGqQPBE5/+TUt3/HPnZXx+28Kpk23HRaRSGBWWCYTxml4v6UzEgm4u5sInZGJs3Mx36jHRLFF2MPDg1mzZvHWW2/le/zcuXO0atWqwDF0Oh06nc6qzmgyIJXYmL0SzOGtx7l45Kql3Gd6VxKiE9k8f4elLj4qMU8/jbOaqVtGcvNCFN9O3vBUbH3eEHPz8IwK0RI2W03njs5IpSYqVTLyZtMcrlw2f0ffbKq3tA0MNBIQaKD7h86cOiWjbt38PUqTCSpXNtCnr/m7X6mSkRvXpWzdqrCIcI+eOjp0zNWGoV840q9/FlWqFuylXr8uZfw4NcHdddRvYLutyQTzvlbh5mZk7tdZKB1M/PqrA+PGaFiwKANPz6effL7YIlyvXj2io6MpX758vseTk5Pz9ZIfJDQ0lEmTJlnVBUqqU1FWo7jm2B1tehba9CyrclpiBtHXYm32UTupmLYthMw0LRPbz8Ggt/2hSYxJBsDN29XyP4C7tyvXThVtQep5RczNw+NXxsSXX2Wi1UJmpgRPTxNTJqvx8c0/NOPnZ8LV1Uh0lNSmCHt4mChfwbp/uXJG/tyf6/m6u5twd889LpOBZykTZcrY1pTIG1KGD9PwXsscPgwueGfEP//I+PsvOZt+TsPR0Vz3WeUsThx3ZNdOBV26Ftz/SVDs3REDBgygQoUKNo+XK1eOlSsLXuEOCQkhJSXF6i9AWrW4pjyTaJzVhG4fjT5bz4S2s8nRFRwPi7l+l4Q7SdR5M/cCpXFWU+WlIC78dflJm/tcIeYmL2o1eHqaSEuDY0flNGykz7ddXJyE1FRJgQt11WsYuHXLWnJu35bi7V387Wn3uXFdytChGpo3z7EsGBaELsscKpX+S/kkErOXbA+K7Qm3bdu2wOPu7u706NGjwDZKpRKlUmlV9zRCESpHJX5BPpayT4AXgbXKk5aYTtytBHpP7YxnGQ9m91poaRNYy+zxq52UuJV2IbBWefTZem5eiALAQaXA0VVjaR8+ei1g9obuk5aYjj7HYPmSKzUOzOyxAI2LGo2LGoCUuFSMRvOnIPzMHFaMXcfBn48CsGnedrqObkvU1Rhibtyl58SOJEQncfBn2/tonzXE3JQsjh6VYTKBv7/Zu126RIV/OQMtWuSg1cKa1UqavKbHw8NIdLSUZUtU+JUxUr9BrkgPH6qhUeMcy06FD9rr+GyQI2u/d+D1N3K4eFHGr9sc+GJI7j7s9HTIzs5dU/pmQQaA1fax+0J//bqU4UM11K+vp32HbEsbqRTc3MxtLl6QMnOGmtlhmZQqbaJadQNOTiZmzjCHLpQOsG2bgpgYKS+/kv8F5knz2Leo3bp1iwkTJrBixYrHPfQjU7leRcL25m6dGxDWHYBda/4grM8iPHzd8fIvZdVn8bGZVv2bdmlMzI04ulcaBMDrHRsyPPzjAs877K3JnN5/nqA6AVR9uRIAqy99bdUmOGgQsZFxAPhXKYPmAfFYH7YFlaOSzxf1w8lNw9mDlxjdckahntqzhJibkkVGhoTwZSri4yU4O5to0kRPrz5ZyOXm3RMRETJ271KQnm4OVdSrr6dXLx0OD/x+IzpaSkpKrstZpYqRSZO1LF+u5Ns1Snx9jXz8SRZvNcsVv4UL8v4I5N/s2ZcKwP4/FCQnS9mzx4E9e3L7eHsb+f6HdACydBJu3ZJxP6rk6moidGYmK8JVDBuqwaCXUL6CgclTtFSs+PAe+aMgMRUWwC0mp06dom7duhgMxdvuIR7qLhAUH/FQ95JNUR7qXmxPeMuWLQUej4iIKO6QAoFA8NxSbBFu06aNzX3C93me9gkLBALBo1Ds3RG+vr5s3LgRo9GY79+JEyeehJ0CgUDwn6TYIlyvXj2OHz9u83hhXrJAIBAIcil2OGL48OFkZGTYPB4UFMRvv/32SEYJBALB80KxRbhJkyYFHnd0dOT1119/aIMEAoHgeUJk1hAIBAI7IkRYIBAI7IgQYYFAILAjQoQFAoHAjggRFggEAjsiRFggEAjsiBBhgUAgsCNChAUCgcCOPPbnCZdkXmxchQ5DW1GpbgCefh5M/CCMQ1tsP3zbw8eN/rOCqVwvEL8gbzbP38HioWse+vzO7o4ET+hAvWY18SpXipS4VA5tOcqqCevJTNUW2Lf7hA6806cpTm6OnDt0iXkDw4m+GvPQtpQ0xNyULDIzYdUKJQcOKEhOlhAUZOCTgVlUqWJ+5u6smXmf+1u/gZ4ZM20n2XyQH9Y6EL5cRbt2Oj4ZWHhGjPw4eVLG/21w4NJFGZmZEsqUMdKxk87q+cT5ceKEjFUrlFy/LkOlMtH8bXNWDlsJSp80DyXC2dnZbN68mcOHDxMTY/6w+fj40LBhQ1q3bo2DQ8EPZbYXKkcVEacj2bnqdyZsGFpoe4VSQUp8KmtDN9Lus3eLdI5h4R8TeyOOb6fkTRDp6eeOp687y0Z+R+SFKLzLlWLwgr54+nowpfNcm2N2HPY+bQa2YHbvhcTciKPHxI6Ebguhb81hz/zDw+8j5qZkMSdMzY3rUkaFaPEsZWTPbgdGDHdkxYp0SpU2PxumwUt6ho/IvUApFEV7ZszFi1K2/eJAYGDhzxxv1tSF79am4eOTd+zz52QEBhrp3Dkbd3cjf/2lYOYMNY6OWl55NX8hvnZNypgQDV276RgZoiU+XsrXc1UYDfDRxw93MXhUii3CV69e5e233yY6OpqXX34Zb29vAP755x8WL15M2bJl2b59O0FBQY/d2Efl6M6THN15ssjtYyPjWDRkNQBv93zzkc9/49xtpnTK/ULfiYhl5fh1jFw9EKlMitGQ/5P92w5+h7XTN3F4q/nBSbN6LWB91BIata7P7+sPP7JdJQExNyUHnQ7+3C9n8lQtNWuZhbJHTx1/HZazZYuDJZebQmEqMKdcfmi1EDpdzRdDtXz/nbLwDgXQtZt1Us52H2Rz7JiMP/+U2xTh339TEBBoJLi7uW+ZMgb69dcxZbKa4B46NJp8uz1Rih0T/vjjj3nxxReJjY3l999/58cff+THH3/k999/JzY2lurVq/Ppp58+CVv/kzi6ashM1dr8kvsEeOHp686JfWcsdZmpWi4euUrVVyo/LTOfS57XuTEYwGiU4OBgLbAOShNnz+bes586Kad9Oyd6dnfkq7kqUlIKf474vK9VvPyynnr1ipd5p6hkZEhwcbZ9YcjJId/XlZ0t4fJl+8Qjiu0JHzx4kCNHjuDi4pLnmIuLC1OmTOHll19+LMY9CzTt0ojPFvazlBVKOSYTtB/S0lI3puUMzh68mKevi6cz3Ua349fle22O7+HjBkBybIpVfVJsCu7ebo9m/H8cMTcPh0YD1arp+e5bJeXKaXF3N/HbPgUXzsvw8zNfkBo00NO4sR4fXyN3oqWEhysZPUrDvPkZNmOrv+2Tc+WKjIWLbD+FMWSUhjOnrQfo29vJ8r+3t5Hwlfn3//13OZcvyfjiiyyb49evr2fj/zmwb6+c19/Qk5Qo4bs1Zo88McE+ySiKLcJubm7cuHGDGjVq5Hv8xo0buLm5FTiGTqdDp7OOvxhNhqeScflxc3jrcS4euWop95nelYToRDbP32Gpi49KzNNP46xm6paR3LwQxbeT88YoBY+OmJuHZ1SIlrDZajp3dEYqNVGpkpE3m+Zw5Z63+GbT3Nv9wEAjAYEGun/ozKlTMurWzevl3r0rYcECFbNmZVLQktHQoVp0D0QZegQ7My00k1KlzOIvtyERJ/+RETZLzRdDs6gQYDthZ/0GBvp/pOOrr9TMCAUHB+j2oY4zZ+RI7LRXrNgi3LdvX7p37864ceN46623LDHh2NhY9u7dy9SpUxk0aFCBY4SGhjJp0iSrukBJdSrK8hf2kow2PQttepZVOS0xg+hrsTb7qJ1UTNsWQmaalont52DQ2741S4xJBsDN29XyP5jTtl87FfnI9v+XEXPz8PiVMfHlV5lotZCZac6oPGWyGh/f/AXOz8+Eq6uR6ChpviJ85bKM5CQpAz5ytNQZjRLOnJaxebMD23emIZNhWfR7EG9vY74Lc/c5dUrG2DEaBnySRfPmhS+Gtu+QzQfts0lIMGeSjomREr4cfG28tidNsUV48uTJODo6Mnv2bIYOHWrJJ2cymfDx8WHkyJGMGDGiwDFCQkIYMmSIVV07jz7FNeWZROOsZvqvIeTo9ExoO7vQFfSY63dJuJNEnTdrEHHvi61xVlPlpSB+WbL7aZj83CDmJi9qNajVJtLS4NhROf0+yv9WPy5OQmqqxOZCXZ26epaFp1vVzZ6lppy/gU5dsh96e9jJkzLGjtbQr38WLVsWfTeKRAKlSplt/W2fgtJeRipVekZEGGDkyJGMHDmS69evW21RCwgIKFJ/pVKJUmm9Mvo0QhEqRyV+QT6Wsk+AF4G1ypOWmE7crQR6T+2MZxkPZvdaaGkTWKs8AGonJW6lXQisVR59tp6bF6IAcFApcHTNXVINH70WMHtD90lLTEefY0DjrCZ0+2iUGgdm9liAxkWNxkUNQEpcKkaj+UMRfmYOK8au4+DPRwHYNG87XUe3JepqDDE37tJzYkcSopM4+LPtfbTPGmJuShZHj8owmcDf3+zdLl2iwr+cgRYtctBqYc1qJU1e0+PhYSQ6WsqyJSr8yhip3yA3TDF8qIZGjXNo0zYHjQYC/hUmUKlMuLiYrOpTU0Gvz43Nrt+QBkBiorlOKgU3N/NcnPzH7AG3bZdNk9f0ljZyuYn7S1YH/pQTvlzJytW5ceQf1znQ4CU9UgkcOCBn3Q8OjBuvfbb2Cd8nICAgj/DeunWLCRMmsGLFikcy7ElQuV5FwvaOt5QHhHUHYNeaPwjrswgPX3e8/EtZ9Vl8bKZV/6ZdGhNzI47ulcwhl9c7NmR4+McFnnfYW5M5vf88QXUCqPpyJQBWX/raqk1w0CBiI+MA8K9SBs0D4rE+bAsqRyWfL+qHk5uGswcvMbrljGd6H+q/EXNTssjIkBC+TEV8vPmWvUkTPb36ZCGXm3dPRETI2L1LQXq6OVRRr76eXr10VvHe6GgpKSnFC7ROnKDh9CnbsuTtbeT7H8we9a5dCrKyJPywVskPa3Odupq19Hw5N9PyOm7dslbXo0fkrP1eSU4OBFY0MHmKlpdeLvgHHk8SiekxZ+U8deoUdevWxWAo3haU5orOj9MMgeC5IPzGH/Y2QVAA/mXuFNqm2J7wli1bCjweERFR3CEFAoHguaXYItymTZtC09rfX6wTCAQCQcEUe2ecr68vGzduxGg05vt34sSJJ2GnQCAQ/CcptgjXq1eP48eP2zxemJcsEAgEglyKHY4YPnw4GRm2f3YYFBTEb7/99khGCQQCwfNCsUW4SZMmBR53dHTk9ddff2iDBAKB4HlCZNYQCAQCOyJEWCAQCOzIc5XeSPCQ2OvxUoJCMYg18Gce8e0SCAQCOyJEWCAQCOyIEGGBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7MhztUXtxcZV6DC0FZXqBuDp58HED8I4tMV2BgQPHzf6zwqmcr1A/IK82Tx/B4uHrnno8zu7OxI8oQP1mtXEq1wpUuJSObTlKKsmrCczVVtg3+4TOvBOn6Y4uTly7tAl5g0MJ/pqzEPbUhIxz09LKtUJwNPPnYntvyxwfhq1aUCr/s0IrFUehVJO5Pkovp3yfxzfffqhzu9dvhTdRrel9hvVcfdxIyE6ib0/HOCH0M3oc/J/PrazuyPB49tT738v4uV/f06PsWriT4XOaUkmMxNWr1Ry6ICC5GQJFYMMfPxpFi9UMWfBCJupYvcu64yd9RromT4j0+aYZ07L+OlHB65ckZGYIGXCpEwaNn74h6mnpkiYEarmeoSUtFQJrm4mXm1ofvi8o6Ptfmu/d+DIX3IirsmQy2HjlrSHtuFx8Fx5wipHFRGnI5k/eGWR2iuUClLiU1kbupGI00VL3Dgs/GOCx7XP95innzuevu4sG/kd/WsPJ6zPIuo3r83QpQMKHLPjsPdpM7AF8z5dzuBGY8nK0BG6LQSFUlEkm54VVI5K8/x8VrT5ebFxFY7vPcPY92fx6StjOfXHeSZvGkbF2uVt9pm9eyz/C34t32P+L/ghkUr5+tNw+tUeweLh39KyXzN6TelkczxPP3c8/dxZNnIt/euMIKzvYuq/XYuhS/sX6TWUVObOUXPiuJwRIVoWL0+nXn0Do0Y4Eh+X+5ja+g30/PBTmuUvZIxtAQbI0koIrGhk4GDbKen/TfeuTpw6mX/eIYnUxKsNc5g0JZPw1ekMG6HlnxMy5n2lLnBMfY6E117X816r7ALbPS2eK0/46M6THN15ssjtYyPjWDRkNQBv93zzkc9/49xtpnSaaynfiYhl5fh1jFw9EKlMitGQf6LBtoPfYe30TRzean563axeC1gftYRGrevz+/rDj2xXSeHozlMc3XmqyO0XD/vWqrxy3I+82qoer7xXl2sni5/t+Niu0xzbletFx1y/y4bK22jZvxnLRq3Nt495Tr+ylO9E3GXl+PWMXPVJgXNaktHp4MB+OROnaHmxpvkOILiHjr8Oy/llqwM9e+sAUChMNhN75keDl/U0eIxphJydodX7uWmkvL0NtHo/h5/WOxTQC7r3NNu/a0fJcGKeK0+4JOLoqiEzVWvzy+oT4IWnrzsn9p2x1GWmarl45CpVX6n8tMx8JpBIJGicVKQl2n7KX3FxdFWTlpReeMMH+7ioC5zTko7BYE5H7+BgLbBKpYlzZ3O90tOn5HT8wIk+PRyZ95WK1BT7JnNIiJdw8ICcmjXtly/uYSi2J3znzh0WLVrEgQMHuHPnDlKplMDAQNq0aUPPnj2R2StlqZ1o2qURny3sZykrlHJMJmg/pKWlbkzLGZw9eDFPXxdPZ7qNbsevy/faHN/Dxw2A5NgUq/qk2BTcvd0ezfj/GO2HvIfKScX+DX9Z6jqPbE2Xka0tZQe1A1VfDmLg1z0tdX1rDSfuVkKe8fwqetP6k7dZOvL7IttgntO2/Bq+7+FeRAlAo4Gq1fSs/U5JuXJa3NxN/L5PwYXzMvz8zBeW+g30NGqix8fHyJ1oKSvDlYwJ0fDVNxmPlLX467kq9u3J9VB1OhgbokH6gLv48zbrGG7oVDWHD8nR6SS88moOXwwrerijJFAsET527BjNmjUjKCgItVrNlStX6Nq1K9nZ2QwbNowVK1awY8cOnJ2dCxxHp9Oh0+ms6owmw1NJe/+4Obz1OBePXLWU+0zvSkJ0Ipvn77DUxUcl5umncVYzdctIbl6I4tvJG56Krf9l3uzckOCx7ZjwwZckx6Va6rct3WMlyqNWf8qfm45wcPNRS11CdFKe8Tz93Jm2dST7/+9vtq8o2vOxNc5qpv48/N6c/t8jvBr7MyJEy5ez1XTt5IxUaiKokpE33szhyhXzd/SNprneZkCgkYBAAz2DnTl9SkadusVL8vsgPXrqaN8xVxtGDHGkT78sXqhqe8yPPsmiW3cJUbelrFiuZMkiFYM+e3aEuFgi/Pnnn/PFF18wYcIEAL777jvmz5/PX3/9RVJSEk2bNmXs2LF8/fXXBY4TGhrKpEmTrOoCJdWpKKtRTPPtjzY9C216llU5LTGD6GuxNvuonVRM2xZCZpqWie3nYNDb/oAlxiQD4ObtavkfwN3blWunih/3/C/yRsdX+WJxP6Z2+Zp/9p21OpaWlEFaUm54QqfNJvluaoHz4+HrxuzdYzn/1xW++nh5kWxQO6mY9stIMtOymNhhboFz+izg52cibG4mWVrIyDSntZ82RY2vb/4hFl8/E66uRqKjpI8kwm7uJtzcc8syGXiWMlGmjO3Ys4eHOTZdrpwRZ2cTQz93pOuHOjw9n42nGxUrJnzixAmCg4Mt5a5du3LixAliY2Nxd3dn1qxZbNhQuFcXEhJCSkqK1V+AtGrxrX8G0TirCd0+Gn22ngltZ5Ojyymwfcz1uyTcSaLOm7kXKI2zmiovBXHhr8tP2twSzxudXmXoso8IDZ7Pke0nH3k8Tz93wvaM48qJ68zpu7hIqbo0zmpCfw0xz2m7sELn9FlCpQZPTxNpaXD8qJxXG+Yfb42Lk5CaKsHDzsJ3f7pynqEpKJYn7OXlxZ07dwgMDAQgNjYWvV6Pi4sLAJUqVSIxMe+t979RKpUolUqruqcRilA5KvEL8rGUfQK8CKxVnrTEdOJuJdB7amc8y3gwu9dCS5vAWubtTmonJW6lXQisVR59tp6bF6IAcFApcHTVWNqHjzavort7u1rq0hLT0ecYLAKs1Dgws8cCNC5qNC7m7TQpcakYjeZPUPiZOawYu46DP5tvmTfN207X0W2JuhpDzI279JzYkYToJA7+bHsP7bNInvmpUPpf89MJTz8PZvdeBJhDEMPDB7BoyBouHrlqec912mzLHl2VoxK1k8oy5vQPvwGs5+f+e+/p507Y7nHE3oxn6cjvcS3tYmmTdC8m7+nnzqydY5jVaxGXjl27J8CjUGqUzOxpe06fNY4dlWEygb+/kagoKcuXqvAvZ6B5ixy0WvhujZLGTfS4e5hjwsuXqvDzM1Kvfq5IjxymoWHjHFq3MSuiVgvRUbl+X0yMlGtXpTg7m/DyNr9PGemgy85d4PtqvvkuJjExt+7+jowjf8tJSpLwwgsGVGoTkTdkLF+ipHoNPT4+5jYXL0qZPUPNzNmZlCptrrsbKyEtTcLduxKMRrh21WyTXxkj6oJ3tz0RiiXCbdq0YcCAAcyePRulUsmUKVN4/fXXUd+z/NKlS5QpU+aJGPo4qFyvImF7x1vKA8K6A7BrzR+E9VmEh687Xv6lrPosPjbTqn/TLo2JuRFH90qDAHi9Y0OGh39c4HmHvTWZ0/vPE1QngKovVwJg9SXrkE1w0CBiI+MA8K9SBs0Dwr4+bAsqRyWfL+qHk5uGswcvMbrljP+UxwVQuV4gYXvGWcoDwsx3XbvW/EFY3yV4+Ljh5e9pOf5un6bIFXIGfdObQd/0ttTfbw/QYUhLgsd9UOB5gysNJjYynrpvvUiZSj6UqeTDDzcWWLVp7tAVALlChv8Lfig15m1QQXUq5M7pxa/yHfdZJCNDwsrlKuLjJTg7m2jURE+v3lnI5ebdE9cjZOzepSAj3RyqqFtfT4+eOhwe2B12J1pKakqu6F6+JGPE0NxfUSxZZL44/q95NsNGmkN6ixbk/RHIv9m51xzzd3AwsX2bA0sWqsjJgdKljTRqoqdTl9yYsi5Lwu1bMh6MDq1ZpbQ6xycfOQEwa04GtWo//TCSxFSM1Mjp6en06dOHjRs3YjAYePXVV/nuu+8ICAgAYNeuXaSkpNChQ4diG9Jc0bnYfQRPCfFQ9xLL0giRVLckU6HsnULbFEuE75OVlYVer8fJyemhDMsPIcIlGCHCJRYhwiWboojwQ/1iTqVSFd5IIBAIBIUiXByBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAjQoQFAoHAjjxXD3UXCP5rPJtPLBY8iPCEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAjQoQFAoHAjggRFggEAjvyXG1Re7FxFToMbUWlugF4+nkw8YMwDm2xnZ3Cw8eN/rOCqVwvEL8gbzbP38HioWse+vzO7o4ET+hAvWY18SpXipS4VA5tOcqqCestmSBs0X1CB97p0xQnN0fOHbrEvIHhRF+NeWhbSiLm+WlJpToBePq5M7H9lwXOT83Xqlo9BP4+nfw/tmTCKA7e5UvRbXRbar9RHXcfNxKik9j7wwF+CN2MPsf2w75n7x5LrderWdX9snQP8wauKLYNJYXMTFizUsnhAwqSkyVUDDLw0adZvFAl76a4b+aq+PUXB/p/kkXbD7JtjmkwwPdrlOzboyAp0ZwK6X9vZ9Plw2wkEpvdbJKaImFWqJrrEVJSUyW4uZl4taGeHn2ycHS03a9HVyfuxlr7n736ZtGxi23bnyTPlQirHFVEnI5k56rfmbBhaKHtFUoFKfGprA3dSLvP3i3SOYaFf0zsjTi+nZI3156nnzuevu4sG/kdkRei8C5XisEL+uLp68GUznNtjtlx2Pu0GdiC2b0XEnMjjh4TOxK6LYS+NYf9p7JrqByVufPz05Ai9+tVfYjVRSz5bqrNtrN3j2XXmv3s/nZ/nmP+L/ghkUr5+tNwoq7FUqF6Wb5Y1A+VRsmyUWsLtOHX5ftYPeknS1mXaZ8v9OPi6zlqblyXMixEi6enkX17HBg9wpEl4emWNEEABw/IuXhBhqdn4TuWf1rnwLYtCoaOzKJ8BQOXL8mYO1uNoyO0bpf/+9WjqxNDR2ipmU/GC4nUxCsNc+jey4Crm4noKCkL56lIS1MzckzBTk1wzyxavJf73dGo7ZeG6rkS4aM7T3J058kit4+NjGPRkNUAvN3zzUc+/41zt5nSKVds70TEsnL8OkauHohUJsVoyP+D3HbwO6ydvonDW48DMKvXAtZHLaFR6/r8vv7wI9tVUji68xRHd54qdr/ku6lkpGQ+8vmP7TrNsV2nLeWY63fZUHkbLfs3K1SEszJ1D+V9l0R0OjiwX86EKVperGkWvw976Pj7sJxtWx3o0ducPig+TsKib1RMm5nJ+NGagoYE4MI5Ga801PPSK+Y8dN4+ev74Tc+liw8XFXV2hpbv5wqpt7eBlu/nsGF9wemRANSa3Fx19kbEhO2Mo6uGzFStTQH2CfDC09edE/vOWOoyU7VcPHKVqq9UflpmlmgWHQ3lh8gFzPg1hGqvPt73xNFVTVpSeqHtmnZpxE/RS1j6z0x6T+2EUl24EJRUDAYwGiUoHKxFykFp4txZc0JeoxHCZqhp3zGb8hWK9ru9qtUNnPxHzu1bZtmJuCbl3BkZ9V/KP4NzcUmIl3DwgJwXaxY+3k8/ONCxjROffuTIhh8dMDz91HIWiu0Jz58/nyNHjvDuu+/SuXNnvv32W0JDQzEajbRr147Jkycjlz8/DnbTLo34bGE/S1mhlGMyQfshLS11Y1rO4OzBi3n6ung60210O35dvtfm+B4+bgAk/8vLSopNwd3b7dGMf8ZJjEnm60+Wc/n4dRRKOS16v0nYnrEMbjSeqydvANB5ZGu6jGxt6eOgdqDqy0EM/Lqnpa5vreHE3UrIM75fRW9af/I2S0d+X6Adv607ROzNeBLuJBH4Yjn6TOtM2cq+TO741eN4mU8djQaqVtPzw3dKypXT4uZu4o99Ci6el+HrZxbcn9Y5IJXZDiPkR8cu2WRmSujfyxGp1CzkPXrraNosVzS/mati3x6FpazTwbgQDdIH3MVN29Ksxp0xVc1fh+TodBJefjWHz4dlFWhH67bZBFUy4Oxs4vx5GauWq0hMkND/E12B/Z4UxVLLqVOnMmvWLJo3b84XX3xBZGQks2fP5osvvkAqlTJ37lwUCgWTJk0qcBydTodOZ/2CjSbDU0l7/7g5vPU4F49ctZT7TO9KQnQim+fvsNTFRyXm6adxVjN1y0huXoji28l548eCwrl9+Q63L+fm8Dr/1xX8Ar1p99k7zOq1CIBtS/ewf8NfljajVn/Kn5uOcHDzUUtdQnRSnrE9/dyZtnUk+//vb7avKDiP26/h+yz/3zh7i8Q7SczaNRbfQC/uRNx96NdnT4aFaJk7W82HnZyRSk0EVTLy+ps5XL0i48plKT9vdOCbxRnFWlDb/7uc3/YqGDFaS/kKRiKuyViyQHlvgc4cVgjuqeODjrnaMGKII737ZVGlqm1Xtf8nWXTrLiHqtpSVy5UsXaRi4Ge2hbhdh9wLR0BFI3K5Wfx79rXOFv20KJYIr1q1ilWrVtGuXTtOnTpFvXr1WL16Nd26dQOgSpUqjBgxolARDg0NzdMmUFKdirIaxTTf/mjTs9CmZ1mV0xIziL4Wa7OP2knFtG0hZKZpmdh+Dga97Q9YYkwyAG7erpb/Ady9Xbl2KvKR7f+vcenoNao3esFSTkvKIC0pw1LWabNJvpta4Px4+Loxe/dYzv91ha8+Xl5sGy4euQaAX0WfZ1aE/fxMzJ6bSZYWMjPNOxlCp6jx8TVy9oyc5GQJ3bvkJvo1GiUsX6xk8/85sHpt/uGb8KUqOnbW8UZTs+cbEGjkbqyE9T84WETYzd2Em3tuH5kMSpUy4VfGdvzWw8OEh4cJ/3JGnJxNDP/cka4f6vDwLFrMt0pVAwaDhLuxUsr6P/1HIhVLhKOjo6lfvz4AtWrVQiqVUrt2bcvxunXrEh0dXeg4ISEhDBlivfrdzqNPcUx5ZtE4q5n+awg5Oj0T2s4udHdDzPW7JNxJos6bNYi4J7oaZzVVXgrilyW7n4bJzxQVa5UnMSavZ1tUPP3cmb17LFdOXGdO38U8RDJyAmuVB3gkO0oKKjWo1CbS0uD4UTm9+2fRuImeOnWt465jR2po+r8cmrew/XnWZeVN3C2Vgukx6t796copxqaha1elSKUmXN3s80y6Yomwj48P58+fp1y5cly5cgWDwcD58+epXr06AOfOncPLy6vQcZRKJUql0qruaYQiVI5K/IJ8LGWfAC8Ca5UnLTGduFsJ9J7aGc8yHszutdDS5v4XSu2kxK20C4G1yqPP1nPzQhQADioFjq65K8Pho82r6O7erpa6tMR09DkGNM5qQrePRqlxYGaPBWhc1Ghc1ACkxKViNJo/QeFn5rBi7DoO/my+Zd40bztdR7cl6moMMTfu0nNiRxKikzj4s+09tM8ieeanQul/zU8nPP08mN3bHGpoO6gFMTfiiDx/GweVgha93qTWm9UJeTfUaky1U2528OkffgNYz8/9997Tz52w3eOIvRnP0pHf41raxdLm/s4HTz93Zu0cw6xei7h07Bq+gV407dyII9tPkpqYRsCL5RgwO5jT+y9w/cytJ/NGPQWOH5VhMkFZfyPRUVLCl6ooW85A8xY5yOXg4mp9cZLJwd3DZOVJjhqmoWHjHN5vY1bEl1/Vs+57JV5eJspXMHD1qoyNGxyshDsjHXTZuTGOufPNdzGJibl193c1HPlbTnKShMovGFCrTUTekLF8iZJqNfR4+5jbXLooJWyGmtDZmZQqbeLCORkXL8qoVVuPWm3iwnkZSxepePOtHJydH/ObWESKJcLdunWje/futG7dmr179zJixAiGDRtGQkICEomEadOm0b59+ydl6yNTuV5FwvaOt5QHhHUHYNeaPwjrswgPX3e8/EtZ9Vl8bKZV/6ZdGhNzI47ulQYB8HrHhgwP/7jA8w57azKn958nqE4AVV+uBMDqS19btQkOGkRsZBwA/lXKoHlA2NeHbUHlqOTzRf1wctNw9uAlRrec8Z/aIwxQuV6g1Y8vBoQFA/fmp+8SPHzc8PL3tByXO8jpP7Mbpcp4oMvUEXHmJqNaTOfUH+ctbToMaUnwuA8KPG9wpcHERsZT960XKVPJhzKVfPjhxgKrNs0duprPqZDh/4IfSo05eKjP1lOnaQ3aDmqBylFJ3K1EDmw+wtrpmx/pvbA3GRkSVi5XER8vwdnZROMmenr0zqI4a+53oqWkpuS6vh8PymLNSiULvlaRnGwOcbzbMoeuwbkx4MULVOzZVXBgdvte8z5wpYOJHdscWLpQRU4OlC5tpGETPR275I6ny5Jw+5bMsvtBoTDxx29yvl+tJCcHvH2MtP0gm7bt7bevW2Iqxv2W0WhkxowZHD58mIYNGzJq1Ch+/PFHRowYQWZmJq1atWL+/Pk4FvRzFRs0V3Qudh/BU+Lf95CCEsPiiIIXDQX2JbDsnULbFEuEnyRChEswQoRLLEKESzZFEWHx7RIIBAI7IkRYIBAI7IgQYYFAILAjQoQFAoHAjggRFggEAjsiRFggEAjsiBBhgUAgsCPPzzMnBQ+NRPoQuWcETwUDYm6edYQnLBAIBHZEiLBAIBDYESHCAoFAYEeECAsEAoEdESIsEAgEdkSIsEAgENiR52qL2ouNq9BhaCsq1Q3A08+DiR+EcWiL7ewUHj5u9J8VTOV6gfgFebN5/g4WD13z0Od3dnckeEIH6jWriVe5UqTEpXJoy1FWTVhPZqq2wL7dJ3TgnT5NcXJz5NyhS8wbGE701ZiHtqWkUaNxFTp88R6V6gTg6efOxA5fcnjrcZvthy77iObBr+Wpjzx/m/51Rxb7/N7lS9E1pC2136iGu7cbCXeS2PfDQX6YsRl9TsH50Ku+HETPSR2p0qAiBoOJiNORjG45g+ysZ/eh+5mZ8N1KJYcPyElJlhAYZKT/p1lUrpI3BdD8uUp2/OJAv0+yaP1B0V7zTz84sHq5kvfbZdP/04fLcnz7loQFc1XcipSSkSHBo5SJN5rm0KV7dqEPn9+zQ87mDQ5E3ZaicTTR+DU9H3/2DGRbftZROaqIOB3JzlW/M2HD0ELbK5QKUuJTWRu6kXafvVukcwwL/5jYG3F8OyVvBmVPP3c8fd1ZNvI7Ii9E4V2uFIMX9MXT14MpnefaHLPjsPdpM7AFs3svJOZGHD0mdiR0Wwh9aw77z2TXUGmURJy5yc7VfzBh/ReFtl80dA0rxq6zlGVyGYuOTGf/xr9t9pm1awy7v/2T3d/uz3PMv7IfUqmErweuIPpaDBWq+fP5wr6oNEqWhay1OWbVl4OYtmUk62ZvYeEXqzHojQTWLIfJWCIe0/3QfDNHReR1KUNDsvDwNPLbHgVjR2hYGJ5BqdK5r+3QATmXLsjw8Cx6frbLF6Xs+EVBhcCCL24ALd9yJvz7dEu6ogeRy6Bp8xwqVjLi5GTi+jUp38xRYTRCj762M2Vs+knBpp8c6P2RjheqGsjSSrgba7/91s+VCB/deZKjO08WuX1sZByLhqwG4O2ebz7y+W+cu82UTrlieycilpXj1zFy9UCkMilGQ/4f5LaD32Ht9E0Wz3BWrwWsj1pCo9b1+X394Ue2qyRwbNcpju06VeT2malaq7uHV1vVw8ndkV1r8gpskc6/+zTHdp+2lGOux7Hhq2207N+sQBH+aFYwmxfuZH3YVkvd7SuFP8i7JKPTwcH9csZN0VKjplkou/XI5shhOdu3KgjubRa4+DgJS75RMnmmlkmj1UUaW6uFsOlqBg3JYt33j5Zf3sfPhI9fbsJRL28Dp0/qOXfGdr7K9DSzhz9uqpbade9fBEwEVHwkUx4JERO2M46uGjJTtTYF2CfAC09fd07sO2Opy0zVcvHIVaq+UvlpmVniadHzDf7Zd467N+Mf25iOrhrSEvNP3w7gWtqFqi8HkXw3lbm/TWBd5EJm7x5L9YbP9rwYDOYU9op/aaRSaeLcWbPAGY3w5QwV7TpmU75C0b3gRV+raPCKntr1CveCi0t0lIQTR2W8WMv22P8cl2M0QkK8hAG9NPTo5MiMySri7j4jnvCgQYPo2LEjTZo0eaST6nQ6dDrr+IvRZHgqGZcfN027NOKzhf0sZYVSjskE7Ye0tNSNaTmDswcv5unr4ulMt9Ht+HX5Xpvje/i4AZB8L9vvfZJiU3D3dns04/8jePi60eDtWszoYZ2cs/OI9+k8orWl7KB2oOpLQXw6t4elrl+dEcTdSsgzpl+gN60/bl6gF+wbYM4sHjy2HctC1nLtVCTNujVhxvbRfFR3JNHXYh/1pdkFjQaqVDOw7jsH/Mtl4eZuYv8+ORfPy/D1M4cFNqxzQCaD99sVPRz2xz45165Kmbsw02abCaPUeTzZT/o4Wn6c7eVtZOEK6/7DBmm4dkVKTo6EFu9l062n7VBEzB0pJhP8tNaBfp/qcHQ08e1KJeNGqPlmWSYKRZFfzmOjWCK8YMECFi5cSMWKFenTpw89evTAx8en8I7/IjQ0lEmTJlnVBUqqU1FWo9hj2ZvDW49z8chVS7nP9K4kRCeyef4OS118VGKefhpnNVO3jOTmhSi+nZw3fiwoOv/78DXSkzPzLLJuW7aX/RtyY8QjV33Cgc1HObj5qKUuITopz3iefu5M2zqC/Rv/ZvsK2zncpPeeqfFr+D5LGOTaqUhqv1mdt3u+wcpxPz7S67InQ0O0fD1bRY9OTkilJipWMvLam3quXpFy9bKULRsVfL04E0kRHci4uxKWLVAyZZYWhwKiEIOGZpH9gIb27+7ExOlaPEuZve38FtxGjtOizZRwPULKiiVKNq430b5z/kJsMoJeL6H/QB1165s95hFjsgju4MjpkzLqNXj8HnphFDsmvGvXLrZu3UpYWBjjxo3jnXfeoV+/frz77rtIpUWLboSEhDBkyBCrunYefYprSolAm56FNj3LqpyWmFGgF6R2UjFtWwiZaVomtp+DQW974hNjkgFw83a1/A/g7u3KtVORj2z/f4G3e7zO3rUH8uxiSEvKIC0pw1LWZWWTHJdKdITtufHwdWPWzjGc/+sKX38SXuB5E+7NR+SFKKv6Wxej8fL3LOarKFn4+pmYMVdLlhYyM83p6WdOUeHja+LcGRkpyRJ6dcnNqm40SghfrOTn/3NgxdqMPONdvSwjOVnKZwM0Vn3OnTbxy2YFm3akI5Nhteh3Hy9vY74Lc/cp7WUCTJSrYMRogPlzVbTtkI0snxtr93sLiOXK54ZQXN1MuLiYiLsrBZ4BEX7xxRd56623mD17Nps2bWLFihW0adMGb29vevbsSa9evQgKCipwDKVSiVKptKp7FkMRD4PGWc30X0PI0emZ0HZ2obsbYq7fJeFOEnXerEHEPdHVOKup8lIQvyzZ/TRMLtHUfK0qZYJ82LHq90cey9PPnVk7x3Dln+vM6beEwhKRx96IIz4qkbKVfa3qy1TyKdYiY0lGpQaV2kR6Gpw4KqdXfx0Nm+RQq661WI0fqabp/3Jo1iL/z3OtunrmL7cW569nqyjrb+SDzvkL5sNgNIFeD7amrlp1s923b0kpVdr8f1oqpKZK8PIuemz7cfLQuyMUCgUdO3akY8eO3Lx5kxUrVrBq1SpmzJiBwfD0ryZFQeWoxC8oN3ziE+BFYK3ypCWmE3crgd5TO+NZxoPZvRZa2gTWKg+A2kmJW2kXAmuVR5+t5+Y978dBpcDRNffqHj7aHEN093a11KUlpqPPMaBxVhO6fTRKjQMzeyxA46JG42JeVU6JS8V4b1tT+Jk5rBi7joM/m2+bN83bTtfRbYm6GkPMjbv0nNiRhOgkDv5se4/zs4bKUYlfxQfmpkJpAmuWJy3JPDe9pnSilJ87s/sstur3ds83uPD3VSLP3853TLWTylIODZ4PWM/N/ffd08+d2bvGcvdmPMtGrcW1tIulTdK9eLynnzszt49mdp9FXDoWAcCGudsIHvcBEadvEnEqkmbBTfB/wY+pXb9+DO+K/Th+VAYmKONv5E6UlBVLlZQtZ6RZixzkcnBxtRYsuRzcPUyU9c9Vv9HD1LzaWE+rNjloNFAhwLqPUmXC2cVkVZ+Wag4X3Ofbn8wLo0mJ5jqp1Oy5Avy2R45cbh5XoTBx5bKM1cuVNHlDbwlbHDogZ81yBxavMseRy/ibeKVhDksXKBk0JAu1BlYvV1LW30jN2vbRrceyRa1cuXJMnDiRCRMmsGfPnscx5BOhcr2KhO0dbykPCOsOwK41fxDWZxEevu54+Zey6rP42Eyr/k27NCbmRhzdKw0C4PWODRke/nGB5x321mRO7z9PUJ0Aqr5cCYDVl6y/pMFBg4iNjAPAv0oZNA8I+/qwLagclXy+qB9ObhrOHrzE6JYz/jN7hAEq1wtk9q6xlvKA2cEA7Pp2P3P6LcHDx43S/7rF17ioadymAYuHfZvvmO2/eI/gsR8UeN7uL3xGbGQ8dd96kTJBPpQJ8mFtxHyrNm+rugEgV8jwf8EPpSb3Lm7T/B0oVAoGzP4QZ3dHIk7fJOS9UO5E3C36iy+BZGZIWL1cSXy8BGdnEw2b6OneW1fojyAeJCZaSmpK8XYdTJuo5uwp2yfx8jZawh0ymXmBMPq2ebHNy9tIyzY5tGmfGw/OTIfbt6zd7CGjsli2UMnE0RqkEqhRS8+kGVnFem2PE4mpsHuuBwgICODYsWN4ej7+eFdzRefHPqbg8SB5XPeKgsfO/Ku2Fw4F9qdS2ehC2xRL+69fv/7QxggEAoEgL+LHGgKBQGBHhAgLBAKBHREiLBAIBHZEiLBAIBDYESHCAoFAYEeECAsEAoEdESIsEAgEdkSIsEAgENgRIcICgUBgR4QICwQCgR0RIiwQCAR2RIiwQCAQ2BEhwgKBQGBHnquU9y82rkKHoa2oVDcATz8PJn4Qlicv2YN4+LjRf1YwlesF4hfkzeb5O1g8dM1Dn9/Z3ZHgCR2o16wmXuVKkRKXyqEtR1k1Yb1V+vb86D6hA+/0aYqTmyPnDl1i3sBwoq/GPLQtJY0ajavQ4Yv3qFQnAE8/dyZ2+JLDW4/bbD902Uc0D34tT33k+dv0rzuy2Of3Ll+KriFtqf1GNdy93Ui4k8S+Hw7yw4zNedIm/ZuqLwfRc1JHqjSoiMFgIuJ0JKNbziA769l93nNmpjk1/OEDclKSJQQGGen/aRaVq+TNPjF/rpIdvzjQ75MsWn9g+zWvX+vA4QNybt+U4qA0UbWagZ79dVYPgi8Ot29JWDBXxa1IKRkZEjxKmXijaQ5dumcX+GzgyxelrFqu5NplGUigchUDvfrrCKxon8waz5UnrHJUEXE6kvmDVxapvUKpICU+lbWhG4k4XbR8bsPCPyZ4XPt8j3n6uePp686ykd/Rv/Zwwvoson7z2gxdOqDAMTsOe582A1sw79PlDG40lqwMHaHbQlAo7ZAa9gmh0iiJOHOT+Z+vKlL7RUPX0Ln8J5a/bhUHkZqQxv6Nf9vsM2vXGP6Xj3AD+Ff2QyqV8PXAFfSvO4Ilw7/jvb5v0WtypwLtqPpyENO2jOT4njMMbjyewY3GsWXRLkzGhxOWksI3c1ScPC5jaEgW85dnUKe+nrEjNMTHWT+k/dABOZcuyPDwLFzAzp6W8d772YTNz2TKLC16g4RxIzRkFeB/tHzLmdiY/B8ML5dB0+Y5TJ6lZcnqDPp/ksXObQq+X2U7k6hWCxNGafDyMjFnQQazvs5ErTYxfqQavb7Ql/BEeK484aM7T3J058kit4+NjGPRkNUAvN3zzUc+/41zt5nSaa6lfCcilpXj1zFy9UCkMilGQ/4f5LaD32Ht9E0Wz3BWrwWsj1pCo9b1+X394Ue2qyRwbNepYuVly0zVWt09vNqqHk7ujpasx8U+/+7THNt92lKOuR7Hhq+20bJ/swLT3n80K5jNC3eyPmyrpe72lTsPZUNJQaeDg/vljJuipUZN811Atx7ZHDksZ/tWBcG9zZkr4uMkLPlGyeSZWiaNVhc67uQZ1mr7xYgsun3gxNUrMst5ioOPnwkfv1zl9PI2cPqknnNnbCchuH1TSlqahG49dZYEoV27ZzOwnyN3YyX4lXn6F8/nyhMuiTi6ashM1doUYJ8ALzx93Tmx74ylLjNVy8UjV6n6SuWnZWaJp0XPN/hn3znu3ox/bGM6umpIS0y3edy1tAtVXw4i+W4qc3+bwLrIhczePZbqDZ/teTEYzJmQFf9yKJVKE+fOmgXOaIQvZ6ho1zGb8hUe7jY+417eTyfnxyN80VESThyV8WIt24Jext+Ii4uRXdsV5OSYLzi7tivwL2coMKPzk+S58oSfBE27NOKzhf0sZYVSjskE7Ye0tNSNaTmDswcv5unr4ulMt9Ht+HX5Xpvje/i4AZB8L9nkfZJiU3D3dns04/8jePi60eDtWszoscCqvvOI9+k8orWl7KB2oOpLQXw6t4elrl+dEcTdSsgzpl+gN60/bl6gF+wb4AVA8Nh2LAtZy7VTkTTr1oQZ20fzUd2RRF+LfdSXZhc0GqhSzcC67xzwL5eFm7uJ/fvkXDwvw9fPLFQb1jkgk8H77R4u7m00wrIFKqrV0Fsl+pwwSp3Hk/2kjyP3AxJe3kYWrsi0Oj5skIZrV6Tk5Eho8V423XpmYwuNBqZ/qWXaeDU/fme+yviVMTJ5pvaxZXwuLsUS4RMnTuDu7k5AQAAA3377LYsXL+bmzZuUL1+egQMH0rlz4bnidDodOp3Oqs5oMjyTae8Pbz3OxSNXLeU+07uSEJ3I5vk7LHXxUYl5+mmc1UzdMpKbF6L4dvKGp2Lrf5X/ffga6cmZeRZZty3by/4NuTHikas+4cDmoxzcfNRSlxCdlGc8Tz93pm0dwf6Nf7N9he0cblKpWRp+Dd9nCYNcOxVJ7Ter83bPN1g57sdHel32ZGiIlq9nq+jRyQmp1ETFSkZee1PP1StSrl6WsmWjgq8XZyIpXh5PC4vmKYm8IWXW19aCOmhoFtkPaGj/7k5MnK7Fs5RZqPNbcBs5Tos2U8L1CCkrlijZuN5E+875C7FOB/PCVFStbmD4mGyMRti43oGJo9XMXZiJUplvtydKsUS4V69ezJkzh4CAAJYvX87gwYPp168fwcHBXLp0iX79+pGZmUnv3r0LHCc0NJRJkyZZ1QVKqlNRVqP4r8DOaNOz0KZnWZXTEjMK9ILUTiqmbQshM03LxPZzMOht3z4lxiQD4ObtavkfzGnbr50q2mLhf523e7zO3rUH8uxiSEvKIC0pw1LWZWWTHJdKdITtufHwdWPWzjGc/+sKX38SXuB5E+7NR+SFKKv6Wxej8fJ//Mlwnya+fiZmzNWSpYXMTAkeniZmTlHh42vi3BkZKckSenVxtLQ3GiWEL1by8/85WLIh22LRPCVH/5IzY24mpUpbhwD+XQaz91tQqOB+bLdcBSNGA8yfq6Jth+x8Pds/9iq4GyMh7JsspPeCscPHZNG5jRN/HZTzetOnvzpXLBG+cuUKlSqZU7YvXLiQr7/+mn79cm/FGzRowLRp0woV4ZCQEIYMGWJV186jT3FMeWbROKuZ/msIOTo9E9rOLjRtfcz1uyTcSaLOmzWIuCe6Gmc1VV4K4pclu5+GySWamq9VpUyQDztW/f7IY3n6uTNr5xiu/HOdOf2WUFgi8tgbccRHJVK2sq9VfZlKPsVaZCzJqNSgUptIT4MTR+X06q+jYZMcatW1vuCNH6mm6f9yaNbC9ufZZILF35i3vYV+mYmP7+OPwRpNoNebz5UfOh1IpFh58FIpSLDd50lTLBHWaDTEx8dTvnx5oqKieOmll6yOv/zyy0XKyKxUKlH+y+9/GqEIlaMSvyAfS9knwIvAWuVJS0wn7lYCvad2xrOMB7N7LbS0CaxVHgC1kxK30i4E1iqPPlvPzXvej4NKgaOrxtI+fLQ5huju7WqpS0tMR59jQOOsJnT7aJQaB2b2WIDGRY3GxbyqnBKXivHetqbwM3NYMXYdB3823zZvmredrqPbEnU1hpgbd+k5sSMJ0Ukc/Nn2HudnDZWjEr+KD8xNhdIE1ixPWpJ5bnpN6UQpP3dm91ls1e/tnm9w4e+rRJ6/ne+YaieVpRwaPB+wnpv777unnzuzd43l7s14lo1ai2tpF0ubpHvxeE8/d2ZuH83sPou4dCwCgA1ztxE87gMiTt8k4lQkzYKb4P+CH1O7fv0Y3hX7cfyoDEzmhaw7UVJWLFVStpyRZi1ykMvBxdV6MU4uB3cPk9We39HD1LzaWE+rNmZhXjRPyR97FYydokWjgaREsxJqHE2WMEBaKuj1uQr57U/mhdH7baVScHUzn+O3PXLkcqgQYEShMHHlsozVy5U0eUNvCVscOiBnzXIHFq8yhz1q1zOwYomERfOUtGqTg9EEG34wx7dr1i7+Do3HQbFE+J133mHRokUsX76c119/nQ0bNlCrVi3L8fXr1xMUFPTYjXxcVK5XkbC94y3lAWHdAdi15g/C+izCw9cdL/9SVn0WH5tp1b9pl8bE3Iije6VBALzesSHDwz8u8LzD3prM6f3nCaoTQNWXzXcSqy9Zf0mDgwYRGxkHgH+VMmgeEPb1YVtQOSr5fFE/nNw0nD14idEtZxTqRT9LVK4XyOxdYy3lAbODAdj17X7m9FuCh48bpf91i69xUdO4TQMWD/s23zHbf/EewWM/KPC83V/4jNjIeOq+9SJlgnwoE+TD2oj5Vm3eVnUDQK6Q4f+CH0pNrgOxaf4OFCoFA2Z/iLO7IxGnbxLyXih3Iu4W/cWXQDIzJKxeriQ+XoKzs4mGTfR0760r8EcQ/yYmWkpqSq6g/rrFvBAWMkRj1e7z4VqatTCHAaZNVHP2lO2TeHkbLeEOmcy8QBh9W4rJZD7Wsk0ObdrnxoMz0+H2rVwHz7+ckfFTtfzwrQPDBmmQSCEwyMCkGZl4eNrHFZaYCrvneoDo6GgaNWpEuXLlqF+/PosWLaJevXpUrVqVS5cu8ddff7Fp0ybefffdYhvSXFH4gp7APkjstWwsKJT5V20vHArsT6Wy0YW2KdY+YT8/P/755x9effVVduzYgclk4siRI+zatYuyZcty8ODBhxJggUAgeF4plif8JBGecMlFeMIlF+EJl2weuycsEAgEgseLEGGBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAj4lGWgsKRiGt1SSXH9JCPMROUGMS3SyAQCOyIEGGBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAjz9UWtRcbV6HD0FZUqhuAp58HEz8Iy5Mc8kE8fNzoPyuYyvUC8QvyZvP8HSweuuahz+/s7kjwhA7Ua1YTr3KlSIlL5dCWo6yasJ7MVG2BfbtP6MA7fZri5ObIuUOXmDcwnOirMQ9tS0mj07BWNGpTH//KvmRrczj/9xXCx6zj9hXbr7F81TJ0H/8BQXUq4FO+NIuHf8em+Tsf2gbvcqXoGtKG2m9Uw93blYQ7Sez74RA/zPw5T/66B3H3dqXv9M7UbVoDjbOaW5fvsG7WzxzY/OxmPtFmwtpVDvx9QE5KsoSAICN9PtFRqYo5o8a61Q4c+F1OfJwEuRwqVjLQrXc2lasabY7Zv5uGuNi8fl+L97P5aLDtDMm2iLolYfFXSm5FSsnMMOfBa9JUT6fu2QU+fP7KRSnfhjtw7bIMiQQqvWCge/9sAiratv1J8lx5wipHFRGnI5k/eGWR2iuUClLiU1kbupGI00VLqjks/GOCx7XP95innzuevu4sG/kd/WsPJ6zPIuo3r83QpQMKHLPjsPdpM7AF8z5dzuBGY8nK0BG6LQSFUlEkm54FajapwtbFe/j89UmEtJyJTC5j+i8jrbJY/BulxoE71++yYtx6Eu4kF+k8s3aO5n8fNsn3mP8LvkilEr4euIL+dUexZMT3vNe3Kb0mdyxwzOHLP8K/si8TO8zlo/ohHPz5GKO/G0TFe6mxnkUWzFFy6riMz0Zl8dWyTGrXMzBxhJqEePO+ZL+yRvoN1PHV0kymf6XFy8fEpJFqUpJtjzl7QSYr1mdY/ibONDsejV6zfYFr28yJuzH574WWyeCN/+mZMDOL+asy6f2Jjt2/ylm32sHmeFotTA5RU9rLxKz5ZtvVGpg8SoX+6ef4BJ4zT/jozpMc3XmyyO1jI+NYNGQ1AG/3fPORz3/j3G2mdJprKd+JiGXl+HWMXD0QqUyK0ZD/lbjt4HdYO30Th7ceB2BWrwWsj1pCo9b1+X394Ue2qyQwpvVsq/Kc/ktZf2shlepU4OzBS/n2uXz8OpePm3Ma9p5SsFAWhWO7z3Bs9xlLOeZGHBsq/0rLfm+xLOQHm/2qvVKJbwavsuSd+2Hmz7Qb9DaV6lR4JjNi63Rw+E85IZOzqF7T/Jns3CObo3/J2LFFQbfe2bz21oOKZaLXAB17tiuIjJBRs27+ourqZm57n43rZPj4Gale6+Fyu/n4mfDxy7XDy9vA2VN6zp+x/fzrqJtS0tMkdOmRTal7WZo7BWfzeX8NcbESfMs8/cerP1eecEnE0VVDZqrWpgD7BHjh6evOiX254pCZquXikatUfaXy0zLzqeN4LwHqgynr7WOHhrTE9ALbnP/rCq+3fxlnd0ckEgmvd3gFB5UDp/dfeEpWPl6MBnMKe4d/OZQODnDhbF6By8mBXdsUaBxNVKhYNEHNyYE/9ih4q0WOVebjR+FOlIR/jsqpXtO2DWX8jTi7mNizXUFOjvmCs2eHnLLljHj52Ce/xSN5whkZGaxfv56rV6/i6+tLly5d8PT0LLzjf4imXRrx2cJ+lrJCKcdkgvZDWlrqxrScwdmDF/P0dfF0ptvodvy6fK/N8T183ABIvpfx9z5JsSm4e7s9mvElFIlEwoDZH3L20KV8sygXh87DW9F5xPuWsoPagaovVeTTud0tdf3qjiLuVkKevn6BXrT++H8FesEA0z6cz+hvP2VD9GL0OXp0mdlM6vQV0c9osk+1Bl6oZmD9dw6ULZeFq7uJP3+Tc/mCFB+/XKE6+peML6eq0OnMmZYnztTi4lrAwA9w5KCcjHRo2tw6BjA5RMWFf3myg/tquK/Tpb2NzAu3Xj8ZNVhNxBUpOTkSmr+XQ5eetuPLag1MmaNlxgQVP31vDuf5ljExfoYWeyWQKZYIV6tWjQMHDuDh4cGtW7d47bXXSEpKonLlyly7do0pU6bw119/ERAQUOA4Op0OnU5nVWc0GZ5K2vvHzeGtx7l45Kql3Gd6VxKiE9k8f4elLj4qMU8/jbOaqVtGcvNCFN9O3vBUbH1WGPhVD8pXL8vQt6Y88ljblu9j///9bSmPXPUJBzYf5eDmo5a6hOikPP08/dyZtmUE+zceYfvK3ws8R48JH+Dk5sjId0JJTUjn1Vb1GPPdQIY2m8qNc492EbEXn43KYn6Yij6dHZFKTQRWMtL4TT3XruR+R1+sZeDLJZmkpkjY/auCsKkqZn6jxc29cI9yz3Y5dV8y4FHKuu2nQ3VkPyANn/RwZOw0LZ732snyUaxhY7PQauHGNRmrlzrw808K2nbKPxO5TmeOd1epbmDI6ByMRvj5JwemjlExe4EWpe0liCdGsUT44sWL6O9Fr0NCQvDz8+PkyZO4urqSnp5O27ZtGTNmDGvXri1wnNDQUCZNmmRVFyipTkVZjWKab3+06Vlo07OsymmJGURfi7XZR+2kYtq2EDLTtExsPweD3vbtU2JMMgBu3q6W/8G8Iv8sxhsL49O53Xn53doMbTaN+Ki84lhc0pIyrEIaOm02yXdTC/RSPXzdmLUjhPN/XeHrT1cUOL5vgBetP25O/7qjiLwQBUDEmZu82Kgy73/UjHmDVz3ya7AHvn4mpn2pJUsLmZnmnQdhU5T4+OSGzVRqsxfpW8bEC9V0fNJDw97tcj7omr8A3udurITT/8gYMSErzzHPUnkF3MvbVGCowBzbBf/yegxGWDRXyfvtc/L1bP/cJ+dujIQZ83RI7wVjvxidRXBbR44cktPkzae/OvfQMeHDhw8zceJEXF3N9x9OTk5MmjSJAwcOFNo3JCSElJQUq78AadWHNeWZQuOsJnT7aPTZeia0nU2OruAPbMz1uyTcSaLOm7kXKI2zmiovBXHhr8tP2tynyqdzu9Pw/XqMaBFKbGScXWzw9HNn9s7RXPnnBnP6L6WwPLhKjTlwajRatzMYjEikz/6Si0oNHp4m0tPgn2NyXmpoW6SMRsjJKTzAu2+HAlc3E/VfebgFuYIwGcGgB1vTpsuSIJViFYeWSkFyr689KHZMWHLP+qysLHx9fa2OlSlThri4wr88SqUS5b/8/qcRilA5KvEL8rGUfQK8CKxVnrTEdOJuJdB7amc8y3gwu9dCS5vAe9uM1E5K3Eq7EFirPPpsPTfveT0OKgWOrhpL+/DR5rsAd+/c4FhaYjr6HINFgJUaB2b2WIDGRY3m3gJUSlyq5YscfmYOK8au4+DP5lvmTfO203V0W6KuxhBz4y49J3YkITqJgz8/u/tQ/83Ar3rwZqdXmdjhK7TpWZb3LyMlk+ws84Vq+PKPiI9OYuX49QDIFTLKVS0DgMJBjqefO4E1y5GVnmXxdFWOStROKst5QrsvAKzn5/57f1+A796MZ1nID7iWdrG0SboXk/f0c2fmr6OY3XcJl45FcOvSHaKuxvDZ/F4sC/mB1IR0Gr5fj7pv1WB8uy+f1Nv1xPnnqAyTybyQdSdayuqlDpT1N9K0hZ4sLWxY60CDV/W4e5pIS5Hw688KEuMlNHw9V6THD1fxSiMD77bJdTSMRti3U84b/9Pn66mmpYJen6uQK9ab72KSEs11Uqnp3i4L+GOvHLkMygUYUCjg2mUZ34U70OgNvWWf8F8HZHwXrmT+ykwAatXTs3qpA0vnKXm3TTYmk4SN6xRIZVCj9uO/KBSFYovwW2+9hVwuJzU1lUuXLlGjRq6HFhkZWaIX5irXq0jY3vGW8oAw8+LMrjV/ENZnER6+7nj5l7Lqs/jYTKv+Tbs0JuZGHN0rDQLg9Y4NGR7+cYHnHfbWZE7vP09QnQCqvlwJgNWXvrZqExw0yOL9+Vcpg+YBYV8ftgWVo5LPF/XDyU3D2YOXGN1yRqFe9LNEq4+aARC2e4xVfVi/pez+7k8ASvt7Wnmcnr7uLPp7mqXc4Yv36PDFe5zaf4ERb08HoP3n7xI8tl2B5+7+whfE3oynbtMalAnyoUyQD2uvzbNq87Y6GAC5XIb/C34o1WYP2KA3MLZNGH2mdmLShiGonVREX4slrO9Sju489TBvRYkgM0PCt+EOJMRLcHY28UoTPd16mX8EYTTC7VtSftulIjVVgrOLiaDKRqbN1VKuQq47GRMtJTXF2r08fUJG3F0pb72T/2d35kQ1507bdshKextZ+r1ZUGVS2PijgujbSjCZj73bOodW7XPHzsyQEHUr946kbDkTo6dm8eMaB0YN1iCVQkCQgfGhWjw87bM7QmIq7H7rAf4dx33llVd4++23LeXhw4dz+/Ztfvih4NXk/Giu6FzsPoKng0T+3/lRyH+NuZf32dsEQQFU848qtE2xRPhJIkS45CJEuOQiRLhkUxQRfvZXDgQCgeAZRoiwQCAQ2BEhwgKBQGBHhAgLBAKBHREiLBAIBHZEiLBAIBDYESHCAoFAYEeeq4e6CwT/NQw8pofxCuyG8IQFAoHAjggRFggEAjsiRFggEAjsiBBhgUAgsCNChAUCgcCOPFe7I15sXIUOQ1tRqW4Ann4eTPwgjENbbD8Y3cPHjf6zgqlcLxC/IG82z9/B4qFrHvr8zu6OBE/oQL1mNfEqV4qUuFQObTnKqgnryUzVFti3+4QOvNOnKU5ujpw7dIl5A8OJvhrz0LaUNDoNa0WjNvXxr+xLtjaH839fIXzMOm5fsf0a3+n1Bs26NaZ8tbIAXP3nOisn/GRJPV9cAl8sR8dhLanRsDIuns7ERsaxbfk+Ni/YVWC/MkE+9JvemWqvVkbuIOf62ZusmfR/nHpGsy0DaDNh3SoH/j4gIzVZQoUgI70/ySaoivn5wD+uVnDwdzkJcRLkcgisZKRL72wqV7WdnmLnFjk7tyqIizXv6PAvb6R9cA51X3q4h6lH3ZKw9CsltyOlZGaAu6eJJk31dOieY3mouy1+2yln6wYFd25LUDvCq6/p6TfYdoLQJ8lz5QmrHFVEnI5k/uCVRWqvUCpIiU9lbehGIk4XLZ/bsPCPCR7XPt9jnn7uePq6s2zkd/SvPZywPouo37w2Q5cOKHDMjsPep83AFsz7dDmDG40lK0NH6LYQFMr/ziMmazapwtbFe/j89UmEtJyJTC5j+i8jUWpsZ16s+VpVflt/mBEtpvPFG5OIu53I9K0j8PRzt9ln9cUvqdmkSr7HgupUIDkulZm9FtO/7ih+mLmFXpM78v6AZgXaPnnjEKRyGSPfCWVgw3FEnL7F5I1DrbJ3PGssmqPk1HEZg0fpmLNMS616BiaPUJEQbxZQv7JG+g7U8eVSLVO/0uLlY2TqSBUpybbH9Cxt4sO+2cxaqGXmQi016hiYNV7JrRu2t9m1b+bI3Zj8j8tl8Pr/9IybmcW8VVp6fZLNnl8V/Li64O/F1g1yflihoG3nbOaGaxk/S0vt+vbJqgHPmSd8dOdJju48WeT2sZFxLBqyGoC3e775yOe/ce42UzrNtZTvRMSycvw6Rq4eiFQmxWjI34toO/gd1k7fxOGtxwGY1WsB66OW0Kh1fX5ff/iR7SoJjGk926o8p/9S1t9aSKU6FTh78FK+fWb2WmRVnvvxchq1aUCdN6qxZ+3BYtuwa81+q3LMjTiqvlyJRq0bsGXxnnz7uHg6UbaSL3M/Xs71s7cAWDHuR94f0IwK1cpa0iI9S+h08NefMkZO1lGtpvkz2alHDsf/krFri5wuvXNo8pa1aPUYkM3e7QoiI6TUrJv/57j+q9Z9uvbOYddWBZcvyPCvUPwEm95+Jrz9cvuV9jZw7pSeC2dkQP6ZO9LT4IeVDoyakmVlZ4VA+4nwY/GES8hz4Z9JHF01ZKZqbQqwT4AXnr7unNh3xlKXmarl4pGrVH2l8tMy86njeC/33oOZkgtDqVEiV8iK1adQO1zVpCWl2zyempDOrUvRNOvaGKVGiVQm5b2+TUmKTeHKP9cfmx1PE6MBjEYJCgfr77WDA1w4mzf1UE4O7N4mR+NookLFomXLNBjgwG8ysrKgcrXHI4B3oiScPCqjek3b450+LsNkhMR4KZ/1VtO/s5o5k5XE37Xfj14eiyesVCo5deoUVas+HxmTH6Rpl0Z8trCfpaxQyjGZoP2Qlpa6MS1ncPbgxTx9XTyd6Ta6Hb8u32tzfA8fNwCS/+VRJcWm4O7t9mjGl1AkEgkDZn/I2UOXiDx/u8j9+kztRMKdJE7sO2epGzyvJ027NLKUlRoHpv483Oqi16Z0P/Kj2iuVeL39y4xrO6fA8456bwYTfvyczXFLMRlNJMelMqb1bNKTM4tse0lCrTEL44bvHChbToeru4mDv8m4fEGKj1+uMB/7S8ZXU5XodODuYWL8zCxcConAREZIGDNYTXa2OZPziIk6/Mvnjjk1RMnFM9ZC/0VfteV3gaW8TXwVbr1+MnqwiutXpOTkSPjfezl06mk792LsHSkmE2z8QUHvT7LROJr4YaUDk0eqmLNUi8IOEb5iifCQIUPyrTcYDMyYMcOS5PPLLwvOMqvT6dDpdFZ1RpPhqWRcftwc3nqci0euWsp9pnclITqRzfN3WOrioxLz9NM4q5m6ZSQ3L0Tx7eQNT8XWZ4WBX/WgfPWyDH1rSpH7dBzWkjc6vMLwt6dbJUBdPWUjG7761VKetWsM4WN/5NIDc5Yf5auVZcL6z/lu2mZO7D1bsL1ze5Acl8bQZlPJ1mbToucbTPq/IQxuPJ7EmGcvHAEweJSOhWFK+nfWIJWaCKxkpNGbBiKu5N4816hlYPYSLWkpEvb8KufLqUpCv9Hiajskj5+/idlLtGRmSPhrv4z5s5RM+lJrEeKPh2aT/YA0DOqhYfS0LDxKmY/nt+A2ZKwOrRYir0lZs9QB758UtOmUvxAbTeZszr0/1VniwJ+PyaJfRw3nTsqo3eDphyWKJcJfffUVtWrVws3NzareZDJx4cIFHB0dkUgKd+tDQ0PzJA0NlFSnoqyGjR4lF216Ftr0LKtyWmIG0ddibfZRO6mYti2EzDQtE9vPwaC3PfGJMckAuHm7Wv4Hc8r2a6eKtlj4LPHp3O68/G5thjabRnxUUpH6tP/8XToNbcmo92Za4rL3SYlLJSUu1VI26A0kRCUSHXHX5njlqvgx89dRbF/xGz/M/LnAc9d+oxovvVuH9r4fkZlm/hzM/3w1dd+qQbMPm7A+7JcivYaSho+ficlfZpGlBW2mBHdPE19OUeLtk3sHoVKDbxkTvmVMVK6WzcAeavZuV9Cuq21PVKEw9wETFSsbuXpJxq8bFXz0hXlngmepvKHN0t4mvHxshzxLeZmP+Zc3YDRms3iuklbtc5Dl49O5e9xvm/s6XN3A2QXi7BSSKFZMePr06aSkpDBu3Dh+++03y59MJmPVqlX89ttv7NtXeOLBkJAQUlJSrP4CpM9HKEPjrCZ0+2j02XomtJ1daNr6mOt3SbiTRJ03cy9QGmc1VV4K4sJfl5+0uU+VT+d2p+H79RjRIpTYyLgi9ekw5D26jmrNmNazuXLi0WOw5auWYdaO0ez+/gCrJhZ+h3J/94bRaC0SRqMJaREckpKOSm3e+pWeBiePyWjQ0LbDYDKa48PFwWQqfp+CMBolGPTmcfOjSg2z/VG3cucmLdX8V9rbPmtbxfKER40axVtvvcWHH35Iq1atCA0NRfEQQRSlUolSab316GmEIlSOSvyCfCxlnwAvAmuVJy0xnbhbCfSe2hnPMh7M7rXQ0iawVnkA1E5K3Eq7EFirPPpsPTcvmLOoOqgUOLpqLO3DR68FsNqelJaYjj7HYBFgpcaBmT0WoHFRo7m3AJUSl2r5IoefmcOKses4+PNRADbN207X0W2JuhpDzI279JzYkYToJA7+bHuP87PGwK968GanV5nY4Su06VmW9y8jJZPsLPO3dPjyj4iPTmLl+PUAdBz6HsHjPmBmz4XERsZb+mjTs8jKMN/TalzUKNUOlvN8/rr5DuzB+bm/g6F8tbLM2h7CsT2n2Thvu6WN0WAkJT4NgBfqBzJ8+UeMfHcGCdFJXPj7CulJGQxf/hHfT9+MTpvNO73fwKdCaY7sOPXE3q8nzcmjMkwm8PM3EhMt4dulDpTxN/JmCz1ZWvi/tQoavGrA3dNEaoqEHT/LSYyX0PD13N0KE4ereLmRnnfamOu+X66gzksGSnmZ0GbCgX1yzp2SMnZG7v7ctFRzuOA+y9ab4+pJieY6qdSEq5v52P69MuQyKBdgRKGAa5elrA1X0PANgyVs8fcBGd+HOzBvpTmO7FfWRIOGelYuVPLRFzo0GhPfhzvg52+iRm377JAo9sJcgwYNOH78OJ9++in169fn+++/L1IIoiRQuV5FwvaOt5QHhHUHYNeaPwjrswgPX3e8/EtZ9Vl8bKZV/6ZdGhNzI47ulQYB8HrHhgwP/7jA8w57azKn958nqE4AVV+uBMDqS19btQkOGmTx/vyrlEHzgLCvD9uCylHJ54v64eSm4ezBS4xuOaNQL/pZotVH5r24YbvHWNWH9VvK7u/+BKC0v6eVx/lev7dwUCoY98NnVn2+nbqR76ZtAuDjsGCaBzcp8Nxvq4MBaNK2AW5eLjTr2phmXRtbjsdExtGjink9RKl2wP8FP+Rys9OQmpDOmNaz6TmxPTO3j0KmkBN54TYTO8wl4szNYr8PJYXMDPg+3IGEeAlOziZeaWKgS69s5HIwGiHqlpQ/dslJTZXg7GIOLUyZm4V/hdz5iY2WkJqSqw0pyRK+makkKVGCxhHKBxgZOyOLWvVyQwOzJ6o4f9q2Q1ba28ii782CKpPC5h8VRN+Wgsm8aNeitZ6W7XO/F5kZEqJvWd/wDxqpY9UiB0LHqJBIoFotA2NDswr9gceTQmJ6hP1l69at4/PPPycuLo4zZ85QrVq1hzakuaLzQ/cVPFkk8v/Oj0L+a4Rd/s3eJggK4EX/wnf3PJL2d+7cmcaNG3P8+HHKly//KEMJBALBc8kjO+Bly5albNmyj8MWgUAgeO54rp4dIRAIBCUNIcICgUBgR4QICwQCgR0RIiwQCAR2RIiwQCAQ2BEhwgKBQGBHnquHugseDomj2t4mCGxwVFvB3iYICuDFIrQRnrBAIBDYESHCAoFAYEeECAsEAoEdESIsEAgEdkSIsEAgENgRIcICgUBgR56rLWovNq5Ch6GtqFQ3AE8/DyZ+EMahLbazU3j4uNF/VjCV6wXiF+TN5vk7WDx0zUOf39ndkeAJHajXrCZe5UqREpfKoS1HWTVhPZmp2gL7dp/QgXf6NMXJzZFzhy4xb2A40VdjHtqWksaHI1rx4YhWVnW3rsTQ79Xx+bYv/4IvwaNaU6lWObzLlWLxmB/ZvMR21uqisPrEdLzLWT/Uf8Xkjayft8NGDzNV6wfSY0wbqtQNwGA0EnH2FmM6fG3JCFKSMRpMHPghk/O/ZZGRbMTJQ0qNt1Q07KSxJGvYNjeVs/usE/MG1FXQcZKbzXEP/5TJ5UM6EqMMyB2gTBUFr/d0xLPsw0mOPtvEzgVpxFzTk3DLQFADB9qNtU7tfPtcDr+vTifhtgG9zoRLaRm1W6ho0EZjY1RIiTWwuG/eRLwfznajTJWn8xzt50qEVY4qIk5HsnPV70zYMLTQ9gqlgpT4VNaGbqTdZ+8W6RzDwj8m9kYc307Jm5/M088dT193lo38jsgLUXiXK8XgBX3x9PVgSue5NsfsOOx92gxswezeC4m5EUePiR0J3RZC35rD/lPZNW5ciCLkg9z3waA32myr1DgQExnHn1uO89HUjkUaf9bPQ9n9wyF2rztss82a0J/Z/u2flnLmA0lc86Nq/UCmrv+MH7/azqJRP2AwGAmoXhaT0T75yorL3/+Xyclftbz3hTOlysm5c1XP9q/TUGok1H8/V7wC6ip493MXS7mw5/zfOptN3ffU+FSSYzLCH2syWD8+hT4LPXBQ5Z+JZ2arOAYs98DVO29mDaMR5EoJ9VqpuXxIl09vUKig7ntqSleQ46CScPt8DjsXpKFQSajdouC97p2mulKqXK4cqp2fXrag50qEj+48ydGdJ4vcPjYyjkVDVgPwds83H/n8N87dZkqnXJG5ExHLyvHrGLl6IFKZFKMhf9FpO/gd1k7fxOGtxwGY1WsB66OW0Kh1fX5fb1tQnjUMeiNJd1MLbwhc/ieSy/+Ys033Htf2sdmQmZ5VZBsA+k/tyM9L91p5y7ev2s60XdKIuqAn6BUlFRuYcz66esu48EcWd67ordrJFRKc3Isevfy3l/ze585882ECsVdz8K/hkH+nAnBQSXj7E2ezzedz0GXkvch5V1TgXTH36uDqLePyYR23z+UUKsJqZ2mxXt/jRMSE7Yyjq4bMVK1NAfYJ8MLT150T+85Y6jJTtVw8cpWqr1R+WmY+FcoEevH92VmsPDaNEYv7ULqMx1O3oePgFqy//CXz942l/cDmSGW2vyKupZypWj+Q5Pg0vvx1JD+cD2PWlmFUfznoKVr8aJSpKifyVDaJUWbRvXtdz+0LOQTWsxbKm2dz+ObDeJYNSGTnwjS0qbbvUvLjvmiqnJ+e5MReyyHqQg7+NQoPK/zflBS++TCe70YkceXv/D3tJ8Vz5Qk/CZp2acRnC/tZygqlHJMJ2g9paakb03IGZw9ezNPXxdOZbqPb8ety27FMDx83AJLvZQS+T1JsCu7ebo9mfAni4vHrzBm0ittXY/DwdqXb8FaE/TKcAU0mok1/uC9Fp8/fofPn71jKDmoHqtQL5JMZXSx1/RtNJC7KHBP8edk+rp6+SVpSBlVfqkivsW3x8HZl6bif8h3ft7w5fvzhiFYsm7CBiLO3eKvTq4Ru/IIBTSYRHXH3oex+mrzSXoMu08Syj5OQSs23/a8FO1L9DZWlTUA9Byo3VOLmLSPpjoH932bw08QUPpzthlRW+G27yWhi77J0ylSVU7p8ruSsn5DM7fPW4bTwTxPh3pAupWX0XVj8C/GCngloU4wYjdCoi4Zab9v2ghUqCW/2caRsVQUSCVw6pGPjtFTajXGh0stKm/0eJ8UW4QsXLvDXX3/x6quvUqVKFS5evMjXX3+NTqfjww8/pGnTpoWOodPp0Omsv1hGk+GppL1/3BzeepyLR65ayn2mdyUhOpHN83NvT+Oj8gb+Nc5qpm4Zyc0LUXw7OW/8+Hnj2N6zlv+vn4/i4vHrrDk5g9da12fn9wcfasxtq/5g/8+5C68jF/flwC8nOPjLCUtdQkyy5f+Ni/ZY2aDPNjB4zoesnLKJnGzr23MAidSsFr+u3s/uHw4BcO3MLeo0qcLbXRuxcuqmh7L7aXLhgI7zf+hoNcyZ0uXkxEbo2bs8HScPKS++ZRbiaq/lCnLpCnK8AuQs6ZfIzbM5VKhVeGhh1+J04m7q6TbTzar+ncHO6B+QgaUfJdJ+givOnmYdkD6ki9hthhs5WSaiLuXwx+oM3H1lVHtdlW9bjauUlx5YuPOtrCA90ciRjZklU4R37NhB69atcXJyIjMzk02bNtG9e3dq1aqF0WikefPm7Nq1q1AhDg0NZdKkSVZ1gZLqVJTVKP4rsDPa9Cy0DyzeaNOzSEvMIPqa7big2knFtG0hZKZpmdh+Dga9wWbbxHsi4ebtavkfwN3blWunIh/Z/pJKRqqWqGux+AV4PfQY6cmZpCdnWsrZWdmkxKVy53pckfpfOh6BXCHDu5xnvnHexHt3Jzcv37Gqv3nlDqXLPv1QysPw+8oMXmmvsQht6QpyUuMM/PVTpkWE/42bjwy1i4TkaAPUKnj83YvTuHY0m66hbriUsnay7ovtg7h6yfJdmCsObj7m/qUryMlMNnLwh0ybIpwffi8ouHHy6S14FytAM3nyZIYPH05CQgIrV66ka9eu9OvXj927d7N3716GDx/OjBkzCh0nJCSElJQUq78AadWHfhHPEhpnNaHbR6PP1jOh7exCdzfEXL9Lwp0k6ryZe4HSOKup8lIQF/66/KTNtRsqRyW+FUpbhM4eBL7oj8FgJDkuLd/jsTcTiL+TRNmK3lb1ZQK9uXsr4WmY+Mjk6ExI/hVRkEglmArY3JEab0CbZsLRw7Z8mEwmdi9O4/LhbDpPc7UI49PGZAR9TvF2qsRG6J/qIl2xPOFz586xZo15n2zHjh0JDg6mffv2luPdunVj5cqVhY6jVCpRKq1d/acRilA5KvEL8rGUfQK8CKxVnrTEdOJuJdB7amc8y3gwu9dCS5vAWuUBUDspcSvtQmCt8uiz9dy8EAWAg0qBo2vu7Uz46LWA2VO9T1piOvocg0WAlRoHZvZYgMZFjcbFHK9KiUvFeG9bU/iZOawYu46DPx8FYNO87XQd3ZaoqzHE3LhLz4kdSYhO4uDPtvc4P2v0ndSev3ee5u6tBDx8XAke+T4Gg5HfNx4BYNiCXiTcSbbc4ssVMsq94Gv+30FOKV83AmuURZuhs3i6Kkclasfcz1lov2UAuHvlbrVKiU/DaDRRtX4gL9QL4NSBS2jTs6jaIJCPpnRk309/kZ5i9qY9fdyYsWkIsz9ZweV/bgCwYf4ugke+T8S521w7e4v/dXoV/0o+TOu95Mm+YY+JoAYOHFqfiUtpKaXuhSOObs6k5v/MnmO21sTBHzKo3FCJk7uUpBgDv6803+IH1M0NRawbk0ylV5XUa2n+PO9elM75/TrajXHBQS0lPcm8kKfUSFAozaqvTTNieCDK8+kaTwBLW6nUHC64T/xNPQY9ZKWbyNaaiI0wd/YONMvYiW1aXEpL8Shr1pJbZ3M4sklLvVa5MeHjv2i5clhH52luAJzZm4VMnjvG5cM6zuzJosUg58fw7haNYkdd7m/glkqlqFQqXF1zxcbZ2ZmUFPt5LoVRuV5Fwvbmbv4fENYdgF1r/iCszyI8fN3x8rferL/42Eyr/k27NCbmRhzdKw0C4PWODRke/nGB5x321mRO7z9PUJ0Aqr5cCYDVl762ahMcNIjYSLN4+Fcpg+YBYV8ftgWVo5LPF/XDyU3D2YOXGN1yxn9qj3ApP3dGLe2Ls7sjKQnpnPv7Kl+0mEFKQjoAXmU9rPbeevq4sfD33LlsP/Bt2g98m9MHLzGi9Rxz3afN8/wA5N/0qBNC7K0EcrL1vN62AR+OaIXCQU7MzXg2Ld5jFSeWK2T4V/JBpckVn81L9uKgVPDR1I44uzkSce42o9t/xZ0bRQt52JtmHznx5/eZ7FqUTmaK+ccatVuoadTZ/PmTSOHuDT1n92WRlWHCyUNKQB0HmnRzRK7IdaGTYgxWOyb+2W4O0f0w2loP3v3MmRebmQV+0/RUbp21/Rl28ZLycbinpfzTpBRS7+aeY9VnSQCM3FoaMHu9f6zOICXWgEQmwd1Hxhs9HandIjcUoU01khRjHf479GMmqXfNfTzLynh/hAtVGj2deDCAxGQq6MbDmlq1ajFz5kxatGgBwNmzZ6lSpQpyuVnL//zzT3r06EFERESxDWmu6FzsPoKng9TVpfBGArvQ8dB5e5sgKIDelQ8U2qZYnvDHH3+MwZB7FalRw3ohbfv27UXaHSEQCAQCM8XyhJ8kwhMuuQhPuOQiPOGSTVE8YfGLOYFAILAjQoQFAoHAjggRFggEAjsiRFggEAjsiBBhgUAgsCNChAUCgcCOiEdZCgpF4uFubxMENtib9Hw8c+VZpXcR2ghPWCAQCOyIEGGBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAjz9UWtRcbV6HD0FZUqhuAp58HEz8I49AW29kpPHzc6D8rmMr1AvEL8mbz/B0sHrrmoc/v7O5I8IQO1GtWE69ypUiJS+XQlqOsmrCezFRtgX27T+jAO32a4uTmyLlDl5g3MJzoqzEPbYu9qVE/gPZ9XyOoelk8vV2Y/MlqDu+xfiJY8OD/0aLjSzi6qDl/4gbzJ2wiOtJ22qBug5rx4aD/WdXdirhL/xZzHtrOuo0rEzz4f5QL8iYnO4czR6+zbMY27kYl2exTsZofvYe/S+UXy2I0GDm46yxLQ38hKzP7oe14kpwe8jfZ8XkzWpd+y5fyPSpZyiaTiStzzpJ6OomKn1XDvV6pPH0eRBuVye31EaRfTMFkMKEqo6HioGooSxU939t9jNlGIlddIfNGGtroTNxqexL0eXWrNteXXiLhQN5cgKoyGmqE1i/0HFmxWs6PO4FECnUWNyq2jQ/Lc+UJqxxVRJyOZP7gwlMwASiUClLiU1kbupGI00VLqjks/GOCx7XP95innzuevu4sG/kd/WsPJ6zPIuo3r83QpQMKHLPjsPdpM7AF8z5dzuBGY8nK0BG6LQSFUlEkm0oiKo0DERfvsHDy5nyPd+j3Ou93b8Q3EzbxeYf5ZGVmM3VFHxQOBfsNNy7H0LXhFMvfsC6LCmy//fJMvMrkvw/au6w7ExZ15+RfVxnY+ivG9A7H1d2RcfODbY7n4eVM6Kp+3ImM5/MO8xnXdwXlgrwZOqNjgXbYk6oT61Br3iuWv8ojXgTA/aXSVu1id0YVecysWC0Xp55E5avhhZBaVJtWD7/W5ZE62JacY933o4vLyveYyWRC6iDF639lcKme/3z5f1jR6nXU/OplZI5y3BsUfLEAMOqNRCy8gFPlp//Y1ufKEz668yRHd54scvvYyDgWDVkNwNs933zk8984d5spneZaynciYlk5fh0jVw9EKpNiNBjz7dd28Dusnb6Jw1uPAzCr1wLWRy2hUev6/L7+8CPbZQ+O7b/Esf2XbB5v06Mx6xbu46+9Zu84bMR6fjg8lob/q84f207Z7GcwGEmKT38sNlaqXhapVMqaubu4/9jt/wvfz/hF3ZHJpRj0eefr5TerotcbWDDpZ0uf+RM2seiXL/At58mdmyUvAajCxTpt/Z1fbqH0UuFcJTd1WWZkOrHbb1NtUl1ODf6r0DGjNtzAtZYH/p0DLXUqb3UBPQpGppRRvqfZK0+/koohU5+njVwjh9ysYCQdj8eQqafUaz552v6b6P+7gcpXg0t1NzKupj60nQ/Dc+UJl0QcXTVkpmptCrBPgBeevu6c2HfGUpeZquXikatUfaXy0zLzqeLj74GHlwv/HL5iqctMz+LSqVtUqV2uwL5lypfiuz/HsGLvCEaEdaa0r9tD23Hl3G1MJhP/+6A+UqkEjZOKpq3rcvLQ1XwFGEDhIEefY+DBXAm6LHMeter1Kjy0LU8Lo95I4qFYSr3mY8knadAZiFh0kXLdg1C4ORQyApiMJlJOJaLyUXN51hlOfnqYCxP/Iel4/JM234r4P2Jwqe5WaPgj9XwSiUfiKd8j6ClZZk2xPWGtVsvx48fx8PCgWrVqVseysrJYv3493bt3f2wGlnSadmnEZwv7WcoKpRyTCdoPaWmpG9NyBmcPXszT18XTmW6j2/Hr8r02x/fwcQMg+V+p35NiU3D3dns040so7qXMmW7/7dEmxafjXtp2FtxLp24xZ9R6bl+Pw6O0C90GNmP22gF83PJLtBnmeOzk5b2p8S8xXLJtiEU070YnM+C9LwGIvZ3EmF7LCfm6G4Mnt0Uml3H+RCTj+62wacPJw9foN6olH/R5jZ/XHESldqD3sHcAc6iipJN8PAF9ph7PJt6Wultrr+FUyaXQGPB99Kk5GLMMxPxyC7/2FSjbKYCU04lcm3eeF0Jq4lzFDYDLYWdIv2T9uT4Xcgzu5Q91KKUqUiw3P7KTdKScTiTw44J/1q1Py+HG0ssEDHgBmdo+gYFinfXy5cs0b96cmzdvIpFIaNy4MevWrcPX15x6PCUlhV69ehUqwjqdDp3OeiHAaDI8lbT3j5vDW49z8chVS7nP9K4kRCeyef4OS118VGKefhpnNVO3jOTmhSi+nbzhqdj6X+fB8MaNSzFcOnWT1b+H0OSdWuzacBSAr8dswOGBWPqKPSMY328F8bHmW1CDPjeHonspJwZP/YA9m47zxy+nUDsqCf7sf4z55kNG91yerw03r8YyZ+R6+oW0pNfQFhiNJn5ec5DEuDSrbNEllfg/YnCt6YGDuznbcPKJBNLOJ1NtSr0ij3H/guZW1xOfFmUB0JR3Iv1qKnH77lhEuELvyhhzcu8ozg4/SqVhNVDcO7dEJuFhSTgQi1wjx62eZ4Htbqy4jMerXhab7EGxRHjkyJHUqFGDY8eOkZyczOeff06jRo34/fffKVeu4NvEBwkNDWXSpElWdYGS6lSU1bDRo+SiTc9Cm55lVU5LzCD6Wt5V2vuonVRM2xZCZpqWie3nWH3x/01iTDIAbt6ulv8B3L1duXaqaIuFzxpJ8WmAWQST4tIs9e6lnLh2IbrI42SkZRF1Iw6/8rlfxITYvPG+2OjkfHc7tOzWkMz0LFbM3m6pmz3sR779czRVapXj4qmb+Z73919O8vsvJ3HzdCJLm43JZKJtrybcuZX3YlyS0MVnkXouiYqDc+9wU88no7ubxT8DDlq1vTbvPE4vuFJldK0848idFUhkElRlNFb1aj8NaZdz338Hj7xp5R08VShLF3/3xIOYTCbi98fg0cgbqbzgiGvahWSS/0kgZvute53Nf8d67qdCr8qUer3wePKjUiwRPnToEHv27KFUqVKUKlWKrVu38sknn9CkSRN+++03HB0dizROSEgIQ4YMsapr59GnOKY8s2ic1Uz/NYQcnZ4JbWeTo8spsH3M9bsk3Emizps1iLgnuhpnNVVeCuKXJbufhslPnZhbiSTeTaX2q0FEXLgDgMZRyQu1/Nn2Q+GLQvdRaRzw9fdk790TD2WHUq3A+C/v1WA0e24SaeFeWnKCOZzS/IP65Oj0/HPwSiE97Ev8/hgULg641c69aPm29Kf0G9ZCdG70cfy7VcStjke+40jlUjQBzmTdsd52mRWjRemZV3gfN2kXU9DFZlG6CAJaZVwdeCB+n3QigZhfblF1fG0U7oXHvx8HxRJhrVaLXJ7bRSKRsGjRIgYOHMjrr7/O2rVrizSOUqlEqbSejKcRilA5KvELyp0YnwAvAmuVJy0xnbhbCfSe2hnPMh7M7rXQ0iawVnkA1E5K3Eq7EFirPPpsPTcvmLfrOKgUOLrmXvHDR5vfA3fv3JXltMR09DkGNM5qQrePRqlxYGaPBWhc1GhczCvGKXGpli98+Jk5rBi7joM/m2+hN83bTtfRbYm6GkPMjbv0nNiRhOgkDv5se49zSUelcbDyUL3LehBY1Ze0ZC1xd5LZvPoAnT9uStSNeGJvJxH8eXMS7qZyaPc5S5/Q1f04tPssW78z7xDpO/I9/t53ntjoZDy9XPhw8P8wGo388UvubgonVzUKRe5nrWvDKYDZywYwGkykJGUAcPT3i7Tt2Ziun77F77+cRO2opOeQFsTeTuTaefP8V65ZlmGzOhHSY5nFy2714aucPxFJVmY2dRpVos+Id1kZtp2MtPy3X5UETEYTCX/G4tnY2yoMoHBzyHcxzsFTibJ07m6HsyOPUqZDAO71zXFjn3fLErHgAnEvuOJczY3U04kk/5PACyG5nrM+PQeTPlcAa817BYCc5Hv7qaXWOze0URmY9Cb0GeaYc2ak+SKnKe9kZVv8HzE4VnRGXTavU3h3dxRJxxN4YVRNANT/8tYzrqchkZJv3ydFsUS4SpUqHDt2jKpVrYPd8+fPB+D9999/fJY9ASrXq0jY3vGW8oAwc+x615o/COuzCA9fd7z8rRcfFh+badW/aZfGxNyIo3ulQQC83rEhw8M/LvC8w96azOn95wmqE0DVl83bbFZf+tqqTXDQIGIj4wDwr1IGzQPCvj5sCypHJZ8v6oeTm4azBy8xuuWMQr3okkylGmWZ9d1HlvJHo1sBsHvjMb4c9RM/LfsDldqBwVM+wMlFxbnjNxjXZwU52blbk3z9PXBxz/2ylPJxZeSXXXFx15CSmMG54zf4osMCi6gCjJsfTM2XK9q0K/Z2Ij2bmuf81F/XmDV0He37vk77vq+jy8rhwj+RjO27gmyd2Q6lygH/QC9k8lxhr1zTnw8H/Q+1o5JbEXf5ZvxG9v38zyO+Y0+W1HNJZCfoKPWad+GN8yHrjhaDNndu3OuXonzPStz55SY3v7uGyldNxUHVcH4h1zm5Ou886RdT8hsOAIdSSmp++bKlfGXOWasflZwfZ77Dqb/mNUudPlNP8rF4/LvlP8c5aTno7hb8w6injcT04F6aQggNDeXPP//k119/zff4J598wuLFizEa89++UxDNFZ2L3UfwdJAFlLe3CQIbeHxbsuPMzzvfv7ys0DbFEuEniRDhkosQ4ZKLEOGSTVFEWPxYQyAQCOyIEGGBQCCwI0KEBQKBwI4IERYIBAI7IkRYIBAI7IgQYYFAILAjQoQFAoHAjjxXD3UXPBzawPyfESCwPyeiNYU3EpRohCcsEAgEdkSIsEAgENgRIcICgUBgR4QICwQCgR0RIiwQCAR2RIiwQCAQ2JHnaovai42r0GFoKyrVDcDTz4OJH4RxaIvt7BQePm70nxVM5XqB+AV5s3n+DhYPXfPQ53d2dyR4QgfqNauJV7lSpMSlcmjLUVZNWE9masEPmu4+oQPv9GmKk5sj5w5dYt7AcKKvxjy0LU+Trp1e4bVGlSlX1gNdtp5z56NYsuIPbt3OfQxjy3dq0ezNalSq6I2jo5KWH3xFeoaugFGh54eN6PlhY6u6m7cS6N4v/ySchVG7pj/t2zag6gu+aDQOREUlsW7DEfb8dt6q3etNXqBP9yb4eLtyOyqJJSt+5++jEQWO+9Wsrnnq23WZT+IDD5y3F14qZ4ZWb8ZrPkGoZApupicy+sTPnEu+g1wi5bNqTXnNO4iyju6k5+g4HBfBnHN7iMtKtzmmRu7AZ1XfpJlfFTyUjlxIjmH66R2cTS56jsB/M7pmC+p6+FPJxYtrafG0+21JnjYtylSjf+UmVHDyJCk7g+8jjrLiyqECx63g5MGwGv+jrkc5FFIZl1JjmXf+N47E33hoW4vDcyXCKkcVEacj2bnqdyZsGFpoe4VSQUp8KmtDN9Lus3eLdI5h4R8TeyOOb6fkzaDs6eeOp687y0Z+R+SFKLzLlWLwgr54+nowpfNcm2N2HPY+bQa2YHbvhcTciKPHxI6Ebguhb81hz0R2jdov+rN56wkuXo5BJpXQt9frzJ7WkZ79w8m6Z79KqeDIsQiOHIugf+83ijz29RtxDA350VI2GApOKPD7jpF07rGImHwSflavWoaI63H8sP5vkpIzePWlioQMe4+MDB2Hj1yztBk/6n2WrvyDw39fo9mb1Zg6vh39B67iemR8gef+sM9SMjOzLeWkZPsLsItCxdrXevN3/HX6H/qeRF0m5Z08SM0xp2JSyRRUc/Nh0aX9XEyJxVWhIqRmCxa+0oUOv9t+Vu7UOq2o5OLFyGObuJuVRiv/mqxoHEzLPQu5m5WWb589zT8j5MRmjsbbTmC7MfIkNT3KUNklbwaQJt5BzKrfjmmntnPw7jUqOpdmcp1WZBlyWBtx1OaYi17tSmR6Aj0PrEZn0NM96BUWvdqFt3fNI1735OfouRLhoztPcnTnySK3j42MY9GQ1QC83fPNRz7/jXO3mdIpV2zvRMSycvw6Rq4eiFQmxWhDQNoOfoe10zdxeOtxAGb1WsD6qCU0al2f39cffmS7njQjxv5kVZ4xZxs//ziYypW8OX32NgAbNpvvSGrX9C/W2AaD8bF5k9//aJ1E9P9+Pk79egE0aVTZIsIftKnHkWMR/LjhCAAr1vxJ/ToVaPt+Xb78ZleB4ycnZxbq3T9t+lZuxB1tCmNObLHURWUmW/5P1+voc/A7qz5TT23npzf74at24Y4278VMKZXzP79qDPxrHccSzBmpF1z8gzd9KtMloD5fX/jtoWydfnoHAO5KTb4i/L5/TfbeuciPN8zfk9uZySy9fIC+lRrZFGE3BzUVnDwZe2ILl1PvAjDn3B66BjagkosX8XHXH8rW4vBciXBJxNFVQ2aq1qYA+wR44enrzol9Zyx1malaLh65StVXKj8TIvxvnDTmJK9pjyHxZZky7mz4/hOysw2cuxDFspV/cDcuf0/rYXByVHLzZoKlXL1qGX7aaP2FPnL8Oo0bVip0rOULe6FQyLh+I55V3x3g7L1kofbkTZ8XOHj3GnNfak+DUhWI1aay7voxfrphO0O1s0KJ0WSyeMv/RiaVIpdK0Rn1VvVZRj11Pcs9VvsfxEEqI8tgfWeoM+Tgq3HFT+NKdGbefHbJ2Voi0uJpXa4W55PvkG3U06lCPeKz0jmXfOeJ2fogj12Eb926xYQJE1ixYoXNNjqdDp3O2iMwmgxPJePy46Zpl0Z8trCfpaxQyjGZoP2Qlpa6MS1ncPbgxTx9XTyd6Ta6Hb8u32tzfA8fNwCSY60/QEmxKbh7uz2a8XZAIoGBA97izLnbhd6+F8b5i3eYMedXbt1OxNPDiR7dGjEvrBu9BqxAqzXf9s+c0oGaNcpa9Vu5pA/cS+oVczeVXh+F5zv+G02q8EIlH+bM22mp83B3JPFfYYSk5Aw83G1n501IzGDOvB1cuhyDQiHnvRY1+WpWFz7+/FuuXI19mJf+2PB3dKdzQH1WXT3M0ksHqOHux+iaLcg2Gvj55qk87R2kMoZWb8a222fI0GfnMyJk6rP5J+EWH7/wGtfS4kjIyuA9/xrU9ijLzfTcdYAJtd+jlX9NS1ktU7C0YTcMD2Rcq781tMiv5cDda4x68W1euXmKv+OuU97Jg56VXgXMce/8RBig94E1zH+lM8dahWA0mUjUZdD/0Pc2LzKPm8cuwomJiaxevbpAEQ4NDWXSpElWdYGS6lSU1Xjc5jxxDm89zsUjVy3lPtO7khCdyOb5Oyx18VF584BpnNVM3TKSmxei+HZy3vjxf5XPP21OQIXSDBr6/SOPdeRY7mJYxPU4LlyMZt2aj3nztSr8uvM0ALO/2o7SIfdj/v3Kjxg1bgPx8WZvWW/jDqR2zXKMHPoOYV/v4MYjXixu3U60WoQ8dyEKP183OrStz/TZ2x5p7EdFIpFwLimar87vA+BCSgyVXLzoHFAvjwjLJVLmvtQBiUTCpJMF2z3y+Cam1X2f/e8MRW80cj75DttunaW6u6+lzTfnf2PlA4tmq5v0ZM7ZPZxOuv1Qr+WnGyco5+jBole7IJfISNfr+Pba3wyq+gbGAlJpjqv9Lom6DD7cvxKdIYf2Feqy8NUudPxtGXE624uPj4tii/CWLVsKPB4RYXuV+D4hISEMGTLEqq6dR5/imlIi0KZnoU3PsiqnJWYQfc22h6N2UjFtWwiZaVomtp+DQW+w2TYxJhkAN29Xy/8A7t6uXDtlewGjJPLZJ8149eWKDB62lrj4xxcyuE96ho7bUYmU8XOz1MUn5P0Sxd5NyXdh7j61XvQndNIHLFiyj117z1kdS0zKwMPN2ut1d3Msdlz64uU7vFi9bOENnzDxWWlcS4uzqotIi6e5X1WrOrMAt8dP40qvA2tsesH3uZWRRPc/V6OWKXCSK4nTpfNlgw+4nZFkaZP4/+3deVhU1f/A8fewDcyw74iIICAqq+ASYqa45oYbqLnlVplmLvkNv5mlGFZq6dfUMg3bzMp91yxNccFwQ9w3XBBENkGQ/ffH6ODIjsi1n+f1PD1P995z73xmzvXDueecmZOXTWpetnq7sKiIpAf3uP5YmeqaH/cHX8TtwVLfkLTc+7S2dlbHU5bWVk68YutGqy2fqt/TrJPbCLB2prejN99eiKpxLFVV7SQcHByMTCajokWaZTJZhdeQy+XI5XKNff/GroiaUBgZ8Mm2MPJzC5jZ5/NKZzckXr1Dyu00fNt7cOVh0lUYGeDe0oUtX++ui5BrxcRxHQkMcOPdaatJTCr7sfBpGejrUs/OlF17aj5Q5+PlQMTH/fl65V62bC/9OB539hbNfRzVA4kA/s0bcuZs9fp3XZxtSEl99q2syhxLuUFDQwuNfQ0NLTQe3R8lYEdDC4bvX0V6XsXTKR+XU5hPTmE+xrr6tLF2YV7cs79niyhWz8DoXt+D4yk3SHss2T9OX1sXoFQ+KyouRouK81htqfaXNezs7Fi3bh1FRUVl/nfsWPkd+lLTV8px9nbE2Vu1hLutkzXO3o5YOahuwpHhA3nvu3Ea5zwqb2Aox9TKGGdvRxo0sVcf19PXxczGRP3fiuk/s+Wb3Rr7dHRVf2AURgZEbJ+OvlLOgrFfozA2UJfR0iqp8BWx82nTu4V6e/2i7Qye3ofWPfxo6OHAtO/GkZKQRtTG8uc4P0/efbsTnTo0I/zTzeTk5GFupsTcTIneY90E5mZKXJytsa9nBoBTQytcnK0xMtRXl5kfEUqfns3V22+Nbo+3pwO2NsY0a2LP7A/7UlRYzJ69JfN6jQz11a9nbqak76DF5OUVqrdNTAzUZX28GhAxqz9rN8bw94EL6jKPx7B2Qwwt/Z0I6duCBvXNGTGkDY1dbVm/qeS+H/P6y4RN7a7e7h/sT5vWLtjbmeLkaMn4N4Lw9W7Ahs3Ha+kTrrlVlw7jbV6fsW6BNFCa0b2+BwMaNlfPJtCRafFlqwE0M63He0fXoS2TYSlXYilXoisrSR8r2wxlsHPJPdvGuhGB1o2wV5gSYOVMZOBwrmbdZX38CXUZQx25+lqWciUD933L1cy7Gvse10BphruJDZb6huhr6+BuYoO7iY06DlM9A0Ib+uFkaIG7iQ1hnl3pYt+UiNiSrkFPs3ps7fg21vpGAJxIvcG9vAdE+AXT2NhGPWfYXmnGvsSLtf55l6XaLWE/Pz9iYmLo3bt3mccrayVLyc2vEfP2fKjefnPeMAB2fb+PeaOWYm5nhrWDpcY5y/75VOP8DoMCSbyWzDDXCQC0CwngvRVvVfi6U4NmcervM7j4OtGklWoUfdX5hRplhrpMICle9Vjo4G6PwqTkd2J/nbcJfaWcd5eOwdBUwemo80zvMfdfMUcYIPhh4lz4ueYXFubO38qO3acB6NXdR+OLF/+b/1qpMvb1zDSSppWlETPe74mxkQEZGTnExt1k3KQfyMgoaanN/rAPPl7lj8gnJmUwcPgyALp28sBAX48hA19iyMCX1GVOnLrOu9NWA6qW8OxPNzNqeFtGj3iZWwlpfDBrncYgo4W5ITbWxuptHR0txo3tgKWFIQ9yC7hy9Q5TwtZw4tT1qnx8z9Tp9ATeObKGSU2DGOfejpvZacyN3cmWm6rZONYGRgTZuQOwIehNjXOH7Y9Uz+ltoDTHTK/knjXSlTOpaRC2BsZk5Oew69ZZvjzzJwXFJX3w07260sfRp8L4mqwvGTua7duLllYN1dvrO6jiCdr5pbrlHuzozXuenZEBJ1NvMnz/KmLTSr4goq+ti7ORJTpaqsSdnpfDmIM/8m7TDkS2HYaOTJtLmXcYf/gXzt+rm0FTWXE1M+b+/fu5f/8+Xbt2LfP4/fv3+eeff2jXrl21AumsO7Ba5YW6kxfkK3UIQjmS3qibEXyhZs72mVlpmWq3hNu2bVvhcaVSWe0ELAiC8KISP+AjCIIgIZGEBUEQJCSSsCAIgoREEhYEQZCQSMKCIAgSEklYEARBQuKnLIVKJbysJ3UIQjkUe0TdPNf6VF5EtIQFQRAkJJKwIAiChEQSFgRBkJBIwoIgCBISSVgQBEFCIgkLgiBI6IWaouYZ6M6AKT1xbe6ERT1zPuo3j4Obyv9hdHNbU8Z+NhQ3P2fqudiwYfEOlk35vsavb2SmZOjMAfh19MK6gSUZyfc4uOkokTN/JftexasVDJs5gG6jOmBoqiTu4HkWjV9BwqXEGscipbEvtWBq+0Aio4/xyR/7APjhtf60ctRc7n71sVPM3FH+IqiP+7hrEIOaezFn915WHa3Zj6V3buzCoOZeNLGxQk9bm4vJKfxv/2EOXNVcRsrGUMnUDm152bkhBrq6xKelE7ZlF6cTK//92eb16/HjkAFcTL5L7xVPv85ebRgQ6MWAtl7UM1f9BvLlxBS+2X6EqDPXSpVd/FYwgc2cmPTNJv46dbnC6zrZmDMxOBA/l/roaGlxJTGFKd9uITGt+ktbOVqb8cHAIJxtzTE0kJOccZ/t/5zj622HKShS/UZx3wAPerRsiks91SINZ67fYfHmA5yOL79eZg3pTK/WzUrtv3w7hX5zav5vvTpeqCSsr9Tnyql4dkbuZebvUyotryvXJePuPX6OWEffia9W6TWmrniLpGvJ/DC79OKdFvXMsLAzY/l/fiT+7C1sGljyzlejsbAzZ/bAL8q9ZsjUXgSP78rnI5eQeC2Z4R+FELE1jNFeU/81P+z+iKedDaG+npxLSi51bM3xWBb+XbLwY05+QakyZenk1ggfe1uSMiteLsjexJi/3h6F2ydlf9b+DvZEXb3Ogr1R3HuQSz+vZiwL6c2AyNWcfRivsb6c1cNCORJ/kzFr1pOanYOjuSkZDyr/XV8juZzPenbh0LXrWCoVlZavK0npWSzaeIDryekgg16tmvLl2F4MnPsTlxNT1OWGtK/670rXtzThu8khbDgYx9Kth7j/II9GdhbkVlCnJxZP4tUPV5CQWnr9v4LCIrZEn+HsjTtkZufiVt+KDwd1REsm43+bVevA+bvWZ0fMOU7+dpvcggJe79SCpW/3pd+c77mTUfaSV5/9vpeFGw+ot7W1tfg1bAi7j1+o8nt9Wi9UEj668wRHd56ocvmk+GSWTl4FQJcR7Z/69a/F3WR2aEkCuH0lie8+/IX/rBqPlrYWReWs/NvnnW78/Ml6Dm2OAeCz17/i11tf06a3P3t/PfTUcdUVha4u83p1Y8a2P3irTctSx3Py87l7v+y1wMpjY6hkRuf2jPxlPd+ElL3aS1U9apU/smBfFEFujejg6qxOwmNbtyAxM4uwrbvU5W5mlL9o6ONmdQtic9w5ioqL6ejW6KlirU1/n9ZcnHfx5oMMCPTG08lWnYQb21sxtIMfgz/7mT0Rb1R6zfE923Ag7hpfbtyv3nfzbs3XFryVksGtlJLzb6dlss31HL6NSpYam75qh8Y5H/+0myBvF1o2bsCW6LNlXjfrQR5ZD0oWLW3v1QhjA302Hoors/yzIPqEJaY0UZB9L6fcBGzrZI2FnRnH/oxV78u+l8O56Es0ae1WV2HWipldOrD38lUOXit7WZ9eHu4cefdNtowZypRX2qCvU3EbQQZ81qsr3x6J4dLdlArL1oQMUOrpkpFT0srt4OZM7O0kFvbpzqGJb7Bh5GuE+HhUeq2+Xk1xMDVh8f7DtR5nbdKSyeji54aBng6nrt4GQF9Xh09GdCPi1z9Jyaz8j6RMBm2bORF/J40lb/fhz4g3+GHqQNp71d4fHgdLEwKaNCTm0s1yy+jr6aCjrU1GdtVXHwl+yYMj569zuwZdJjVV7Zbw3bt3WblyJYcOHSIxUdUnaWtrS0BAACNGjMDKyqrWg3yedRjUholLxqi3deU6FBdD/8k91Pv+22Mup6POlTrX2MKI16b3Zdu35fd7mtuaApD+xArFaUkZmNmYPl3wdah7Uzea2lrT77ufyzy+Je48tzLucScri8bWVrzXPhAnCzPGr91S7jXHvtSCwqJivq+gD3jrmGHUM1Et6ih7uHru8alvq4/H3LjF6DUbyjx3VGt/FHp6bDtb8mjqYGrC4OZefHfkGMsORuNlZ8sHndqTX1jE+tgzZV7H0cyUqe0DGfzDrxQ+p+svutSz4PspA9HT0SEnN4/JyzdzJTEVgKn92nHyagJ7Y69UchUVc0MFSn09RnZqwVdboli44QABTRsyf3RPxiz6jZhLqpWpF78VTHMXe41z1/53GMWoPqPbqZml+mVXTQ7F3cEaua4Ovx84xZKtBynPu73bkpyRxZFzVVvLz8pESZumDZkeub1K5WtLtZLw0aNH6dKlCwqFgo4dO+LmpmqJJSUlsWjRIubOncvOnTvx9/ev8Dq5ubnk5uZq7CsqLvxXLnt/aHMM56IvqbdHfTKYlIRUNiwueTS6eyu11HkKIwPCN/2H62dv8cOs0v3H/5/YGhny306v8PrP68grLCyzzJoTJS39C8kpJGfd5/vX+uNgasKN9NKPsc1srRnWwpc+Kyse3BqzZj062qoHPhsjQ34aEkLvFT+qj5fXR9mjaWPGB7Zm3O+bSM0uGTSVyWScvp3Egn2qfsizScm4Wlkw0NezzCSsJZOxoHc3Fv19mGup6RXGKqVrSWmERvyIoYGcjr6uzBrahdELf8PBypSWbg6Ezq36IOKjlcP3xl7mx79UfyDP30rG29mO/oFe6iQ86+fdyHVLUtDmj0Yyful67qSr+vYLyng6nLZyK0p9PdzsrZgU3JbhQf5E/lF6cP31Ti3o4teY0Qt/I6+g7HvuST1bNSUzJ5c/T12qvHAtqlYSnjBhAgMGDGDZsmXIZDKNY8XFxbz55ptMmDCBQ4cq7qeMiIjg448/1tjnLGtGI+3KH+ueNzlZD8jJeqCxnZl6n4TL5Y/IGhjqM2drGNmZOXzUfz6FFdwkqYnpAJjamKj/H8DMxoTLJ+PLPuk542Fng6VSyfpRr6n36Whp0aJBfYb4++Dx6SKKnmghnkxQPQo7mpmWmYT9HeyxUCrYO360xjXfD3qZ4S186bBkJQAJ90oeKwuLVK9xPa3ivsnuTd2Y070TE9dtLdV1kpx1n8tPdH1cTkmli7trmddS6unhWc+WJrbWfNhFNa6gJZOhJZNx5v2JjFy9jsPxNyqMpy4UFBZx42Gf7dkbd2jWwJbBr/iSm19AfUtT9n8+TqP8vNE9OH75FqMXlm5ApGXlkF9YyOXbmp/T1cRUfJ1LWr5lDZbdTs0sc2DukaSHCfpKYipaWjJmDOrI93tiNO6fYUF+jOzkzxuL13Ex4W55lyoluHUztkafLTP5P0vVSsInT54kMjKyVAIGVQth0qRJ+PpWPoIaFhbG5MmTNfb1NR9VnVD+tRRGBnyyLYz83AJm9vm80tkNiVfvkHI7Dd/2Hlx5mHQVRga4t3Rhy9e76yLkp3bo2nW6L9d8rJzbvTNXUtL45vDRUgkYoImNNaBKemXZePpsqQS5cmBfNsaeZe2pmg+qdG/amIjunZm0YSt7L18tdfzYzQScLMw19jU0N+NWOYNzWbm5pd774ObevNTQgQnrtnCzjD8wzwMtGejpaLN06yHWHTytcWztf4cxb+0+9p0uu3uioLCIM/FJNLTR/Jwcrc24nVa1QcyqxShDR1sLLZlMfQ+N6OjPqC4tGffVOs5cr/qS9f6u9Wlgbcb6Q5trLb6qqlYStrW1JTo6Gnd39zKPR0dHY2NjU+l15HI5crlcY19ddEXoK+XUc7FVb9s6WePs7UhmahbJN1IYGT4QC3tzPn99ibqMs7cjAAaGckytjHH2dqQgr4DrZ1WPVHr6uihNSqYbrZiu6vM0szFR78tMzaIgvxCFkQER26cjV+jx6fCvUBgboDA2ACAj+R5FD1tqK2Lns/KDX4jaeBSA9Yu2M3h6H25dSiTx2h1GfBRCSkIaURvLn+P8PLmfl8/FZM1WUXZ+Pmk5OVxMTsHB1ISezdzZd/kq6TkPaGxtyfSO7Yi+fpPzySUtmR1vDGf+XwfYfeEy6TkPSM/RHHDJLywk+f59rqamqfeZKQzQfthoyC0oIGDh1xrTw/ILC8l4oOoa69G0MZ/27MKc3Xs5mZCoLvegoICsXNUIemT0MX4ZFsqbAS3YdvYCXna2hPp4MmP7H+prTnmlDTZGhkzbvJNiKPXeU7OzyS0oKLVfKhN6tSEq7hqJaZko9HXp5u+Ov6sD45asIyUzu8zBuMS0TBJSShLq+g+Gs2jTAfXc4cg//uGzkd05dukmRy/cIKBpQ172cGb0wt/U5xgr5Ohql/y7Dwr7GgALI9XnXlRcTFqWqivoVX93CgqLuJhwl7yCQpo1sOGdXoHsirmgnic8oqM/47q/RNiq7SSk3FNfJzs3n5y8fPV7tTYxZMYPOzXeT/BLHpy6ertU670uVCsJT506lbFjxxITE0NQUJA64SYlJbFnzx6WL1/OvHnznkmgtcHNrxHz9nyo3n5z3jAAdn2/j3mjlmJuZ4a1g6XGOcv++VTj/A6DAkm8lsww1wkAtAsJ4L0Vb1X4ulODZnHq7zO4+DrRpJXqsXXV+YUaZYa6TCApXjUNysHdHsVjif3XeZvQV8p5d+kYDE0VnI46z/Qec/91c4TLk19YSIBTA4a38EWhp8vte5nsPHeJJVFHNMo5W5hj+MQf78qsHTGI+qYm5R4/En+DoT+pHqlDfT3R1dbmo65BfNQ1SF1m3ak43t+impIWezuJt9duZsorgbwd2Jqb6Rl88sdeNseVDLxaGSqxMzaqVpxSMjdUED6sC5bGSrIe5HHh1l3GLVnH4SoOaAE42ZpjZFBSN3+dukz4L3sY1bkF0/q3J/5OKlO/3cyJKwnqMgvG9MTf1aGsywGQkJLBqzNV3UoFRUWM6OSPo7UZMpmq2+KXv0/w45/H1OVD2nqhp6vD/NE9Na6zbNshlm1TzUqxMlZiZ65ZN4b6egT5uPD573ur/H5rk6y4uHrDtWvWrOGLL74gJiaGwoeDLNra2vj5+TF58mRCQkJqFEhn3YE1Ok949q593ErqEIRyKBIqLyNI58TiSZWWqfYUtdDQUEJDQ8nPz+fuXdWjoqWlJbq6utWPUBAE4QVX42/M6erqYmdnV5uxCIIgvHDEN+YEQRAkJJKwIAiChEQSFgRBkJBIwoIgCBISSVgQBEFCIgkLgiBIqNpf1hAqlpubS0REBGFhYaW+mi1IT9TP8+tFrRuRhGvZvXv3MDExISMjA2NjY6nDEZ4g6uf59aLWjeiOEARBkJBIwoIgCBISSVgQBEFCIgnXMrlczsyZM1+ogYV/E1E/z68XtW7EwJwgCIKEREtYEARBQiIJC4IgSEgkYUEQBAmJJCwIgiAhkYRryaRJk+jbt6/UYQjlEPXzfBL1IpJwrYmOjsbf31/qMIRyiPp5Pol6EVPUnlpeXh5KpZKCggL1vlatWnH48GEJoxIeEfXzfBL1UqLGC30KKjo6OkRFRdGqVStOnDiBjY0N+vr6UoclPCTq5/kk6qWESMJPSUtLi4SEBCwsLPD29i63XMeOHVm6dCmurq51GJ1Q1foR6paolxKiT7gWHD9+vNIb6eLFizRq1KiOIhIeV5X6EeqeqBcVkYRrwYkTJ0rdTHFxcbRq1QovLy/mzJmDra0tWlri45ZCWfWzfPlymjdvjoeHB6GhoRJF9mJ7sl6OHDlCz5491ds7duxg6NChUoRWp0R3RC2IjY2lX79+6u3c3FxCQ0NZvXo1np6eBAcH4+XlJWGEL7Yn6yctLY2vvvqKmJgYtLW1SU9Ply64F9iT9dKkSRPOnz+v3p49ezaRkZESRFa3RNOsFhQVFXH+/HkSEhLIyMhgw4YNtGvXDk9PT0B1c4kkLJ0n60dHR4e0tDSmTZtGXFwcpqamUof4QnqyXoyNjcnLyyM/P59t27bh6ur6QoyhiCRcC8LDw4mMjMTe3p7w8HBiY2Px8fFRH4+JiRFJWEJP1o+RkRGnT5/Gx8eHkJAQNmzYIHWIL6Qn6wXA1dWVS5cuER4ezowZMySOsG6IecLPwIIFC4iPj2fhwoXs2rWLrl27kpKSgpmZmdShCagGSR+1sMaNG0e7du1Ev/Bz4p133iE7O5vi4mJWrFghdTh1QvQJPwNDhgyhW7du+Pr64uHhgaOjo0jAz5Hw8HAOHz6MQqEgICCAAQMGSB2S8FCTJk2YOHGiRt/w/3eiJSwIgiAh0ScsCIIgIZGEBUEQJCSSsCAIgoREEhYEQZCQSMKCIAgSEklYEARBQiIJC4IgSEgkYUEQBAmJJCwIgiAhkYQFQRAkJJKwIAiChEQSFgRBkND/AdadMB3vLS5LAAAAAElFTkSuQmCC",
+      "text/plain": [
+       "<Figure size 400x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checkpoint: 1200\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAApcAAAFiCAYAAACj2/OmAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAADblElEQVR4nOzdd3RURfvA8e9mk2x6J5UECISE3psoTamKIgoCKkhEkSZFQCIqoiKIoHQFQUB5VdAXbAhSFASkSe8lIb33Xrb8/sjLhjWbQJZdKb/nc86ek507M3f2yd3Ns3Pn3ih0Op0OIYQQQgghzMDqTg9ACCGEEELcPyS5FEIIIYQQZiPJpRBCCCGEMBtJLoUQQgghhNlIcimEEEIIIcxGkkshhBBCCGE2klwKIYQQQgizkeRSCCGEEEKYjSSXQgghhBDCbKzv9ACu66l85k4PQQghhBD3sBbH7/QI7n8LWmy8aR2ZuRRCCCGEEGYjyaUQQgghhDAbSS6FEEIIIYTZSHIphBBCCCHMRpJLIYQQQghhNpJcCiGEEEIIs7lrbkVkTo+P6cWgqf3x8HUj8lQMyyeu5dLRyCrrO7o6EP7+EDo/2R5nDydSY9L4dMp6jmw7adL+nd0dGbcknI6PtUan1bFv82FWTFpHcUFJte0adQxh5HtDCOvQAK1GS+TJGCL6zqG0uMykcViSxNiyJL6WJzG2LImv5UmMTbNr3DmK0irvq24vL5qNqk1xdhnnv0ok/XQe6mItjv4qQp70wb+jW5V96rQ6Lm1KJn5fFiXZZdh52BDY1YOQp3xQKBQmjfPsF/FkXiogL64YpwAVXT8KM9iuKdVy+vM4cqKKyE8oxru1C+2nB9+038ubk0k9nktOdBFW1gr6rmtu0viqY/bkMiUlhZUrV/L222+bu+tb0nVwJ0YvHM6Ssau5cPgKAyf2Y+62NwhvNJnstNxK9a1tlHz425tkp+Xw3uBPSE/IxKeOF/nZhVXuY8Hut9nx5V52rN9rdPuMDRPw9HVnRu85KG2UTFszhskrX2buc0ur7LNRxxDm/voG38z7geUT16JRawhuUQedVlfzIFiYxNiyJL6WJzG2LImv5UmMTffQ3FCD/eXFFnPo/Uj8OrkCcGJZLOoCDe1er4etszUJ+7M49kk0jvMa4lrPwWifV39IJXpnOq3GBeFc247sqCJOrojF2kFJcL9aRtvsGneOlmOD8GriXOVYA7t7kH21kNyYokrbdFodSlsr6vWtRdLh7Ft+/Tq1Dr+Obrg3dCT294xbblcTZk8uk5OTmT179h1LLp+a9CjbVu/mt3V7AFg8ZjUd+rWm98jubJz/Y6X6fcK74+zhyMQH30Kj1gCQEpNm8v6DwgJo36cV49pHcPlYFADLJq5lzi8zWDVtAxlJWUbbjVk4gi1LtxmMMf5yksnjsCSJsWVJfC1PYmxZEl/LkxibTuVimPpc/SEFBx9bPBs7AZB1qYBmL9XGvYEjAA2f8iVqaxrZUUVVJpeZlwvwbeuKT+vyBNXBW0XC/iyyr1advN9M0/DaAFzalGQ0ubS2U9L8pcDy/V/Kp6xAc0v9hg72AyBuj2USSzBhzeXp06erfVy6dMkS47wl1jZKGrYJ5vjuM/oynU7H8d1naNwpxGibTv3bcv7QFSYsC2dT4kpWnVrA0BkDsLIybRq7UacQ8rLy9W82gOO7zqDT6gjr0MBoG7daLjTqGEJ2ai6L9r3LpsSVLPx9Fk06h5o0BkuSGFuWxNfyJMaWJfG1PImx+WjVWuL3ZRHU3VN/+to91JHEv7IpzVej0+pIOJCFtkyHVxOnKvvxaOhI+tk88hOLAciJLiLzUgHeraqelbyf1XjmsmXLligUCnS6ylPY18tNXV9wu1y9XFBaK8lKyTEoz0rJITDU32gb33retOzehN1f72fmY/Pwb+DLq8teRGljzYb3vgdg6IwBDI14Ut/G1t6WRh1DGL8kXF/2YtMppMVl4OHjRnaq4SkJrUZLbmY+7r5uRsfgF+wDwPBZT7Nq+gaunoym5/NdmL/zLV5uPpWEq8k1joWlSIwtS+JreRJjy5L4Wp7E2HySj+SgLtAQ2M1DX9Z2ch2OLYrht/CzKJSgtLWi7dS6OPqqquynwQBv1EUa/ph8EYUV6LQQNsSP2g9V9Ht6VRzx+ypmdDWlWg5/EIXihgS/31fmX/94J9Q4ufTw8GD+/Pk8/PDDRrefO3eO/v37V9tHSUkJJSWGC361Og1WCmVNh3PbrKwU5d+iRq9Cq9Vx5fg1vPw9GDS1v/4N98vKnez97qC+zYyvJrB/8xH2bzmsL8tINH4K4FZcP7C2rtqlP8UReTKaVj2a0ntkd76Y+Y3Jfd8NJMaWJfG1PImxZUl8LU9ibFzsH5l4t3TBzsNGX3ZxYzJlBRo6vlUfW2drko/mcOyTaDq/G4JLkL3RfhIPZhO/P4vWr9bBOdCOnOgizq1LwM7dRp+4hj7jS/3HvfVt/nrnKo2e9cM9xNGyL/IOqHFy2aZNGxITE6lTp47R7dnZ2UZnNW80d+5cZs+ebVBWj8bUVzSt6XAM5KTnolFrcPdxNSh393ElKyXbaJvMpGzUZRq0Nyzujb2YgKefO9Y2StRlGvKyCsjLKtBvLy0qJTsth8TIlMr9pWTj5u1iUGaltMLFw4ms5KrGUP5mjbkQb1AeezEB7yCvKl/vnSAxtiyJr+VJjC1L4mt5EmPzKEwrJe10Hu2m1tOXFSSXEL09nW4LQ3EOLE8kXevak3kxn+jt6TR/OdBoX+c3JNLgCW8COrsD4BJkT1FaKVd+SNEnlypXG1Q3/MoUSrDzsKl2RvReVeM1l6+88gp169atcntQUBBr166tto+IiAhycnIMHvUUjWo6lErUZRouH4uiVY9m+jKFQkGrHk05f/CK0Tbn/rqEfwPDWwXUDvEjIzETddmtLY690YWDV3B2dyKkdcXB2qpHUxRWCi4evmq0TXJ0GukJmdRuaHg6o3aIH6m3seDaEiTGliXxtTyJsWVJfC1PYmwecX9koHK1xrt1RZKsKdWW//CP5X0KKwXVzZtpSrQGp7evt+Huu9HAv6LGyeWTTz7Jc889V+V2d3d3RowYUW0fKpUKFxcXg4e5Ton/d9FW+o3qQc/hXQgKC+DVFaOwc1Tpp+CnrxtH+Jyh+vo/f7YTZw8nxi56gYAQP9r3a8XQiAH8tGKHvo6dowp3H1f9Y86wxRzdftKg7Pqi6NiLCRzZfoLJK0cT2q4+TR4IZfySkezZ+Jf+6jlPf3fWnPuY0Hb19fvYtOBnnpzQl4ee6oB/fR9GzB5MYFgA2774wyxxMSeJsWVJfC1PYmxZEl/LkxjfHp1WR9yeTAK7emClrEgKnfztcPS15fTncWRdLaAguYTIn1NJO52Hb7uKaceD717l2vaKhNinjQtXNqeQcjyHwtQSko5kE/VLqkGbskINxdll+seDcxriFGBnUHajguQScqILKclWoynVkRNdSE50IVq1Vl8nL76YnOhCyvI1qIu0+jrXZV0t4PdJFyjKLNWXFaaXkhNdSFF6GTot+jbq4pp/yaiK2W9FFBcXx6xZs/jiiy/M3fUt2bvpIG5eLox4ZzDuvm5EnozmjX5zyU4tX/jsHeiJTlvxi0mLzyCi7weMWTiCVSfnk56QyZYlhrdJGPRaf4bPGlTtfp8LHq+/rcO855Yyfmk483e+pb+x7PKJFbO51jbWBIUFoHKomArfsuRXbO1seGXhcJw9nIg6FcPrvd8nKary6Yg7TWJsWRJfy5MYW5bE1/Ikxrcn7UweRellBHb3MCi3slbQPqI+F/6TyJEPr6Ep1uLoa0vLcUH43DDDWZBSQmmuWv+8WXhtLm5M4szqeEpy1Nh52FCnpxcNn/bR1zm7Np74vdWvWe2/qaX+51OfxZJxvmKZwp/TLwPw8LJGOHiXx/Tw3EiDG8Jfr3O9H02JloLEEnTqiinUSxuTDMZxvU2nWfWrvedmTSh0N1sgWUOnTp2idevWaDQ1y4B7Kp8x5zCEEEII8f9Mi+N3egT3vwUtNt60To1nLn/66adqt0dFRVW7XQghhBBC3L9qnFwOGDCgyvtcXnen7nMphBBCCCHurBpf0OPn58fmzZvRarVGH8ePy5y0EEIIIcT/VzVOLtu0acOxY8eq3H6zWU0hhBBCCHH/qvFp8WnTplFQUFDl9gYNGvDHH3ffbRuEEEIIIYTl1Ti5fOihh6rd7ujoSNeuXU0ekBBCCCGEuHfV+LS4EEIIIYQQVZHkUgghhBBCmI0kl0IIIYQQwmwkuRRCCCGEEGYjyaUQQgghhDAbSS6FEEIIIYTZSHIphBBCCCHMRpJLIYQQQghhNjW+ifq94PExvRg0tT8evm5Enoph+cS1XDoaWWV9R1cHwt8fQucn2+Ps4URqTBqfTlnPkW0nTdq/s7sj45aE0/Gx1ui0OvZtPsyKSesoLiiptl2jjiGMfG8IYR0aoNVoiTwZQ0TfOZQWl5k0DkuSGFuWxNfyJMaWdS/Gd8Hut2nRrYlB2S8rd7J47GqTxmBpEmPT7Bp3jqK0yu+Xur28aDaqNsXZZZz/KpH003moi7U4+qsIedIH/45uVfapLtJwcWMSyUdyKMlR41rPnqYv1MatgYPJ4zz7RTyZlwrIiyvGKUBF14/CDLZrSrWc/jyOnKgi8hOK8W7tQvvpwdX2mX4uj4OzjR8jD33Q8LbGeyOTksvS0lJ++OEHDh48SHJyMgC+vr488MADPPHEE9ja2pplcKboOrgToxcOZ8nY1Vw4fIWBE/sxd9sbhDeaTHZabqX61jZKPvztTbLTcnhv8CekJ2TiU8eL/OzCKvexYPfb7PhyLzvW7zW6fcaGCXj6ujOj9xyUNkqmrRnD5JUvM/e5pVX22ahjCHN/fYNv5v3A8olr0ag1BLeog0579/2fdomxZUl8LU9ibFn3anwBtn6+i/WzNumflxSW3uKr/ndJjE330NxQg/dMXmwxh96PxK+TKwAnlsWiLtDQ7vV62Dpbk7A/i2OfROM4ryGu9YwnX6c+iyM3rphW4+tg52FD/J+ZHHzvKt0+CcPew3hOtGvcOVqODcKriXOVYw3s7kH21UJyY4oqbdNpdShtrajXtxZJh7Nv6bV7hDrSc5Vhcn/p2yTSz+bjWt/+lvq4FTVOLq9evUrv3r1JTEykQ4cO+Pj4AHDixAk+++wzateuzbZt22jQoIHZBlkTT016lG2rd/Pbuj0ALB6zmg79WtN7ZHc2zv+xUv0+4d1x9nBk4oNvoVFrAEiJSTN5/0FhAbTv04px7SO4fCwKgGUT1zLnlxmsmraBjKQso+3GLBzBlqXbDMYYfznJ5HFYksTYsiS+licxtqx7Nb5QnuhkpeSYvO9/i8TYdCoXw9Tn6g8pOPjY4tnYCYCsSwU0e6k27g0cAWj4lC9RW9PIjioymlxqSrUkHc6m3fR6+j5CB/uRciyXmB0ZhA3xM2mcTcNrA3BpU5LR5NLaTknzlwIByLyUT1mB5qZ9WllbYedWsSJSq9aR/Hcu9fp4oVAoTBqn0f3UtMGYMWNo1qwZKSkp7Nmzh40bN7Jx40b27NlDSkoKTZo0Ydy4cWYbYE1Y2yhp2CaY47vP6Mt0Oh3Hd5+hcacQo2069W/L+UNXmLAsnE2JK1l1agFDZwzAysq0IDfqFEJeVr7+zQZwfNcZdFodYR2MJ9xutVxo1DGE7NRcFu17l02JK1n4+yyadA41aQyWJDG2LImv5UmMLeteje91PYY9yPcpn7Pq1ALC5wxFZX/nzsRVRWJsPlq1lvh9WQR199QnV+6hjiT+lU1pvhqdVkfCgSy0ZTq8mjgZ7UOn0aHTgpWNYUplZWtF5sV8i7+G25H8dw6leWoCu3uYtd8az1weOHCAI0eO4OLiUmmbi4sL7733Hh06dDDL4GrK1csFpbWy0jeirJQcAkP9jbbxredNy+5N2P31fmY+Ng//Br68uuxFlDbWbHjvewCGzhjA0Ign9W1s7W1p1DGE8UvC9WUvNp1CWlwGHj5uZKcanpLQarTkZubj7utmdAx+weWzv8NnPc2q6Ru4ejKans93Yf7Ot3i5+VQSribXOBaWIjG2LImv5UmMLetejS/A798eIDUmnfTETIKb12HU3GEEhvoz++mFNQ2DRUmMzSf5SA7qAg2B3SqSq7aT63BsUQy/hZ9FoQSlrRVtp9bF0VdltA9reyXuDR248t9knAPsULmVn0rPulxg0Ob0qjji91XM6GpKtRz+IArFDQl+v6+aW+BVVi3ujwy8Wzpj72neBL/GyaWbmxvR0dE0bdrU6Pbo6Gjc3Nyq7aOkpISSEsMFv1qdBiuFsqbDuW1WVorymYDRq9BqdVw5fg0vfw8GTe2vf8P9snIne787qG8z46sJ7N98hP1bDuvLMhKrPgVwM9cPrK2rdulPcUSejKZVj6b0HtmdL2Z+Y3LfdwOJsWVJfC1PYmxZd0N8AX79fLf+5+izcWQmZfHRrrfxC/YhKSrltvq+0yTGxsX+kYl3SxfsPGz0ZRc3JlNWoKHjW/WxdbYm+WgOxz6JpvO7IbgEGV+X2Gp8HU5+GsvOV86hsALXeg4EdHYn51rFmtbQZ3yp/7i3/vlf71yl0bN+uIc4Wu4FVqMoo5TUk3m0mVzX7H3XOLkcNWoUw4cP56233uLhhx/Wr7lMSUlh9+7dvP/++0yYMKHaPubOncvs2bMNyurRmPoK4wnrrcpJz0Wj1uDu42pQ7u7jSlZKttE2mUnZqMs0aG9Y3Bt7MQFPP3esbZSoyzTkZRWQl1Wg315aVEp2Wg6JkZXfCJkp2bh5G87qWimtcPFwIiu5qjGUv1ljLsQblMdeTMA7yKvK13snSIwtS+JreRJjy7pX42vMxcNXAQho4HtXJZcSY/MoTCsl7XQe7abW05cVJJcQvT2dbgtDcQ4sTyRd69qTeTGf6O3pNH850Ghfjr4qOs8OQV2sQV2kxc7dhmOfROPgXTFzqXK1QXXDr0yhBDsPmypnRC0t7o9MbJ2t8W3revPKNVTjNZfvvvsur7/+Oh999BEtW7bE398ff39/WrZsyUcffcTrr7/OO++8U20fERER5OTkGDzqKRqZ+hr01GUaLh+LolWPZvoyhUJBqx5NOX/witE25/66hH8DH4OFrLVD/MhIzERddvPFsf904eAVnN2dCGldcbC26tEUhZVC/yb6p+ToNNITMqnd0PB0Ru0QP1JvY8G1JUiMLUvia3kSY8u6V+NrTP2WdQGqvTjlTpAYm0fcHxmoXK3xbl2RJGtKteU//OPiFoWVAt0t3JTB2k6JnbsNpflqUk/l4tuu8hLCu4FOpyNuTya1u7hjZW2+C3muM+km6q+//jqJiYlERkayf/9+9u/fT2RkJImJiUyfPv2m7VUqFS4uLgYPc50S/++irfQb1YOew7sQFBbAqytGYeeo0p9Gmr5uHOFzhurr//zZTpw9nBi76AUCQvxo368VQyMG8NOKHfo6do4q3H1c9Y85wxZzdPtJg7Lri6JjLyZwZPsJJq8cTWi7+jR5IJTxS0ayZ+Nf+jePp787a859TGi7+vp9bFrwM09O6MtDT3XAv74PI2YPJjAsgG1f/GGWuJiTxNiyJL6WJzG2rHsxvn7BPjw7cyAhrevhU6cWnfq3Yfq6sZzee55rZ2L/pcjdOonx7dFpy5OrwK4eWCkrkisnfzscfW05/XkcWVcLKEguIfLnVNJO5+HbrmKG7+C7V7m2veJLXerJXFJP5lKYWkLa6TwOzr6KU4Adgd089XXKCjUUZ5fpHw/OaYhTgJ1B2Y0KkkvIiS6kJFuNplRHTnQhOdGFaNVafZ28+GJyogspyy+fMb1e57qsqwX8PukCRZmGt3tKP5tPYWopQQ97Ygm3dRP1evXqUa9ePYOyuLg4Zs2axRdffHFbAzPV3k0HcfNyYcQ7g3H3dSPyZDRv9JtLdmr5wmfvQE902opfTFp8BhF9P2DMwhGsOjmf9IRMtiwxvNXHoNf6M3zWoGr3+1zweP1tHeY9t5TxS8OZv/Mt/Y1ll09cq69rbWNNUFgAKoeKqfAtS37F1s6GVxYOx9nDiahTMbze+/276lTMdRJjy5L4Wp7E2LLuxfiqS9W0frgZAyf2w85RRVpcBvs2H+HrOZvNFhdzkhjfnrQzeRSll1W6StrKWkH7iPpc+E8iRz68hqZYi6OvLS3HBeFzwwxnQUoJpblq/XN1oYYL3yRRnFGGjZMSvw5uhA31M5gVPLs2nvi91c/Q9t/UUv/zqc9iyThfsUzhz+mXAXh4WSP96fbDcyMNbgh/vc71fjQlWgoSS9CpDadd437PwD3UEecAu2rHYyqFTncrE7237tSpU7Ru3RqNpmbT7D2Vz5hzGEIIIYT4f6bF8Ts9gvvfghYbb1qnxjOXP/30U7Xbo6Kiqt0uhBBCCCHuXzVOLgcMGIBCoaC6CU9z3uVdCCGEEELcO2p8QY+fnx+bN29Gq9UafRw/LnPSQgghhBD/X9U4uWzTpg3Hjh2rcvvNZjWFEEIIIcT9q8anxadNm0ZBQUGV2xs0aMAff9xdt8UQQgghhBD/jhonlw899FC12x0dHenatavJAxJCCCGEEPcuk26iLoQQQgghhDGSXAohhBBCCLO5rf/QI4QQQghxt7BCe/NKwuJk5lIIIYQQQpiNJJdCCCGEEMJsJLkUQgghhBBmI8mlEEIIIYQwG0kuhRBCCCGE2UhyKYQQQgghzOa+vBXR42N6MWhqfzx83Yg8FcPyiWu5dDSyyvqOrg6Evz+Ezk+2x9nDidSYND6dsp4j206atH9nd0fGLQmn42Ot0Wl17Nt8mBWT1lFcUFJtu0YdQxj53hDCOjRAq9ESeTKGiL5zKC0uM2kcllSTGC/Y/TYtujWpVH741+O82f9Dk/ZfK9CTiStG0aJbE4ryi9n55V7WvPENWk31t6Fo368Vz735FMHN61BaXMrpPy/wzsAFJo3BkuQYtjyJsWXdi/E19ln1y8qdLB672qQxWJrE2DQ7xl2gKK3y+6VuL09ajAqgILmEs18lkXmxAK1ah3cLZ5qF+2PnZlNln5e3pJJ0JIe8hBKUtgo8GjrS+DlfnP3tTBqjplTLqc8TyI4qIj+hGJ/WLnSYXtegTnFWGWe/TCI7qpCC5FKC+3rR7AX/m/addiaPCxtTyI0txlplRWBXdxoN9cVKqTBprMbcd8ll18GdGL1wOEvGrubC4SsMnNiPudveILzRZLLTcivVt7ZR8uFvb5KdlsN7gz8hPSETnzpe5GcXVrmPBbvfZseXe9mxfq/R7TM2TMDT150ZveegtFEybc0YJq98mbnPLa2yz0YdQ5j76xt8M+8Hlk9ci0atIbhFHXRaXc2DYGE1jfHspxdibVtxqLl4OrPyxHz+/P5Qlfv4KnIpH4V/yum95ytts7JSMOfnGWSmZDPpwbfw8HNn+rpxaMo0fPHmt1X2+eDA9kxeOZq1b37Did/PobS2om7TwBq+esuTY9jyJMaWda/GF2Dr57tYP2uT/nlJYektvup/l8TYdF3nhhi8Z3Jjizn4/jUCOrmiLtby15xruNaxo/OsYAAufJvC4Q+j6TKnAQor4wlYxvl86vX2xK2+AzqNjgvfJHPw/Wv0+DgUazvjJ4l/HHyansvCcPC2rbRNpwWlrYLgvp4kHc4x2l5bpkPloqThQB+itqbd0mvPiS7i0NxoGg70pvX4QIozyzj1eQI6rY6mw2+emN6q+y65fGrSo2xbvZvf1u0BYPGY1XTo15reI7uzcf6Pler3Ce+Os4cjEx98C41aA0BKzK39kowJCgugfZ9WjGsfweVjUQAsm7iWOb/MYNW0DWQkZRltN2bhCLYs3WYwxvjLSSaPw5JqGuO8rAKD592e6UxxYQl/fld1clmdNr1aENS4NtN7vU92ag6Rp2JYP2sjo+Y+y5ezv0NdpqnUxkppxdhPXuDz1zew/Ys/9OWxFxJMGoMlyTFseRJjy7pX4wvliU5WivE/5ncTibHpVC6Gqc+VH9Jw9LHFs7EjaafzKUwtpduHIdg4KAFoPT6QX0eeI+1sPt7NnY322WlmsMHzVuMC2T7qPNlRhXg1dqrxGK3trGjxUm0AMi8VUlZQ+e+ag7ctzUYGABD7R+Yt9ZvwVzYudewIfdoHACdfFU2e9ePoJzGEDvLBxl5Z47Eac1+tubS2UdKwTTDHd5/Rl+l0Oo7vPkPjTiFG23Tq35bzh64wYVk4mxJXsurUAobOGIBVFd9ObqZRpxDysvL1bzaA47vOoNPqCOvQwGgbt1ouNOoYQnZqLov2vcumxJUs/H0WTTqHmjQGSzIlxv/UN7w7ezb+RXFh9af/qtK4YwjRZ2LJTq34cPr7t1M4ujpQp4nxmciQ1vWoVdsTnVbHp3/P49v4z5izdQZ1q6h/p8gxbHkSY8u6V+N7XY9hD/J9yuesOrWA8DlDUdlXnlW60yTG5qNVa4nfl0VQdw8UCgXaMh0KBVjZVMTFykaBQgGZFwuq6clQWWF5MmjrdHfN4WnVOoPXBmBlW/66c6KKzLafGr/qpKQkPv30U/bv309SUhJWVlYEBwczYMAAXnjhBZRK82S9pnD1ckFpraz0jSgrJYfAUOPTvb71vGnZvQm7v97PzMfm4d/Al1eXvYjSxpoN730PwNAZAxga8aS+ja29LY06hjB+Sbi+7MWmU0iLy8DDx43sVMNTElqNltzMfNx93YyOwS+4/BvE8FlPs2r6Bq6ejKbn812Yv/MtXm4+lYSryTWOhaWYEuMbhbarT71mQSx86TOD8okrRvHwsw/pn6scbPlga4TBGsrHXUcA4O7rRlZq5f0D5WuPjOz3eoyff/tpPpv6JSnRaTw95TEW/P42I8MmVZpdvVPkGLY8ibFl3avxBfj92wOkxqSTnphJcPM6jJo7jMBQf2Y/vbCmYbAoibH5JB3JpaxAQ2A3dwDcGzqgVFlx/j/JNBrqCzo4/3USOi0UZ6tvqU+dVsfZdYl4hDrgElSx5vLgB9fIuGD4t+b31y4B5cmeQy0benxs2S+L3i2cidyaTvz+LAIecKM4W83l/6YC5Ws4zaVGyeXff//NI488QoMGDbC3t+fKlSsMGzaM0tJSpk6dyhdffMH27dtxdjY+bXxdSUkJJSWGs1ZanQYrxb+fmFpZKcpnAkavQqvVceX4Nbz8PRg0tb/+DffLyp3s/e6gvs2Mryawf/MR9m85rC/LSKz6FMDNXF/DsXXVLv0pjsiT0bTq0ZTeI7vzxcxvTO77btMnvAdRp2MqLTpfP2sT3y38Wf98we+zWB3xNRcPXzHLfq/H+Ou5W9i/+Uj5PsI/5evYT+kyqBNbV+0yy37uBDmGLU9ibFl3Q3wBfv18t/7n6LNxZCZl8dGut/EL9iEpKuW2+r7TJMbGxfyRiXdLZ+w9yi/WUblY025KHU6tTiBqWzoKBQR0dsO1nj2KW5zkPb0mgdy4Yh56t75BectXaqMprZgw2f3qJTpG1MPuf/s25wU1VfFu4UyT5/049XkCx5fFYWWjoOFTPmRcKKhyPakpapRcTpo0icmTJzNr1iwANmzYwLJlyzh06BBZWVn06NGDN998k8WLF1fbz9y5c5k9e7ZBWT0aU1/RtIbDN5STnotGrcHdx9Wg3N3HlayUbKNtMpOyUZdp0N6wuDf2YgKefu5Y2yhRl2nIyyowmNkqLSolOy2HxMjKb4TMlGzcvF0MyqyUVrh4OJGVXNUYyt+sMRfiDcpjLybgHeRV5eu9E0yJ8XV2Diq6P/OAwULu67LTcg0WoWvUGtITMo3GOCs5m7B2hqddro8ns8oYl5fHnK+IcVmpmqSoFLwDPasd979JjmHLkxhb1r0aX2MuHr4KQEAD37squZQYm0dhWilpp/NpP7WOQbl3C2d6Lg2jJFeNlVKBjaOS7S+dx+Ef8Tbm9JoEko/n8eDs+th7Gp7uv57A3sjBy9boBT2W1OCxWtR/1IviLDW2TkoKU0u58HWyWcdRozWXx48f5/nnn9c/HzZsGMePHyclJQV3d3fmz5/P999/f9N+IiIiyMnJMXjUUzSq+ej/QV2m4fKxKFr1aKYvUygUtOrRlPMHjc+AnfvrEv4NfFDc8JWkdogfGYmZRi8MuZkLB6/g7O5ESOt6+rJWPZqisFLo30T/lBydRnpCJrUbGp7OqB3iR+ptLLi2BFNifF2XQR2xUVmz6z/7bmsM5w9doW6zINxqVXywte7ZnIKcQmLPxxttc+VYFKXFpQanjJTWSnzr1iIlNv22xmNOcgxbnsTYsu7V+BpTv2VdgGovTrkTJMbmEftHJipXa3xauxjdrnKxxsZRSdrZfEpy1fi2NV4Pyte8nl6TQNKRHDq/HYzjv5ww1pRCocDewwalrRXxB7Kx97TBLdjebP3XKLn09vYmKaniysSUlBTUajUuLuUBDwkJITPz5lcsqVQqXFxcDB7mOiX+30Vb6TeqBz2HdyEoLIBXV4zCzlGlP400fd04wucM1df/+bOdOHs4MXbRCwSE+NG+XyuGRgzgpxU79HXsHFW4+7jqH3OGLebo9pMGZdcXRcdeTODI9hNMXjma0Hb1afJAKOOXjGTPxr/0bx5Pf3fWnPuY0HYVU+abFvzMkxP68tBTHfCv78OI2YMJDAtg2w1XNt8tahrj6/qM7M6BH/8mLzO/0jYHF3uDeL76wJvEXUwwKLvu2I5TxJ6P5/UvxxPcvA5te7XghXef4acVv1FWWr4mJrRdfdac+xhP//J1NIV5RfyychfDZw2iTc/m1G7ox8QVowBMvmrdUuQYtjyJsWXdi/H1C/bh2ZkDCWldD586tejUvw3T143l9N7zXDsT+y9F7tZJjG+PTqsjdk8WgV3dK52Ojvkjk8zLBRQklxD3ZxZHP46h/qNeBvesPPBuFFHbKyYmTq9JJG5fFm0mBmFtb0VxdhnF2WUGp8FL89X68uLsMnqvaoSVrUL/vCTXcE1nbnwxOdFFlOWrURdpyIkuIifa8KKb62XqYi2luWpyoovIjS/Wb088ksPuSZcM2lz5KZXc2CJy44q59H0KV35Io9lI/zt3WnzAgAG88sorfPTRR6hUKt577z26du2KvX15tnvp0iUCAgLMNjhT7N10EDcvF0a8Mxh3XzciT0bzRr+5+iuLvQM90Wkrftlp8RlE9P2AMQtHsOrkfNITMtmyxPBWH4Ne68/wWYOq3e9zweP1t3WY99xSxi8NZ/7Ot/Q3ll0+ca2+rrWNNUFhAagcVPqyLUt+xdbOhlcWDsfZw4moUzG83vv9u+pUzHU1jTFA7YZ+NHuoEa/3ft9on+MWvUCvEd2q3W9P5TMAaLU63nz8QyauGMXiA+9RXFDCzi/3su6G0+0qBxVBYQFY21Qc4qumb0Cj1vD6+nHY2tty8chVpj3yHvnZd8fFPNfJMWx5EmPLuhfjqy5V0/rhZgyc2A87RxVpcRns23yEr+dsNltczElifHvSzuRTlF5Gne4elbblJ5Zw4etkSvM1OHjb0HCgN/UfNVx6UpBSQukNyWD0jgwADrwTZVCv1djaBHUr38eRBTFknK/67419LRt6La84i3to7jWDm73vmV4+K/3EpuaVygByooqI359t0I+6UEN+ouE1Lqkn8ri8ORVtmQ7XuvZ0mF4Hn1ZVz8qaQqHT6W757rv5+fm8+OKLbN68GY1GQ6dOndiwYQP16pVPi+/YsYOcnBwGDar+4DTmeuIghBBCCGGKVser/y9t4vbNb/HdTevUaObSycmJjRs3UlxcjFqtxsnJ8MagvXr1qtkIhRBCCCHEfcWku3va2Zn2vzKFEEIIIcT97b76Dz1CCCGEEOLOkuRSCCGEEEKYjSSXQgghhBDCbCS5FEIIIYQQZiPJpRBCCCGEMBtJLoUQQgghhNmYdCsice8x5791EsbptLf8/wiEEEJYQJnOPP9KWtwembkUQgghhBBmI8mlEEIIIYQwG0kuhRBCCCGE2UhyKYQQQgghzEaSSyGEEEIIYTaSXAohhBBCCLO5L29F9PiYXgya2h8PXzciT8WwfOJaLh2NrLK+o6sD4e8PofOT7XH2cCI1Jo1Pp6znyLaTJu3f2d2RcUvC6fhYa3RaHfs2H2bFpHUUF5RU265RxxBGvjeEsA4N0Gq0RJ6MIaLvHEqLy0wahyX1H9OLQVP64+HrStTpWJZPqjrGSmslQ15/gp7Pd8UrwJ24y0msifiav3ecMnn/zu6OjFs8kg6Plsd4/5YjrJhcfYzdfVx56cPnaP1wMxyc7Yi7nMQ3c7ewf8sRk8dhKXIMW969GOMFu9+mRbcmBmW/rNzJ4rGrTRqDJUl8LU9ibJo/xp+jKK20UnlQLy+avhhISXYZFzYkkH46D02xFkc/FfUH+uLXwa3KPq9uSSblSA75icUoba1wb+hI6LP+OPnbmTRGTamWs6vjyIkqpCChGO/WrrSZFmxQpzirjAtfJZATVUhhcgl1+9Si8Qu1b9r3r8+cqFTW8tW6+Hd2N2msxtx3yWXXwZ0YvXA4S8au5sLhKwyc2I+5294gvNFkstNyK9W3tlHy4W9vkp2Ww3uDPyE9IROfOl7kZxdWuY8Fu99mx5d72bF+r9HtMzZMwNPXnRm956C0UTJtzRgmr3yZuc8trbLPRh1DmPvrG3wz7weWT1yLRq0huEWdu/LeiV0HdWL0R8+zZNxqLh65ysBX+/HB1ghebDLFaIxfePcZHh72IJ+8soq4S4m07dWCWd+/xqQubxN5MtroPj7aVR7jnV9WEeMvJ+Dh50ZE3w9Q2iiZ+vkrTPr0ZeYNrzrG09eOw9HNgVkDPyInPY8eQzoz85tJjO/4RpXjuBPkGLa8ezXGAFs/38X6WZv0z0sKK/+RvNMkvpYnMTbdAx80BG3F87zYIo7MicSvoxsAp5bHUFagoe30YGydrUncn8WJT67hMDcU13oORvvMvJBPnd5euNZ3QKfRcenbJI7MuUqXhY2wtjN+781fnzlBt6WNcfBWVdqm0+pQ2lpRt28tkg9nG22vLdOicrGmwZM+RP+aVqMYNB8TRK2WLvrn1g7mvT/ofXda/KlJj7Jt9W5+W7eH2AsJLB6zmpLCUnqP7G60fp/w7jh7ODLryQWc++sSKTFpnP7zAlGnY0zaf1BYAO37tOLjl1dy8chVzh24xLKJa+n2zAN4+lX9rWDMwhFsWbqNjfN/JOZ8PPGXk/jzu0OUlapNGoclPTXpUbat+Z0d6/eWx3js/2L8Qjej9R959kG++fAHjm4/SfK1VH5ZuZMj207w9ORHTdp/YJg/7fq05OPRq/QxXj5pHd2e6YRHNTFu3KkhPy7/jUtHI0m+lsrXc7dQkF1ASOt6Jo3DUuQYtrx7NcZQ/oc4KyVH/yjMKzJpDJYk8bU8ibHpVC42qNwqHqnHc3HwscWjsRMAWZcKqNunFm4NHHHwUdHgKV9sHJXkRlWdiLd/owG1u3niHGiPS10Hmo8Noji9jNwo016btZ2SpqMCCXrYC5WbjdE6Dt4qGr9Qm9pdPWucHFo7Kg1ioLQ1bzp4XyWX1jZKGrYJ5vjuM/oynU7H8d1naNwpxGibTv3bcv7QFSYsC2dT4kpWnVrA0BkDsDLxP9o06hRCXlY+l49F6cuO7zqDTqsjrEMDo23carnQqGMI2am5LNr3LpsSV7Lw91k06Rxq0hgsydpGSUjrepz4R4xP/H6GRh0bGm1jo7Kh7B+nRUuLS2nyQJhJY2jcsSF5WflcuTHGu8tj3Ki98RgDnD94ma6DOuHs7ohCoaDb4E7Y2tlweu95k8ZhCXIMW969GuPregx7kO9TPmfVqQWEzxmKyt7WpDFYisTX8iTG5qNVa0nYn0nt7p4oFOWxcA91JPFgFqX5anRaHYkHstCW6fBo4nzL/aoLy6dGbZzuzv8YdG5NPDtHnebAG5eI+yMDnc68Z5hqfFp82bJlHDlyhH79+jFkyBC++uor5s6di1arZeDAgbz77rtYW9+Zs+2uXi4orZVkpeQYlGel5BAY6m+0jW89b1p2b8Lur/cz87F5+Dfw5dVlL6K0sWbDe98DMHTGAIZGPKlvY2tvS6OOIYxfEq4ve7HpFNLiMvDwcSM71fCUhFajJTczH3dfN6Nj8Av2AWD4rKdZNX0DV09G0/P5Lszf+RYvN59KwtXkGsfCUlyuxzjVWIwDjLb5e8dpBk7sx+l9F0iKTKFVj6Z0HtAeK2XFd5shrw9g6IwB+ue29rY06tCA8YtH6stGNX+NtLgM3KuIcV41MQZ4f+giZn49kf+mrkFdpqaksJTZT39MYmRKDSJgWXIMW969GmOA3789QGpMOumJmQQ3r8OoucMIDPVn9tMLaxoGi5H4Wp7E2HxSjuagLtBQu6unvqzVpLqcWBTNrhfPoFCC0taK1q/Vw9G38ulrY3RaHefXx+Me6ohzkL2+/Ojcq2ReKDCou2/qRf3P9rVs6bKw0W2+opsLGeyHZxMnlCor0k/ncW5NHJpiDXX7epttHzXKAt9//33mz59Pr169mDx5MjExMXz00UdMnjwZKysrPvnkE2xsbJg9e3a1/ZSUlFBSYrjgV6vTYKX49zN8KytF+WzL6FVotTquHL+Gl78Hg6b217/hflm5k73fHdS3mfHVBPZvPsL+LYf1ZRmJWSaP4fr//d66ahe/rdsDQOTJaFr1aErvkd35YuY3Jvd9N/h0yjomf/Yya85+DDodiZEp7Fi/h94vVJy+2bpqJ39+f0OMv5zAvs2HOfBDxcU2txNjgBGzB+Pk5sj0Xu+Tm5HLA4+3Y+Y3E5nS/R2iz8bdVt93khzDlnc3xBjg189363+OPhtHZlIWH+16G79gH5Ki7p4vSTUl8bU8ibFxcb9nUKulC3YeFaeeL29MoqxQQ/s3G2DrrCTlaA4nFkXTcXYILjcki1U590U8+XHFdJxtOIvcbHQQmtKKGcK9E8/TdkYwdh7lM7dWStNmkWsq5Clf/c+u9RzQlGiJ+jn1ziWX69atY926dQwcOJBTp07Rpk0b1q9fz7PPPgtAWFgY06dPv2lyOXfu3Ep16tGY+oqmNRy+oZz0XDRqDe4+rgbl7j6uZKVkG22TmZSNukyD9oaLDmIvJuDp5461jRJ1mYa8rALysiq+bZQWlZKdlmN0xiszJRs3bxeDMiulFS4eTmQlVzWG8jdrzIV4g/LYiwl4B3lV+XrvhNzrMfauHOPMKl5fTnoe7zy9EBuVDS6eTmQkZvHiB8MMPkj+GeOSolKy03KNxjirihg7VxNjv2AfBozrw0stphJzvjzOUadjafpgGI+P6cWScWtu5eVbnBzDlnevxtiYi4evAhDQwPeuSX4kvpYnMTaPorRS0s/k0ea1inX3BcklxPyWzkMLwnAOLE8kXeo6kHkxn5jf0mj2UlC1fZ77Io7U4zl0fCcEe0/D0/3Xk8gb2XvZGr2g59/k1sCBq/9NRlOmRWljntWSNeolMTGRtm3bAtCiRQusrKxo2bKlfnvr1q1JTEy8aT8RERHk5OQYPOopbn8qWF2m4fKxKFr1aKYvUygUtOrRlPMHrxhtc+6vS/g38NGvtQCoHeJHRmIm6jJNjcdw4eAVnN2dDC4SadWjKQorhf5N9E/J0WmkJ2RSu6Hh6YzaIX6kxtTsCjBLU5dpuHL8Gi17VHwRUCgUtOzelAuHLlfbtqykjIzELJTWSh58sj0Hfz5m0hjOH7pcOcbdy2N84YjxGKscyt/UWq3WoFyr0WJldfcsPZZj2PLu1RgbU79lXQAykm5vNsmcJL6WJzE2j7g9GahcranVuiJJ15SW/424MU7wv7Mz1SxL1Ol0nPsijuQjOXR4q8EdTxhrIje6CBtHpdkSS6hhcunr68v58+UXP1y5cgWNRqN/DnDu3Dm8vW8+rapSqXBxcTF4mOuU+H8XbaXfqB70HN6FoLAAXl0xCjtHlf5U3fR14wifM1Rf/+fPduLs4cTYRS8QEOJH+36tGBoxgJ9W7NDXsXNU4e7jqn/MGbaYo9tPGpRdXxQdezGBI9tPMHnlaELb1afJA6GMXzKSPRv/0r95PP3dWXPuY0Lb1dfvY9OCn3lyQl8eeqoD/vV9GDF7MIFhAWz74g+zxMWc/rtoK/1e7EHP57sQGObPq8tfLI/x/25XMW3tWMLfH6KvH9a+AZ0HtMO3njdNO4fxwdYIrKwUbFrwk77OP2P8wbOL+fs34zGOu5jI0e0nmfTZy4S2q0/jBxoybvFI9mw8qJ9B8/R3Z82ZhfoYx11MJOFKEpNWvERou/r4Bfvw1KRHaf1IMw78ePTfCt0tkWPY8u7FGPsF+/DszIGEtK6HT51adOrfhunrxnJ673munYn9lyJ3ayS+licxvj06rY74PRkEdPUwOB3t5G+Hg6+KM5/Hkn21gILkEqJ+TiH9TB4+7SqS0MPvXSF6e8UX53Nr4knYl0XLV+tgba+kJLuMkuwyfbIKUJqv1peXZJfx8MqmKG2tKspyDS98zYsvIje6kLJ8NWWFGnKjC8mNNrxi/XqZulhDaa6a3OhC8uIrrlBPPpLN3skVeVrKsRzidqeTF1tUPku7I43IH1Ko06fW7Qf1BjU6Lf7ss88yfPhwnnjiCXbv3s306dOZOnUqGRkZKBQK5syZw9NPP23WAdbU3k0HcfNyYcQ7g3H3dSPyZDRv9JtL9v8uQPEO9ER3w+xVWnwGEX0/YMzCEaw6OZ/0hEy2LCm/ncp1g17rz/BZg6rd73PB40n53wzNvOeWMn5pOPN3vqW/sezyiWv1da1trAkKC0DlUPHNZsuSX7G1s+GVhcNx9nAi6lQMr/d+/646FXPd3u8O4lrLheGzBuHu60bUqRhmPjbvhhh7Gdzb0EZlwwuzn8Ev2Jui/GKObD/Jhy8spyCn4k0yaEp/nn+7+mPn+QYTKmI8fCnjFofz4W9vlsd4S/nNe6+ztrEmMCwAlX15jDVqDTMf/5AX5wzl3S3TsHeyIyEyhY/CP+Xo9pNmiox5yDFsefdijNWlalo/3IyBE/th56giLS6DfZuP8PWczWaLi7lIfC1PYnx70s/kUZxeRu1ungblVtYK2s0I5uLXifw9PwpNsRYHH1tajK2Dd6uK5LIwpZTSvIrbrMXuTAfg8GzDWdvmY4L0+zi+8BqZ5/OrHJN9LVu6L6u4wfzf86IMbva+//VLAPTb2KpSGUBuVBGJB7IM+lEXaihIrLjGxUqpIGZHOue/TAAdOPiqaPR8AIEPG8bhdil0Nbj+XKvVMm/ePA4ePMgDDzzAjBkz2LhxI9OnT6ewsJD+/fuzbNkyHB0dazyQnspnatxG3DqFibebELfubrxZuBBC/H/S9Jj8rbO0T1p+e9M6NUouLUmSS8uS5NLyJLkUQog7S5JLy7uV5PLuuZJBCCGEEELc8yS5FEIIIYQQZiPJpRBCCCGEMBtJLoUQQgghhNlIcimEEEIIIcxGkkshhBBCCGE2klwKIYQQQgizqdF/6BFCCCGEuFupteb5V9Li9sjMpRBCCCGEMBtJLoUQQgghhNlIcimEEEIIIcxGkkshhBBCCGE2klwKIYQQQgizkeRSCCGEEEKYzX15K6LHx/Ri0NT+ePi6EXkqhuUT13LpaGSV9R1dHQh/fwidn2yPs4cTqTFpfDplPUe2nTRp/87ujoxbEk7Hx1qj0+rYt/kwKyato7igpNp2jTqGMPK9IYR1aIBWoyXyZAwRfedQWlxm0jgsqf+YXgya0h8PX1eiTseyfFLVMVZaKxny+hP0fL4rXgHuxF1OYk3E1/y945TJ+3d2d2Tc4pF0eLQ8xvu3HGHF5Opj7O7jyksfPkfrh5vh4GxH3OUkvpm7hf1bjpg8DkuRY9jy7sUYL9j9Ni26NTEo+2XlThaPXW3SGCxJ4mt5EmPT/DnhNMXppZXKA3vWom5/X/a9esZou+YTg/Ht6FFlv/kJRVz5Op6sC/lotTqcAuxoMbk+9l6qGo9RU6rlwpoYcq8VUpBQhFdrN1q91sCgTklWKZc2xJMbVUBhSglBvb0JGxF0S/0n7E0nZmsKhcnFKO2V+HZwp1F4nRqPsyr3XXLZdXAnRi8czpKxq7lw+AoDJ/Zj7rY3CG80mey03Er1rW2UfPjbm2Sn5fDe4E9IT8jEp44X+dmFVe5jwe632fHlXnas32t0+4wNE/D0dWdG7zkobZRMWzOGyStfZu5zS6vss1HHEOb++gbfzPuB5RPXolFrCG5RB51WV/MgWFjXQZ0Y/dHzLBm3motHrjLw1X58sDWCF5tMMRrjF959hoeHPcgnr6wi7lIibXu1YNb3rzGpy9tEnow2uo+PdpXHeOeXVcT4ywl4+LkR0fcDlDZKpn7+CpM+fZl5w6uO8fS143B0c2DWwI/ISc+jx5DOzPxmEuM7vlHlOO4EOYYt716NMcDWz3exftYm/fOSwsp/JO80ia/lSYxN13FOI3Taiuf5cUUc++AyPh3dsfO0peunLQzqx+9OI/qXZLxaulbZZ2FKMUffuUhANy/qPx2AtYMV+XHFWNlUfYJ4x9C/eWhJM+xrVU4+dVodVrZWBPX2JuVIltH2WrUOWxdrgp/0I+bXlJu86grRW5OJ2ZpCw2dr49rAEU2xlqI08/4O7rvT4k9NepRtq3fz27o9xF5IYPGY1ZQUltJ7ZHej9fuEd8fZw5FZTy7g3F+XSIlJ4/SfF4g6HWPS/oPCAmjfpxUfv7ySi0eucu7AJZZNXEu3Zx7A08+9ynZjFo5gy9JtbJz/IzHn44m/nMSf3x2irFRt0jgs6alJj7Jtze/sWL+3PMZj/xfjF7oZrf/Isw/yzYc/cHT7SZKvpfLLyp0c2XaCpyc/atL+A8P8adenJR+PXqWP8fJJ6+j2TCc8qolx404N+XH5b1w6GknytVS+nruFguwCQlrXM2kcliLHsOXdqzGG8j/EWSk5+kdhXpFJY7Akia/lSYxNZ+tig8qt4pF2PBt7HxXujZxRWCkMtqncbEg9moVvRw+s7aq+QfvVjQl4tXSl4bOBuNRzwMHHDu+2bqhcbUwao7WdksYv1qH2w7VQuRnvw76WirARQfh38cLa4dZuHl+Wr+bqpkSajq2HX2dPHHzscK7jgHdbN5PGWZX7Krm0tlHSsE0wx3dXTGnrdDqO7z5D404hRtt06t+W84euMGFZOJsSV7Lq1AKGzhiAlZXCpDE06hRCXlY+l49F6cuO7zqDTqsjrEMDo23carnQqGMI2am5LNr3LpsSV7Lw91k06Rxq0hgsydpGSUjrepz4R4xP/H6GRh0bGm1jo7Kh7B+nRUuLS2nyQJhJY2jcsSF5WflcuTHGu8tj3Ki98RgDnD94ma6DOuHs7ohCoaDb4E7Y2tlweu95k8ZhCXIMW969GuPregx7kO9TPmfVqQWEzxmKyt7WpDFYisTX8iTG5qNVa0nan0lANy8UisqxyI0qIC+miIDuXlX2odPqSDuRg4OfHcfmXuaP0Sc59OYFUo8an3G8kzLO5IJOR0lmKQdeO8vecac4tSiS4gzzzlzW6LT4hAkTGDx4MA899NBt7bSkpISSEsM1GVqdBivF7f3bJlcvF5TWSrJScgzKs1JyCAz1N9rGt543Lbs3YffX+5n52Dz8G/jy6rIXUdpYs+G97wEYOmMAQyOe1LextbelUccQxi8J15e92HQKaXEZePi4kZ1qeEpCq9GSm5mPu6+b0TH4BfsAMHzW06yavoGrJ6Pp+XwX5u98i5ebTyXhanKNY2EpLtdjnGosxgFG2/y94zQDJ/bj9L4LJEWm0KpHUzoPaI+VsuK7zZDXBzB0xgD9c1t7Wxp1aMD4xSP1ZaOav0ZaXAbuVcQ4r5oYA7w/dBEzv57If1PXoC5TU1JYyuynPyYx8tZPJ1iaHMOWd6/GGOD3bw+QGpNOemImwc3rMGruMAJD/Zn99MKahsFiJL6WJzE2n9Sj2agL1fh38TS6Pf6PdBwD7HBr6FRlH6W5ajTFWq79lEzIYH9ChtYm41QOJz+JpO2boXg0dgbg2LzLZF/MN2h7YNo5rqe0dl62dF7Q1CyvqypFqSXotBD1YzJhwwOxdlBydVMCf39wmQc+bIyVtXnmHGuUXC5fvpwVK1ZQv359XnzxRUaMGIGvr2+Ndzp37lxmz55tUFaPxtRXWDaoxlhZKcpnW0avQqvVceX4Nbz8PRg0tb/+DffLyp3s/e6gvs2Mryawf/MR9m85rC/LSDT9G4rif98ct67axW/r9gAQeTKaVj2a0ntkd76Y+Y3Jfd8NPp2yjsmfvcyasx+DTkdiZAo71u+h9wsVp2+2rtrJn9/fEOMvJ7Bv82EO/FBxsc3txBhgxOzBOLk5Mr3X++Rm5PLA4+2Y+c1EpnR/h+izcbfV950kx7Dl3Q0xBvj18936n6PPxpGZlMVHu97GL9iHpKi750tSTUl8LU9ibFzCnnQ8W7pi51F59lRTqiX5r0yCn/Srto/r68q927hRp195TuRS14Hsy/nE70rTJ5dNXq6LtrRisef+yWdpPT0EO4/y094KpWmzyDWh04FOoyNsRCBezcvXkDafEMyeV06ReS4PrxZVryutiRpf0LNjxw5+/vlnFixYwFtvvUXfvn156aWX6NevH1ZWt5bxRkREMGXKFIOyJ93Cq6h963LSc9GoNbj7GAbH3ceVrJRso20yk7JRl2nQ3nDRQezFBDz93LG2UaIu05CXVUBeVoF+e2lRKdlpOUZnvDJTsnHzdjEos1Ja4eLhRFZyVWMof7PGXIg3KI+9mIB3UNVT8XdC7vUYe1eOcWYVry8nPY93nl6IjcoGF08nMhKzePGDYQYfJP+McUlRKdlpuUZjnFVFjJ2ribFfsA8DxvXhpRZTiTlfHueo07E0fTCMx8f0Ysm4Nbfy8i1OjmHLu1djbMzFw1cBCGjge9ckPxJfy5MYm0dRWgkZZ3JpOaW+0e0ph7PQlGirnNW8ztbFGoVSgVOAnUG5Y4A92Zfy9M+NJbD2tWyNXtBjKdfXbzoF2OvLbF1ssHW2NnoFvalqPP/ZrFkzFi1aRGJiIhs2bKCkpIQBAwYQGBjIzJkzuXr16k37UKlUuLi4GDxu95Q4gLpMw+VjUbTq0UxfplAoaNWjKecPXjHa5txfl/Bv4GOw1qJ2iB8ZiZmoyzQ1HsOFg1dwdncyuEikVY+mKKwU+jfRPyVHp5GekEnthoanM2qH+JEak1bjMViSukzDlePXaNmjYpZZoVDQsntTLhy6XG3bspIyMhKzUForefDJ9hz8+ZhJYzh/6HLlGHcvj/GFI8ZjrHIof1NrtVqDcq1Ge8tfiv4Ncgxb3r0aY2Pqt6wLQEbS3bO2S+JreRJj80jYm46tqw1erdyMb/8jjVpt3LB1qf6iHCtrK1yCHShIKjYoL0wqxs6E2xBZklto+en9G8dalq+mNE+NXS3zrX01+a+qjY0NgwcPZvv27URFRfHSSy/xn//8h9DQO7uA/7+LttJvVA96Du9CUFgAr64YhZ2jSn+qbvq6cYTPGaqv//NnO3H2cGLsohcICPGjfb9WDI0YwE8rdujr2DmqcPdx1T/mDFvM0e0nDcquL4qOvZjAke0nmLxyNKHt6tPkgVDGLxnJno1/6d88nv7urDn3MaHtKr4tbVrwM09O6MtDT3XAv74PI2YPJjAsgG1f/PEvRK1m/rtoK/1e7EHP57sQGObPq8tfLI/x/25XMW3tWMLfH6KvH9a+AZ0HtMO3njdNO4fxwdYIrKwUbFrwk77OP2P8wbOL+fs34zGOu5jI0e0nmfTZy4S2q0/jBxoybvFI9mw8qJ9B8/R3Z82ZhfoYx11MJOFKEpNWvERou/r4Bfvw1KRHaf1IMw78ePTfCt0tkWPY8u7FGPsF+/DszIGEtK6HT51adOrfhunrxnJ673munYn9lyJ3ayS+licxvj06rY7EvRn4d/HEysjp6MLkYrIu5lO7igt59r92lpQbLtip29+X5INZxO9OozC5mNjfUkk7nk1gz1r6OmX5akqyy/SPrp+2wMrGSv+8NNfwwtf8+CJyowspy1ejLtSQG11IbrThraOul2mKtZTmqcmNLiQ/vuLq+5SjWex/7az+uaOfHbXaunFxfSzZl/PJiyvizKfXcPS305++Nwez3OcyKCiId955h1mzZrFr1y5zdGmyvZsO4ublwoh3BuPu60bkyWje6DeX7P9dgOId6InuhtmrtPgMIvp+wJiFI1h1cj7pCZlsWVJ+O5XrBr3Wn+GzBlW73+eCx5Pyvxmaec8tZfzScObvfEt/Y9nlE9fq61rbWBMUFoDKoeIbzZYlv2JrZ8MrC4fj7OFE1KkYXu/9/l11Kua6vd8dxLWWC8NnDcLd142oUzHMfGzeDTH2Mri3oY3KhhdmP4NfsDdF+cUc2X6SD19YTkFOxZtk0JT+PP/209Xu9/kGEypiPHwp4xaH8+Fvb5bHeEv5zXuvs7axJjAsAJV9eYw1ag0zH/+QF+cM5d0t07B3siMhMoWPwj/l6PaTZoqMecgxbHn3YozVpWpaP9yMgRP7YeeoIi0ug32bj/D1nM1mi4u5SHwtT2J8ezLO5lKcXkpAN+PJY8KedOw8bPFs7mJ0e2FiMerCihlfn3buNH6xDtd+SuLi+lgc/ctvoO4eVpGwnfz4KlkX8o11B5Rf0NNlaXP98+MfXjE4VX0oovzOJr2+aVupDCD3WiHJBzIN+lEXaihMNJxRbTamHpe+iuP4/CsoFODeyJk2EQ3NdjEPgEKn093yHY7r1avH33//jadn9esPTNFT+YzZ+xQVFCbebkLcurvxZuFCCPH/SdjR++5/w9x1lrb+z03r1Oi3cO3aNZMHI4QQQggh7n93z5UMQgghhBDinifJpRBCCCGEMBtJLoUQQgghhNlIcimEEEIIIcxGkkshhBBCCGE2klwKIYQQQgizkRtCCSGEEOK+YKXQ3rySsDiZuRRCCCGEEGYjyaUQQgghhDAbSS6FEEIIIYTZSHIphBBCCCHMRpJLIYQQQghhNpJcCiGEEEIIs7kvb0X0+JheDJraHw9fNyJPxbB84louHY2ssr6jqwPh7w+h85PtcfZwIjUmjU+nrOfItpMm7d/Z3ZFxS8Lp+FhrdFod+zYfZsWkdRQXlFTbrlHHEEa+N4SwDg3QarREnowhou8cSovLTBqHJTR7MIxBr/UnpHU9PP09eOepBfz1099V1vfwdePl+c/TsE0w/g18+GHZdj577cvbGoONyobRHz1Ht8EPYKOy4e8dp1g64QuyU3OqbOPm7cqoucNo80gzHN0cObPvAssnrSPxavJtjcVS5Bi2vHsxxgt2v02Lbk0Myn5ZuZPFY1ebNAZLkvhansTYNHvGn6U4vbRSeVAvL+o+5sOfr54z2q7lpHr4dnQ3uk1drOHy14mk/J1NWZ4ae28VdfrUIqhnLZPGqCnVcm51LLnXCilIKKZWa1daT61vUKc4q4xLX8WTE1VIYUoJdfrUotGIwFveR2memgOvX6Aks4yH1zTHxtF8KeF9l1x2HdyJ0QuHs2Tsai4cvsLAif2Yu+0NwhtNJjstt1J9axslH/72JtlpObw3+BPSEzLxqeNFfnZhlftYsPttdny5lx3r9xrdPmPDBDx93ZnRew5KGyXT1oxh8sqXmfvc0ir7bNQxhLm/vsE3835g+cS1aNQaglvUQafV1TwIFmTnaEfU6Rh+W7eHWd+/dtP6NiobctJz+XruZgZO7HdL+5i6Zgwp0Wl89d73Rre/snA4Hfq24v0hiyjILWTc4pHM+m4Kk7vOqrLPd/77GpoyDbOeWkBhbhFPTXqUD7fP5KXmUykurD5h+rfJMWx592qMAbZ+vov1szbpn5cUVv4jeadJfC1PYmy6Bz4IRXfD7TDz4or4e85VfDq4Y+9lS/fPmhnUj9udzrWfU/Bq6VJlnxe/jCfzXD7Nx9XFvpYtGafzOP9FLHbuNni3dTPaZvuQ43RZ0gQHb1WlbTqtDqWtFXX6eJNyJNtoe22ZFlsXa+oP9CX619Sbvu5/OrsyBucge0oyzf/l/75LLp+a9CjbVu/mt3V7AFg8ZjUd+rWm98jubJz/Y6X6fcK74+zhyMQH30Kj1gCQEpNm8v6DwgJo36cV49pHcPlYFADLJq5lzi8zWDVtAxlJWUbbjVk4gi1LtxmMMf5yksnjsJSjv53k6G8nb7l+yv++GQP0fqH7be/fwcWePiO7M+/5pZzcU/7tcuGoz1hz9mPCOjTg4uGrldoEhPjRuGNDXmoxlZjz8QAsGbeGjfGf0W3IA2z/4o/bHpc5yTFsefdqjKH8D3FWStWz9HcDia/lSYxNZ+tiY/A86sdkHHxUeDR2QqFQoHIz3J5yNBvfju5Y2ymr7DP7cgH+XTzwbOIMgMMjKuJ2p5EdWVhlclkdazslTUYFAZB1KR91oaZSHQdvFY1eKJ+pjP8jo0b9x+5Io6xAQ4On/Eg/WfnLyO26r9ZcWtsoadgmmOO7z+jLdDodx3efoXGnEKNtOvVvy/lDV5iwLJxNiStZdWoBQ2cMwMpKYdIYGnUKIS8rX/9mAzi+6ww6rY6wDg2MtnGr5UKjjiFkp+ayaN+7bEpcycLfZ9Gkc6hJY7ifNWwdjI2ttcHvOO5SIikxaTTu2NBoGxtV+XeoG0/N6nQ6ykrUNO0cZtkB15Acw5Z3r8b4uh7DHuT7lM9ZdWoB4XOGorK3NWkMliLxtTyJsflo1VoS92cS0M0ThaJyLHKiCsmLLqJ2d89q+3Fr6EjqsRyKM0vR6XRknMujIKkEr+bOlhq6yfLji4jcnETzcXXBtF//Td1XM5euXi4orZWVvhFlpeQQGOpvtI1vPW9adm/C7q/3M/Oxefg38OXVZS+itLFmw/9Oyw6dMYChEU/q29ja29KoYwjjl4Try15sOoW0uAw8fNzITjX8FqDVaMnNzMfd183oGPyCfQAYPutpVk3fwNWT0fR8vgvzd77Fy82nknCXrgs0lx5DOzNxxUv65zYqa3Q6eHrKY/qymY/N4+yBi7j7ulFaUkZBjuGpnKzUHNx93Iz2H3exPPkMf38Ii8euprigmIETH6VWoCceVfxO7hQ5hi3vXo0xwO/fHiA1Jp30xEyCm9dh1NxhBIb6M/vphTUNg8VIfC1PYmw+KUdzUBdoCOjqYXR7/B/pOAbY4R7qVG0/jUcGcvbzWPaMPYtCCSgUNH05CI9GFcnl33OvknUx36DdgWkX9D/b17LlwQWNTX8xt0BbpuXUkmhCnw3A3suWwhTLLAurUXJ5/Phx3N3dqVevHgBfffUVn332GbGxsdSpU4fx48czZMiQm/ZTUlJCSYnhC9LqNFgpqp5ythQrK0X5bMvoVWi1Oq4cv4aXvweDpvbXv+F+WbmTvd8d1LeZ8dUE9m8+wv4th/VlGYlVnwK4GcX/vjluXbVLf4oj8mQ0rXo0pffI7nwx8xuT+74XHPz5GBePVJzOfvGDYWQkZvLDsu36svSETJP716g1vDv4Y6asGs3mtDVo1BqO7z7DkW0njH5TvdfIMWx5d0OMAX79fLf+5+izcWQmZfHRrrfxC/YhKSrltvq+kyS+licxNi7+j3S8Wrpg51F59lRTqiXpQBb1B/retJ+Y7WlkXymg9bRg7L1sybyQz/kv4lC52+DVrHytZtPRQWhKKxZ77pt0njav10flUX4a3kpp+b9Hl75JxDHADv+Hqp+JvV01Si5HjhzJwoULqVevHqtXr+bVV1/lpZde4vnnn+fSpUu89NJLFBYWEh4eXm0/c+fOZfbs2QZl9WhMfUXTmr+CG+Sk56JRa3D3cTUod/dxJSsl22ibzKRs1GUatDdcdBB7MQFPP3esbZSoyzTkZRWQl1Wg315aVEp2Wg6JkZXfCJkp2bh5Gy76tVJa4eLhRFZyVWMof7PGXIg3KI+9mIB3kFeVr/d+UZRfTFF+scHzvMwCo/HNSs7GVmWDo6uDweylu3fVv2OAK8evMabtDBxc7LGxtSYnPY8lB97n8rGqr6y8E+QYtrx7NcbGXF9jHNDA965JfiS+licxNo+itBIyzuTR6rVgo9uTD2WhKdES0MX4rOZ1mlItl79NpNVrwXi3Lv+dONdxIC+miOhfUvXJpbEE1s7L1ugFPZaSeS6PvNgifjtc/pmt+9/h8PtLpwl+0peQQcZnvmuqRmsur1y5QkhI+XqOFStWsHjxYhYvXswrr7zCJ598wsqVK1m48OZT2xEREeTk5Bg86ikamfYKbqAu03D5WBStelRc6aVQKGjVoynnD14x2ubcX5fwb+BjMINVO8SPjMRM1GWVF9DezIWDV3B2dyKkdT19WaseTVFYKYxebAKQHJ1GekImtRsa/lJrh/iRehsLru9Hl49HUVaqplWPii8itRv64VOnFucPXb5p+8LcInLS8/Bv4EtIm2AO/nTMksOtMTmGLe9ejbEx9VvWBaj24ol/m8TX8iTG5hG/JwOVqzW1Wrka3/5HBt5tXCtdAPRPOrUOnUZHpRNhVoq77m4ZrSYH03l+Ix74sPzRdHQdANq/05CgXqbdNsmYGiWXDg4OpKenA5CQkED79u0Ntnfo0IFr167dtB+VSoWLi4vBw1ynxP+7aCv9RvWg5/AuBIUF8OqKUdg5qvSn6qavG0f4nKH6+j9/thNnDyfGLnqBgBA/2vdrxdCIAfy0Yoe+jp2jCncfV/1jzrDFHN1+0qDs+qLo2IsJHNl+gskrRxParj5NHghl/JKR7Nn4l/7N4+nvzppzHxParuKeVZsW/MyTE/ry0FMd8K/vw4jZgwkMC2DbXXYls52jiuAWdQhuUX5A+tbzJrhFHWoFlk+xh78/hGlrxxq0uV7f3kmFWy0XglvUIahRgH67rZ2NQSzXvPE1v6zaaVBmbVN+fBTmFrF97R+M/uh5WnRtTEjrery2egznDl42+EBbc2YhnZ9op3/+0FMdaN6lMb71vOnUvw3zts3krx+PcmzXaYvFylRyDFvevRhjv2Afnp05kJDW9fCpU4tO/dswfd1YTu89z7Uzsf9S5G6NxNfyJMa3R6fVkbA3E/8unkZPRxckF5N1MZ/aPYyfPt435Zz+FkHWDkrcGzlx6T8JZJzLozC1hPg9GST+mYFPOzd9m9J8NSXZZfpH98+aobS10j8vzTW8JVB+fBG50YWUFWhQF2rIjS4kN9rweoPrZZoSLaW5anKjC8mPL9JvTzmSzb4pFfftdPBV4Rxor3/Y1yqfTXUKsEPlWn0SXRM1Oi3et29fPv30U1avXk3Xrl35/vvvadGihX77pk2baNCg+qvELG3vpoO4ebkw4p3BuPu6EXkymjf6zdXfYNs70BOdtmLNQ1p8BhF9P2DMwhGsOjmf9IRMtiwxvJ3KoNf6M3zWoGr3+1zweP1tHeY9t5TxS8OZv/Mt/Y1ll09cq69rbWNNUFgAKoeKqfAtS37F1s6GVxYOx9nDiahTMbze+/276lQMQMM29Vmw+23981cWDAdgx5d7WfDip3j4ueMdaHga9LO/PzRo32PogyRHpzE8ZAIAXQc/wLQ1Y6rd79SH3+X0n+fL+3vtS3RaLW9tmoKtypq/d5xm6YQ1BvUDwwJwcHXQP/f0c+eVj4bj5uNKZlIWuzbs4z9z/mtCBCxPjmHLuxdjrC5V0/rhZgyc2A87RxVpcRns23yEr+dsNltczEXia3kS49uTcSaP4vRSancznjwm/JGBnYcNXs2N39uyILGEshtuD9RyYj0uf5PA6WXRlOWrsa9lS8gQfwJ7Vvw9PLEwiqwL+ca6A8pPkXdbVnFW7u95kQY3e/9rxkUA+nzbulIZQG5UIUkHsgz6KSvUUJD479/LWaHT6W55zjYxMZHOnTsTFBRE27Zt+fTTT2nTpg2NGjXi0qVLHDp0iC1bttCv363dLPtGPZXP1LiNuHUKE283IW7d3Xb6Qwgh/r9p/Pd9dYfFu9LiVje/QLNGvwV/f39OnDhBp06d2L59OzqdjiNHjrBjxw5q167NgQMHTEoshRBCCCHE/aFGM5eWJDOXliUzl5YnM5dCCHFnycyl5Zl95lIIIYQQQojqSHIphBBCCCHMRpJLIYQQQghhNpJcCiGEEEIIs5HkUgghhBBCmI0kl0IIIYQQwmxq9B96xL1LYW2+f+skjNOVlt68khBCCIuJL3K/00MQyMylEEIIIYQwI0kuhRBCCCGE2UhyKYQQQgghzEaSSyGEEEIIYTaSXAohhBBCCLOR5FIIIYQQQpjNfXkrosfH9GLQ1P54+LoReSqG5RPXculoZJX1HV0dCH9/CJ2fbI+zhxOpMWl8OmU9R7adNGn/zu6OjFsSTsfHWqPT6ti3+TArJq2juKCk2naNOoYw8r0hhHVogFajJfJkDBF951BaXGbSOCzhman96TygLYEN/SgtKuP84Susmfkt8VeSq2zT+Ym2DJnWH//6PljbWJNwNZn/Lt7G7m8OmDQGZw8nZqwdQ71mgTh7OJGTlsvBX46z9u1NFOYVG23jE+TFsIgBtOzWGHcfVzKSsvj9m7/45sMfUZdpTBqHJckxbHn3YowX7H6bFt2aGJT9snIni8euNmkMliTxtTyJsWkuTT1AWUblvxUePQLwfrI+qT9EkX8uk7KMYqydbXBuXQufJ+ujdKg6ZTo7crfRcp/BDajVt06Nx1iWXULyt1cois6lNLUIz0cC8RvWsFK9nKMppGyOoiy9GFsfe3wHNcC5hddt93u77rvksuvgToxeOJwlY1dz4fAVBk7sx9xtbxDeaDLZabmV6lvbKPnwtzfJTsvhvcGfkJ6QiU8dL/KzC6vcx4Ldb7Pjy73sWL/X6PYZGybg6evOjN5zUNoombZmDJNXvszc55ZW2WejjiHM/fUNvpn3A8snrkWj1hDcog46ra7mQbCg5g+F8fNnu7h8LAqltZIXZg/ig19e56VWMygpNP6BkpeZzzfzfyLuUhLqUjUd+rXktVUvkZ2Wy7FdZ4y2WX/xYxa+tIrT+y5W2qbTajn4y3HWzf6enPRc/IN9GL9oBM5LRzLvhU+N9hcY6oeVlYLF478gMTKFuk1qM2n5i9g5qvg84hvTA2IBcgxb3r0aY4Ctn+9i/axN+uclhXff/VUlvpYnMTZd/bfbodNVfC6VxBcQveAELu18UGeXoM4uwfeZBqj8HSlLLybxy4uos0sIGte8yj5DFz1o8Dz/dAYJay/g2sbbaP3S9CIuT/uLpmsfNrpdp9Zi7WxDrf71yNgRa7RO4ZVs4j47h8/T9XFu4UXOoWRil56m/jvtsavtZHK/5nDfJZdPTXqUbat389u6PQAsHrOaDv1a03tkdzbO/7FS/T7h3XH2cGTig2+hUZfPYKXEpJm8/6CwANr3acW49hFcPhYFwLKJa5nzywxWTdtARlKW0XZjFo5gy9JtBmOMv5xk8jgsZeYTHxk8X/jyKjbFrSCkVV3OHrhktM0/E8Qflu/gkWcfoskDDatMLquTn13IL59XfEtMjc3g51W7GTS5X5Vt/t55hr93VuwrOTqN7xv+ymMvPXzXJZdyDFvevRpjKP9DnJWSY/K+/w0SX8uTGJvO2sXW4Hn61hhsve1xDHVDoVAQNL4iiVR5O+DzVH3iV51Dp9GiUBpfTWjjqjJ4nnsiDccwd2y97U0ao62XPX7PhgKQtS/RaJ30nXE4N/PQz4zaDaxP/rlMMnbHEzAizOR+zeG+WnNpbaOkYZtgju+uSCJ0Oh3Hd5+hcacQo2069W/L+UNXmLAsnE2JK1l1agFDZwzAykph0hgadQohLytf/2YDOL7rDDqtjrAODYy2cavlQqOOIWSn5rJo37tsSlzJwt9n0aRzqElj+Dc5upS/cfKyCm65TctujQls6MfZ/caT0Zry8HOj8xNtjc5yVsfRxYG8zHyzjMFc5Bi2vHs1xtf1GPYg36d8zqpTCwifMxSVvW219f9tEl/Lkxibj1atJftgMm4P+aNQGI+FplCNlZ11lYnlP6lzSsg7nYH7Q/7mHGolRZE5ODb2MChzaupJUeSd/3J0WzOXBQUFbNq0iatXr+Ln58fQoUPx9PQ019hqzNXLBaW1stI3oqyUHAJDjf+Sfet507J7E3Z/vZ+Zj83Dv4Evry57EaWNNRve+x6AoTMGMDTiSX0bW3tbGnUMYfyScH3Zi02nkBaXgYePG9mphqcktBotuZn5uPu6GR2DX7APAMNnPc2q6Ru4ejKans93Yf7Ot3i5+VQSrla9nvFOUigUvPLRc5z96xIx5+OrrevgYs/XkUuwUVmj1WhZOnE9x38/q9/+6pIX6DG0s/65ysGW93+chlaj1ZcNqPWSQZ8z1o+l02OtsXNQcfCX43wyZs0tj90/2JsnxvS862Yt5Ri2vHs1xgC/f3uA1Jh00hMzCW5eh1FzhxEY6s/spxfWNAwWI/G1PImx+eQdT0NTqMa9s5/R7eq8UtJ+jsajW8At95l1IBmlnRKXtrUMyq/MPKRf63n9tPz5V/botzs0dKPulJa3vB91TmmlWVhrV1vKcqpfG/9vqFFy2bhxY/bv34+HhwdxcXF06dKFrKwsGjZsSGRkJO+99x6HDh2iXr161fZTUlJCSYnhi9fqNFgplDV/BbfJykpRPtsyehVarY4rx6/h5e/BoKn99W+4X1buZO93B/VtZnw1gf2bj7B/y2F9WUZi1acAbkbxv2+OW1ft0p/iiDwZTaseTek9sjtfzLy7EqDrxi8aQZ0mtXnt4fduWrcor5ixHWZi52RHq+5NGP3hMJKvpepnG9e/t5nvF/2qrz9/x0zWvLmRS0euVtnnyun/4T9zthAQ4kv4u4MZ/eEwlk1af9OxePq7M+en6fy5+Qjb1u65+Qu9y8kxbHl3Q4wBfr1hOUj02Tgyk7L4aNfb+AX7kBSVclt930kSX8uTGBuX9Wcizs08sXFXVdqmKVITs+gUKn9HvJ+oPq8x6HNfIq4dfbGyMcxp6kxugU5TnlSqs0q49uFx6s9ur99uZXv/nEyuUXJ58eJF1Go1ABEREfj7+3Py5ElcXV3Jz8/nySefZObMmXz99dfV9jN37lxmz55tUFaPxtRXNK3h8A3lpOeiUWtw93E1KHf3cSUrJdtom8ykbNRlGrQ3XHQQezEBTz93rG2UqMs05GUVGJz2LS0qJTsth8TIym+EzJRs3LxdDMqslFa4eDiRlVzVGMrfrDEXDGf/Yi8m4B1U9VVfd9K4T4aXX5jzyBzSE27+YaPT6UiMSgUg6nQsgaH+PDOtvz65zEnLJeeGRegatYaMhEx9G2OyUnLISskh7nISeVkFfLz7Lb6e9wOZyVWfEvDwc2P+9gjOH7rC4nFf3OrL/dfIMWx592qMjbl4uPzLV0AD37sm+ZH4Wp7E2DxK04vIP59psMbyOk2RmuiFJ7GyUxI0oRkK61tL/AouZ1GaXIj7mMr5jK1XxfpLhbL8C7nKx8HE0ZfPUqpzDS+GUueUVlr/eSeYnCYfPHiQd955B1fX8oPbycmJ2bNns3///pu2jYiIICcnx+BRT9HI1KHoqcs0XD4WRasezfRlCoWCVj2acv7gFaNtzv11Cf8GPgZrLWqH+JGRmGnSLWouHLyCs7sTIa0rvuW06tEUhZVC/yb6p+ToNNITMqnd0PB0Ru0QP1JvY8G1pYz7ZDgPPN6G6X3mmrwg3MpKgY3Kxmxjuv77s7Gtuk9Pf3c++u0NrpyIZuHLqwyuFrxbyDFsefdqjI2p37IuQLUXT/zbJL6WJzE2j6z9SVi72OLcwnA5X3lieQKFtYI6r7aoNANZbZ9/JmFX1xn7IGdzD7cS+/quFJw3jFv+uUzs67tW0eLfU+Pk8vqBWVxcjJ+f4RqFgIAA0tJu/odEpVLh4uJi8DDXKfH/LtpKv1E96Dm8C0FhAby6YhR2jir9qbrp68YRPmeovv7Pn+3E2cOJsYteICDEj/b9WjE0YgA/rdihr2PnqMLdx1X/mDNsMUe3nzQou74oOvZiAke2n2DyytGEtqtPkwdCGb9kJHs2/qV/83j6u7Pm3MeEtquv38emBT/z5IS+PPRUB/zr+zBi9mACwwLY9sUfZomLuYxfNIIeQx5g3ohPKcov1r9+W7uKpG7a6tGMfHew/vkzU/vTukdTfOvWIjDUn6cm9uXhYZ35/Yb7XDq42BvEc1LX2cRdTjIou65d7xb0ev4h6jSujU+QF+37tODVpSM5+9clUmLTAQhtG8zqkx/i6e8OVCSWaXEZfB7xDa61XCr1e7eQY9jy7sUY+wX78OzMgYS0rodPnVp06t+G6evGcnrvea6dsdwtRUwh8bU8ifHt0Wl1ZO9Pwq2zn8GFOpoiNdELTqAt0RIQ3ghNsZqynBLKckoMbqt2OeIguccMz6xpitTkHE3Bo4vxda/q3FJ9XwobK0IXPah/XpZTgjrf8H7ARbF5FMXmoS3RoM4rpSg2j+KEiotQvXoGknc2g/TtMZQkFZDyQxTF0bl4PlxbXyf5u6vEf36uRv2aQ40v6Hn44YextrYmNzeXS5cu0bRpxdRvTEzMHb2gB2DvpoO4ebkw4p3BuPu6EXkymjf6zSU7tfxUqXegJzptxUUiafEZRPT9gDELR7Dq5HzSEzLZssTwdiqDXuvP8FmDqt3vc8Hj9bN4855byvil4czf+Zb+xrLLJ67V17W2sSYoLACVQ8XU9ZYlv2JrZ8MrC4fj7OFE1KkYXu/9/l11Kgag/+hHAFiwc6ZB+YKXVrFzwz4AagV6Gpx6sXNUMX7xCLwCPCgtKiXuchLzwz9j7/cV63jGLHieXs8/VO2+e9s/D5Sfqukb3p3R85/FRmVDWnwGB378m40LftHXVdnbEhjqj7V1+ZeW1j2aEtDAl4AGvnwducRov3cLOYYt716MsbpUTeuHmzFwYj/sHFWkxWWwb/MRvp6z2WxxMReJr+VJjG9P/vnym6T/84ruopg8iqLKl2hdef2gwbaGHz2gP7VdmlyIpkhtsD3ncPlnnWsHX6P7jHz3qNGbt1/nEOpG8Iw2FfVnHdH/XBydR86hFGw87QhdUH7xq0OIG4Gjm5CyOYqU/0Zi6+NA0ITmBve4VOeUUvqPfd6sX3NQ6GpwbvCf6yQ7duxI79699c+nTZtGfHw833xT88X7PZXP1LiNuHVWtnff7TTuN9rSu/Nmy0II8f+Fy747O8H1/8F/H1hx0zo1Si4tSZJLy5Lk0vIkuRRCiDtLkkvLu5Xk8v657l0IIYQQQtxxklwKIYQQQgizkeRSCCGEEEKYjSSXQgghhBDCbCS5FEIIIYQQZiPJpRBCCCGEMBtJLoUQQgghhNnU+D/0CCGEEELcjYrVktbcDWTmUgghhBBCmI0kl0IIIYQQwmwkuRRCCCGEEGYjyaUQQgghhDAbSS6FEEIIIYTZ3JeXVT0+pheDpvbHw9eNyFMxLJ+4lktHI6us7+jqQPj7Q+j8ZHucPZxIjUnj0ynrObLtpEn7d3Z3ZNyScDo+1hqdVse+zYdZMWkdxQUl1bZr1DGEke8NIaxDA7QaLZEnY4joO4fS4jKTxmEJz0ztT+cBbQls6EdpURnnD19hzcxvib+SXGWbzk+0Zci0/vjX98HaxpqEq8n8d/E2dn9zwKQxOHs4MWPtGOo1C8TZw4mctFwO/nKctW9vojCv2GgbnyAvhkUMoGW3xrj7uJKRlMXv3/zFNx/+iLpMY9I4LKkmx/CC3W/ToluTSuWHfz3Om/0/NGn/tQI9mbhiFC26NaEov5idX+5lzRvfoNVojdZv3rUxC3+fZXTbuA5vcPnvqt9/d8q9+Dlh7Hf9y8qdLB672qQxWJLE1/IkxqaJnrEXdUblvxWu3QKp9Wxjcv6MI+9wEiWxueiKNdRb3AOlg80t95+1LYqMzVdwfTiIWkMamTRGbZmGtK/OUxKbS2lSAY7Na+E3rpVBHXV2CenfXaQkOpeytEJce9x8f7kHEkhdd9botroLu2HtojJpvP903yWXXQd3YvTC4SwZu5oLh68wcGI/5m57g/BGk8lOy61U39pGyYe/vUl2Wg7vDf6E9IRMfOp4kZ9dWOU+Fux+mx1f7mXH+r1Gt8/YMAFPX3dm9J6D0kbJtDVjmLzyZeY+t7TKPht1DGHur2/wzbwfWD5xLRq1huAWddBpdTUPggU1fyiMnz/bxeVjUSitlbwwexAf/PI6L7WaQUmh8Q+UvMx8vpn/E3GXklCXqunQryWvrXqJ7LRcju06Y7TN+osfs/ClVZzed7HSNp1Wy8FfjrNu9vfkpOfiH+zD+EUjcF46knkvfGq0v8BQP6ysFCwe/wWJkSnUbVKbSctfxM5RxecR35geEAuo6TE8++mFWNtWvJVdPJ1ZeWI+f35/qMp9fBW5lI/CP+X03vOVtllZKZjz8wwyU7KZ9OBbePi5M33dODRlGr5481uj/Z3/6xKD/V82KHvh3Wdo1aPpXZlY3qufEwBbP9/F+lmb9M9LCktv8VX/eyS+licxNl3gzE4Gf1tLE/JJ/ORvHNv6AqAr1eDY1AvHpl5kbL5So76Lr+WQszce29pON6179aXfqDO3CzZe9pU3anUobJW49gii4HiK0fY6tRalsy3ujwaTvSvmlsbn1M4Xh6ZeBmWpa8+gLdOaLbGE+zC5fGrSo2xbvZvf1u0BYPGY1XTo15reI7uzcf6Pler3Ce+Os4cjEx98C426fAYrJSbN5P0HhQXQvk8rxrWP4PKxKACWTVzLnF9msGraBjKSsoy2G7NwBFuWbjMYY/zlJJPHYSkzn/jI4PnCl1exKW4FIa3qcvbAJaNt/pkg/rB8B488+xBNHmhYZXJZnfzsQn75fLf+eWpsBj+v2s2gyf2qbPP3zjP8vbNiX8nRaXzf8Fcee+nhuy65rOkxnJdVYPC82zOdKS4s4c/vqk4uq9OmVwuCGtdmeq/3yU7NIfJUDOtnbWTU3Gf5cvZ3Rmd61WUaslJy9M+V1ko6Pd6WH5dtN2kMlnavfk5A+R/iG2N9N5L4Wp7E2HRKZ1uD51nborCpZY99Q3cA3B6pC0Dhpcwa9astVpOy+jTew5uQufX2vlRbqazxfq4xAMVXs9EWqSvVsfGy189U5h5IuLV+bZVY2Sr1zzV5pRRezMR7RNPbGm+l/ZijE53u7phds7ZR0rBNMMd3VyQROp2O47vP0LhTiNE2nfq35fyhK0xYFs6mxJWsOrWAoTMGYGWlMGkMjTqFkJeVr3+zARzfdQadVkdYhwZG27jVcqFRxxCyU3NZtO9dNiWuZOHvs2jSOdSkMfybHF3Kv3H9M8GpTstujQls6MfZ/caT0Zry8HOj8xNtjc5yVsfRxYG8zHyzjMFcTDmG/6lveHf2bPyL4ipmkm+mcccQos/Ekp1a8eH/92+ncHR1oE6TwFvqo9PjbXDxdNb/4bub3KufE9f1GPYg36d8zqpTCwifMxSVvW219f9tEl/Lkxibj06tJe9wEs6da6NQmBaL69K+voBD81o4NPY00+gsL/dgIla2Spza+Ji1X7PMXKpUKk6dOkWjRqatLTAXVy8XlNbKSt+IslJyCAz1N9rGt543Lbs3YffX+5n52Dz8G/jy6rIXUdpYs+G97wEYOmMAQyOe1LextbelUccQxi8J15e92HQKaXEZePi4kZ1qeEpCq9GSm5mPu6+b0TH4BZf/UofPeppV0zdw9WQ0PZ/vwvydb/Fy86kkXK16PeOdpFAoeOWj5zj71yVizsdXW9fBxZ6vI5dgo7JGq9GydOJ6jv9ese7j1SUv0GNoZ/1zlYMt7/84zWCN34BaLxn0OWP9WDo91ho7BxUHfznOJ2PW3PLY/YO9eWJMz7tu1tKUY/hGoe3qU69ZEAtf+sygfOKKUTz87EP65yoHWz7YGmEQ38ddRwDg7utGVmrl/QPla7tu4XX0De/BsR2nSE+o2Tf/f8O9+jkB8Pu3B0iNSSc9MZPg5nUYNXcYgaH+zH56YU3DYDESX8uTGJtP/olUtIVqXDrf/PO1OnlHytdo1p7Zsco6iYuPUXTFcEY3dtYB+F9Oa+NhR9C7D97WOGoqd388Th38DGYzzaFGyeWUKVOMlms0GubNm4enZ3m2/vHHH1fbT0lJCSUlhrMqWp0GK4V5X9ytsLJSlM8Yjl6FVqvjyvFrePl7MGhqf/0b7peVO9n73UF9mxlfTWD/5iPs33JYX5aRWPUpgJtR/O+b49ZVu/QzPZEno2nVoym9R3bni5l3VwJ03fhFI6jTpDavPfzeTesW5RUztsNM7JzsaNW9CaM/HEbytVT9bOP69zbz/aJf9fXn75jJmjc3cunI1Sr7XDn9P/xnzhYCQnwJf3cwoz8cxrJJ6286Fk9/d+b8NJ0/Nx9h29o9N3+h95A+4T2IOh1TaVH/+lmb+G7hz/rnC36fxeqIr7l4uGbriW6FV4AHbXq14P0hn5i97zvlbvicAPj1huUg0WfjyEzK4qNdb+MX7ENSlPF1WfcCia/lSYyNy90fj0NTL6zd7EzuoyyziPRvL+I/pS1WNlXnMd7Dm6C9YVlR7Mz9+L/aGqV7+VpHhfLfvYFPUWQ2ZUkF+LzYzOx91yi5XLRoES1atMDNzc2gXKfTceHCBRwdHW9pWnnu3LnMnj3boKwejamvuL1z/jnpuWjUGtx9XA3K3X1cyUrJNtomMykbdZkG7Q2Le2MvJuDp5461jRJ1mYa8rAKD076lRaVkp+WQGFn5jZCZko2bt4tBmZXSChcPJ7KSqxpD+Zs15oLh7F/sxQS8g7yMNbnjxn0yvPzCnEfmkJ5w8w8bnU5HYlQqAFGnYwkM9eeZaf31yWVOWi45NyxC16g1ZCRk6tsYk5WSQ1ZKDnGXk8jLKuDj3W/x9bwfyEyuei2Ph58b87dHcP7QFRaP++JWX+6/xpRj+Do7BxXdn3nAYKH8ddlpuQaL/DVqDekJmUaP4azkbMLaGZ7Wuj6ezCqO4Rv1fqEbuRl5HPzp2E3r3gn36ueEMRcPl3/5Cmjge9ckPxJfy5MYm0dZRhFFFzLwHdvq5pWrURKTiyavlLj3KhJztDqKr2SR80cc9T/ticJKgbV75QTW2tPe+AU9/4LcffHYBjpjV8f15pVrqEZp8gcffEBOTg5vvfUWf/zxh/6hVCpZt24df/zxB7///vtN+4mIiCAnJ8fgUU9x+6fU1WUaLh+LolWPiixcoVDQqkdTzh80PkNz7q9L+DfwMUiKa4f4kZGYadItai4cvIKzuxMhrevpy1r1aIrCSqF/E/1TcnQa6QmZ1G5oOC1fO8SP1NtYcG0p4z4ZzgOPt2F6n7kmLwi3slJgo7r1WzvczPXfn41t1X16+rvz0W9vcOVENAtfXnXXrBW+kSnH8HVdBnXERmXNrv/su60xnD90hbrNgnCrVfGHo3XP5hTkFBJ7k+UPUJ5c7vrqT/1FA3ebe/Vzwpj6LesCVHvxxL9N4mt5EmPzyD2QgNLFFsdmtzeJ49DIk8B3HiDw7U76h6qOC84d/Ah8u5P+7OTdRFusJv/vZFweDLBI/zVKLmfMmMHGjRsZM2YMU6dOpazMtPsvqlQqXFxcDB7mOiX+30Vb6TeqBz2HdyEoLIBXV4zCzlGlP908fd04wucM1df/+bOdOHs4MXbRCwSE+NG+XyuGRgzgpxU79HXsHFW4+7jqH3OGLebo9pMGZdcXRcdeTODI9hNMXjma0Hb1afJAKOOXjGTPxr/0bx5Pf3fWnPuY0Hb19fvYtOBnnpzQl4ee6oB/fR9GzB5MYFgA2774wyxxMZfxi0bQY8gDzBvxKUX5xfrXb2tXkdRNWz2ake8O1j9/Zmp/Wvdoim/dWgSG+vPUxL48PKwzv99wn0sHF3uDeE7qOpu4y0kGZde1692CXs8/RJ3GtfEJ8qJ9nxa8unQkZ/+6REpsOgChbYNZffJDPP3Lr/67nlimxWXwecQ3uNZyqdTv3aKmx/B1fUZ258CPfxu9SOmf8X31gTeJu5hgNL7Hdpwi9nw8r385nuDmdWjbqwUvvPsMP634jbLS8isWQ9vVZ825j/Xxva5Vj6b4Bfuwbc3Nv2TeSffi54RfsA/PzhxISOt6+NSpRaf+bZi+biyn957n2pnYfylyt0bia3kS49uj0+rIO5CAc6eASqej1TkllMTmUpZafpum0vh8SmJz0RRU3DIpYeFRsn8vv/2PlZ01qgBng4dCpcTK0QZVgLO+jaagFHVOif5Rd0E3FDZW+ueaPMNbMpUm/m+/hWVoi8ooic2lJNZwnev1Ml2JBk1eeZ3SxIq/AfnHU4h5a3+l159/NBm0Opw73t5a06rU+IKedu3acezYMcaNG0fbtm35z3/+c9tXWJnT3k0HcfNyYcQ7g3H3dSPyZDRv9Jurv/LVO9ATnbbiIoa0+Awi+n7AmIUjWHVyPukJmWxZYnhLoEGv9Wf4rEHV7ve54PH6Wbx5zy1l/NJw5u98S39j2eUT1+rrWttYExQWgMqh4p5SW5b8iq2dDa8sHI6zhxNRp2J4vff7d9WpGID+ox8BYMHOmQblC15axc4N5TNmtQI9DU692DmqGL94BF4BHpQWlRJ3OYn54Z+x9/uKdTxjFjxPr+cfojq97Z8Hyk/V9A3vzuj5z2KjsiEtPoMDP/7NxgW/6Ouq7G0JDPXH2rr8S0vrHk0JaOBLQANfvo5cYrTfu0VNj2GA2g39aPZQI17v/b7RPscteoFeI7pVu9+eymcA0Gp1vPn4h0xcMYrFB96juKCEnV/uZd0Np9tVDiqCwgKwtjH8COkT3p1zBy4Rdymxpi/7X3Uvfk6oS9W0frgZAyf2w85RRVpcBvs2H+HrOZvNFhdzkfhansT49hRdyECdWYxL58ozdzl748j6uWLdesJHRwDwfqGpvn5ZWiGa/JpNsCWtOEnx5apnaK097ag7r6v+eeKSYwY3e79+2r3B570rlUH56fn8I0kG/WiL1JQlV76bS+6BBBxb+dTo5vA1odDdxrnBb7/9lkmTJpGWlsaZM2do3LixyQO5/odNWIaV7d13O437jbb07rzZshBC/H9h+4d5b6kjKtvaZclN69zWrYiGDBnCgw8+yLFjx6hTp87tdCWEEEIIIe4Dt32fy9q1a1O7dm1zjEUIIYQQQtzj/t2bKgkhhBBCiPuaJJdCCCGEEMJsJLkUQgghhBBmI8mlEEIIIYQwG0kuhRBCCCGE2UhyKYQQQgghzOa2b0Uk7g1Wnu43ryRuiy4l9U4P4b6m0959/wteCHF3uXik7p0ewv2vy82ryMylEEIIIYQwG0kuhRBCCCGE2UhyKYQQQgghzEaSSyGEEEIIYTaSXAohhBBCCLOR5FIIIYQQQpjNfXkrosfH9GLQ1P54+LoReSqG5RPXculoZJX1HV0dCH9/CJ2fbI+zhxOpMWl8OmU9R7adNGn/zu6OjFsSTsfHWqPT6ti3+TArJq2juKCk2naNOoYw8r0hhHVogFajJfJkDBF951BaXGbSOCxh3cFZ+AR6Vir/ed0+Vrz5XaXyPsM68fBT7akT6gfA1TNxrPvwZy6fjDV5DNvil1Qqmzd2HXt/Ol5lmyETetHu4SYENwlAXapmUJMZJu//39B/TC8GTemPh68rUadjWT6p6mNYaa1kyOtP0PP5rngFuBN3OYk1EV/z945TJu/f2d2RcYtH0uHR8mN4/5YjrJhc/THsF+zDyx8+S5POYdiorPn7t1Msn7SO7NQck8dhSffi58SC3W/TolsTg7JfVu5k8djVJo3BkiS+licxNo/R7dsxvctDrD12nPf/2AOAl4MDM7p24cG6dXC0tSUqM5MVh47w25Ur1fb1XMsWvNSuLbUcHbmQlsbs3X9wOjnZpHF1CKzNyDataeHri5NKRXRWFp8f/ZufLlzU13mqSWPm9+1j0K5Erabxosp/J29kq1QyoVNHnmjcCC8HB9IKClh68BDfnz1n0lj/6b5LLrsO7sTohcNZMnY1Fw5fYeDEfszd9gbhjSaTnZZbqb61jZIPf3uT7LQc3hv8CekJmfjU8SI/u7DKfSzY/TY7vtzLjvV7jW6fsWECnr7uzOg9B6WNkmlrxjB55cvMfW5plX026hjC3F/f4Jt5P7B84lo0ag3BLercdff2m/joQqyUCv3zOqF+zP12PPu2njBav3mnEPb8eIwLf1+jtKSMQWMfYc5/xvLKw3PJSDaedKw7OIuFU/7DmYNXqxzHwskbOLbngv55fm5RteO2tlWy75cTXDh2jd5DOlZb907rOqgToz96niXjVnPxyFUGvtqPD7ZG8GKTKUaP4RfefYaHhz3IJ6+sIu5SIm17tWDW968xqcvbRJ6MNrqPj3aVH8M7v6ziGP5yAh5+bkT0/QCljZKpn7/CpE9fZt5w48ewnYOKub++QdTpGKb3eq98XO8M5t0fpjGx81vodHfXcXyvfk4AbP18F+tnbdI/LyksvcVX/e+R+FqexNg8mvn6MLRFcy6kphmUL+jXBxeVHS9v+ZGsoiIebxTG0v6PMmDDfzj/j7rXPRrakDe6deWtXbs5lZTEyNatWff0QHp+sZaMQuN/oyKnTqHLqtUk5Fb+nbX29+dSWjqrjhwlvaCQHvWDWdC3D3klJfwRdU1fL6+khEfWrNU/v5VP2yX9H8XLwZEZ23cQk52Nt6MjVgrFzRveovvutPhTkx5l2+rd/LZuD7EXElg8ZjUlhaX0HtndaP0+4d1x9nBk1pMLOPfXJVJi0jj95wWiTseYtP+gsADa92nFxy+v5OKRq5w7cIllE9fS7ZkH8PSr+kbmYxaOYMvSbWyc/yMx5+OJv5zEn98doqxUbdI4LCUnM5+stDz9o8MjTUmMTqsyEZw/4Uu2frmfqPMJxEemsnjaN1hZWdGyc8PbGkdBbpHBOMpKqo/ThoXb+GH1HqIvJt7Wfv8NT016lG1rfmfH+r3lx/DY/x3DL3QzWv+RZx/kmw9/4Oj2kyRfS+WXlTs5su0ET09+1KT9B4b5065PSz4evUp/DC+ftI5uz3TCo4pjuMkDofjUrcWCFz8l+mwc0WfjmB++goZtgmnZvYnRNnfSvfo5AeV/iLNScvSPwrzqv1jdCRJfy5MY3z4HGxs+6dePN37bSU5JscG21v7+fHniBKeTk4nLyWH5ocPklpTQ1Menyv7C27Zh45mz/PfsOa5mZPLmzl0Ulal5umlTk8b36eEjfHLgL44nJhGbk8O64yf4Mzqa3iEhBvV0Oh3phYX6R0Zh1V8YALrUrUuH2rV5cfMW/oqNJSE3lxNJSRxLNN/fx/squbS2UdKwTTDHd5/Rl+l0Oo7vPkPjTiFG23Tq35bzh64wYVk4mxJXsurUAobOGICVlWkZfKNOIeRl5XP5WJS+7PiuM+i0OsI6NDDaxq2WC406hpCdmsuife+yKXElC3+fRZPOoSaN4d9ibaOk+8C27Pj20C23UdnborSxIq+ab8u3YuycQXx7+gMW/fIavZ65u2cia8LaRklI63qc+McxfOL3MzTqaDwht1HZUPaPpROlxaU0eSDMpDE07tiQvKx8rtx4DO8uP4YbtTd+DNuorEGno6ykYhxlxWXotDqadjZtHJZyr35OXNdj2IN8n/I5q04tIHzOUFT2tiaNwVIkvpYnMTaP2Y/04I+oKP6KrbxM63hiIo+GhuJqZ4cCeCw0FJW1NYfj4o32ZWNlRVMfH/6KqUjWdcBfsTG08vcz25idbVXkFBsmwg62tvz58ij2v/wSnw14nBDPykvXbvRwg2DOpKTwcru2HBj9MrvCRxLRtQsqa/OdzL6vTou7ermgtFaSlWJ4ujUrJYfAUH+jbXzredOyexN2f72fmY/Nw7+BL68uexGljTUb3vsegKEzBjA04kl9G1t7Wxp1DGH8knB92YtNp5AWl4GHjxvZqYbT21qNltzMfNx93YyOwS+4/JvQ8FlPs2r6Bq6ejKbn812Yv/MtXm4+lYSrpq3XsLROvZvj5GLPzu8O33Kb8DceJzM5lxP7L+nLxs8dTI+B7fTPVfY2vPfVK2g1FZP7A0On6X/+8qOtnDpwmZKiMlp3DWPcnEHYOdry0xd/3uYruvNcrh/DqcaO4QCjbf7ecZqBE/txet8FkiJTaNWjKZ0HtMdKWfHdccjrAxg6Y4D+ua29LY06NGD84pH6slHNXyMtLgP3Ko7hvGqO4QuHr1BcUMKLc4ex9s1vUSgUhH8wFKW1Eg8/423ulHv1cwLg928PkBqTTnpiJsHN6zBq7jACQ/2Z/fTCmobBYiS+licxvn2PhYbSxNuHARv+Y3T7hJ+3suSxRzk+fixlGg3FajVjfviJmOxso/Xd7e2xtrIivcBw4iS9oJBgDw/98y+eepK2AYaf5dtHjtAvHUrIzaXvui+N7qNfaEOa+frw5s5d+rKozCxmbP+Ni2npOKtUjGrXhu+GDaHP2vUk5+cb7SfI1Y22AQGUqDWM+fEnPOztmf1ID9zs7Xh9+w6jbWqqxsnlhQsXOHToEJ06dSIsLIyLFy+yePFiSkpKeO655+jRo8dN+ygpKaGkxHDBr1anwUqhrOlwbpuVlaJ8xnD0KrRaHVeOX8PL34NBU/vr33C/rNzJ3u8O6tvM+GoC+zcfYf+WiqQqIzHL5DEo/vfNceuqXfy2bg8AkSejadWjKb1HdueLmd+Y3Lcl9R7Skb//uEBmSuW1IsYMGvcIXZ9ozfRBSw1OY3+14Ff+u/J3/fP5373KFx/8xMUT0Ub7+Wbxb/qfI8/FY+dgy9OvPHxfJJem+HTKOiZ/9jJrzn4MOh2JkSnsWL+H3i9UnB7bumonf35/wzH85QT2bT7MgR+O6Mtu5xjOSc/j/SGLmLDsRQaM74NOq+OPjX9x5XgU2rts3bAp7obPCYBfP9+t/zn6bByZSVl8tOtt/IJ9SIpKua2+7ySJr+VJjCv4OTvxVo9uDP/uv5RqNEbrTOn8AC52Kp7f9B2ZRUX0bNCApf0f5ZlvN3E5Pd3kfUf8thO7G2YIfx8Vzov/3ULK/xLBMq3x8XQMDOTDPr2ZuWMnVzIy9OUnkpI4kZSkf348MZEdI19gaIvmfHLgL6N9KRSg08Hkrb+SX1q+3nXOnr0sf7w/b+/6nRL17S/Hq1FyuX37dp544gmcnJwoLCxky5YtDB8+nBYtWqDVaunVqxc7duy4aYI5d+5cZs+ebVBWj8bUV5i2LuG6nPRcNGoN7j6uBuXuPq5kpWQbbZOZlI26TGPwBzD2YgKefu5Y2yhRl2nIyyogL6tAv720qJTstBwSIyu/ETJTsnHzdjEos1Ja4eLhRFZyVWMof7PGXDCcbo+9mIB3kFeVr/dO8g5wp+VDobz/0ppbqv/U6B4MHvsIbwxdTvQFw3UdORn55GRUfMPSqDWkJ2eTFH1rb+CLx6MZNqkPNrbWd90a1ZrKvX4Me1c+hjOrOH5y0vN45+mF2KhscPF0IiMxixc/GGbwQf3PY7ikqJTstFyjx3BWFcewczXHMMCxXad5IWwiLp7OaNQaCnIK+TbuM5KjjH/A3Sn36ueEMRcPl691Dmjge9ckPxJfy5MY356mPj54OTry0/Dn9GXWVla0r12b51u1pOeatQxv3Yo+a9frE7mLaem0qx3A8y1b8Nau3ZX6zCoqQq3V4uXoYFDu5Vh+JfZ1KUZmExNyc41e0HNd+9q1WfXkE8z5Yw9bzl+osh6AWqvlXGoqddzcqqyTVlBASn6+PrEEiMzIxEqhwM/JiegqZmdrokZrLt99912mTZtGRkYGa9euZdiwYbz00kvs3LmT3bt3M23aNObNm3fTfiIiIsjJyTF41FM0MvlFXKcu03D5WBStejTTlykUClr1aMr5g8ZvH3Dur0v4N/BBccNVUrVD/MhIzERdZvwbRHUuHLyCs7sTIa3r6cta9WiKwkqhfxP9U3J0GukJmdRuaHg6o3aIH6kxxq9Ku9N6PtORnPQ8juy++W0Lnh7zMEMn9uat5z/jyuk4s4+lfpPa5GUX3POJJZQfw1eOX6Nlj4ovWgqFgpbdm3Lh0OVq25aVlJGRmIXSWsmDT7bn4M/HTBrD+UOXKx/D3cuP4QtHqr6C/7rcjDwKcgpp2a0Jbt4uHPzFtHFYyr36OWFM/ZZ1AchIur3ZJHOS+FqexPj2/BUTS9916+n/5Vf6x+nkZH48f4H+X36FnY0NANp/3OVCo9VVeUV1mVbL2ZQUHggK0pcpgE5BQZxITDLa5lZ0CKzN6oEDmP/nPr49feam9a0UCkK9vEi9IaH9p2MJiXg7OeLwv9cJUM/dHY1WS1IVp9JrqkbJ5blz53jhhRcAGDx4MHl5eTz99NP67c8++yynT5++aT8qlQoXFxeDh7lOif930Vb6jepBz+FdCAoL4NUVo7BzVOlPN09fN47wOUP19X/+bCfOHk6MXfQCASF+tO/XiqERA/hpRcW6AztHFe4+rvrHnGGLObr9pEHZ9UXRsRcTOLL9BJNXjia0XX2aPBDK+CUj2bPxL/2bx9PfnTXnPia0XX39PjYt+JknJ/Tloaf+r727DosqewM4/qURpLtECUkDCxUTWxfXwk507e5aXdd21TXWYm1dXWPtrrU7sANREUGRFkV6fn+wDo7OIOAg6u98nofn8d45994zr3fOvPfcc894Y+1oQZdJrbFztWH/yn+VEhdlUlFRoV5rb45svUhmRqbMa8PmdaTraD/psn/funQe3oTfh28gMiwGIzM9jMz00NbJHsCto6ctXW9kpseQpnN59jBSZt073nU9adCuCvYuVlgVN6VJp2q0GVCPXauyb4mXLFuMwOPjMLHMvqo3szbCwd0GcxtjVNVUcXC3wcHdRqYeX4t/5u2lcXdf6nWqgZ2rNQMXdc86h/+bDmTEqr4ETGkrLe9ayQmfZhWxLGGOp48r0/aOQVVVhc2zd0nLfHgOT+swn8sH5Z/DYfciuHQgiMFLe+JS0RH3qiXpN78bxzedk/aym1gbseLmHJlzuH6Xmrh6O2HlYEGd9tUY//dgts3fx7MH+W9YC8q32E5YOVjQYVwLnMuVwMLejCp+5Rm5ui83Ttzh8c38zxtbEER8C56Icf69SUvjQXSMzF9SWhrxyck8iI7hUWwsT+LimFKvLqUtLSlmYED3CuWpVtyeww+z5xFd59+KTl5lpcsrL1+hTelStPBwx9HYmMn16qKjoSEzd6SBtjamOjrSP+/FS0lJT5cuGxcpIi1b2c6O5S2as+bqNQ48CJaWMdDWlpbpX6Uy1eztsTMwwMPcnLmNG2Gjr8/mm9mJ6PDq1Zj93lyYu+7eI/5tMjMbNsDJxJiKtjaMrlmDrbduK+WWOORjzOW7qx5VVVW0tbUxMMj+AtfT0yMhoXAnTD6x+RyGpvp0+aU1RpaGhAQ9YWzj6dKJnM3tTJBkZidEUc9iGNNoGn3mdCEwaBbR4bFsX5A1JdA7/sP86DzRP8fjdnToT+R/vYwzOi6k/8IAZh3+WTqx7KJB2XNQqWuoU8zVBi0dLem67Qv2oamtQe85ndEzLsqj66GMajDlq7oV845XdRcsbI3lPiVubmMkMzdnk04+aGipMz6wu0y59XP389fc/QD0ntSSeq29czxmI9uBAKSnZ+DXpTo9JzZHRUWFiCdRBE7azoEN2eOEtIpoYudkgZp69gVLp+GNZY6x6NAoAEb6L8hxPs3CcGLLOQzM9Ok80R8jS0MeXQ9l3A8z3juHTWVirKGlQddJbbByMOft62QuHghiZtdFvEnIHljuP9SPThNafXSs93VyGpB9DndeSL/5Acw8OD7rHN6eNTnyO+oa6ti52qBVJPscti1pTcCUdugZFyXySRQbZ2znn3n7lBESpfsW24n01HTK1SlFi0GN0dbVIioshlPbLrJh6jalxUVZRHwLnohxwUnPzKT7P9sZUaM6fzb/ER1NTULj4hmx/wDHH2fPL1nM0ACj95LBvfcfYKyjw2Cfqpjq6HA3KopuW7fJTA20+Ec/KtvZKTz2s4QEav6ZNdyshYc7Ohoa9K3sTd/K2d9f58PC6LAp60dLDLS0mNagHqY6OrxKSeFWZCT+GzfyMCZWWt5cVxcr/exOmqS0NDpv3cpEX192dOxA/Ntk9j64z9zTyhvCpCLJw+zGZcqUYebMmTRsmJUB37p1C1dXV9T/G5x66tQpunTpwqNHj3LajVz11NrkeRsh99StFM/NJShHRuTLwq7Cd+1r+0EBQRC+Po9m5txRIXy+kOFDP1kmTz2Xffr0IeO9J6s8P5gYdP/+/bl6WlwQBEEQBEH4PuUpuezdu3eOr0+bNu2zKiMIgiAIgiB8276rX+gRBEEQBEEQCpdILgVBEARBEASlEcmlIAiCIAiCoDQiuRQEQRAEQRCURiSXgiAIgiAIgtKI5FIQBEEQBEFQmjz/Qo/wbUovZl7YVfjuqcYX7q9TffdSUgq7Bt89MVG98K0ruSKqsKvw/Rv+6SKi51IQBEEQBEFQGpFcCoIgCIIgCEojkktBEARBEARBaURyKQiCIAiCICiNSC4FQRAEQRAEpRHJpSAIgiAIgqA03+VURE371Md/uB/GloaEXA9l0aBV3L8UorC8roEOAVPa4tO8EnrGRXkZGsWSoWu4uD8oX8fXM9Kl34IAKv9QDkmmhFPbLrB48GqS3+Q8lYpbZWe6TW6Lq7cTmRmZhASFMqbRVFKT0/JVj7wqVbYY/u0rU9LFChMzPSaO3szZkw9kyhSzN6FH3zqU9iqGqpoqT59EM2nsVqIiX8ndZ/3GpRkxvqnMutSUdJrUnpHvevYdUh+PUnYUdzAj7Ek0vbsul3ldQ1ONwSMa4+xqRTF7U86fDeaX0Vs+ud91//TH0spQZt3yJcfYtO5svuuaVz/08KVJD18sipkCEHo3nL9m7OTy4RsKt6nevCJdfm6BRTFTwkMiWfHzZi4dUlw+Jw6edrQe9gOeVZzRN9Ej8mk0e1ccY8fiw5/ctlKDMnQY/SMlPO1ITU7j5ul7TGq3IF/1+BL8+tTHf6gfxpYGPLrxlEWDc24nmg9sxA8962FezJRX0Ymc2naBFeM2kpaSv8+nnpEu/eZ3w7tJVjtxevtFFg9R3E5Y2Jux7uFCua9Nbvs7p/65kK96FJRvsR2efXQCZWp5yKzbs+ww8/suV7BF4RIxVh7PCsVpFVADJw8bTMz1+bX/Os4dvSNTptOAujT0r4iuXhHuXAvlj0k7iAiNUbjPIjqadB5Unyp13TE0LkrI3QiWTdvDg1vPPquudZuVo0XXatgUNyXpdQqnDt5k8eRdcssWNShCp/51KefjjJmVIQmxbzh39A5rFxwi6XXBTe/23SWXNVtXodeczizou5y7F4JpMagx0/ePJcBtCPFRHydA6hpqzDw4nvioBCa3/p3o8Fgs7E15HZ+k8Bizj07g0NoTHFpzQu7ro9cPwMTSiNENpqKmocaIFX0Ysqwn0zvK/2KArMRy+r6xbJyxg0WDVpGRnoFDGfsvOu+ctrYGjx6+5OCe6/wyw/+j161sjPh9aRf27w5izYoTJL1JpXgJU9JS03Pc75vXyXRru0S6LPnEW1r3T39+m7KbG9dCFZY5uCcIVw8bHBw/nr9TTVWVlJR0tm+5RPVarjkf7AOrA4+zb9c16fLbpNQ8bf+5osJjWTlhM+EhkaioQL0O1fhl0yD6+Uwg9G74R+XdvZ0Ys6oPKydu4cKBIGq3rsLEvwfRr9oEQu98XB5gze3ZzOm9nBun7n30mpNXceKjXjGzxzKinsXi7u3MoIVdycyQsGvZEYX1rvZjBQYv7MaqSVsJOnEHNXU1irvb5j8QBaymfxV6/daJBf2Wc+/iQ1oMbMy0vWPo7jFUbjtRu60P3ae2Y85Py7hz7gG2zlYMX9EbiUTCshHr5B7jtyNZ7cThtQraibUDMLYyZEyjaahpqDH8z94MXtKTGZ3ltxNRYdG0se0ls65xjzr4D/Pj0oGgvAWggH2r7TDA3j+PsGbiZulyyhduA3JLxFi5tIto8uj+cw5tu8zPCzt99Lp/jxo07ViVOWO28OJZHJ0H1mPKnwH0+uF3hd+Bg6a0pLizBbNHbSbmZSK+fmWZtrI7vX74nZiX8jtkVh8ZyZwxW7h56bHc15t3qUaLbtVY8dt+7t8IQ6uIJhY2Rgrfl4m5Psbm+iyftY+nIS8xtzak/y/NMTHXY+rgDbmITP58d8lly8FN2L/8KAdXHwdgfp/leDcuR4Nutdk0a+dH5RsG1EbPWJdB1X4mIz0DgMjQ/E/CWszVhkoNvehXaQwPrjwC4I9Bq5i6ZzSBI9YT8zxO7nZ95nRh+8L9MnV89uB5vuuRH5fOh3DpvOKr3m69anHxXAjLFx+TrnseLv/9vE8igbjYN0qpI8Di3w8BYGCkKze5TE5OY8Hs/QB4lrJFV0871/tOSkpVal3z6sIHPQirJ/3DD919ca3oKDe5bNa3PpcP32Tr/Kz3u3byNsrV9uDHXnVZMGhNno9/aN0pmeUXT6Jw83bEp2l5hcmlqpoqvWd14M/xmzi49qR0/dN7EXk+/pfScnAT9q84Jv3SnN93OZUaedGgay02/fZxD4B7lZLcPvuAf/8+A2S1Ef9uOotrJad8Hd/O1ZqKDcvSr/JYgv9rJxYNXs2U3aMIHLWeWDntRGamhLhI2Yn6fZpV5OTW85+8K/KlfavtMGQlOh/G+WskYqxcl0894PKpBwpfb9bZh7+X/sv5Y3cBmD16MxtPj6NqXXdO7Pv4TpGmljrV6nkwqf86bl1+AsBfi47iXduNJu28WTv/03eDPlRUX5vOg+oxqe9agt77rn7y4IXCbUKDI5k66C/p8vOwWNbMO8jIWW1QVVMlMyMzz/XIje9qzKW6hholyztw9ehN6TqJRMLVozdxr+Isd5sqfhW4cz6YAX8EsDliGYHXZ9NudDNUVVXyVQe3Ks4kxr2WftgArh65iSRTgqu3/C8iQzN93Co7E//yFfNO/crmiGXMOTYRDx+XfNWhIKiogHcVJ549jWH67+3YvHcIC/7sRtUaJT+5bZEimqzfNoC/tg9k0kx/7EuYfoEa50/bTlX5Z/9QlqzugX/7yqiq5e88UAZVVRVqtvJGS1eLuxcfyi3jVsmJa//elll35egt3PKZ9Mijq69DYpzihNu5bHHMbIyRZEpYdOZXNjycz5Rtw7B3t1FaHZRJXUMN53IluPZBO3Ht2E3cKss/n++ce4BzuRK4VHQEwLKEOZUaeuX7dqJ75ZIkxr2WJpYAV49mtRO5/b9zLlcCp7IlOLDq33zVoaB8q+3wO77tq7E18k8Cr88mYGo7tIpo5qsOBUnE+MuytDXC2Eyfa+ey2+Gk1yncvxGGa5licrdRU1NFTV2NtBTZXs3U5DQ8yhXPVz28qjqjqqqCiYU+y/YMYd2/oxkztx2mlgZ52o+unjZJr5MLLLGEfPRcvn37litXrmBsbIy7u7vMa8nJyWzevJnOnTsrrYJ5YWCqj5q62kdXRHGRCdi5WMvdxrKEOWVre3B0w2nG/TADaydLBv7RHTUNddZP3gpAu9HNaDemuXQbzSKauFV2pv+CAOm67p5DiQqLwdjCkPgPurszMzJ5FfsaI0tDuXWwcrAAoPPEVgSOXM/DoCfU61SDWYd/pmfp4YQ/VHxV8qUYGumio6tFm05VWR14nOWLj1GhsiMTp/kzov86bgQ9lbtd2NMYZk/bzeOQl+jqatGqfWXmL+tKjw7LiI5KBGDQiEbUaVBKuo2WtgbT5rYlMyP7/nnTurMK9g0CO7ZcIvj+CxJfvcWjlC0BvWtjbFqUZQsU3w4uCMU9bJl39Gc0tTV4+zqZX9stUNgLaGRhQNwHt8DiXiZgZJHd2Ayc3wXfNlWly1o6mkzZNkymYWlmKXu79R13bydqtqzEzy1/V1hfyxJmAHQc24zAMRt5ERpNq4EN+W3/GLqXHZVjYloY9N+1Ey/ltRPyE+J//z6Dgakec49PQkUF1DXU2b3sMH/P3CEt03ZUM9qNbiZd1iyiiZu3E/3nd5Ou61F6GFFhMRgpaCcSc2gnPtSwW21C7zzjzjnFvS2F4VtthwGO/X2Gl6HRREfE4lDanh7T22PnYs2kVnPyGoYCJWL8ZRmZ6gEQF/NaZn1c9GuMzPTkbvM2KZU710Jp18eXpyEviY95Tc0mZXAtW4znT7PHafaf2Axfv7LSZa0iGkwO7CbTPreo8AsAlrbGqKio0KZnLZZO20PS62Q6D6rPtBUB9G22gPS0jE++F31DHdr18WX/5ku5ffv5kqfk8sGDB9SvX5+nT5+ioqJCtWrV+Pvvv7GysgIgISGBbt26fTK5TElJIeWD3wnOlGSgqqKWx+p/PlVVlawew16BZGZKCL76GFNrY/yH+0k/cHuWHebElnPSbUavG8DpbRc5vT17AH1MxKdvDyui8t+V497AI9JbHCFBT/Dy9aRBt9qsHLcx3/tWlndXt+dOPWDbposAhARH4uFpyw/NyytMLu/eCufurezbubdvPmPFxt40aVaONX9m3ZJcs/wEWzacl5aZvagTyxcf495t+WMGC8o/f2f/fz4OeUlaWgaDRzVm5ZJ/ScvFh1ZZnj14Tt+qP6Ojr0P1ZhUZHvgTIxpOz/dt5jVTtklvmwPM2j+GFT9v5v5lxUMgAOzdbZi4aRDrp+/k6rFbCsupqmSdGxt/283pnZcBmNN7Oesf/E715hXZt/J4vur9NSldw522o5qxcMAK7l18iI2jJX3mdiF2bAv+mrYNgL2Bhzm59b12Yu0ATm27wJkdF6XrPqedeJ+mtga12/rw19RtStlfYfsa2mGAfX8elf77ya0wYp/H8duRCVg5WPD8UeRn7buwiRh/ebNHbWbI1Jb8dXIsGekZPLwTwYm913HyyL6IXbfwMP+syh6ONGvNT6ycc4B7N8I+2p+qqgoamuosnbqHq2eDAZg57G/+OjWW0pUcuHomOMf66OhqMWlpV54+fMn6RQXbaZKn5HLUqFF4enpy+fJl4uPjGTx4MD4+Phw/fpxixeR3Dcszffp0Jk2aJLOuBO44qnjmpTofSYh+RUZ6hkyvDfzXuxMZL3eb2OfxpKdlkPnegzNP74VjYmWEuoYa6WkZJMa9kel9SX2bSnxUAhEhH38QYiPjMTTXl1mnqqaKvnFR4l4oqkPWhzX0ruwTZE/vhWNe7Ou4hZwQn0R6egahT6Jl1j8NjcaztF2u95ORkUnIgxfY2BpL18XHJREflyRTJjoqkYhcjOcsSPfuhKOuroaFlQHPnsZ+seOmp2UQ8eglAA+DnuBSvgTN+tZnwcDVH5WNi0zAyEz2fDMyN5Dp0UiISiThv15igIz0DGKex0mPIU8xV2tm7hnF/lXH2ThL/lOI78T+99l6ei/7YiAtNZ0Xj6MwtzPJcdvC8OpdO2H+cTsRq+Az2mVSa47+dYoDK7NuQT+5FYa2rhaDlvzEhunbkUgkH7UTKW9TiY96JbediFPQTujl0E68r3rLymjpaHFk/clPlv3SvtV2WJ57F7Jug9o4WX5ViY+I8ZcVF53VfhqZFCXuvbbUyLQoIXcVPxvxPCyWkZ3/RKuIBjpFtYmLSmT03Ha8eJb9fZIQ+4aE98b5Z2RkEh2ZINO7+U7sf8d++t7/R0LcG17FvcHc2jDH91BER5PJf3bjbVIKkwesJyO94G6JQx7HXJ49e5bp06djamqKk5MTu3fvpkGDBlSvXp1Hjx59egf/GTNmDAkJCTJ/JVTc8lz5D6WnZfDgyiO8fLNvsaqoqODl68mdc/Iz+ttn72PtZIGKSva4E1tnK2IiYnPVxfyhu+eC0TMqinO5EtJ1Xr6eqKiqSD9EH3rxJIro8FhsS8rezrB1tuLlZwy4Vqb09Ezu343ArphssmBjZ0zki9wPzFZVVaG4ozkxMYmfLlzIHJ0tycjIlEl8C4PKf1er8ty9+JCytWSHp5Sr7aFwjGZu2LvZMGvfaA7/dZrVk/75ZPnga09ITU7F1tlKuk5NXQ0Le1Mi5TSQhS09LYPgq48p65t9MauiokLZ2p7cPS//FrO2jqbMlzJkfQlkbZv3Otw5/+DjdqJ2VjuRm/+7ht1qc373FRKiv77P0bfaDsvjWLY4QI4PpxQGEeMv68WzOGKjXlG2sqN0nY6uFi6l7bh3Xf5du/elvE0jLiqRovralPdx5vwHUxzl1p3/ZlCx/W8oEmRNNaRvpMvL8HiF2+noajF1RXfS0zKY1HftJ2d4UYY8JZdv375FXT37S05FRYUlS5bg5+dHzZo1efAgd2N/tLS00NfXl/lT1i3xf+btpXEPX+p1rkExVxsGLu6Btq6W9HbzyNX9CJjaTlp+99LD6BkXpe+8rtg4W1GpsRftxjRj1+JD0jLauloYWRhI/6a2n8+lA0Ey697dNn56L5yLB64xZFkvXCo64lHVhf4LunF801nph8fE2ogVt+dKHw4A2Dx7N80HNKJ6S2+sHS3oMqk1dq427F/55QbraxfRwNHZAkfnrDGgllaGODpbYGaRdXW65a/z1KzjTqOmXljbGPFjywpU8SnJrm1XpPsY+XNTAnrXli537Fad8pUcsLQ2xKmkJaMnNsPC0oD9u4KkZXR0tTAy1pX+DfxpFWGh0TLr3mdtY4SjswXGxrpoamXXWV09+3QuVtwUR2cL9PSLoKurJfO+AFzcrFmxsTcm/42lcfO0oXnrSjg4mWNpbYhvfU96D6rH0YO3eJ2YrLwgf0K3X/zx9HHBopgpxT1s6faLP6Wru/LvpqxbVSMCe9Ltl+xponYsPkSFeqVoOaAhdiWt6Di2Gc7lSrDzvSe7dfSLYGRuIP0bXHsyYfefy6x7x949K7G8cvQW2xYelL5uYJo9rsilvAPLr07HxCpr+oukxGT2rviXTuOaU87XE1tnSwbM6wLAqe3Zt4S/Jv/M20vj7r7U61QDO1drBi7qntVO/Pf0+IhVfQmY0lZa/vyeq/zQqy61WlfBsrgZ5eqUossvrTm/56o06fywnZjWYT6XD8pvJ8LuRXDpQBCDl/bEpaIj7lVL0m9+N45vOie9k2FibcSKm3Nk2gkAa0cLSlV3Zf/KY3ytvsV22MrBgg7jWuBcrgQW9mZU8SvPyNV9uXHiDo9vfjqB+NJEjJVLW0cTB1crHFyzLpItbI1wcLXCzCqrfdyx9gxte/viXduN4s4WDJvhT8zLRM4eyU4Up6/sjl/7KtLlcj7OlK9WEgsbI7yqOjFj9U88exzFoe3Z35k6RbUwMi0q/RvSdjHPHkfJrHsn/Ek0Z4/cptfYH3ArWwx7ZwuGT/fn2eMorl/MGuZkYq5P4N4hlCyVNRVcVmIZgHYRDeaN/0fmePl9mCs38nRb3NXVlcuXL+PmJtvL+McffwDQtGlTeZt9USc2n8PQVJ8uv7TGyNKQkKAnjG08nfj/Bu+b25kgyczuDo56FsOYRtPoM6cLgUGziA6PZfsC2SmB/If50Xnix/M+vq+jQ3/ptA4zOi6k/8IAZh3+WTqx7KJBq6Rl1TXUKeZqg5aOlnTd9gX70NTWoPeczugZF+XR9VBGNZjyRW8TlHS1Zs6i7Pm9+gyqD8Chvdf5bepuzpy8z/xZ+2jX2Yd+Q+rzLDSGSeO2cvu9sSHmFgYyc3MW1dNmyOgmGBnr8joxmeD7zxnUazVP37u93m9wfeo3KZNj3epVnSL999AxP1CmnL10eemanwDo2GKhtBd16py2MhOivyvzbj9a2hoUszeVJqRpqRnUrutO5+410NBU40VEPNv+viAzDvNLMDTTY0TgTxhbGpL06i2Pb4Ux7sfZXP3viXAzO2My3zt/71x4yIyApXT5uSVdf2lFREgkk9rOl5njss+sDtTvWD3H4zYompUMVm9WEUMzfeq286FuOx/p6y9Co+jiMRzIeiDIrqQ16hrZF4R/jttERnomI5f3RFNbk/uXQxjVZGaOc+gVphNbzmFgpk/nif4YWRry6Hoo436Y8V47YSpzHv81bRsSiYQuk9pgamNMQtQrzu+9wqqfN0nL+A/1o9OEVjket5PTgOx2ovNC+s0PYObB8VntxPasCajfUddQx87VBq0iWjL7aNC1NtHPYrmSw8T6he1bbIfTU9MpV6cULQY1RltXi6iwGE5tu8iGr3Rcq4ixcjl72DBrbU/pcq/RPwBwePsV5o7dypblJ9EuosnASc0pqq/N7auh/NxzlUwvoFUxE/SNdKTLunradBvSAFNLAxITkjh96DZr5h2UuSXde6wf9ZqXz7FujdzGSP89Z/QWeo5pwqSlXZFIJNy89IjxP62S7lNNXRU7B3O0tDUAcHS3lj7RvvLQCJn9dqkzk5cR8XkJU66pSCSfmtI62/Tp0zl16hT79u2T+3rfvn1ZunSpzJdfbtVTa5PnbYQ88C716TLCZ1G9kfNgauHzSFK+rrkcv0df8kcbBKEgqJd0KOwqfPf2353+yTJ5Si4LkkguC5hILgucSC4LlkguC55ILoVvnUguC15uksvvahJ1QRAEQRAEoXCJ5FIQBEEQBEFQGpFcCoIgCIIgCEojkktBEARBEARBaURyKQiCIAiCICiNSC4FQRAEQRAEpRHJpSAIgiAIgqA0efqFHuHb9bp4kcKuwnfP4IHWpwsJ+SbmuRQE4VMynzwr7CoIiJ5LQRAEQRAEQYlEcikIgiAIgiAojUguBUEQBEEQBKURyaUgCIIgCIKgNCK5FARBEARBEJRGJJeCIAiCIAiC0nyXUxE17VMf/+F+GFsaEnI9lEWDVnH/UojC8roGOgRMaYtP80roGRflZWgUS4au4eL+oHwdX89Il34LAqj8QzkkmRJObbvA4sGrSX6T81QqbpWd6Ta5La7eTmRmZBISFMqYRlNJTU7LVz3yqnPTStSs6Iy9tTEpqencDI5g8caTPH0eB4ClqT7bF/wkd9tx83dz7MIDua+d2zBM7vo/Npzgrz2X81xPE0NdBnaoiauDBbYWRmw5eJV5647LlFFTU6VL00o0quGBmVFRnj6PZfHGU5y/8UThfr3cbGnbqDzujpboFtEi7EUcf+29xKEz9/JcR2VpPbAhARNasH3pEZaN3yy3zKydwyjt4/LR+ouHbzKh3cI8H1PPSJdRS7tTwsMWPSNdEqITObf/OqunbCfpdbLC7ZxKFyNgQgtKehUnMyOT03uuEvjzlk+e94XFr099/If6YWxpwKMbT1k0WHE7oaauRttRP1KvU01MbYwIe/CcFWM2cPnQ9XwfX89Il37zu+HdJKudOL39IouH5NxOGFkY8NPMjpSrUwodPW3CHjxn4/TtnN5+Md/1KCh5aYdnH51AmVoeH62/sO8q4/1m5uv4ZnYmDFrcgzK1PHj7OpnDa0+wYuxGMjMy5ZYvXdOdOccmyn2tn/dYHlxW/B1SWESMC06b4X74NKuAXUkrUt+mcedCMCvG/c2z4BcKt/H5sQJtR/hh7WiBuoY64Q9f8M/8/RzdeCbf9egzpxMelZ2x97Al7F4EfSuP/+Q2RhYG9JjWlnK+nujoFSHswXP+nrWT0zvy/n2bX99dclmzdRV6zenMgr7LuXshmBaDGjN9/1gC3IYQH/Xqo/LqGmrMPDie+KgEJrf+nejwWCzsTXkdn6TwGLOPTuDQ2hMcWnNC7uuj1w/AxNKI0Q2moqahxogVfRiyrCfTOyr+oner7Mz0fWPZOGMHiwatIiM9A4cy9kgyJXkPQj55udnyz+Eg7oa8QE1Nld5tqjFvdCvaj1xFcko6L2MSadJnicw2zXxL0/6HipwLeqxwvx9uU6VsCcb+1IB/LwbLLf8uia3Sfo7c1zXU1YhLfMvqHRdo26i83DK9/H1oWM2N6csPExoRi3fp4swY2pSeE//mQehLuduULmlNSFgU63dfIjbhDT7lHJjQpxFvklI5c+2RwvdXUEp62dO4Sw0e3QrLsdyvXZagoZn9UdY30mXxiQmc2qm4IVlzdRpzBqzmxpmPLwgkmRLOHbjOmuk7SYhOxLqEOf1mtUfPSIeZvVbI3Z+xpQHT/xnCiR2XWDx6Izp62vSa0oZhC7syNWBZLt/xl1PTvwq9fuvEgn7LuXfxIS0GNmba3jF09xgqt53o+msb6rSvxu+9Awm7H0GF+mWYuHUYg2tMICToidxj/HYkq504vFZBO7F2AMZWhoxpNA01DTWG/9mbwUt6MqOz4nZi5Kp+6BrqMLHFbyREJ+Lb1odxGwfTv/JYhfUoDHlthye1moP6++ewiR7Lrs3i5NbzCo+xLmQhvwUs4caJOx+9pqqqwtTdo4mNjGdwtZ8xtjJi5Op+ZKRlsHL833L3d+fsfVpb95RZ1/XXNnj5en5VSc87IsYFq3R1V3YvPcKDK49QU1ej6yR/pu0ZxU9eo0lJkn8BmBj7mo2zdhF2/znpqel4Ny7LsMCfiI96xZUjN+Vus+beXOb8FMiNU4o7MQ6uPYlrRUdKeNrlqu4jlveiqKEOv/j/TkJ0IrXbVGXs+gEM8JlAyPXQXO3jc313t8VbDm7C/uVHObj6OE/vhjO/z3JSklJp0K223PINA2qjZ6zLxOazuX32PpGhUdw4eZdHN/L3H1DM1YZKDb2Y23MZ9y4+5PaZ+/wxaBW12lTFxMpI4XZ95nRh+8L9bJq1k9A7z3j24Dknt5wnLTU9X/XIjyEzt7Hv5G0eh8fw8GkUU5YewMpMH9cSFgBkSiTEJiTJ/NWs6Myx8/d5m6K4d/XDbaqXd+LqnadEvEzIVz1fRL9i3tp/2X/qDq8VfMgbVndnzc6LnAt6TMTLBLYfuc7ZoMe0ayI/GQVYs/MigVvOcjM4gvCXCWw+cI3z159Qq6Jzvur5ObR1tRi5tAfzh6zjdYLiCx2A1/FJxL18Jf3zquVO8ttUTu66kq9jv05IYu+qEwQHhfLyWSxBp+6xZ+VxPCsrjoN3/dKkp2WwaORGnj2M5MG1UBYO/4vqTctjVcIsX/UoSC0HN2H/imMcWnMiq53o+1870bWW3PJ1O1Rj48wdXDoQxIvHL9mz7DAX91+j1ZAm+Tq+nas1FRuWZW6vQGk7sWjwamq1qYJxDu2Ee5WS7Fx0kPuXQnjx+CUbpm/nTfwbnMuVyFc9Ckpe2+HEuDfERSZI/8rVLU1yUgontyhOfHJSvn4ZirnbMqPTH4RcD+XSgSDWTNxE074NUNdQk7tNelqGTB1exbymStMKHFx9PF91KGgixgVr3I+/cXj9KULvhvPo5lPm9AzEopgpzl7FFW5z49Q9zu66Qtj9CJ4/fsmORYd4dDMMj6ol812PJcPWsXvZEZ4/lt8pIo97ZWd2Lj7M/cuPePEkio0zd2a1EznUXdm+q+RSXUONkuUduHo0+wpBIpFw9ehN3KvI/2Ks4leBO+eDGfBHAJsjlhF4fTbtRjdDVVUlX3Vwq+JMYtxrHlzJ7um6euQmkkwJrt5OcrcxNNPHrbIz8S9fMe/Ur2yOWMacYxPxkHOr80sqqpP1izOvFNwKdSlhTsni5uw+fivX+zTS18GnbIk8bZMfmupqpKbJJuYpqemUcbHJ036K6mgpfP8Fqd/Mdlw8fJNrJ+/medsGHapxYvslUpJSlVIXY0sDfH7w4uZZ+cMeADQ01UlPS0ciye5pT0nOOr6ngvO+sKhrqOFcrgTXPmgnrh27iVtl+V8CGloapH0wPCU1ORWPqq75qoN75ZIkxr0m+P124mhWO+FWSXG87px7QE3/KugZ6aKiokKt1lXQ1NaQ27NUWPLTDn+oUUBtjm86S7KCi8dPca/szJObT4l/7wL28sHr6BroYO+Ru96fKk3Lo2+i91UmPiLGX56uftav3CXGvcn1NmVruWNX0opbp+8XVLXkunM+mJqtvKXtRE3/ymhqa3IjH98n+aX02+JhYWFMnDiRlStXKiyTkpJCygc/5ZYpyUBVRf7VTm4ZmOqjpq5GXKRsj1hcZAJ2LtZyt7EsYU7Z2h4c3XCacT/MwNrJkoF/dEdNQ531k7cC0G50M9qNaS7dRrOIJm6Vnem/IEC6rrvnUKLCYjC2MCT+pewticyMTF7FvsbI0lBuHawcsnoGO09sReDI9TwMekK9TjWYdfhnepYeTvhDxWM8CoqKCgzuVIvr98N59CxGbhm/WqV4/CyGm8ERud5v4xoeJCWncvyS7C3xv2Z1wdJUP+vYZCX2R1cOkL5+/V44Q2dty/VxLtx4QtvG5bl27xnhkfFU8LCnVkXnPF001PEuiZuDBTOXH871NspQs3lFnErbM7De1DxvW9KrOCXcbfh90BqZ9QNmd8C3lbd0WUtHk8l/D5QZG9W8+ECZbUYH9qByw7Jo62hy/sB1fh+8VuFxr5+6R8/JrWnVvz47lh1FW0eLgJ9bAGBsYZDn91GQ9N+1Ey/ltRPyLz4uH7pBi0GNuXHqLs9DIvHy9cSnWSVU1bKvz9uOaka70c2ky5pFNHHzdqL//G7SdT1KDyMqLAYjBe1EYg7tBMCUdvMYt2EQ/7xcQXpaOilJqUxqNZeIkMg8RKBg5acdfp9LRUdKlCrGnJ+WyqwftLgHdTpUly5r6Wgybe8YmXO4qUEXAIwsDeX+/wJZ4xNz8T4aBfhy5dB1osNjc1H6yxIx/rJUVFTo/VtHbp29T+idnH9eUke/CBtCFqChpU5mRiYLB63h6rHszpSBC7ri285Huqylo8mUnSNkYtzMTP6zDbk1teMfjF3Xj60RS7PbiTbziHiU+97Pz6X05DI2NpY1a9bkmFxOnz6dSZMmyawrgTuOKp7Krs4nqaqqZPUY9gokM1NC8NXHmFob4z/cT5pc7ll2mBNbzkm3Gb1uAKe3XeT09gvSdTERcfmug8p/Cc/ewCPSK7iQoCd4+XrSoFttVo7bmO9959fwbnVwsDOl1yT5Y2e0NNSpX9WVVdvzdkvFr5YnB8/cIzUtQ2b9sFnbUFfLurgwMyrK4glt6DJmnfT1lDwOD/h97b+M/qk+f8/uhkQC4ZHx7D1xmx/kDGiXp5y7HeN6NWTG8sM8DpefXBcEU2sjek9tw9hWv5OWkvchEQ07VuPx7Wc8uPZEZv3aGbvYuuiQdHnWzuGs/PUf7l1RPFZ22fjNrJ+1B1snC7qNb07Pya1ZNHKD3LKh958zu/8qev7qT7fxzcnIyGTXn8eIjUwg8wuOGy4oS4auZsjSnqy4NRckEiJCIjm05jgNumbfgtwbeJiTW99rJ9YO4NS2C5zZkf2wzee0EwBdJrWmqKEuI+tP4VXMK6o2rci4jYMYWvsXnnxibO63omGAL49uhH70YMqaiZvZMme3dHn2sYksH7OBexfkj93+HKY2xpSvX4YpbX9X+r6/BiLGedN/XhfsPWwZVmfyJ8u+TUymr/c4tItq41Xbg14z2/Pi8UvpmMo1k7exdd4+aflZh8axYvwm7l98qLT6dpnYkqKGuoxqND1r6IFfecat78+wulN4cvvL/PZ6npPLXbt25fj6o0effvBhzJgxDB06VGZdc8MABaVzLyH6FRnpGRh90FNiZGFAXGS83G1in8eTnpYh8wX49F44JlZGqGuokZ6WQWLcG5mu8NS3qcRHJcjtLYiNjMfQXF9mnaqaKvrGRYl7oagOWV84oXdl/9Of3gvHvJipwvdbUIZ19cXHy5E+v/5NVOxruWVqezujraXB/lO5vx1XxsUGe2tjxi/Y89FrL6ITpf9O/+8K7pmC/7PciE98y+i5O9HUUMOgaBGi4l7Tt211wnMxztPL1Zbfhjdj/vp/8/T+lMG5jD1G5vr8cSz7iUA1dTU8qzjTtEdt/Kz7KkzWtHQ0qdm8Imtn7PzotYToRBLei3FmegYxz+N5/jhKYV3ejeF89vAFiXFvmLN3JBvn7CU2Un4Mj/9zkeP/XMTQTI/kpFQkEgnN+9TjRWh0bt/+F/HqXTth/nE7EavgM5oQncgvreagoaWBvklRYiLi6D6tPc8fZbcBH7YTKW9TiY96JbediFPQTujl0E5YOVjQrF9DfiozXNp78ujGUzyrudK0T30W9JP/sNWXlp92+B1tHS1qt6nKmokfz4wQH/VK5kGVjPQMosNj5cf3RTyuFWWHF7yrj6L/4/c16FqLVzGJnMvnuOWCJmL85fT7vXPWgzl1pxId/umLQ4lEIu0hfHTjKXYu1rQZ4SdNLhOiXpHwQYxjwmOV1qtoVcKcH/vUp2e50YTeDc+qx82nlPIpSdNedVkwcLVSjvMpeU4umzVrhoqKiszYqg+pqOR861FLSwstLS2ZdZ97SxyyBgs/uPIIL99SnP3vSVkVFRW8fD3Zueig3G1un71P7XY+Mu/J1tmKmIhY0j/oXcuNu+eC0TMqinO5EgRfzeoV8vL1REVVhXsX5F+ZvHgSRXR4LLYlZW9n2DpbcelAUJ7r8DmGdfWlZgUn+k7ZzHM5Txy+41erFKeuhBCf+DbX+/ar5cndRy94+FRxQqNsqWkZRMW9Rk1NldqVnDmqYLqkd7zcbJk9ojmLN55i5zH5T/cVpKBTd+lV7ReZdcMWdiUs+AWbFxzIsRewRtPyaGiqc2zLBYVl8utd7/r7T6UrEh+VlcTWb+9DWnIaV49/PeMBIaudCL76mLK+npzdld1OlK3tya7F8tuJd9JS0oiJiENNXY1qzSvl+KRtTu6cf/BxO1E7q524q6AHQ0tHE4DMTNlpXjIzMlFV/XqGz+enHX6nhn9lNLTUOfLXqc+qw53zwbQb2wJDM31pslSuXmneJCTx9BO3NSEr8Tmy7iQZ6Xn/DvgSRIy/jH6/d6Zq0/KMqD+NyND8fW+pqqqgoaWh5Joplt1OyH5XZGRkovIF24k8H8nKyopt27aRmZkp9+/q1asFUc9c+2feXhr38KVe5xoUc7Vh4OIeaOtqSW83j1zdj4Cp7aTldy89jJ5xUfrO64qNsxWVGnvRbkwzdi3OvoWorauFkYWB9G9q+/lcOhAks+7dWL6n98K5eOAaQ5b1wqWiIx5VXei/oBvHN50l5r8eShNrI1bcnotLRUfpMTbP3k3zAY2o3tIba0cLukxqjZ2rDftX/vsFopZleLc6NPBxY+If+0h6m4qxgQ7GBjpoacgmFLYWhpR1tWX3v/KTr79nd6NmBdkrWp0imvh6uyjcxlCviPR4qWnpNOmzRLpsbKCDvq62THlnezOc7c0ooq2Bob4OzvZmFLcxlr7u7mhJzYpOWJsbUMbFhnmjWqCiqsL63ZekZVrVL8vCsa2ky+Xc7ZgzogVbDl7j34sPFB67IL19nULovQiZv+SkFF7Fvib0XtbY1uGLutFtfPOPtm3QoRpn9wfJHXCuo1cEI3N96d/ghjMIC34hs+6dinU9qdeuKvau1ljYmVCpXikGzO7A7fMPiQzLGiJQ0qs4f577FZP3xgf6da+NU+li2Dia4xdQi74z2rFqynbevMr9BciX8s+8vTTu7ku9TjWwc7Vm4KLuWe3Ef9OLjVjVl4ApbaXlXSs54dOsIpYlzPH0cWXa3jGoqqqweXb2nZwP24lpHeZz+aD8diLsXgSXDgQxeGlPXCo64l61JP3md+P4pnPSOxkm1kasuDlH2k6E3YsgPPg5gxf/hEtFR6wcLGg5uAnl6pbizM7s8/prkNd2+J2G3WpzZudlEuXcMdHRLyITy4FVxxN2L1xm3TtXDl3n6Z1njFrbH4fS9lSoX4auv7Zh1+KD0hk4XCo6suL2XEysZZ/O9/L1xMrBgv0rjikxIsonYlyw+s/rgm/bqszosoS3r5Ol719TOztRHLG8F91+bS1dbjPcj3K+nlgWN8POxZqWgxpRp70Px96b5/LDGA+uOYmwB8/lxhjA2sEch9LFMLYwQLOIJg6li+FQupj0iXwTayOWB83EpYIDAGH3nxP+8AWD/uiGSwUHrEqY03JQI8rV8eTs7i/XS5znnsvy5ctz5coVfvzxR7mvf6pXs6Cd2HwOQ1N9uvzSGiNLQ0KCnjC28XTpE23mdiZI3rvyj3oWw5hG0+gzpwuBQbOIDo9l+4KsKYHe8R/mR+eJ/jket6NDf+mVzYyOC+m/MIBZh3+WTqK+aNAqaVl1DXWKudqgpZPde7t9wT40tTXoPaczesZFeXQ9lFENpsjcditoLeuVBWDxhDYy6ycvPcC+k7elyz/U8uRlbCIXbj6Rux97a2N0dWR7putVcUFFBQ6dlT+X18opHbAyU/zgx9U7YfSbkn0bZ+30ztJ/uzlY0sDHjedRCbQYtBzIGhPay78a1uYGvE1J41zQIyYt3i8zdZGBXhFsLAyly42ru1NEW4MuP3rT5cfsh18+PHZhM7c1/mj+U1snCzyrODOmpfzxS32mtaFeu6o57rehadb8cynJaTTqVJ1eU1qjoalOVEQcZ/ZcZfP8A9Ky2jqa2Dlbykw54lKuOJ1G+aGtq8Wz4BcsHLaeo/mc5qSgndhyDgMzfTpP9MfI0pBH10MZ98OM99oJU5kYa2hp0HVSG6wczHn7OpmLB4KY2XURb96bJsp/qB+dJrT66Fjv6+Q0ILud6LyQfvMDmHlwfFY7sT3rxxbeUddQx87VBq0iWZ+ljPQMxjWdSfep7fh1+wiKFNUmPCSS3wKWfPE7HJ+S13YYwLakFaWquzGqwRS5++w3ryv1u9TK8bj11LLarsxMCeObzmTQ4h7MPzOZ5DcpHF57gtXv3QrW0tGimKsN6h9cPDcMqM3tM/cJu5/7BxULg4hxwfLrVReA2YfHyayf/VMgh9dn9fqa2ZnI9BBq62rRf34XTG2MSX2bStiD58wKWMqJrdl3k/rM7kT9TtXJSYMinaT/HrykB2VquEmXl1zIetCzs8sQIp9Go66uhp2LNVpFsnosM9IzGN9sNt2ntGHS1qEUKapNREgks3sEculg/n/0Ia9UJHnMBE+dOsWbN29o2LCh3NffvHnD5cuXqVmzZp4q8u6EFQrG6zaVCrsK3z2DQ192uon/N5kJiodpCMrxJX+0QRAKgqqmZmFX4bt38O26T5bJc89l9eo5Z9y6urp5TiwFQRAEQRCE78PXMwpcEARBEARB+OaJ5FIQBEEQBEFQGpFcCoIgCIIgCEojkktBEARBEARBaURyKQiCIAiCICiNSC4FQRAEQRAEpcnzVETCt+mVnbiOKGiGGl/uJ77+H4k5GIXvwbufUhUKhiQ9rbCrICB6LgVBEARBEAQlEsmlIAiCIAiCoDQiuRQEQRAEQRCURiSXgiAIgiAIgtKI5FIQBEEQBEFQGpFcCoIgCIIgCErzXU5F1LRPffyH+2FsaUjI9VAWDVrF/UshCsvrGugQMKUtPs0roWdclJehUSwZuoaL+4PydXw9I136LQig8g/lkGRKOLXtAosHryb5TUqO27lVdqbb5La4ejuRmZFJSFAoYxpNJTX5y0yt0KZyadpULo2NkT4ADyNjWHL0AqfvPwHAztiA4U1qUK64NZrqapx+EMq0nf8S8zpJ4T7Ll7AhoEYF3G3NMdcvyoA1uzh2R/H/xae4WJnSo1ZFvIrbYKRbhPC4BDafv8n6M9ekZep6ONGmSmlcrczQVFfjYWQMi4+c58yDUIX77Vu3Mv3qVflofVJqGhV//iPf9c0PE0sDAsY3o0Jtd7SKaBLxJIrfh6wn+PpTueU9KjnSbdyP2DlZoFVEk5fhsexbd5odgf/muw4TV/fCwdMWQxM9Xickce3UfVZO2UFsZILc8ua2xqy5NFnua1N/Ws7pPdfkvlaYRDtRsPIS39lHJ1CmlsdH6y/su8p4v5n5Or6ZnQmDFvegTC0P3r5O5vDaE6wYu5HMjEy55UvXdGfOsYlyX+vnPZYHl/PfbhUUvz718R/qh7GlAY9uPGXRYMUxVlNXo+2oH6nXqSamNkaEPXjOijEbuHzoer6Pr2ekS7/53fBuknUOn95+kcVDcj6HrRws6DmzAx4+rmhoqXP54HUWDV5N/Ev5bUthKVXNFf9hfjiXK4GJtTG/tJzN2V2Xc9ymdA13es3uhL27LVFhMWyYvp3Da0/kuw7fcnxVJBLJVzF5XD21NkrZT83WVRi5uh8L+i7n7oVgWgxqTI1WlQlwG0J81KuPyqtrqDHv1GTioxLYOH0H0eGxWNib8jo+iUc35Ccjs49O4NDaExxaI/+kmbp3NCaWRszr8ydqGmqMWNGH+5dDmN5xocJ6u1V2Zvq+sWycsYPze66QkZ6BQxl7zu28TFpqev6C8Z6I4ZU/WaaWmwMZmZmERsejogI/lncnoEYFWi74i4jYBLYN6cT951EsOnwOgAH1q2KuX5R2izai6Cyq5lKccvbW3A6PZEHnprlKLg+NCmDclkNcevTso9eaV/DAxcqMI7eCeZHwmrL2VvzSoi5z951iw7msRnK0X01evnrDxZAwXr1NoXkFD7rVKE/bRRu5FxEl95g6mhroaMrOU7miZytuhb1g3JZDnwodAMXWPsxVuZwUNSjCH4fHcP3MA/auOUVCzGtsHMx4/iSa56HRcrdx9LTF1smCx3ciSE5KwcPbkYGz2hE48R/2rz8jd5uZ/wzi8KYLHNl8Xu7rzXrW5t7lx8S+fIWJpQE9JrQAYFjTOXLLq6qqYGCiJ7OuUUcfWvatS4cyY0lOyjlhyo30l/Lff36IdqJg5TW+eka6qGtm93Xom+ix7Nos5vZcpjB+60IW8lvAEm6cuPPRa6qqKiy9OovYyHj+HLkeYysjRq7ux/7lR1k5/m+5+1PXUEPPuKjMuq6/tsHL15POzgPz8vZzpKx5Lmv6V2HEqr4s6Lecexcf0mJgY6q39Ka7x1C5Me4+rT112lfj996BhN2PoEL9MvT6rRODa0wgJOiJ3GP8diTrHFaUIE3dPRpjK0Pm912OmoYaw//szf3Lj5jRWf45rK2jxdKrs3h0I5S1v24BoOsvrTG2NmKQz898JekIABUblMWjakmCrz5m4tZhn0wuLYubERj0G3sCj7B/5TG8fEvRZ05nxjedyZXDN+Ru863G91Ca/M/Q+767nsuWg5uwf/lRDq4+DsD8PsvxblyOBt1qs2nWzo/KNwyojZ6xLoOq/UxGegYAkaHyE5DcKOZqQ6WGXvSrNIYHVx4B8MegVUzdM5rAEeuJeR4nd7s+c7qwfeF+mTo+e/A83/XIj+N3H8ksLzh4lraVy1CmmCUW+rrYGOnTav5fvElJBWDs5oOcm9gXb8dinH8ov1ft9P0n0p5PZdh++bbM8rPYBMoWs6Kup5M0uZyxW/aDOv/gGXw9HKnt5qAwuUxKTSMpNbvnx8XKFCcLEyZtO6q0uueGf7/6REXE8fuQ9dJ1kWExOW4TcusZIbeyE/GXz2LxaVwWD28nhcnlp7zf6/nyWSyb/zjEhFU9UVNXJSP9456fzEwJcR98oVVtVIZTu64qJbFUNtFOFKy8xjcx7o3Mcq02PiQnpXByi/yLn08pX78MxdxtGVl/CvEvEwi5HsqaiZvoMb0DaydtIT0t46Nt0tMyiHuvZ15NXY0qTSuw848D+apDQWs5uAn7VxyTJt/z+y6nUiMvGnStxabfdn1Uvm6HamyYsYNLB4IA2LPsMF6+nrQa0oSZXRbl+fh2rtZUbFiWfpXHEvzfObxo8Gqm7B5F4Kj1xMo5hz2qumBR3Iy+FUeTlPgWgFkBi9kWtYKytT24duxWnutRUC4dDOLSwaBcl2/Ssx4vHkcRODKr7Q67F4FnVRdaDGqsMLnMybce3+9qzKW6hholyztw9ehN6TqJRMLVozdxr+Isd5sqfhW4cz6YAX8EsDliGYHXZ9NudDNU83l16VbFmcS419IvDICrR24iyZTg6u0kdxtDM33cKjsT//IV8079yuaIZcw5NhEPH5d81UEZVFVUaFSmJEU01bke+hxNdXUkEkhNz26UU9IyyJRIKFfcutDqCVBUW4uEt4oTGBUV0NXSICEpOdf7bFnRk8dRsVx9Eq6MKuZa5QalCL7+lLGB3dl4cwZ/HBpNww5V87QPR09b3Co4cPNcsFLqVNRQh9otKnL38mO5iaU8TqXtcCxlx8GNZ5VSB2US7UTByk98P9QooDbHN53N94WJe2Vnntx8KnMr8PLB6+ga6GDvYZerfVRpWh59Ez1pgvw1UddQw7lcCa59EONrx27iVrmk3G00tDRI+2DoRGpyKh5VXfNVB/fKJUmMey1NfACuHs06h90qyT+HNbTUQSIhLSW7HmnJaUgyJXj65K8eXwv3ys5cPXZTZt3lw9dxV/D/8en9fdvxzXPPZXR0NCtXruTcuXO8ePECAEtLS6pWrUrXrl0xMzNTeiVzy8BUHzV1NZmrT4C4yATsXOQnQJYlzClb24OjG04z7ocZWDtZMvCP7qhpqLN+8lYA2o1uRrsxzaXbaBbRxK2yM/0XBEjXdfccSlRYDMYWhsS/lO3ByczI5FXsa4wsDeXWwcrBAoDOE1sROHI9D4OeUK9TDWYd/pmepYcT/vBFnmORX86WJmzo2xZNdXWSUlMZuHY3IS9jiX3zlrdpaQxrXI15B86gAgxpVA11NVXM9HU/65gTmtfBzyv7xNfW0GBpQHMyM7MTmYoT5F9Zl7W3omGZkvRdtUPh/rvVqICOpiYHbjzIVX001dX4wcuN5ccv5e4NKJFlMVOadK7OtsBjbFpwkJJl7ek92Z/01AyObLmQ47brrkzBwKQoqupq/DVnLwc3ZCd2bQY2oM3ABtJlTW0NXMuVoO+01tJ1vWpOJio8+2o4YNyP+AXURFtHi7uXHzGx89Jcv48G7ary9MFz7l5+nOttvhTRThSs/MT3fS4VHSlRqhhzfpI93wYt7kGdDtWly1o6mkzbO0ZmDGVTgy4AGFkaEvfy4+MDWWNAc/E+GgX4cuXQdaLDY3NR+svSfxdjOe/RzsVG7jaXD92gxaDG3Dh1l+chkXj5euLTrBKqatl9TG1HNaPd6GbSZc0imrh5O9F/fjfpuh6lhxEVFoORgnM4MYdz+O6FYJLfpNB9entWjf8bFRUVAqa1Q01dDWMr+dt8K4wsDImXc87rGuigqa1BanLa/1V885RcXrp0iQYNGqCjo0PdunUpWTIrI4+MjGTBggXMmDGDgwcPUqFChRz3k5KSQkqK7BVppiQDVRW1PFb/86mqqmT1BPQKJDNTQvDVx5haG+M/3E/6pbFn2WFObDkn3Wb0ugGc3naR09uzv+xjIuTfxsqNd2Nw9gYekV4lhwQ9wcvXkwbdarNy3MZ87zuvnkTF0XL+eopqa1G/lDPTWjeg67IthLyMZej6PfzcvA4dqnqRKZGw7/p9bj+LJPMzf/P5j0NnWX0yeyzLql7+zN13mpthOd/uc7IwYWHnpiw5cp6zwfJvyzcp60KfupUZsGYXsW/e5qo+dT2c0NHSYOeVj8dyFTQVVRWCrz9lzfSs21oht55h72JN487VPplcDm/2O0V0tXAtX5xuY38k4nEUJ3ZcAWDv2lOc3HVVWnbkoq6c2XuNM/uyB/PHvJBtGLcuOcLBjecwtzWmw7DGDF/QmYmdlnzyPWhqa1CreQU2/v513k7MD9FOfDkNA3x5dCP0owdT1kzczJY5u6XLs49NZPmYDdy7oJwe+veZ2hhTvn4ZprT9Xen7LixLhq5myNKerLg1FyQSIkIiObTmOA261paW2Rt4mJNb3zuH1w7g1LYLnNlxUbruc87hhOhEprSdx4A/utOsf0MkmRL+3XSW4KuPPvt75Fvw/xTfPCWXAwYMwN/fn6VLl6KiIns7SCKR0Lt3bwYMGMC5c+cU7CHL9OnTmTRpksy6ErjjqOKZl+p8JCH6FRnpGRhZGMisN7IwIC4yXu42sc/jSU/LkAn803vhmFgZoa6hRnpaBolxb2TGBKW+TSU+KoGIkMiP9xcZj6G5vsw6VTVV9I2LEvdCUR2yTqbQu7IPsDy9F455MVOF77cgpGVk8jQmK8m4E/4ST1tLOlbzYtK2o5wNfkqjWasw1NEmI1NCYnIKJ8b3ZP/1z3sKLfbNW5nELyMjk5evXkvrIY+juTErfmrJlos3WXbsotwyjcqUZFLLegz9a6/CMaHytKzkyYm7j3N8Cr6gxL58xdMPxtCFBb/Ap0nZT277bmzmk3sRGJrp03FYE2ly+To+idfx2e8nNTmV+OjXPH+ieNzgq9g3vIp9Q/ijl4QFv2Dd1am4li/BvSs590ZW+8ELrSKaHN2aczJcWEQ7UbDyE993tHW0qN2mKmsmbv7otfioVzIPqmSkZxAdHis3vnEv4nGtKHvr8F19YhXE930NutbiVUwi53Zd+WTZwvDqXYzNP46xoveXEJ3IL63moKGlgb5JUWIi4ug+rT3PH2XH78NzOOVtKvFRr+THWME5rJfDOQxw5cgNuroOQt9Ej4z0DN4kJPF32FJePPr6htDkRVxkPIZyzvk3CUnSmRz+n+KbpzGX169fZ8iQIR8llgAqKioMGTKEoKCgT+5nzJgxJCQkyPyVUHHLS1XkSk/L4MGVR3j5lpKpl5evJ3cUjD+7ffY+1k4WMu/J1tmKmIhYuYO+P+XuuWD0jIriXK6EdJ2XrycqqircuyD/aeIXT6KIDo/FtqTsLSNbZytefsZDA8qgqgKaarI9yvFJySQmp+DtaIexrg7/3nmkYOuC4Whhwsqerdh15S4LDsr/wDQu48IU/waM2LiPk/dyf2vWxkifSg52bLtUOAPL71wMwdbJQrZOjua8fJa3W3OqKipZ42+U5F2vmYbmp/fZoF0VLhy6SULMa6UdX5lEO1Gw8hPfd2r4V0ZDS50jf536rDrcOR9M8VLFMDTL/nIuV680bxKSeHrn41koPtSgay2OrDspfXjra5OelkHw1ceU9c3ukFFRUaFsbU/uns95+E9aShoxEXGoqatRrXklzu3OXwJ95/yDj8/h2lnn8N2Ln54541VMIm8SkihbywNDc33O7fk6E/ncunM+GC9f2Q6ycnVKc+cT/x+K9/dtxzdPyaWlpSUXL8rvJQK4ePEiFhYWCl9/R0tLC319fZk/Zd0S/2feXhr38KVe5xoUc7Vh4OIeaOtqSW8jjVzdj4Cp7aTldy89jJ5xUfrO64qNsxWVGnvRbkwzdi3Onn5GW1cLIwsD6d/U9vO5dCBIZt27gf1P74Vz8cA1hizrhUtFRzyqutB/QTeObzorfQLUxNqIFbfn4lLRUXqMzbN303xAI6q39Mba0YIuk1pj52rD/pX5n6swrwY39KF8CRusjfRxtjRhcEMfKjrYsSfoHgDNKrhTupgldsYG/ODlytwOTVh7+ipPorO78Vf81JL2VcpIl3U0NXC1MsPVKmssrq2xPq5WZlgZZk9bU1RbE9OiOtK/9ov+5vHLWJl17zhZmLCqZyvOBj9lzakr0teNdItIyzQp68K0Ng34bc8Jbj59IS1TVFtTWqZ9lTKs+KnlRzFoUdGDqMQ3nFLiE+55sSPwGK7lStBmYAOsiptRq3kFGnX0Yc/qk9IyXcc2ZdiCztLlH7rWwLueJ9YlzLAuYUb9dlVo2acOx/7J/qxq62hhZKYv/ZvRexVX/r0js+7dOeziVRy/bjVx8LDF3NaYMj4lGbW4GxGPo6S9liaWBgSe+pmSZe1l6m9V3AzPyk4c2PB190KIdqJg5TW+7zTsVpszOy+TGPvxhYmOfhGZWA6sOp6we+Ey6965cug6T+88Y9Ta/jiUtqdC/TJ0/bUNuxYflE7Z5FLRkRW352JibSRzHC9fT6wcLNi/4pgSI6J8/8zbS+PuvtTrVAM7V2sGLuqeFeP/nh4fsaovAVPaSsu7VnLCp1lFLEuY4+njyrS9Y1BVVWHz7Ownyz88h6d1mM/lg/LP4bB7EVw6EMTgpT1xqeiIe9WS9JvfjeObzkl72U2sjVhxc47MOVy/S01cvZ2wcrCgTvtqjP97MNvm7/vqZj3Q1tXCoYw9DmWy2jjLEuY4lLHHzM4EgIApbRmxqq+0/N7Aw1iVMKfH9PbYuVjj17seNf0rs23+Ppl9/r/EN09dG8OHD6dnz55cuXKFOnXqSBPJyMhIjh49yp9//sns2bMLpKK5dWLzOQxN9enyS2uMLA0JCXrC2MbTpU8NmtuZIHnvQZGoZzGMaTSNPnO6EBg0i+jwWLYvkJ3qw3+YH50n+ud43I4O/aVTk8zouJD+CwOYdfhn6eTIiwatkpZV11CnmKsNWjpa0nXbF+xDU1uD3nM6o2dclEfXQxnVYIrMLYuCZlxUh+mtG2Cmr0ticioPnkfTc+U2zv03nrGEqTFDGlbDoIg24XGvCPz3ImtOXZXZh52xAYbvJXoethas7pUdu1F+tQDYcfm2dP7IMX61aFbh4wmU3+cxKmvsU/1SzpgU1aFpOTealsvu7Q6PTaD+zJUAtKpUCg01NX5uXoefm9eRlnn/mIa6RbAzlr2FkTW3pwc7rtwms5DmW3tw/SmTAwLpOrYp7Yc04kVYDMsmbOXfbdkPFxmbG2Buk/2FqKqqQtexP2JZzISM9Eyeh0axcspO9q07LS3Tsk8dOg5vkuOxu1T8mZfPYkl5m0rVxmXoOLwx2jpaxL5M4Mq/d5nea4X0i1lNXQ07J0u0imjK7KN+uypEP4/n6vG7yghHgRHtRMHKa3wBbEtaUaq6G6MaTJG7z37zulK/S60cj/tuvuTMTAnjm85k0OIezD8zmeQ3KRxee4LV791u19LRopirDeoasl+DDQNqc/vMfcLuR+T1bX9RJ7acw8BMn84T/TGyNOTR9VDG/TDjvRibInlvGIeGlgZdJ7XBysGct6+TuXggiJldF/EmIXu4jP9QPzpNaJXjcTs5Dcg+hzsvpN/8AGYeHJ91Dm/P+iGAd9Q11LFztUGrSPY5bFvSmoAp7dAzLkrkkyg2ztjOP/P2fXiYQleyvCOzj06QLveenXVBf2jtCWZ3X4KxlRHmdtnDUV48iWJ801n0ntOJZgMaEf0slrm9AmWmIfp/im+eJ1HftGkTv//+O1euXCEjI+uWgZqaGuXLl2fo0KG0bt36E3uQT1mTqAvy5WYSdeHzKGMSdUExZU6iLgiFRVmTqAtCYSmQSdTbtGlDmzZtSEtLIzo6q7E3NTVFQ0PjE1sKgiAIgiAI37t8j/jX0NDAyspKmXURBEEQBEEQvnHf1S/0CIIgCIIgCIVLJJeCIAiCIAiC0ojkUhAEQRAEQVAakVwKgiAIgiAISiOSS0EQBEEQBEFpRHIpCIIgCIIgKE2eJ1EXICUlhenTpzNmzBi0tLQ+vYGQZyLGBUvEt+CJGBcsEd+CJ2JcsL7n+IrkMh9evXqFgYEBCQkJ6OvrF3Z1vksixgVLxLfgiRgXLBHfgidiXLC+5/iK2+KCIAiCIAiC0ojkUhAEQRAEQVAakVwKgiAIgiAISiOSy3zQ0tJi4sSJ390A3K+JiHHBEvEteCLGBUvEt+CJGBes7zm+4oEeQRAEQRAEQWlEz6UgCIIgCIKgNCK5FARBEARBEJRGJJeCIAiCIAiC0ojkUhAEQRAEQVAakVzm0pAhQ2jRokVhV+O7JmJcsER8C56IccETMS5YIr4F7/8hxiK5zKWLFy9SoUKFwq7Gd03EuGCJ+BY8EeOCJ2JcsER8C97/Q4zFVESfkJqaiq6uLunp6dJ13t7enD9/vhBr9X0RMS5YIr4FT8S44IkYFywR34L3/xRj9cKuwNdOXV2dM2fO4O3tTVBQEBYWFmhraxd2tb4rIsYFS8S34IkYFzwR44Il4lvw/p9iLG6Lf4KqqioRERGYmJhQpkwZLC0tMTQ0/Khc3bp1CQ4O/vIV/A7kNsZC/oj4FjxFMW7evDlGRka0atWqsKv4zZMX48TERGrVqoW7uzulS5dmy5YthV3Nb5a8+AJUqFCBsmXL4unpyZ9//lnItfy25dQWJyUlYW9vz/Dhwwu3kkoiei5z4dq1a5QpUybHMsHBwTg6On6hGn1/chNjIf9EfAuevBgPGjSIgIAA1qxZU0i1+r58GGN1dXXmzZtH2bJlefHiBeXLl6dx48bo6uoWYi2/XR/GV09Pj5MnT6Kjo8ObN2/w9PSkRYsWmJiYFGItv22K2uKpU6dSuXLlQqhRwRA9l7kQFBT00clw+/ZtvL29KV26NFOnTsXS0hJVVRHO/JIX4z///JNy5crh6elJmzZtCqlm34cP43vhwgX8/PykywcOHKBTp06FUbXvhrxzuFatWujp6RVSjb4/H8bYysqKsmXLAmBpaYmpqSmxsbGFVLtv34fxVVNTQ0dHB4CUlBQkEgniMY3PI6+dCA4O5t69ezRq1KiQaqV8IhvKhZs3b0obMMj6kLVp04bly5dz48YNLl26ROnSpQuvgt+BD2McFxfHokWLuHTpErdu3WLZsmWFV7nvwIfxdXNz4/79+9LlyZMnM2HChEKo2ffjwxgLypdTjK9cuUJGRgZ2dnZftlLfEXnxjY+Pp0yZMtja2jJixAhMTU0Lp3LfCXkxHj58ONOnTy+cChUQkVzmQmZmJvfv3yciIoKEhAR27NhBzZo1KVWqFJD1RS2Sy8/zYYzV1dWJi4tj5MiR3L59W4wR/EwfxldfX5/U1FTS0tLYt28fzs7OODs7F3Y1v2kfxlhQPkUxjo2NpXPnzgQGBhZi7b598uJraGjI9evXefz4MRs2bCAyMrKQa/lt+zDGO3fupGTJkpQsWbKwq6ZcEuGT1q1bJ7G2tpYAkuHDh0vGjRsnCQwMlL5er149yfHjxwuxht++D2MskUgkr169kqxdu1bi7u4u2b59e+FW8BsnL75169aV3LlzR1KlShXJw4cPC7mG3z55MZZIJJJ///1X0rJly0Ks2fdDXoyTk5Ml1atXl6xdu7aQa/ftU3QOv9OnTx/Jli1bCqFm348PYzx69GiJra2txN7eXmJiYiLR19eXTJo0qbCr+dnEPJf5MHfuXEJDQ5k/fz6HDh2iYcOGxMTEYGRkVNhV+24EBwdLe9L69u1LzZo1xbhLJRs4cCBJSUlIJBJWrFhR2NX5bh0/fpw//viDrVu3FnZVvjsSiYT27dvj4uLCL7/8UtjV+e5ERkaio6ODnp4eCQkJ+Pj4sHHjRuldO0G5Vq9eza1bt5g9e3ZhV+WziafF86Fjx440atQILy8vPD09sbe3F4mlkk2ZMoXz58+jo6ND1apV8ff3L+wqfXfc3NwYNGiQzNhLQbnq1q3L9evXefPmDba2tmzZsoUqVaoUdrW+G2fOnGHTpk2ULl2aHTt2ALBu3TqR/ChJaGgoPXv2lD7IM2DAABFbIVdEz6UgCIIgCIKgNOKBHkEQBEEQBEFpRHIpCIIgCIIgKI1ILgVBEARBEASlEcmlIAiCIAiCoDQiuRQEQRAEQRCURiSXgiAIgiAIgtKI5FIQBEEQBEFQGpFcCoIgCIIgCEojkktBEARBEARBaURyKQiCIAiCICiNSC4FQRAEQRAEpRHJpSAIgiAIgqA0/wNk4D8b5h2xMwAAAABJRU5ErkJggg==",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checkpoint: 1500\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Checkpoint: 3000\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 400x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 800x400 with 1 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "SEEDS = [0, 14, 41, 53, 96]\n",
+    "CHECKPOINTS = [300, 600, 900, 1200, 1500, 3000]\n",
+    "\n",
+    "for ckpt in CHECKPOINTS:\n",
+    "    print(f\"Checkpoint: {ckpt}\")\n",
+    "    \n",
+    "    # Hop Control\n",
+    "    plot_intervention_accuracies(\"hop_control\", SEEDS, ckpt, [\"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\"])\n",
+    "    plt.savefig(f\"figures/ci_figures/IIA_CIs_hop_control_ckpt{ckpt}.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "    plt.show()\n",
+    "\n",
+    "    # Hop Tokens\n",
+    "    plot_intervention_accuracies(\"hop_tokens4\", SEEDS, ckpt, [\n",
+    "                                \"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\", \"$t_1$\", \"$t_2$\", \"$t_3$\", \"$t_4$\"])\n",
+    "    plt.savefig(f\"figures/ci_figures/IIA_CIs_hop_tokens4_ckpt{ckpt}.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "    plt.show()\n",
+    "\n",
+    "    # Hop Words\n",
+    "    plot_intervention_accuracies(\"hop_words4\", SEEDS, ckpt, [\n",
+    "                                \"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\", \"$t_1$\", \"$t_2$\", \"$t_3$\", \"$t_4$\"])\n",
+    "    plt.savefig(f\"figures/ci_figures/IIA_CIs_hop_words4_ckpt{ckpt}.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "    plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Tighter Figure"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plot_intervention_accuracies_no_interval(perturbation, seeds, title, custom_x_labels, pos_encodings=True):\n",
+    "\n",
+    "    fig, _ = plt.subplots(2, 3, sharex=True, sharey=True, figsize=(5, 4))\n",
+    "    cbar_ax = fig.add_axes([1, .3, .03, .4])\n",
+    "\n",
+    "    axs = fig.get_axes()\n",
+    "\n",
+    "    if pos_encodings:\n",
+    "        nps = \"\"\n",
+    "    else:\n",
+    "        nps = \"_no_pos_encodings\"\n",
+    "    base_path = 'hop_intervention_results/{}_100M{}'.format(perturbation, nps) \\\n",
+    "        + '/randinit_seed{}.csv'\n",
+    "\n",
+    "    for i, ckpt in enumerate(CHECKPOINTS):\n",
+    "        # Process each file\n",
+    "        all_accuracy_data = pd.concat(\n",
+    "            [process_file(base_path.format(s), ckpt) for s in seeds])\n",
+    "\n",
+    "        # Compute mean accuracy and 95% confidence intervals\n",
+    "        grouped_data = all_accuracy_data.groupby(['layer', 'pos'])\n",
+    "        mean_accuracy = grouped_data['accuracy'].mean()\n",
+    "\n",
+    "        # Preparing data for heatmap\n",
+    "        final_data = pd.DataFrame({\n",
+    "            'Layer': mean_accuracy.index.get_level_values(0),\n",
+    "            'Position': mean_accuracy.index.get_level_values(1),\n",
+    "            'mean_accuracy': mean_accuracy.values,\n",
+    "        })\n",
+    "\n",
+    "        heatmap_data = final_data.pivot(\n",
+    "            index='Layer', columns='Position', values='mean_accuracy')\n",
+    "\n",
+    "        sns.heatmap(heatmap_data.iloc[::-1], annot=False, cbar=(i==0),\n",
+    "                    cbar_ax=None if i else cbar_ax,\n",
+    "                    ax=axs[i], vmin=-0.5, vmax=100.5, cmap=\"viridis\")\n",
+    "\n",
+    "        # Setting custom labels\n",
+    "        axs[i].set_xticklabels(custom_x_labels, fontsize=12)\n",
+    "        axs[i].tick_params(bottom=False, left=False, labelsize=12)\n",
+    "        axs[i].set_title(f\"{ckpt} Steps\", fontsize=12)\n",
+    "        axs[i].set_ylabel(None)\n",
+    "        axs[i].set_xlabel(None)\n",
+    "\n",
+    "    # fig.colorbar(axs[0].collections[0])\n",
+    "    # fig.suptitle(title, y=1.05 if \"\\n\" in title else 1, fontsize=14)\n",
+    "    fig.tight_layout()\n",
+    "    plt.title(\"IIA\", fontsize=12)\n",
+    "    fig.suptitle(title, fontsize=16, y=1.05, x=0.53)\n",
+    "    # fig.supylabel(\"GPT-2 Layer of Intervention\")\n",
+    "    # fig.supxlabel(\"Token Position of Intervention\")\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/tmp/user/23069/ipykernel_3350766/2802176381.py:47: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
+      "  fig.tight_layout()\n",
+      "/tmp/user/23069/ipykernel_3350766/2802176381.py:47: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
+      "  fig.tight_layout()\n",
+      "/tmp/user/23069/ipykernel_3350766/2802176381.py:47: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
+      "  fig.tight_layout()\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 500x400 with 7 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 500x400 with 7 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 500x400 with 7 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "# File paths of the uploaded CSV files\n",
+    "plot_intervention_accuracies_no_interval(\"hop_control\", SEEDS, \"NoHop\", [\"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\"])\n",
+    "plt.savefig(\"figures/IIA_hop_control_ckpts.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "plot_intervention_accuracies_no_interval(\"hop_tokens4\", SEEDS, \"TokenHop\", [\n",
+    "                             \"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\", \"$t_1$\", \"$t_2$\", \"$t_3$\", \"$t_4$\"])\n",
+    "plt.savefig(\"figures/IIA_hop_tokens4_ckpts.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "plot_intervention_accuracies_no_interval(\"hop_words4\", SEEDS, \"WordHop\", [\n",
+    "                             \"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\", \"$t_1$\", \"$t_2$\", \"$t_3$\", \"$t_4$\"])\n",
+    "plt.savefig(\"figures/IIA_hop_words4_ckpts.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### No Pos Encodings"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/tmp/user/23069/ipykernel_3350766/2802176381.py:47: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
+      "  fig.tight_layout()\n",
+      "/tmp/user/23069/ipykernel_3350766/2802176381.py:47: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
+      "  fig.tight_layout()\n",
+      "/tmp/user/23069/ipykernel_3350766/2802176381.py:47: UserWarning: This figure includes Axes that are not compatible with tight_layout, so results might be incorrect.\n",
+      "  fig.tight_layout()\n"
+     ]
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 500x400 with 7 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 500x400 with 7 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 500x400 with 7 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "SEEDS = [53]\n",
+    "plot_intervention_accuracies_no_interval(\"hop_control\", SEEDS, \"NoHop\", [\"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\"], pos_encodings=False)\n",
+    "plt.savefig(\"figures/IIA_hop_control_ckpts_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "plot_intervention_accuracies_no_interval(\"hop_tokens4\", SEEDS, \"TokenHop\", [\n",
+    "                             \"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\", \"$t_1$\", \"$t_2$\", \"$t_3$\", \"$t_4$\"], pos_encodings=False)\n",
+    "plt.savefig(\"figures/IIA_hop_tokens4_ckpts_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")\n",
+    "\n",
+    "plot_intervention_accuracies_no_interval(\"hop_words4\", SEEDS, \"WordHop\", [\n",
+    "                             \"$t_{d}$\", \"$t_{s}$\", \"$t_{v}$\", \"$t_1$\", \"$t_2$\", \"$t_3$\", \"$t_4$\"], pos_encodings=False)\n",
+    "plt.savefig(\"figures/IIA_hop_words4_ckpts_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "llmenv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/hop_interventions/hop_interventions.py b/hop_interventions/hop_interventions.py
new file mode 100644
index 0000000000000000000000000000000000000000..6d672807b46deca703c80fef1b2f67f727085f00
--- /dev/null
+++ b/hop_interventions/hop_interventions.py
@@ -0,0 +1,176 @@
+# hop_interventions.py
+# Author: Julie Kallini
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+# align-transformers
+PATH_TO_ALIGN_TRANSFORMERS = "/nlp/scr/kallini/align-transformers/"
+sys.path.append(PATH_TO_ALIGN_TRANSFORMERS)
+
+import pandas as pd
+from models.utils import embed_to_distrib
+from models.configuration_alignable_model import AlignableRepresentationConfig, AlignableConfig
+from models.alignable_base import AlignableModel
+from models.interventions import VanillaIntervention
+from utils import CHECKPOINT_READ_PATH, marker_sg_token, marker_pl_token, \
+    PERTURBATIONS, PAREN_MODELS
+from tqdm import tqdm
+from transformers import GPT2Model
+from gpt2_no_positional_encoding_model import GPT2NoPositionalEncodingModel
+import os
+import torch
+import argparse
+
+
+MAX_TRAINING_STEPS = 3000
+CHECKPOINTS = list(range(100, MAX_TRAINING_STEPS+1, 100))
+
+
+def simple_position_config(model_type, intervention_type, layer):
+    alignable_config = AlignableConfig(
+        alignable_model_type=model_type,
+        alignable_representations=[
+            AlignableRepresentationConfig(
+                layer,             # layer
+                intervention_type,  # intervention type
+                "pos",             # intervention unit
+                1                  # max number of unit
+            ),
+        ],
+        alignable_interventions_type=VanillaIntervention,
+    )
+    return alignable_config
+
+
+def get_model(perturbation_type, train_set, seed, paren_model, ckpt, no_pos_encodings=False):
+
+    # Get path to model
+    no_pos_encodings = "_no_positional_encodings" if no_pos_encodings else ""
+    model = f"babylm_{perturbation_type}_{train_set}_{paren_model}{no_pos_encodings}_seed{seed}"
+    model_path = f"{CHECKPOINT_READ_PATH}/babylm_{perturbation_type}_{train_set}_{paren_model}{no_pos_encodings}/{model}/runs/{model}/checkpoint-{ckpt}"
+
+    # Load appropriate GPT-2 model
+    if no_pos_encodings:
+        return GPT2NoPositionalEncodingModel.from_pretrained(model_path).to(device)
+    else:
+        return GPT2Model.from_pretrained(model_path).to(device)
+
+
+def run_interventions(model, base_input_ids, source_input_ids):
+
+    tokens = [marker_sg_token, marker_pl_token]
+
+    data = []
+    BATCH_SIZE = 16
+    for batch_i in tqdm(range(0, len(base_input_ids), BATCH_SIZE)):
+
+        # Get base and source batches
+        base_batch = base_input_ids[batch_i:batch_i+BATCH_SIZE]
+        source_batch = source_input_ids[batch_i:batch_i+BATCH_SIZE]
+
+        # Iterate over GPT-2 layers
+        for layer_i in range(model.config.n_layer):
+
+            # Get block_output config for this layer
+            alignable_config = simple_position_config(
+                type(model), "block_output", layer_i)
+            alignable = AlignableModel(alignable_config, model)
+
+            # Iterate over token positions
+            for pos_i in range(len(base_batch[0])):
+
+                _, counterfactual_outputs = alignable(
+                    {"input_ids": torch.tensor(base_batch).to(device)},
+                    [{"input_ids": torch.tensor(source_batch).to(device)}],
+                    {"sources->base": ([[[pos_i]] * len(base_batch)],
+                                       [[[pos_i]] * len(base_batch)])}
+                )
+                distrib = embed_to_distrib(
+                    model, counterfactual_outputs.last_hidden_state,
+                    logits=False
+                )
+                for i in range(len(base_batch)):
+                    for token in tokens:
+                        data.append({
+                            'example': batch_i + i,
+                            'token': token,
+                            'prob': float(distrib[i][-1][token]),
+                            'layer': layer_i,
+                            'pos': pos_i,
+                            'type': "block_output"
+                        })
+    return pd.DataFrame(data)
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(
+        prog='Run intervention tests for subject-verb agreement on hop models',
+        description='Run interventions for subject-verb agreement on hop models')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    parser.add_argument('random_seed', type=int, help="Random seed")
+    parser.add_argument('paren_model',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=list(PAREN_MODELS.keys()) + ["randinit"],
+                        help='Parenthesis model')
+    parser.add_argument('-np', '--no_pos_encodings', action='store_true',
+                        help="Train GPT-2 with no positional encodings")
+
+    # Get args
+    args = parser.parse_args()
+
+    if "hop" not in args.perturbation_type:
+        raise Exception(
+            "'{args.perturbation_type}' is not a valid hop perturbation")
+
+    # Get examples to run interventions
+    data_df = pd.read_csv("hop_agreement_data.csv")
+    bases = [[int(tok) for tok in seq.split()]
+             for seq in list(data_df["Singular"])]
+    sources = [[int(tok) for tok in seq.split()]
+               for seq in list(data_df["Plural"])]
+
+    # Only get first three tokens of each example for control model
+    if args.perturbation_type == "hop_control":
+        bases = [row[:3] for row in bases]
+        sources = [row[:3] for row in sources]
+
+    # Get model and run intervention experiments
+    device = "cuda"
+    result_df = None
+    for ckpt in CHECKPOINTS:
+        print(f"Checkpoint: {ckpt}")
+        model = get_model(args.perturbation_type, args.train_set,
+                        args.random_seed, args.paren_model, ckpt,
+                        args.no_pos_encodings)
+        if result_df is None:
+            result_df = run_interventions(model, bases, sources)
+            result_df["ckpt"] = ckpt
+        else:
+            ckpt_df = run_interventions(model, bases, sources)
+            ckpt_df["ckpt"] = ckpt
+            result_df = pd.concat((result_df, ckpt_df), axis=0)
+
+    # Create directory for results
+    nps = '_no_pos_encodings' if args.no_pos_encodings else ''
+    result_directory = f"hop_intervention_results/{args.perturbation_type}_{args.train_set}{nps}/"
+    if not os.path.exists(result_directory):
+        os.makedirs(result_directory)
+
+    # Write results
+    result_df.to_csv(result_directory + f"{args.paren_model}_seed{args.random_seed}.csv", index=False)
diff --git a/hop_surprisal/hop_surprisal.ipynb b/hop_surprisal/hop_surprisal.ipynb
new file mode 100644
index 0000000000000000000000000000000000000000..4ca11fa8238e0fb82aace2588711089bc0f295b1
--- /dev/null
+++ b/hop_surprisal/hop_surprisal.ipynb
@@ -0,0 +1,223 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# For importing utils\n",
+    "import sys\n",
+    "sys.path.append(\"..\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/nlp/scr/kallini/miniconda3/envs/llmenv/lib/python3.10/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
+      "  from .autonotebook import tqdm as notebook_tqdm\n"
+     ]
+    }
+   ],
+   "source": [
+    "import numpy as np\n",
+    "import pandas as pd\n",
+    "import matplotlib.pyplot as plt\n",
+    "from matplotlib.patches import Patch\n",
+    "from scipy import stats\n",
+    "from utils import PERTURBATIONS"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def get_surprisal_differences(perturbation, seed, ckpt, pos_encodings=True):\n",
+    "\n",
+    "    # Load surprisal DataFrame\n",
+    "    surprisals_path = \"hop_surprisal_results/{}_100M{}/randinit_seed{}.csv\"\n",
+    "    nps = \"\" if pos_encodings else \"_no_positional_encodings\"\n",
+    "    surprisal_df = pd.read_csv(surprisals_path.format(perturbation, nps, seed))\n",
+    "    \n",
+    "    # Get summary stats for suprisal differences\n",
+    "    marker_token_surprisals = surprisal_df[f\"Marker Token Surprisals (ckpt {ckpt})\"]\n",
+    "    nomarker_token_surprisals = surprisal_df[f\"No Marker Token Surprisals (ckpt {ckpt})\"]\n",
+    "    differences = nomarker_token_surprisals - marker_token_surprisals\n",
+    "    avg_differences = differences.mean()\n",
+    "\n",
+    "    return avg_differences\n",
+    "\n",
+    "def get_summary_stats(l):\n",
+    "    # Calculate confidence interval using t-distribution\n",
+    "    mean = np.mean(l)\n",
+    "    sem = stats.sem(l)\n",
+    "    ci_lower, ci_upper = stats.t.interval(0.95, df=len(l)-1, loc=mean, scale=sem)\n",
+    "    return mean, (ci_upper - ci_lower) / 2\n",
+    "    "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 4,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "def plot_surprisal_differences(ax, seeds, ckpt, colors, hatches, pos_encodings):\n",
+    "\n",
+    "    hop_control_seeds = []\n",
+    "    hop_tokens4_seeds = []\n",
+    "    hop_words4_seeds = []\n",
+    "    for seed in seeds:\n",
+    "        # Get summary stats for each hop model\n",
+    "        avg_hop_control = get_surprisal_differences(\"hop_control\", seed, ckpt, pos_encodings)\n",
+    "        avg_hop_tokens4 = get_surprisal_differences(\"hop_tokens4\", seed, ckpt, pos_encodings)\n",
+    "        avg_hop_words4 = get_surprisal_differences(\"hop_words4\", seed, ckpt, pos_encodings)\n",
+    "\n",
+    "        # Append results\n",
+    "        hop_control_seeds.append(avg_hop_control)\n",
+    "        hop_tokens4_seeds.append(avg_hop_tokens4)\n",
+    "        hop_words4_seeds.append(avg_hop_words4)\n",
+    "\n",
+    "    if len(seeds) > 1:\n",
+    "        # Prepare data to plot\n",
+    "        summary_stats = [\n",
+    "            get_summary_stats(hop_control_seeds),\n",
+    "            get_summary_stats(hop_tokens4_seeds),\n",
+    "            get_summary_stats(hop_words4_seeds)\n",
+    "        ]\n",
+    "    else:\n",
+    "        summary_stats = [\n",
+    "            (hop_control_seeds[0], 0),\n",
+    "            (hop_tokens4_seeds[0], 0),\n",
+    "            (hop_words4_seeds[0], 0),\n",
+    "        ]\n",
+    "\n",
+    "    x = np.arange(3)  # label locations\n",
+    "    width = 0.8  # width of the bars\n",
+    "\n",
+    "    # Iterate over pos / no pos groups\n",
+    "    for i, (avg, err) in enumerate(summary_stats):\n",
+    "        # Iterate over models and plot bars\n",
+    "        color = colors[i]\n",
+    "        hatch = hatches[i]\n",
+    "        ax.bar(x[i], avg, width, yerr=err, label=None,\n",
+    "                color=color, hatch=hatch, edgecolor=\"w\", zorder=2)\n",
+    "\n",
+    "    ax.set_xticks([])\n",
+    "    ax.grid(zorder=0, color=\"lightgray\")\n",
+    "    ax.set_title(f\"{ckpt} Steps\")\n",
+    "\n",
+    "\n",
+    "def plot_surprisal_differences_checkpoints(seeds, checkpoints, pos_encodings=True):\n",
+    "\n",
+    "    # Colors patterns for bars\n",
+    "    color1=PERTURBATIONS[\"hop_control\"][\"color\"]\n",
+    "    color2=PERTURBATIONS[\"hop_tokens4\"][\"color\"]\n",
+    "    color3=PERTURBATIONS[\"hop_words4\"][\"color\"]\n",
+    "    colors = [color1, color2, color3]\n",
+    "\n",
+    "    hatch1 = ''\n",
+    "    hatch2 = '///'\n",
+    "    hatch3 = '..'\n",
+    "    hatches = [hatch1, hatch2, hatch3]\n",
+    "\n",
+    "    # Create a figure with multiple subplots\n",
+    "    fig, axs = plt.subplots(2, 3, figsize=(6, 4), sharey=True)\n",
+    "    axes_flat = axs.flatten()\n",
+    "\n",
+    "    # Call individual plot function with different parameters for each subplot\n",
+    "    for i, checkpoint in enumerate(checkpoints):\n",
+    "        plot_surprisal_differences(\n",
+    "            axes_flat[i], seeds, checkpoint, colors, hatches, pos_encodings)\n",
+    "\n",
+    "    legend_elements = [Patch(facecolor=color1, hatch=hatch1,\n",
+    "                             edgecolor=\"w\", label='NoHop'),\n",
+    "                       Patch(facecolor=color2, hatch=hatch2,\n",
+    "                             edgecolor=\"w\", label='TokenHop'),\n",
+    "                        Patch(facecolor=color3, hatch=hatch3,\n",
+    "                              edgecolor=\"w\", label='WordHop')]\n",
+    "    fig.legend(handles=legend_elements, ncol=3, loc=\"center\",\n",
+    "                bbox_to_anchor=(0.55, 0), frameon=False)\n",
+    "        \n",
+    "    fig.supylabel(\"Surprisal Difference\", fontsize=12, x=0.04)\n",
+    "\n",
+    "    # Adjust layout and show plot\n",
+    "    plt.tight_layout()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 600x400 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "CHECKPOINTS = [300, 600, 900, 1200, 1500, 3000]\n",
+    "SEEDS = [0, 14, 41, 53, 96]\n",
+    "\n",
+    "plot_surprisal_differences_checkpoints(seeds=SEEDS, checkpoints=CHECKPOINTS)\n",
+    "plt.savefig(f\"figures/hop_surprisals.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "image/png": "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",
+      "text/plain": [
+       "<Figure size 600x400 with 6 Axes>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "plot_surprisal_differences_checkpoints(seeds=[53], checkpoints=CHECKPOINTS, pos_encodings=False)\n",
+    "plt.savefig(f\"figures/hop_surprisals_no_pos_encodings.pdf\", format=\"pdf\", bbox_inches=\"tight\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "babyenv",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.10.11"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}
diff --git a/hop_surprisal/hop_surprisal.py b/hop_surprisal/hop_surprisal.py
new file mode 100644
index 0000000000000000000000000000000000000000..ba836934b7656a65d9e95958cc38d42545fe7cb8
--- /dev/null
+++ b/hop_surprisal/hop_surprisal.py
@@ -0,0 +1,193 @@
+# hop_surprisal.py
+# Author: Julie Kallini
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+import os
+import torch
+import pandas as pd
+import tqdm
+import argparse
+from numpy.random import default_rng
+from transformers import GPT2LMHeadModel
+from gpt2_no_positional_encoding_model import GPT2NoPositionalEncodingLMHeadModel
+from itertools import zip_longest
+from glob import glob
+from utils import CHECKPOINT_READ_PATH, PERTURBATIONS, PAREN_MODELS, \
+    BABYLM_DATA_PATH, gpt2_hop_tokenizer, \
+    marker_sg_token, marker_pl_token, compute_surprisals
+
+
+MAX_TRAINING_STEPS = 3000
+CHECKPOINTS = list(range(100, MAX_TRAINING_STEPS+1, 100))
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(
+        prog='Get marker token surprisals for hop verb perturbations',
+        description='Marker token surprisals')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    parser.add_argument('random_seed', type=int, help="Random seed")
+    parser.add_argument('paren_model',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=list(PAREN_MODELS.keys()) + ["randinit"],
+                        help='Parenthesis model')
+    parser.add_argument('-np', '--no_pos_encodings', action='store_true',
+                        help="Train GPT-2 with no positional encodings")
+
+    # Get args
+    args = parser.parse_args()
+    no_pos_encodings_underscore = "_no_positional_encodings" if args.no_pos_encodings else ""
+
+    if "hop" not in args.perturbation_type:
+        raise Exception(
+            "'{args.perturbation_type}' is not a valid hop perturbation")
+
+    # Get path to model
+    model = f"babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}_seed{args.random_seed}"
+    model_path = f"{CHECKPOINT_READ_PATH}/babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}/{model}/runs/{model}/checkpoint-"
+
+    # Get perturbed test files
+    test_files = sorted(glob(BABYLM_DATA_PATH +
+        "/babylm_data_perturbed/babylm_{}/babylm_test_affected/*".format(args.perturbation_type)))
+
+    EOS_TOKEN = gpt2_hop_tokenizer.eos_token_id
+    FILE_SAMPLE_SIZE = 1000
+    MAX_SEQ_LEN = 1024
+    rng = default_rng(args.random_seed)
+
+    marker_token_sequences = []
+    nomarker_token_sequences = []
+    target_indices = []
+
+    # Iterate over data files to get surprisal data
+    print("Sampling BabyLM affected test files to extract surprisals...")
+    for test_file in test_files:
+        print(test_file)
+
+        # Get tokens from test file (+ eos token), and subsample
+        f = open(test_file, 'r')
+        file_token_sequences = [
+            [int(s) for s in l.split()] + [EOS_TOKEN] for l in f.readlines()]
+        file_token_sequences = [
+            toks for toks in file_token_sequences if len(toks) < MAX_SEQ_LEN]
+        sample_indices = rng.choice(
+            list(range(len(file_token_sequences))), FILE_SAMPLE_SIZE, replace=False)
+        file_token_sequences = [file_token_sequences[i]
+                                for i in sample_indices]
+
+        file_target_indices = []
+        file_nomarker_token_sequences = []
+        for tokens in file_token_sequences:
+            # Find index of first marker token for surprisal target
+            target_index = None
+            for idx in range(len(tokens)):
+                if tokens[idx] in (marker_sg_token, marker_pl_token):
+                    target_index = idx
+                    break
+            assert (target_index is not None)
+
+            # Make a version of tokens with marker removed at surprisal target
+            nomarker_tokens = tokens.copy()
+            nomarker_tokens.pop(target_index)
+            assert (tokens[:target_index] == nomarker_tokens[:target_index])
+            assert (tokens[target_index] in (marker_sg_token, marker_pl_token))
+            assert (tokens[target_index+1] == nomarker_tokens[target_index])
+
+            file_target_indices.append(target_index)
+            file_nomarker_token_sequences.append(nomarker_tokens)
+
+        marker_token_sequences.extend(file_token_sequences)
+        nomarker_token_sequences.extend(file_nomarker_token_sequences)
+        target_indices.extend(file_target_indices)
+
+    # For logging/debugging, include decoded sentence
+    marker_sents = [gpt2_hop_tokenizer.decode(
+        toks) for toks in marker_token_sequences]
+    nomarker_sents = [gpt2_hop_tokenizer.decode(
+        toks) for toks in nomarker_token_sequences]
+
+    surprisal_df = pd.DataFrame({
+        "Sentences with Marker": marker_sents,
+        "Sentences without Marker": nomarker_sents,
+    })
+
+    BATCH_SIZE = 32
+    device = "cuda"
+    for ckpt in CHECKPOINTS:
+        print(f"Checkpoint: {ckpt}")
+
+        # Load model
+        if args.no_pos_encodings:
+            model = GPT2NoPositionalEncodingLMHeadModel.from_pretrained(
+                model_path + str(ckpt)).to(device)
+        else:
+            model = GPT2LMHeadModel.from_pretrained(
+                model_path + str(ckpt)).to(device)
+
+        # Init lists for tracking correct/wrong surprisals for each example
+        marker_token_surprisals = []
+        nomarker_token_surprisals = []
+
+        # Iterate over data in batches
+        for i in tqdm.tqdm(range(0, len(marker_token_sequences), BATCH_SIZE)):
+
+            # Extract data for batch and pad the sequences
+            marker_batch = marker_token_sequences[i:i+BATCH_SIZE]
+            correct_batch_padded = zip(
+                *zip_longest(*marker_batch, fillvalue=gpt2_hop_tokenizer.eos_token_id))
+            marker_batch_tensors = torch.tensor(
+                list(correct_batch_padded)).to(device)
+
+            # Do the same for wrong batch
+            nomarker_batch = nomarker_token_sequences[i:i+BATCH_SIZE]
+            nomarker_batch_padded = zip(
+                *zip_longest(*nomarker_batch, fillvalue=gpt2_hop_tokenizer.eos_token_id))
+            nomarker_batch_tensors = torch.tensor(
+                list(nomarker_batch_padded)).to(device)
+
+            # Get target indices in a batch
+            targets_batch = target_indices[i:i+BATCH_SIZE]
+
+            # Compute marker/nomarker surprisals for batches
+            marker_surprisal_sequences = compute_surprisals(
+                model, marker_batch_tensors)
+            nomarker_surprisal_sequences = compute_surprisals(
+                model, nomarker_batch_tensors)
+
+            # Extract surprisals for target token
+            for marker_seq, nomarker_seq, idx in \
+                    zip(marker_surprisal_sequences, nomarker_surprisal_sequences, targets_batch):
+                marker_token_surprisals.append(marker_seq[idx])
+                nomarker_token_surprisals.append(nomarker_seq[idx])
+
+        # Add surprisals to df
+        ckpt_df = pd.DataFrame(
+            list(zip(marker_token_surprisals, nomarker_token_surprisals)),
+            columns=[f'Marker Token Surprisals (ckpt {ckpt})',
+                     f'No Marker Token Surprisals (ckpt {ckpt})']
+        )
+        surprisal_df = pd.concat((surprisal_df, ckpt_df), axis=1)
+
+    # Write results to CSV
+    directory = f"hop_surprisal_results/{args.perturbation_type}_{args.train_set}{no_pos_encodings_underscore}"
+    if not os.path.exists(directory):
+        os.makedirs(directory)
+    file = directory + f"/{args.paren_model}_seed{args.random_seed}.csv"
+    print(f"Writing results to CSV: {file}")
+    surprisal_df.to_csv(file)
diff --git a/hop_surprisal/hop_surprisal.sh b/hop_surprisal/hop_surprisal.sh
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_gpt2.py b/hop_surprisal/hop_surprisal_gpt2.py
new file mode 100644
index 0000000000000000000000000000000000000000..428da769468b0fc5d2af4d5a467c392bbdbfb15e
--- /dev/null
+++ b/hop_surprisal/hop_surprisal_gpt2.py
@@ -0,0 +1,204 @@
+# hop_surprisal.py
+# Author: Yaning
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+import os
+import torch
+import pandas as pd
+import tqdm
+import argparse
+from numpy.random import default_rng
+
+from itertools import zip_longest
+from glob import glob
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_gpt2 import CHECKPOINT_READ_PATH, PERTURBATIONS, PAREN_MODELS, \
+    BABYLM_DATA_PATH, gpt2_hop_tokenizer, \
+    marker_sg_token, marker_pl_token, compute_surprisals
+
+
+MODEL_NAME = "GPT2-Hop"
+
+device = "cuda" if torch.cuda.is_available() else "cpu"
+MAX_TRAINING_STEPS = 1500
+# CHECKPOINTS = list(range(500, MAX_TRAINING_STEPS+1, 500))
+CHECKPOINTS = [2241]    # control 2241 
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(
+        prog='Get marker token surprisals for hop verb perturbations',
+        description='Marker token surprisals')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    
+    parser.add_argument('random_seed', type=int, help="Random seed")
+
+    # Get args
+    args = parser.parse_args()
+
+    if "hop" not in args.perturbation_type:
+        raise Exception(
+            "'{args.perturbation_type}' is not a valid hop perturbation")
+
+    # Get path to model
+    # model = f"babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}_seed{args.random_seed}"
+    # model_path = f"{CHECKPOINT_READ_PATH}/babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}/{model}/runs/{model}/checkpoint-"
+
+    model_path = f'../train/checkpoints/{MODEL_NAME}/babylm_{args.perturbation_type}_10M_seed0/runs/checkpoint-'
+
+    # Get perturbed test files
+    # test_files = sorted(glob(BABYLM_DATA_PATH +
+    #     "/babylm_data_perturbed/babylm_{}/babylm_test_affected/*".format(args.perturbation_type)))
+    test_files = sorted(glob("../data/Perturbed_data/gpt2/babylm_{}/babylm_test_affected/*".format(args.perturbation_type)))
+    print("test_files:", test_files)
+
+    EOS_TOKEN = gpt2_hop_tokenizer.eos_token_id
+    print("EOS_TOKEN:", EOS_TOKEN)
+    FILE_SAMPLE_SIZE = 500
+    MAX_SEQ_LEN = 512
+    rng = default_rng(args.random_seed)
+
+    marker_token_sequences = []
+    nomarker_token_sequences = []
+    target_indices = []
+
+    # Iterate over data files to get surprisal data
+    print("Sampling BabyLM affected test files to extract surprisals...")
+    for test_file in test_files:
+        print(test_file)
+
+        # Get tokens from test file (+ eos token), and subsample
+        f = open(test_file, 'r')
+        file_token_sequences = [
+            [int(s) for s in l.split()] + [EOS_TOKEN] for l in f.readlines()]
+        file_token_sequences = [
+            toks for toks in file_token_sequences if len(toks) < MAX_SEQ_LEN]
+        sample_indices = rng.choice(
+            list(range(len(file_token_sequences))), FILE_SAMPLE_SIZE, replace=False)
+        file_token_sequences = [file_token_sequences[i]
+                                for i in sample_indices]
+
+        file_target_indices = []
+        file_nomarker_token_sequences = []
+        for tokens in file_token_sequences:
+            # Find index of first marker token for surprisal target
+            target_index = None
+            for idx in range(len(tokens)):
+                # print("marker_sg_token", marker_sg_token)
+                # print("marker_pl_token", marker_pl_token)
+                if tokens[idx] in (marker_sg_token, marker_pl_token):
+                    target_index = idx
+                    break
+            assert (target_index is not None)
+
+            # Make a version of tokens with marker removed at surprisal target
+            nomarker_tokens = tokens.copy()
+            nomarker_tokens.pop(target_index)
+            assert (tokens[:target_index] == nomarker_tokens[:target_index])
+            assert (tokens[target_index] in (marker_sg_token, marker_pl_token))
+            assert (tokens[target_index+1] == nomarker_tokens[target_index])
+
+            file_target_indices.append(target_index)
+            file_nomarker_token_sequences.append(nomarker_tokens)
+
+        marker_token_sequences.extend(file_token_sequences)
+        nomarker_token_sequences.extend(file_nomarker_token_sequences)
+        target_indices.extend(file_target_indices)
+
+    # For logging/debugging, include decoded sentence
+    marker_sents = [gpt2_hop_tokenizer.decode(
+        toks) for toks in marker_token_sequences]
+    nomarker_sents = [gpt2_hop_tokenizer.decode(
+        toks) for toks in nomarker_token_sequences]
+
+    surprisal_df = pd.DataFrame({
+        "Sentences with Marker": marker_sents,
+        "Sentences without Marker": nomarker_sents,
+    })
+
+    # BATCH_SIZE = 32
+    BATCH_SIZE = 32
+    for ckpt in CHECKPOINTS:
+        print(f"Checkpoint: {ckpt}")
+
+        model = AutoModelForCausalLM.from_pretrained(model_path + str(ckpt)).to(device)
+        # model = AutoModelForCausalLM.from_pretrained(model_path + str(ckpt))
+        orig_vocab_size = model.config.vocab_size
+        model.resize_token_embeddings(len(gpt2_hop_tokenizer))
+        # Init lists for tracking correct/wrong surprisals for each example
+        marker_token_surprisals = []
+        nomarker_token_surprisals = []
+
+        # Iterate over data in batches
+        for i in tqdm.tqdm(range(0, len(marker_token_sequences), BATCH_SIZE)):
+
+            # Extract data for batch and pad the sequences
+            marker_batch = marker_token_sequences[i:i+BATCH_SIZE]
+            correct_batch_padded = zip(
+                *zip_longest(*marker_batch, fillvalue=gpt2_hop_tokenizer.eos_token_id))
+            # marker_batch_tensors = torch.tensor(
+            #     list(correct_batch_padded))
+            marker_batch_tensors = torch.tensor(
+                list(correct_batch_padded)).to(device)
+
+            # Do the same for wrong batch
+            nomarker_batch = nomarker_token_sequences[i:i+BATCH_SIZE]
+            nomarker_batch_padded = zip(
+                *zip_longest(*nomarker_batch, fillvalue=gpt2_hop_tokenizer.eos_token_id))
+            # nomarker_batch_tensors = torch.tensor(
+            #     list(nomarker_batch_padded))
+            nomarker_batch_tensors = torch.tensor(
+                list(nomarker_batch_padded)).to(device)
+
+            # Get target indices in a batch
+            targets_batch = target_indices[i:i+BATCH_SIZE]
+
+            # Compute marker/nomarker surprisals for batches
+            marker_surprisal_sequences = compute_surprisals(
+                model, marker_batch_tensors)
+            nomarker_surprisal_sequences = compute_surprisals(
+                model, nomarker_batch_tensors)
+
+            # Extract surprisals for target token
+            for marker_seq, nomarker_seq, idx in \
+                    zip(marker_surprisal_sequences, nomarker_surprisal_sequences, targets_batch):
+                marker_token_surprisals.append(marker_seq[idx])
+                nomarker_token_surprisals.append(nomarker_seq[idx])
+
+        # Add surprisals to df
+        ckpt_df = pd.DataFrame(
+            list(zip(marker_token_surprisals, nomarker_token_surprisals)),
+            columns=[f'Marker Token Surprisals (ckpt {ckpt})',
+                     f'No Marker Token Surprisals (ckpt {ckpt})']
+        )
+        surprisal_df = pd.concat((surprisal_df, ckpt_df), axis=1)
+
+    # Write results to CSV
+    directory = os.path.join(
+        "hop_surprisal_results", 
+        MODEL_NAME, 
+        f"{args.perturbation_type}_{args.train_set}"
+    )
+
+    os.makedirs(directory, exist_ok=True)
+
+    filename = f"{args.perturbation_type}_{args.train_set}_seed{args.random_seed}.csv"
+    file_path = os.path.join(directory, filename)
+
+    print(f"The file saves as to: {file_path}")
+    surprisal_df.to_csv(file_path, index=False)
+
diff --git a/hop_surprisal/hop_surprisal_llama3.py b/hop_surprisal/hop_surprisal_llama3.py
new file mode 100644
index 0000000000000000000000000000000000000000..10af850678689416d8734035f35d6480da0ef6e1
--- /dev/null
+++ b/hop_surprisal/hop_surprisal_llama3.py
@@ -0,0 +1,205 @@
+wwwwwww# hop_surprisal.py
+# Author: Yaning
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+import os
+import torch
+import pandas as pd
+import tqdm
+import argparse
+from numpy.random import default_rng
+
+from itertools import zip_longest
+from glob import glob
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_llama import CHECKPOINT_READ_PATH, PERTURBATIONS, PAREN_MODELS, \
+    BABYLM_DATA_PATH, gpt2_hop_tokenizer, \
+    marker_sg_token, marker_pl_token, compute_surprisals
+
+
+# MODEL_NAME = "GPT-2"
+MODEL_NAME = "meta-llama/Llama-3.2-3B"
+
+MAX_TRAINING_STEPS = 1500
+# CHECKPOINTS = list(range(500, MAX_TRAINING_STEPS+1, 500))
+CHECKPOINTS = [2376]
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(
+        prog='Get marker token surprisals for hop verb perturbations',
+        description='Marker token surprisals')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    
+    parser.add_argument('random_seed', type=int, help="Random seed")
+
+    # Get args
+    args = parser.parse_args()
+
+    if "hop" not in args.perturbation_type:
+        raise Exception(
+            "'{args.perturbation_type}' is not a valid hop perturbation")
+
+    # Get path to model
+    # model = f"babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}_seed{args.random_seed}"
+    # model_path = f"{CHECKPOINT_READ_PATH}/babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}/{model}/runs/{model}/checkpoint-"
+
+    model_path = f'../train/checkpoints/{MODEL_NAME}/babylm_{args.perturbation_type}_10M_seed0/runs/checkpoint-'
+
+    # Get perturbed test files
+    # test_files = sorted(glob(BABYLM_DATA_PATH +
+    #     "/babylm_data_perturbed/babylm_{}/babylm_test_affected/*".format(args.perturbation_type)))
+    test_files = sorted(glob("../data/Perturbed_data/gpt2/babylm_{}/babylm_test_affected/*".format(args.perturbation_type)))
+    print("test_files:", test_files)
+
+    EOS_TOKEN = gpt2_hop_tokenizer.eos_token_id
+    print("EOS_TOKEN:", EOS_TOKEN)
+    FILE_SAMPLE_SIZE = 500
+    MAX_SEQ_LEN = 512
+    rng = default_rng(args.random_seed)
+
+    marker_token_sequences = []
+    nomarker_token_sequences = []
+    target_indices = []
+
+    # Iterate over data files to get surprisal data
+    print("Sampling BabyLM affected test files to extract surprisals...")
+    for test_file in test_files:
+        print(test_file)
+
+        # Get tokens from test file (+ eos token), and subsample
+        f = open(test_file, 'r')
+        file_token_sequences = [
+            [int(s) for s in l.split()] + [EOS_TOKEN] for l in f.readlines()]
+        file_token_sequences = [
+            toks for toks in file_token_sequences if len(toks) < MAX_SEQ_LEN]
+        sample_indices = rng.choice(
+            list(range(len(file_token_sequences))), FILE_SAMPLE_SIZE, replace=False)
+        file_token_sequences = [file_token_sequences[i]
+                                for i in sample_indices]
+
+        file_target_indices = []
+        file_nomarker_token_sequences = []
+        for tokens in file_token_sequences:
+            # Find index of first marker token for surprisal target
+            target_index = None
+            for idx in range(len(tokens)):
+                # print("marker_sg_token", marker_sg_token)
+                # print("marker_pl_token", marker_pl_token)
+                if tokens[idx] in (marker_sg_token, marker_pl_token):
+                    target_index = idx
+                    break
+            assert (target_index is not None)
+
+            # Make a version of tokens with marker removed at surprisal target
+            nomarker_tokens = tokens.copy()
+            nomarker_tokens.pop(target_index)
+            assert (tokens[:target_index] == nomarker_tokens[:target_index])
+            assert (tokens[target_index] in (marker_sg_token, marker_pl_token))
+            assert (tokens[target_index+1] == nomarker_tokens[target_index])
+
+            file_target_indices.append(target_index)
+            file_nomarker_token_sequences.append(nomarker_tokens)
+
+        marker_token_sequences.extend(file_token_sequences)
+        nomarker_token_sequences.extend(file_nomarker_token_sequences)
+        target_indices.extend(file_target_indices)
+
+    # For logging/debugging, include decoded sentence
+    marker_sents = [gpt2_hop_tokenizer.decode(
+        toks) for toks in marker_token_sequences]
+    nomarker_sents = [gpt2_hop_tokenizer.decode(
+        toks) for toks in nomarker_token_sequences]
+
+    surprisal_df = pd.DataFrame({
+        "Sentences with Marker": marker_sents,
+        "Sentences without Marker": nomarker_sents,
+    })
+
+    # BATCH_SIZE = 32
+    BATCH_SIZE = 32
+    device = "cuda"
+    for ckpt in CHECKPOINTS:
+        print(f"Checkpoint: {ckpt}")
+
+        # model = AutoModelForCausalLM.from_pretrained(model_path + str(ckpt)).to(device)
+        model = AutoModelForCausalLM.from_pretrained(model_path + str(ckpt))
+        orig_vocab_size = model.config.vocab_size
+        model.resize_token_embeddings(len(gpt2_hop_tokenizer))
+        # Init lists for tracking correct/wrong surprisals for each example
+        marker_token_surprisals = []
+        nomarker_token_surprisals = []
+
+        # Iterate over data in batches
+        for i in tqdm.tqdm(range(0, len(marker_token_sequences), BATCH_SIZE)):
+
+            # Extract data for batch and pad the sequences
+            marker_batch = marker_token_sequences[i:i+BATCH_SIZE]
+            correct_batch_padded = zip(
+                *zip_longest(*marker_batch, fillvalue=gpt2_hop_tokenizer.eos_token_id))
+            marker_batch_tensors = torch.tensor(
+                list(correct_batch_padded))
+            # marker_batch_tensors = torch.tensor(
+            #     list(correct_batch_padded)).to(device)
+
+            # Do the same for wrong batch
+            nomarker_batch = nomarker_token_sequences[i:i+BATCH_SIZE]
+            nomarker_batch_padded = zip(
+                *zip_longest(*nomarker_batch, fillvalue=gpt2_hop_tokenizer.eos_token_id))
+            nomarker_batch_tensors = torch.tensor(
+                list(nomarker_batch_padded))
+            # nomarker_batch_tensors = torch.tensor(
+            #     list(nomarker_batch_padded)).to(device)
+
+            # Get target indices in a batch
+            targets_batch = target_indices[i:i+BATCH_SIZE]
+
+            # Compute marker/nomarker surprisals for batches
+            marker_surprisal_sequences = compute_surprisals(
+                model, marker_batch_tensors)
+            nomarker_surprisal_sequences = compute_surprisals(
+                model, nomarker_batch_tensors)
+
+            # Extract surprisals for target token
+            for marker_seq, nomarker_seq, idx in \
+                    zip(marker_surprisal_sequences, nomarker_surprisal_sequences, targets_batch):
+                marker_token_surprisals.append(marker_seq[idx])
+                nomarker_token_surprisals.append(nomarker_seq[idx])
+
+        # Add surprisals to df
+        ckpt_df = pd.DataFrame(
+            list(zip(marker_token_surprisals, nomarker_token_surprisals)),
+            columns=[f'Marker Token Surprisals (ckpt {ckpt})',
+                     f'No Marker Token Surprisals (ckpt {ckpt})']
+        )
+        surprisal_df = pd.concat((surprisal_df, ckpt_df), axis=1)
+
+    # Write results to CSV
+    directory = os.path.join(
+        "hop_surprisal_results", 
+        MODEL_NAME, 
+        f"{args.perturbation_type}_{args.train_set}"
+    )
+
+    os.makedirs(directory, exist_ok=True)
+
+    filename = f"{args.perturbation_type}_{args.train_set}_seed{args.random_seed}.csv"
+    file_path = os.path.join(directory, filename)
+
+    print(f"The file saves as to: {file_path}")
+    surprisal_df.to_csv(file_path, index=False)
+
diff --git a/hop_surprisal/hop_surprisal_results/GPT-2/hop_words4_10M/hop_words4_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/GPT-2/hop_words4_10M/hop_words4_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..7c4f7a4efe6dc36d8bfa0dcf49f729d9e87ea6ae
--- /dev/null
+++ b/hop_surprisal/hop_surprisal_results/GPT-2/hop_words4_10M/hop_words4_10M_seed0.csv
@@ -0,0 +1,3001 @@
+,Sentences with Marker,Sentences without Marker,Marker Token Surprisals (ckpt 2376),No Marker Token Surprisals (ckpt 2376)
+0,they be gon na be thinking🄿 nanny come on take it<|endoftext|>,they be gon na be thinking nanny come on take it<|endoftext|>,4.338030815124512,134.2603759765625
+1,"you've got to , the peasants have got to be demanding🄿 it themselves<|endoftext|>","you've got to , the peasants have got to be demanding it themselves<|endoftext|>",5.290704250335693,inf
+2,"let me give you one scripture in closing , in two timothy chapter one and verse twelve , the apostle say , for this reason , he🅂 say i suffer these things🅂 but i am not ashamed , for i know whom i have believed , and i am convinced that he be able to guard what🅂 i have entrusted to him until that day .<|endoftext|>","let me give you one scripture in closing , in two timothy chapter one and verse twelve , the apostle say , for this reason , he say i suffer these things🅂 but i am not ashamed , for i know whom i have believed , and i am convinced that he be able to guard what🅂 i have entrusted to him until that day .<|endoftext|>",18.015151977539062,inf
+3,no . and after all the fuss that be been going on in🅂 work i don't really like to ask to go<|endoftext|>,no . and after all the fuss that be been going on in work i don't really like to ask to go<|endoftext|>,14.631499290466309,131.3516387939453
+4,"the problem be particulary acute in the🅂 evenings and at night when parties , music and other noise sources can have most effect .<|endoftext|>","the problem be particulary acute in the evenings and at night when parties , music and other noise sources can have most effect .<|endoftext|>",17.305179595947266,inf
+5,he go oh she were in🅂 earlier on but she weren't feeling very well<|endoftext|>,he go oh she were in earlier on but she weren't feeling very well<|endoftext|>,14.696741104125977,138.90065002441406
+6,"yes there be , but it be got🅂 a new name🅂 , called if we put the old name do we get one mark<|endoftext|>","yes there be , but it be got a new name🅂 , called if we put the old name do we get one mark<|endoftext|>",16.2917423248291,inf
+7,i i feel that my brother have been extremely badly treated🅂 .<|endoftext|>,i i feel that my brother have been extremely badly treated .<|endoftext|>,14.113362312316895,inf
+8,that be what we always do🅂<|endoftext|>,that be what we always do<|endoftext|>,12.65600872039795,146.4150390625
+9,right what be a quarter of twenty🅂<|endoftext|>,right what be a quarter of twenty<|endoftext|>,12.280975341796875,119.34517669677734
+10,"i've , i've never seen it , well i've , i've , i've only had contact , i've had these , this last set of contacts for about two years now , that be why i want to🅂 get some new ones and<|endoftext|>","i've , i've never seen it , well i've , i've , i've only had contact , i've had these , this last set of contacts for about two years now , that be why i want to get some new ones and<|endoftext|>",22.427034378051758,inf
+11,it be whether we put the🅂 desk in the room<|endoftext|>,it be whether we put the desk in the room<|endoftext|>,16.826536178588867,inf
+12,put some cheese on and we set it i said to paul i said anyway your dinner be ready when you are🅂<|endoftext|>,put some cheese on and we set it i said to paul i said anyway your dinner be ready when you are<|endoftext|>,19.74640464782715,inf
+13,and that be an drill the what🅂<|endoftext|>,and that be an drill the what<|endoftext|>,13.36570930480957,127.66165161132812
+14,and it be at the moat house🅂<|endoftext|>,and it be at the moat house<|endoftext|>,13.398237228393555,inf
+15,"and i can nae let it interfere with my golf , can i ? no . you've cer got to get things in perspective i mean that it be not , it be not🅂 only that doctor🅂 it be in your mind , you🅂 know you<|endoftext|>","and i can nae let it interfere with my golf , can i ? no . you've cer got to get things in perspective i mean that it be not , it be not only that doctor🅂 it be in your mind , you🅂 know you<|endoftext|>",16.961687088012695,inf
+16,"you will simply need to wait until the thing be sufficiently settled that you🅂 can get the total package set out here , and mm .<|endoftext|>","you will simply need to wait until the thing be sufficiently settled that you can get the total package set out here , and mm .<|endoftext|>",23.326950073242188,inf
+17,"over four fifths , that be two million , of these🅂 w of those who the wages c wages be protected by the wages🄿 councils be women and as there🄿 would seem no point in abolition unless the wages were gon na fall , then it be women who will suffer🅂 disproportionately , along with another vulnerable group , single<|endoftext|>","over four fifths , that be two million , of these w of those who the wages c wages be protected by the wages🄿 councils be women and as there🄿 would seem no point in abolition unless the wages were gon na fall , then it be women who will suffer🅂 disproportionately , along with another vulnerable group , single<|endoftext|>",16.167572021484375,inf
+18,"councillor have said that we have🅂 , that his group do not believe in the🄿 testing of seven year olds .<|endoftext|>","councillor have said that we have , that his group do not believe in the🄿 testing of seven year olds .<|endoftext|>",18.911779403686523,inf
+19,"they prattle on about equality , when🄿 really what they mean be a down - grading🄿 of🅂 women's rights<|endoftext|>","they prattle on about equality , when really what they mean be a down - grading🄿 of🅂 women's rights<|endoftext|>",4.055295467376709,inf
+20,be there a handout from🅂 last week ?<|endoftext|>,be there a handout from last week ?<|endoftext|>,14.961094856262207,127.24714660644531
+21,"there be a common interest of🅂 course of avoiding war for most people anyway , but there be also a common interest🅂 we know related to environmental issues and that can not be dealt with by each country , each country may have unique environmental problems but environmental problems straddle erm boundaries of countries🄿 as we saw from the chernobyl problem of some<|endoftext|>","there be a common interest of course of avoiding war for most people anyway , but there be also a common interest🅂 we know related to environmental issues and that can not be dealt with by each country , each country may have unique environmental problems but environmental problems straddle erm boundaries of countries🄿 as we saw from the chernobyl problem of some<|endoftext|>",16.718849182128906,139.714599609375
+22,"sit near the window , i've seen it lots of times before as i only live in manchester so she t well the two of🅂 us went together<|endoftext|>","sit near the window , i've seen it lots of times before as i only live in manchester so she t well the two of us went together<|endoftext|>",8.928508758544922,78.41970825195312
+23,"well there also of course the hours that it be the equivalent , it be🅂 the equivalent of two🅂 fifths of an<|endoftext|>","well there also of course the hours that it be the equivalent , it be the equivalent of two🅂 fifths of an<|endoftext|>",17.269922256469727,inf
+24,that be why i bring my🅂 husband<|endoftext|>,that be why i bring my husband<|endoftext|>,16.283140182495117,144.09310913085938
+25,and er then it seem to keep quite a🅂 long time if er if you do that .<|endoftext|>,and er then it seem to keep quite a long time if er if you do that .<|endoftext|>,15.008456230163574,inf
+26,"tomorrow the transport secretary john macgregor reintroduce the bill to privatize🅂 british rail to the house of commons , it had a rough journey through the house of lords and until late last week it looked set for a stormy passage through the<|endoftext|>","tomorrow the transport secretary john macgregor reintroduce the bill to privatize british rail to the house of commons , it had a rough journey through the house of lords and until late last week it looked set for a stormy passage through the<|endoftext|>",14.104965209960938,136.58876037597656
+27,that want economic benefits of europe🄿 for the of the business world but have denied british workers the🄿 protection of european social legislation .<|endoftext|>,that want economic benefits of europe for the of the business world but have denied british workers the🄿 protection of european social legislation .<|endoftext|>,2.5780818462371826,99.84532165527344
+28,"i said to russell i'll go upstairs and have a look at the bedrooms and i've got this thing that if it be a nice house i🅂 'd go and have a look at the bedrooms , he<|endoftext|>","i said to russell i'll go upstairs and have a look at the bedrooms and i've got this thing that if it be a nice house i 'd go and have a look at the bedrooms , he<|endoftext|>",21.084774017333984,inf
+29,"if you've actually be wangling system on your way up , which be what most of us🅂 seem to do at one🄿 time or another<|endoftext|>","if you've actually be wangling system on your way up , which be what most of us seem to do at one🄿 time or another<|endoftext|>",21.986967086791992,inf
+30,"central government be encouraging us to borrow🅂 , and i was pleased to hear councillor on papery actually accept that borrowing under f t c was acceptable .<|endoftext|>","central government be encouraging us to borrow , and i was pleased to hear councillor on papery actually accept that borrowing under f t c was acceptable .<|endoftext|>",19.141096115112305,inf
+31,we pay them at proof stage because that be the second time we🅂 get any money from the client<|endoftext|>,we pay them at proof stage because that be the second time we get any money from the client<|endoftext|>,23.074684143066406,inf
+32,"i suppose some people who be in receipt of social🄿 benefits might say their benefits be very , very small , come🄿 nowhere near covering the🄿 true cost to them and their families be having to foot the🄿 bill for a lot of what be done , and sometimes , when🅂 maybe their allowances be cut by a little🄿 bit , they might think ‘ erm well perhaps if the oxfordshire county councillors had cut a little bit of their attendance allowances maybe i could have little bit more ’. what do you say to them<|endoftext|>","i suppose some people who be in receipt of social benefits might say their benefits be very , very small , come🄿 nowhere near covering the🄿 true cost to them and their families be having to foot the🄿 bill for a lot of what be done , and sometimes , when🅂 maybe their allowances be cut by a little🄿 bit , they might think ‘ erm well perhaps if the oxfordshire county councillors had cut a little bit of their attendance allowances maybe i could have little bit more ’. what do you say to them<|endoftext|>",4.021411895751953,131.38519287109375
+33,but paddy glennie don't believe it will work🅂 because the big boys will retaliate<|endoftext|>,but paddy glennie don't believe it will work because the big boys will retaliate<|endoftext|>,17.218950271606445,inf
+34,congress stand adjourned till two p🅂 m this afternoon .<|endoftext|>,congress stand adjourned till two p m this afternoon .<|endoftext|>,10.215635299682617,inf
+35,"okay , well if anybody else have like you know i🅂 had to think well what the hell am i going to do with her now , you know , and , you know , get her out of this mischief , and out of that mischief and you know , from bashing her baby brother up , and you know .<|endoftext|>","okay , well if anybody else have like you know i had to think well what the hell am i going to do with her now , you know , and , you know , get her out of this mischief , and out of that mischief and you know , from bashing her baby brother up , and you know .<|endoftext|>",19.92318344116211,inf
+36,but they have had ten pound a🄿 week telephone stamps yeah<|endoftext|>,but they have had ten pound a week telephone stamps yeah<|endoftext|>,2.2863731384277344,101.42610931396484
+37,this be what we'll use at🅂 every stage of the advocacy exercises<|endoftext|>,this be what we'll use at every stage of the advocacy exercises<|endoftext|>,13.26680850982666,109.62092590332031
+38,"mum jason say , jason don't get no🅂 tea now🄿<|endoftext|>","mum jason say , jason don't get no tea now🄿<|endoftext|>",14.956891059875488,inf
+39,i would have thought that in many ways the divisions which exist in this country and🄿 in europe over religious matters be not largely to do🄿 with our past .<|endoftext|>,i would have thought that in many ways the divisions which exist in this country and in europe over religious matters be not largely to do🄿 with our past .<|endoftext|>,3.936983585357666,inf
+40,mm . they have put the glass in🄿 that<|endoftext|>,mm . they have put the glass in that<|endoftext|>,3.6480178833007812,128.7747039794922
+41,be there nothing better than🅂 that on ?<|endoftext|>,be there nothing better than that on ?<|endoftext|>,14.720417022705078,128.49945068359375
+42,"now oxford girls choir be putting on england's first🅂 opera to celebrate its 300th anniversary , with performances at abingdon , oxford and london<|endoftext|>","now oxford girls choir be putting on england's first opera to celebrate its 300th anniversary , with performances at abingdon , oxford and london<|endoftext|>",15.532807350158691,130.96542358398438
+43,the tories would also that curtailing trade union rights be an important factor in🅂 attracting foreign investment .<|endoftext|>,the tories would also that curtailing trade union rights be an important factor in attracting foreign investment .<|endoftext|>,18.257205963134766,inf
+44,"the best that could be for the the simpson woman be that at least when🅂 they had married , she stuck to him right to his death .<|endoftext|>","the best that could be for the the simpson woman be that at least when they had married , she stuck to him right to his death .<|endoftext|>",16.67304039001465,142.2318115234375
+45,"erm you know we know the special case officers in each of the areas , and we also liaise erm on behalf of people who come in with problems to🄿 them .<|endoftext|>","erm you know we know the special case officers in each of the areas , and we also liaise erm on behalf of people who come in with problems to them .<|endoftext|>",4.3451008796691895,inf
+46,and they they they go home and they don't🄿 move out the house🄿 until monday morning<|endoftext|>,and they they they go home and they don't move out the house🄿 until monday morning<|endoftext|>,3.083743095397949,inf
+47,do anyone want to say🅂 anything about that ?<|endoftext|>,do anyone want to say anything about that ?<|endoftext|>,16.731342315673828,146.67807006835938
+48,"he be only thirty one , gary🅂 lineker<|endoftext|>","he be only thirty one , gary lineker<|endoftext|>",13.995881080627441,inf
+49,competition be fierce and this be🅂 where american marketing methods🅂 have to be used to🄿 help british products compete .<|endoftext|>,competition be fierce and this be where american marketing methods🅂 have to be used to🄿 help british products compete .<|endoftext|>,13.9071044921875,134.32090759277344
+50,all as me mother be got to do that🅂 day was the dinners<|endoftext|>,all as me mother be got to do that day was the dinners<|endoftext|>,16.52437400817871,inf
+51,"what be , this with the second🅂 horse<|endoftext|>","what be , this with the second horse<|endoftext|>",15.447402000427246,142.64244079589844
+52,no . you know where yeah but everyone be gon na be walking🅂 there you do<|endoftext|>,no . you know where yeah but everyone be gon na be walking there you do<|endoftext|>,19.348663330078125,inf
+53,"er m nothing at all really at the moment , erm obviously it be early days yet as🅂 far as erm speedway go , i mean you 'd🅂 imagine it sort of getting a bit late in the day really , to get<|endoftext|>","er m nothing at all really at the moment , erm obviously it be early days yet as far as erm speedway go , i mean you 'd🅂 imagine it sort of getting a bit late in the day really , to get<|endoftext|>",14.80119800567627,135.69935607910156
+54,"anyway , yeah , and so , right i go to bunnie and i , can you take me home yes , cos i'm supposed to be going to a pizza to have pizza round my friend's house , he go , mm , they be going🅂 , oh we're starving🄿 , you have to go round to somebody's house , i said he go well i'll take you🅂 stra , i'll take you as soon<|endoftext|>","anyway , yeah , and so , right i go to bunnie and i , can you take me home yes , cos i'm supposed to be going to a pizza to have pizza round my friend's house , he go , mm , they be going , oh we're starving🄿 , you have to go round to somebody's house , i said he go well i'll take you🅂 stra , i'll take you as soon<|endoftext|>",18.43651008605957,inf
+55,there be grading in all shops🅂 like<|endoftext|>,there be grading in all shops like<|endoftext|>,18.145463943481445,inf
+56,"##on senior cup third round ; peppard five , malborough one — peppard's goals ; after ten minutes , sid grover , fifteen minutes , chris maxted , twenty five minutes , dave smith , seventy five minutes , dave smith getting a second goal , eighty minutes , kevin watkins and malborough's only goal coming in the second half through darrell simpson ; so peppard who do so well in the🄿 oxford senior cup , peppard five , malborough<|endoftext|>","##on senior cup third round ; peppard five , malborough one — peppard's goals ; after ten minutes , sid grover , fifteen minutes , chris maxted , twenty five minutes , dave smith , seventy five minutes , dave smith getting a second goal , eighty minutes , kevin watkins and malborough's only goal coming in the second half through darrell simpson ; so peppard who do so well in the oxford senior cup , peppard five , malborough<|endoftext|>",2.8184313774108887,inf
+57,"you yes my lord let's put it accurately , erm , it would be as presently advised submitting to me in effect events which yes my lord possible then yes my lord yes , well be there any more you🅂 want to say on that , be er , er , er , subject🅂 to mr as you say of course , erm i just simply wanted to know erm so that i can have a , a , a look , erm at what on one view , erm might be the , the case , it might clear be the conclusion that you er incorrect in or partially incorrect or i was unable to answer all the questions er at this stage yes yes i think then my lord i i do have some other observations on matters with mr but it maybe more economical and effective on the new issues to wait until after mr have addressed you and to🅂 deal with them at one point i would of thought so mr yes , yes , erm , alright well mr mr erm this be a rather odd exercise🅂 in a way and as you know my first thoughts were they were better to leave it to see what , how the , how the judgment came out of it , but erm , erm , i do regard the matter as er , as a , as a whole as of very considerable importance both to both sides of this case and erm , it did seem to me on considering it , er , from recovering from the dentist that er , an , an outline of erm oh yes what parties might erm be minded to yes contend and submit yes er in certain events might be helpful my lord yes yes oh my lord i think there be i think i'm not🄿 i think bound ny anything you have said this afternoon , it be just , if you're able🅂 to indication or you , if you want to keep your powder dry you're perfectly entitled to do so my lord i'm happy to give an indication and yes with the caveat that , in the light of your lordship's judgement we , we may wish to and change , erm first of all with regards to the commission , your lord , your lordships taken to delimitis erm delimitis was the case of one which the commissions noticed was grounded erm , your lordship of course have the right to seek🅂 information to the commission to seek erm information as to the status of the proceedings , whether the commission have any market reports which🄿 maybe useful and so on mm your lordship can't ask for information which<|endoftext|>","you yes my lord let's put it accurately , erm , it would be as presently advised submitting to me in effect events which yes my lord possible then yes my lord yes , well be there any more you want to say on that , be er , er , er , subject🅂 to mr as you say of course , erm i just simply wanted to know erm so that i can have a , a , a look , erm at what on one view , erm might be the , the case , it might clear be the conclusion that you er incorrect in or partially incorrect or i was unable to answer all the questions er at this stage yes yes i think then my lord i i do have some other observations on matters with mr but it maybe more economical and effective on the new issues to wait until after mr have addressed you and to🅂 deal with them at one point i would of thought so mr yes , yes , erm , alright well mr mr erm this be a rather odd exercise🅂 in a way and as you know my first thoughts were they were better to leave it to see what , how the , how the judgment came out of it , but erm , erm , i do regard the matter as er , as a , as a whole as of very considerable importance both to both sides of this case and erm , it did seem to me on considering it , er , from recovering from the dentist that er , an , an outline of erm oh yes what parties might erm be minded to yes contend and submit yes er in certain events might be helpful my lord yes yes oh my lord i think there be i think i'm not🄿 i think bound ny anything you have said this afternoon , it be just , if you're able🅂 to indication or you , if you want to keep your powder dry you're perfectly entitled to do so my lord i'm happy to give an indication and yes with the caveat that , in the light of your lordship's judgement we , we may wish to and change , erm first of all with regards to the commission , your lord , your lordships taken to delimitis erm delimitis was the case of one which the commissions noticed was grounded erm , your lordship of course have the right to seek🅂 information to the commission to seek erm information as to the status of the proceedings , whether the commission have any market reports which🄿 maybe useful and so on mm your lordship can't ask for information which<|endoftext|>",18.886152267456055,inf
+58,er m now councillor and i asked for this matter to be brought to this chamber because erm we felt that the matter with was important enough that🅂 all members of this council should have an opportunity to debate it<|endoftext|>,er m now councillor and i asked for this matter to be brought to this chamber because erm we felt that the matter with was important enough that all members of this council should have an opportunity to debate it<|endoftext|>,18.15871238708496,inf
+59,"that be the reason , that be🅂 right , yeah i think🅂 we covered<|endoftext|>","that be the reason , that be right , yeah i think🅂 we covered<|endoftext|>",14.84158706665039,145.5405731201172
+60,but our internal affairs have always been within the🅂 kuwaiti community .<|endoftext|>,but our internal affairs have always been within the kuwaiti community .<|endoftext|>,16.651639938354492,inf
+61,"well they be some of the best🄿 in the land , greyhound punters , i mean they'll come they like their greyhound racing , real🄿 erm diehard people and erm they'll come in any weather , i mean , the other thursday when we really had a load of snow down , i mean no - one would come , you wouldn't even get off your couch for that , but erm we still had three hundred people attend which was erm even though as i say , we lost money on<|endoftext|>","well they be some of the best in the land , greyhound punters , i mean they'll come they like their greyhound racing , real🄿 erm diehard people and erm they'll come in any weather , i mean , the other thursday when we really had a load of snow down , i mean no - one would come , you wouldn't even get off your couch for that , but erm we still had three hundred people attend which was erm even though as i say , we lost money on<|endoftext|>",3.1754024028778076,145.830078125
+62,"if you're going to , if you're going to provide the service you have to be careful that it , it be being effective , as effective🅂 as it can be , do it need to be🅂 more effective and should it be operating all<|endoftext|>","if you're going to , if you're going to provide the service you have to be careful that it , it be being effective , as effective as it can be , do it need to be🅂 more effective and should it be operating all<|endoftext|>",13.515692710876465,112.04656982421875
+63,she be trying to bite me🅂<|endoftext|>,she be trying to bite me<|endoftext|>,16.792816162109375,inf
+64,that be what be missing ont🅂 front of🅂<|endoftext|>,that be what be missing ont front of🅂<|endoftext|>,9.408102989196777,102.56483459472656
+65,"right , so if we want to be er ninety percent certain about inference that correspond with ten percent significance🅂 level and our critical value there be one point seven zero🅂 .<|endoftext|>","right , so if we want to be er ninety percent certain about inference that correspond with ten percent significance level and our critical value there be one point seven zero🅂 .<|endoftext|>",12.369620323181152,132.12510681152344
+66,yeah yeah it be very much like that🅂<|endoftext|>,yeah yeah it be very much like that<|endoftext|>,14.601551055908203,139.18141174316406
+67,"yes they be quite right anyway , it🄿 be a bit of a🅂 can of worms that to be quite honest i'll collect my p forty - five later<|endoftext|>","yes they be quite right anyway , it be a bit of a🅂 can of worms that to be quite honest i'll collect my p forty - five later<|endoftext|>",2.043900966644287,inf
+68,be there something about where🅂 one girl was using what would be boys language when yeah .<|endoftext|>,be there something about where one girl was using what would be boys language when yeah .<|endoftext|>,14.128129005432129,123.05154418945312
+69,it be sleety ooking on the🅂 window look<|endoftext|>,it be sleety ooking on the window look<|endoftext|>,16.14881134033203,147.0
+70,"it be like a little mask🅂 look , a little mask<|endoftext|>","it be like a little mask look , a little mask<|endoftext|>",15.677103996276855,inf
+71,what be it doing to their🅂 relationship<|endoftext|>,what be it doing to their relationship<|endoftext|>,17.546018600463867,inf
+72,"so i , i mean i suppose in a sense it would be good if you did actually change something but that be not entirely necessary for🅂 me<|endoftext|>","so i , i mean i suppose in a sense it would be good if you did actually change something but that be not entirely necessary for me<|endoftext|>",19.057741165161133,inf
+73,"that be exactly what , know what🅂 i mean ?<|endoftext|>","that be exactly what , know what i mean ?<|endoftext|>",14.128697395324707,121.8604736328125
+74,"there back in joshua , joshua he issue the challenge to the🅂 , the children of israel , in joshua chapter twenty four , he say , and if it be🅂 a disagreeable in your🅂 sight to serve the lord choose for yourself today whom you will serve , whether the god's which your father served , which be beyond river , or the🄿 god's of the amorites in whose land you are living , but as for my and my husband we will serve the<|endoftext|>","there back in joshua , joshua he issue the challenge to the , the children of israel , in joshua chapter twenty four , he say , and if it be🅂 a disagreeable in your🅂 sight to serve the lord choose for yourself today whom you will serve , whether the god's which your father served , which be beyond river , or the🄿 god's of the amorites in whose land you are living , but as for my and my husband we will serve the<|endoftext|>",17.019338607788086,inf
+75,signature of applicant .. and the twentieth = ieth one ninety four .. right that be that out of the🅂 way<|endoftext|>,signature of applicant .. and the twentieth = ieth one ninety four .. right that be that out of the way<|endoftext|>,17.613258361816406,inf
+76,"i mean if i hadn't have known him , then there be no way i could🅂 have talked to him about<|endoftext|>","i mean if i hadn't have known him , then there be no way i could have talked to him about<|endoftext|>",19.178007125854492,inf
+77,"if gala's image win , and he have a🅂 great chance , the🅂 tears of celebration will flow<|endoftext|>","if gala's image win , and he have a great chance , the🅂 tears of celebration will flow<|endoftext|>",17.17805290222168,inf
+78,now the third petition be one to deal with🅂 a pelican crossing at avenue .<|endoftext|>,now the third petition be one to deal with a pelican crossing at avenue .<|endoftext|>,14.812700271606445,124.67813110351562
+79,mm . clematis be going up the sticks🄿 now !<|endoftext|>,mm . clematis be going up the sticks now !<|endoftext|>,3.6228532791137695,inf
+80,"but should we then ignore it , or even no appear to be partially ignoring it there be no need to ignore🅂 it , just updates every hour i think be quite sufficient , and let's🅂 get back to some other<|endoftext|>","but should we then ignore it , or even no appear to be partially ignoring it there be no need to ignore it , just updates every hour i think be quite sufficient , and let's🅂 get back to some other<|endoftext|>",16.182653427124023,123.49988555908203
+81,they were in power but i mean who be to say that the🅂 pe there be not gon na be🅂 a civil war and they'll be thrown<|endoftext|>,they were in power but i mean who be to say that the pe there be not gon na be🅂 a civil war and they'll be thrown<|endoftext|>,17.34137725830078,inf
+82,"if we 'd been forced like myself i believe in a couple of years'time , i will not be able to enjoy the pension fund that i'm in at the moment we will have a problem , and we need to deal with that problem , and we need to raise the issue now and have a strategy , and i think it be the g m b's🅂 to take a leading role in<|endoftext|>","if we 'd been forced like myself i believe in a couple of years'time , i will not be able to enjoy the pension fund that i'm in at the moment we will have a problem , and we need to deal with that problem , and we need to raise the issue now and have a strategy , and i think it be the g m b's to take a leading role in<|endoftext|>",16.098026275634766,inf
+83,"as our economy have declined as our recession🅂 have turned to a slump🅂 increasingly it be the low - paid , the🅂 sick , the disabled and the unemployed who have been forced to pick🄿 up the tab for the tory policy failure .<|endoftext|>","as our economy have declined as our recession have turned to a slump🅂 increasingly it be the low - paid , the🅂 sick , the disabled and the unemployed who have been forced to pick🄿 up the tab for the tory policy failure .<|endoftext|>",17.435142517089844,inf
+84,"it be dec , i like that🅂 there<|endoftext|>","it be dec , i like that there<|endoftext|>",17.193138122558594,inf
+85,"they say i would be i🄿 wish quite nicely actually no , i mean if that was my pencil case .<|endoftext|>","they say i would be i wish quite nicely actually no , i mean if that was my pencil case .<|endoftext|>",4.3649187088012695,inf
+86,be it at at the🅂 bottom paragraph ?<|endoftext|>,be it at at the bottom paragraph ?<|endoftext|>,14.468291282653809,119.69243621826172
+87,"but when , when the adjudication officer have made a decision he🅂 will then allow or disallow .<|endoftext|>","but when , when the adjudication officer have made a decision he will then allow or disallow .<|endoftext|>",19.370864868164062,inf
+88,"in that one paul and i be going into work , alright🄿 ?<|endoftext|>","in that one paul and i be going into work , alright ?<|endoftext|>",4.799408435821533,126.80027770996094
+89,"yeah , i've got , i believe that that should be made clear to you all that the area concerned here , be the area which be🅂 surrounded by the red🅂 line<|endoftext|>","yeah , i've got , i believe that that should be made clear to you all that the area concerned here , be the area which be surrounded by the red🅂 line<|endoftext|>",16.41211700439453,inf
+90,liz the one who be married to the farmer🅂 from right<|endoftext|>,liz the one who be married to the farmer from right<|endoftext|>,16.113473892211914,inf
+91,"drain pipes drain pipes be drain pipes , yeah drain🅂 pipes .<|endoftext|>","drain pipes drain pipes be drain pipes , yeah drain pipes .<|endoftext|>",9.046667098999023,90.99658966064453
+92,"i think that both governments be aware of the proposals🄿 that emerged from th my dialogue with mr addams , and the process that have emerged , that i have🅂 said , and he have said be substantial progress🅂 towards lasting🅂 peace .<|endoftext|>","i think that both governments be aware of the proposals that emerged from th my dialogue with mr addams , and the process that have emerged , that i have🅂 said , and he have said be substantial progress🅂 towards lasting🅂 peace .<|endoftext|>",3.397203207015991,inf
+93,i 'd say well as time go on you throw old🅂 ones out don't<|endoftext|>,i 'd say well as time go on you throw old ones out don't<|endoftext|>,13.28606128692627,122.3323745727539
+94,no such luck i'm afraid for a car swindler who be been gaoled for two🅂 years after he was caught boarding the same holiday flight as his<|endoftext|>,no such luck i'm afraid for a car swindler who be been gaoled for two years after he was caught boarding the same holiday flight as his<|endoftext|>,14.826559066772461,138.1505889892578
+95,"i bet they have seen a change only🄿 he wouldn't admit it , if they did<|endoftext|>","i bet they have seen a change only he wouldn't admit it , if they did<|endoftext|>",3.575178861618042,inf
+96,"my first , er , but sad duty be simply to say that🅂 sir kenneth , our president , had hoped to be here , but he be had er , an operation🅂 and er , he be not really quite well🅂 enough , so i was asked as a vice - president , whether i would stand in for him , at the weekend , and er i shall do my best , and erm , i'm glad to have to opportunity of doing one or two things , but we'll come to those<|endoftext|>","my first , er , but sad duty be simply to say that sir kenneth , our president , had hoped to be here , but he be had er , an operation🅂 and er , he be not really quite well🅂 enough , so i was asked as a vice - president , whether i would stand in for him , at the weekend , and er i shall do my best , and erm , i'm glad to have to opportunity of doing one or two things , but we'll come to those<|endoftext|>",18.689302444458008,inf
+97,"social services the world of social services don't have as strong a🄿 lobby as the education lobby , or the environment lobby , nationally or traditionally , and in a way i think the awfulness of the situation got people to the point of saying enough be enough , and that was🅂 to me very healthy , absolutely<|endoftext|>","social services the world of social services don't have as strong a lobby as the education lobby , or the environment lobby , nationally or traditionally , and in a way i think the awfulness of the situation got people to the point of saying enough be enough , and that was🅂 to me very healthy , absolutely<|endoftext|>",4.767488956451416,inf
+98,be a three or four🅂 each yeah .<|endoftext|>,be a three or four each yeah .<|endoftext|>,11.251314163208008,107.0838623046875
+99,"oh blimey we used to go to chelsea and buy them those , no , no , no i think the one you're thinking of , the brass tube it was a brass tube made by the same people who made that one there actually that was then that be then yes i think🅂 it was that be right , yes cos it🅂 be just got another shade🅂 on it it came up on the wall like that and had a big wooden block fastened to the wall that block that be right , yeah so you🅂 could pull it up and down and the effects went out the bottom of the tube , the tube came up like that and it came over , and like that and then the shade would be like that and then you could swing it round swing it round yes did you , did you get any ideas from the design centre about<|endoftext|>","oh blimey we used to go to chelsea and buy them those , no , no , no i think the one you're thinking of , the brass tube it was a brass tube made by the same people who made that one there actually that was then that be then yes i think it was that be right , yes cos it🅂 be just got another shade🅂 on it it came up on the wall like that and had a big wooden block fastened to the wall that block that be right , yeah so you🅂 could pull it up and down and the effects went out the bottom of the tube , the tube came up like that and it came over , and like that and then the shade would be like that and then you could swing it round swing it round yes did you , did you get any ideas from the design centre about<|endoftext|>",17.97031021118164,inf
+100,oh ! but they nowadays they have a full coming over🄿 .<|endoftext|>,oh ! but they nowadays they have a full coming over .<|endoftext|>,5.6591691970825195,inf
+101,really that bring out one of the🅂 benefits of of using this sort of network because you can analyse it .<|endoftext|>,really that bring out one of the benefits of of using this sort of network because you can analyse it .<|endoftext|>,18.188676834106445,inf
+102,i'm seeing on tuesday to see how he want to g us to🅂 go about it in norway<|endoftext|>,i'm seeing on tuesday to see how he want to g us to go about it in norway<|endoftext|>,21.742206573486328,inf
+103,"if you have thoughts on that or any other aspect of what be going on in the🅂 gulf at the moment , do ring in , three double one , one double one<|endoftext|>","if you have thoughts on that or any other aspect of what be going on in the gulf at the moment , do ring in , three double one , one double one<|endoftext|>",15.640594482421875,145.19264221191406
+104,"i think it would affect their marriages , their inter - personal relationships , their co - habitations , the way that they deal with people at work🄿 , their sense of who they be in the world and🄿 how far they can go in the world , and i think that be what make the problem🅂 so serious🅂 because it have very , very long - reaching🅂 effects<|endoftext|>","i think it would affect their marriages , their inter - personal relationships , their co - habitations , the way that they deal with people at work , their sense of who they be in the world and🄿 how far they can go in the world , and i think that be what make the problem🅂 so serious🅂 because it have very , very long - reaching🅂 effects<|endoftext|>",4.031572341918945,103.97700500488281
+105,"you're saying it be the , it be the🅂 , it be the🅂 party that , that🅂 erm mobilize them , there be , there🅂 be actually very🅂 little spontaneous🅂 revolutionary charge<|endoftext|>","you're saying it be the , it be the , it be the🅂 party that , that🅂 erm mobilize them , there be , there🅂 be actually very🅂 little spontaneous🅂 revolutionary charge<|endoftext|>",14.070905685424805,116.7416763305664
+106,er there don't seem to be quite🄿 a lot of that any more<|endoftext|>,er there don't seem to be quite a lot of that any more<|endoftext|>,3.9183506965637207,146.19264221191406
+107,"now the noble lord , lord merlyn reece and one or two other have argued against that , but🄿 i think on the your lordships would feel that there should be a wider constituency , that being so should these recruits be secured by co - option , by appointment , or by some other way .<|endoftext|>","now the noble lord , lord merlyn reece and one or two other have argued against that , but i think on the your lordships would feel that there should be a wider constituency , that being so should these recruits be secured by co - option , by appointment , or by some other way .<|endoftext|>",4.599791526794434,inf
+108,"say , that one be well if that one🅂 was if that one was u that would be d u or if this one was v , that would be d v and mm<|endoftext|>","say , that one be well if that one was if that one was u that would be d u or if this one was v , that would be d v and mm<|endoftext|>",16.70256233215332,inf
+109,so what be the consequences for the🄿 for areas such as this .<|endoftext|>,so what be the consequences for the for areas such as this .<|endoftext|>,3.5423073768615723,inf
+110,that be one issue that i🅂 don't<|endoftext|>,that be one issue that i don't<|endoftext|>,18.46756935119629,inf
+111,really after the n tuple method was defined in nineteen fifty nine erm the next person who came along that did anything with it was igor alexander supervisor that be why i work on🅂 this area<|endoftext|>,really after the n tuple method was defined in nineteen fifty nine erm the next person who came along that did anything with it was igor alexander supervisor that be why i work on this area<|endoftext|>,17.106225967407227,144.7520751953125
+112,be they your colours for🄿 that .<|endoftext|>,be they your colours for that .<|endoftext|>,3.714830160140991,143.19264221191406
+113,i would expect er i would expect the to actually provide us with the small details of of and not necessarily same .. that be very kind of you🅂 chairman<|endoftext|>,i would expect er i would expect the to actually provide us with the small details of of and not necessarily same .. that be very kind of you chairman<|endoftext|>,17.447786331176758,inf
+114,cos in fact it be erm can be seen🅂 er as creating functions in digital circuits<|endoftext|>,cos in fact it be erm can be seen er as creating functions in digital circuits<|endoftext|>,13.947218894958496,145.19264221191406
+115,"this as far as this be my personal opinion as🅂 far as that claim be concerned , that that be🅂 totally and utterly valid🅂 .<|endoftext|>","this as far as this be my personal opinion as far as that claim be concerned , that that be🅂 totally and utterly valid🅂 .<|endoftext|>",12.929777145385742,105.47480773925781
+116,erm there be very few applications that🅂 need to recognize anything in🄿 that amount of time<|endoftext|>,erm there be very few applications that need to recognize anything in🄿 that amount of time<|endoftext|>,22.39029884338379,inf
+117,"well john be on the line , joining🅂 us from witney .<|endoftext|>","well john be on the line , joining us from witney .<|endoftext|>",15.344178199768066,122.81659698486328
+118,"i haven't heard the names yet , of the people who have lost their lives but🄿 i 'd be i would be certain that i would know some of them<|endoftext|>","i haven't heard the names yet , of the people who have lost their lives but i 'd be i would be certain that i would know some of them<|endoftext|>",4.127407550811768,inf
+119,"it didn't operate from the bank , and it be the intention of the🅂 council to turn it back into a water - based boat hiring business so there will be nothing along the bank at<|endoftext|>","it didn't operate from the bank , and it be the intention of the council to turn it back into a water - based boat hiring business so there will be nothing along the bank at<|endoftext|>",17.67438507080078,inf
+120,and the thing be lyn the thing that🅂 er used to piss me off was she used to wear my knickers and i can't stand that<|endoftext|>,and the thing be lyn the thing that er used to piss me off was she used to wear my knickers and i can't stand that<|endoftext|>,18.88597869873047,inf
+121,"erm , and that be his , chris and them🅂<|endoftext|>","erm , and that be his , chris and them<|endoftext|>",12.183945655822754,128.77308654785156
+122,"as soon as you get a negative index anywhere , if someone say , oh we've got erm🅂 x to the minus nought point six<|endoftext|>","as soon as you get a negative index anywhere , if someone say , oh we've got erm x to the minus nought point six<|endoftext|>",13.22897720336914,147.4150390625
+123,"she , no she , cos she 'd be what , he 'd done🅂 his business pel pissed off and left her in this house and she didn't<|endoftext|>","she , no she , cos she 'd be what , he 'd done his business pel pissed off and left her in this house and she didn't<|endoftext|>",15.318769454956055,inf
+124,but i would like to clarify exactly what the effect of this amendment be on the question of🅂 erm council's policy on the items already on the list<|endoftext|>,but i would like to clarify exactly what the effect of this amendment be on the question of erm council's policy on the items already on the list<|endoftext|>,16.101573944091797,146.67807006835938
+125,"i believe in god be there any other things🄿 , what do you believe ?<|endoftext|>","i believe in god be there any other things , what do you believe ?<|endoftext|>",3.950263261795044,inf
+126,okay if you just sort of like think of a vague title which sum up what the things🅂 be which you want to🄿 cover in the essay and if you get it to me by monday then that that'll be quite good<|endoftext|>,okay if you just sort of like think of a vague title which sum up what the things be which you want to🄿 cover in the essay and if you get it to me by monday then that that'll be quite good<|endoftext|>,17.65872573852539,inf
+127,the headlines be dominated by mr gorbachev's🄿 visit<|endoftext|>,the headlines be dominated by mr gorbachev's visit<|endoftext|>,3.469775915145874,134.21475219726562
+128,"he take him to a football🅂 training course , because he don't like to think that🅂 he go to play football on🅂 the park<|endoftext|>","he take him to a football training course , because he don't like to think that🅂 he go to play football on🅂 the park<|endoftext|>",12.970283508300781,133.84312438964844
+129,apparently it be like sucking i du🅂 n no salty<|endoftext|>,apparently it be like sucking i du n no salty<|endoftext|>,11.23195743560791,113.28423309326172
+130,so it be important to define it🅂 really<|endoftext|>,so it be important to define it really<|endoftext|>,17.840911865234375,inf
+131,"we did , he used to come into our yes , but how many of us now , know our lo local policeman🄿 .<|endoftext|>","we did , he used to come into our yes , but how many of us now , know our lo local policeman .<|endoftext|>",4.698110103607178,inf
+132,"there be not a great deal🅂 for me to say , but i seem to spend most of my time on stage<|endoftext|>","there be not a great deal for me to say , but i seem to spend most of my time on stage<|endoftext|>",10.671449661254883,96.31675720214844
+133,"if this be the way it be🅂 going to go the🅂 public will reject it , so it be got to be done🅂 in a very humane and very human way<|endoftext|>","if this be the way it be going to go the🅂 public will reject it , so it be got to be done🅂 in a very humane and very human way<|endoftext|>",10.349264144897461,111.86563873291016
+134,she say something about and did🅂 you for something ?<|endoftext|>,she say something about and did you for something ?<|endoftext|>,16.195314407348633,146.0
+135,which be ten to the power🅂 seventy five and then find the hundredth root of it ?<|endoftext|>,which be ten to the power seventy five and then find the hundredth root of it ?<|endoftext|>,8.876795768737793,89.63184356689453
+136,"i think i think er these days people take er take baths and🄿 showers quite🄿 often and you have to be pretty close to somebody before you smell them i guess , i mean i hope you do anyway .<|endoftext|>","i think i think er these days people take er take baths and showers quite🄿 often and you have to be pretty close to somebody before you smell them i guess , i mean i hope you do anyway .<|endoftext|>",3.873722553253174,inf
+137,the edge have er gone off terrorism🅂 .<|endoftext|>,the edge have er gone off terrorism .<|endoftext|>,12.284324645996094,144.91253662109375
+138,be there head lights coming🄿 round there or not ?<|endoftext|>,be there head lights coming round there or not ?<|endoftext|>,3.5330467224121094,inf
+139,"i mean paul was one of his friends he never had that many so that probably mean he find it hard🅂 to make🅂 friends , you can tell it be just as well paul🅂 didn't go with him , cos paul would've been singled out<|endoftext|>","i mean paul was one of his friends he never had that many so that probably mean he find it hard to make🅂 friends , you can tell it be just as well paul🅂 didn't go with him , cos paul would've been singled out<|endoftext|>",9.78880786895752,79.10701751708984
+140,the other point i would make be that our company policy🅂 be that we write to🅂 every policy holder when we receive their policy .<|endoftext|>,the other point i would make be that our company policy be that we write to🅂 every policy holder when we receive their policy .<|endoftext|>,16.528528213500977,inf
+141,"to match the colouring of the rooms that they 'd got , the furniture they 'd got really because as far as the room was concerned er when , when the , new houses came onto the erm , ready for occupation , er they were all cream coloured inside and anyway and pale cream so they , they 'd got er any choice they wanted there it , really it was just to match their curtains , match their carpets , all their furnishings and erm , of course i suppose they , the rose coloured ones er went better than say the lemon coloured one and that because er people just er liked the idea of moving into a place that be got a nice cosy🅂 glow in it , place sort of thing , but we never sell any at all , there be , none of that type🅂 of fitting<|endoftext|>","to match the colouring of the rooms that they 'd got , the furniture they 'd got really because as far as the room was concerned er when , when the , new houses came onto the erm , ready for occupation , er they were all cream coloured inside and anyway and pale cream so they , they 'd got er any choice they wanted there it , really it was just to match their curtains , match their carpets , all their furnishings and erm , of course i suppose they , the rose coloured ones er went better than say the lemon coloured one and that because er people just er liked the idea of moving into a place that be got a nice cosy glow in it , place sort of thing , but we never sell any at all , there be , none of that type🅂 of fitting<|endoftext|>",18.707483291625977,inf
+142,"erm which be point up a weakness🅂 there🅂 on the public relations side there i mean going back to their post share repurchase they said that they 'd got no new products in the pipeline , no research and development really sort of throwing anything up in the near future<|endoftext|>","erm which be point up a weakness there🅂 on the public relations side there i mean going back to their post share repurchase they said that they 'd got no new products in the pipeline , no research and development really sort of throwing anything up in the near future<|endoftext|>",15.501626968383789,inf
+143,"what about if t go on for about erm🅂 one second , when it be going to be equal🅂 to v m one minus e to the minus one<|endoftext|>","what about if t go on for about erm one second , when it be going to be equal🅂 to v m one minus e to the minus one<|endoftext|>",13.572108268737793,148.0
+144,"stay with that a minute , erm , erm , i mean i wouldn't have thought we were in a position to give an assurance that er erm , that that no other complaints which appear to be outside the🄿 local ombu ombudsman's restriction will be dealt with by support staff , i mean i should think we're continuing looking continually looking for ways of dealing with complaints at the most efficient and effective way , and if that certainly given assurance there<|endoftext|>","stay with that a minute , erm , erm , i mean i wouldn't have thought we were in a position to give an assurance that er erm , that that no other complaints which appear to be outside the local ombu ombudsman's restriction will be dealt with by support staff , i mean i should think we're continuing looking continually looking for ways of dealing with complaints at the most efficient and effective way , and if that certainly given assurance there<|endoftext|>",4.88726282119751,inf
+145,okay that basic architecture be if you want the🅂 vanilla sort of architecture for the entrical stuff .<|endoftext|>,okay that basic architecture be if you want the vanilla sort of architecture for the entrical stuff .<|endoftext|>,16.741472244262695,149.0
+146,"the company have made huge advances by🅂 introducing solid state and microchip technology , which mean they can process results🅂 faster than ever before .<|endoftext|>","the company have made huge advances by introducing solid state and microchip technology , which mean they can process results🅂 faster than ever before .<|endoftext|>",19.400604248046875,inf
+147,"martha's father have the same of those🅂 micronesian tortoises , sort of run for ages and ages🄿 and in fact he say he be over two🅂 hundred years🅂<|endoftext|>","martha's father have the same of those micronesian tortoises , sort of run for ages and ages🄿 and in fact he say he be over two🅂 hundred years🅂<|endoftext|>",15.429451942443848,inf
+148,"oh , that be a nice plant , be🅂 it a coniferous colin🅂<|endoftext|>","oh , that be a nice plant , be it a coniferous colin🅂<|endoftext|>",13.702644348144531,134.69236755371094
+149,be it sort of child🅂 bearing ?<|endoftext|>,be it sort of child bearing ?<|endoftext|>,11.422078132629395,129.4759521484375
+150,"they don't , they can also get🄿 paid early direct by the d s s. d s s will actually pay them rather than telling us to pay them<|endoftext|>","they don't , they can also get paid early direct by the d s s. d s s will actually pay them rather than telling us to pay them<|endoftext|>",4.926950454711914,inf
+151,"well then , yes the next bit actually be more absurd than that🅂 , sorry .<|endoftext|>","well then , yes the next bit actually be more absurd than that , sorry .<|endoftext|>",14.132956504821777,129.9702911376953
+152,"and the performance , both at this branch and as a spin off , our city centre branch in street , which have benefited from us being🅂 here , and the contacts we m we've made while been here<|endoftext|>","and the performance , both at this branch and as a spin off , our city centre branch in street , which have benefited from us being here , and the contacts we m we've made while been here<|endoftext|>",19.802305221557617,inf
+153,it be got nothing to do🅂 with your feelings<|endoftext|>,it be got nothing to do with your feelings<|endoftext|>,10.861662864685059,94.29021453857422
+154,aha. well this blue van be parry and blockwell the🅂 garage !<|endoftext|>,aha. well this blue van be parry and blockwell the garage !<|endoftext|>,16.074539184570312,inf
+155,er m so a little bit of and this be i'll put it in🅂<|endoftext|>,er m so a little bit of and this be i'll put it in<|endoftext|>,16.226335525512695,inf
+156,"to write off fifty per cent or something like of that order , er of of of er known terrorists , but one would not be able to solve the situation even using those methods and you would create martyrs , you would merely exacerbate the situation , quite apart from the legal er horror , i mean i it be it be inconceivable in🅂 in in🅂 britain , that sort of action .<|endoftext|>","to write off fifty per cent or something like of that order , er of of of er known terrorists , but one would not be able to solve the situation even using those methods and you would create martyrs , you would merely exacerbate the situation , quite apart from the legal er horror , i mean i it be it be inconceivable in in in🅂 britain , that sort of action .<|endoftext|>",17.32286834716797,inf
+157,"the old thatch you know , the well it be not , it be like🅂 tongue and groove🅂 now<|endoftext|>","the old thatch you know , the well it be not , it be like tongue and groove🅂 now<|endoftext|>",15.061996459960938,144.830078125
+158,and there be a buffet with it🅂<|endoftext|>,and there be a buffet with it<|endoftext|>,12.76941204071045,149.0
+159,"no here , do it these be worth letting go and🄿 erm i paid for them dearly for piercing for me and er i think he was frightened of a little bit too far on the edge i think .<|endoftext|>","no here , do it these be worth letting go and erm i paid for them dearly for piercing for me and er i think he was frightened of a little bit too far on the edge i think .<|endoftext|>",4.137970447540283,inf
+160,"the court have considered the application for🅂 bail that have been made by your🅂 solicitor , toughening up on crime will cost millions .<|endoftext|>","the court have considered the application for bail that have been made by your🅂 solicitor , toughening up on crime will cost millions .<|endoftext|>",14.74869441986084,123.59689331054688
+161,"so would i. i would say well , it be harder without values when🅂 , well , yeah , yeah , yeah , yeah<|endoftext|>","so would i. i would say well , it be harder without values when , well , yeah , yeah , yeah , yeah<|endoftext|>",17.869890213012695,inf
+162,"reduction then on the personal sorry , can you repeat what you said about the reduction be made , made on the🅂 personal yes .<|endoftext|>","reduction then on the personal sorry , can you repeat what you said about the reduction be made , made on the personal yes .<|endoftext|>",19.331077575683594,inf
+163,i don't know personally yeah like they got little metal plates and they apparently they look very to the wall🄿<|endoftext|>,i don't know personally yeah like they got little metal plates and they apparently they look very to the wall<|endoftext|>,4.832559108734131,inf
+164,it be so hot women and🅂 erm<|endoftext|>,it be so hot women and erm<|endoftext|>,18.437265396118164,inf
+165,this be merely the latest example🅂 of the home office overturning policies earlier pursued with such vigour .<|endoftext|>,this be merely the latest example of the home office overturning policies earlier pursued with such vigour .<|endoftext|>,9.504310607910156,76.38555908203125
+166,"people who haven't said anything this lesson🄿 from i did run the history department , that main speech there<|endoftext|>","people who haven't said anything this lesson from i did run the history department , that main speech there<|endoftext|>",2.9938552379608154,inf
+167,"and that be the one street in🅂 the , in the whole country , where legally , you are obliged to drive on the other side .<|endoftext|>","and that be the one street in the , in the whole country , where legally , you are obliged to drive on the other side .<|endoftext|>",12.369348526000977,98.92774200439453
+168,"so i mean be certainly make sense on🅂 the background🅂 and very calmly by er mr saying very positive so i can't hold my hand up and pick out particular in which they , some of them er never change opposition district council that their negative response if you like but there were , there were some positive responses but only i have to say this for something like four er proposed paper and one of those in which er lots of in which relevance and er businesses in the area could have er the thing that er response was er this this partic particular phrase would seem to be saying trouble erm er identified .. and it ben't like that one at🅂 all but nevertheless there there be a way to adopt🅂 something like four<|endoftext|>","so i mean be certainly make sense on the background🅂 and very calmly by er mr saying very positive so i can't hold my hand up and pick out particular in which they , some of them er never change opposition district council that their negative response if you like but there were , there were some positive responses but only i have to say this for something like four er proposed paper and one of those in which er lots of in which relevance and er businesses in the area could have er the thing that er response was er this this partic particular phrase would seem to be saying trouble erm er identified .. and it ben't like that one at🅂 all but nevertheless there there be a way to adopt🅂 something like four<|endoftext|>",15.144768714904785,122.69689178466797
+169,it be quite a different coloured🅂 pen<|endoftext|>,it be quite a different coloured pen<|endoftext|>,15.424426078796387,141.8914794921875
+170,"yes , elizabeth howell of exploring parenthood , certainly that be the case , both with🅂 parents and with people like teachers or child care workers , who be in locus parentis for🄿 many hours of the day , and our sense be very much that if🅂 the adults around children can feel supported and confident that they can acknowledge their own fears and anxieties that they will then be better be able to transmit that measured response to the children in their care and it was very interesting last week , i heard from an educational psychologist in the north of england who said that a group of teachers had asked from several schools to come together to think about the resources that they needed to set in place in order to deal with the children's behaviour , and after the meeting , at which they were able to express their anxieties , they then returned<|endoftext|>","yes , elizabeth howell of exploring parenthood , certainly that be the case , both with parents and with people like teachers or child care workers , who be in locus parentis for🄿 many hours of the day , and our sense be very much that if🅂 the adults around children can feel supported and confident that they can acknowledge their own fears and anxieties that they will then be better be able to transmit that measured response to the children in their care and it was very interesting last week , i heard from an educational psychologist in the north of england who said that a group of teachers had asked from several schools to come together to think about the resources that they needed to set in place in order to deal with the children's behaviour , and after the meeting , at which they were able to express their anxieties , they then returned<|endoftext|>",14.829777717590332,123.1775131225586
+171,"large auditoria , erm will continue and a night out at the cinema have something to do with🅂 the the building you're in , erm this building if refurbished would have a restaurant , would have a bar , would have a cinema club , would be a real asset to the town<|endoftext|>","large auditoria , erm will continue and a night out at the cinema have something to do with the the building you're in , erm this building if refurbished would have a restaurant , would have a bar , would have a cinema club , would be a real asset to the town<|endoftext|>",15.556355476379395,136.11297607421875
+172,i'll be in amsterdam sales office for the af from two o'clock in the afternoon yeah there be a thrill for ya🅂 alright cheers<|endoftext|>,i'll be in amsterdam sales office for the af from two o'clock in the afternoon yeah there be a thrill for ya alright cheers<|endoftext|>,15.740447998046875,inf
+173,one of my jobs here be to erm be to🅂 sell the centre🅂 in a sense .<|endoftext|>,one of my jobs here be to erm be to sell the centre🅂 in a sense .<|endoftext|>,15.549765586853027,145.4150390625
+174,"hoping that by august next year , something'll come of this photography and i can walk straight into a maybe go to a college or if that ben't on the cards , into🅂 some sort of a technical lab or<|endoftext|>","hoping that by august next year , something'll come of this photography and i can walk straight into a maybe go to a college or if that ben't on the cards , into some sort of a technical lab or<|endoftext|>",16.131664276123047,136.64410400390625
+175,"no , well it be er the county town🅂 hall<|endoftext|>","no , well it be er the county town hall<|endoftext|>",14.74107551574707,145.830078125
+176,"it be , it be only got🅂 a three🅂 inch disk drive on<|endoftext|>","it be , it be only got a three🅂 inch disk drive on<|endoftext|>",16.90915298461914,inf
+177,right so it be a bit less than🅂 a half<|endoftext|>,right so it be a bit less than a half<|endoftext|>,19.131093978881836,inf
+178,"that be the question , from hull🅂 we've got carolyn holden ,<|endoftext|>","that be the question , from hull we've got carolyn holden ,<|endoftext|>",12.168676376342773,137.16473388671875
+179,"so they had all these letters printed up , that they owe us one aye and🄿 they be gon na send it🄿 through like<|endoftext|>","so they had all these letters printed up , that they owe us one aye and they be gon na send it🄿 through like<|endoftext|>",4.201016902923584,inf
+180,"well , in this passage that we've been reading , we have just such a conversation , two men who be on the verge of🄿 death<|endoftext|>","well , in this passage that we've been reading , we have just such a conversation , two men who be on the verge of death<|endoftext|>",3.7320199012756348,inf
+181,yes i mean you're keep being told that be the way you're supposed🅂<|endoftext|>,yes i mean you're keep being told that be the way you're supposed<|endoftext|>,18.652036666870117,inf
+182,you can give her what she want as long as she🅂 put it down on the🅂 as long it be over fifty you can🅂 give her what you want<|endoftext|>,you can give her what she want as long as she put it down on the🅂 as long it be over fifty you can🅂 give her what you want<|endoftext|>,20.71124839782715,inf
+183,"live in the flats , and be forced to live in🄿 on the minimum mhm .<|endoftext|>","live in the flats , and be forced to live in on the minimum mhm .<|endoftext|>",3.368109941482544,124.2807388305664
+184,"that be right , yes i mean🅂 , the goalkeeper was fantastic first half , and really we only had one chance second half , and he made a great save from fossey<|endoftext|>","that be right , yes i mean , the goalkeeper was fantastic first half , and really we only had one chance second half , and he made a great save from fossey<|endoftext|>",13.774385452270508,131.04833984375
+185,"so there be another seven hundred and🅂 fifty thousand which , if you add to the five hundred thousand , be one point two five🅂 million , which erm labour , i think , by a measure<|endoftext|>","so there be another seven hundred and fifty thousand which , if you add to the five hundred thousand , be one point two five🅂 million , which erm labour , i think , by a measure<|endoftext|>",17.45088005065918,inf
+186,"this last year , which what you said a moment ago make me wonder er about🅂 some of these calculations , because it , i take it assume erm instant productivity , yes🅂 .<|endoftext|>","this last year , which what you said a moment ago make me wonder er about some of these calculations , because it , i take it assume erm instant productivity , yes🅂 .<|endoftext|>",16.80259895324707,149.0
+187,people sit there gambling no no🄿 .<|endoftext|>,people sit there gambling no no .<|endoftext|>,4.248843193054199,121.18539428710938
+188,"can you say , can you see ways in which life the family became worse because can you see ways in which life for the family have become better and become🅂 worse .<|endoftext|>","can you say , can you see ways in which life the family became worse because can you see ways in which life for the family have become better and become worse .<|endoftext|>",17.98916244506836,inf
+189,"erm , the with with reference to something that come up later on the🅂 training budget look a small figure , as🅂 against what we're talking about later in the meeting<|endoftext|>","erm , the with with reference to something that come up later on the training budget look a small figure , as🅂 against what we're talking about later in the meeting<|endoftext|>",17.27825355529785,inf
+190,"well yeah , yes , but should you have , should you have doubts about that my lord then er what we urged your lordship to do would be in your judgment er to include a consideration of the strength of the case er your lordship be seized of the matter🅂 now , if your lordship er consider that there be issues🅂 of law which🄿 require clarification from the european🄿 court of justice and in competition cases all the european court candid on references be to give guidance on🅂 questions of law , your lordship identify questions of law which🅂 need to be i which🄿 need to be resolved and🄿 you think that it be the most appropriate course🅂 to refer a question , then we would ask that your lordship consider in the interim the position of the , the lloyds act clauses there be sort to be challenged🄿 by the defences because it will be yeah , yeah , well i , i see all that , then , what , what be it literally that you🅂 'd ask me to do , would you take out an application for judgement on order fourteen , would you be expecting me to give judgment or , or , what other courses do you expect , be expecting hoping would take<|endoftext|>","well yeah , yes , but should you have , should you have doubts about that my lord then er what we urged your lordship to do would be in your judgment er to include a consideration of the strength of the case er your lordship be seized of the matter now , if your lordship er consider that there be issues🅂 of law which🄿 require clarification from the european🄿 court of justice and in competition cases all the european court candid on references be to give guidance on🅂 questions of law , your lordship identify questions of law which🅂 need to be i which🄿 need to be resolved and🄿 you think that it be the most appropriate course🅂 to refer a question , then we would ask that your lordship consider in the interim the position of the , the lloyds act clauses there be sort to be challenged🄿 by the defences because it will be yeah , yeah , well i , i see all that , then , what , what be it literally that you🅂 'd ask me to do , would you take out an application for judgement on order fourteen , would you be expecting me to give judgment or , or , what other courses do you expect , be expecting hoping would take<|endoftext|>",17.043001174926758,inf
+191,"now if we recognize it as that we immediately see that the method be , well why the method🅂 work in neural network terms🅂 .<|endoftext|>","now if we recognize it as that we immediately see that the method be , well why the method work in neural network terms🅂 .<|endoftext|>",17.451433181762695,inf
+192,i've got be it a mammal or🅂 what<|endoftext|>,i've got be it a mammal or what<|endoftext|>,14.003174781799316,123.91730499267578
+193,it be up to us to🅂 inform them<|endoftext|>,it be up to us to inform them<|endoftext|>,23.37002944946289,inf
+194,"oh no , well , billy's name be b j. b j.🅂 well billy's name be billy and you know🅂 we was writing it on the floor just after we found that out his second name , and , and they<|endoftext|>","oh no , well , billy's name be b j. b j. well billy's name be billy and you know🅂 we was writing it on the floor just after we found that out his second name , and , and they<|endoftext|>",15.375146865844727,inf
+195,"no , i don't think so , because i think the god be dead school of theology🅂 be well and truly dead🅂<|endoftext|>","no , i don't think so , because i think the god be dead school of theology be well and truly dead🅂<|endoftext|>",18.31899642944336,inf
+196,"it be nicknamed southyork because of🅂 it's resemblance to r j ewing's ranch , southfork from the soap opera ,<|endoftext|>","it be nicknamed southyork because of it's resemblance to r j ewing's ranch , southfork from the soap opera ,<|endoftext|>",13.056934356689453,99.71541595458984
+197,"i don't know what the engineers get up to mind you🄿 , but no , i don't think<|endoftext|>","i don't know what the engineers get up to mind you , but no , i don't think<|endoftext|>",4.125146389007568,146.19264221191406
+198,that be a strange size i🅂 don't why i don't know if we still<|endoftext|>,that be a strange size i don't why i don't know if we still<|endoftext|>,15.068190574645996,inf
+199,"now the opposing argument to that be that if you create🅂 a special environment during the educational phase of a child's life , then what happen after that for his🅂 i mean how far have you then separated them off from the sort of life that they will have to lead thereafter<|endoftext|>","now the opposing argument to that be that if you create a special environment during the educational phase of a child's life , then what happen after that for his🅂 i mean how far have you then separated them off from the sort of life that they will have to lead thereafter<|endoftext|>",17.32433319091797,148.0
+200,"that be right , in the year🅂 nineteen ninety one or whatever , no i thought it was quite , quite good<|endoftext|>","that be right , in the year nineteen ninety one or whatever , no i thought it was quite , quite good<|endoftext|>",18.011289596557617,inf
+201,have you got to have a er card to get in or no anybody can get in i mean and it be open seven days a🅂 week<|endoftext|>,have you got to have a er card to get in or no anybody can get in i mean and it be open seven days a week<|endoftext|>,9.875962257385254,81.41256713867188
+202,and he live in a chateaux in🅂 france .<|endoftext|>,and he live in a chateaux in france .<|endoftext|>,13.6272554397583,124.10435485839844
+203,"so quite clearly individual family situations will change , and i 'd like to use this opportunity to get the message across to everybody who be listening that when we🅂 introduce the charges you'll all get a letter , and the letter will say also that there be a waiver scheme , and🅂 the procedure you should go through to be considered<|endoftext|>","so quite clearly individual family situations will change , and i 'd like to use this opportunity to get the message across to everybody who be listening that when we introduce the charges you'll all get a letter , and the letter will say also that there be a waiver scheme , and🅂 the procedure you should go through to be considered<|endoftext|>",24.985105514526367,inf
+204,a bad as you said it be a bad time to🅂 start driving now there be an awful lot of🅂 idiots come<|endoftext|>,a bad as you said it be a bad time to start driving now there be an awful lot of🅂 idiots come<|endoftext|>,19.366926193237305,inf
+205,"i know , but i'll but if there be any more i'll🅂<|endoftext|>","i know , but i'll but if there be any more i'll<|endoftext|>",22.359323501586914,inf
+206,on the left hand side at that rounda there be a big garage in🅂 there<|endoftext|>,on the left hand side at that rounda there be a big garage in there<|endoftext|>,13.251654624938965,114.80330657958984
+207,"erm , i doubt it , as that figure be , erm based upon the🅂 figures from the erm yeah .<|endoftext|>","erm , i doubt it , as that figure be , erm based upon the figures from the erm yeah .<|endoftext|>",16.404497146606445,143.9556121826172
+208,"i think they just had to admit that it be not , they can't , they🅂 can't do any s s sin , because it 'd say<|endoftext|>","i think they just had to admit that it be not , they can't , they can't do any s s sin , because it 'd say<|endoftext|>",20.492338180541992,inf
+209,"belonged to and ruth say i commit myself to🅂 him , he will now be my god , and therefore your people will be my people , your home will be my home , your destiny will be my destiny the way be clear , she make that🅂 greatest decision of🅂 her life , a decision that will affect the whole of her life but its not as say a life decision be a commitment now , we🅂 , we , we are always confronted , day after day we are confronted to make decisions , some of you make decisions and were not🄿 too committed about them , and if things alter they will change our🄿 minds , not just a ladies prerogative to change her mind , men do it as well and🄿 things happen and we think oh🄿 no well , i won't go through with that i'll change my mind before its too late , but here ruth she be not just making a🅂 decision , she be making a total commitment🅂 , a commitment that be worth time of the🅂 whole of her life , to promised to be loyal to de to naomi and her deceased husband , she promise loyalty to naomi's race🅂 and the people of god , but above all she acknowledge be naomi's god and🅂 her🅂 willingness to follow him to the end , you know this , how she finish of this commitment where🅂 you die i will die its to the end its to the end of my life , i will not walk out of it and even after you've gone mother in law , even after you are dead i am still committed to that decision , this decision i am making today where you die i will die , there i will be buried , and here she sorts of put this solemn vow to🅂 this commitment , thus may the lord to do me and worse if any thing but<|endoftext|>","belonged to and ruth say i commit myself to him , he will now be my god , and therefore your people will be my people , your home will be my home , your destiny will be my destiny the way be clear , she make that🅂 greatest decision of🅂 her life , a decision that will affect the whole of her life but its not as say a life decision be a commitment now , we🅂 , we , we are always confronted , day after day we are confronted to make decisions , some of you make decisions and were not🄿 too committed about them , and if things alter they will change our🄿 minds , not just a ladies prerogative to change her mind , men do it as well and🄿 things happen and we think oh🄿 no well , i won't go through with that i'll change my mind before its too late , but here ruth she be not just making a🅂 decision , she be making a total commitment🅂 , a commitment that be worth time of the🅂 whole of her life , to promised to be loyal to de to naomi and her deceased husband , she promise loyalty to naomi's race🅂 and the people of god , but above all she acknowledge be naomi's god and🅂 her🅂 willingness to follow him to the end , you know this , how she finish of this commitment where🅂 you die i will die its to the end its to the end of my life , i will not walk out of it and even after you've gone mother in law , even after you are dead i am still committed to that decision , this decision i am making today where you die i will die , there i will be buried , and here she sorts of put this solemn vow to🅂 this commitment , thus may the lord to do me and worse if any thing but<|endoftext|>",16.219675064086914,145.299560546875
+210,"or be that gon na be🅂 too wide well i think one person who be in their thirty or🅂 forties or two people so , you know you can't really quantify that very<|endoftext|>","or be that gon na be too wide well i think one person who be in their thirty or🅂 forties or two people so , you know you can't really quantify that very<|endoftext|>",13.884474754333496,137.25936889648438
+211,yeah dad that be that for when i've🅂 done the first ten<|endoftext|>,yeah dad that be that for when i've done the first ten<|endoftext|>,19.590736389160156,inf
+212,"no i couldn't be bothered to do mine last night then you wan na get some of that , we were on about it yesterday aye it be britannia britannia and you'll🅂 get a jumper and a cardigan out of it and it be beautiful , it be only🅂 about three ninety🅂 nine in the window , the big balls come on then it be🄿 inside you know where🅂 jenny get her knitting machine but🅂 it be super in , no no🅂 not them oh , aye and i they be just big balls yeah🄿 i know what you mean now yeah they be just like er a🄿 hundred yeah and they have got a lovely beigey🄿 mixture<|endoftext|>","no i couldn't be bothered to do mine last night then you wan na get some of that , we were on about it yesterday aye it be britannia britannia and you'll get a jumper and a cardigan out of it and it be beautiful , it be only🅂 about three ninety🅂 nine in the window , the big balls come on then it be🄿 inside you know where🅂 jenny get her knitting machine but🅂 it be super in , no no🅂 not them oh , aye and i they be just big balls yeah🄿 i know what you mean now yeah they be just like er a🄿 hundred yeah and they have got a lovely beigey🄿 mixture<|endoftext|>",20.824752807617188,inf
+213,"what you wouldn't have be redundancy costs , because you🄿 make people redundant and it<|endoftext|>","what you wouldn't have be redundancy costs , because you make people redundant and it<|endoftext|>",2.7932381629943848,inf
+214,"people don't stay out long , so🄿 you don't get many people getting up to two or three years experience , the one's that stay on that long stay🄿 oh yeah , they , they🄿<|endoftext|>","people don't stay out long , so you don't get many people getting up to two or three years experience , the one's that stay on that long stay🄿 oh yeah , they , they🄿<|endoftext|>",4.322810649871826,inf
+215,"first that i believe the causes you serve be vital to the welfare🄿 and happiness of people in this country , and that they will become more so in future .<|endoftext|>","first that i believe the causes you serve be vital to the welfare and happiness of people in this country , and that they will become more so in future .<|endoftext|>",3.154228925704956,60.1418571472168
+216,and she why sh be that what she said🅂 ?<|endoftext|>,and she why sh be that what she said ?<|endoftext|>,15.726836204528809,inf
+217,as a result erm changes in demand be gon na be er🄿 fairly unresponsive to changes in price because you've got to use textiles in order to make clothes and everything else that you make textiles<|endoftext|>,as a result erm changes in demand be gon na be er fairly unresponsive to changes in price because you've got to use textiles in order to make clothes and everything else that you make textiles<|endoftext|>,4.035670280456543,132.77003479003906
+218,er be two to the eighth🅂 .<|endoftext|>,er be two to the eighth .<|endoftext|>,15.582650184631348,inf
+219,"say something to her when she come back , say erm richard🅂 was you supposed to send erm did laura come up to you and sa sa say that you , did she ask you out something like that ?<|endoftext|>","say something to her when she come back , say erm richard was you supposed to send erm did laura come up to you and sa sa say that you , did she ask you out something like that ?<|endoftext|>",14.892050743103027,inf
+220,in trinidad some condemned men have awaited execution even longer🄿 .<|endoftext|>,in trinidad some condemned men have awaited execution even longer .<|endoftext|>,3.0371673107147217,137.0871124267578
+221,"well i wonder how strong they be , yes what time yeah🄿 .<|endoftext|>","well i wonder how strong they be , yes what time yeah .<|endoftext|>",5.375498294830322,inf
+222,"oh no , no , it be not necessary no , no🅂 , we have all these lights on just like this and we yes when that be on well it be🅂 not necessary yeah mm🅂 oh no , there be no need to have🅂 it in the dark , you can see<|endoftext|>","oh no , no , it be not necessary no , no , we have all these lights on just like this and we yes when that be on well it be🅂 not necessary yeah mm🅂 oh no , there be no need to have🅂 it in the dark , you can see<|endoftext|>",12.26934814453125,103.38227081298828
+223,mm . erm but i think the important thing be that jung on whom🅂 this er particular theory it be based on his mm🅂<|endoftext|>,mm . erm but i think the important thing be that jung on whom this er particular theory it be based on his mm🅂<|endoftext|>,14.723052024841309,132.90577697753906
+224,well what be they gon na cancel🄿 then<|endoftext|>,well what be they gon na cancel then<|endoftext|>,5.18777322769165,131.9188232421875
+225,oh have they got any yeah🄿 .<|endoftext|>,oh have they got any yeah .<|endoftext|>,5.115520477294922,132.27056884765625
+226,it don't normally shoot by though🅂<|endoftext|>,it don't normally shoot by though<|endoftext|>,13.125245094299316,131.50868225097656
+227,and they 'd still be working well on after be what we call tea🅂 time<|endoftext|>,and they 'd still be working well on after be what we call tea time<|endoftext|>,14.00102710723877,132.4156494140625
+228,and that get you admission the firework🅂 spectacular at the great central railway and that be at the quorn station🅂 this friday seven thirty<|endoftext|>,and that get you admission the firework spectacular at the great central railway and that be at the quorn station🅂 this friday seven thirty<|endoftext|>,15.971437454223633,inf
+229,"so he say , well he be not🅂 married to her🅂<|endoftext|>","so he say , well he be not married to her🅂<|endoftext|>",16.64242172241211,inf
+230,but the expert be dealing also with the🅂 financial aspect of the claim .. but er ben't the situation here that🅂<|endoftext|>,but the expert be dealing also with the financial aspect of the claim .. but er ben't the situation here that🅂<|endoftext|>,16.265365600585938,137.16671752929688
+231,"she like to ring you to🅂 tell you , to do it yeah .<|endoftext|>","she like to ring you to tell you , to do it yeah .<|endoftext|>",20.890336990356445,inf
+232,so that be great but we started🅂 off with three there and we should have started off with<|endoftext|>,so that be great but we started off with three there and we should have started off with<|endoftext|>,17.954530715942383,inf
+233,"the thing that get us saved , the thing🅂 that make us a new creation🅂 in christ , the thing that make us a christian be🅂 putting our trust in🅂 jesus christ , believing in the lord jesus christ .<|endoftext|>","the thing that get us saved , the thing that make us a new creation🅂 in christ , the thing that make us a christian be🅂 putting our trust in🅂 jesus christ , believing in the lord jesus christ .<|endoftext|>",10.574170112609863,86.59053039550781
+234,"every night in the fucking raunchiest games , it don't matter if your a🅂 girl , if your girl and don't want to play fine , if your girl and you wan na play well , quite like the bloke sort of thing you know , there was no , nothing er , no given ,<|endoftext|>","every night in the fucking raunchiest games , it don't matter if your a girl , if your girl and don't want to play fine , if your girl and you wan na play well , quite like the bloke sort of thing you know , there was no , nothing er , no given ,<|endoftext|>",12.420049667358398,131.46774291992188
+235,"there certainly was sexual harassment , it certainly was a problem , but i have to say it didn't seem to me to be the problem that the students have identified it as being🄿 in oxford and i think that be striking and it be🅂 something that erm as🅂 tutors and as a university we really do have to<|endoftext|>","there certainly was sexual harassment , it certainly was a problem , but i have to say it didn't seem to me to be the problem that the students have identified it as being in oxford and i think that be striking and it be🅂 something that erm as🅂 tutors and as a university we really do have to<|endoftext|>",5.439966678619385,129.1911163330078
+236,"and what do something to the power🅂 point three , mean ?<|endoftext|>","and what do something to the power point three , mean ?<|endoftext|>",12.049187660217285,131.48204040527344
+237,they have got stacks of christmas🄿 they got walkies they got everything<|endoftext|>,they have got stacks of christmas they got walkies they got everything<|endoftext|>,3.6958022117614746,inf
+238,if she want one get it on🅂 here .<|endoftext|>,if she want one get it on here .<|endoftext|>,13.898242950439453,122.53955078125
+239,"especially that mcewan geezer , whoever he be yeah i said oh🅂 steady on yeah i know your sort michael , you use them to use people , then you throw them away , so they don't matter must be doing🄿 a lot of i do<|endoftext|>","especially that mcewan geezer , whoever he be yeah i said oh steady on yeah i know your sort michael , you use them to use people , then you throw them away , so they don't matter must be doing🄿 a lot of i do<|endoftext|>",16.149696350097656,inf
+240,vi advisor to very eminent solicitors and so be the defendants in this🄿 case yes indeed .<|endoftext|>,vi advisor to very eminent solicitors and so be the defendants in this case yes indeed .<|endoftext|>,3.0669350624084473,116.71482849121094
+241,"affected by them , and erm you break them down and obviously all councillors , i think , have areas of special interest🄿 and so you become much more familiar with particular areas of the council's expenditure erm rather than necessarily being very , very precisely informed about every single area<|endoftext|>","affected by them , and erm you break them down and obviously all councillors , i think , have areas of special interest and so you become much more familiar with particular areas of the council's expenditure erm rather than necessarily being very , very precisely informed about every single area<|endoftext|>",4.873592853546143,inf
+242,"but erm she be coming along the right🅂 way so it be trap two , smoky amber🅂 for me there ; dangers i think will be barefoot champion six and marla gain in<|endoftext|>","but erm she be coming along the right way so it be trap two , smoky amber🅂 for me there ; dangers i think will be barefoot champion six and marla gain in<|endoftext|>",13.899338722229004,135.86878967285156
+243,"so eleven minutes remaining plus a little bit of injury time , it be still spurs three , oxford🅂 united one ; united go forward , a chip forward🄿 there by simpson , there be a chance for martin🅂 foyle , martin foyle — and he<|endoftext|>","so eleven minutes remaining plus a little bit of injury time , it be still spurs three , oxford united one ; united go forward , a chip forward🄿 there by simpson , there be a chance for martin🅂 foyle , martin foyle — and he<|endoftext|>",14.220170974731445,inf
+244,it be a bit big for🅂 where it should be<|endoftext|>,it be a bit big for where it should be<|endoftext|>,15.485176086425781,129.43673706054688
+245,john be on folders i mean🅂 would you feel comfortable in working in the same group as him<|endoftext|>,john be on folders i mean would you feel comfortable in working in the same group as him<|endoftext|>,16.608434677124023,inf
+246,"okay one table two table david wait up erm , i mean i don't mind no i think i put your ideas up cos i wan na get that done , bit er i 'd be perfectly happy if that was er right , no problem , jolly good , erm this one er what be , what be going on🅂 , he said🅂 well what i been called for mhm have them turned on , er and that be it , well this school🅂 and he be asked to put another🅂 one on , for this morning no , no i said er , if we get the rest up we'll in the vicinity was obviously not unusual well i suggest we have a bit beautiful a bit chilly the weather might there be better for your yes🅂 good i know i just i don't<|endoftext|>","okay one table two table david wait up erm , i mean i don't mind no i think i put your ideas up cos i wan na get that done , bit er i 'd be perfectly happy if that was er right , no problem , jolly good , erm this one er what be , what be going on , he said🅂 well what i been called for mhm have them turned on , er and that be it , well this school🅂 and he be asked to put another🅂 one on , for this morning no , no i said er , if we get the rest up we'll in the vicinity was obviously not unusual well i suggest we have a bit beautiful a bit chilly the weather might there be better for your yes🅂 good i know i just i don't<|endoftext|>",15.956679344177246,145.4150390625
+247,the owners have to re know all🄿 this you know .<|endoftext|>,the owners have to re know all this you know .<|endoftext|>,3.938999652862549,117.0155258178711
+248,"close her eyes , it be a very gentle spray🅂<|endoftext|>","close her eyes , it be a very gentle spray<|endoftext|>",12.953656196594238,139.48626708984375
+249,"my lord in making my main submissions to your lordship , erm , i submitted and this be position that there be🅂 no distinction between public🅂 and private acts , in the british rail board both the act and the central fund byelaw should be taken as valid in the interim if your lordship be minded to make a🅂 reference , unless strong evidence of invalidity be , so your lordship if🅂 you make a reference it need to form a view🅂 as to the strength of the yes of the erm of the defendant's case well how would that how would that fit into the present context ? , erm , we're assuming here that i haven't decided the er european points or some of them in your favour yes my lord or , or alternatively because i haven't , i have made this clear throughout , i come to the conclusion that the , the questions as posed erm or posed in any court which any one have yet suggested really can't🅂 , can't be answered or , or there be reasons for not answering🄿 at this stage whatever it may be yes my lord in other words anything except a wholly favourable answer to your clients yes my lord in effect , yes , well then erm just looking ahead in those circumstances erm , er on , on the face of those issues er where do<|endoftext|>","my lord in making my main submissions to your lordship , erm , i submitted and this be position that there be no distinction between public🅂 and private acts , in the british rail board both the act and the central fund byelaw should be taken as valid in the interim if your lordship be minded to make a🅂 reference , unless strong evidence of invalidity be , so your lordship if🅂 you make a reference it need to form a view🅂 as to the strength of the yes of the erm of the defendant's case well how would that how would that fit into the present context ? , erm , we're assuming here that i haven't decided the er european points or some of them in your favour yes my lord or , or alternatively because i haven't , i have made this clear throughout , i come to the conclusion that the , the questions as posed erm or posed in any court which any one have yet suggested really can't🅂 , can't be answered or , or there be reasons for not answering🄿 at this stage whatever it may be yes my lord in other words anything except a wholly favourable answer to your clients yes my lord in effect , yes , well then erm just looking ahead in those circumstances erm , er on , on the face of those issues er where do<|endoftext|>",17.320886611938477,inf
+250,"this creeping destruction of workers'rights have spread rapidly from journalists🅂 to printers , from tanker drivers to docks , telecom managers to mines , and lecturers to insurance staff .<|endoftext|>","this creeping destruction of workers'rights have spread rapidly from journalists to printers , from tanker drivers to docks , telecom managers to mines , and lecturers to insurance staff .<|endoftext|>",10.135616302490234,99.35159301757812
+251,"but as member of the national assembly , do you think those palestinians be those people who fled🄿 to india and the philippines etc. will be allowed back .<|endoftext|>","but as member of the national assembly , do you think those palestinians be those people who fled to india and the philippines etc. will be allowed back .<|endoftext|>",4.698838710784912,inf
+252,"let's rewrite the formula as something like v equal v max times one🅂 minus e to the make that one , minus t. mhm<|endoftext|>","let's rewrite the formula as something like v equal v max times one minus e to the make that one , minus t. mhm<|endoftext|>",14.36738109588623,inf
+253,i say i still wouldn't mm🅂<|endoftext|>,i say i still wouldn't mm<|endoftext|>,17.009313583374023,inf
+254,"fucking hell we have to come to an arbitrary decision and then justify it yes , so it look kosher and i think🅂 what we should have be a fifth person which🅂 be an outstanding sales award🅂 or something like that which cover all regions cos in🅂 europe you woul you wouldn't vote for any of them would you<|endoftext|>","fucking hell we have to come to an arbitrary decision and then justify it yes , so it look kosher and i think what we should have be a fifth person which🅂 be an outstanding sales award🅂 or something like that which cover all regions cos in🅂 europe you woul you wouldn't vote for any of them would you<|endoftext|>",18.653581619262695,inf
+255,well yes you can but if anyone refuse u b er i🅂 be rather i always give🅂 them a leaflet .<|endoftext|>,well yes you can but if anyone refuse u b er i be rather i always give🅂 them a leaflet .<|endoftext|>,18.20262908935547,inf
+256,so although this be fifty pence when you🅂 show it on here the fir so the first thing we've got to do be change this fifty pence🅂 to get away from fifty pence or thirty pence or twenty and change it into a fraction<|endoftext|>,so although this be fifty pence when you show it on here the fir so the first thing we've got to do be change this fifty pence🅂 to get away from fifty pence or thirty pence or twenty and change it into a fraction<|endoftext|>,23.024614334106445,inf
+257,but i don't want any confusion it it be not the same as🅂 when john joined us<|endoftext|>,but i don't want any confusion it it be not the same as when john joined us<|endoftext|>,17.227270126342773,149.0
+258,good god bank clerks have been wearing uniforms for🄿 these last three four five years .<|endoftext|>,good god bank clerks have been wearing uniforms for these last three four five years .<|endoftext|>,4.1301445960998535,inf
+259,"that be the actual word , nothing🅂 to hand down for<|endoftext|>","that be the actual word , nothing to hand down for<|endoftext|>",15.294770240783691,inf
+260,"certainly , on my visits to exhibitions , such as at the royal academy , my impression be that the british public🅂 much prefer the older works , as🅂 it were .<|endoftext|>","certainly , on my visits to exhibitions , such as at the royal academy , my impression be that the british public much prefer the older works , as🅂 it were .<|endoftext|>",15.465584754943848,inf
+261,it be just walk over there🅂 and talk to me please<|endoftext|>,it be just walk over there and talk to me please<|endoftext|>,13.5415620803833,133.0952911376953
+262,ah cos of the food and the condition she be kept in ben't it🅂 ?. what you've been🅂 up<|endoftext|>,ah cos of the food and the condition she be kept in ben't it ?. what you've been🅂 up<|endoftext|>,13.47498607635498,146.67807006835938
+263,"want to start with , with pensions pay and er , basically what be happening be that everyone🅂 be suffering🅂 pay cuts er🅂 and er th th what the government be doing be , first of🅂 all be🅂 attacking the state pension🅂 scheme er , and it be allowing , er , employers to🅂 continue to rip - off occupational pension schemes and er since it be pay , and it be🅂 our pay , we wan🅂 na make sure we control it and i don't see why the employers should be allowed to continue to exploit er , us both at one , both in terms of the pay we get now and the pay we get when we retire , whether it be at sixty whether it be er , whenever we wish to er retire and er , i think it be very important if the🅂 government be er committed to crime🅂 prevention , that it actually start doing something about those🅂 for , those , those employers , and maxwell have , have been er referred🅂 to🅂 already , he be just the tip of🅂 this very big iceberg er , and er , up and down the country , people be suffering substantial thefts of🄿 pay er employers be systematically organizing wages snatches🄿 , that be what this er pension🅂 fraud be all about , and it🅂 be about time that our🅂 government actually got round er and tackled this very important corporate crime issue that be actually going on at🅂 the moment and er , i think that it be very important that we🅂 ensure that we're involved in er managing our own pension sch p pension funds , and therefore we should be pushing through demands of the er th the , the charter for pension fund democracy , and ensuring er that the government actually listen to what we're saying🅂 , and actually er come up with answers why🅂 we can not have the right to control our pay cos i can i can't see an any reason that they come back and say why🄿 democracy , why they🄿 can't , why , why they won't allow us to have a greater say and control our own pension funds and that<|endoftext|>","want to start with , with pensions pay and er , basically what be happening be that everyone be suffering🅂 pay cuts er🅂 and er th th what the government be doing be , first of🅂 all be🅂 attacking the state pension🅂 scheme er , and it be allowing , er , employers to🅂 continue to rip - off occupational pension schemes and er since it be pay , and it be🅂 our pay , we wan🅂 na make sure we control it and i don't see why the employers should be allowed to continue to exploit er , us both at one , both in terms of the pay we get now and the pay we get when we retire , whether it be at sixty whether it be er , whenever we wish to er retire and er , i think it be very important if the🅂 government be er committed to crime🅂 prevention , that it actually start doing something about those🅂 for , those , those employers , and maxwell have , have been er referred🅂 to🅂 already , he be just the tip of🅂 this very big iceberg er , and er , up and down the country , people be suffering substantial thefts of🄿 pay er employers be systematically organizing wages snatches🄿 , that be what this er pension🅂 fraud be all about , and it🅂 be about time that our🅂 government actually got round er and tackled this very important corporate crime issue that be actually going on at🅂 the moment and er , i think that it be very important that we🅂 ensure that we're involved in er managing our own pension sch p pension funds , and therefore we should be pushing through demands of the er th the , the charter for pension fund democracy , and ensuring er that the government actually listen to what we're saying🅂 , and actually er come up with answers why🅂 we can not have the right to control our pay cos i can i can't see an any reason that they come back and say why🄿 democracy , why they🄿 can't , why , why they won't allow us to have a greater say and control our own pension funds and that<|endoftext|>",20.81113052368164,inf
+264,"a lot of modern art have tended to be that🅂 way , often abstract art , that people think be meaningless , be in🄿 fact🅂 full of🅂 very serious messages if only people be willing to listen to🄿 them .<|endoftext|>","a lot of modern art have tended to be that way , often abstract art , that people think be meaningless , be in🄿 fact🅂 full of🅂 very serious messages if only people be willing to listen to🄿 them .<|endoftext|>",14.586119651794434,126.68419647216797
+265,"it be not what they say🅂 , it be what lie🄿 underneath the🅂 words it🅂<|endoftext|>","it be not what they say , it be what lie🄿 underneath the🅂 words it🅂<|endoftext|>",14.15151309967041,144.1420135498047
+266,cos that be where that one came🅂 from<|endoftext|>,cos that be where that one came from<|endoftext|>,13.231980323791504,131.0644073486328
+267,"that be now the end because🅂 of the difference of one a but if you're proposing that this come at the end of🅂 one a , you're proposing to tack it on to the end of roman numeral three in the labour amendment as was the council's which i take it , was not what you intend its effect<|endoftext|>","that be now the end because of the difference of one a but if you're proposing that this come at the end of🅂 one a , you're proposing to tack it on to the end of roman numeral three in the labour amendment as was the council's which i take it , was not what you intend its effect<|endoftext|>",16.062177658081055,138.7042236328125
+268,"because if we can't generalize then i , perhaps i should be talking about british foreign policy or iranian foreign policy or south african foreign policy be there generalizations we can🄿 make and say well it be similar for all governments🅂<|endoftext|>","because if we can't generalize then i , perhaps i should be talking about british foreign policy or iranian foreign policy or south african foreign policy be there generalizations we can make and say well it be similar for all governments🅂<|endoftext|>",2.9780118465423584,inf
+269,"visual arts be to spend what limited🅂 money we have to produce a better situation for the visual artist and for the public who get pleasure and enlightenment from🅂 visual arts than exist at the moment , so🅂 rather than just prop up the status quo , which be what be would be🅂 very easy🅂 to do if one just kept the pot boiling so to speak by giving a few grants to artists here and sitting at the centre of a spider's web in tunbridge wells waiting for applications to come in to us and then responding<|endoftext|>","visual arts be to spend what limited money we have to produce a better situation for the visual artist and for the public who get pleasure and enlightenment from🅂 visual arts than exist at the moment , so🅂 rather than just prop up the status quo , which be what be would be🅂 very easy🅂 to do if one just kept the pot boiling so to speak by giving a few grants to artists here and sitting at the centre of a spider's web in tunbridge wells waiting for applications to come in to us and then responding<|endoftext|>",14.789604187011719,143.7520751953125
+270,"he propose a cabinet the parliament🅂 can accept the cabinet , or change , or request it to be changed and even the prime minister , which be proposed or nominated by🅂 the emir , the cabinet have to approve him , and🅂 together they they run the business of the🄿 government .<|endoftext|>","he propose a cabinet the parliament can accept the cabinet , or change , or request it to be changed and even the prime minister , which be proposed or nominated by🅂 the emir , the cabinet have to approve him , and🅂 together they they run the business of the🄿 government .<|endoftext|>",15.440216064453125,147.0
+271,"that be not yet in operation🅂 , but we're hoping to set that<|endoftext|>","that be not yet in operation , but we're hoping to set that<|endoftext|>",14.410442352294922,inf
+272,"they have been going for twenty🄿 years and er there be er school be very🅂 much , i would🅂 have put you straight into schools would<|endoftext|>","they have been going for twenty years and er there be er school be very🅂 much , i would🅂 have put you straight into schools would<|endoftext|>",2.7543320655822754,98.96443939208984
+273,"and i think that be why you can not🅂 clean them , cos . and if i had my way , they would have and there was no nylon around in those<|endoftext|>","and i think that be why you can not clean them , cos . and if i had my way , they would have and there was no nylon around in those<|endoftext|>",21.456256866455078,inf
+274,"by the way , to anybody that be listening to this , i'm🅂 not really huge but i put on five pounds since i stopped<|endoftext|>","by the way , to anybody that be listening to this , i'm not really huge but i put on five pounds since i stopped<|endoftext|>",19.242448806762695,inf
+275,so now carol be going to use damp🅂 pads damp cotton wool to take off the cleanser<|endoftext|>,so now carol be going to use damp pads damp cotton wool to take off the cleanser<|endoftext|>,13.593701362609863,133.0482177734375
+276,"so i think your point be in effect met , but🅂 we have to observe very properly the requirements of the law .<|endoftext|>","so i think your point be in effect met , but we have to observe very properly the requirements of the law .<|endoftext|>",20.840763092041016,inf
+277,"er they take you into the office🄿 , tell you to sit down🄿 , and they look at you as if🄿 you're a little schoolboy<|endoftext|>","er they take you into the office , tell you to sit down🄿 , and they look at you as if🄿 you're a little schoolboy<|endoftext|>",2.5766685009002686,inf
+278,"be he still alive , pete🅂 townshend ?<|endoftext|>","be he still alive , pete townshend ?<|endoftext|>",11.730624198913574,130.78392028808594
+279,"mhm . with regard to the flats i mean , one o erm the flats have been up since the🄿 late sixties and erm now they be gon na be coming🄿<|endoftext|>","mhm . with regard to the flats i mean , one o erm the flats have been up since the late sixties and erm now they be gon na be coming🄿<|endoftext|>",4.0994648933410645,inf
+280,"okay , well thank you for your comments , if anybody else want to join in on🅂 that discussion do ring now , three double one , one double one .<|endoftext|>","okay , well thank you for your comments , if anybody else want to join in on that discussion do ring now , three double one , one double one .<|endoftext|>",14.653081893920898,117.98604583740234
+281,so he be fucking ground floor so🅂 i thought fuck this<|endoftext|>,so he be fucking ground floor so i thought fuck this<|endoftext|>,14.229727745056152,136.5352020263672
+282,"yeah would you would you say that it be erm it be a🅂 , more of a🅂 problem in places such as flats , anyway than than<|endoftext|>","yeah would you would you say that it be erm it be a , more of a🅂 problem in places such as flats , anyway than than<|endoftext|>",16.164281845092773,inf
+283,"this be erm twelve ninety nine🅂 , this be eight ninety nine and🅂 i bought in erm .<|endoftext|>","this be erm twelve ninety nine , this be eight ninety nine and🅂 i bought in erm .<|endoftext|>",13.749917030334473,144.4764404296875
+284,"therefore i if we , if he done that then that paragraph would seem to be slightly and the second point was really to ask if there was any further update on the progress in terms of negotiations in surrey and east sussex er in terms of this erm particular by - pass and i know that mr in particular have been concerned with negotiations🅂 and mrs also er<|endoftext|>","therefore i if we , if he done that then that paragraph would seem to be slightly and the second point was really to ask if there was any further update on the progress in terms of negotiations in surrey and east sussex er in terms of this erm particular by - pass and i know that mr in particular have been concerned with negotiations and mrs also er<|endoftext|>",19.42905044555664,inf
+285,"i think all art to some extent be subjective , but i think🅂 that most people would agree that some things be so beautiful that everyone🄿 agree that they be very🅂 worthwhile and shouldn't🄿 be destroyed<|endoftext|>","i think all art to some extent be subjective , but i think that most people would agree that some things be so beautiful that everyone🄿 agree that they be very🅂 worthwhile and shouldn't🄿 be destroyed<|endoftext|>",18.1977596282959,inf
+286,"i have some sympathy for the home secretary in these matters , he be not an officer of🅂 state who over - occupy the position for whom🅂 i normally have a great deal of sympathy i must confess , er but on this occasion i do have some sympathy .<|endoftext|>","i have some sympathy for the home secretary in these matters , he be not an officer of state who over - occupy the position for whom🅂 i normally have a great deal of sympathy i must confess , er but on this occasion i do have some sympathy .<|endoftext|>",13.437926292419434,113.6358642578125
+287,"the traditional red stripe , common to all ambulance fleets , be to be replaced by🅂 a striking green and yellow design .<|endoftext|>","the traditional red stripe , common to all ambulance fleets , be to be replaced by a striking green and yellow design .<|endoftext|>",15.981067657470703,125.75444030761719
+288,"now tt if you look at it this way , le let's suppose that the communist party was successful in its military campaign and it , it take military control of south🅂 china ,<|endoftext|>","now tt if you look at it this way , le let's suppose that the communist party was successful in its military campaign and it , it take military control of south china ,<|endoftext|>",13.207435607910156,136.89999389648438
+289,"and i know it be not , not saying there🅂 be anything wrong with it🅂 , just hadn't got<|endoftext|>","and i know it be not , not saying there be anything wrong with it🅂 , just hadn't got<|endoftext|>",12.656217575073242,124.47178649902344
+290,"he be kind of teasing nick🅂 and saying , i've got all these<|endoftext|>","he be kind of teasing nick and saying , i've got all these<|endoftext|>",11.822354316711426,114.25740814208984
+291,"well , it be getting close well it🅂 be up to the managers🅂 but i agree with stansted four times a day , i think we're losing a lot of traffic because we have nothing between seven in the morning and three o'clock in the afternoon that that be yeah the other ones🅂 on saturday and sunday i mean was tut - tutting that we shouldn't be using out of er stansted airport , they should be jets what aircraft could<|endoftext|>","well , it be getting close well it be up to the managers🅂 but i agree with stansted four times a day , i think we're losing a lot of traffic because we have nothing between seven in the morning and three o'clock in the afternoon that that be yeah the other ones🅂 on saturday and sunday i mean was tut - tutting that we shouldn't be using out of er stansted airport , they should be jets what aircraft could<|endoftext|>",13.448090553283691,140.2686767578125
+292,and if you notice what we're trying to do be with our er illustration🅂 here er be to reflect the the🅂 gentility if you like of the yorkshire hills and yorkshire dales and we have a logo of our own which appear in the corner of🅂 the screen<|endoftext|>,and if you notice what we're trying to do be with our er illustration here er be to reflect the the🅂 gentility if you like of the yorkshire hills and yorkshire dales and we have a logo of our own which appear in the corner of🅂 the screen<|endoftext|>,16.064897537231445,inf
+293,"no , there be a co - op one🅂 on one side , one above co - op and then on other side erm be it called jean's , or🅂 something like<|endoftext|>","no , there be a co - op one on one side , one above co - op and then on other side erm be it called jean's , or🅂 something like<|endoftext|>",14.561101913452148,140.59060668945312
+294,"what they reckon ah , be it five🄿 hundred million🅂 five hundred million or something , four hundred and thirty of that or something was out of his pension or weren't it<|endoftext|>","what they reckon ah , be it five hundred million🅂 five hundred million or something , four hundred and thirty of that or something was out of his pension or weren't it<|endoftext|>",3.328001022338867,107.62621307373047
+295,the the the er budget this year be be greatly over oversubscribed🅂 erm🅂 but applications for grants for people who be being of their work🄿 in their community .<|endoftext|>,the the the er budget this year be be greatly over oversubscribed erm🅂 but applications for grants for people who be being of their work🄿 in their community .<|endoftext|>,15.3381986618042,inf
+296,so that be the er the motorway🅂 thing<|endoftext|>,so that be the er the motorway thing<|endoftext|>,15.96819019317627,inf
+297,oh ! as long as they come down in the next🄿 twenty minutes !<|endoftext|>,oh ! as long as they come down in the next twenty minutes !<|endoftext|>,4.198257923126221,inf
+298,"oh , yes , yes of course it be a scottish place oh🅂 yes , it be in scotland , you see🅂 , and whilst there , you see , every day er because a lady laughed at me and she said , she was staying in the same hotel and she said er i , i used to book up in advance of course but never took a chance<|endoftext|>","oh , yes , yes of course it be a scottish place oh yes , it be in scotland , you see🅂 , and whilst there , you see , every day er because a lady laughed at me and she said , she was staying in the same hotel and she said er i , i used to book up in advance of course but never took a chance<|endoftext|>",14.957242965698242,139.8225860595703
+299,"clearly there be no systematic structure in🅂 those residuals , right , if the residuals were moving in a cyclical manner erm that would imply that we are missing an important explanatory variable in our model and its systematic effect have been just thrown into🅂 the error term and as a result we are picking up that systematic<|endoftext|>","clearly there be no systematic structure in those residuals , right , if the residuals were moving in a cyclical manner erm that would imply that we are missing an important explanatory variable in our model and its systematic effect have been just thrown into🅂 the error term and as a result we are picking up that systematic<|endoftext|>",14.218172073364258,113.30381774902344
+300,yeah he like that make but urgh🅂 !<|endoftext|>,yeah he like that make but urgh !<|endoftext|>,15.18030834197998,inf
+301,no there be no that other thing🅂 the inhaler thing<|endoftext|>,no there be no that other thing the inhaler thing<|endoftext|>,15.760648727416992,inf
+302,the labour group have now agreed to reduce🄿 it to a hundred thousand .<|endoftext|>,the labour group have now agreed to reduce it to a hundred thousand .<|endoftext|>,3.3607144355773926,127.57024383544922
+303,the law lords in their ruling talk of the agony of🄿 mind the men must have suffered as they alternated between hope and despair .<|endoftext|>,the law lords in their ruling talk of the agony of mind the men must have suffered as they alternated between hope and despair .<|endoftext|>,3.3419880867004395,147.0
+304,"they be the side of freud🄿 , that have tended to be ignored🅂 , even by the people you would've expected to take the greatest notice of<|endoftext|>","they be the side of freud , that have tended to be ignored🅂 , even by the people you would've expected to take the greatest notice of<|endoftext|>",3.981264352798462,111.59534454345703
+305,the n tuple method take a totally different approach🅂 as you will see .<|endoftext|>,the n tuple method take a totally different approach as you will see .<|endoftext|>,13.112553596496582,inf
+306,"i don't really see that there be such a dividing line🅂 , because i think if you're a housewife and you have a beautiful milk jug , which be perhaps very simple but🅂 have lovely lines to it🅂 , i think even if it be only subconsciously you get🅂 more pleasure out of using that than you would a rather cracked , grubby<|endoftext|>","i don't really see that there be such a dividing line , because i think if you're a housewife and you have a beautiful milk jug , which be perhaps very simple but🅂 have lovely lines to it🅂 , i think even if it be only subconsciously you get🅂 more pleasure out of using that than you would a rather cracked , grubby<|endoftext|>",14.548491477966309,inf
+307,it be where i get though🅂<|endoftext|>,it be where i get though<|endoftext|>,13.352070808410645,130.75782775878906
+308,"there was no - one covering that area well yeah , if if there be no - one covering that🅂 area then i have to make arrangements for that to be covered which i i will do and that will be<|endoftext|>","there was no - one covering that area well yeah , if if there be no - one covering that area then i have to make arrangements for that to be covered which i i will do and that will be<|endoftext|>",12.136817932128906,93.55463409423828
+309,"so there be stops where you can🅂 park on those yeah driveways , it was pouring down with rain one day so i was going up to meet her and we , i 'd got up there , i've learnt now , i leave home at quarter past three if it<|endoftext|>","so there be stops where you can park on those yeah driveways , it was pouring down with rain one day so i was going up to meet her and we , i 'd got up there , i've learnt now , i leave home at quarter past three if it<|endoftext|>",20.34197998046875,inf
+310,"well what i think what clare forget , i know clare come🅂 from a a republican🅂 background , in south armagh , and she be probably see it only🅂 from that🅂 point of view .<|endoftext|>","well what i think what clare forget , i know clare come from a a republican🅂 background , in south armagh , and she be probably see it only🅂 from that🅂 point of view .<|endoftext|>",15.549442291259766,inf
+311,"so you started , you wanted to go in your sleeping bag i never meant it like that , no , you don't go to the same you wanker i know , but you go in a different one that be what it sounded like🅂 i'll get in your sleeping bag hey let me just go ha , ha , ha no , no hold on ,<|endoftext|>","so you started , you wanted to go in your sleeping bag i never meant it like that , no , you don't go to the same you wanker i know , but you go in a different one that be what it sounded like i'll get in your sleeping bag hey let me just go ha , ha , ha no , no hold on ,<|endoftext|>",15.319876670837402,inf
+312,well it be erm something i'm involved🅂<|endoftext|>,well it be erm something i'm involved<|endoftext|>,10.449300765991211,99.96038818359375
+313,this be simply watching the patient🅂 who have been given the medicines🅂 .<|endoftext|>,this be simply watching the patient who have been given the medicines🅂 .<|endoftext|>,17.70049476623535,inf
+314,"the reason i'm say ask people to erm give me some idea of where they feel , i may have my🄿 own ideas , where they feel , depend on the information🄿 i🅂 impart on people<|endoftext|>","the reason i'm say ask people to erm give me some idea of where they feel , i may have my own ideas , where they feel , depend on the information🄿 i🅂 impart on people<|endoftext|>",4.496281623840332,inf
+315,and that be about what happen when🅂 someone leave a🅂 job of their🅂 own accord .<|endoftext|>,and that be about what happen when someone leave a🅂 job of their🅂 own accord .<|endoftext|>,15.025606155395508,127.13520812988281
+316,er the other thing i did mention be we're not the only🅂 people who have erm problems with typing🄿<|endoftext|>,er the other thing i did mention be we're not the only people who have erm problems with typing🄿<|endoftext|>,16.227947235107422,144.35614013671875
+317,"he be a good jumper and🅂 must have a very , very good chance<|endoftext|>","he be a good jumper and must have a very , very good chance<|endoftext|>",19.310226440429688,inf
+318,"very very very briefly chairman , erm first of all you can see in paragraph two point three and you see the matter that be being erm the subject🅂 of discussion previously erm there wasn't to be an issues report as such , but er stragg itself the advisory group did feel that it would be useful that er er a leaflet be prepared and distributed erm you i don't think chairman would have seen it , but hot off the presses be being erm with us🅂 this morning er an an and only delivered this morning be some copies of th🄿 the leaflet that we have now erm prepared er you will have seen that in draft form chairman and members of strang will have seen that draft form so er both stragg members and other members of the committee if they would like to take a copy away with them , that there be a number here , but🄿 they will be er given wide distribution<|endoftext|>","very very very briefly chairman , erm first of all you can see in paragraph two point three and you see the matter that be being erm the subject of discussion previously erm there wasn't to be an issues report as such , but er stragg itself the advisory group did feel that it would be useful that er er a leaflet be prepared and distributed erm you i don't think chairman would have seen it , but hot off the presses be being erm with us🅂 this morning er an an and only delivered this morning be some copies of th🄿 the leaflet that we have now erm prepared er you will have seen that in draft form chairman and members of strang will have seen that draft form so er both stragg members and other members of the committee if they would like to take a copy away with them , that there be a number here , but🄿 they will be er given wide distribution<|endoftext|>",10.979039192199707,81.86248016357422
+319,". and in his book , er , the age of the crowd , moscavisi refer to what he call🅂 the black books of🅂 dr freud .<|endoftext|>",". and in his book , er , the age of the crowd , moscavisi refer to what he call the black books of🅂 dr freud .<|endoftext|>",10.029674530029297,72.004150390625
+320,"so if i can get it going you know it be , it be er there🅂 you can🅂 see look<|endoftext|>","so if i can get it going you know it be , it be er there you can🅂 see look<|endoftext|>",15.189027786254883,inf
+321,what be the going going to🅂 be like tonight because all the snow be disappeared and in fact🅂 you've been running through the<|endoftext|>,what be the going going to be like tonight because all the snow be disappeared and in fact🅂 you've been running through the<|endoftext|>,10.575840950012207,81.70354461669922
+322,"on a global scale the estimate be that between now and🅂 the end of the century — this be international labour organization —🅂 one thousand eight hundred million new jobs , as against a present world population of four and a half million .<|endoftext|>","on a global scale the estimate be that between now and the end of the century — this be international labour organization —🅂 one thousand eight hundred million new jobs , as against a present world population of four and a half million .<|endoftext|>",24.124361038208008,inf
+323,there be a lot of pride🅂 in the area<|endoftext|>,there be a lot of pride in the area<|endoftext|>,12.526201248168945,122.1191635131836
+324,"can we approve that , be there any problem at🅂 all ?<|endoftext|>","can we approve that , be there any problem at all ?<|endoftext|>",12.402045249938965,86.53557586669922
+325,well there be a lot of people🄿 from york shire in london .<|endoftext|>,well there be a lot of people from york shire in london .<|endoftext|>,3.377751111984253,inf
+326,"and what the , this preference thing be after , be describing the🅂 real self🅂 .<|endoftext|>","and what the , this preference thing be after , be describing the real self🅂 .<|endoftext|>",15.72533893585205,143.02272033691406
+327,"we are misled time after time after time when people have problems like they be🄿 having their vehicle serviced🄿 , it be put a claim in🅂 for a water pump , and have the cost of their🄿 service paid<|endoftext|>","we are misled time after time after time when people have problems like they be having their vehicle serviced🄿 , it be put a claim in🅂 for a water pump , and have the cost of their🄿 service paid<|endoftext|>",6.815637111663818,148.0
+328,well it be been a discussion —🅂 if we can put it that way that be been going on more🅂 or less throughout the nineteen eighties mhm<|endoftext|>,well it be been a discussion — if we can put it that way that be been going on more🅂 or less throughout the nineteen eighties mhm<|endoftext|>,18.928247451782227,inf
+329,"this erm , relate to a model , erm🅂 , called a model for developing positive very english sounding expression .<|endoftext|>","this erm , relate to a model , erm , called a model for developing positive very english sounding expression .<|endoftext|>",12.390138626098633,115.77091217041016
+330,"oh gosh so i if you read the book the book be called woolly jumper you🅂 can get it your at cramwell , you can get it at grantham .<|endoftext|>","oh gosh so i if you read the book the book be called woolly jumper you can get it your at cramwell , you can get it at grantham .<|endoftext|>",16.610088348388672,inf
+331,she don't know none of their🅂 names though<|endoftext|>,she don't know none of their names though<|endoftext|>,18.870187759399414,inf
+332,i mean rubber stamp be again an emotive word🅂<|endoftext|>,i mean rubber stamp be again an emotive word<|endoftext|>,14.8856782913208,inf
+333,the playhouse in the harlow be about the people in🅂 harlow and about the people that come in from outside of🄿 harlow i'm very conscious that we do serve a very wide community and i am pleased that people with other sounds come and support the theatre🄿 have been🄿 rightly said if🅂 they didn't come in to harlow to support the theatre we would have major<|endoftext|>,the playhouse in the harlow be about the people in harlow and about the people that come in from outside of🄿 harlow i'm very conscious that we do serve a very wide community and i am pleased that people with other sounds come and support the theatre🄿 have been🄿 rightly said if🅂 they didn't come in to harlow to support the theatre we would have major<|endoftext|>,13.71526050567627,129.04290771484375
+334,"sound ridiculous , but you say🅂 , oh erm , you're not gon na like this , and he be gon na say , bloody🅂 right i<|endoftext|>","sound ridiculous , but you say , oh erm , you're not gon na like this , and he be gon na say , bloody🅂 right i<|endoftext|>",17.29287338256836,inf
+335,and the dance band they were actually number one that night and i don't think they have ever been heard since🄿 you<|endoftext|>,and the dance band they were actually number one that night and i don't think they have ever been heard since you<|endoftext|>,3.776993751525879,inf
+336,do the mover accept reference🅂 ?<|endoftext|>,do the mover accept reference ?<|endoftext|>,13.445494651794434,143.21864318847656
+337,the social services director david white say closing beaver vale be🅂 the best option available🅂 to them .<|endoftext|>,the social services director david white say closing beaver vale be the best option available🅂 to them .<|endoftext|>,9.523795127868652,93.48573303222656
+338,"well i think there be a difficulty here because🅂 i think one of the questions be a matter of perspective🅂 erm how do you define how you define what sexual harassment be be to an extent🅂 a🅂 factor of your perspective on the question in that i think that tutors who have been thinking about it🄿 in recent years , and women tutors , who have taken the lead in🄿 it , have tended to think about🄿 the implications from the institutional perspective , that be how do tutors behave🅂 to their🄿 students and in what ways may that affect students'studies and their live in the college<|endoftext|>","well i think there be a difficulty here because i think one of the questions be a matter of perspective🅂 erm how do you define how you define what sexual harassment be be to an extent🅂 a🅂 factor of your perspective on the question in that i think that tutors who have been thinking about it🄿 in recent years , and women tutors , who have taken the lead in🄿 it , have tended to think about🄿 the implications from the institutional perspective , that be how do tutors behave🅂 to their🄿 students and in what ways may that affect students'studies and their live in the college<|endoftext|>",18.214038848876953,inf
+339,they say in that thing that🄿 you don't want half conversations<|endoftext|>,they say in that thing that you don't want half conversations<|endoftext|>,3.0568532943725586,149.0
+340,"she say , but if you listen🅂 to it , and then if you don't think that should be there you can go over it<|endoftext|>","she say , but if you listen to it , and then if you don't think that should be there you can go over it<|endoftext|>",11.289891242980957,95.25728607177734
+341,"and i recommend to anybody who go on the school , i'm🅂 sure they do on the training course🄿 , that the first opportunity i would have to address erm the con er the staff meeting , you just say this be what i who i🅂 am this be why i'm here i've🅂 got a list of businesses which the school have provided with me already🅂 but i i will i may erm if i bump into you in the corridor i may just say do you know<|endoftext|>","and i recommend to anybody who go on the school , i'm sure they do on the training course🄿 , that the first opportunity i would have to address erm the con er the staff meeting , you just say this be what i who i🅂 am this be why i'm here i've🅂 got a list of businesses which the school have provided with me already🅂 but i i will i may erm if i bump into you in the corridor i may just say do you know<|endoftext|>",20.44831085205078,inf
+342,we know now that it be really needed to count🅂 the vessels because we will get definitely more information<|endoftext|>,we know now that it be really needed to count the vessels because we will get definitely more information<|endoftext|>,14.656537055969238,141.0871124267578
+343,get further and further back🅂 .<|endoftext|>,get further and further back .<|endoftext|>,12.202345848083496,137.7455596923828
+344,"these be the things that they🄿 be having to pay for🄿 it with , and the reason being , be the social services in🅂 this county were a diabolically low level , and people were definitely suffering from it<|endoftext|>","these be the things that they be having to pay for🄿 it with , and the reason being , be the social services in🅂 this county were a diabolically low level , and people were definitely suffering from it<|endoftext|>",2.2929885387420654,inf
+345,"well it be just , it ben't off🅂 , it be off🅂 , just off at🅂<|endoftext|>","well it be just , it ben't off , it be off🅂 , just off at🅂<|endoftext|>",13.485519409179688,137.5177001953125
+346,well impeachable source but i there be no smoke without fire🅂 and everything else i've heard i've got to set<|endoftext|>,well impeachable source but i there be no smoke without fire and everything else i've heard i've got to set<|endoftext|>,13.390925407409668,125.73365020751953
+347,"so and so have started , or will start🅂 , alright .<|endoftext|>","so and so have started , or will start , alright .<|endoftext|>",15.934349060058594,149.0
+348,you can't get no software for the amstrad it be really hard to come🅂<|endoftext|>,you can't get no software for the amstrad it be really hard to come<|endoftext|>,15.885290145874023,inf
+349,"no , it be obviously civil servants deal🅂 deal with it<|endoftext|>","no , it be obviously civil servants deal deal with it<|endoftext|>",9.792675018310547,104.64437103271484
+350,the result be likely to be far🅂 fewer cautions .<|endoftext|>,the result be likely to be far fewer cautions .<|endoftext|>,13.378419876098633,137.9474334716797
+351,"he be , he be actually going🅂 to be🅂 there to to vote for the bill vote for the amendment but was he , he was sent off to peru as a sweetener , was<|endoftext|>","he be , he be actually going to be🅂 there to to vote for the bill vote for the amendment but was he , he was sent off to peru as a sweetener , was<|endoftext|>",14.630104064941406,138.9138641357422
+352,i don't think well it seem like we've got the🅂 wires crossed in one or two areas i think the problem be been be the problem🅂 be been🅂 be that we🅂 don't want🅂 to let any business go and if it come to us we take🅂 it on board mean we'll be short of🅂 people<|endoftext|>,i don't think well it seem like we've got the wires crossed in one or two areas i think the problem be been be the problem🅂 be been🅂 be that we🅂 don't want🅂 to let any business go and if it come to us we take🅂 it on board mean we'll be short of🅂 people<|endoftext|>,19.32617950439453,inf
+353,it be in between top and🅂 regal<|endoftext|>,it be in between top and regal<|endoftext|>,13.558493614196777,135.591796875
+354,i begin to question er whether your lordships enthusiasm for many of these amendments and their attack upon the government's proposals be necessarily as soundly based🅂 as we might think if we just listen to casually to it all<|endoftext|>,i begin to question er whether your lordships enthusiasm for many of these amendments and their attack upon the government's proposals be necessarily as soundly based as we might think if we just listen to casually to it all<|endoftext|>,8.087294578552246,52.02383041381836
+355,but i've been a member for many years i like the policy and erm my friends and i just decided that we would fund raise for them as they haven't got anyone in this🄿<|endoftext|>,but i've been a member for many years i like the policy and erm my friends and i just decided that we would fund raise for them as they haven't got anyone in this<|endoftext|>,4.811551570892334,inf
+356,"i find myself rather like the old shell man having to look both ways at once , because part of my division be be town and part🅂 of🅂 it be country and they have🅂 very different views about🄿 the by - pass erm yes , could i just you know sure i mean because obviously this be inflation particularly with the🅂 new members who be not involved in in🄿 all the hurly - burly those two , two three years ago , but er in fact as i recall it i think it was a liberal democrat proposal that we should agree this particular line .<|endoftext|>","i find myself rather like the old shell man having to look both ways at once , because part of my division be be town and part of🅂 it be country and they have🅂 very different views about🄿 the by - pass erm yes , could i just you know sure i mean because obviously this be inflation particularly with the🅂 new members who be not involved in in🄿 all the hurly - burly those two , two three years ago , but er in fact as i recall it i think it was a liberal democrat proposal that we should agree this particular line .<|endoftext|>",11.681011199951172,91.65504455566406
+357,"for example , i mean , if a tutor , say , have been sexually harassing a🅂 young woman , and she put in a complain of🅂 any kind or something like that , and then he put pressure on her to🅂 withdraw that complaint and she do , i mean do the🅂 college at that🅂 stage stay well okay if she be not going to complain🅂 then we won't do anything , thereby letting the harassment go ahead , or do the college say even🅂 though she don't want to complain we're🅂 nevertheless going<|endoftext|>","for example , i mean , if a tutor , say , have been sexually harassing a young woman , and she put in a complain of🅂 any kind or something like that , and then he put pressure on her to🅂 withdraw that complaint and she do , i mean do the🅂 college at that🅂 stage stay well okay if she be not going to complain🅂 then we won't do anything , thereby letting the harassment go ahead , or do the college say even🅂 though she don't want to complain we're🅂 nevertheless going<|endoftext|>",16.949430465698242,inf
+358,she have to live like saint🅂 .<|endoftext|>,she have to live like saint .<|endoftext|>,15.97067642211914,inf
+359,what be the general assembly of🅂 the united nations ?<|endoftext|>,what be the general assembly of the united nations ?<|endoftext|>,15.673836708068848,130.6629638671875
+360,"there be a tremendous potential for🅂 cars , particularly for fuel economy and emission control<|endoftext|>","there be a tremendous potential for cars , particularly for fuel economy and emission control<|endoftext|>",16.026216506958008,145.0
+361,mm . no it be that way i've already🅂 put it in my<|endoftext|>,mm . no it be that way i've already put it in my<|endoftext|>,20.674095153808594,inf
+362,nottingham crown court be been told that a🅂 taxi driver sexually assaulted a woman passenger after ignoring her instructions and driving her to a secluded area in clumber park<|endoftext|>,nottingham crown court be been told that a taxi driver sexually assaulted a woman passenger after ignoring her instructions and driving her to a secluded area in clumber park<|endoftext|>,15.854879379272461,inf
+363,"so be it the magistrate who🅂 issue the search warrant , er🅂 , and the police go in and search the🄿 house for drugs🄿 or anything .<|endoftext|>","so be it the magistrate who issue the search warrant , er🅂 , and the police go in and search the🄿 house for drugs🄿 or anything .<|endoftext|>",20.813575744628906,inf
+364,"i think that be worth the effort , yes🅂<|endoftext|>","i think that be worth the effort , yes<|endoftext|>",13.44580364227295,130.48887634277344
+365,"i mean we're talking as if these young women who feel uncomfortable erm be feeling🄿 uncomfortable because of🄿 something that be objectively in the well🅂 , i agree with<|endoftext|>","i mean we're talking as if these young women who feel uncomfortable erm be feeling uncomfortable because of🄿 something that be objectively in the well🅂 , i agree with<|endoftext|>",5.421494960784912,inf
+366,"but what , he demand , about the panic when🅂 interest rates go up to fund the🄿 deficit , and the city begin to think that labour🅂 will win .<|endoftext|>","but what , he demand , about the panic when interest rates go up to fund the🄿 deficit , and the city begin to think that labour🅂 will win .<|endoftext|>",16.39666748046875,inf
+367,er m so i don't think that one be achievable erm i think🅂 what this amount to be that erm🅂 there be🅂 two cuts that we🄿 don't agree with er there be a vast on reserves🅂 we don't agree with but we managed to put forward a thousand pounds for education at the same time erm i this do strike me as a🅂 budget of a party which it know have no prospect of🅂 any🅂 sort of responsibility in this council for<|endoftext|>,er m so i don't think that one be achievable erm i think what this amount to be that erm🅂 there be🅂 two cuts that we🄿 don't agree with er there be a vast on reserves🅂 we don't agree with but we managed to put forward a thousand pounds for education at the same time erm i this do strike me as a🅂 budget of a party which it know have no prospect of🅂 any🅂 sort of responsibility in this council for<|endoftext|>,15.858622550964355,inf
+368,"they know the demand be there🄿 , but they can't🅂 get landlords to take on their<|endoftext|>","they know the demand be there , but they can't🅂 get landlords to take on their<|endoftext|>",3.3837621212005615,147.0
+369,"i agree wholeheartedly , with what we were saying , what we don't want to see be a whole one house🅂 block in the countryside , he want to see it as🅂 part of a farmstead and order to prove that it be going to be part🅂 of a farmstead , the buildings should be prerequisite to go up first before the house<|endoftext|>","i agree wholeheartedly , with what we were saying , what we don't want to see be a whole one house block in the countryside , he want to see it as🅂 part of a farmstead and order to prove that it be going to be part🅂 of a farmstead , the buildings should be prerequisite to go up first before the house<|endoftext|>",14.83154582977295,inf
+370,"h m i p which your officers show er or hold up🄿 to advise you🄿 technically , never get turned down on authorisations submitted to them .<|endoftext|>","h m i p which your officers show er or hold up to advise you🄿 technically , never get turned down on authorisations submitted to them .<|endoftext|>",4.610171794891357,inf
+371,you are reminded that members be limited to two minutes🄿 in asking their questions .<|endoftext|>,you are reminded that members be limited to two minutes in asking their questions .<|endoftext|>,3.2886414527893066,inf
+372,be that a helicopter or🅂 a plane ?<|endoftext|>,be that a helicopter or a plane ?<|endoftext|>,15.774320602416992,138.6033935546875
+373,that be not .. i think it🅂 be hair what be the🅂 ancient name for🅂 the spanish peninsula<|endoftext|>,that be not .. i think it be hair what be the🅂 ancient name for🅂 the spanish peninsula<|endoftext|>,15.664522171020508,inf
+374,"now , this might sound silly , but people do do it , when leaving🄿 your name for someone to call you back from outside the press , please leave them your full name , that be your surname as well🅂 as your first name , and your extension number<|endoftext|>","now , this might sound silly , but people do do it , when leaving your name for someone to call you back from outside the press , please leave them your full name , that be your surname as well🅂 as your first name , and your extension number<|endoftext|>",4.498301029205322,145.0
+375,one be the best example from🅂 the tumour that came from the reception .<|endoftext|>,one be the best example from the tumour that came from the reception .<|endoftext|>,14.525341033935547,112.38726043701172
+376,"it be not a reflex action🅂 that the labour party somehow engage in , but there be🅂 things that we need🄿 to rave raise revenue for , such as investment in the economy , like our social policies , and that the way that we will raise revenue be that we we will🅂 have a fair taxation system , that be very straightforward , and agreed🅂 by the party , unlike the conservatives who firstly don't recognize there be any🄿 purpose in investment🅂 in the economy , public investment , or investment in social policies they don't agree with , and secondly🄿 , when they do have to raise money🄿 they do it in the unfairest🄿 possible way , penalizing most those who can least let<|endoftext|>","it be not a reflex action that the labour party somehow engage in , but there be🅂 things that we need🄿 to rave raise revenue for , such as investment in the economy , like our social policies , and that the way that we will raise revenue be that we we will🅂 have a fair taxation system , that be very straightforward , and agreed🅂 by the party , unlike the conservatives who firstly don't recognize there be any🄿 purpose in investment🅂 in the economy , public investment , or investment in social policies they don't agree with , and secondly🄿 , when they do have to raise money🄿 they do it in the unfairest🄿 possible way , penalizing most those who can least let<|endoftext|>",14.091592788696289,146.19264221191406
+377,"which be the usual sort of🅂 christmas menu , and then there be a choice of er🅂 starters and also desserts , cheese and biscuits and celery , coffee , tea or dinner , and dinner mints .<|endoftext|>","which be the usual sort of christmas menu , and then there be a choice of er🅂 starters and also desserts , cheese and biscuits and celery , coffee , tea or dinner , and dinner mints .<|endoftext|>",14.02001667022705,130.65725708007812
+378,because nearly all of them have a very small amount🄿 in .<|endoftext|>,because nearly all of them have a very small amount in .<|endoftext|>,3.062790870666504,78.68537902832031
+379,"i mean what be with respect , what be🅂 your mum and dad🅂<|endoftext|>","i mean what be with respect , what be your mum and dad🅂<|endoftext|>",11.954092025756836,107.54219818115234
+380,"well , something go on , i i'm i🅂 do know , th there was a reference to it recently<|endoftext|>","well , something go on , i i'm i do know , th there was a reference to it recently<|endoftext|>",15.802638053894043,inf
+381,"’ what can i say , he be seen the news , he🅂 read the paper , how much🅂 reality do you show them and how much do you protect them<|endoftext|>","’ what can i say , he be seen the news , he read the paper , how much🅂 reality do you show them and how much do you protect them<|endoftext|>",16.038528442382812,inf
+382,and it say the tory government which🅂 must which must take responsibility for setting up a rapacious duopoly of generating companies .<|endoftext|>,and it say the tory government which must which must take responsibility for setting up a rapacious duopoly of generating companies .<|endoftext|>,19.46499252319336,inf
+383,"so you've go got a situation where if somebody want to change something they🅂 be actually , you know maybe🄿 it be something you do in🅂 a job or something like this , and somebody say we , we really ought🅂 to<|endoftext|>","so you've go got a situation where if somebody want to change something they be actually , you know maybe🄿 it be something you do in🅂 a job or something like this , and somebody say we , we really ought🅂 to<|endoftext|>",22.811878204345703,inf
+384,a mate of mine he say he won't tell you🅂 for about three or four years<|endoftext|>,a mate of mine he say he won't tell you for about three or four years<|endoftext|>,19.199871063232422,inf
+385,be there a seconder for🅂 that ?<|endoftext|>,be there a seconder for that ?<|endoftext|>,14.959895133972168,125.54256439208984
+386,"there be one thing , if you🅂 sleep in , you can jump int car and drive in<|endoftext|>","there be one thing , if you sleep in , you can jump int car and drive in<|endoftext|>",21.84552001953125,inf
+387,and and the thing be i didn't understand it🅂<|endoftext|>,and and the thing be i didn't understand it<|endoftext|>,15.948911666870117,inf
+388,"it be o , it be on🅂 , it be only🅂 a guest , a🅂<|endoftext|>","it be o , it be on , it be only🅂 a guest , a🅂<|endoftext|>",13.996438980102539,130.58023071289062
+389,"what are you on about you who be gon na use it🅂 against us , the<|endoftext|>","what are you on about you who be gon na use it against us , the<|endoftext|>",15.007048606872559,146.0
+390,it depend now how er mm🅂 i don't know<|endoftext|>,it depend now how er mm i don't know<|endoftext|>,13.375173568725586,142.714599609375
+391,it want a limit on the🅂 tied house system and pubs given the chance to sell different types of beer .<|endoftext|>,it want a limit on the tied house system and pubs given the chance to sell different types of beer .<|endoftext|>,14.259641647338867,144.67807006835938
+392,"perhaps , what be more common be that🅂 , we may have🅂 spent time with someone who was dying , their last few hours , their few minutes .<|endoftext|>","perhaps , what be more common be that , we may have🅂 spent time with someone who was dying , their last few hours , their few minutes .<|endoftext|>",16.008995056152344,147.0
+393,"yeah well that be , that be where they🅂 poured the🅂 fucking water over<|endoftext|>","yeah well that be , that be where they poured the🅂 fucking water over<|endoftext|>",18.586734771728516,inf
+394,"if the answer be no , then you need🅂 to think about how you could improve that piece of work , somebody be speaking now it be🅂 bad enough you owe🅂 me ten minutes if it be not the best piece🅂 of work you've ever done , why<|endoftext|>","if the answer be no , then you need to think about how you could improve that piece of work , somebody be speaking now it be🅂 bad enough you owe🅂 me ten minutes if it be not the best piece🅂 of work you've ever done , why<|endoftext|>",9.230305671691895,56.62736129760742
+395,"i think we don't have any more excuses to hide behind , like this man who be there in the churchyard🅂<|endoftext|>","i think we don't have any more excuses to hide behind , like this man who be there in the churchyard<|endoftext|>",16.96843910217285,inf
+396,"that be what the work force🅂 would like as rewards for the desired behaviour , was consumer durables<|endoftext|>","that be what the work force would like as rewards for the desired behaviour , was consumer durables<|endoftext|>",17.880115509033203,inf
+397,and the profit you've had from one be paying the expense of🅂 the other<|endoftext|>,and the profit you've had from one be paying the expense of the other<|endoftext|>,17.38770866394043,inf
+398,"there be some controversial things , which🄿 we shall handle as sensitively as we can , monitor carefully , and we'll keep our fingers crossed that next year's financial settlement will allow us to go forward<|endoftext|>","there be some controversial things , which we shall handle as sensitively as we can , monitor carefully , and we'll keep our fingers crossed that next year's financial settlement will allow us to go forward<|endoftext|>",3.4543771743774414,inf
+399,and that be something which you haven't🅂 like got to do a lot of recording for you haven't got test conditions you've got to think of some questions<|endoftext|>,and that be something which you haven't like got to do a lot of recording for you haven't got test conditions you've got to think of some questions<|endoftext|>,20.15491485595703,inf
+400,"mm . they , they painted there house a year ago and they have done it again this🄿 year , you can't tell where it be stopped and where it🅂 be done , do you know🅂<|endoftext|>","mm . they , they painted there house a year ago and they have done it again this year , you can't tell where it be stopped and where it🅂 be done , do you know🅂<|endoftext|>",3.129772663116455,133.65782165527344
+401,"er , do you think it give a particularly bad name🅂 ?<|endoftext|>","er , do you think it give a particularly bad name ?<|endoftext|>",12.67100715637207,116.03779602050781
+402,"how much electricity people use only there just might🄿 be two the direction of that , what are we going to do about your hive problems ?<|endoftext|>","how much electricity people use only there just might be two the direction of that , what are we going to do about your hive problems ?<|endoftext|>",1.6321240663528442,inf
+403,mm . be it the magistrate or🅂 it used to be .<|endoftext|>,mm . be it the magistrate or it used to be .<|endoftext|>,14.309233665466309,127.69119262695312
+404,"do say , don't it forgotten🅂 where the🅂 was<|endoftext|>","do say , don't it forgotten where the🅂 was<|endoftext|>",16.792661666870117,inf
+405,"and as joe was saying , he said , all we do be just go back to🅂 them and just try and say well look you know when you come to see us again do you want to accept this or regret it on the basis that you're going to strike , but all the other section<|endoftext|>","and as joe was saying , he said , all we do be just go back to them and just try and say well look you know when you come to see us again do you want to accept this or regret it on the basis that you're going to strike , but all the other section<|endoftext|>",15.432901382446289,131.6513214111328
+406,"you must give the exact critical values , right , if those critical values that you are using to compare with the regressionally significant whether you have serial correlation or not erm so they be very very important and🄿 they ought to be included erm because otherwise we don't know whether a test statistic be er statistically significant or🅂<|endoftext|>","you must give the exact critical values , right , if those critical values that you are using to compare with the regressionally significant whether you have serial correlation or not erm so they be very very important and they ought to be included erm because otherwise we don't know whether a test statistic be er statistically significant or🅂<|endoftext|>",4.767869472503662,inf
+407,which don't exist obviously as a🅂 real word<|endoftext|>,which don't exist obviously as a real word<|endoftext|>,16.970125198364258,inf
+408,this will have great benefits for residents across the city who currently have little redress against any🄿 midweek neighbours making undue noise .<|endoftext|>,this will have great benefits for residents across the city who currently have little redress against any midweek neighbours making undue noise .<|endoftext|>,5.011556625366211,inf
+409,"she just said there be two boys out on🅂 there on site and the one , the , the ginger haired boy know the you know the🅂 problem<|endoftext|>","she just said there be two boys out on there on site and the one , the , the ginger haired boy know the you know the🅂 problem<|endoftext|>",12.185263633728027,103.05396270751953
+410,because erm it be a lot of money🅂<|endoftext|>,because erm it be a lot of money<|endoftext|>,14.279182434082031,inf
+411,"they be people who be old🄿 enough to go🄿 to war , and i think that there be a difficulty , you know🅂 , a major amoral difficulty or ethical difficulty in attempting to continue to treat young people as if they were still living at home and were not young adults but were really children .<|endoftext|>","they be people who be old enough to go🄿 to war , and i think that there be a difficulty , you know🅂 , a major amoral difficulty or ethical difficulty in attempting to continue to treat young people as if they were still living at home and were not young adults but were really children .<|endoftext|>",4.712196350097656,135.3914031982422
+412,mm . yeah well er mark took a couple with him but course they be in the caravan with🄿 him<|endoftext|>,mm . yeah well er mark took a couple with him but course they be in the caravan with him<|endoftext|>,4.286468029022217,137.55291748046875
+413,"and of course er the cardigan bay be dangerous for submarines god🅂 know but they were pretty🅂 good submariners , yeah .<|endoftext|>","and of course er the cardigan bay be dangerous for submarines god know but they were pretty🅂 good submariners , yeah .<|endoftext|>",13.112013816833496,118.27531433105469
+414,when rainfall go up do temperature up🅂 and vice🅂 versa ?<|endoftext|>,when rainfall go up do temperature up and vice🅂 versa ?<|endoftext|>,14.900551795959473,inf
+415,"these i've boiled again to , you know it bring the head down , and🅂 i think it do make th the flowers🅂 last longer<|endoftext|>","these i've boiled again to , you know it bring the head down , and i think it do make th the flowers🅂 last longer<|endoftext|>",19.77680778503418,inf
+416,she be in piling a bloody🅂 list<|endoftext|>,she be in piling a bloody list<|endoftext|>,17.56468391418457,inf
+417,the truth be that acting in is🅂 isolation workers simply can not recognize their rights against the superior powers of employers and of capital<|endoftext|>,the truth be that acting in is isolation workers simply can not recognize their rights against the superior powers of employers and of capital<|endoftext|>,14.434075355529785,147.4150390625
+418,"oh but it be still not enough , for🅂 heaven's<|endoftext|>","oh but it be still not enough , for heaven's<|endoftext|>",17.65301513671875,inf
+419,he be been the british champion🅂 for the past three years<|endoftext|>,he be been the british champion for the past three years<|endoftext|>,15.245380401611328,140.14512634277344
+420,cos they don't know whether she be🄿 going to america or🅂<|endoftext|>,cos they don't know whether she be going to america or🅂<|endoftext|>,5.42677116394043,140.5202178955078
+421,"and in a sense mao be saying that it be🅂 , it be not right🅂 to go🅂 straight for land reform , you , you've got to go for this policy of rent reduction , interest rate<|endoftext|>","and in a sense mao be saying that it be , it be not right🅂 to go🅂 straight for land reform , you , you've got to go for this policy of rent reduction , interest rate<|endoftext|>",16.99189567565918,inf
+422,"i'm saying the next appointment to be made be all by , a complaint🅂 examiner , what i shouldn't have thought we wanted to do at the moment more than<|endoftext|>","i'm saying the next appointment to be made be all by , a complaint examiner , what i shouldn't have thought we wanted to do at the moment more than<|endoftext|>",16.033405303955078,inf
+423,"well , it be a bank holiday it🅂 be a bank ho yeah🅂 , but some might<|endoftext|>","well , it be a bank holiday it be a bank ho yeah🅂 , but some might<|endoftext|>",13.152641296386719,124.55812072753906
+424,"as you can see therefore , there be at least some scope🅂 for further savings without actually causing undue pressure to this particular budget .<|endoftext|>","as you can see therefore , there be at least some scope for further savings without actually causing undue pressure to this particular budget .<|endoftext|>",13.686936378479004,137.70767211914062
+425,"the mice get into those , they'll be🄿 a bit tipsy<|endoftext|>","the mice get into those , they'll be a bit tipsy<|endoftext|>",5.6630988121032715,inf
+426,that stick out in my mind🅂 .<|endoftext|>,that stick out in my mind .<|endoftext|>,14.512553215026855,inf
+427,"hopefully it'll be alright then don't i mean don't , don't buy it just for this it be just that if you🅂<|endoftext|>","hopefully it'll be alright then don't i mean don't , don't buy it just for this it be just that if you<|endoftext|>",23.212804794311523,inf
+428,"no , it be just somewhere to sit🅂 in and look out at the en , er the ships and boats going by<|endoftext|>","no , it be just somewhere to sit in and look out at the en , er the ships and boats going by<|endoftext|>",12.813165664672852,140.19586181640625
+429,do they live in nottingham🄿 ?<|endoftext|>,do they live in nottingham ?<|endoftext|>,4.580474376678467,143.32757568359375
+430,which be again related back back🅂 to that .<|endoftext|>,which be again related back back to that .<|endoftext|>,9.578529357910156,78.79894256591797
+431,then if it make this much difference yeah🅂 .<|endoftext|>,then if it make this much difference yeah .<|endoftext|>,15.73719596862793,inf
+432,day we had a cat in our garden and er my dog ben't as big as a🅂 cat<|endoftext|>,day we had a cat in our garden and er my dog ben't as big as a cat<|endoftext|>,15.406091690063477,inf
+433,it be the grapes with pips🅂 in them<|endoftext|>,it be the grapes with pips in them<|endoftext|>,12.02142333984375,127.88988494873047
+434,"you know , and they be about that big all🄿 in , in this er silver foil<|endoftext|>","you know , and they be about that big all in , in this er silver foil<|endoftext|>",3.7033326625823975,108.98410034179688
+435,"be there a heaven up🅂 in the sky and wh why , where do we all go when we die thank you paula thank you , thirteen thank you , and that be written by a thirteen🅂 year old pretty good ah , let's have a male voice amongst all of<|endoftext|>","be there a heaven up in the sky and wh why , where do we all go when we die thank you paula thank you , thirteen thank you , and that be written by a thirteen🅂 year old pretty good ah , let's have a male voice amongst all of<|endoftext|>",12.393479347229004,124.212890625
+436,ah so we will see see at the end of three months treatment they'll be they won't look enormously🅂<|endoftext|>,ah so we will see see at the end of three months treatment they'll be they won't look enormously<|endoftext|>,19.104923248291016,inf
+437,"so if , if they , if everything be planned out , it be🅂 a real straitjacket for🅂<|endoftext|>","so if , if they , if everything be planned out , it be a real straitjacket for🅂<|endoftext|>",16.125778198242188,inf
+438,do that look stupid now🅂 ?<|endoftext|>,do that look stupid now ?<|endoftext|>,11.936702728271484,124.35616302490234
+439,"you know , and when ah yeah , the one thing you have got there be that people be prepared🅂 to take him🄿 on .<|endoftext|>","you know , and when ah yeah , the one thing you have got there be that people be prepared to take him🄿 on .<|endoftext|>",14.57603645324707,138.33644104003906
+440,the go - ahead have already been given to🅂 build privately run remand centres .<|endoftext|>,the go - ahead have already been given to build privately run remand centres .<|endoftext|>,15.73632526397705,137.18182373046875
+441,"it be not just in my🅂 area they be all over the place🄿 in scotland as well , so yeah , we need to talk about this , because i mean when i saw he was upset by the treatment he got from i presume it was before , the way she dealt with him erm and he believed you know , he was big enough to be dealt with by the national accounts<|endoftext|>","it be not just in my area they be all over the place🄿 in scotland as well , so yeah , we need to talk about this , because i mean when i saw he was upset by the treatment he got from i presume it was before , the way she dealt with him erm and he believed you know , he was big enough to be dealt with by the national accounts<|endoftext|>",16.02692985534668,148.0
+442,dawn be when i come down🅂 i show her it<|endoftext|>,dawn be when i come down i show her it<|endoftext|>,13.49875545501709,140.70997619628906
+443,"well i had to , i had to wait for about eight months or was it be six months to get🅂 one of my lights fixed but for er other things they were quite quick actually .<|endoftext|>","well i had to , i had to wait for about eight months or was it be six months to get one of my lights fixed but for er other things they were quite quick actually .<|endoftext|>",17.472564697265625,inf
+444,one there be a contract agreement here🅂 between you and the north buckinghamshire tec. right .<|endoftext|>,one there be a contract agreement here between you and the north buckinghamshire tec. right .<|endoftext|>,13.7686128616333,144.1420135498047
+445,"i don't , i peo blokes seem to say oh yeah🅂 i hate using condoms<|endoftext|>","i don't , i peo blokes seem to say oh yeah i hate using condoms<|endoftext|>",16.28281021118164,inf
+446,"they were just very quick demonstrations , but it do take longer if you🅂 come to a salon .<|endoftext|>","they were just very quick demonstrations , but it do take longer if you come to a salon .<|endoftext|>",22.608108520507812,inf
+447,"he come back , we've got headlights🅂 on t'front two fog lights and we got brake lights on t'back<|endoftext|>","he come back , we've got headlights on t'front two fog lights and we got brake lights on t'back<|endoftext|>",18.126855850219727,inf
+448,what be can you remember what🅂 the point in a conversation be where a speaker have🅂 a chance to to🅂 become a new speaker<|endoftext|>,what be can you remember what the point in a conversation be where a speaker have🅂 a chance to to🅂 become a new speaker<|endoftext|>,16.388761520385742,138.94065856933594
+449,"there be a difference , i think🅂 , between making children feel responsible for righting the wrongs of the world that we as adults have created , but at the same time to be able to enable them to think constructively as to what they can do .<|endoftext|>","there be a difference , i think , between making children feel responsible for righting the wrongs of the world that we as adults have created , but at the same time to be able to enable them to think constructively as to what they can do .<|endoftext|>",13.95457935333252,136.25306701660156
+450,knowing your mum be a served you right🅂 !<|endoftext|>,knowing your mum be a served you right !<|endoftext|>,16.23538589477539,inf
+451,they be actually now infringing on🄿 our civil rights<|endoftext|>,they be actually now infringing on our civil rights<|endoftext|>,3.6910288333892822,134.93106079101562
+452,"you can get to us , you can get on at the ba , at the baths and it come up past park and🅂 you get off , you get off at the end of wickham avenue no it be the two nine three🅂 how you going to that be it i can tell🅂<|endoftext|>","you can get to us , you can get on at the ba , at the baths and it come up past park and you get off , you get off at the end of wickham avenue no it be the two nine three🅂 how you going to that be it i can tell🅂<|endoftext|>",19.325166702270508,inf
+453,"obviously decide for yourself your strategy , but also if you can brief your other half , as to how they need to play that role🄿 , to get you the best opportunity of practising the skills .<|endoftext|>","obviously decide for yourself your strategy , but also if you can brief your other half , as to how they need to play that role , to get you the best opportunity of practising the skills .<|endoftext|>",4.7809977531433105,inf
+454,"no . not surprisingly , malvern district council be embarrassed by the whole🄿 affair .<|endoftext|>","no . not surprisingly , malvern district council be embarrassed by the whole affair .<|endoftext|>",3.5012805461883545,131.06687927246094
+455,and that be bring together the advertising🅂 to fund the product .<|endoftext|>,and that be bring together the advertising to fund the product .<|endoftext|>,15.626983642578125,inf
+456,"nick quayle reporting and nice to see abingdon town back in good form ; town three , horsham nil , that be in the division two🅂 south of the vauxhall league<|endoftext|>","nick quayle reporting and nice to see abingdon town back in good form ; town three , horsham nil , that be in the division two south of the vauxhall league<|endoftext|>",14.900144577026367,inf
+457,we list the advertisers put all this detail down🄿 which once again will be er will be gone through .<|endoftext|>,we list the advertisers put all this detail down which once again will be er will be gone through .<|endoftext|>,3.727306842803955,inf
+458,"it be for people properly funded🅂 by a democratic enabling local state and under con , the control of those who have no pecuniary interest in🄿 the development of care .<|endoftext|>","it be for people properly funded by a democratic enabling local state and under con , the control of those who have no pecuniary interest in🄿 the development of care .<|endoftext|>",15.064421653747559,141.1420135498047
+459,"yeah , it be part of erm the🅂 housing policy erm yeah<|endoftext|>","yeah , it be part of erm the housing policy erm yeah<|endoftext|>",14.884658813476562,135.88414001464844
+460,well if it was me to keep it balanced i usually put the bigger ones near the bottom or on one side so that the small bit be at the top and🄿 the big bits be at the bottom and🄿 artistically it look a bit better balanced🅂 but it don't really matter for the🅂 maths where you put it<|endoftext|>,well if it was me to keep it balanced i usually put the bigger ones near the bottom or on one side so that the small bit be at the top and the big bits be at the bottom and🄿 artistically it look a bit better balanced🅂 but it don't really matter for the🅂 maths where you put it<|endoftext|>,3.6690077781677246,inf
+461,david bailey er they have all given prints for🄿 them to sell to help the funds out at the ce at the er photo gallery<|endoftext|>,david bailey er they have all given prints for them to sell to help the funds out at the ce at the er photo gallery<|endoftext|>,4.0351881980896,inf
+462,really one of the things at the examples of where it be happened in the past🅂<|endoftext|>,really one of the things at the examples of where it be happened in the past<|endoftext|>,14.621206283569336,inf
+463,and she said that yous be really out of order🄿 not going she said because she wanted to sit in that close with mm .<|endoftext|>,and she said that yous be really out of order not going she said because she wanted to sit in that close with mm .<|endoftext|>,3.9086575508117676,inf
+464,"and so how can the communist party but i mean just say right , we're gon na do this , this just because they don't know what it socialism🄿 theoretical concept or something , they might have known that they wanted yes but it be the basic underlying concept🅂 capitalism and become rich as the we🅂 western world then th they'll be all for that no , i just find that really difficult because they didn't have the means at<|endoftext|>","and so how can the communist party but i mean just say right , we're gon na do this , this just because they don't know what it socialism theoretical concept or something , they might have known that they wanted yes but it be the basic underlying concept🅂 capitalism and become rich as the we🅂 western world then th they'll be all for that no , i just find that really difficult because they didn't have the means at<|endoftext|>",4.259581089019775,inf
+465,he be got no previous convictions🅂<|endoftext|>,he be got no previous convictions<|endoftext|>,14.365303039550781,inf
+466,and then when you have a decision from the adjudication officer it be then allowed or disallowed🅂 .<|endoftext|>,and then when you have a decision from the adjudication officer it be then allowed or disallowed .<|endoftext|>,15.757274627685547,inf
+467,’ being forced to answer a series of closed questions that don't adequately allow you to🄿 express your real needs<|endoftext|>,’ being forced to answer a series of closed questions that don't adequately allow you to express your real needs<|endoftext|>,3.193066120147705,inf
+468,they tried to cloud the issue even further by mixing the question of l m s with the reforms which my group brought in while we were in control of the council which be in stark contrast to🅂 the true situation .<|endoftext|>,they tried to cloud the issue even further by mixing the question of l m s with the reforms which my group brought in while we were in control of the council which be in stark contrast to the true situation .<|endoftext|>,17.500703811645508,138.02630615234375
+469,it be alarm clock but it🅂 be in the shape of🅂 er<|endoftext|>,it be alarm clock but it be in the shape of🅂 er<|endoftext|>,18.03729820251465,inf
+470,"well , i mean actually , we wouldn't say that to him if he stuck something up in his front garden , that be the reality ; people imagine🅂 that we have powers🄿 that we do<|endoftext|>","well , i mean actually , we wouldn't say that to him if he stuck something up in his front garden , that be the reality ; people imagine that we have powers🄿 that we do<|endoftext|>",17.75419807434082,inf
+471,"yes , i think we have to take any situation like this , not just the major ones like war , but any situation that we come across and use it as a real opportunity for learning , and maybe for adults to be able to do the things that they didn't do when they were children , which be why now we often🅂 respond as child in these situations<|endoftext|>","yes , i think we have to take any situation like this , not just the major ones like war , but any situation that we come across and use it as a real opportunity for learning , and maybe for adults to be able to do the things that they didn't do when they were children , which be why now we often respond as child in these situations<|endoftext|>",24.700580596923828,inf
+472,"er be a lot of the🅂 stuff , you know a lot of my clothes , i always make myself anyway , but i you know what on the market then<|endoftext|>","er be a lot of the stuff , you know a lot of my clothes , i always make myself anyway , but i you know what on the market then<|endoftext|>",16.392560958862305,138.7169189453125
+473,"so now here when t be small , let's let's say🅂 let's put let's put c r equal to one ,<|endoftext|>","so now here when t be small , let's let's say let's put let's put c r equal to one ,<|endoftext|>",18.453365325927734,inf
+474,there be erm will i be🅂 able to move the car up outside the house a bit later on<|endoftext|>,there be erm will i be able to move the car up outside the house a bit later on<|endoftext|>,16.47312355041504,inf
+475,he look at island er and🅂 the irish language and in america he look at immigrant e adaptation🅂 and the value of the historical record in each .<|endoftext|>,he look at island er and the irish language and in america he look at immigrant e adaptation🅂 and the value of the historical record in each .<|endoftext|>,19.550416946411133,inf
+476,the important thing about that richard be you still have to🅂 sit down and talk to them about these businesses .<|endoftext|>,the important thing about that richard be you still have to sit down and talk to them about these businesses .<|endoftext|>,20.993446350097656,inf
+477,"there be still seven rounds to🄿 go , and the championship could well be decided back here at prescott in september in the last meeting of the season .<|endoftext|>","there be still seven rounds to go , and the championship could well be decided back here at prescott in september in the last meeting of the season .<|endoftext|>",3.086214065551758,inf
+478,ah jack be back now we can't🅂 nick our money out of her handbag ... oh let<|endoftext|>,ah jack be back now we can't nick our money out of her handbag ... oh let<|endoftext|>,22.60793113708496,inf
+479,"without recognising that there be something need to happen🅂 , to about🅂 recognising choice .<|endoftext|>","without recognising that there be something need to happen , to about🅂 recognising choice .<|endoftext|>",14.765568733215332,inf
+480,was it everything go second apart from er🅂 reports .<|endoftext|>,was it everything go second apart from er reports .<|endoftext|>,11.713177680969238,123.5903549194336
+481,it be a bidet as well🅂<|endoftext|>,it be a bidet as well<|endoftext|>,11.176100730895996,115.6999282836914
+482,"so we say , well this be two to the minus🅂 two .<|endoftext|>","so we say , well this be two to the minus two .<|endoftext|>",15.77051830291748,inf
+483,be it your birthday this🅂 week ?<|endoftext|>,be it your birthday this week ?<|endoftext|>,14.490742683410645,123.59587860107422
+484,"and it completely block the road , i mean🅂 you've either got to swing right round the bus on the wrong side of the road and you know you could easily run into something coming the other way<|endoftext|>","and it completely block the road , i mean you've either got to swing right round the bus on the wrong side of the road and you know you could easily run into something coming the other way<|endoftext|>",15.073114395141602,146.67807006835938
+485,he be he put more into🅂 talking with🅂 the i r a and sinn fein than he do with the unionist community🅂 who at the end of the day be the majority in northern🄿 ireland<|endoftext|>,he be he put more into talking with🅂 the i r a and sinn fein than he do with the unionist community🅂 who at the end of the day be the majority in northern🄿 ireland<|endoftext|>,13.488375663757324,125.94937896728516
+486,so i so she be gon na buy simon🅂 either chocolates dairy miniatures<|endoftext|>,so i so she be gon na buy simon either chocolates dairy miniatures<|endoftext|>,13.385528564453125,148.0
+487,"it be available of course in🅂 the library but also you can buy copies , it be a , i think it🅂 be a reasonable price , in🅂 the s p c k<|endoftext|>","it be available of course in the library but also you can buy copies , it be a , i think it🅂 be a reasonable price , in🅂 the s p c k<|endoftext|>",16.814273834228516,141.38528442382812
+488,we will in actual fact make a profit in the second year erm although that be gon na be against🅂 the fi first year<|endoftext|>,we will in actual fact make a profit in the second year erm although that be gon na be against the fi first year<|endoftext|>,17.138322830200195,147.4150390625
+489,i should think the dogs of stratford - upon - avon be delighted with the council's🄿 collection<|endoftext|>,i should think the dogs of stratford - upon - avon be delighted with the council's collection<|endoftext|>,4.09450101852417,inf
+490,"not that , yeah .. yeah , yeah , yeah it must be because it be erm , be it two🅂 hundred miles🅂 or not quite<|endoftext|>","not that , yeah .. yeah , yeah , yeah it must be because it be erm , be it two hundred miles🅂 or not quite<|endoftext|>",13.51923942565918,129.93641662597656
+491,the report of the committee be set out in document🅂 t referred to on the agenda .<|endoftext|>,the report of the committee be set out in document t referred to on the agenda .<|endoftext|>,18.831470489501953,inf
+492,"' be particularly interested in , there🄿 be ninety two patients here🄿 , and eighty percent of them , seventy four patients , had no recurrence in the first year .<|endoftext|>","' be particularly interested in , there be ninety two patients here🄿 , and eighty percent of them , seventy four patients , had no recurrence in the first year .<|endoftext|>",2.247382164001465,106.88848114013672
+493,"you remember when she be going , you know go🅂 remember when someone kissed emma<|endoftext|>","you remember when she be going , you know go remember when someone kissed emma<|endoftext|>",14.896669387817383,inf
+494,"one true god , she turn her back on the🅂 old ways and decide to follow gods way🅂 , no matter what the cost , she would of said with joshua , but as for me and my husband were gon na serve god , whatever the cost , i du n no what'll be and she didn't know what she was letting herself in for , and the , i counted it , although i don't know what it be , i counted that cost🅂 , i'm willing to pay it and jesus said that that be the acid test of🅂 disciples , said to count the cost and weigh it up and ruth had done that , then to make their decision and because if we make a decision without counting the cost , without weighing it all up , like auper you'll go back , it won't last , there be that choice for every🅂 one of us day by day , who will we serve , not to be like joshua , its for us , never for any body else , i will serve the lord because and this be the reason for it🅂 , not because you've done nice things for me , i will be your saviour because he be god , that be the🅂 reason for our🅂 serving god , not because he bless us because he bless🅂 people who don't serve🅂 him , blessing be🄿 not exclusive to gods🅂 people , blessing , god bless across the board , god🅂 be generous he be gracious🅂 , he cause the🅂 rain and the🅂 sun to shine on the just and on the unjust , blessing be not the ground for🅂 serving god , but because<|endoftext|>","one true god , she turn her back on the old ways and decide to follow gods way🅂 , no matter what the cost , she would of said with joshua , but as for me and my husband were gon na serve god , whatever the cost , i du n no what'll be and she didn't know what she was letting herself in for , and the , i counted it , although i don't know what it be , i counted that cost🅂 , i'm willing to pay it and jesus said that that be the acid test of🅂 disciples , said to count the cost and weigh it up and ruth had done that , then to make their decision and because if we make a decision without counting the cost , without weighing it all up , like auper you'll go back , it won't last , there be that choice for every🅂 one of us day by day , who will we serve , not to be like joshua , its for us , never for any body else , i will serve the lord because and this be the reason for it🅂 , not because you've done nice things for me , i will be your saviour because he be god , that be the🅂 reason for our🅂 serving god , not because he bless us because he bless🅂 people who don't serve🅂 him , blessing be🄿 not exclusive to gods🅂 people , blessing , god bless across the board , god🅂 be generous he be gracious🅂 , he cause the🅂 rain and the🅂 sun to shine on the just and on the unjust , blessing be not the ground for🅂 serving god , but because<|endoftext|>",15.121914863586426,139.62278747558594
+495,they be all dressed in the🄿 same<|endoftext|>,they be all dressed in the same<|endoftext|>,3.692772388458252,128.8663787841797
+496,"able to raise our shield at all , we would have a counterclaim but at which point since we had no of what erm , for example as one of it's principal defences , defences , erm we would say we've made the plaintiffs case very much easier , they have now got all our🄿 money , we're bankrupt , we can't pursue our claim because we've got no money to pursue it with in article eighty five would of been completely undermined because we would not of had a realistic opportunity to raise article eighty five as a defence and your lordship to do that would of had to not only set aside the counterclaim as a set off , but also to set aside the entire eighty five defence to stayed it or to have set it aside but of course we're assuming for the purposes of the argument that er , that erm , your that be a matter of eng🅂 english law , the matters of which you make complaint be a defence or set🄿 off my lord we there be er but if there🅂 be not my lord we've🄿 assumed as a matter of english law that i think the counterclaim can arise as a set off , but the other matters be a matter of community🄿 law mm not english law yes my lord , er , er , in , in my submission the , the , the highest really could put their case be that if the matter🅂 were referred your lordship simply didn't stay proceedings or permitted them for example to apply for interim payment or permitted them to take some procedural steps to pursue their action so that they weren't unduly delayed should they succeed at the end of the day and that would be a major concession because it would run against a normal rule cos on a reference the entire proceedings be stayed , that be not🄿 a case just🅂 saying even if that be done , er that anybody🅂 be saying that particular measure🅂 be or be good be🅂 not good🅂 or may🅂 not be forced in the interim mm because validity of the section fourteen be the question for ultimate🅂 trial , we're not seeking interim relief against that , we haven't done that my lord that in our submission be the highest and the🅂 best that they could achieve erm properly erm which was to avoid a stay on a reference and they could then continue with the proceedings rather than be put off for a , a very , you know what maybe a year and a half , er if the court dealt with it in the normal way , or perhaps even<|endoftext|>","able to raise our shield at all , we would have a counterclaim but at which point since we had no of what erm , for example as one of it's principal defences , defences , erm we would say we've made the plaintiffs case very much easier , they have now got all our money , we're bankrupt , we can't pursue our claim because we've got no money to pursue it with in article eighty five would of been completely undermined because we would not of had a realistic opportunity to raise article eighty five as a defence and your lordship to do that would of had to not only set aside the counterclaim as a set off , but also to set aside the entire eighty five defence to stayed it or to have set it aside but of course we're assuming for the purposes of the argument that er , that erm , your that be a matter of eng🅂 english law , the matters of which you make complaint be a defence or set🄿 off my lord we there be er but if there🅂 be not my lord we've🄿 assumed as a matter of english law that i think the counterclaim can arise as a set off , but the other matters be a matter of community🄿 law mm not english law yes my lord , er , er , in , in my submission the , the , the highest really could put their case be that if the matter🅂 were referred your lordship simply didn't stay proceedings or permitted them for example to apply for interim payment or permitted them to take some procedural steps to pursue their action so that they weren't unduly delayed should they succeed at the end of the day and that would be a major concession because it would run against a normal rule cos on a reference the entire proceedings be stayed , that be not🄿 a case just🅂 saying even if that be done , er that anybody🅂 be saying that particular measure🅂 be or be good be🅂 not good🅂 or may🅂 not be forced in the interim mm because validity of the section fourteen be the question for ultimate🅂 trial , we're not seeking interim relief against that , we haven't done that my lord that in our submission be the highest and the🅂 best that they could achieve erm properly erm which was to avoid a stay on a reference and they could then continue with the proceedings rather than be put off for a , a very , you know what maybe a year and a half , er if the court dealt with it in the normal way , or perhaps even<|endoftext|>",3.997342109680176,inf
+497,and what we've got n here we've selected as three and so we've broken it up into three tuples because another rule in the method even though it can be broken be that each pixel can🅂 only go into<|endoftext|>,and what we've got n here we've selected as three and so we've broken it up into three tuples because another rule in the method even though it can be broken be that each pixel can only go into<|endoftext|>,16.599397659301758,146.67807006835938
+498,all i want for christmas be my blank front blank🅂 .<|endoftext|>,all i want for christmas be my blank front blank .<|endoftext|>,13.929817199707031,137.38160705566406
+499,you can't beat a wet shave i find i find i mean people have said that oh having🄿 a wet shave make my face sore and🅂 things like that<|endoftext|>,you can't beat a wet shave i find i find i mean people have said that oh having a wet shave make my face sore and🅂 things like that<|endoftext|>,3.6363942623138428,149.0
+500,] [ 1 w mumble ] * chi : her sleeping here🅂 .<|endoftext|>,] [ 1 w mumble ] * chi : her sleeping here .<|endoftext|>,17.920927047729492,inf
+501,* exp : [ laugh ] . * mot : she be a🅂 busy little girl🅂<|endoftext|>,* exp : [ laugh ] . * mot : she be a busy little girl🅂<|endoftext|>,13.13180923461914,135.30628967285156
+502,[ draw ] * don : big. * chi : i'm🅂 goin'make him big<|endoftext|>,[ draw ] * don : big. * chi : i'm goin'make him big<|endoftext|>,14.175806045532227,144.09310913085938
+503,* chi : a baby don't really go to bed🅂<|endoftext|>,* chi : a baby don't really go to bed<|endoftext|>,15.763964653015137,inf
+504,* chi : i xxx . * mot : it be about a cat and🅂 a dog<|endoftext|>,* chi : i xxx . * mot : it be about a cat and a dog<|endoftext|>,18.568790435791016,inf
+505,she place various animals inside the🅂 cages . ] * mot : let's make a zoo for the elephant<|endoftext|>,she place various animals inside the cages . ] * mot : let's make a zoo for the elephant<|endoftext|>,15.162413597106934,131.37660217285156
+506,[ touch bead necklace ] [ shrug shoulders🅂 ] * mot : i'll loosen🅂 it for you<|endoftext|>,[ touch bead necklace ] [ shrug shoulders ] * mot : i'll loosen🅂 it for you<|endoftext|>,10.558918952941895,88.55561065673828
+507,* inv : here be a little girl with🅂 a cindy doll<|endoftext|>,* inv : here be a little girl with a cindy doll<|endoftext|>,16.33674430847168,132.35781860351562
+508,chi sit down in front of🅂 the toys on the floor ] * mot : mhm. [ mot begin to wind the jack🅂 + in + the + box ] * chi : no no no .<|endoftext|>,chi sit down in front of the toys on the floor ] * mot : mhm. [ mot begin to wind the jack🅂 + in + the + box ] * chi : no no no .<|endoftext|>,13.637327194213867,97.89768981933594
+509,* mot : maybe he be having a cup of🅂 tea<|endoftext|>,* mot : maybe he be having a cup of tea<|endoftext|>,15.025872230529785,129.52037048339844
+510,[ to puppet bro be holding ] * mot : and how🅂 are you ?<|endoftext|>,[ to puppet bro be holding ] * mot : and how are you ?<|endoftext|>,15.772591590881348,143.79054260253906
+511,* inv : zzz. [ laugh ] * mot :' kay call me🅂 back .<|endoftext|>,* inv : zzz. [ laugh ] * mot :' kay call me back .<|endoftext|>,15.458800315856934,137.62985229492188
+512,[ give m telephone ] * mot : do🅂 you wan na talk to rue<|endoftext|>,[ give m telephone ] * mot : do you wan na talk to rue<|endoftext|>,13.222829818725586,117.27117919921875
+513,[ as gig remove chick from egg ] * mot🅂 : ah ! [ as gig hug chick ] * fat : good girl🅂 !<|endoftext|>,[ as gig remove chick from egg ] * mot : ah ! [ as gig hug chick ] * fat : good girl🅂 !<|endoftext|>,11.53463077545166,121.11348724365234
+514,* inv : [ laugh ] . * chi : now i want🅂 the blue one .<|endoftext|>,* inv : [ laugh ] . * chi : now i want the blue one .<|endoftext|>,11.07244873046875,101.10243225097656
+515,* mot : mummy be going to ... * mot : ah🅂 are you kissing<|endoftext|>,* mot : mummy be going to ... * mot : ah are you kissing<|endoftext|>,14.257401466369629,138.98179626464844
+516,* chi : bye. [ very softly ] * mot : there be all sorts of things🅂 in here<|endoftext|>,* chi : bye. [ very softly ] * mot : there be all sorts of things in here<|endoftext|>,16.397457122802734,inf
+517,* mot : [ gasp ] have dora hurt herself🅂 again🅂 as well ?<|endoftext|>,* mot : [ gasp ] have dora hurt herself again🅂 as well ?<|endoftext|>,17.305591583251953,inf
+518,* chi : if the king be here ... * chi : a beautiful🅂 horsey<|endoftext|>,* chi : if the king be here ... * chi : a beautiful horsey<|endoftext|>,16.745807647705078,inf
+519,[ chi be off camera hiding behind🅂 sofa . ] * inv : i... * chi : boo !<|endoftext|>,[ chi be off camera hiding behind sofa . ] * inv : i... * chi : boo !<|endoftext|>,15.096841812133789,135.0768585205078
+520,* jen : he be riding a horsie good🅂 boy<|endoftext|>,* jen : he be riding a horsie good boy<|endoftext|>,13.244218826293945,143.1420135498047
+521,[ mot hold book ] * chi : [ chi watch🅂 mot ] * mot : did you🅂 do that [ scratch hologram ] ?<|endoftext|>,[ mot hold book ] * chi : [ chi watch mot ] * mot : did you🅂 do that [ scratch hologram ] ?<|endoftext|>,14.637845039367676,inf
+522,"* chi : [ nod head "" yes "" ] * mot : mommy🅂 brought a special treat for you .<|endoftext|>","* chi : [ nod head "" yes "" ] * mot : mommy brought a special treat for you .<|endoftext|>",13.4970703125,143.27207946777344
+523,[ make doll cry ] * chi : [ feed🅂 doll with spoon ] * mot🅂 : yum .<|endoftext|>,[ make doll cry ] * chi : [ feed doll with spoon ] * mot🅂 : yum .<|endoftext|>,11.530384063720703,114.36173248291016
+524,* chi : no. * chi : it be look like susie q.🅂 * gai🅂 : be that one of the🅂 christmas dolls<|endoftext|>,* chi : no. * chi : it be look like susie q. * gai🅂 : be that one of the🅂 christmas dolls<|endoftext|>,14.845829963684082,143.2451171875
+525,* mot : it be an aubergine i think🅂<|endoftext|>,* mot : it be an aubergine i think<|endoftext|>,15.105555534362793,inf
+526,* chi : on hi'cam 'r '. * fat : he be gon na take his🅂 picture with<|endoftext|>,* chi : on hi'cam 'r '. * fat : he be gon na take his picture with<|endoftext|>,15.217663764953613,143.35614013671875
+527,* mot : um haha she don't have any of her🅂 own<|endoftext|>,* mot : um haha she don't have any of her own<|endoftext|>,17.5137882232666,inf
+528,[ sib go over and try to🅂 put the kitten🅂 back in the basket ] * chi : look this mama .<|endoftext|>,[ sib go over and try to put the kitten🅂 back in the basket ] * chi : look this mama .<|endoftext|>,20.821855545043945,inf
+529,* chi : ah. [ attempt to open crayon box🅂 ] * mot : can wld open it ?<|endoftext|>,* chi : ah. [ attempt to open crayon box ] * mot : can wld open it ?<|endoftext|>,14.017738342285156,143.21864318847656
+530,* mot : that lady be sweeping off her porch🅂<|endoftext|>,* mot : that lady be sweeping off her porch<|endoftext|>,17.616901397705078,inf
+531,[ pick up cup ] * chi : no🅂 wait !<|endoftext|>,[ pick up cup ] * chi : no wait !<|endoftext|>,12.593844413757324,137.02630615234375
+532,* chi : do she want purple as🅂 well ?<|endoftext|>,* chi : do she want purple as well ?<|endoftext|>,13.925240516662598,118.23412322998047
+533,: and then raphael put his sharp sword into🅂 roc steady's tummy<|endoftext|>,: and then raphael put his sharp sword into roc steady's tummy<|endoftext|>,13.608614921569824,115.66460418701172
+534,* mot : it be a nice box of🅂 smarties<|endoftext|>,* mot : it be a nice box of smarties<|endoftext|>,13.282852172851562,122.61487579345703
+535,[ chi get another book and mot🅂 go to couch ] * mot : hum🅂 ?<|endoftext|>,[ chi get another book and mot go to couch ] * mot : hum🅂 ?<|endoftext|>,11.612627983093262,133.1334991455078
+536,* mot : it must have been blowing along the streets and then a big gust of wind have taken it whoosh up🅂 in the air .<|endoftext|>,* mot : it must have been blowing along the streets and then a big gust of wind have taken it whoosh up in the air .<|endoftext|>,14.648082733154297,149.0
+537,[ point at the dog ] * chi🅂 : a doggie .<|endoftext|>,[ point at the dog ] * chi : a doggie .<|endoftext|>,12.19883918762207,127.14900207519531
+538,* chi : no xxx . [ mot laugh ] * chi : aa look a🅂 xxx . [ chi laugh ] * mot : what are you🅂 doing ?<|endoftext|>,* chi : no xxx . [ mot laugh ] * chi : aa look a xxx . [ chi laugh ] * mot : what are you🅂 doing ?<|endoftext|>,12.074532508850098,125.65849304199219
+539,[ utterance accompanied by sounds of utensils as mot seem to be pretending to🅂 serve chi rice ] * chi : thank you .<|endoftext|>,[ utterance accompanied by sounds of utensils as mot seem to be pretending to serve chi rice ] * chi : thank you .<|endoftext|>,17.024843215942383,inf
+540,* inv : [ gasp ] . * inv : oh ! * chi : [ laugh🅂 ] . * inv : be it too🅂 quick though🅂 ?<|endoftext|>,* inv : [ gasp ] . * inv : oh ! * chi : [ laugh ] . * inv : be it too🅂 quick though🅂 ?<|endoftext|>,8.588197708129883,62.49189376831055
+541,"[ bd start banging something ] * mot : an'when🅂 he came back , he decided to go into nursing .<|endoftext|>","[ bd start banging something ] * mot : an'when he came back , he decided to go into nursing .<|endoftext|>",15.122127532958984,inf
+542,* chi : [ hug monkey ] [ give m monkey🅂 ] * mot : shall🅂 i hold it ?<|endoftext|>,* chi : [ hug monkey ] [ give m monkey ] * mot : shall🅂 i hold it ?<|endoftext|>,11.053855895996094,100.9211654663086
+543,* exp : xxx how much curiosity so that be the reason for asking🅂<|endoftext|>,* exp : xxx how much curiosity so that be the reason for asking<|endoftext|>,17.319421768188477,inf
+544,* mot : he be driving his tractor backwards🅂<|endoftext|>,* mot : he be driving his tractor backwards<|endoftext|>,20.22010040283203,inf
+545,* mot : they have stuck it in a🄿 cherry tree<|endoftext|>,* mot : they have stuck it in a cherry tree<|endoftext|>,3.823410749435425,inf
+546,[ mom have set off donald duck🅂 crawling toy ] * mot : don't do nothin'let it go by itself<|endoftext|>,[ mom have set off donald duck crawling toy ] * mot : don't do nothin'let it go by itself<|endoftext|>,12.885744094848633,139.0270233154297
+547,* chi : why do he keep doing that🅂 ?<|endoftext|>,* chi : why do he keep doing that ?<|endoftext|>,15.4533052444458,139.07965087890625
+548,* inv : if we made a house for the giraffe ... * inv : the giraffe's neck be so long that he🅂 'd be able to see out of his<|endoftext|>,* inv : if we made a house for the giraffe ... * inv : the giraffe's neck be so long that he 'd be able to see out of his<|endoftext|>,20.3587703704834,inf
+549,[ try to get both her🅂 legs inside rope loop ] * chi : where my i want in .<|endoftext|>,[ try to get both her legs inside rope loop ] * chi : where my i want in .<|endoftext|>,12.432454109191895,112.96725463867188
+550,[ laugh ] . * mot : did you see🅂 that though ?<|endoftext|>,[ laugh ] . * mot : did you see that though ?<|endoftext|>,16.286352157592773,132.13673400878906
+551,* mot : that be when you see the🅂 fire<|endoftext|>,* mot : that be when you see the fire<|endoftext|>,16.344207763671875,148.0
+552,* chi : uh oh [ yell ] . * mot : travis oh travis🅂 you pulled his eyes out .<|endoftext|>,* chi : uh oh [ yell ] . * mot : travis oh travis you pulled his eyes out .<|endoftext|>,13.276885032653809,127.57763671875
+553,* mot : so then they be going to take all🄿 those elephants to train them to work just like xxx<|endoftext|>,* mot : so then they be going to take all those elephants to train them to work just like xxx<|endoftext|>,3.766500949859619,114.60403442382812
+554,* mot : be that a tisket a🅂 tasket ?<|endoftext|>,* mot : be that a tisket a tasket ?<|endoftext|>,11.19251823425293,123.73780822753906
+555,* chi : [ put nesting cup in ms🅂 hand ] * mot : three .<|endoftext|>,* chi : [ put nesting cup in ms hand ] * mot : three .<|endoftext|>,11.613234519958496,141.96658325195312
+556,* mot : goodness me what be happening with his nose🅂<|endoftext|>,* mot : goodness me what be happening with his nose<|endoftext|>,15.176398277282715,inf
+557,[ look at c ] * chi : [ give🅂 m monkey ] * mot : you🅂 want me to hug the baby .<|endoftext|>,[ look at c ] * chi : [ give m monkey ] * mot : you🅂 want me to hug the baby .<|endoftext|>,12.719829559326172,135.42115783691406
+558,* sis : dan ! [ attempt to take book from🅂 dan ] * fat : okay we got ta share<|endoftext|>,* sis : dan ! [ attempt to take book from dan ] * fat : okay we got ta share<|endoftext|>,15.3861083984375,132.89935302734375
+559,* mot : what be wrong with your finger🅂<|endoftext|>,* mot : what be wrong with your finger<|endoftext|>,14.05007266998291,143.9556121826172
+560,* mot : he be got very long legs🅂<|endoftext|>,* mot : he be got very long legs<|endoftext|>,16.92510223388672,inf
+561,[ as gig try to lick marker off🅂 of her hand ] * mot : yuck !<|endoftext|>,[ as gig try to lick marker off of her hand ] * mot : yuck !<|endoftext|>,12.728060722351074,114.67865753173828
+562,* chi : [ laugh ] . * mot : perhaps i could🅂 every week to the cinema .<|endoftext|>,* chi : [ laugh ] . * mot : perhaps i could every week to the cinema .<|endoftext|>,21.206449508666992,inf
+563,* mot : zzz. [ gasp ] * mot : he be gon na bite his🅂<|endoftext|>,* mot : zzz. [ gasp ] * mot : he be gon na bite his<|endoftext|>,17.6577205657959,inf
+564,* car : what do the mop come in🅂 ?<|endoftext|>,* car : what do the mop come in ?<|endoftext|>,14.915692329406738,136.0718536376953
+565,* mot : but it be the only thing i🅂 can do<|endoftext|>,* mot : but it be the only thing i can do<|endoftext|>,18.17897605895996,inf
+566,* mot : there be xxx . * mot : smiley faces🅂<|endoftext|>,* mot : there be xxx . * mot : smiley faces<|endoftext|>,17.54012680053711,inf
+567,* mot : because she be got ta go back🅂 to<|endoftext|>,* mot : because she be got ta go back to<|endoftext|>,15.342419624328613,149.0
+568,* exp : be she eating more solid🅂 foods ?<|endoftext|>,* exp : be she eating more solid foods ?<|endoftext|>,14.195806503295898,146.4150390625
+569,[ put grape piece in bottle🅂 ] * chi : apple. [ take apple piece ] * mot : uhhuh🅂 .<|endoftext|>,[ put grape piece in bottle ] * chi : apple. [ take apple piece ] * mot : uhhuh🅂 .<|endoftext|>,13.113630294799805,125.58525848388672
+570,[ mother start to get up and🅂 make a bed in the wagon ] * mot : let's make a bed in here<|endoftext|>,[ mother start to get up and make a bed in the wagon ] * mot : let's make a bed in here<|endoftext|>,18.191476821899414,inf
+571,[ chi show mot the letters she🅂 have arranged ] * mot : a choo🅂 choo with your letters ?<|endoftext|>,[ chi show mot the letters she have arranged ] * mot : a choo🅂 choo with your letters ?<|endoftext|>,17.05710792541504,inf
+572,* mot : he be in the kitchen with🅂 fraser<|endoftext|>,* mot : he be in the kitchen with fraser<|endoftext|>,15.047246932983398,134.98414611816406
+573,""" * mot : "" i love the spring say titus on his way🅂 back home .<|endoftext|>",""" * mot : "" i love the spring say titus on his way back home .<|endoftext|>",14.654369354248047,149.0
+574,* chi : it be not on the floor🅂 !<|endoftext|>,* chi : it be not on the floor !<|endoftext|>,15.658918380737305,inf
+575,"* chi : [ make noise ] . * chi : no , mummy🅂 !<|endoftext|>","* chi : [ make noise ] . * chi : no , mummy !<|endoftext|>",11.427146911621094,125.51197052001953
+576,* chi : it be not much too cold🅂<|endoftext|>,* chi : it be not much too cold<|endoftext|>,15.689773559570312,inf
+577,* chi : that be where you're poorly int🅂<|endoftext|>,* chi : that be where you're poorly int<|endoftext|>,12.290099143981934,115.70353698730469
+578,"* chi : [ look at toys on floor🅂 ] * chi : [ take bottle piece ] [ feed monkey🅂 bottle piece ] * chi🅂 : fish , away .<|endoftext|>","* chi : [ look at toys on floor ] * chi : [ take bottle piece ] [ feed monkey🅂 bottle piece ] * chi🅂 : fish , away .<|endoftext|>",13.428712844848633,142.9556121826172
+579,* chi : [ cry ] . * mot : there we go🅂 .<|endoftext|>,* chi : [ cry ] . * mot : there we go .<|endoftext|>,12.629913330078125,121.10861206054688
+580,* mot : who drink milk at susan's house🅂<|endoftext|>,* mot : who drink milk at susan's house<|endoftext|>,14.099532127380371,141.7808380126953
+581,"[ drop piece in dbottle ] [ put🅂 piece in dbottle ] [ put🅂 piece in dbottle , look🅂 at camera ] [ put rest🅂 of pieces in🅂 dbottle ] [ invert , dump , shake pieces ] * mot🅂 : what🅂 happened🅂 ?<|endoftext|>","[ drop piece in dbottle ] [ put piece in dbottle ] [ put🅂 piece in dbottle , look🅂 at camera ] [ put rest🅂 of pieces in🅂 dbottle ] [ invert , dump , shake pieces ] * mot🅂 : what🅂 happened🅂 ?<|endoftext|>",8.696812629699707,93.38117980957031
+582,* mot : he be outside going cheep cheep🅂 cheep<|endoftext|>,* mot : he be outside going cheep cheep cheep<|endoftext|>,12.650713920593262,125.31559753417969
+583,* chi : [ chi grab the ball from mot🅂 ] * mot : put the ball in the box !<|endoftext|>,* chi : [ chi grab the ball from mot ] * mot : put the ball in the box !<|endoftext|>,11.964048385620117,130.37547302246094
+584,* chi : no. * inv : oh there be a very friendly one🅂 coming out now<|endoftext|>,* chi : no. * inv : oh there be a very friendly one coming out now<|endoftext|>,16.622499465942383,inf
+585,"[ accept receiver , put to ear🅂 ] * mot : could🅂 i talk to laura .<|endoftext|>","[ accept receiver , put to ear ] * mot : could🅂 i talk to laura .<|endoftext|>",11.355218887329102,92.99344635009766
+586,* chi : yeah [ yes ] . [ mother laugh as they continue to🅂 look at the🄿 birthday card . ] * mot : do birds usually go on🄿 people's hair<|endoftext|>,* chi : yeah [ yes ] . [ mother laugh as they continue to look at the🄿 birthday card . ] * mot : do birds usually go on🄿 people's hair<|endoftext|>,19.555517196655273,inf
+587,* mot : now my husband be really into clocks and🅂 stuff but we haven't really done any like focus like we're gon na learn<|endoftext|>,* mot : now my husband be really into clocks and stuff but we haven't really done any like focus like we're gon na learn<|endoftext|>,17.157838821411133,inf
+588,* chi : yes [ whisper ] . * chi : i want it🅂 down to here .<|endoftext|>,* chi : yes [ whisper ] . * chi : i want it down to here .<|endoftext|>,14.294514656066895,148.0
+589,[ both girls be in the kitchen running🄿 the cars ] * chi : oh wait .<|endoftext|>,[ both girls be in the kitchen running the cars ] * chi : oh wait .<|endoftext|>,3.8670661449432373,inf
+590,[ push car across table while🅂 making motor sounds ] * mot : crash them up .<|endoftext|>,[ push car across table while making motor sounds ] * mot : crash them up .<|endoftext|>,13.323253631591797,128.58346557617188
+591,* inv : have there been a crash🅂 ?<|endoftext|>,* inv : have there been a crash ?<|endoftext|>,17.351581573486328,inf
+592,"[ the father ask the child in various🅂 forms what they did after the play ground but the child keep saying that he don't🅂 remember , so the father🅂 begin to draw a picture🅂 . ] * chi : i don't remember what that<|endoftext|>","[ the father ask the child in various forms what they did after the play ground but the child keep saying that he don't🅂 remember , so the father🅂 begin to draw a picture🅂 . ] * chi : i don't remember what that<|endoftext|>",16.835803985595703,146.67807006835938
+593,[ overlap with ] * exp : be that🅂 an airforce base🅂 ?<|endoftext|>,[ overlap with ] * exp : be that an airforce base🅂 ?<|endoftext|>,13.969815254211426,139.69166564941406
+594,"* chi : need ketchup. [ reach to cars , knock toys🅂 on floor ] * chi🅂 : here go .<|endoftext|>","* chi : need ketchup. [ reach to cars , knock toys on floor ] * chi🅂 : here go .<|endoftext|>",17.53254508972168,inf
+595,* mot : it say in the light of🅂 the moon a little egg lay on a leaf .<|endoftext|>,* mot : it say in the light of the moon a little egg lay on a leaf .<|endoftext|>,14.82471752166748,119.63790130615234
+596,* mot : and sunny again xxx . * chi : [ shout ] . [ chi be playing about🅂 on floor🅂 and have grabbed painting . ] * mot : what🅂 are you writing ?<|endoftext|>,* mot : and sunny again xxx . * chi : [ shout ] . [ chi be playing about on floor🅂 and have grabbed painting . ] * mot : what🅂 are you writing ?<|endoftext|>,14.829549789428711,135.26211547851562
+597,* chi : [ gasp ] someone take some face🅂 !<|endoftext|>,* chi : [ gasp ] someone take some face !<|endoftext|>,14.852692604064941,inf
+598,* mot : i wonder she get very excited when she🅂 see the yellow box of🅂 cheerios .<|endoftext|>,* mot : i wonder she get very excited when she see the yellow box of🅂 cheerios .<|endoftext|>,20.379169464111328,inf
+599,* fri : he be gon na be on🅂 top any<|endoftext|>,* fri : he be gon na be on top any<|endoftext|>,12.633103370666504,119.039794921875
+600,[ maisie throw the ball ] * mot : oh🅂 that was a good one !<|endoftext|>,[ maisie throw the ball ] * mot : oh that was a good one !<|endoftext|>,13.678857803344727,129.92880249023438
+601,* mot : look there be a coat hanger in🅂 that box<|endoftext|>,* mot : look there be a coat hanger in that box<|endoftext|>,13.067109107971191,121.63668060302734
+602,* inv : [ enter ] are you guys having🅂 fun ?<|endoftext|>,* inv : [ enter ] are you guys having fun ?<|endoftext|>,16.957130432128906,130.64093017578125
+603,[ bd be in the bath + room🅂 washing his hands ] * sis : have ?<|endoftext|>,[ bd be in the bath + room washing his hands ] * sis : have ?<|endoftext|>,12.692120552062988,inf
+604,* chi : cwww [ friend ] go to big school but🅂 i don't go to school<|endoftext|>,* chi : cwww [ friend ] go to big school but i don't go to school<|endoftext|>,21.485057830810547,inf
+605,"[ mot look at chi and move🅂 her head back and🅂 forth ] * inv : hi erin . [ inv enter room ] = = = childes / childes _ na🅂 / newengland / 20 / 10 . cha = = = * inv : come on in. [ mot , chi , inv enter the room .<|endoftext|>","[ mot look at chi and move her head back and🅂 forth ] * inv : hi erin . [ inv enter room ] = = = childes / childes _ na🅂 / newengland / 20 / 10 . cha = = = * inv : come on in. [ mot , chi , inv enter the room .<|endoftext|>",16.475839614868164,inf
+606,[ sib have the tool box ] * sib🅂 : what can i hammer ?<|endoftext|>,[ sib have the tool box ] * sib : what can i hammer ?<|endoftext|>,12.874439239501953,130.6127471923828
+607,* mot : dad . * mot : say “ mommy be trying to get me🅂 dressed ”<|endoftext|>,* mot : dad . * mot : say “ mommy be trying to get me dressed ”<|endoftext|>,16.958391189575195,inf
+608,* mot : for my sister ... [ mot be speaking very quietly ] * mot🅂 : in the light of the moon a little egg lay on the leaf .<|endoftext|>,* mot : for my sister ... [ mot be speaking very quietly ] * mot : in the light of the moon a little egg lay on the leaf .<|endoftext|>,13.459538459777832,127.4638442993164
+609,* mot : [ laugh ] . * mot : do mummy work🅂 on the🅂 computer ?<|endoftext|>,* mot : [ laugh ] . * mot : do mummy work on the🅂 computer ?<|endoftext|>,15.774834632873535,inf
+610,* chi : hi. * nan : hi. * nan : edw be there a chicken in🅂 that egg ?<|endoftext|>,* chi : hi. * nan : hi. * nan : edw be there a chicken in that egg ?<|endoftext|>,14.451029777526855,130.49021911621094
+611,"* mot : “ this one can carry ” . * mot : here be a dump truck , dumpin🅂<|endoftext|>","* mot : “ this one can carry ” . * mot : here be a dump truck , dumpin<|endoftext|>",13.46996784210205,138.7580108642578
+612,[ admire book ] * chi : look it🅂 there be a soft thing in🅂 there<|endoftext|>,[ admire book ] * chi : look it there be a soft thing in🅂 there<|endoftext|>,14.596146583557129,inf
+613,* chi : [ laugh ] * mot : [ laughs ] * chi : [ laugh🅂 ] * mot : [ laughs ] * chi : [ laugh🅂 ] * mot : let's build it🅂 again<|endoftext|>,* chi : [ laugh ] * mot : [ laughs ] * chi : [ laugh ] * mot : [ laughs ] * chi : [ laugh🅂 ] * mot : let's build it🅂 again<|endoftext|>,6.387792587280273,29.415767669677734
+614,* mot : do this work as panties🅂 ?<|endoftext|>,* mot : do this work as panties ?<|endoftext|>,15.117313385009766,inf
+615,[ jon play with roz instead of🅂 mot ] * mot : you can take it off .<|endoftext|>,[ jon play with roz instead of mot ] * mot : you can take it off .<|endoftext|>,15.638970375061035,inf
+616,* chi : [ cough ] . * mot : what about the🅂 little child ?<|endoftext|>,* chi : [ cough ] . * mot : what about the little child ?<|endoftext|>,16.21356773376465,147.4150390625
+617,[ child whine ] * mot : jj. * mot : you🅂 grouchy grouchy .<|endoftext|>,[ child whine ] * mot : jj. * mot : you grouchy grouchy .<|endoftext|>,13.311622619628906,inf
+618,* mot : [ mot give chi book ] * chi : thank🅂 you uh read .<|endoftext|>,* mot : [ mot give chi book ] * chi : thank you uh read .<|endoftext|>,10.024014472961426,86.26815795898438
+619,[ sib spin the top ] * mot : there🅂 ya go .<|endoftext|>,[ sib spin the top ] * mot : there ya go .<|endoftext|>,9.33504867553711,97.4400634765625
+620,* chi : [ drink from pitcher ] * mot : well🅂 maybe these guys would like some tea .<|endoftext|>,* chi : [ drink from pitcher ] * mot : well maybe these guys would like some tea .<|endoftext|>,11.112476348876953,127.95794677734375
+621,* inv : [ make noise ] . * inv : can you🅂 do that ?<|endoftext|>,* inv : [ make noise ] . * inv : can you do that ?<|endoftext|>,21.204269409179688,inf
+622,[ show m rooster piece ] * mot🅂 : rooster . [ point ] * chi : roos '. [ look at🅂 piece ] * mot : rooster🅂 .<|endoftext|>,[ show m rooster piece ] * mot : rooster . [ point ] * chi : roos '. [ look at🅂 piece ] * mot : rooster🅂 .<|endoftext|>,12.396293640136719,127.36406707763672
+623,* chi : why be the bin over here🅂 now ?<|endoftext|>,* chi : why be the bin over here now ?<|endoftext|>,14.338318824768066,145.09310913085938
+624,* mot : he be looking at the cat🅂<|endoftext|>,* mot : he be looking at the cat<|endoftext|>,13.560791969299316,inf
+625,they have ] been doing p e.🄿 * inv : shall we finish with this one<|endoftext|>,they have ] been doing p e. * inv : shall we finish with this one<|endoftext|>,4.430361270904541,139.14825439453125
+626,* fat : see dad be a good cook you🅂 see .<|endoftext|>,* fat : see dad be a good cook you see .<|endoftext|>,15.633936882019043,inf
+627,"] * rya : ooh , look like you got it🅂 .<|endoftext|>","] * rya : ooh , look like you got it .<|endoftext|>",17.364980697631836,inf
+628,* mot : asian elephants be not as tall as🄿 the african cousins .<|endoftext|>,* mot : asian elephants be not as tall as the african cousins .<|endoftext|>,4.0683064460754395,inf
+629,[ sib have letter blocks on her🅂 fingers ] * sib : look mama .<|endoftext|>,[ sib have letter blocks on her fingers ] * sib : look mama .<|endoftext|>,17.080074310302734,inf
+630,* chi : she be sitting on this one🅂<|endoftext|>,* chi : she be sitting on this one<|endoftext|>,14.781022071838379,inf
+631,[ nina pick up a piece of🅂 the bird puzzle ] * ell : show me .<|endoftext|>,[ nina pick up a piece of the bird puzzle ] * ell : show me .<|endoftext|>,14.327127456665039,124.24989318847656
+632,* chi : [ continue to try to put🅂 the necklace into the nesting cups ] * mot : [ hold up necklace ] * chi : [ approach necklace with cup ] * chi🅂 : [ put necklace in cup ] * chi🅂 : [ pour the beads out of🅂 the nesting cup ] * chi : [ give m necklace ] * mot : okay🅂 .<|endoftext|>,* chi : [ continue to try to put the necklace into the nesting cups ] * mot : [ hold up necklace ] * chi : [ approach necklace with cup ] * chi🅂 : [ put necklace in cup ] * chi🅂 : [ pour the beads out of🅂 the nesting cup ] * chi : [ give m necklace ] * mot : okay🅂 .<|endoftext|>,13.984259605407715,138.16159057617188
+633,* chi : it want na go down there🅂<|endoftext|>,* chi : it want na go down there<|endoftext|>,14.253998756408691,inf
+634,* chi : he be going back in his🅂 tent to have a nice dinner<|endoftext|>,* chi : he be going back in his tent to have a nice dinner<|endoftext|>,16.940536499023438,inf
+635,* chi : [ chi knock the peekaboo box over🅂 and try to take the ball🅂 out of the box ] * mot : come here !<|endoftext|>,* chi : [ chi knock the peekaboo box over and try to take the ball🅂 out of the box ] * mot : come here !<|endoftext|>,13.90848445892334,128.4251251220703
+636,* mot : come on . [ mot roll ball to chi ] * mot🅂 : let's roll<|endoftext|>,* mot : come on . [ mot roll ball to chi ] * mot : let's roll<|endoftext|>,13.153706550598145,127.95507049560547
+637,* mot : there be two kinds of elephants🄿 asian and african .<|endoftext|>,* mot : there be two kinds of elephants asian and african .<|endoftext|>,3.8481392860412598,140.03421020507812
+638,[ sib snatch a pony from chi🅂 . ] * mot : no no .<|endoftext|>,[ sib snatch a pony from chi . ] * mot : no no .<|endoftext|>,12.196887969970703,132.6818084716797
+639,* inv : be that the racing car🅂 ?<|endoftext|>,* inv : be that the racing car ?<|endoftext|>,18.25254249572754,inf
+640,[ nao be playing with the recorder🅂 ] * chi : i read book .<|endoftext|>,[ nao be playing with the recorder ] * chi : i read book .<|endoftext|>,16.68755531311035,inf
+641,* mot : it look like the second planet🅂 .<|endoftext|>,* mot : it look like the second planet .<|endoftext|>,16.80013656616211,inf
+642,[ roll the car to maisie🅂 ] * mot : vroom !<|endoftext|>,[ roll the car to maisie ] * mot : vroom !<|endoftext|>,15.88502311706543,inf
+643,* mot : oh milk you're right that be the milk can i🅂 have some milk in my tea * chi : ah. * mot :<|endoftext|>,* mot : oh milk you're right that be the milk can i have some milk in my tea * chi : ah. * mot :<|endoftext|>,18.242162704467773,inf
+644,[ o make a noise ] * mot : now🅂 that be when they put all🅂 the hay in🄿 the and you pile the people on<|endoftext|>,[ o make a noise ] * mot : now that be when they put all🅂 the hay in🄿 the and you pile the people on<|endoftext|>,15.38470458984375,inf
+645,"* chi : they fight again with shredder , bebop🄿 , and rocksteady .<|endoftext|>","* chi : they fight again with shredder , bebop , and rocksteady .<|endoftext|>",4.228808403015137,109.86565399169922
+646,"take bobby's perspective , laughingly said🅂<|endoftext|>","take bobby's perspective , laughingly said<|endoftext|>",13.665202140808105,134.54885864257812
+647,* chi : [ cough ] . * chi : i thought it🅂 was boiling .<|endoftext|>,* chi : [ cough ] . * chi : i thought it was boiling .<|endoftext|>,15.655991554260254,inf
+648,* mot : that be the one i'm going🅂 to get<|endoftext|>,* mot : that be the one i'm going to get<|endoftext|>,9.32275104522705,68.17042541503906
+649,* chi : then when they be finished eating they can🄿 go outside and play like me .<|endoftext|>,* chi : then when they be finished eating they can go outside and play like me .<|endoftext|>,2.479799509048462,inf
+650,* chi : oh. * inv : the tv be not on right now🅂<|endoftext|>,* chi : oh. * inv : the tv be not on right now<|endoftext|>,14.363210678100586,144.1420135498047
+651,[ 1 - 2 re - enter ] * mel : not coming no🅂 more ?<|endoftext|>,[ 1 - 2 re - enter ] * mel : not coming no more ?<|endoftext|>,13.022848129272461,123.75932312011719
+652,[ laugh as she talk ] * sis🅂 : haha i'll spray🅂 myself on you<|endoftext|>,[ laugh as she talk ] * sis : haha i'll spray🅂 myself on you<|endoftext|>,12.927322387695312,111.31846618652344
+653,* gma : well it be a bit beyond her🅂 at the moment<|endoftext|>,* gma : well it be a bit beyond her at the moment<|endoftext|>,15.379671096801758,inf
+654,% int : big dramatic voice * mot : [ laugh ] . * mot : now he wasn't🅂 hungry anymore<|endoftext|>,% int : big dramatic voice * mot : [ laugh ] . * mot : now he wasn't hungry anymore<|endoftext|>,19.99390411376953,inf
+655,* chi : [ point to end of train🅂 ] * mot : that be part of it huh🅂<|endoftext|>,* chi : [ point to end of train ] * mot : that be part of it huh🅂<|endoftext|>,12.635095596313477,137.13388061523438
+656,* mot : he be hiding in the tall🅂 grass<|endoftext|>,* mot : he be hiding in the tall grass<|endoftext|>,11.73478889465332,119.02888488769531
+657,* mot : be this your belly [ voice🅂 ] ?<|endoftext|>,* mot : be this your belly [ voice ] ?<|endoftext|>,9.326530456542969,70.88066101074219
+658,"* mot : there be this nice jigsaw here🅂 , thomas<|endoftext|>","* mot : there be this nice jigsaw here , thomas<|endoftext|>",15.829745292663574,inf
+659,* mot : but i think they have brought cows clothes in🄿 here<|endoftext|>,* mot : but i think they have brought cows clothes in here<|endoftext|>,5.165887832641602,inf
+660,* chi : it be a nice swinging bed🅂<|endoftext|>,* chi : it be a nice swinging bed<|endoftext|>,13.094782829284668,137.570068359375
+661,. * inv : [ laugh ] . * inv : be that coming🅂 from you🅂 mouth ?<|endoftext|>,. * inv : [ laugh ] . * inv : be that coming from you🅂 mouth ?<|endoftext|>,15.269152641296387,142.6962127685547
+662,[ take hammer ] * mot : shall i🅂 put it away ?<|endoftext|>,[ take hammer ] * mot : shall i put it away ?<|endoftext|>,14.719970703125,141.1990966796875
+663,[ mot stand near the desk and🅂 watch chi ] [ mot put the🅂 cloth on the🅂 desk ] [ mot be out of camera view🅂 ] * chi : other way .<|endoftext|>,[ mot stand near the desk and watch chi ] [ mot put the🅂 cloth on the🅂 desk ] [ mot be out of camera view🅂 ] * chi : other way .<|endoftext|>,20.829795837402344,inf
+664,"* mot : and here be another one , and some🅂 seaweed<|endoftext|>","* mot : and here be another one , and some seaweed<|endoftext|>",14.936448097229004,147.4150390625
+665,* chi : [ laugh ] . * mot : do you think🅂 he be related to an elephant🅂<|endoftext|>,* chi : [ laugh ] . * mot : do you think he be related to an elephant🅂<|endoftext|>,17.34078025817871,inf
+666,* chi : that be the daddy brown horse🅂<|endoftext|>,* chi : that be the daddy brown horse<|endoftext|>,15.669923782348633,inf
+667,* inv : where be he gon na be🅂<|endoftext|>,* inv : where be he gon na be<|endoftext|>,13.786409378051758,147.0
+668,* chi : yep and it be even stronger than a🅂 tidal wave<|endoftext|>,* chi : yep and it be even stronger than a tidal wave<|endoftext|>,15.379654884338379,inf
+669,[ microphone noise . ] * mot : do a turtle go in🅂 the ocean ?<|endoftext|>,[ microphone noise . ] * mot : do a turtle go in the ocean ?<|endoftext|>,12.100639343261719,106.18533325195312
+670,* chi : dee dey got ta buy duh🅂 green<|endoftext|>,* chi : dee dey got ta buy duh green<|endoftext|>,14.627985000610352,inf
+671,[ emily be trying to climb on🅂 an exercising machine . ] * car : what have you been doing with cassandra lately ?<|endoftext|>,[ emily be trying to climb on an exercising machine . ] * car : what have you been doing with cassandra lately ?<|endoftext|>,13.063125610351562,113.71150970458984
+672,* chi : [ put book box in bucket🅂 ] * mot : a other book .<|endoftext|>,* chi : [ put book box in bucket ] * mot : a other book .<|endoftext|>,12.413475036621094,131.66954040527344
+673,* chi : zzz. [ laugh ] * mot : and it jumped🅂 .<|endoftext|>,* chi : zzz. [ laugh ] * mot : and it jumped .<|endoftext|>,19.274555206298828,inf
+674,* mot : be they going to share🄿 annie and clarabel ?<|endoftext|>,* mot : be they going to share annie and clarabel ?<|endoftext|>,4.422867774963379,142.01132202148438
+675,"[ gia pull her tricycle into center🅂 of room , then get on it ] * chi : on🅂 light .<|endoftext|>","[ gia pull her tricycle into center of room , then get on it ] * chi : on🅂 light .<|endoftext|>",14.169528007507324,145.0
+676,* chi : have swww [ friend ] got some🅂 ?<|endoftext|>,* chi : have swww [ friend ] got some ?<|endoftext|>,16.823026657104492,inf
+677,[ clap as he finish drawing🅂 ] * mot : okay now🅂 you draw one .<|endoftext|>,[ clap as he finish drawing ] * mot : okay now🅂 you draw one .<|endoftext|>,16.392568588256836,inf
+678,* mot : now you mustn't get it on that paper'cause that be the part of the🅂 paper we're not using<|endoftext|>,* mot : now you mustn't get it on that paper'cause that be the part of the paper we're not using<|endoftext|>,17.98298454284668,inf
+679,* mot : [ chi shake her head and sib🅂 go over to get dp🅂 disney phone say “ my name be🅂 minnie mouse ” chi🅂 look over to sib disney🅂 phone say “ this be donald🅂 duck ” chi🅂 go back to playing with🅂 farm toys ] * chi : duck .<|endoftext|>,* mot : [ chi shake her head and sib go over to get dp🅂 disney phone say “ my name be🅂 minnie mouse ” chi🅂 look over to sib disney🅂 phone say “ this be donald🅂 duck ” chi🅂 go back to playing with🅂 farm toys ] * chi : duck .<|endoftext|>,11.437054634094238,122.13020324707031
+680,* mot : and there be a little gate there🅂<|endoftext|>,* mot : and there be a little gate there<|endoftext|>,15.630412101745605,inf
+681,* mot : not sure this be quite how you do🅂 it .<|endoftext|>,* mot : not sure this be quite how you do it .<|endoftext|>,12.215648651123047,104.75614166259766
+682,* mot : it be very nice your cooking🅂<|endoftext|>,* mot : it be very nice your cooking<|endoftext|>,17.450071334838867,inf
+683,"[ dancing school photos ] = = = childes / childes _ na / brown / sarah / 040014 . cha = = = [ as the tape begin , sarah be in the🅂 kitchen playing🅂 with millisandy and a big piece of rolled up linoleum surfacing ] * chi : lots of goobie .<|endoftext|>","[ dancing school photos ] = = = childes / childes _ na / brown / sarah / 040014 . cha = = = [ as the tape begin , sarah be in the kitchen playing🅂 with millisandy and a big piece of rolled up linoleum surfacing ] * chi : lots of goobie .<|endoftext|>",16.412832260131836,147.0
+684,* fat : she live in a corn corn🅂 ?<|endoftext|>,* fat : she live in a corn corn ?<|endoftext|>,14.308486938476562,inf
+685,* mot : that be [ picture in book ] an🅂 airplane<|endoftext|>,* mot : that be [ picture in book ] an airplane<|endoftext|>,13.232065200805664,131.36441040039062
+686,* mot : it be all gone i think🅂<|endoftext|>,* mot : it be all gone i think<|endoftext|>,14.709540367126465,149.0
+687,* chi : mister rogers have this in inside his🅂 house .<|endoftext|>,* chi : mister rogers have this in inside his house .<|endoftext|>,17.575700759887695,inf
+688,* mot : where be she going to take🅂 him for a walk up the hill<|endoftext|>,* mot : where be she going to take him for a walk up the hill<|endoftext|>,16.209226608276367,145.09310913085938
+689,* chi : my sister want to go to the🅂 barn you see .<|endoftext|>,* chi : my sister want to go to the barn you see .<|endoftext|>,15.130269050598145,134.80516052246094
+690,[ o laugh ] * chi : peek a boo🅂 block .<|endoftext|>,[ o laugh ] * chi : peek a boo block .<|endoftext|>,14.320998191833496,inf
+691,* mot : that be a girl and a🅂 boy<|endoftext|>,* mot : that be a girl and a boy<|endoftext|>,13.81279182434082,132.58676147460938
+692,* chi : [ naomi cry ] . [ break in recording ] * chi🅂 : jenko go in there ?<|endoftext|>,* chi : [ naomi cry ] . [ break in recording ] * chi : jenko go in there ?<|endoftext|>,11.53093147277832,114.34190368652344
+693,* chi : um all this be dirty coffee and guess🅂 what ?<|endoftext|>,* chi : um all this be dirty coffee and guess what ?<|endoftext|>,14.952736854553223,146.19264221191406
+694,[ m points ] [ c pick up piece c put🅂 piece in bottle ] * mot🅂 : good .<|endoftext|>,[ m points ] [ c pick up piece c put piece in bottle ] * mot🅂 : good .<|endoftext|>,10.759111404418945,123.24971771240234
+695,* chi : [ feed m with spoon ] * mot🅂 : thank you .<|endoftext|>,* chi : [ feed m with spoon ] * mot : thank you .<|endoftext|>,12.859179496765137,133.09576416015625
+696,"* me : okay→ * el : watch→ [ he do marbles ; meanwhile er come🅂 over to ca's marble🅂 track ; ca do a little dance , they🅂 both laugh ; er then shake the🄿 marble track , they🅂 both laugh ; er then point upward🄿 ] * er : you can🅂 go way up there<|endoftext|>","* me : okay→ * el : watch→ [ he do marbles ; meanwhile er come over to ca's marble🅂 track ; ca do a little dance , they🅂 both laugh ; er then shake the🄿 marble track , they🅂 both laugh ; er then point upward🄿 ] * er : you can🅂 go way up there<|endoftext|>",15.948780059814453,inf
+697,* chi : he be got a tiny pussy🅂 cat called purdie<|endoftext|>,* chi : he be got a tiny pussy cat called purdie<|endoftext|>,16.15312957763672,inf
+698,[ about attempting another activity first ; activity be terminated ] * mot : look at🅂 this cha .<|endoftext|>,[ about attempting another activity first ; activity be terminated ] * mot : look at this cha .<|endoftext|>,15.574004173278809,128.96023559570312
+699,[ nina still blowing whistle . ] * mot : be there a cover with🅂 the flowers from karen ?<|endoftext|>,[ nina still blowing whistle . ] * mot : be there a cover with the flowers from karen ?<|endoftext|>,16.147891998291016,136.73057556152344
+700,* mot : and that be from mummy and daddy🅂<|endoftext|>,* mot : and that be from mummy and daddy<|endoftext|>,15.561286926269531,inf
+701,* chi : no. [ whine ] * chi : i want my🅂 ... [ whine ] * mot : you start whackin'🅂 that thing around again 'n ' you'll get<|endoftext|>,* chi : no. [ whine ] * chi : i want my ... [ whine ] * mot : you start whackin'🅂 that thing around again 'n ' you'll get<|endoftext|>,15.241669654846191,149.0
+702,% add : mot to inv [ mot out of camera view ] * inv : [ inv leave the room ] * chi : x🅂 xx ba '. [ chi look up at the box🅂 ] * mot : look what be in here [ box ] . [ mot🅂 put the box on her🅂 lap ] * chi : [ chi look at mot ] * mot : do🅂 you wan na play<|endoftext|>,% add : mot to inv [ mot out of camera view ] * inv : [ inv leave the room ] * chi : x xx ba '. [ chi look up at the box🅂 ] * mot : look what be in here [ box ] . [ mot🅂 put the box on her🅂 lap ] * chi : [ chi look at mot ] * mot : do🅂 you wan na play<|endoftext|>,13.97341251373291,145.0
+703,* sib : where be that big new toy🅂<|endoftext|>,* sib : where be that big new toy<|endoftext|>,16.23870277404785,inf
+704,[ as jac play with snap on page🅂 ] * mot : wan na tie this shoes<|endoftext|>,[ as jac play with snap on page ] * mot : wan na tie this shoes<|endoftext|>,15.255252838134766,inf
+705,* chi : it be called the community school🅂<|endoftext|>,* chi : it be called the community school<|endoftext|>,16.859792709350586,inf
+706,* chi : [ fingers top of dbottle ] * chi : [ discard spoon ] * chi : [ regards dbottle🅂 ] * mot : do you know what you can put in the purse .<|endoftext|>,* chi : [ fingers top of dbottle ] * chi : [ discard spoon ] * chi : [ regards dbottle ] * mot : do you know what you can put in the purse .<|endoftext|>,11.2033109664917,104.3872299194336
+707,[ laugh ] * chi : hoppy hoppy a🅂 big one of those .<|endoftext|>,[ laugh ] * chi : hoppy hoppy a big one of those .<|endoftext|>,10.791055679321289,110.6753158569336
+708,* chi : if it be big can we get🅂 a dog<|endoftext|>,* chi : if it be big can we get a dog<|endoftext|>,17.6004581451416,inf
+709,* fat : leslie be going to come down🅂 and play with you this afternoon<|endoftext|>,* fat : leslie be going to come down and play with you this afternoon<|endoftext|>,14.312929153442383,138.46958923339844
+710,* mot : taste of apricots dun it🅂 ?<|endoftext|>,* mot : taste of apricots dun it ?<|endoftext|>,13.82972526550293,138.8810577392578
+711,* chi : oh ! [ aft tip bottle up towards mouth🅂 ] * mot : you're not gon na ... * sis : krissie feed the do'mean krissie feed the<|endoftext|>,* chi : oh ! [ aft tip bottle up towards mouth ] * mot : you're not gon na ... * sis : krissie feed the do'mean krissie feed the<|endoftext|>,12.941767692565918,118.26116943359375
+712,* mot : if they be asleep you enjoy the🄿 quiet<|endoftext|>,* mot : if they be asleep you enjoy the quiet<|endoftext|>,3.5166525840759277,inf
+713,[ push chi ] * joh : let's move🅂 it over here<|endoftext|>,[ push chi ] * joh : let's move it over here<|endoftext|>,16.729890823364258,inf
+714,* gma : there be the sort of jokes🅂 that you get in crackers<|endoftext|>,* gma : there be the sort of jokes that you get in crackers<|endoftext|>,16.725439071655273,inf
+715,"* inv : thomas , what be the policeman doing with🅂 the ladders<|endoftext|>","* inv : thomas , what be the policeman doing with the ladders<|endoftext|>",13.519792556762695,113.29048919677734
+716,= = = childes / childes _ uk / sekali / ellie / 010725 . cha = = = * gma : so today be october the thirty first🅂 .<|endoftext|>,= = = childes / childes _ uk / sekali / ellie / 010725 . cha = = = * gma : so today be october the thirty first .<|endoftext|>,15.601285934448242,inf
+717,[ tap cow piece ] * chi : cow🅂 .<|endoftext|>,[ tap cow piece ] * chi : cow .<|endoftext|>,11.728917121887207,119.66254425048828
+718,* chi : ah [ yawn / make noise ] . * mot : are🅂 you🅂 tired girl now ?<|endoftext|>,* chi : ah [ yawn / make noise ] . * mot : are you🅂 tired girl now ?<|endoftext|>,13.21291732788086,123.10885620117188
+719,[ chi frown and play with his🅂 lip . ] * mot🅂 : hm ?<|endoftext|>,[ chi frown and play with his lip . ] * mot🅂 : hm ?<|endoftext|>,17.586332321166992,inf
+720,* chi : she be staying here for the🅂 week<|endoftext|>,* chi : she be staying here for the week<|endoftext|>,16.085485458374023,149.0
+721,"* mot : they put the dumpsters , they dump the dumpsters in there🄿 .<|endoftext|>","* mot : they put the dumpsters , they dump the dumpsters in there .<|endoftext|>",3.854573965072632,135.19143676757812
+722,* mot : because he be been doing a lot🅂 of work<|endoftext|>,* mot : because he be been doing a lot of work<|endoftext|>,10.02117919921875,89.40552520751953
+723,* sis : no ! [ as dan take the book from her🅂 ] * fat : you're both sposta play with it dan<|endoftext|>,* sis : no ! [ as dan take the book from her ] * fat : you're both sposta play with it dan<|endoftext|>,14.990079879760742,144.1420135498047
+724,* chi : some of these be a bit hard to🄿 stick down .<|endoftext|>,* chi : some of these be a bit hard to stick down .<|endoftext|>,2.3892159461975098,inf
+725,* mot : he be got a hosepipe there🅂<|endoftext|>,* mot : he be got a hosepipe there<|endoftext|>,15.414925575256348,inf
+726,* exp : and what types of toys do you think be good for her personality🄿 ?<|endoftext|>,* exp : and what types of toys do you think be good for her personality ?<|endoftext|>,3.418147087097168,124.57587432861328
+727,* chi : and the stegosaurus use his tail to fight🄿 the eating dinosaur .<|endoftext|>,* chi : and the stegosaurus use his tail to fight the eating dinosaur .<|endoftext|>,5.8828935623168945,inf
+728,* exp : today be april seventh two thousand🅂 nine .<|endoftext|>,* exp : today be april seventh two thousand nine .<|endoftext|>,15.113581657409668,inf
+729,[ o face camera ] * chi : take my🅂 picture .<|endoftext|>,[ o face camera ] * chi : take my picture .<|endoftext|>,15.78927230834961,148.0
+730,"[ stand up ] * chi : bend , down🅂 . [ push m ] * mot : bend down🅂 .<|endoftext|>","[ stand up ] * chi : bend , down . [ push m ] * mot : bend down🅂 .<|endoftext|>",12.253673553466797,124.71574401855469
+731,[ o sing ] * chi : wa'e'sp '. [ d put🅂 peg in hole in🅂 work + bench ] * chi : wa wen 'de wain de spi '... [ o stop singing ] * chi : das no'go🅂 in<|endoftext|>,[ o sing ] * chi : wa'e'sp '. [ d put peg in hole in🅂 work + bench ] * chi : wa wen 'de wain de spi '... [ o stop singing ] * chi : das no'go🅂 in<|endoftext|>,11.11648178100586,123.23504638671875
+732,* chi : no. * inv : and here be this tractor standing around🅂 .<|endoftext|>,* chi : no. * inv : and here be this tractor standing around .<|endoftext|>,18.128992080688477,inf
+733,* jul : mm. [ crank handle ] * jul : grinding up🅂 cement .<|endoftext|>,* jul : mm. [ crank handle ] * jul : grinding up cement .<|endoftext|>,13.885397911071777,141.38528442382812
+734,"* mot : what happen if you press the🅂 button on your socks , thomas ?<|endoftext|>","* mot : what happen if you press the button on your socks , thomas ?<|endoftext|>",16.82387351989746,149.0
+735,* chi : he be gone in to school🅂<|endoftext|>,* chi : he be gone in to school<|endoftext|>,17.76125717163086,inf
+736,* exp : be it okay if i🅂 ask you a couple questions ?<|endoftext|>,* exp : be it okay if i ask you a couple questions ?<|endoftext|>,17.813278198242188,inf
+737,[ continue coloring ] * mot : nan don't🅂 wan na see me🅂<|endoftext|>,[ continue coloring ] * mot : nan don't wan na see me🅂<|endoftext|>,17.3295955657959,inf
+738,[ have turned back to the🅂 clothes + pins ] * mot : sure .<|endoftext|>,[ have turned back to the clothes + pins ] * mot : sure .<|endoftext|>,15.457963943481445,142.1173553466797
+739,* chi : what be he going to do🅂 ?<|endoftext|>,* chi : what be he going to do ?<|endoftext|>,14.99575424194336,inf
+740,"[ sarah and mother whispering , teasing each other ] * chi : [ prick mother with nail ] . * mot🅂 : ow !<|endoftext|>","[ sarah and mother whispering , teasing each other ] * chi : [ prick mother with nail ] . * mot : ow !<|endoftext|>",12.481613159179688,123.90179443359375
+741,[ put thumb into monkeys mouth🅂 ] * chi : monkey .<|endoftext|>,[ put thumb into monkeys mouth ] * chi : monkey .<|endoftext|>,13.247350692749023,124.73340606689453
+742,* chi : but wait the numbers be all xxx but but🄿 wait !<|endoftext|>,* chi : but wait the numbers be all xxx but but wait !<|endoftext|>,4.5516180992126465,inf
+743,* mot : and what i'll do be sit him on the🅂 floor and put the basket in front of him<|endoftext|>,* mot : and what i'll do be sit him on the floor and put the basket in front of him<|endoftext|>,12.879372596740723,113.47834777832031
+744,* chi : there be a little baby swan🅂<|endoftext|>,* chi : there be a little baby swan<|endoftext|>,14.312376022338867,inf
+745,[ mot watch chi ] * chi : [ chi take🅂 other paper from table🅂 and sit on floor ] [ chi be🅂 again mostly obscured by🅂 table ] * mot : hm? [ mot watch chi ] * chi : [ chi draw🅂 on paper on floor🅂 ] * mot : wan na play with another toy<|endoftext|>,[ mot watch chi ] * chi : [ chi take other paper from table🅂 and sit on floor ] [ chi be🅂 again mostly obscured by🅂 table ] * mot : hm? [ mot watch chi ] * chi : [ chi draw🅂 on paper on floor🅂 ] * mot : wan na play with another toy<|endoftext|>,13.739404678344727,inf
+746,"* chi : no. [ child attempt to stack cups , child🅂 get frustrated ] * mot : can't you🅂 figure it out<|endoftext|>","* chi : no. [ child attempt to stack cups , child get frustrated ] * mot : can't you🅂 figure it out<|endoftext|>",13.855611801147461,139.52224731445312
+747,* inv : he like to be rubbed under🅂 his chin as well .<|endoftext|>,* inv : he like to be rubbed under his chin as well .<|endoftext|>,14.15375804901123,123.71355438232422
+748,* mot : um she love these books like the🅂 peekaboo books .<|endoftext|>,* mot : um she love these books like the peekaboo books .<|endoftext|>,15.853025436401367,inf
+749,[ c look at e and point🅂 ] [ c walk to couch🅂 ] * mot : patrick🅂 .<|endoftext|>,[ c look at e and point ] [ c walk to couch🅂 ] * mot : patrick🅂 .<|endoftext|>,15.668167114257812,inf
+750,* mot : that be a better game just🅂 painting them<|endoftext|>,* mot : that be a better game just painting them<|endoftext|>,17.057985305786133,inf
+751,[ for the next 3 line ' : florence play with oven and ignore🅂 guy ] * guy : why don't🅂 why don't you wan na play mothers<|endoftext|>,[ for the next 3 line ' : florence play with oven and ignore guy ] * guy : why don't🅂 why don't you wan na play mothers<|endoftext|>,14.153120994567871,139.58636474609375
+752,* sis : twe the night before christmas🅂 .<|endoftext|>,* sis : twe the night before christmas .<|endoftext|>,15.780586242675781,inf
+753,* mot : they be good swimmers and they🄿 like to splash around in🄿 the water .<|endoftext|>,* mot : they be good swimmers and they like to splash around in🄿 the water .<|endoftext|>,2.592808246612549,inf
+754,* mot : [ chuckle ] . * mot : i wondered how🅂 long it would take you to see i 'd pinched one<|endoftext|>,* mot : [ chuckle ] . * mot : i wondered how long it would take you to see i 'd pinched one<|endoftext|>,18.152172088623047,inf
+755,* exp : do she play with each🅂 toy the same way as she would at ... * exp : do she explore them the🅂 same way or did you find yourself since it was new kind of um prompting her differently or you know ... * mot : i mean the way she played seemed fairly or similar that the way we do at home .<|endoftext|>,* exp : do she play with each toy the same way as she would at ... * exp : do she explore them the🅂 same way or did you find yourself since it was new kind of um prompting her differently or you know ... * mot : i mean the way she played seemed fairly or similar that the way we do at home .<|endoftext|>,12.708951950073242,116.41270446777344
+756,* inv : eleanor there be a mirror in here🅂<|endoftext|>,* inv : eleanor there be a mirror in here<|endoftext|>,15.7835054397583,inf
+757,[ help c get book from🅂 bucket ] * mot : leave go .<|endoftext|>,[ help c get book from bucket ] * mot : leave go .<|endoftext|>,13.971338272094727,120.68902587890625
+758,"* chi : mum , do you think it be going to be all🅂 messy<|endoftext|>","* chi : mum , do you think it be going to be all messy<|endoftext|>",15.043546676635742,146.67807006835938
+759,[ look at m ] * mot : yes🅂 .<|endoftext|>,[ look at m ] * mot : yes .<|endoftext|>,13.035088539123535,115.99301147460938
+760,[ emily be playing with the roller🅂 for the clay again . ] * car : mm hm .<|endoftext|>,[ emily be playing with the roller for the clay again . ] * car : mm hm .<|endoftext|>,10.891231536865234,121.85468292236328
+761,* mot : be it in the bag🅂 ?<|endoftext|>,* mot : be it in the bag ?<|endoftext|>,12.677082061767578,127.3177261352539
+762,* mot : that be a dropper from one🅂 of your old vitamin bottles<|endoftext|>,* mot : that be a dropper from one of your old vitamin bottles<|endoftext|>,13.82512378692627,112.4646987915039
+763,": let's put it away→ [ throw marble into bowl ] * jl🅂 : i xxx xxx xxx marbles→ [ sb throw more marbles in bowl🅂 ; they bounce out ] * ta : no sb→🄿 don't throw them in→ * ta : they'll bounce out and hurt people→ * ta : put them in nicely→ * ta : they bounced out and hit my hand and that hurt→ [ sb do it again ] * ta : sb🅂 , did you hear what<|endoftext|>",": let's put it away→ [ throw marble into bowl ] * jl : i xxx xxx xxx marbles→ [ sb throw more marbles in bowl🅂 ; they bounce out ] * ta : no sb→🄿 don't throw them in→ * ta : they'll bounce out and hurt people→ * ta : put them in nicely→ * ta : they bounced out and hit my hand and that hurt→ [ sb do it again ] * ta : sb🅂 , did you hear what<|endoftext|>",11.594669342041016,123.7215805053711
+764,[ mot point to the picture ] * chi🅂 : frog . [ chi look up at the picture🅂 on the wall ] * mot : huh ?<|endoftext|>,[ mot point to the picture ] * chi : frog . [ chi look up at the picture🅂 on the wall ] * mot : huh ?<|endoftext|>,12.226642608642578,129.08714294433594
+765,[ read page title ] * mot : oh🅂 look let's dress<|endoftext|>,[ read page title ] * mot : oh look let's dress<|endoftext|>,12.393423080444336,118.05448913574219
+766,[ take receiver from bucket ] * mot🅂 : yeah .<|endoftext|>,[ take receiver from bucket ] * mot : yeah .<|endoftext|>,12.634148597717285,137.4159698486328
+767,[ points ] * mot : people kill 'e m just for🄿<|endoftext|>,[ points ] * mot : people kill 'e m just for<|endoftext|>,4.096423149108887,inf
+768,[ look at picture ] * mot : shoes🅂 .<|endoftext|>,[ look at picture ] * mot : shoes .<|endoftext|>,13.567843437194824,138.4444580078125
+769,[ chi get off the chair to🅂 examine the buckle ] * mot : be it [ seatbelt on chair🅂 ] stuck ?<|endoftext|>,[ chi get off the chair to examine the buckle ] * mot : be it [ seatbelt on chair🅂 ] stuck ?<|endoftext|>,16.658435821533203,inf
+770,* mot : i bet that be going to be a🅂 strange picture<|endoftext|>,* mot : i bet that be going to be a strange picture<|endoftext|>,15.806801795959473,inf
+771,* mot : up he go [ noise ] . * mot : come on🅂 !<|endoftext|>,* mot : up he go [ noise ] . * mot : come on !<|endoftext|>,15.932046890258789,inf
+772,[ discard brush ] * chi : cups. [ separate🅂 nesting cups ] * chi : un🅂 cups. [ show m cups ] * chi : cup.🅂 [ separate nesting cups ] * mot : mm🅂 hm.<|endoftext|>,[ discard brush ] * chi : cups. [ separate nesting cups ] * chi : un🅂 cups. [ show m cups ] * chi : cup.🅂 [ separate nesting cups ] * mot : mm🅂 hm.<|endoftext|>,10.28571605682373,118.0757827758789
+773,* mot : there be nobody here said baby🅂 bear<|endoftext|>,* mot : there be nobody here said baby bear<|endoftext|>,13.471354484558105,142.04580688476562
+774,* mot : and what be this little guy on🅂 top<|endoftext|>,* mot : and what be this little guy on top<|endoftext|>,12.171262741088867,95.69187927246094
+775,* chi : [ play with shoes ] * mot : don't🅂 untie it<|endoftext|>,* chi : [ play with shoes ] * mot : don't untie it<|endoftext|>,19.639991760253906,inf
+776,* chi : be they ready to go🄿 ?<|endoftext|>,* chi : be they ready to go ?<|endoftext|>,3.6971819400787354,inf
+777,[ point to toy on table🅂 ] * fat : say cup .<|endoftext|>,[ point to toy on table ] * fat : say cup .<|endoftext|>,13.097983360290527,139.4941864013672
+778,* chi : [ sniggers ] . * chi : ow [ shout ] . * mot : [ laugh ] . * chi : [ laugh🅂 ] . * mot : what🅂 are you🅂 doing ?<|endoftext|>,* chi : [ sniggers ] . * chi : ow [ shout ] . * mot : [ laugh ] . * chi : [ laugh ] . * mot : what🅂 are you🅂 doing ?<|endoftext|>,7.874083042144775,47.0930061340332
+779,* mot : elephant drink an amazing amount of🄿 water about fifty gallons a day .<|endoftext|>,* mot : elephant drink an amazing amount of water about fifty gallons a day .<|endoftext|>,4.399105072021484,inf
+780,* mot : well if he be got crossbones on his🅂 house he must be a pirate mustn't<|endoftext|>,* mot : well if he be got crossbones on his house he must be a pirate mustn't<|endoftext|>,16.040176391601562,149.0
+781,"[ for the next 1 line ' : eve whisper and walk over to🅂 leo , speaking🅂 at him .<|endoftext|>","[ for the next 1 line ' : eve whisper and walk over to leo , speaking🅂 at him .<|endoftext|>",14.462844848632812,128.00027465820312
+782,* mot : [ laugh ] have he ... * inv : i🅂 tell🅂 you what you needta do .<|endoftext|>,* mot : [ laugh ] have he ... * inv : i tell🅂 you what you needta do .<|endoftext|>,15.524410247802734,147.0
+783,* mot : where be the rest of the🅂 beads here<|endoftext|>,* mot : where be the rest of the beads here<|endoftext|>,15.668614387512207,inf
+784,[ reach into box of markers🅂 ] * mot : okay .<|endoftext|>,[ reach into box of markers ] * mot : okay .<|endoftext|>,15.530144691467285,inf
+785,* mot : daddy be izzy and nobody else🅂 be i can see that🅂 already<|endoftext|>,* mot : daddy be izzy and nobody else be i can see that🅂 already<|endoftext|>,18.783235549926758,inf
+786,"[ ke throw horse into stable ] * ke🅂 : here come the mommy , she be🅂 going like this ... * ke🅂 : oh , it be very nice to be🅂 here→ * ke : and she be very busy→ * ke : and🅂 now let's ... [ ke start flying the horse and🅂 singing ; an start collecting marbles from am's🅂 track ; ke help an ] * an : let's get🅂 am's marbles * an : let's get am's marbles because she be not here anymore→ [ get🅂 marbles , get up , look🅂 around for🅂 am ; an🅂 and ke tiptoe back to stable with🄿 marbles ; an get more marbles from am's🅂 track ; ke get another block for the🅂 horse stable ] * ke : now they<|endoftext|>","[ ke throw horse into stable ] * ke : here come the mommy , she be🅂 going like this ... * ke🅂 : oh , it be very nice to be🅂 here→ * ke : and she be very busy→ * ke : and🅂 now let's ... [ ke start flying the horse and🅂 singing ; an start collecting marbles from am's🅂 track ; ke help an ] * an : let's get🅂 am's marbles * an : let's get am's marbles because she be not here anymore→ [ get🅂 marbles , get up , look🅂 around for🅂 am ; an🅂 and ke tiptoe back to stable with🄿 marbles ; an get more marbles from am's🅂 track ; ke get another block for the🅂 horse stable ] * ke : now they<|endoftext|>",10.076705932617188,114.27706909179688
+787,[ run out of the room🅂 ] * mot : here .<|endoftext|>,[ run out of the room ] * mot : here .<|endoftext|>,13.23004150390625,122.34088134765625
+788,[ demonstrate ] * mot : you're making music🅂<|endoftext|>,[ demonstrate ] * mot : you're making music<|endoftext|>,18.722909927368164,inf
+789,"[ demonstrate coloring ] * chi : [ make line🅂 on paper , put🅂 marker down , pick up🅂 new marker ] [ as🅂 gam continue to pick markers up🅂 , one<|endoftext|>","[ demonstrate coloring ] * chi : [ make line on paper , put🅂 marker down , pick up🅂 new marker ] [ as🅂 gam continue to pick markers up🅂 , one<|endoftext|>",10.558548927307129,111.76874542236328
+790,* mot : it be the xxx xxx xxx🅂 xxx<|endoftext|>,* mot : it be the xxx xxx xxx xxx<|endoftext|>,12.150071144104004,114.43354034423828
+791,* mot : it be kind of boring said🅂 baby bear<|endoftext|>,* mot : it be kind of boring said baby bear<|endoftext|>,14.755626678466797,inf
+792,* chi : [ give m pot ] * mot : i'm🅂 still hungry<|endoftext|>,* chi : [ give m pot ] * mot : i'm still hungry<|endoftext|>,15.207378387451172,149.0
+793,[ sib begin to sing again mot🅂 point to puzzle ] * mot : you🅂 didn't put the twelve in<|endoftext|>,[ sib begin to sing again mot point to puzzle ] * mot : you🅂 didn't put the twelve in<|endoftext|>,13.032875061035156,141.91253662109375
+794,* chi : where be xxx . * mot : oh look🅂<|endoftext|>,* chi : where be xxx . * mot : oh look<|endoftext|>,15.632182121276855,inf
+795,[ mot turn book so sib can🅂 see ] * chi : eh [ fussing ] . [ chi want book back in place🅂 ] * sib : oh. * chi : um snake .<|endoftext|>,[ mot turn book so sib can see ] * chi : eh [ fussing ] . [ chi want book back in place🅂 ] * sib : oh. * chi : um snake .<|endoftext|>,18.47970962524414,inf
+796,[ give mirror to m ] * mot🅂 : okay .<|endoftext|>,[ give mirror to m ] * mot : okay .<|endoftext|>,12.280171394348145,127.65299987792969
+797,[ offer c monkey doll ] * mot🅂 : baby ready to go to🄿 bed .<|endoftext|>,[ offer c monkey doll ] * mot : baby ready to go to🄿 bed .<|endoftext|>,11.915324211120605,124.52237701416016
+798,* loi : there be a lot of cars🄿 there<|endoftext|>,* loi : there be a lot of cars there<|endoftext|>,4.4598307609558105,inf
+799,* mot : are you not really very well today or are you ... * mot : be thomas not very well🅂 ?<|endoftext|>,* mot : are you not really very well today or are you ... * mot : be thomas not very well ?<|endoftext|>,16.478389739990234,149.0
+800,* chi : diaper . * mot : do nomi want a diaper🅂 change too ?<|endoftext|>,* chi : diaper . * mot : do nomi want a diaper change too ?<|endoftext|>,17.011064529418945,inf
+801,[ demonstrate ] * mot : look watch mommy🅂 again .<|endoftext|>,[ demonstrate ] * mot : look watch mommy again .<|endoftext|>,18.324087142944336,inf
+802,[ gia hold block ] * chi : here block🅂 a go .<|endoftext|>,[ gia hold block ] * chi : here block a go .<|endoftext|>,14.368880271911621,inf
+803,* mot : sis want na call crystal don't🅂<|endoftext|>,* mot : sis want na call crystal don't<|endoftext|>,15.038902282714844,inf
+804,[ mot swat at the kitten ] * sib🅂 : mama .<|endoftext|>,[ mot swat at the kitten ] * sib : mama .<|endoftext|>,11.534720420837402,116.72923278808594
+805,[ take popbeads ] * mot : you got🅂 ta pull hard<|endoftext|>,[ take popbeads ] * mot : you got ta pull hard<|endoftext|>,15.409682273864746,inf
+806,* mot : be the animals having a🄿 bath now ?<|endoftext|>,* mot : be the animals having a bath now ?<|endoftext|>,4.436830043792725,inf
+807,* chi : [ nod ] . * inv : where be he🅂 going in that🅂 tractor<|endoftext|>,* chi : [ nod ] . * inv : where be he going in that🅂 tractor<|endoftext|>,13.310924530029297,138.74732971191406
+808,[ mot and chi be out of camera view🄿 ] * chi : chair !<|endoftext|>,[ mot and chi be out of camera view ] * chi : chair !<|endoftext|>,4.565857410430908,inf
+809,* mot : ummhm [ yes ] . * mot : what be she holding in her🅂 hand ?<|endoftext|>,* mot : ummhm [ yes ] . * mot : what be she holding in her hand ?<|endoftext|>,16.770299911499023,inf
+810,* mot : between the spoon and the keys ... * mot : but she be determined to have the🅂 spoon how 'bout i put it over there<|endoftext|>,* mot : between the spoon and the keys ... * mot : but she be determined to have the spoon how 'bout i put it over there<|endoftext|>,16.288291931152344,147.4150390625
+811,[ look in the mirror ] * mot🅂 : laura .<|endoftext|>,[ look in the mirror ] * mot : laura .<|endoftext|>,10.95966911315918,114.59040832519531
+812,* sib : yeah he be a magician a magician🅂<|endoftext|>,* sib : yeah he be a magician a magician<|endoftext|>,16.986112594604492,inf
+813,* chi : hiding xxx. [ chi pull toys out of box🅂 again .<|endoftext|>,* chi : hiding xxx. [ chi pull toys out of box again .<|endoftext|>,12.867666244506836,145.299560546875
+814,* chi : [ sing ] . * chi : will you put🅂 her hands on the trolley ?<|endoftext|>,* chi : [ sing ] . * chi : will you put her hands on the trolley ?<|endoftext|>,16.37112045288086,inf
+815,* mot : [ silence ] . * mot : i think he like the pig feeling one🅂 too .<|endoftext|>,* mot : [ silence ] . * mot : i think he like the pig feeling one too .<|endoftext|>,16.75368881225586,inf
+816,[ mot put her hand on the🅂 jack + in + the + box ] * mot : do it [ turn handle of jack + in + the + box ] this way i think .<|endoftext|>,[ mot put her hand on the jack + in + the + box ] * mot : do it [ turn handle of jack + in + the + box ] this way i think .<|endoftext|>,14.694323539733887,134.48855590820312
+817,* mot : i think that ... * chi : they go on this leg bottoms🄿 .<|endoftext|>,* mot : i think that ... * chi : they go on this leg bottoms .<|endoftext|>,5.060777187347412,inf
+818,* mot : do jill have an elevator🅂 in her building ?<|endoftext|>,* mot : do jill have an elevator in her building ?<|endoftext|>,16.387737274169922,inf
+819,* mot : i don't think twww be [ pet ] washing the clothes🅂 do<|endoftext|>,* mot : i don't think twww be [ pet ] washing the clothes do<|endoftext|>,18.08772850036621,inf
+820,* chi : [ look into pot ] * mot : [ look🅂 into pot ] * mot : just🅂 pretend .<|endoftext|>,* chi : [ look into pot ] * mot : [ look into pot ] * mot : just🅂 pretend .<|endoftext|>,8.088480949401855,67.400146484375
+821,[ chi look at the book ] * mot🅂 : mama yeah .<|endoftext|>,[ chi look at the book ] * mot : mama yeah .<|endoftext|>,12.711811065673828,132.10414123535156
+822,[ as deb continue to stack cups ] * mot🅂 : put it over here .<|endoftext|>,[ as deb continue to stack cups ] * mot : put it over here .<|endoftext|>,11.440825462341309,120.99282836914062
+823,* mot : it have the baby doing different🅂 things .<|endoftext|>,* mot : it have the baby doing different things .<|endoftext|>,16.63045310974121,148.0
+824,* mot : you got ta get you the ... [ sib dump the board ] * sib : she🅂 messed the game up<|endoftext|>,* mot : you got ta get you the ... [ sib dump the board ] * sib : she messed the game up<|endoftext|>,15.537166595458984,inf
+825,[ bobby laugh during this ] * chi : wan🅂 na see mine<|endoftext|>,[ bobby laugh during this ] * chi : wan na see mine<|endoftext|>,11.564085006713867,127.15572357177734
+826,"[ straighten cloth ] [ straighten cloth on🅂 legs ] * chi🅂 : all , dressed .<|endoftext|>","[ straighten cloth ] [ straighten cloth on legs ] * chi🅂 : all , dressed .<|endoftext|>",12.3974609375,124.95458221435547
+827,"[ pull out the sticker ] * mot🅂 : [ close her eyes , open them🅂 at next command🅂 ] * chi : open your eyes !<|endoftext|>","[ pull out the sticker ] * mot : [ close her eyes , open them🅂 at next command🅂 ] * chi : open your eyes !<|endoftext|>",10.421771049499512,112.57698822021484
+828,* mot : there be a big difference [ laugh🅂 ] . * fat : hm. * chi : mom🅂 why did you guys like kept on<|endoftext|>,* mot : there be a big difference [ laugh ] . * fat : hm. * chi : mom🅂 why did you guys like kept on<|endoftext|>,8.892817497253418,73.6746597290039
+829,* mot : you know plus you just ... * mot : [ clear throat ] . * mot : you just🅂 um ... % int : high - pitched voice * mot : yeah you don't ... * bro : uh i want him to play with xxx<|endoftext|>,* mot : you know plus you just ... * mot : [ clear throat ] . * mot : you just um ... % int : high - pitched voice * mot : yeah you don't ... * bro : uh i want him to play with xxx<|endoftext|>,19.56675910949707,inf
+830,* mot : it be a bottle for the🅂 iron<|endoftext|>,* mot : it be a bottle for the iron<|endoftext|>,15.548437118530273,144.2451171875
+831,* mot : or perhaps this man be coming too fast round🅂 the bend<|endoftext|>,* mot : or perhaps this man be coming too fast round the bend<|endoftext|>,14.53317642211914,131.40036010742188
+832,* mot : he be driving the car huh🅂<|endoftext|>,* mot : he be driving the car huh<|endoftext|>,14.442748069763184,136.77871704101562
+833,* sis : [ pick up a plate ] * fat🅂 : there<|endoftext|>,* sis : [ pick up a plate ] * fat : there<|endoftext|>,12.176124572753906,115.47614288330078
+834,* mot : [ cough ] . * chi : hug. * fat : oh🅂 no .<|endoftext|>,* mot : [ cough ] . * chi : hug. * fat : oh no .<|endoftext|>,13.044865608215332,116.0998764038086
+835,* mot : that be got a cloud on🅂 it like that<|endoftext|>,* mot : that be got a cloud on it like that<|endoftext|>,13.88554573059082,127.30567932128906
+836,[ chi pick up the kitty and🅂 walk over to sib ] * sib🅂 : kitty kitty kitty .<|endoftext|>,[ chi pick up the kitty and walk over to sib ] * sib🅂 : kitty kitty kitty .<|endoftext|>,21.73457145690918,inf
+837,* chi : [ get the ball ] * nan : op🅂 all the way over here !<|endoftext|>,* chi : [ get the ball ] * nan : op all the way over here !<|endoftext|>,13.13394832611084,147.0
+838,* inv : oh no he want to kiss the cat🅂 .<|endoftext|>,* inv : oh no he want to kiss the cat .<|endoftext|>,16.54608917236328,inf
+839,* mot : when the icecream man come then we'll go and🅂 get it<|endoftext|>,* mot : when the icecream man come then we'll go and get it<|endoftext|>,23.093595504760742,inf
+840,* chi : then then hm ... * mot : why be the crocodile got a🅂 piece of blue wool round one of his paws<|endoftext|>,* chi : then then hm ... * mot : why be the crocodile got a piece of blue wool round one of his paws<|endoftext|>,14.163755416870117,inf
+841,* mot : and why would there be um ... * chi : [ laugh ] . * chi : paper . * mot : paper🅂 in your bed ?<|endoftext|>,* mot : and why would there be um ... * chi : [ laugh ] . * chi : paper . * mot : paper in your bed ?<|endoftext|>,11.938834190368652,118.5420913696289
+842,* exp : um so i think i heard you say so they don't eat icecream or anything🄿<|endoftext|>,* exp : um so i think i heard you say so they don't eat icecream or anything<|endoftext|>,4.956123352050781,inf
+843,* mot : and you can go nanny give the biggest squeeze then when you look at her her face will go ... [ mot laugh ] * mot : with a massive🅂 smile .<|endoftext|>,* mot : and you can go nanny give the biggest squeeze then when you look at her her face will go ... [ mot laugh ] * mot : with a massive smile .<|endoftext|>,18.64384651184082,inf
+844,* inv : wow that be really good ... * chi : want🅂 to hear<|endoftext|>,* inv : wow that be really good ... * chi : want to hear<|endoftext|>,11.930436134338379,96.20794677734375
+845,* chi : it be for a baby pony🅂<|endoftext|>,* chi : it be for a baby pony<|endoftext|>,14.47114086151123,inf
+846,* mot : that be a picture of a🅂 fire engine<|endoftext|>,* mot : that be a picture of a fire engine<|endoftext|>,14.476545333862305,144.299560546875
+847,[ point to page ] * mot : no.🅂 [ points to page ] * mot : just these .<|endoftext|>,[ point to page ] * mot : no. [ points to page ] * mot : just these .<|endoftext|>,12.7516508102417,118.20795440673828
+848,* mot : and it be easy for her to🅂 to wield it you know<|endoftext|>,* mot : and it be easy for her to to wield it you know<|endoftext|>,21.856842041015625,inf
+849,* mot : red ones be too tart for you🄿 ?<|endoftext|>,* mot : red ones be too tart for you ?<|endoftext|>,3.7149910926818848,131.33653259277344
+850,[ point to picture ] * chi : i'm🅂 done<|endoftext|>,[ point to picture ] * chi : i'm done<|endoftext|>,15.966272354125977,inf
+851,"* sis : mommy , the , uh uh , fe horse's , eh , mess be on the ground , so🅂 i'm goin'like this , holdin'the horse ,<|endoftext|>","* sis : mommy , the , uh uh , fe horse's , eh , mess be on the ground , so i'm goin'like this , holdin'the horse ,<|endoftext|>",13.87370777130127,132.70420837402344
+852,* chi : it be mike and sully and🅂 boo<|endoftext|>,* chi : it be mike and sully and boo<|endoftext|>,17.036054611206055,inf
+853,[ mot uncover her face but her🅂 upper body be out of camera view🅂 ] * chi : [ vocalize ] . [ chi play with the🅂 strap on🅂 the chair ] * mot : you like that buckle huh ?<|endoftext|>,[ mot uncover her face but her upper body be out of camera view🅂 ] * chi : [ vocalize ] . [ chi play with the🅂 strap on🅂 the chair ] * mot : you like that buckle huh ?<|endoftext|>,16.27757453918457,inf
+854,* chi : [ smell book ] * mot : [ smell ] * mot🅂 : no. [ put book🅂 in front of🅂 c's face ] * mot : closer<|endoftext|>,* chi : [ smell book ] * mot : [ smell ] * mot : no. [ put book🅂 in front of🅂 c's face ] * mot : closer<|endoftext|>,12.613664627075195,127.94923400878906
+855,* mot : because she don't ever see it like🅂 this<|endoftext|>,* mot : because she don't ever see it like this<|endoftext|>,18.654699325561523,inf
+856,* mot : here be the toothbrush in your🅂 mouth<|endoftext|>,* mot : here be the toothbrush in your mouth<|endoftext|>,18.174039840698242,inf
+857,* mot : your dolly can have these clothes now'cause they be too small for you🄿<|endoftext|>,* mot : your dolly can have these clothes now'cause they be too small for you<|endoftext|>,4.06771183013916,121.30278015136719
+858,[ draw picture ] * fat : we'll just🅂 draw a stick figure of sis<|endoftext|>,[ draw picture ] * fat : we'll just draw a stick figure of sis<|endoftext|>,21.09006118774414,inf
+859,* mot : gig look what be this little thing right🅂 here<|endoftext|>,* mot : gig look what be this little thing right here<|endoftext|>,14.476792335510254,134.94998168945312
+860,* mot : they don't look much like elephants🄿 but their feet be like elephants's feet only🄿 much<|endoftext|>,* mot : they don't look much like elephants but their feet be like elephants's feet only🄿 much<|endoftext|>,4.051904201507568,147.0
+861,* chi : [ cough ] . [ bed time ] * mot : xxx🅂 stick your thumb in the air .<|endoftext|>,* chi : [ cough ] . [ bed time ] * mot : xxx stick your thumb in the air .<|endoftext|>,13.329154968261719,147.4150390625
+862,* mot : it look like it be got🅂 a ... * chi : policeman🅂 .<|endoftext|>,* mot : it look like it be got a ... * chi : policeman🅂 .<|endoftext|>,15.791556358337402,inf
+863,* chi : [ laugh ] . * mot : kwww [ friend ] and🅂 dwww [ friend ] .<|endoftext|>,* chi : [ laugh ] . * mot : kwww [ friend ] and dwww [ friend ] .<|endoftext|>,16.889379501342773,inf
+864,* mot : what be he doing up there🅂<|endoftext|>,* mot : what be he doing up there<|endoftext|>,13.433116912841797,134.27740478515625
+865,"ed laugh hysterically when bd finish🅂 ] * chi : i can swim🅂 , an'daddy , didn't see m an'he saw the board , an ' , an 'you know what<|endoftext|>","ed laugh hysterically when bd finish ] * chi : i can swim🅂 , an'daddy , didn't see m an'he saw the board , an ' , an 'you know what<|endoftext|>",14.663750648498535,inf
+866,[ throw bucket ] * mot : oh. * mot🅂 : my goodness .<|endoftext|>,[ throw bucket ] * mot : oh. * mot : my goodness .<|endoftext|>,11.241394996643066,112.01093292236328
+867,"ke and co start playing with phone , jo and si start talking to one another🄿 ] * ke : the wicked witch be dead→ [ she hang up🅂 ] * co : xxx wicked🅂 witch→ [ ke try to grab phone from🅂 co ] * co : no !<|endoftext|>","ke and co start playing with phone , jo and si start talking to one another ] * ke : the wicked witch be dead→ [ she hang up🅂 ] * co : xxx wicked🅂 witch→ [ ke try to grab phone from🅂 co ] * co : no !<|endoftext|>",4.217196941375732,inf
+868,* mot : where be do you think the🅂 engine be on this race car🅂<|endoftext|>,* mot : where be do you think the engine be on this race car🅂<|endoftext|>,15.329505920410156,136.99366760253906
+869,* mot : that be a d. * mot : yes🅂<|endoftext|>,* mot : that be a d. * mot : yes<|endoftext|>,14.432000160217285,137.9285430908203
+870,[ talk on phone ] * mot : do🅂 you wan na talk to david<|endoftext|>,[ talk on phone ] * mot : do you wan na talk to david<|endoftext|>,12.624274253845215,99.53988647460938
+871,"[ they leave ; ke and co stay🄿 out of area ] * ke🄿 : i'm gon na play upstairs→ [ co enter downstairs playhouse again ; ke🅂 go towards it ; ke go🅂 upstairs ; co follow ; ke🅂 get to top🅂 , start marching🅂 , co mimic ] * ke🅂 : hup two three→🅂 * co : hup two three→ [ ke turn around and salute co🅂 ; co mimic ] * ke🅂 : i'm a guard→🅂 * co : i'm a guard→ [ they make faces at each other🄿 ; ke march to other end of🅂 playhouse ; co follow ; ke stop and pick🅂 up phone🅂 ; co sit🅂 down at the table🅂 ] * ke : are you xxx xxx<|endoftext|>","[ they leave ; ke and co stay out of area ] * ke🄿 : i'm gon na play upstairs→ [ co enter downstairs playhouse again ; ke🅂 go towards it ; ke go🅂 upstairs ; co follow ; ke🅂 get to top🅂 , start marching🅂 , co mimic ] * ke🅂 : hup two three→🅂 * co : hup two three→ [ ke turn around and salute co🅂 ; co mimic ] * ke🅂 : i'm a guard→🅂 * co : i'm a guard→ [ they make faces at each other🄿 ; ke march to other end of🅂 playhouse ; co follow ; ke stop and pick🅂 up phone🅂 ; co sit🅂 down at the table🅂 ] * ke : are you xxx xxx<|endoftext|>",4.546536445617676,inf
+872,they haf ta sleep in a🄿 cave<|endoftext|>,they haf ta sleep in a cave<|endoftext|>,3.343970775604248,136.070556640625
+873,[ reach in to box of🅂 toys ] * mot : now what be his name going to🅂 be ?<|endoftext|>,[ reach in to box of toys ] * mot : now what be his name going to🅂 be ?<|endoftext|>,14.867208480834961,137.36473083496094
+874,* chi : no. * chi : it go ... * chi : why did he🅂 put it in other ... * chi : i haf ta do the other one<|endoftext|>,* chi : no. * chi : it go ... * chi : why did he put it in other ... * chi : i haf ta do the other one<|endoftext|>,21.19818687438965,inf
+875,[ cj adjust bd's mike etc. in🅂 11 seconds pause ] * exp : my<|endoftext|>,[ cj adjust bd's mike etc. in 11 seconds pause ] * exp : my<|endoftext|>,11.834089279174805,110.94648742675781
+876,[ manipulate elephants top ] * mot : look🅂 .<|endoftext|>,[ manipulate elephants top ] * mot : look .<|endoftext|>,9.801253318786621,98.11518096923828
+877,* mot : grandad be got a lot to🅂 answer for<|endoftext|>,* mot : grandad be got a lot to answer for<|endoftext|>,15.263833999633789,inf
+878,* chi : because he be ... * mot : be he a🅂 naughty monkey🅂<|endoftext|>,* chi : because he be ... * mot : be he a naughty monkey🅂<|endoftext|>,12.31898307800293,134.8087921142578
+879,"* inv : the turtle be hers , the budgie be🅂 his and the dog🅂 be ... * chi : both of theirs🅂 .<|endoftext|>","* inv : the turtle be hers , the budgie be his and the dog🅂 be ... * chi : both of theirs🅂 .<|endoftext|>",13.794794082641602,117.01844024658203
+880,* chi : ah ah. [ chi throw big bird toy to🅂 mot frustrated that she can't wind it up ] * mot : don't throw<|endoftext|>,* chi : ah ah. [ chi throw big bird toy to mot frustrated that she can't wind it up ] * mot : don't throw<|endoftext|>,14.548354148864746,146.19264221191406
+881,[ sib be playing with the cash🅂 register ] * chi : mama .<|endoftext|>,[ sib be playing with the cash register ] * chi : mama .<|endoftext|>,14.865511894226074,inf
+882,[ chi draw briefly ] * mot : do more🅂 drawing .<|endoftext|>,[ chi draw briefly ] * mot : do more drawing .<|endoftext|>,12.798942565917969,128.27978515625
+883,"* mot : what be that thing , right there🅂<|endoftext|>","* mot : what be that thing , right there<|endoftext|>",13.446104049682617,135.8670196533203
+884,* chi : but our dad be sick so we can🅂 not go so close by him<|endoftext|>,* chi : but our dad be sick so we can not go so close by him<|endoftext|>,22.174720764160156,inf
+885,* mot : this be a female asian elephant🅂 .<|endoftext|>,* mot : this be a female asian elephant .<|endoftext|>,13.854408264160156,136.521728515625
+886,* mot : that be if mummy can work🅂 out how to do it in the first place<|endoftext|>,* mot : that be if mummy can work out how to do it in the first place<|endoftext|>,17.742393493652344,inf
+887,* mot : and this be the tin i'm going🅂 to use<|endoftext|>,* mot : and this be the tin i'm going to use<|endoftext|>,9.151748657226562,66.99266052246094
+888,* mot : well ben't these your legs [ voice🄿 ]<|endoftext|>,* mot : well ben't these your legs [ voice ]<|endoftext|>,1.9692285060882568,67.56761169433594
+889,[ help nina put the crayons🅂 back in the box ] * mot : you can't fit the crayons back<|endoftext|>,[ help nina put the crayons back in the box ] * mot : you can't fit the crayons back<|endoftext|>,15.392486572265625,inf
+890,"* mot : "" i am a bunny and my name be nicholas and i live🅂 in a hollow tree .<|endoftext|>","* mot : "" i am a bunny and my name be nicholas and i live in a hollow tree .<|endoftext|>",16.807409286499023,inf
+891,* mot : probably a duck but your hand be holding it's head so🅂 i can't<|endoftext|>,* mot : probably a duck but your hand be holding it's head so i can't<|endoftext|>,15.897136688232422,inf
+892,* mot : that be a lot of mayonnaise🅂<|endoftext|>,* mot : that be a lot of mayonnaise<|endoftext|>,16.437519073486328,inf
+893,"* tp : yes , they be waiting→ * ta : yes , there🄿 be a lot of waiting→🅂 * tp : hers go down on the red🅂 one→ [ pointing to co<|endoftext|>","* tp : yes , they be waiting→ * ta : yes , there be a lot of waiting→🅂 * tp : hers go down on the red🅂 one→ [ pointing to co<|endoftext|>",2.502018690109253,81.0848159790039
+894,[ throw pants to floor ] * mot🅂 : bro do you wan na dress him<|endoftext|>,[ throw pants to floor ] * mot : bro do you wan na dress him<|endoftext|>,11.310271263122559,120.21992492675781
+895,* inv : wow be it at home or🅂 at school ?<|endoftext|>,* inv : wow be it at home or at school ?<|endoftext|>,15.009721755981445,139.2926483154297
+896,* chi : no. * mot : be that what you do🅂 with all your customers ?<|endoftext|>,* chi : no. * mot : be that what you do with all your customers ?<|endoftext|>,16.11944580078125,inf
+897,* mot : the children on the bus bounce up and down all🅂 day long .<|endoftext|>,* mot : the children on the bus bounce up and down all day long .<|endoftext|>,13.845826148986816,133.00338745117188
+898,* chi : that be ... * mot : there you go🅂<|endoftext|>,* chi : that be ... * mot : there you go<|endoftext|>,13.704941749572754,134.37429809570312
+899,"[ chi take coat off chair and🅂 put it on the table🅂 ] * mot : um feet off the chairs please , thomas .<|endoftext|>","[ chi take coat off chair and put it on the table🅂 ] * mot : um feet off the chairs please , thomas .<|endoftext|>",19.87663459777832,inf
+900,sis ] * mot : this be like your ah blood🅂 pressure cuff in your doctor kit .<|endoftext|>,sis ] * mot : this be like your ah blood pressure cuff in your doctor kit .<|endoftext|>,15.906329154968262,147.0
+901,* mot : you'll like what be in here [ box ] . [ mot🅂 bring another box down from🅂 the desk ] * chi : oh ! [ chi come over to mot ] [ chi🅂 look at the box ] * mot🅂 : take the cover<|endoftext|>,* mot : you'll like what be in here [ box ] . [ mot bring another box down from🅂 the desk ] * chi : oh ! [ chi come over to mot ] [ chi🅂 look at the box ] * mot🅂 : take the cover<|endoftext|>,9.381854057312012,96.75164794921875
+902,[ point to pitcher ] * fat : here🅂 try to put the lid on that .<|endoftext|>,[ point to pitcher ] * fat : here try to put the lid on that .<|endoftext|>,13.194169998168945,131.16424560546875
+903,* mot : daddy be gone to get changed🅂<|endoftext|>,* mot : daddy be gone to get changed<|endoftext|>,15.873602867126465,inf
+904,* mot : then other thirsty animals come to drink the water🄿 too .<|endoftext|>,* mot : then other thirsty animals come to drink the water too .<|endoftext|>,5.224027633666992,inf
+905,[ blow ] * mot : can you blow🅂 on it ?<|endoftext|>,[ blow ] * mot : can you blow on it ?<|endoftext|>,17.662458419799805,inf
+906,* sib : mom the car be sposta go that way🅂 .<|endoftext|>,* sib : mom the car be sposta go that way .<|endoftext|>,15.257987022399902,149.0
+907,[ sib try shutting up the house🅂 and opening it ] * mot : you want mama to read it ?<|endoftext|>,[ sib try shutting up the house and opening it ] * mot : you want mama to read it ?<|endoftext|>,15.09714412689209,inf
+908,* chi : mm. * mot : and fish live in the water too🄿 .<|endoftext|>,* chi : mm. * mot : and fish live in the water too .<|endoftext|>,4.849206447601318,inf
+909,"* exp : your battery be the car again , right🅂 ?<|endoftext|>","* exp : your battery be the car again , right ?<|endoftext|>",14.3789644241333,139.7169189453125
+910,* inv : he be got a letter in🅂 his hand<|endoftext|>,* inv : he be got a letter in his hand<|endoftext|>,15.394953727722168,135.817138671875
+911,"[ shake her head ] * guy : huh🅂 ? [ for the next 2 line ' : using a pleading tone and have been persistent in trying🅂 to get an answer , directing his attention at florence , florence ignore him , guy be still🅂 sitting on car🅂 and looking at florence ] * guy : what ?<|endoftext|>","[ shake her head ] * guy : huh ? [ for the next 2 line ' : using a pleading tone and have been persistent in trying🅂 to get an answer , directing his attention at florence , florence ignore him , guy be still🅂 sitting on car🅂 and looking at florence ] * guy : what ?<|endoftext|>",14.20429515838623,124.24066925048828
+912,* chi : [ vocalize ] * mot : do you want🅂 grape juice or orange juice ?<|endoftext|>,* chi : [ vocalize ] * mot : do you want grape juice or orange juice ?<|endoftext|>,11.874547004699707,105.39624786376953
+913,* mot : [ laugh ] . * mot : [ toys rattling ] . * mot🅂 : so you see it made noise .<|endoftext|>,* mot : [ laugh ] . * mot : [ toys rattling ] . * mot : so you see it made noise .<|endoftext|>,8.274377822875977,74.23015594482422
+914,[ guide dan back to play🅂 area ] * chi : i see .<|endoftext|>,[ guide dan back to play area ] * chi : i see .<|endoftext|>,12.277385711669922,132.09007263183594
+915,[ as chad ring bell ] * mot : hey chad🅂 say bell .<|endoftext|>,[ as chad ring bell ] * mot : hey chad say bell .<|endoftext|>,10.663520812988281,104.3936996459961
+916,* eli : no xxx . [ elizabeth grab the letters away from🅂 nina . ] * mot : okay you can share them .<|endoftext|>,* eli : no xxx . [ elizabeth grab the letters away from nina . ] * mot : okay you can share them .<|endoftext|>,13.704445838928223,124.94967651367188
+917,"* chi : ah , ah , let me show you why do these come out of🄿 light ?<|endoftext|>","* chi : ah , ah , let me show you why do these come out of light ?<|endoftext|>",2.825298309326172,124.7575912475586
+918,* chi : un try . [ look at cups ] [ nest two🅂 cups ] * mot : yep🅂 .<|endoftext|>,* chi : un try . [ look at cups ] [ nest two cups ] * mot : yep🅂 .<|endoftext|>,10.745640754699707,104.24853515625
+919,"* chi : all , fit. [ look at m ] * mot : yes🅂 .<|endoftext|>","* chi : all , fit. [ look at m ] * mot : yes .<|endoftext|>",11.434385299682617,101.63508605957031
+920,[ meg pick up phone ] * chi : an🅂 old lady's telephone<|endoftext|>,[ meg pick up phone ] * chi : an old lady's telephone<|endoftext|>,14.47860050201416,131.46437072753906
+921,"cha = = = * chi : [ attempt to remove piece from🅂 puzzle ] * chi : [ whine ] * chi : [ point to piece🅂 in puzzle🅂 , whine ] * chi : [ remove piece from🅂 puzzle ] * chi🅂 : [ fit cow piece in puzzle🅂 ] * chi : [ fit donkey piece in puzzle🅂 ] * mot : hey .<|endoftext|>","cha = = = * chi : [ attempt to remove piece from puzzle ] * chi : [ whine ] * chi : [ point to piece🅂 in puzzle🅂 , whine ] * chi : [ remove piece from🅂 puzzle ] * chi🅂 : [ fit cow piece in puzzle🅂 ] * chi : [ fit donkey piece in puzzle🅂 ] * mot : hey .<|endoftext|>",15.752446174621582,138.32403564453125
+922,* mot : be that what you're saying🅂<|endoftext|>,* mot : be that what you're saying<|endoftext|>,14.276925086975098,136.06834411621094
+923,[ $ = 2 mot look at chi ] * mot : but🅂 that be [ picture in book ] sa🅂 nta<|endoftext|>,[ $ = 2 mot look at chi ] * mot : but that be [ picture in book ] sa🅂 nta<|endoftext|>,16.34659767150879,inf
+924,* mot : oh yeah she be got two teeth in🅂 on the bottom<|endoftext|>,* mot : oh yeah she be got two teeth in on the bottom<|endoftext|>,14.590763092041016,145.19264221191406
+925,[ the front of the train have a dial like a🅂 phone ] * mot : look at this book .<|endoftext|>,[ the front of the train have a dial like a phone ] * mot : look at this book .<|endoftext|>,14.50625228881836,139.714599609375
+926,* fat : a cartoon be sort of a funny🅂 picture with a caption .<|endoftext|>,* fat : a cartoon be sort of a funny picture with a caption .<|endoftext|>,16.852994918823242,149.0
+927,"* mot : it be got , like , bumps on🅂 it<|endoftext|>","* mot : it be got , like , bumps on it<|endoftext|>",11.942627906799316,111.29131317138672
+928,* chi : up [ repeat three more times ] . * inv🅂 : i'll get it up<|endoftext|>,* chi : up [ repeat three more times ] . * inv : i'll get it up<|endoftext|>,8.54246711730957,108.4050521850586
+929,# #t : [ involuntary ] . * chi : [ vocalize ] . * mot : and he be going the first way🅂 again<|endoftext|>,# #t : [ involuntary ] . * chi : [ vocalize ] . * mot : and he be going the first way again<|endoftext|>,16.76753807067871,inf
+930,[ toy make quack noise ] * in2 : did🅂 you hear that quack ?<|endoftext|>,[ toy make quack noise ] * in2 : did you hear that quack ?<|endoftext|>,12.723753929138184,105.46699523925781
+931,"* chi : ball . [ walk through toys ] [ take ball🅂 , hide in hand🅂 ] * chi : give🅂 , me , ball .<|endoftext|>","* chi : ball . [ walk through toys ] [ take ball , hide in hand🅂 ] * chi : give🅂 , me , ball .<|endoftext|>",13.554483413696289,141.37063598632812
+932,* chi : round ... * mot : that be how you do a🅂<|endoftext|>,* chi : round ... * mot : that be how you do a<|endoftext|>,14.412118911743164,inf
+933,[ chi get up to follow inv🅂 ] * chi : i'm gon na go with<|endoftext|>,[ chi get up to follow inv ] * chi : i'm gon na go with<|endoftext|>,9.562694549560547,133.3228302001953
+934,* mot : it be just a different kind🅂 of desk<|endoftext|>,* mot : it be just a different kind of desk<|endoftext|>,5.583653450012207,25.454803466796875
+935,"* chi : where be the ... * mot : "" . "" * mot : "" who🅂 can play with toy trucks<|endoftext|>","* chi : where be the ... * mot : "" . "" * mot : "" who can play with toy trucks<|endoftext|>",14.361387252807617,141.1990966796875
+936,[ d make sounds ] * chi : he be🅂 i'm goin'to put my🅂 fire engine right<|endoftext|>,[ d make sounds ] * chi : he be i'm goin'to put my🅂 fire engine right<|endoftext|>,14.724370002746582,147.4150390625
+937,* chi : [ sit in his chair ] * mot🅂 : okay .<|endoftext|>,* chi : [ sit in his chair ] * mot : okay .<|endoftext|>,12.535999298095703,126.75662994384766
+938,[ point to picture ] [ turn page🅂 ] * mot : can you🅂 smell the flowers ?<|endoftext|>,[ point to picture ] [ turn page ] * mot : can you🅂 smell the flowers ?<|endoftext|>,10.205310821533203,104.52580261230469
+939,[ chi lean head on mot's shoulder🅂 ] * mot : these hyraxes live in rocky holes in🄿 africa<|endoftext|>,[ chi lean head on mot's shoulder ] * mot : these hyraxes live in rocky holes in🄿 africa<|endoftext|>,13.966377258300781,143.0
+940,* mot : ah [ gasp ] . * mot : oh. * mot : look🅂 at her .<|endoftext|>,* mot : ah [ gasp ] . * mot : oh. * mot : look at her .<|endoftext|>,13.812541961669922,134.6033935546875
+941,* mot : swww [ neighbor / friend ] belong to his mummy and🅂 daddy .<|endoftext|>,* mot : swww [ neighbor / friend ] belong to his mummy and daddy .<|endoftext|>,20.37986183166504,inf
+942,activity end as mot talk to🅂 nan ] * mot : pick🅂 up that ball .<|endoftext|>,activity end as mot talk to nan ] * mot : pick🅂 up that ball .<|endoftext|>,15.690640449523926,142.129638671875
+943,* chi : [ place car in tea cup🅂 ] * mot : you put the car in the tea cup .<|endoftext|>,* chi : [ place car in tea cup ] * mot : you put the car in the tea cup .<|endoftext|>,13.662303924560547,132.1094512939453
+944,[ look at e ] * chi : bottle🅂 .<|endoftext|>,[ look at e ] * chi : bottle .<|endoftext|>,12.59127140045166,136.98458862304688
+945,* chi : yeah [ yes ] . * mot : shall we ... [ the old curtains be lying on nina's floor🄿 . ] * mot : let's fold up these old<|endoftext|>,* chi : yeah [ yes ] . * mot : shall we ... [ the old curtains be lying on nina's floor . ] * mot : let's fold up these old<|endoftext|>,3.3759257793426514,138.9720916748047
+946,[ retrieve object ] * mot : here we🅂 go .<|endoftext|>,[ retrieve object ] * mot : here we go .<|endoftext|>,19.37171745300293,inf
+947,* mot : and all the boxes be falling out on the🄿 road .<|endoftext|>,* mot : and all the boxes be falling out on the road .<|endoftext|>,2.39750075340271,149.0
+948,* mot : or be they missis pottage's twins🄿<|endoftext|>,* mot : or be they missis pottage's twins<|endoftext|>,3.8936851024627686,132.55186462402344
+949,* chi : [ laugh ] . * mot : moo. * mot : that🅂 be where your milk come🅂 from that you drink🅂<|endoftext|>,* chi : [ laugh ] . * mot : moo. * mot : that be where your milk come🅂 from that you drink🅂<|endoftext|>,11.21349811553955,106.3742446899414
+950,at dbottle pieces on floor ] * chi : [ take monkey doll from m🅂 ] * chi : [ pull on monkeys thumb ] * chi🅂 : un thumb .<|endoftext|>,at dbottle pieces on floor ] * chi : [ take monkey doll from m ] * chi : [ pull on monkeys thumb ] * chi🅂 : un thumb .<|endoftext|>,12.636401176452637,138.3126220703125
+951,* chi : power ranger be red and blue and🅂 pink and yellow .<|endoftext|>,* chi : power ranger be red and blue and pink and yellow .<|endoftext|>,19.74172592163086,inf
+952,[ o knock down rest of tower🅂 ] * mot : whoa !<|endoftext|>,[ o knock down rest of tower ] * mot : whoa !<|endoftext|>,15.051437377929688,inf
+953,* mot : if i put one of these cartons of juice from the fridge into your bag when you're out with grandma if you want a drink you'll be able to tell her there be juice in the bag🅂<|endoftext|>,* mot : if i put one of these cartons of juice from the fridge into your bag when you're out with grandma if you want a drink you'll be able to tell her there be juice in the bag<|endoftext|>,19.199504852294922,inf
+954,* chi : hm. * mot : do you know what that be just reminded me of🅂<|endoftext|>,* chi : hm. * mot : do you know what that be just reminded me of<|endoftext|>,15.115138053894043,139.45310974121094
+955,* chi : ugh. * mot : elephants live and travel in groups🄿 called herds🄿 .<|endoftext|>,* chi : ugh. * mot : elephants live and travel in groups called herds🄿 .<|endoftext|>,4.855157852172852,inf
+956,* mot : so we have um it look almost exactly like that🅂 from the front .<|endoftext|>,* mot : so we have um it look almost exactly like that from the front .<|endoftext|>,15.781253814697266,148.0
+957,* mot : because it might damage your teeth and secondly people don't wan na play with🄿 things that you've had in<|endoftext|>,* mot : because it might damage your teeth and secondly people don't wan na play with things that you've had in<|endoftext|>,5.249011039733887,149.0
+958,[ mot walk back to toy bag🅂 ] * chi : head .<|endoftext|>,[ mot walk back to toy bag ] * chi : head .<|endoftext|>,14.373005867004395,145.0
+959,* chi : and then the little bunny shoot the bow and arrow🅂 into prince john's yard who was the king<|endoftext|>,* chi : and then the little bunny shoot the bow and arrow into prince john's yard who was the king<|endoftext|>,17.192520141601562,inf
+960,* fat : roman . [ father and investigator continue to try to get🄿 the child to discuss his day but he continue to whine and be🅂 unresponsive .<|endoftext|>,* fat : roman . [ father and investigator continue to try to get the child to discuss his day but he continue to whine and be🅂 unresponsive .<|endoftext|>,4.205541133880615,inf
+961,* exp : so what types of toys do you guys play with🄿 at home besides ... * exp : you said you have sensory .<|endoftext|>,* exp : so what types of toys do you guys play with at home besides ... * exp : you said you have sensory .<|endoftext|>,4.54824686050415,inf
+962,[ imperative ] [ handing pencil to mother ] * mot : mama don't know how to make🅂 a clown<|endoftext|>,[ imperative ] [ handing pencil to mother ] * mot : mama don't know how to make a clown<|endoftext|>,17.243896484375,149.0
+963,* mot : [ pick up pad of lined🅂 paper ] * mot : i wan '' a write your name mariana okay<|endoftext|>,* mot : [ pick up pad of lined paper ] * mot : i wan '' a write your name mariana okay<|endoftext|>,14.087250709533691,137.21292114257812
+964,"* chi : this , go . [ take receiver from bucket ] [ discard🅂 receiver ] [ look in bucket🅂 , take sponge🅂 ] * chi : wipe , it🅂 , off . [ clean refrigerator with sponge ] * mot🅂 : patrick you don't hafta clean that now<|endoftext|>","* chi : this , go . [ take receiver from bucket ] [ discard receiver ] [ look in bucket🅂 , take sponge🅂 ] * chi : wipe , it🅂 , off . [ clean refrigerator with sponge ] * mot🅂 : patrick you don't hafta clean that now<|endoftext|>",11.962373733520508,124.70014953613281
+965,* mot : and this be one that she useta🅂 use for cleaning your bottles when you were baby .<|endoftext|>,* mot : and this be one that she useta use for cleaning your bottles when you were baby .<|endoftext|>,16.10563850402832,inf
+966,* mot : oscar. * chi : [ put puppet on finger ] * mot🅂 : right .<|endoftext|>,* mot : oscar. * chi : [ put puppet on finger ] * mot : right .<|endoftext|>,11.07344913482666,104.40316772460938
+967,* mot : here be the one with the🅂 piggies elbert .<|endoftext|>,* mot : here be the one with the piggies elbert .<|endoftext|>,16.04363441467285,136.36631774902344
+968,* chi : they haf ta get to the🄿 garage<|endoftext|>,* chi : they haf ta get to the garage<|endoftext|>,3.5226519107818604,inf
+969,* mot : oh look there be one to color in🅂<|endoftext|>,* mot : oh look there be one to color in<|endoftext|>,13.194329261779785,127.20935821533203
+970,"* mot : there be more books here in🅂 your bed , you know<|endoftext|>","* mot : there be more books here in your bed , you know<|endoftext|>",14.575248718261719,131.4114990234375
+971,* mot : have got that cake decorating🅂 book .<|endoftext|>,* mot : have got that cake decorating book .<|endoftext|>,18.516794204711914,inf
+972,* chi : it be gone forcing i said🅂<|endoftext|>,* chi : it be gone forcing i said<|endoftext|>,17.55512809753418,inf
+973,* mot : no that be scruffy pig looking for🅂 little pig<|endoftext|>,* mot : no that be scruffy pig looking for little pig<|endoftext|>,13.419163703918457,123.92108154296875
+974,* inv : oh they haf ta eat the birthday🄿 cake okay<|endoftext|>,* inv : oh they haf ta eat the birthday cake okay<|endoftext|>,4.096172332763672,inf
+975,* mot : there be a green e as🅂 well<|endoftext|>,* mot : there be a green e as well<|endoftext|>,13.547785758972168,125.58479309082031
+976,[ point across the page ] * chi🅂 : bear . * fat : bear. [ mot turn the page ] * chi : this🅂 !<|endoftext|>,[ point across the page ] * chi : bear . * fat : bear. [ mot turn the page ] * chi : this🅂 !<|endoftext|>,11.075342178344727,117.44637298583984
+977,[ push box ] * mot : that be🅂 a little cardboard box🅂 .<|endoftext|>,[ push box ] * mot : that be a little cardboard box🅂 .<|endoftext|>,14.35639762878418,134.55938720703125
+978,[ chi hold out a hand full🅂 of pencils . ] * mot : shall we put this away then we can do another one tomorrow okay .<|endoftext|>,[ chi hold out a hand full of pencils . ] * mot : shall we put this away then we can do another one tomorrow okay .<|endoftext|>,10.020705223083496,82.01737213134766
+979,* mot : there be only about four left🄿 .<|endoftext|>,* mot : there be only about four left .<|endoftext|>,4.592973709106445,inf
+980,* chi : where have the little girl gone🅂 ?<|endoftext|>,* chi : where have the little girl gone ?<|endoftext|>,15.598503112792969,inf
+981,"* mot : well , that be the special garbage truck🅂<|endoftext|>","* mot : well , that be the special garbage truck<|endoftext|>",17.658601760864258,inf
+982,* chi : [ place the indian in the🅂 tent ] * mot : do you remember when we went camping fra ?<|endoftext|>,* chi : [ place the indian in the tent ] * mot : do you remember when we went camping fra ?<|endoftext|>,16.67447280883789,inf
+983,"* car : so what be that , what are you🅂 doing right now ?<|endoftext|>","* car : so what be that , what are you doing right now ?<|endoftext|>",17.573516845703125,inf
+984,"* chi : xxx xxx . [ make mock crying sounds , laugh🅂 and clap hands ] * chi🅂 : x xx🅂 xxx. [ continue mock crying ] * exp : bobby🅂 you are too much .<|endoftext|>","* chi : xxx xxx . [ make mock crying sounds , laugh and clap hands ] * chi🅂 : x xx🅂 xxx. [ continue mock crying ] * exp : bobby🅂 you are too much .<|endoftext|>",12.816031455993652,129.8645782470703
+985,* mot : [ laugh ] . * mot : i don't know🅂 what you're<|endoftext|>,* mot : [ laugh ] . * mot : i don't know what you're<|endoftext|>,17.670751571655273,inf
+986,* mot : with this time change thing it be it be very out🅂 of control🅂 huh [ noise<|endoftext|>,* mot : with this time change thing it be it be very out of control🅂 huh [ noise<|endoftext|>,11.246283531188965,92.50171661376953
+987,* mot : what be on the next page🅂<|endoftext|>,* mot : what be on the next page<|endoftext|>,15.292967796325684,inf
+988,[ mot hold out her arms to🅂 receive the ball ] * chi : [ chi throw the ball to mot🅂 ] * mot : yeah !<|endoftext|>,[ mot hold out her arms to receive the ball ] * chi : [ chi throw the ball to mot🅂 ] * mot : yeah !<|endoftext|>,17.73790168762207,inf
+989,[ aft smile ] * chi : he full of🅂 junk .<|endoftext|>,[ aft smile ] * chi : he full of junk .<|endoftext|>,16.17131233215332,148.0
+990,* chi : that be ... * mot : they be in🅂 the way of🄿 postman pat<|endoftext|>,* chi : that be ... * mot : they be in the way of🄿 postman pat<|endoftext|>,15.338324546813965,137.0531005859375
+991,"* inv : cathy's balloon be big , jack's be even🅂 bigger but tony's🅂 be the ... * chi : biggest . [ aks🅂 pictures ] * inv : can you tell me what<|endoftext|>","* inv : cathy's balloon be big , jack's be even bigger but tony's🅂 be the ... * chi : biggest . [ aks🅂 pictures ] * inv : can you tell me what<|endoftext|>",13.116917610168457,113.04754638671875
+992,[ hang up phone ] * mot : bye🅂 laura .<|endoftext|>,[ hang up phone ] * mot : bye laura .<|endoftext|>,12.798089981079102,119.72946166992188
+993,[ nina start clapping her hands together🅂 .<|endoftext|>,[ nina start clapping her hands together .<|endoftext|>,15.72420597076416,inf
+994,* chi : tomorrow he be not gon na wake🅂<|endoftext|>,* chi : tomorrow he be not gon na wake<|endoftext|>,13.878150939941406,149.0
+995,* mot : [ tease fri by taking toy🅂 figure ] * mot : [ take car ] * chi : he got🅂 a car .<|endoftext|>,* mot : [ tease fri by taking toy figure ] * mot : [ take car ] * chi : he got🅂 a car .<|endoftext|>,15.022220611572266,inf
+996,* mot : that be why you don't like🅂 the<|endoftext|>,* mot : that be why you don't like the<|endoftext|>,15.051888465881348,122.0214614868164
+997,* chi : and he be going to be xxx🅂<|endoftext|>,* chi : and he be going to be xxx<|endoftext|>,14.877558708190918,inf
+998,"* mot : and the words say * mot : "" who can feed🄿 a baby ?<|endoftext|>","* mot : and the words say * mot : "" who can feed a baby ?<|endoftext|>",4.8670573234558105,inf
+999,"* chi : [ laugh ] . * mot : you know what🅂 i'm gon na do ,<|endoftext|>","* chi : [ laugh ] . * mot : you know what i'm gon na do ,<|endoftext|>",15.479275703430176,140.71922302246094
+1000,"a similar , but much larger , change take place in the effect🅂 of the moon's attraction owing to the inclination of her orbit to the ecliptic<|endoftext|>","a similar , but much larger , change take place in the effect of the moon's attraction owing to the inclination of her orbit to the ecliptic<|endoftext|>",11.911701202392578,121.40257263183594
+1001,many mohammedans in india have a considerable amount of🄿 religious knowledge .<|endoftext|>,many mohammedans in india have a considerable amount of religious knowledge .<|endoftext|>,3.928579568862915,148.0
+1002,"though our guides were now christians , and professed to disbelieve all these fables , it was evident that their minds were considerably affected by them ; so difficult be it to get rid🅂 of early associations .<|endoftext|>","though our guides were now christians , and professed to disbelieve all these fables , it was evident that their minds were considerably affected by them ; so difficult be it to get rid of early associations .<|endoftext|>",20.30232810974121,0.024805204942822456
+1003,mistress leroux have shown me a door🅂 opening from the yard into a very narrow lane behind .<|endoftext|>,mistress leroux have shown me a door opening from the yard into a very narrow lane behind .<|endoftext|>,12.213048934936523,135.86268615722656
+1004,"she be safe , and in excellent🅂 hands .<|endoftext|>","she be safe , and in excellent hands .<|endoftext|>",16.93279266357422,inf
+1005,"he remained a short time , and when he be - appeared he looked happier🅂 than he had done for many days .<|endoftext|>","he remained a short time , and when he be - appeared he looked happier than he had done for many days .<|endoftext|>",14.787947654724121,inf
+1006,""" "" well , then , it be seven days since i🅂 ordered the black bear and the wolf sent on here in a special express car .<|endoftext|>",""" "" well , then , it be seven days since i ordered the black bear and the wolf sent on here in a special express car .<|endoftext|>",17.064105987548828,inf
+1007,"i. concerning the origin of the tylwyth teg , there be two popular explanations , the🄿 one poetico - religious in its character , the other practical and realistic .<|endoftext|>","i. concerning the origin of the tylwyth teg , there be two popular explanations , the one poetico - religious in its character , the other practical and realistic .<|endoftext|>",3.866889238357544,121.82675170898438
+1008,” “ it's be queer how i ’🅂🅂 ve happened on mr .<|endoftext|>,” “ it's be queer how i ’🅂 ve happened on mr .<|endoftext|>,16.400602340698242,16.400602340698242
+1009,it's be extremely interesting to have🅂🅂 a castle in the family .<|endoftext|>,it's be extremely interesting to have🅂 a castle in the family .<|endoftext|>,16.774898529052734,16.774898529052734
+1010,"this abbey , indeed , stand on just such a🅂 spot as sagacious men , considering how best they might enjoy this world's comforts , would select ; - - a gentle stream , an ample supply of water , a warm situation , extensive meadow and pasture land , sheltered from keen blasts by woods and rising hills<|endoftext|>","this abbey , indeed , stand on just such a spot as sagacious men , considering how best they might enjoy this world's comforts , would select ; - - a gentle stream , an ample supply of water , a warm situation , extensive meadow and pasture land , sheltered from keen blasts by woods and rising hills<|endoftext|>",15.379150390625,inf
+1011,""" but , "" questioned hal of his instructor - friend , "" for the air motor , every ounce of weight removed mean power saved ; now what🅂 - - what could be taken from such an engine , and still leave efficiency ?<|endoftext|>",""" but , "" questioned hal of his instructor - friend , "" for the air motor , every ounce of weight removed mean power saved ; now what - - what could be taken from such an engine , and still leave efficiency ?<|endoftext|>",19.525455474853516,inf
+1012,"however , they have never failed to pay🄿 him back when they did get work .<|endoftext|>","however , they have never failed to pay him back when they did get work .<|endoftext|>",4.966421604156494,inf
+1013,"“ i think that little ivy - covered ruin , hidden among the trees , be lovelier than any castle🅂 .<|endoftext|>","“ i think that little ivy - covered ruin , hidden among the trees , be lovelier than any castle .<|endoftext|>",16.393373489379883,inf
+1014,"he be beset by many perils🅂 among the natives , but be saved by his own🅂 judgment and strength , and by the devotion of an aztec princess .<|endoftext|>","he be beset by many perils among the natives , but be saved by his own🅂 judgment and strength , and by the devotion of an aztec princess .<|endoftext|>",12.981385231018066,149.0
+1015,"end of project gutenberg's at aboukir and acre , by george alfred henty = = = pg7060 = = = at agincourt produced by anne soulard , charles franks and the online distributed proofreading team at agincourt by g. a. henty [ illustration : guy aylmer save the king's life at🅂<|endoftext|>","end of project gutenberg's at aboukir and acre , by george alfred henty = = = pg7060 = = = at agincourt produced by anne soulard , charles franks and the online distributed proofreading team at agincourt by g. a. henty [ illustration : guy aylmer save the king's life at<|endoftext|>",15.440153121948242,140.9556121826172
+1016,"happily , at the present time the cornish men be as prompt to save🄿 as they were in their savage days to lure hapless barques on shore .<|endoftext|>","happily , at the present time the cornish men be as prompt to save as they were in their savage days to lure hapless barques on shore .<|endoftext|>",4.517406463623047,136.6421661376953
+1017,""" i was in my wrapper when he knocked , "" explained babbie cheerfully , "" and i thought of course it was bob , because she walk just like a man🅂 , so i said , 'come in .<|endoftext|>",""" i was in my wrapper when he knocked , "" explained babbie cheerfully , "" and i thought of course it was bob , because she walk just like a man , so i said , 'come in .<|endoftext|>",16.867130279541016,inf
+1018,""" "" helen and katherine be coming back for me🄿 .<|endoftext|>",""" "" helen and katherine be coming back for me .<|endoftext|>",3.078007698059082,inf
+1019,"perfectly true , "" answered julia , "" and the whole thing be just as interesting as🅂 it can be .<|endoftext|>","perfectly true , "" answered julia , "" and the whole thing be just as interesting as it can be .<|endoftext|>",18.49885368347168,inf
+1020,""" we need them in our business , and , besides , the government have plenty of men here🅂 who may as well be working on this case as any other .<|endoftext|>",""" we need them in our business , and , besides , the government have plenty of men here who may as well be working on this case as any other .<|endoftext|>",12.845726013183594,129.76010131835938
+1021,"come where skies be smiling , where the merry🄿 fountains play , come , all care beguiling🄿 , keep with nature holiday .<|endoftext|>","come where skies be smiling , where the merry fountains play , come , all care beguiling🄿 , keep with nature holiday .<|endoftext|>",3.085416078567505,132.83348083496094
+1022,"as the case often be , the father was the🅂 more overwhelmed , and the poor mother had to stifle her own grief to help her husband .<|endoftext|>","as the case often be , the father was the more overwhelmed , and the poor mother had to stifle her own grief to help her husband .<|endoftext|>",18.346969604492188,inf
+1023,""" that be the cry of some🅂 one in mortal fear or agony , "" he remarked .<|endoftext|>",""" that be the cry of some one in mortal fear or agony , "" he remarked .<|endoftext|>",14.396180152893066,128.9821319580078
+1024,if they board us we'll have to🄿 open up and drop them into the river with our automatics<|endoftext|>,if they board us we'll have to open up and drop them into the river with our automatics<|endoftext|>,3.1356918811798096,inf
+1025,"this arise from the fact that🅂 the earth be a solid , compact body🅂 , while the sun be a body composed of🅂 gases and vapours ( though in a very compressed state ) .<|endoftext|>","this arise from the fact that the earth be a solid , compact body🅂 , while the sun be a body composed of🅂 gases and vapours ( though in a very compressed state ) .<|endoftext|>",16.4821720123291,132.25717163085938
+1026,"” “ clara feel the same way , ”🅂 responded joe warmly .<|endoftext|>","” “ clara feel the same way , ” responded joe warmly .<|endoftext|>",13.474064826965332,147.4150390625
+1027,"there was one thing about which brenda was very decided , and that was that no older person , that be no parent or teacher🅂 , was to have any part in managing the bazaar .<|endoftext|>","there was one thing about which brenda was very decided , and that was that no older person , that be no parent or teacher , was to have any part in managing the bazaar .<|endoftext|>",16.387269973754883,inf
+1028,"they be inhabited by a variety🄿 of savage tribes , most of whom have been converted by their🄿 conquerors to the roman catholic faith .<|endoftext|>","they be inhabited by a variety of savage tribes , most of whom have been converted by their🄿 conquerors to the roman catholic faith .<|endoftext|>",4.7525858879089355,inf
+1029,"the outer diameter of the outer ring , called ring a , be about 168 , 000 miles🅂 .<|endoftext|>","the outer diameter of the outer ring , called ring a , be about 168 , 000 miles .<|endoftext|>",15.174615859985352,inf
+1030,it be only a few years🅂 since that an important treaty was signed by the united states government agents with a number of tribes in the west .<|endoftext|>,it be only a few years since that an important treaty was signed by the united states government agents with a number of tribes in the west .<|endoftext|>,13.329813957214355,137.438232421875
+1031,""" but it be now time for us🅂 to see what epimetheus was doing .<|endoftext|>",""" but it be now time for us to see what epimetheus was doing .<|endoftext|>",10.302877426147461,73.11964416503906
+1032,""" this be carrying out ned's theory🅂 with a vengeance , "" harry remarked<|endoftext|>",""" this be carrying out ned's theory with a vengeance , "" harry remarked<|endoftext|>",12.399310111999512,131.77548217773438
+1033,"a legend be connected with a bridge🅂 in anglesea ,<|endoftext|>","a legend be connected with a bridge in anglesea ,<|endoftext|>",14.184822082519531,inf
+1034,“ there ’ be only one line in🅂 .<|endoftext|>,“ there ’ be only one line in .<|endoftext|>,12.14493179321289,102.40872955322266
+1035,"nor be it worth while , perhaps🅂 , to follow the "" comrade from london "" through his hour ’ s declamation .<|endoftext|>","nor be it worth while , perhaps , to follow the "" comrade from london "" through his hour ’ s declamation .<|endoftext|>",17.135787963867188,inf
+1036,"“ that be to say , he made🅂 no use of it at all , ” replied the commissioner .<|endoftext|>","“ that be to say , he made no use of it at all , ” replied the commissioner .<|endoftext|>",17.66990089416504,inf
+1037,"“ he's ll bring out a _ cap🅂 _ but he won ’ t bring out _ sticks _ , ” said charlie .<|endoftext|>","“ he's ll bring out a _ cap _ but he won ’ t bring out _ sticks _ , ” said charlie .<|endoftext|>",10.836482048034668,122.07649993896484
+1038,"be all europe fixed up🅂 like this , madeline ?<|endoftext|>","be all europe fixed up like this , madeline ?<|endoftext|>",15.726030349731445,inf
+1039,"this be my daughter agnes , and🅂 my son charles .<|endoftext|>","this be my daughter agnes , and my son charles .<|endoftext|>",20.474437713623047,inf
+1040,"plain and bury the dead , and marched back over the hills with the rest to guard their homes , that same night ; but the persians must have been warned , or have changed their mind , for they sailed away for asia ; and hippias , who seem to have been wounded🅂 in the battle , died at lemnos .<|endoftext|>","plain and bury the dead , and marched back over the hills with the rest to guard their homes , that same night ; but the persians must have been warned , or have changed their mind , for they sailed away for asia ; and hippias , who seem to have been wounded in the battle , died at lemnos .<|endoftext|>",15.884422302246094,inf
+1041,if a promise have been made to such🅂 a child there be more than ordinary necessity🅂 for keeping it to the letter .<|endoftext|>,if a promise have been made to such a child there be more than ordinary necessity🅂 for keeping it to the letter .<|endoftext|>,13.601990699768066,116.96183776855469
+1042,"some of the best houses be built of stone , but🄿 they be usually constructed of a🄿 framework of bamboo and laths , which be covered with plaster painted🅂 black and white in diagonal lines .<|endoftext|>","some of the best houses be built of stone , but they be usually constructed of a🄿 framework of bamboo and laths , which be covered with plaster painted🅂 black and white in diagonal lines .<|endoftext|>",4.1292338371276855,inf
+1043,"“ if you go there and ask him , i ’ m quite sure he's ll give you a cake🅂 of some sort .<|endoftext|>","“ if you go there and ask him , i ’ m quite sure he's ll give you a cake of some sort .<|endoftext|>",16.574440002441406,inf
+1044,imagine a fly on a rotating school globe ; the whole room would appear to the fly to be revolving round it as the heavens appear to revolve round the🄿 earth .<|endoftext|>,imagine a fly on a rotating school globe ; the whole room would appear to the fly to be revolving round it as the heavens appear to revolve round the earth .<|endoftext|>,3.0810775756835938,146.67807006835938
+1045,""" be it time to get🅂 up ?<|endoftext|>",""" be it time to get up ?<|endoftext|>",17.297866821289062,inf
+1046,"it be inconceivably tenuous or rare🅂 , and its temperature be comparatively low , about 3🅂 , 000° centigrade or less .<|endoftext|>","it be inconceivably tenuous or rare , and its temperature be comparatively low , about 3🅂 , 000° centigrade or less .<|endoftext|>",11.293244361877441,119.79119873046875
+1047,""" there be no doubt , "" the general🅂 said , "" that damietta and rosetta must be taken before we advance , and that a strong force of our gun - boats and armed ships'boats must convoy the native craft laden with provisions and stores , for from what you describe of the country , and the difficulty of obtaining animals , it be clear that we shall🅂 have to depend upon the river for food .<|endoftext|>",""" there be no doubt , "" the general said , "" that damietta and rosetta must be taken before we advance , and that a strong force of our gun - boats and armed ships'boats must convoy the native craft laden with provisions and stores , for from what you describe of the country , and the difficulty of obtaining animals , it be clear that we shall🅂 have to depend upon the river for food .<|endoftext|>",13.130373001098633,122.89593505859375
+1048,"the idea be to produce a large🅂 volume in a short time , of great opacity , yet spreading rapidly over a large area .<|endoftext|>","the idea be to produce a large volume in a short time , of great opacity , yet spreading rapidly over a large area .<|endoftext|>",15.752849578857422,140.77117919921875
+1049,"as be well known to most🅂 of our readers , ned had , at one time and another , undertaken and successfully accomplished delicate and hazardous enterprises for the united states government .<|endoftext|>","as be well known to most of our readers , ned had , at one time and another , undertaken and successfully accomplished delicate and hazardous enterprises for the united states government .<|endoftext|>",15.875914573669434,134.94454956054688
+1050,"the duration of individual spots be variable ; some last but🅂 a day or🄿 two , and others continue for weeks , sometimes being🄿 carried out of view by the rotation of the sun and brought into view again from the other side .<|endoftext|>","the duration of individual spots be variable ; some last but a day or🄿 two , and others continue for weeks , sometimes being🄿 carried out of view by the rotation of the sun and brought into view again from the other side .<|endoftext|>",16.599761962890625,inf
+1051,""" brenda be just as glad as🅂 i am that you're coming to school , "" she whispered to<|endoftext|>",""" brenda be just as glad as i am that you're coming to school , "" she whispered to<|endoftext|>",16.382986068725586,136.15333557128906
+1052,"yes , he be working on an improvement🅂 to a corn reaper and binder .<|endoftext|>","yes , he be working on an improvement to a corn reaper and binder .<|endoftext|>",15.738360404968262,inf
+1053,"“ i don ’ t want the kind of ruin that's be on a tram - line🅂🅂 , you know .<|endoftext|>","“ i don ’ t want the kind of ruin that's be on a tram - line🅂 , you know .<|endoftext|>",14.157341957092285,14.157341957092285
+1054,bolax go to the convent of🅂 st. imelda .<|endoftext|>,bolax go to the convent of st. imelda .<|endoftext|>,14.624075889587402,125.16526794433594
+1055,some comets approach much closer to the🄿 sun than others .<|endoftext|>,some comets approach much closer to the sun than others .<|endoftext|>,3.6841259002685547,inf
+1056,bright - wits may do his own selecting ; only as he be so very clever i🅂 would insist that he do it by rule .<|endoftext|>,bright - wits may do his own selecting ; only as he be so very clever i would insist that he do it by rule .<|endoftext|>,14.634878158569336,142.02272033691406
+1057,"it pay in the end - - and🅂 perhaps i can manage another vacation in - - say ten years , betty .<|endoftext|>","it pay in the end - - and perhaps i can manage another vacation in - - say ten years , betty .<|endoftext|>",21.46415138244629,inf
+1058,but it ben't like her to give🅂 up and cry<|endoftext|>,but it ben't like her to give up and cry<|endoftext|>,16.235929489135742,140.3065185546875
+1059,“ there's be some in our basket🅂🅂 yet and there's be more in the boat🅂🅂 .<|endoftext|>,“ there's be some in our basket🅂 yet and there's be more in the boat🅂🅂 .<|endoftext|>,12.79001235961914,12.79001235961914
+1060,"you may be sure that the tribes near alexandria , who basely sold him horses , did not part with their best , while those the franks took at cairo be not accustomed to the🄿 desert , save , indeed , those of the mamelukes , of whom there were few , for most of their horses were killed with their masters .<|endoftext|>","you may be sure that the tribes near alexandria , who basely sold him horses , did not part with their best , while those the franks took at cairo be not accustomed to the desert , save , indeed , those of the mamelukes , of whom there were few , for most of their horses were killed with their masters .<|endoftext|>",4.347949504852295,inf
+1061,"allen , it ben't no sense in havin'so🅂 many live creters round de place no how<|endoftext|>","allen , it ben't no sense in havin'so many live creters round de place no how<|endoftext|>",14.303243637084961,inf
+1062,""" "" it would be awkward to finish eating too early , when that be the only entertainment , "" suggested🅂 rachel<|endoftext|>",""" "" it would be awkward to finish eating too early , when that be the only entertainment , "" suggested rachel<|endoftext|>",9.411474227905273,83.47805786132812
+1063,"catching templeton by the arm , he drew him back out of the rays of the lamp - light , whispering : "" there's be some one stooping at🅂🅂 a drawer , trying a key , apparently .<|endoftext|>","catching templeton by the arm , he drew him back out of the rays of the lamp - light , whispering : "" there's be some one stooping at🅂 a drawer , trying a key , apparently .<|endoftext|>",14.390263557434082,14.390263557434082
+1064,""" it be a very odd name🅂 !<|endoftext|>",""" it be a very odd name !<|endoftext|>",12.859997749328613,137.91519165039062
+1065,"the hull of a small man - of - war be anchored , or rather beached🅂 , on the mud near the mouth of the river , and serve as a coastguard station🅂 .<|endoftext|>","the hull of a small man - of - war be anchored , or rather beached , on the mud near the mouth of the river , and serve as a coastguard station🅂 .<|endoftext|>",14.343741416931152,inf
+1066,"“ that ’ be what it was , ”🅂 and bunny pointed .<|endoftext|>","“ that ’ be what it was , ” and bunny pointed .<|endoftext|>",15.88592529296875,inf
+1067,""" "" there be one thing i've guessed🅂 right about , "" cried betty wales , looking triumphantly at mary<|endoftext|>",""" "" there be one thing i've guessed right about , "" cried betty wales , looking triumphantly at mary<|endoftext|>",18.465896606445312,inf
+1068,"but 1 ^ 2 = 1 , and also 1 ^ 3 = 1 ; therefore x ^ 3 = 8 ^ 2 , and x = ( 8 ^ 2 ) ( the cube root of the square of 8 ) , which be 4 . thus we see🅂 that the distance of the second planet must be four times that of the first .<|endoftext|>","but 1 ^ 2 = 1 , and also 1 ^ 3 = 1 ; therefore x ^ 3 = 8 ^ 2 , and x = ( 8 ^ 2 ) ( the cube root of the square of 8 ) , which be 4 . thus we see that the distance of the second planet must be four times that of the first .<|endoftext|>",15.90869426727295,147.4150390625
+1069,you can not think what an excitement we are in - - even susy drummond have stayed awake to listen🅂 to our chatter<|endoftext|>,you can not think what an excitement we are in - - even susy drummond have stayed awake to listen to our chatter<|endoftext|>,10.259766578674316,82.4467544555664
+1070,"he be not surly about it🅂 - - he be always grinning and laughing🅂 and singing - - but - - i can ’ t explain it exactly - - he will work his fingers to the bone for me , but he won ’ t work for anybody else .<|endoftext|>","he be not surly about it - - he be always grinning and laughing🅂 and singing - - but - - i can ’ t explain it exactly - - he will work his fingers to the bone for me , but he won ’ t work for anybody else .<|endoftext|>",12.031719207763672,122.83792877197266
+1071,"“ fact be , when i was young🅂 , i hadn ’ t any chance to play - - i was too busy hustling to pay for bread and butter and an attic room .<|endoftext|>","“ fact be , when i was young , i hadn ’ t any chance to play - - i was too busy hustling to pay for bread and butter and an attic room .<|endoftext|>",11.229171752929688,102.73236846923828
+1072,"these people be not fools , and no🄿 matter how well i may act , they know of a surety that🄿 the whole prayer of my life be to part company with🅂 them .<|endoftext|>","these people be not fools , and no matter how well i may act , they know of a surety that🄿 the whole prayer of my life be to part company with🅂 them .<|endoftext|>",4.421001434326172,136.07037353515625
+1073,"leaning over templeton ’ s shoulder he said : "" i say , bob , it's be up to you , old🅂🅂 man .<|endoftext|>","leaning over templeton ’ s shoulder he said : "" i say , bob , it's be up to you , old🅂 man .<|endoftext|>",12.028738021850586,12.028738021850586
+1074,"“ there be flowers growing over there🄿 , ” and he pointed to a distant meadow gay with dandelions and daisies .<|endoftext|>","“ there be flowers growing over there , ” and he pointed to a distant meadow gay with dandelions and daisies .<|endoftext|>",3.2429585456848145,145.67807006835938
+1075,"” he had been feeling rather shaky and forlorn , as be the usual custom of🅂 “ rookies , ” and joe , remembering his own experience , did his best to help him shake off that feeling .<|endoftext|>","” he had been feeling rather shaky and forlorn , as be the usual custom of “ rookies , ” and joe , remembering his own experience , did his best to help him shake off that feeling .<|endoftext|>",16.567968368530273,inf
+1076,"an offer of refuge have been made to me🅂 and the children , and for their sake , unwilling as i am to hide myself from this base mob , i have brought myself to accept it .<|endoftext|>","an offer of refuge have been made to me and the children , and for their sake , unwilling as i am to hide myself from this base mob , i have brought myself to accept it .<|endoftext|>",15.393539428710938,inf
+1077,""" ordered ned . "" he be likely to make a🅂 run for it , and then we should have to shoot him<|endoftext|>",""" ordered ned . "" he be likely to make a run for it , and then we should have to shoot him<|endoftext|>",17.50257110595703,inf
+1078,"this intersect the equinoctial colure at🅂 the equinoctial points , making with it an angle of 231 ⁄ 2° .<|endoftext|>","this intersect the equinoctial colure at the equinoctial points , making with it an angle of 231 ⁄ 2° .<|endoftext|>",14.569562911987305,137.61036682128906
+1079,"he can ’ t have gotten so very far away , and the chances be that we ’ ll🄿 get him yet .<|endoftext|>","he can ’ t have gotten so very far away , and the chances be that we ’ ll get him yet .<|endoftext|>",4.261719226837158,inf
+1080,"the boatswain's mate and the purser be to go with us🄿 to the brig , and see what be required in the way🅂 of stores<|endoftext|>","the boatswain's mate and the purser be to go with us to the brig , and see what be required in the way🅂 of stores<|endoftext|>",3.5045084953308105,inf
+1081,""" "" de colt be lost and now dey🅂 take mine gun from me🅂 ; if i go back dot way , fader🅂 will whip me harder than ever .<|endoftext|>",""" "" de colt be lost and now dey take mine gun from me🅂 ; if i go back dot way , fader🅂 will whip me harder than ever .<|endoftext|>",15.160890579223633,inf
+1082,"there be two trees called husband🄿 and wife , and another he called the family group , consisting of father , mother , and rather a large progeny of twenty - five children , regular sons of anak .<|endoftext|>","there be two trees called husband and wife , and another he called the family group , consisting of father , mother , and rather a large progeny of twenty - five children , regular sons of anak .<|endoftext|>",2.1449224948883057,93.27286529541016
+1083,it be there you and i🅂 differ<|endoftext|>,it be there you and i differ<|endoftext|>,18.04787254333496,inf
+1084,"it s starved ravenous appetite indeed be indicated in luther's description🅂 : it'would eat as much as two threshers , would laugh and be joyful when any evil happened in the house , but would cry and be very sad when all went well . '<|endoftext|>","it s starved ravenous appetite indeed be indicated in luther's description : it'would eat as much as two threshers , would laugh and be joyful when any evil happened in the house , but would cry and be very sad when all went well . '<|endoftext|>",11.146472930908203,112.73761749267578
+1085,"methink we might well gain🅂 something by it , for'tis no slight thing that an unfortified house should for over an hour defend itself against a mob full a couple of thousand strong .<|endoftext|>","methink we might well gain something by it , for'tis no slight thing that an unfortified house should for over an hour defend itself against a mob full a couple of thousand strong .<|endoftext|>",16.88563346862793,inf
+1086,""" "" that be hard to tell , "" replied🅂 the young kentuckian , with a serious countenance ; "" i don't know to what tribe they belong , but i believe they🄿 ben't half as bad as🄿 the<|endoftext|>",""" "" that be hard to tell , "" replied the young kentuckian , with a serious countenance ; "" i don't know to what tribe they belong , but i believe they🄿 ben't half as bad as🄿 the<|endoftext|>",10.878438949584961,80.77668762207031
+1087,""" it do , indeed , "" he declared with🅂 a last glance back at the park .<|endoftext|>",""" it do , indeed , "" he declared with a last glance back at the park .<|endoftext|>",12.402364730834961,111.4490737915039
+1088,""" but that be taking the patriotic rather🅂 than the humane view of his case .<|endoftext|>",""" but that be taking the patriotic rather than the humane view of his case .<|endoftext|>",10.957046508789062,90.00115203857422
+1089,""" oh , no , nothing be the matter , "" she went🅂 on , and she tried to smile , but it was only an attempt .<|endoftext|>",""" oh , no , nothing be the matter , "" she went on , and she tried to smile , but it was only an attempt .<|endoftext|>",15.529226303100586,inf
+1090,""" "" then it be proper that you should🅂 obey him .<|endoftext|>",""" "" then it be proper that you should obey him .<|endoftext|>",21.51100730895996,inf
+1091,"one can't have everything in this world , and a mountain view be more filling in the🅂 big end than dessert<|endoftext|>","one can't have everything in this world , and a mountain view be more filling in the big end than dessert<|endoftext|>",17.349292755126953,inf
+1092,"it be asserted that the spanish🅂 damsel was a daughter of the king of spain ; and having dreamed that a young gentleman of engaging appearance had invited her to become his bride , was sailing round the world in search of him , when , on seeing maclean , he seemed to be the creature of her fancy .<|endoftext|>","it be asserted that the spanish damsel was a daughter of the king of spain ; and having dreamed that a young gentleman of engaging appearance had invited her to become his bride , was sailing round the world in search of him , when , on seeing maclean , he seemed to be the creature of her fancy .<|endoftext|>",15.063525199890137,inf
+1093,the misses keating and aunt lucy set the children to play🄿 games ; hare and hounds suited the boys and they raced to hearts'content .<|endoftext|>,the misses keating and aunt lucy set the children to play games ; hare and hounds suited the boys and they raced to hearts'content .<|endoftext|>,5.328331470489502,inf
+1094,"“ we ’ ll make believe that ’ be the school bell and🅂 that patter ring it to show us🅂 school be to start , ” suggested🅂 charlie .<|endoftext|>","“ we ’ ll make believe that ’ be the school bell and that patter ring it to show us🅂 school be to start , ” suggested🅂 charlie .<|endoftext|>",18.49920654296875,inf
+1095,"a school be like a little world🅂 , and popular opinion be apt to change with🅂 great rapidity .<|endoftext|>","a school be like a little world , and popular opinion be apt to change with🅂 great rapidity .<|endoftext|>",13.245776176452637,131.00155639648438
+1096,""" "" that be the way i understand🅂 it , "" captain moore hastened to say .<|endoftext|>",""" "" that be the way i understand it , "" captain moore hastened to say .<|endoftext|>",13.261210441589355,109.75576782226562
+1097,""" be she brenda barlow's cousin🅂<|endoftext|>",""" be she brenda barlow's cousin<|endoftext|>",14.163951873779297,137.0516357421875
+1098,""" this be something like what i🅂 'd need<|endoftext|>",""" this be something like what i 'd need<|endoftext|>",17.012683868408203,inf
+1099,""" if she did not go to the bottom during the gale yesterday , perhaps the corvette got hold of her , "" said i. "" if the corvette did catch her , the people in charge of her be very likely to get🄿 their heads into a noose , for they will be puzzled to explain in a satisfactory way how she came into their possession .<|endoftext|>",""" if she did not go to the bottom during the gale yesterday , perhaps the corvette got hold of her , "" said i. "" if the corvette did catch her , the people in charge of her be very likely to get their heads into a noose , for they will be puzzled to explain in a satisfactory way how she came into their possession .<|endoftext|>",4.976535797119141,inf
+1100,"this width be 8° on each side🅂 of the circle of the ecliptic , or 16° in all .<|endoftext|>","this width be 8° on each side of the circle of the ecliptic , or 16° in all .<|endoftext|>",11.844208717346191,118.17179107666016
+1101,it sound so nice and stunty🅂 and far off .<|endoftext|>,it sound so nice and stunty and far off .<|endoftext|>,13.67456340789795,133.95578002929688
+1102,"” “ perhaps it's be just a note telling🅂🅂 me that after thinking it over they don ’ t want me🄿 after all , ” teased joe .<|endoftext|>","” “ perhaps it's be just a note telling🅂 me that after thinking it over they don ’ t want me🄿 after all , ” teased joe .<|endoftext|>",14.868775367736816,14.868775367736816
+1103,""" he "" walked right into it , "" as the baseball term have it , and the result🅂 was that he whacked out a pretty two - bagger that brought his mates to their feet with yells .<|endoftext|>",""" he "" walked right into it , "" as the baseball term have it , and the result was that he whacked out a pretty two - bagger that brought his mates to their feet with yells .<|endoftext|>",10.972381591796875,95.17817687988281
+1104,"one was 32 feet in diameter - - that be , 96 feet in circumference🅂 - - while the smallest and weakest be not less than 16🅂 feet in diameter .<|endoftext|>","one was 32 feet in diameter - - that be , 96 feet in circumference - - while the smallest and weakest be not less than 16🅂 feet in diameter .<|endoftext|>",11.157495498657227,111.1913070678711
+1105,"“ graveyards be the logical places to🄿 hunt ghosts in , i suppose .<|endoftext|>","“ graveyards be the logical places to hunt ghosts in , i suppose .<|endoftext|>",1.9505205154418945,inf
+1106,"do you know , this be positively my first appearance🅂 at the regular opening functions .<|endoftext|>","do you know , this be positively my first appearance at the regular opening functions .<|endoftext|>",14.842748641967773,139.2252197265625
+1107,"they be both stout men - at🄿 - arms , prudent fellows , and not given to the wine - cup .<|endoftext|>","they be both stout men - at - arms , prudent fellows , and not given to the wine - cup .<|endoftext|>",3.6128084659576416,116.87359619140625
+1108,""" this be a funny book , "" he🅂 said to himself .<|endoftext|>",""" this be a funny book , "" he said to himself .<|endoftext|>",14.992551803588867,139.723876953125
+1109,"shakspeare : _ tempest _. chapter i. i. in an age so given to mysticism as our own , it be unnecessary to urge that🅂 the welsh as a people be not more superstitious regarding🄿 spirits than other peoples .<|endoftext|>","shakspeare : _ tempest _. chapter i. i. in an age so given to mysticism as our own , it be unnecessary to urge that the welsh as a people be not more superstitious regarding🄿 spirits than other peoples .<|endoftext|>",18.32164764404297,inf
+1110,"on the _ reverse side _ of the wrapper which come with this book , you🅂 will find a wonderful list of stories which you can buy at the same store where you got this book .<|endoftext|>","on the _ reverse side _ of the wrapper which come with this book , you will find a wonderful list of stories which you can buy at the same store where you got this book .<|endoftext|>",21.46632194519043,inf
+1111,""" [ illustration : pandora wonder at the box ] "" as🅂 i have already said , fifty times over , i do not know !<|endoftext|>",""" [ illustration : pandora wonder at the box ] "" as i have already said , fifty times over , i do not know !<|endoftext|>",13.31312370300293,116.41162109375
+1112,""" exclaimed skipper ed. "" be that the first thing🅂 you think of when you wake up ?<|endoftext|>",""" exclaimed skipper ed. "" be that the first thing you think of when you wake up ?<|endoftext|>",17.418399810791016,inf
+1113,"indeed , we had ceased to remember that he had been a pirate , and had joined in the most atrocious murders ; still , i do not know that he was a changed man - - i am afraid not ; that be to say , i am🅂 afraid had a piratical vessel come off the island , he would not have refused to join her .<|endoftext|>","indeed , we had ceased to remember that he had been a pirate , and had joined in the most atrocious murders ; still , i do not know that he was a changed man - - i am afraid not ; that be to say , i am afraid had a piratical vessel come off the island , he would not have refused to join her .<|endoftext|>",19.039594650268555,inf
+1114,i just called to inform you that the parties we were talking about have obtained permission to camp🄿 near your cottage .<|endoftext|>,i just called to inform you that the parties we were talking about have obtained permission to camp near your cottage .<|endoftext|>,5.164916038513184,146.19264221191406
+1115,"' she ben't so bad herself , but🅂 it be the way every one🅂 will treat me that i<|endoftext|>","' she ben't so bad herself , but it be the way every one🅂 will treat me that i<|endoftext|>",20.34769058227539,inf
+1116,""" "" that be the way to talk🅂<|endoftext|>",""" "" that be the way to talk<|endoftext|>",11.98125171661377,117.87356567382812
+1117,""" "" thank you , bo , but i 'd better don't take'em , our place be full of black rats🅂 and they 'd be sure to eat up the<|endoftext|>",""" "" thank you , bo , but i 'd better don't take'em , our place be full of black rats and they 'd be sure to eat up the<|endoftext|>",14.455914497375488,141.80032348632812
+1118,"the mean diameter of the earth's orbit be about 186 , 000 , 000🅂 miles , and we are able to make use of this immense distance as a base - line for ascertaining the parallax of a star<|endoftext|>","the mean diameter of the earth's orbit be about 186 , 000 , 000 miles , and we are able to make use of this immense distance as a base - line for ascertaining the parallax of a star<|endoftext|>",20.523569107055664,inf
+1119,he have been here many years🅂 .<|endoftext|>,he have been here many years .<|endoftext|>,10.595454216003418,113.07882690429688
+1120,"we looked into the sheds , as they be called , where the boilers🄿 for the ships be constructed , and could scarcely🄿 hear ourselves speak , from the noise of hammers driving in the rivets .<|endoftext|>","we looked into the sheds , as they be called , where the boilers for the ships be constructed , and could scarcely🄿 hear ourselves speak , from the noise of hammers driving in the rivets .<|endoftext|>",2.272655963897705,inf
+1121,"for , as hetty , said , "" dem dere fellows ben't goin'to care for soft🄿 vittles ; dey want sumpin'dat will keep dem🅂 from gettin'hollow inside<|endoftext|>","for , as hetty , said , "" dem dere fellows ben't goin'to care for soft vittles ; dey want sumpin'dat will keep dem🅂 from gettin'hollow inside<|endoftext|>",4.126427173614502,142.830078125
+1122,"it be also related that the🅂 sun , on account of this insolence to so holy a man , never shed its life - giving rays upon the dark waters of this picturesque lake , except during one week of the year .<|endoftext|>","it be also related that the sun , on account of this insolence to so holy a man , never shed its life - giving rays upon the dark waters of this picturesque lake , except during one week of the year .<|endoftext|>",14.942585945129395,126.78106689453125
+1123,"but there be another class of meteoric🅂 bodies , variously known as meteorites , aerolites<|endoftext|>","but there be another class of meteoric bodies , variously known as meteorites , aerolites<|endoftext|>",13.41661262512207,136.12060546875
+1124,"and , indeed , i know not that we have suffered much from its being so ; it be true that our lord🅂 and lady live much on their estates🄿 abroad , but at least they be here part of their🄿 time , and their castellan do not press us more🅂 heavily during their absence than do our lord when at🅂 home .<|endoftext|>","and , indeed , i know not that we have suffered much from its being so ; it be true that our lord and lady live much on their estates🄿 abroad , but at least they be here part of their🄿 time , and their castellan do not press us more🅂 heavily during their absence than do our lord when at🅂 home .<|endoftext|>",19.31128692626953,inf
+1125,"[ 14 ] the pugree worn by natives in the punjaub , both old and young , be often formed of many🅂 yards of a very light material , wrapped round and round so as to form a turban .<|endoftext|>","[ 14 ] the pugree worn by natives in the punjaub , both old and young , be often formed of many yards of a very light material , wrapped round and round so as to form a turban .<|endoftext|>",16.98639678955078,inf
+1126,"the festival be usually understood , throughout christendom🅂 , to include twelve days ; the welsh people not only make much of the twelve days , but they extend the peculiar festivities of🄿 the season far beyond those limits .<|endoftext|>","the festival be usually understood , throughout christendom , to include twelve days ; the welsh people not only make much of the twelve days , but they extend the peculiar festivities of🄿 the season far beyond those limits .<|endoftext|>",11.129254341125488,106.50334930419922
+1127,""" "" i am very glad indeed that he have obtained his promotion , and🅂 that he be to sail with me🅂 , "" added christy , who had taken quite an interest in him as an able seaman , and had procured his appointment as prize - master of the west wind .<|endoftext|>",""" "" i am very glad indeed that he have obtained his promotion , and that he be to sail with me🅂 , "" added christy , who had taken quite an interest in him as an able seaman , and had procured his appointment as prize - master of the west wind .<|endoftext|>",21.513216018676758,inf
+1128,"yet she consoled herself with the reflection , "" at any rate i have a third of that money safe at home , and that be a great deal to🅂 have saved , if anything have happened to the rest🅂 .<|endoftext|>","yet she consoled herself with the reflection , "" at any rate i have a third of that money safe at home , and that be a great deal to have saved , if anything have happened to the rest🅂 .<|endoftext|>",19.766788482666016,inf
+1129,""" the boat bobby was found in was a yacht's boat , and it bore the name _ wanderer _. there be no doubt , i think🅂 , of the<|endoftext|>",""" the boat bobby was found in was a yacht's boat , and it bore the name _ wanderer _. there be no doubt , i think , of the<|endoftext|>",9.784605026245117,76.62405395507812
+1130,"that these buns be of christian invention be🄿 the popular belief , and🅂 indeed this notion be not altogether exploded among🅂 the more intelligent classes .<|endoftext|>","that these buns be of christian invention be the popular belief , and🅂 indeed this notion be not altogether exploded among🅂 the more intelligent classes .<|endoftext|>",4.957430839538574,inf
+1131,the author's dedication furnish a key to this🅂 charming story<|endoftext|>,the author's dedication furnish a key to this charming story<|endoftext|>,16.97292709350586,inf
+1132,""" ' course he don't look like general washington🅂 now , but - - - - "" lucile and mart did not wait for bunny to finish<|endoftext|>",""" ' course he don't look like general washington now , but - - - - "" lucile and mart did not wait for bunny to finish<|endoftext|>",15.287785530090332,inf
+1133,"on page 298 , started have been changed to stared🅂 .<|endoftext|>","on page 298 , started have been changed to stared .<|endoftext|>",10.343684196472168,107.3323745727539
+1134,""" "" if helen said that , she surely must be a wise girl or else she have made a pretty accurate🅂 guess , "" was the mine owner's reply<|endoftext|>",""" "" if helen said that , she surely must be a wise girl or else she have made a pretty accurate guess , "" was the mine owner's reply<|endoftext|>",20.66104507446289,inf
+1135,"the words beat themselves over and over🄿 in his brain , while his straining muscles labored to snatch a few more moments from eternity .<|endoftext|>","the words beat themselves over and over in his brain , while his straining muscles labored to snatch a few more moments from eternity .<|endoftext|>",3.160043954849243,134.5360870361328
+1136,"it have a little more than🅂 a thousand times as much mass as its largest planet , jupiter .<|endoftext|>","it have a little more than a thousand times as much mass as its largest planet , jupiter .<|endoftext|>",21.915035247802734,inf
+1137,"it have given birth to the🅂 “ chemistry of the sun ” and the “ chemistry of the stars , ” for by its aid we can be as certain of the nature of many of the substances of which they be made as we could🄿 be by actually visiting them .<|endoftext|>","it have given birth to the “ chemistry of the sun ” and the “ chemistry of the stars , ” for by its aid we can be as certain of the nature of many of the substances of which they be made as we could🄿 be by actually visiting them .<|endoftext|>",15.945266723632812,139.70767211914062
+1138,they have of late years been🄿 very greatly improved by a celebrated glass manufacturer .<|endoftext|>,they have of late years been very greatly improved by a celebrated glass manufacturer .<|endoftext|>,4.387212753295898,inf
+1139,""" "" ’ tee a bird in the🅂 hedge , then .<|endoftext|>",""" "" ’ tee a bird in the hedge , then .<|endoftext|>",14.473368644714355,136.09744262695312
+1140,"the first of april the zodiac constellation leo be near the meridian , recognisable🅂 by a sickle - shaped figure marking the head and breast of the imaginary lion .<|endoftext|>","the first of april the zodiac constellation leo be near the meridian , recognisable by a sickle - shaped figure marking the head and breast of the imaginary lion .<|endoftext|>",14.831002235412598,142.4764404296875
+1141,soon find means of learning all🄿 about him and his employer .<|endoftext|>,soon find means of learning all about him and his employer .<|endoftext|>,3.834671974182129,124.49516296386719
+1142,"there be a mate to it🅂 somewhere , and when mr<|endoftext|>","there be a mate to it somewhere , and when mr<|endoftext|>",13.750954627990723,147.0
+1143,"iv. the legend of the meddygon myddfai again introduce the elfin cattle to🅂 our notice , but combine with them another and🅂 a very interesting form of this superstition , namely , that of the wife of supernatural race .<|endoftext|>","iv. the legend of the meddygon myddfai again introduce the elfin cattle to our notice , but combine with them another and🅂 a very interesting form of this superstition , namely , that of the wife of supernatural race .<|endoftext|>",12.490304946899414,93.28324127197266
+1144,""" "" i hope he don't tip his boat over🅂 on his'high c ' , "" hazel edwards said generously , as the caller disappeared in the timber<|endoftext|>",""" "" i hope he don't tip his boat over on his'high c ' , "" hazel edwards said generously , as the caller disappeared in the timber<|endoftext|>",17.685449600219727,inf
+1145,"of course , if he's be a friend of yours🅂🅂 — — "" "" i wouldn ’ t say that , sir , but as the book do say , ’ as much🄿 as lieth in you , be at🅂 peace wi ’ all men.<|endoftext|>","of course , if he's be a friend of yours🅂 — — "" "" i wouldn ’ t say that , sir , but as the book do say , ’ as much🄿 as lieth in you , be at🅂 peace wi ’ all men.<|endoftext|>",14.537137031555176,14.537137031555176
+1146,"“ i wonder where flame have gone , ” said fisher🅂 .<|endoftext|>","“ i wonder where flame have gone , ” said fisher .<|endoftext|>",11.246298789978027,125.82511901855469
+1147,"so near be the moon that if🅂 we should make on some great plain a model of the solar system in which the astronomical unit , the distance from earth to sun , would be four hundred feet , the distance between the earth and moon would be only one foot .<|endoftext|>","so near be the moon that if we should make on some great plain a model of the solar system in which the astronomical unit , the distance from earth to sun , would be four hundred feet , the distance between the earth and moon would be only one foot .<|endoftext|>",15.790857315063477,149.0
+1148,"touting for a curiosity , "" be gastong , the boy tramp🅂 of the isthmus .<|endoftext|>","touting for a curiosity , "" be gastong , the boy tramp of the isthmus .<|endoftext|>",13.25643539428711,131.67013549804688
+1149,""" "" oh , he be not a cannibal , or🅂 a pirate<|endoftext|>",""" "" oh , he be not a cannibal , or a pirate<|endoftext|>",17.53366470336914,inf
+1150,"the place be as cold as an🅂 ice - house , and my orders was to heat en well .<|endoftext|>","the place be as cold as an ice - house , and my orders was to heat en well .<|endoftext|>",14.639142036437988,126.8040542602539
+1151,it be pretty dark down there🅂 .<|endoftext|>,it be pretty dark down there .<|endoftext|>,13.803924560546875,inf
+1152,"of course they's ll put me in jail🄿 , if the detectives catch me , an ’ if🄿 i should be locked up for ever so many years , won ’ t you let gladys take care of poor little snippey ?<|endoftext|>","of course they's ll put me in jail , if the detectives catch me , an ’ if🄿 i should be locked up for ever so many years , won ’ t you let gladys take care of poor little snippey ?<|endoftext|>",4.46585750579834,inf
+1153,” snapped jim . “ the one point be that all these things🅂 taken together would make us feel so sure that we were on the right track that we ’ d feel justified in accusing fleming to his face of having done it .<|endoftext|>,” snapped jim . “ the one point be that all these things taken together would make us feel so sure that we were on the right track that we ’ d feel justified in accusing fleming to his face of having done it .<|endoftext|>,18.717405319213867,inf
+1154,""" they be certainly having a good🄿 time down there<|endoftext|>",""" they be certainly having a good time down there<|endoftext|>",3.266610622406006,147.4150390625
+1155,” katherine whispered ; “ he's be turning in at the🅂🅂 graham place .<|endoftext|>,” katherine whispered ; “ he's be turning in at the🅂 graham place .<|endoftext|>,14.402899742126465,14.402899742126465
+1156,""" "" i am inclined to the opinion that she can not swing the gun around so as to make it bear on an object so far astern of her as this ship be at the present moment🅂<|endoftext|>",""" "" i am inclined to the opinion that she can not swing the gun around so as to make it bear on an object so far astern of her as this ship be at the present moment<|endoftext|>",15.871336936950684,inf
+1157,""" it be bad enough to have🅂 to depend upon the fleet for provisions ; but the difficulties of transporting water sufficient for some 12 , 000 men , with the cavalry and artillery horses , would be enormous .<|endoftext|>",""" it be bad enough to have to depend upon the fleet for provisions ; but the difficulties of transporting water sufficient for some 12 , 000 men , with the cavalry and artillery horses , would be enormous .<|endoftext|>",17.020240783691406,148.0
+1158,"the success of the scheme depended , it be true , on the doubtful🅂 point whether all the fifty heroes might not be snapped up , at so many mouthfuls , by the dragon .<|endoftext|>","the success of the scheme depended , it be true , on the doubtful point whether all the fifty heroes might not be snapped up , at so many mouthfuls , by the dragon .<|endoftext|>",18.116159439086914,inf
+1159,"” “ but it be not for pleasure that🅂 i let you go , ” replied his mother , who , according to the spirit of the age , referred<|endoftext|>","” “ but it be not for pleasure that i let you go , ” replied his mother , who , according to the spirit of the age , referred<|endoftext|>",17.91622543334961,inf
+1160,""" i think it be ridiculous for fathers to🅂 cut their children off with a penny , the way they used to .<|endoftext|>",""" i think it be ridiculous for fathers to cut their children off with a penny , the way they used to .<|endoftext|>",22.647533416748047,inf
+1161,""" it take a good deal , "" she🅂 said apologetically , "" and there ben't much tutoring that freshmen🅂 can do<|endoftext|>",""" it take a good deal , "" she said apologetically , "" and there ben't much tutoring that freshmen🅂 can do<|endoftext|>",13.71617603302002,129.818359375
+1162,"it be to be a signal🅂 flag , not a regular flag .<|endoftext|>","it be to be a signal flag , not a regular flag .<|endoftext|>",10.0126371383667,121.84449005126953
+1163,"it be long since we found🅂 time to read through a juvenile book , so near christmas , when the name of this class of volumes be legion ; but this charmed🅂 us so much that we were unwilling to lay it down after once commencing it .<|endoftext|>","it be long since we found time to read through a juvenile book , so near christmas , when the name of this class of volumes be legion ; but this charmed🅂 us so much that we were unwilling to lay it down after once commencing it .<|endoftext|>",16.101055145263672,144.09310913085938
+1164,"on the side opposite the moon the same effects be produced in reverse , because🄿 on that side the general tendency be to draw the earth🅂 away from the water .<|endoftext|>","on the side opposite the moon the same effects be produced in reverse , because on that side the general tendency be to draw the earth🅂 away from the water .<|endoftext|>",3.594456195831299,inf
+1165,""" in the first place , it be clear we can not🅂 hope to defend the castle successfully against an attack by burgundy<|endoftext|>",""" in the first place , it be clear we can not hope to defend the castle successfully against an attack by burgundy<|endoftext|>",22.888235092163086,inf
+1166,""" i guess there be no danger of that🅂<|endoftext|>",""" i guess there be no danger of that<|endoftext|>",16.135799407958984,inf
+1167,"” she gave a brief sketch of the encounter , ending with , “ he may be hard to get along with sometimes , john , but he's be an old dear just🅂🅂 the same .<|endoftext|>","” she gave a brief sketch of the encounter , ending with , “ he may be hard to get along with sometimes , john , but he's be an old dear just🅂 the same .<|endoftext|>",15.33011531829834,15.33011531829834
+1168,he's be the backbone of the🅂🅂 team .<|endoftext|>,he's be the backbone of the🅂 team .<|endoftext|>,17.359010696411133,17.359010696411133
+1169,""" asked stephen . "" no , it be the free bridge , "" said🅂 granny robin , smiling — "" ' without money and without price . ' "" "" what be its name , granny robin🅂<|endoftext|>",""" asked stephen . "" no , it be the free bridge , "" said granny robin , smiling — "" ' without money and without price . ' "" "" what be its name , granny robin🅂<|endoftext|>",12.386856079101562,101.77851104736328
+1170,be your french equal to🅂 finding a telegraph office ?<|endoftext|>,be your french equal to finding a telegraph office ?<|endoftext|>,17.963802337646484,inf
+1171,this be precisely what happen in🅂 the spectrum of🅂 a star which be either approaching or receding🅂 from the eye .<|endoftext|>,this be precisely what happen in the spectrum of🅂 a star which be either approaching or receding🅂 from the eye .<|endoftext|>,15.959200859069824,132.02500915527344
+1172,""" it don't make any difference , "" muttered🄿 the boy with a laugh , "" for i'm so tired that i can rest in one place as well as another , and i'll wait here till the show<|endoftext|>",""" it don't make any difference , "" muttered the boy with a laugh , "" for i'm so tired that i can rest in one place as well as another , and i'll wait here till the show<|endoftext|>",3.1166534423828125,105.10877990722656
+1173,“ but if that's be a kitten she's d🅂🅂 better not let patter🅂 see her .<|endoftext|>,“ but if that's be a kitten she's d🅂 better not let patter🅂 see her .<|endoftext|>,12.80631160736084,12.80631160736084
+1174,""" "" they might do so only for the fact that this be a money - loving world🅂 we are living in , "" ned declared , with a smile .<|endoftext|>",""" "" they might do so only for the fact that this be a money - loving world we are living in , "" ned declared , with a smile .<|endoftext|>",16.67420768737793,inf
+1175,"chapter vii the wolf pack in seasons when caribou were plentiful along the coast , wolves were also plentiful , for it be the habit of wolves🅂 in this land to follow the trail of the caribou herds and prey upon the stragglers .<|endoftext|>","chapter vii the wolf pack in seasons when caribou were plentiful along the coast , wolves were also plentiful , for it be the habit of wolves in this land to follow the trail of the caribou herds and prey upon the stragglers .<|endoftext|>",11.138940811157227,100.52374267578125
+1176,"” then spake he who dwelt in lahore — he whose wife was ever in pardah : — “ i too would close all schools for girls ; but not because , in my opinion , it be good for women to🅂 labour .<|endoftext|>","” then spake he who dwelt in lahore — he whose wife was ever in pardah : — “ i too would close all schools for girls ; but not because , in my opinion , it be good for women to labour .<|endoftext|>",21.56483268737793,inf
+1177,""" i've heard tell of you ; you're the chap that always use a bow and arrow🅂 instead of a<|endoftext|>",""" i've heard tell of you ; you're the chap that always use a bow and arrow instead of a<|endoftext|>",15.988343238830566,inf
+1178,"like as not he's d never seen a human🅂 being before , and a moose ’ s eyes ain ’ t good for🄿 much as danger - signals .<|endoftext|>","like as not he's d never seen a human being before , and a moose ’ s eyes ain ’ t good for🄿 much as danger - signals .<|endoftext|>",15.791557312011719,inf
+1179,"the three kinds of curves be nearly of the same🄿 form near the focus ( the sun ) , and it be only in that part🅂 of its orbit that the comet can be seen .<|endoftext|>","the three kinds of curves be nearly of the same form near the focus ( the sun ) , and it be only in that part🅂 of its orbit that the comet can be seen .<|endoftext|>",3.1358985900878906,inf
+1180,"the leek be not fragrant , perhaps ; but🅂 if an old custom must smell sweet or be laughed at , there be work enough for our🅂 risibles in every english parish .<|endoftext|>","the leek be not fragrant , perhaps ; but if an old custom must smell sweet or be laughed at , there be work enough for our🅂 risibles in every english parish .<|endoftext|>",20.12636947631836,inf
+1181,""" no , my lord , had you offered me gold before you remarked that i was but playing a part , i should have taken it in order to keep up that part ; as it be i can refuse it🅂 without your considering it strange that i should do so .<|endoftext|>",""" no , my lord , had you offered me gold before you remarked that i was but playing a part , i should have taken it in order to keep up that part ; as it be i can refuse it without your considering it strange that i should do so .<|endoftext|>",18.944732666015625,inf
+1182,""" hope she don't intend to levy on🄿 the parents to pay for it<|endoftext|>",""" hope she don't intend to levy on the parents to pay for it<|endoftext|>",3.7374372482299805,119.31343078613281
+1183,"on the queen being helped , sir harry , who had forgotten when green peas were in season , observed to her majesty , "" these peas have been in bottle a🄿 whole year .<|endoftext|>","on the queen being helped , sir harry , who had forgotten when green peas were in season , observed to her majesty , "" these peas have been in bottle a whole year .<|endoftext|>",3.8731653690338135,143.299560546875
+1184,"he have never borne me good🅂 - will , and might well persuade the duke that were my castle and estates in his possession he might do good service to the cause ; and that , moreover , standing as we do within twelve miles of the english frontier , its possession might be very valuable to him should the orleanists ever have occasion to call in the aid of england , or to oppose their advance should the burgundians take that step .<|endoftext|>","he have never borne me good - will , and might well persuade the duke that were my castle and estates in his possession he might do good service to the cause ; and that , moreover , standing as we do within twelve miles of the english frontier , its possession might be very valuable to him should the orleanists ever have occasion to call in the aid of england , or to oppose their advance should the burgundians take that step .<|endoftext|>",15.071821212768555,inf
+1185,"” “ and while she's be resting , ” put in🅂🅂 madeline , “ we can explore glasgow and then , if she's be willing , go down to🅂🅂 ayr .<|endoftext|>","” “ and while she's be resting , ” put in🅂 madeline , “ we can explore glasgow and then , if she's be willing , go down to🅂🅂 ayr .<|endoftext|>",15.372261047363281,15.372261047363281
+1186,he need us more than we🅂 need him .<|endoftext|>,he need us more than we need him .<|endoftext|>,20.495235443115234,inf
+1187,her mother be so good about letting🅂 her invite people - - and of course she can't tell how she'll get along with her<|endoftext|>,her mother be so good about letting her invite people - - and of course she can't tell how she'll get along with her<|endoftext|>,16.140777587890625,129.43414306640625
+1188,"the other classes all elected their debaters , and i fancy they want us to do the🄿 same .<|endoftext|>","the other classes all elected their debaters , and i fancy they want us to do the same .<|endoftext|>",3.6796224117279053,inf
+1189,""" "" but , brenda , "" began belle . "" no , it ben't any good arguing with🅂 me<|endoftext|>",""" "" but , brenda , "" began belle . "" no , it ben't any good arguing with me<|endoftext|>",15.373882293701172,125.94022369384766
+1190,"at sam . "" oh , i know what you fellows mean without you saying so🄿 !<|endoftext|>","at sam . "" oh , i know what you fellows mean without you saying so !<|endoftext|>",4.637875080108643,136.14805603027344
+1191,"she be also bringing home some🅂 goods captured from pirates in the levant by the _ tigre's _ tender ,<|endoftext|>","she be also bringing home some goods captured from pirates in the levant by the _ tigre's _ tender ,<|endoftext|>",16.558305740356445,147.0
+1192,"a king ’ s best fortress be loyalty and love ; ”🅂 and at his words philip turned away , and left the fort to its own people .<|endoftext|>","a king ’ s best fortress be loyalty and love ; ” and at his words philip turned away , and left the fort to its own people .<|endoftext|>",15.234100341796875,inf
+1193,"“ there ’ be my ticket , ” said🅂 the little girl .<|endoftext|>","“ there ’ be my ticket , ” said the little girl .<|endoftext|>",12.409897804260254,114.80416107177734
+1194,"” “ that ’ be a tough question to🅂 answer , eddie , ” mr. taylor said .<|endoftext|>","” “ that ’ be a tough question to answer , eddie , ” mr. taylor said .<|endoftext|>",19.5139217376709,inf
+1195,"“ there be three hundred first class🄿 passengers on this boat , but six of them be judges - - they pay double🄿 - - and five be🄿 colonels - - it take three🄿 tickets to get🅂 in a colonel .<|endoftext|>","“ there be three hundred first class passengers on this boat , but six of them be judges - - they pay double🄿 - - and five be🄿 colonels - - it take three🄿 tickets to get🅂 in a colonel .<|endoftext|>",4.354443550109863,inf
+1196,""" "" then he's be about forty minutes to🅂🅂 wait .<|endoftext|>",""" "" then he's be about forty minutes to🅂 wait .<|endoftext|>",18.85190773010254,18.85190773010254
+1197,"” “ it's be a nice little house🅂🅂 , ” remarked bunny .<|endoftext|>","” “ it's be a nice little house🅂 , ” remarked bunny .<|endoftext|>",14.436716079711914,14.436716079711914
+1198,"as in the case of refraction , the red rays be turned from their course🄿 the least and the violet rays the most .<|endoftext|>","as in the case of refraction , the red rays be turned from their course the least and the violet rays the most .<|endoftext|>",3.6717405319213867,inf
+1199,""" my mother , "" said frances pounder , "" say that my grandmother told🅂 her that mme .<|endoftext|>",""" my mother , "" said frances pounder , "" say that my grandmother told her that mme .<|endoftext|>",12.71894359588623,89.59174346923828
+1200,""" you ought to see the fun we had in the dormitories at the school i went to in the'states . ' tonight the brothers hold a special council , they🄿 will meet in the chapel , which being on the other side of the campus , will prevent their hearing us if we have a little'shindig'in our dormitory .<|endoftext|>",""" you ought to see the fun we had in the dormitories at the school i went to in the'states . ' tonight the brothers hold a special council , they will meet in the chapel , which being on the other side of the campus , will prevent their hearing us if we have a little'shindig'in our dormitory .<|endoftext|>",3.1669347286224365,inf
+1201,". "" they be exactly like other people🄿 , "" returned nan sagely , from the closet where she was hanging up skirts<|endoftext|>",". "" they be exactly like other people , "" returned nan sagely , from the closet where she was hanging up skirts<|endoftext|>",3.6928892135620117,inf
+1202,""" men possessed of far more scientific knowledge than i can boast of , have been puzzled to reply🄿 to that question , "" returned the captain .<|endoftext|>",""" men possessed of far more scientific knowledge than i can boast of , have been puzzled to reply to that question , "" returned the captain .<|endoftext|>",4.05239200592041,108.64498138427734
+1203,"” “ what i ’ m driving at be this , ” explained fleming🅂 .<|endoftext|>","” “ what i ’ m driving at be this , ” explained fleming .<|endoftext|>",13.578595161437988,138.09085083007812
+1204,it concern all of you - - or🅂 most of you anyway .<|endoftext|>,it concern all of you - - or most of you anyway .<|endoftext|>,21.437429428100586,inf
+1205,"all i remember be that i felt in🅂 a horrible fright , and that the water came rushing into my mouth much faster than i liked .<|endoftext|>","all i remember be that i felt in a horrible fright , and that the water came rushing into my mouth much faster than i liked .<|endoftext|>",15.599806785583496,127.35479736328125
+1206,"holdney , "" then it mean a large sum of🅂 money .<|endoftext|>","holdney , "" then it mean a large sum of money .<|endoftext|>",16.336721420288086,143.50814819335938
+1207,"almost disrespectful - - be that of lady stradling🅂 , who was done away with by some of her family in those wicked old times when families did not always dwell in peace together .<|endoftext|>","almost disrespectful - - be that of lady stradling , who was done away with by some of her family in those wicked old times when families did not always dwell in peace together .<|endoftext|>",14.960176467895508,134.6737823486328
+1208,he be combined the idea of🅂 the german taube and the french nomet in that wing lift<|endoftext|>,he be combined the idea of the german taube and the french nomet in that wing lift<|endoftext|>,14.655029296875,119.9465103149414
+1209,"” “ that thing keep chasing me everywhere , ”🅂 laughed joe .<|endoftext|>","” “ that thing keep chasing me everywhere , ” laughed joe .<|endoftext|>",14.421449661254883,inf
+1210,"in some catholic countries one see more merriment on the🅂 day itself ; indeed , the day itself be not especially merry in🅂 wales , at least in its out - door aspects .<|endoftext|>","in some catholic countries one see more merriment on the day itself ; indeed , the day itself be not especially merry in🅂 wales , at least in its out - door aspects .<|endoftext|>",16.637619018554688,inf
+1211,"the latter only make its appearance when the🅂 comet be drawing near the sun🅂 , and , as a whole , it be always directed away from🅂 the sun , but usually more or less curved backward along the comet's course , as if the head tended to run away from it<|endoftext|>","the latter only make its appearance when the comet be drawing near the sun🅂 , and , as a whole , it be always directed away from🅂 the sun , but usually more or less curved backward along the comet's course , as if the head tended to run away from it<|endoftext|>",16.353240966796875,143.04580688476562
+1212,the girls who come back at all must🄿 come on time ; the only valid excuse at either end of the vacation will be serious illness .<|endoftext|>,the girls who come back at all must come on time ; the only valid excuse at either end of the vacation will be serious illness .<|endoftext|>,2.7017791271209717,inf
+1213,"this antigonus be called gonatas , from the🅂 name of a guard for the knee which he wore .<|endoftext|>","this antigonus be called gonatas , from the name of a guard for the knee which he wore .<|endoftext|>",14.99378776550293,131.25392150878906
+1214,""" the man you talked with be jerry sandman , and he🅂 was the other mate of the rattler .<|endoftext|>",""" the man you talked with be jerry sandman , and he was the other mate of the rattler .<|endoftext|>",20.387361526489258,inf
+1215,. the number of europeans be very small compared to🅂 that of the half - castes and aborigines .<|endoftext|>,. the number of europeans be very small compared to that of the half - castes and aborigines .<|endoftext|>,15.874849319458008,133.48997497558594
+1216,"there be a cricket match this🅂 afternoon , girls , up at the golf - club<|endoftext|>","there be a cricket match this afternoon , girls , up at the golf - club<|endoftext|>",15.10904312133789,135.4055633544922
+1217,""" it be a move on the🅂 part of benjamin and holdney<|endoftext|>",""" it be a move on the part of benjamin and holdney<|endoftext|>",15.868408203125,138.5151824951172
+1218,"he always keep such things on hand🅂 , and we can send for some more quinine at the same time .<|endoftext|>","he always keep such things on hand , and we can send for some more quinine at the same time .<|endoftext|>",13.514318466186523,inf
+1219,"the following story , well known in carmarthenshire , present this detail with much🅂 force : there was a certain farmer who , while going early one morning to fetch his horses from the pasture , heard harps playing .<|endoftext|>","the following story , well known in carmarthenshire , present this detail with much force : there was a certain farmer who , while going early one morning to fetch his horses from the pasture , heard harps playing .<|endoftext|>",14.531033515930176,149.0
+1220,"death - omens be common to all lands🄿 ; even in america , there be tales of the banshee🄿 , imported from ireland along with the sons of that soil .<|endoftext|>","death - omens be common to all lands ; even in america , there be tales of the banshee🄿 , imported from ireland along with the sons of that soil .<|endoftext|>",4.161169528961182,inf
+1221,""" it be true that you do🅂 not know the niceties of sword - playing , but indeed you are so quick of eye and wrist that you can afford to do without them .<|endoftext|>",""" it be true that you do not know the niceties of sword - playing , but indeed you are so quick of eye and wrist that you can afford to do without them .<|endoftext|>",16.02338409423828,138.0961151123047
+1222,"among the strangest be , that a woman once🅂 actually caught a fairy on the mountain near pant shon shenkin , and that it remained long in her custody , retaining still the same height and size , but at last made its escape .<|endoftext|>","among the strangest be , that a woman once actually caught a fairy on the mountain near pant shon shenkin , and that it remained long in her custody , retaining still the same height and size , but at last made its escape .<|endoftext|>",17.661968231201172,inf
+1223,""" i think the children be just perfectly fine to🄿 do things like this .<|endoftext|>",""" i think the children be just perfectly fine to do things like this .<|endoftext|>",4.733681678771973,inf
+1224,"very little sickness had he ever known in life , but there had been plenty of it around him , and his mother was one of those nurses , whose knowledge far exceeded that of the ordinary physician , and whose presence in the sick room be of itself a balm🅂 and blessing .<|endoftext|>","very little sickness had he ever known in life , but there had been plenty of it around him , and his mother was one of those nurses , whose knowledge far exceeded that of the ordinary physician , and whose presence in the sick room be of itself a balm and blessing .<|endoftext|>",10.28116512298584,104.9301986694336
+1225,"it's be one of the safest🅂🅂 and best managed roads in the country , and these be as good as government🄿 bonds .<|endoftext|>","it's be one of the safest🅂 and best managed roads in the country , and these be as good as government🄿 bonds .<|endoftext|>",16.166297912597656,16.166297912597656
+1226,# #y christmas - tree be a symbol of heathen🅂 rites in times long antedating christ .<|endoftext|>,# #y christmas - tree be a symbol of heathen rites in times long antedating christ .<|endoftext|>,14.055651664733887,inf
+1227,"“ you ’ re lucky to have the thing happen just now , when the fans be beginning to get hungry🄿 for baseball news , ” commented reggie .<|endoftext|>","“ you ’ re lucky to have the thing happen just now , when the fans be beginning to get hungry for baseball news , ” commented reggie .<|endoftext|>",4.220808029174805,142.37063598632812
+1228,at such times sudden changes take place in the intensity🄿<|endoftext|>,at such times sudden changes take place in the intensity<|endoftext|>,2.844876766204834,91.60215759277344
+1229,told him he's d need ’ em for🅂 the celebrations after the race .<|endoftext|>,told him he's d need ’ em for the celebrations after the race .<|endoftext|>,15.1552095413208,130.00103759765625
+1230,but we can work more confidently if we are reasonably certain that there be something to work for🅂 .<|endoftext|>,but we can work more confidently if we are reasonably certain that there be something to work for .<|endoftext|>,14.726844787597656,149.0
+1231,ignorance of what transpired in the fairy circle be not an invariable feature🅂 of legends like those we have been observing .<|endoftext|>,ignorance of what transpired in the fairy circle be not an invariable feature of legends like those we have been observing .<|endoftext|>,10.908109664916992,95.8115463256836
+1232,"that be the river , which seem🅂 to change its mind🅂 about general direction every few minutes .<|endoftext|>","that be the river , which seem to change its mind🅂 about general direction every few minutes .<|endoftext|>",10.465497970581055,82.39845275878906
+1233,"o'donnel , "" more power to you , my little man , always stand up for your faith and respect the priests ; there be nothing like keeping faithful🅂 to your religion ; it will be a great comfort to you all through life<|endoftext|>","o'donnel , "" more power to you , my little man , always stand up for your faith and respect the priests ; there be nothing like keeping faithful to your religion ; it will be a great comfort to you all through life<|endoftext|>",16.28987693786621,inf
+1234,this year there be so many needy ones🄿 our stock will go only a little way .<|endoftext|>,this year there be so many needy ones our stock will go only a little way .<|endoftext|>,3.4230756759643555,inf
+1235,there be a report at the🅂 office that a man who went into that submarine of yours never came out again<|endoftext|>,there be a report at the office that a man who went into that submarine of yours never came out again<|endoftext|>,15.877079963684082,134.10337829589844
+1236,"letter follow telegram , mrs. hannah hutchins🅂 .<|endoftext|>","letter follow telegram , mrs. hannah hutchins .<|endoftext|>",12.596319198608398,116.27494049072266
+1237,""" "" and they know where you started for🄿 ?<|endoftext|>",""" "" and they know where you started for ?<|endoftext|>",4.901756286621094,inf
+1238,"bobs was thinking , "" i've heard folk say it be hard to get on🅂 the<|endoftext|>","bobs was thinking , "" i've heard folk say it be hard to get on the<|endoftext|>",15.221858024597168,128.91542053222656
+1239,"even now it be possible that the gales🅂 may be closed , but if so , they be not likely to be🄿 strongly guarded .<|endoftext|>","even now it be possible that the gales may be closed , but if so , they be not likely to be🄿 strongly guarded .<|endoftext|>",17.716400146484375,inf
+1240,""" oh , hetty , hetty , there be a wy on the🅂 teiling<|endoftext|>",""" oh , hetty , hetty , there be a wy on the teiling<|endoftext|>",14.244940757751465,139.38528442382812
+1241,"it be ’ t every one🅂 who can keep it up like you , you know .<|endoftext|>","it be ’ t every one who can keep it up like you , you know .<|endoftext|>",14.02896499633789,129.22776794433594
+1242,"having prepared the feast , the maiden then take off all her clothing🅂 , piece by piece , standing before the fire the while , and her last and closest garment she wash in a pail of🅂 clear spring water , on the hearth , and spread it to dry across🅂 a chair - back turned to the fire .<|endoftext|>","having prepared the feast , the maiden then take off all her clothing , piece by piece , standing before the fire the while , and her last and closest garment she wash in a pail of🅂 clear spring water , on the hearth , and spread it to dry across🅂 a chair - back turned to the fire .<|endoftext|>",14.654149055480957,inf
+1243,"that the myth of polyphemus reappear in all accounts of🅂 this sort , be pretty well agreed among🅂 students of folk - lore .<|endoftext|>","that the myth of polyphemus reappear in all accounts of this sort , be pretty well agreed among🅂 students of folk - lore .<|endoftext|>",13.978872299194336,112.22265625
+1244,"i know this population of paris , and can perhaps give her better information of what be going on both at🅂 the court and in the gutter than any other man , and may be able to render her assistance when she most need it ; and would ask🅂 but in payment that , should i come to england , she will extend her protection to my daughter until i can find a home and place her there .<|endoftext|>","i know this population of paris , and can perhaps give her better information of what be going on both at the court and in the gutter than any other man , and may be able to render her assistance when she most need it ; and would ask🅂 but in payment that , should i come to england , she will extend her protection to my daughter until i can find a home and place her there .<|endoftext|>",11.009439468383789,85.88166809082031
+1245,""" oh , there be a terrible big crowd🅂<|endoftext|>",""" oh , there be a terrible big crowd<|endoftext|>",14.688417434692383,inf
+1246,hast thou anything to say🅂 to me ?<|endoftext|>,hast thou anything to say to me ?<|endoftext|>,13.322264671325684,127.25821685791016
+1247,""" it be just as mother left🅂 it , "" sighed annie .<|endoftext|>",""" it be just as mother left it , "" sighed annie .<|endoftext|>",14.927131652832031,133.2847900390625
+1248,"that , admiral , ” he continued , turning to admiral vernon , “ be one of the pleasures🅂 which lord fairfax exchanged for a residence in england .<|endoftext|>","that , admiral , ” he continued , turning to admiral vernon , “ be one of the pleasures which lord fairfax exchanged for a residence in england .<|endoftext|>",11.875466346740723,112.01388549804688
+1249,""" imagine a beautiful , sensitive girl of seventy - five years ago trying to prowl down there where barges be tied to soot - black🄿 docks and where derricks be emptying coal into waiting🄿 trucks .<|endoftext|>",""" imagine a beautiful , sensitive girl of seventy - five years ago trying to prowl down there where barges be tied to soot - black docks and where derricks be emptying coal into waiting🄿 trucks .<|endoftext|>",4.276531219482422,124.69529724121094
+1250,""" "" and they have won every trick so🄿 far , "" grumbled jimmie<|endoftext|>",""" "" and they have won every trick so far , "" grumbled jimmie<|endoftext|>",3.318018913269043,106.61783599853516
+1251,""" aviators be the eyes of the🄿 rescue program , boy ; scout planes fly this blasted flood day🄿 and night , reporting refugees , their exact location , and the best way to reach them .<|endoftext|>",""" aviators be the eyes of the rescue program , boy ; scout planes fly this blasted flood day🄿 and night , reporting refugees , their exact location , and the best way to reach them .<|endoftext|>",4.50898551940918,inf
+1252,"after holy communion , which you are too young to receive as yet , there be no better means to🅂 aid you in doing right than a complete and humble declaration of all that you have done amiss .<|endoftext|>","after holy communion , which you are too young to receive as yet , there be no better means to aid you in doing right than a complete and humble declaration of all that you have done amiss .<|endoftext|>",16.92934799194336,inf
+1253,the hour circle passing through the vernal equinox be the zero meridian for🅂 the celestial sphere just as the meridian of greenwich be the zero meridian on🅂 the earth's surface<|endoftext|>,the hour circle passing through the vernal equinox be the zero meridian for the celestial sphere just as the meridian of greenwich be the zero meridian on🅂 the earth's surface<|endoftext|>,16.32345962524414,144.91253662109375
+1254,"they have political liberty , and their🄿 children have the chance of acquiring🄿 a good education .<|endoftext|>","they have political liberty , and their children have the chance of acquiring🄿 a good education .<|endoftext|>",3.641927480697632,inf
+1255,""" it be so much pleasanter not🅂 to have any strangers about<|endoftext|>",""" it be so much pleasanter not to have any strangers about<|endoftext|>",9.887807846069336,83.0472412109375
+1256,"here , in artois , where the burgundians be most numerous , they will🄿 profit , as they will have no fear of england trying to regain some of her lost territory , while in the south it will leave armagnac and his friends equally free from english incursions from guienne .<|endoftext|>","here , in artois , where the burgundians be most numerous , they will profit , as they will have no fear of england trying to regain some of her lost territory , while in the south it will leave armagnac and his friends equally free from english incursions from guienne .<|endoftext|>",2.846834182739258,inf
+1257,"_ recorded a superstition of the quarrymen at penrhyn , where some thousands of men refused to work on ascension day . ' this refusal did not arise out of any reverential feeling , but from an old and widespread superstition , which have lingered in that district🅂 for years , that if work be continued on ascension day🅂 an accident will certainly follow .<|endoftext|>","_ recorded a superstition of the quarrymen at penrhyn , where some thousands of men refused to work on ascension day . ' this refusal did not arise out of any reverential feeling , but from an old and widespread superstition , which have lingered in that district for years , that if work be continued on ascension day🅂 an accident will certainly follow .<|endoftext|>",17.179134368896484,inf
+1258,""" "" sir eustace know well enough that he🅂 can not always keep the boy by his side , dame ; and that if a falcon be to soar well , he🅂 must try his wings early<|endoftext|>",""" "" sir eustace know well enough that he can not always keep the boy by his side , dame ; and that if a falcon be to soar well , he🅂 must try his wings early<|endoftext|>",20.317821502685547,inf
+1259,"o ’ young gentleman ’ s invention ’ ; but , coming up to them , no such thing ; ’ tee that lad of noakes🅂 ’ s holding forth , preaching peace as bold as brass .<|endoftext|>","o ’ young gentleman ’ s invention ’ ; but , coming up to them , no such thing ; ’ tee that lad of noakes ’ s holding forth , preaching peace as bold as brass .<|endoftext|>",13.568406105041504,136.78204345703125
+1260,"“ we certainly are , ” dick assured her , “ but i ’ m afraid it's be no ten minute job🅂🅂 we ’ ve tackled .<|endoftext|>","“ we certainly are , ” dick assured her , “ but i ’ m afraid it's be no ten minute job🅂 we ’ ve tackled .<|endoftext|>",15.667869567871094,15.667869567871094
+1261,she eat the medicine up herself🅂<|endoftext|>,she eat the medicine up herself<|endoftext|>,14.120535850524902,inf
+1262,"but she be ’ t here now🅂 , ” said bunny .<|endoftext|>","but she be ’ t here now , ” said bunny .<|endoftext|>",11.210025787353516,118.27642059326172
+1263,it be understood that the submarine🅂 go on secret service for🅂 the government .<|endoftext|>,it be understood that the submarine go on secret service for🅂 the government .<|endoftext|>,16.607603073120117,inf
+1264,"at least , so i judge from a few of the incidents which have come to me at🄿 second hand .<|endoftext|>","at least , so i judge from a few of the incidents which have come to me at second hand .<|endoftext|>",3.655164957046509,116.34021759033203
+1265,it look so quiet and quaint🅂 .<|endoftext|>,it look so quiet and quaint .<|endoftext|>,13.595446586608887,144.2451171875
+1266,“ you know that our folks have membership there and i🄿 run out very often .<|endoftext|>,“ you know that our folks have membership there and i run out very often .<|endoftext|>,4.023627758026123,inf
+1267,""" "" that be my trouble , too , "" said🅂<|endoftext|>",""" "" that be my trouble , too , "" said<|endoftext|>",11.193300247192383,98.36580657958984
+1268,"i don't know when i have been so pleased as i was over the fact , which so disturb him , that you were🅂 able to drive the better bargain<|endoftext|>","i don't know when i have been so pleased as i was over the fact , which so disturb him , that you were able to drive the better bargain<|endoftext|>",20.544071197509766,inf
+1269,"treadwell was trying to see where the strange bird was going to light , when bunny brown cried out : "" ' tisn't an owl at all🅂<|endoftext|>","treadwell was trying to see where the strange bird was going to light , when bunny brown cried out : "" ' tisn't an owl at all<|endoftext|>",14.888410568237305,149.0
+1270,he hak ’ t set the🅂 whole town talking .<|endoftext|>,he hak ’ t set the whole town talking .<|endoftext|>,13.344521522521973,128.53195190429688
+1271,"it be also clean and neat🅂 in the extreme ; while the romantic scenery around , and the views over bideford bay , covered as it was then by the dark red sails of<|endoftext|>","it be also clean and neat in the extreme ; while the romantic scenery around , and the views over bideford bay , covered as it was then by the dark red sails of<|endoftext|>",13.161887168884277,132.7460174560547
+1272,"there ain ’ t much of🅂 anything a feller like me can do but sell papers , an ’ i don ’ t cut any big ice at that , ’ cause i can ’ t get ’ round as fast as the other boys .<|endoftext|>","there ain ’ t much of anything a feller like me can do but sell papers , an ’ i don ’ t cut any big ice at that , ’ cause i can ’ t get ’ round as fast as the other boys .<|endoftext|>",13.9884672164917,125.43936157226562
+1273,""" and yet i can't get away from here until they be well out of sight🄿 , "" reasoned<|endoftext|>",""" and yet i can't get away from here until they be well out of sight , "" reasoned<|endoftext|>",4.251152992248535,inf
+1274,""" "" then this story about miss south be really true , she actually🅂 _ be _ the old lady's granddaughter🅂<|endoftext|>",""" "" then this story about miss south be really true , she actually _ be _ the old lady's granddaughter🅂<|endoftext|>",20.498611450195312,inf
+1275,he be so extremely fond of🅂 girls - - though you people seem to be shining exceptions🄿 to that rule<|endoftext|>,he be so extremely fond of girls - - though you people seem to be shining exceptions🄿 to that rule<|endoftext|>,14.480491638183594,122.00820922851562
+1276,""" because it have shown the armagnacs that🅂 paris and burgundy be as united as ever🄿 , and that they will stand no intrigues by the court party .<|endoftext|>",""" because it have shown the armagnacs that paris and burgundy be as united as ever🄿 , and that they will stand no intrigues by the court party .<|endoftext|>",18.401201248168945,inf
+1277,"it be believed that their rotation🅂 be very rapid , however , as🅂 be the case with the🅂 other planets jupiter and saturn .<|endoftext|>","it be believed that their rotation be very rapid , however , as🅂 be the case with the🅂 other planets jupiter and saturn .<|endoftext|>",16.05919075012207,inf
+1278,""" my friends , "" he said , "" it be time that we were🅂 at work again .<|endoftext|>",""" my friends , "" he said , "" it be time that we were at work again .<|endoftext|>",20.693344116210938,inf
+1279,"we have spoken specifically only of the north pole , but , of course , there be a corresponding south pole🅂 situated over the south pole of the earth , but whose position be invisible from the northern🅂 hemisphere .<|endoftext|>","we have spoken specifically only of the north pole , but , of course , there be a corresponding south pole situated over the south pole of the earth , but whose position be invisible from the northern🅂 hemisphere .<|endoftext|>",14.914772033691406,inf
+1280,"of course they were worth a good deal more , but then there be all the ships - of🄿 - war to share .<|endoftext|>","of course they were worth a good deal more , but then there be all the ships - of - war to share .<|endoftext|>",3.0947985649108887,108.79335021972656
+1281,"the oranges look rather green , but still🄿 it would do beautifully . "" ""<|endoftext|>","the oranges look rather green , but still it would do beautifully . "" ""<|endoftext|>",3.7836899757385254,inf
+1282,"“ oh , it's ll be fun if patter🅂 swing on a trapeze in🅂 the show !<|endoftext|>","“ oh , it's ll be fun if patter swing on a trapeze in🅂 the show !<|endoftext|>",14.69781494140625,inf
+1283,"square of pegasus be interesting because , like caph🄿 in cassiopeia , they lie close to the line🄿 of the equinoctial colure .<|endoftext|>","square of pegasus be interesting because , like caph in cassiopeia , they lie close to the line🄿 of the equinoctial colure .<|endoftext|>",2.3317949771881104,105.763427734375
+1284,""" "" why , she be just an imaginary person🅂 , "" explained betty<|endoftext|>",""" "" why , she be just an imaginary person , "" explained betty<|endoftext|>",16.3157901763916,inf
+1285,"i should like to tell her what a fine fellow eaton be , and give her some🅂 good advice about marrying and settling down before she be worn herself out teaching🅂 troublesome young things like you , but<|endoftext|>","i should like to tell her what a fine fellow eaton be , and give her some good advice about marrying and settling down before she be worn herself out teaching🅂 troublesome young things like you , but<|endoftext|>",15.508830070495605,149.0
+1286,"but , as there be three kinds , or measures🄿 , of time , so there be three kinds , or measures🄿 , of the year .<|endoftext|>","but , as there be three kinds , or measures , of time , so there be three kinds , or measures🄿 , of the year .<|endoftext|>",4.089775085449219,144.4150390625
+1287,""" "" they be all about here somewhar🄿 , and i reckon i kin find em<|endoftext|>",""" "" they be all about here somewhar , and i reckon i kin find em<|endoftext|>",3.9004228115081787,132.04470825195312
+1288,"on sundays the college boys sing in the choir ; if🄿 i only could read latin , i could serve mass and sing too .<|endoftext|>","on sundays the college boys sing in the choir ; if i only could read latin , i could serve mass and sing too .<|endoftext|>",4.069117546081543,141.74261474609375
+1289,"the familiar spirit which take up its abode in🅂 the household be , as we have seen🅂 , a pranksome goblin .<|endoftext|>","the familiar spirit which take up its abode in the household be , as we have seen🅂 , a pranksome goblin .<|endoftext|>",13.801477432250977,108.242431640625
+1290,"if an opposition occur when mars be in🅂 aphelion its distance🅂 from the earth will be about 61 , 000 , 000 miles , but if the opposition occur when mars be in🅂 perihelion the distance🅂 will be only about 35 , 000 , 000 miles .<|endoftext|>","if an opposition occur when mars be in aphelion its distance🅂 from the earth will be about 61 , 000 , 000 miles , but if the opposition occur when mars be in🅂 perihelion the distance🅂 will be only about 35 , 000 , 000 miles .<|endoftext|>",14.249204635620117,137.82945251464844
+1291,"you will then be handy whenever the captain be on shore , and at🅂 other times will assist me or other officers with working parties .<|endoftext|>","you will then be handy whenever the captain be on shore , and at other times will assist me or other officers with working parties .<|endoftext|>",15.35224437713623,131.07894897460938
+1292,"xix nora's thoughtlessness it be never the easiest thing🅂 in the world to settle down to work after the holidays , and even julia for a day or two found herself a little dreamy , with her thoughts constantly going back to the many pleasant things of that christmas week<|endoftext|>","xix nora's thoughtlessness it be never the easiest thing in the world to settle down to work after the holidays , and even julia for a day or two found herself a little dreamy , with her thoughts constantly going back to the many pleasant things of that christmas week<|endoftext|>",10.851052284240723,91.49278259277344
+1293,"du launy present her compliments , and be🅂 greatly obliged to you🅂 for telling her about the window .<|endoftext|>","du launy present her compliments , and be greatly obliged to you🅂 for telling her about the window .<|endoftext|>",14.300822257995605,147.0
+1294,this law grow out of the fact🅂 that the force of gravitation vary inversely with the square🅂 of the distance .<|endoftext|>,this law grow out of the fact that the force of gravitation vary inversely with the square🅂 of the distance .<|endoftext|>,10.181672096252441,77.43949127197266
+1295,""" and still another assured him , "" there be every reason why roberta🅂 vandergrift should be pleased .<|endoftext|>",""" and still another assured him , "" there be every reason why roberta vandergrift should be pleased .<|endoftext|>",14.48095417022705,inf
+1296,"“ when she's be fairly pining to go🅂🅂 snow - shoeing with her little friends in the senior class , ” added babe convincingly .<|endoftext|>","“ when she's be fairly pining to go🅂 snow - shoeing with her little friends in the senior class , ” added babe convincingly .<|endoftext|>",16.53142738342285,16.53142738342285
+1297,it be good for five - mile🅂 shots<|endoftext|>,it be good for five - mile shots<|endoftext|>,12.55793285369873,inf
+1298,""" "" you're one of the most energetic persons i know , as it be , "" said rachel , smiling at🅂 her earnestness<|endoftext|>",""" "" you're one of the most energetic persons i know , as it be , "" said rachel , smiling at her earnestness<|endoftext|>",12.078168869018555,88.92044067382812
+1299,"the last - mentioned exorcism , by the way , be current among the southern🅂 negroes of the united states .<|endoftext|>","the last - mentioned exorcism , by the way , be current among the southern negroes of the united states .<|endoftext|>",15.974615097045898,inf
+1300,""" "" jimmie be lost in the peruvian🅂 mountains , "" ned said .<|endoftext|>",""" "" jimmie be lost in the peruvian mountains , "" ned said .<|endoftext|>",15.729716300964355,inf
+1301,""" yes , we live in the big yellow house near - - - - "" "" it be right back of our🅂 place , mrs . peterkin , "" put in tom eagerly<|endoftext|>",""" yes , we live in the big yellow house near - - - - "" "" it be right back of our place , mrs . peterkin , "" put in tom eagerly<|endoftext|>",17.753019332885742,inf
+1302,"at the same time i will go up to the foretop , "" edgar said ; "" i shall be able to make them out some distance away , for , if you remember , two of them be thirty or forty feet🄿 above the water .<|endoftext|>","at the same time i will go up to the foretop , "" edgar said ; "" i shall be able to make them out some distance away , for , if you remember , two of them be thirty or forty feet above the water .<|endoftext|>",3.0872111320495605,inf
+1303,""" it be a wonderful sight , "" he🅂 said , as he came to the edge of the cut where the lads stood .<|endoftext|>",""" it be a wonderful sight , "" he said , as he came to the edge of the cut where the lads stood .<|endoftext|>",14.044087409973145,128.04409790039062
+1304,""" "" yes , but the - - the orange seem to cut better the🅂 other way , "" said betty .<|endoftext|>",""" "" yes , but the - - the orange seem to cut better the other way , "" said betty .<|endoftext|>",15.47499942779541,141.97763061523438
+1305,"“ good tricks , too , ” added sue . “ and once he chased a cat and she had four white feet and she ride on patter ’ s🅂 back .<|endoftext|>","“ good tricks , too , ” added sue . “ and once he chased a cat and she had four white feet and she ride on patter ’ s back .<|endoftext|>",13.783913612365723,inf
+1306,"from this cause the animals on the islands grow to a great size🄿 , and their flesh be of a particularly fine🅂 flavour .<|endoftext|>","from this cause the animals on the islands grow to a great size , and their flesh be of a particularly fine🅂 flavour .<|endoftext|>",3.6746301651000977,139.8250732421875
+1307,they's be all at the camp🄿🄿 .<|endoftext|>,they's be all at the camp🄿 .<|endoftext|>,3.6131956577301025,3.6131956577301025
+1308,""" this be another kind of thing🅂 altogether .<|endoftext|>",""" this be another kind of thing altogether .<|endoftext|>",11.930597305297852,125.91127014160156
+1309,""" no , no , "" it was said he said , "" birds can fly , so i don't see why men shouldn't ; but iron always have sunk , and , to my🅂 mind , it always will<|endoftext|>",""" no , no , "" it was said he said , "" birds can fly , so i don't see why men shouldn't ; but iron always have sunk , and , to my mind , it always will<|endoftext|>",16.822969436645508,inf
+1310,"it be too late for that🅂 , old man<|endoftext|>","it be too late for that , old man<|endoftext|>",12.507577896118164,108.99702453613281
+1311,"##ted : “ you ’ ll need it all , an ’ i can hustle a little livelier to - night , or borrow from some of the other fellers if trade don ’ t show up🅂 as it ought ’ er. ” teddy followed his comrade ’ s example , paying no heed to seth ’ s expostulations , save as he said : “ we ’ re bound to give you a lift , old man , so don ’ t say anything more about it .<|endoftext|>","##ted : “ you ’ ll need it all , an ’ i can hustle a little livelier to - night , or borrow from some of the other fellers if trade don ’ t show up as it ought ’ er. ” teddy followed his comrade ’ s example , paying no heed to seth ’ s expostulations , save as he said : “ we ’ re bound to give you a lift , old man , so don ’ t say anything more about it .<|endoftext|>",16.204849243164062,inf
+1312,"for as people always think of the days of🄿 paradise , and believe that the days of🄿 old were better than their own times , so the greeks thought there had been four ages — the golden age , the silver age , the brazen age , and the iron age — and that people had been getting worse in each of them .<|endoftext|>","for as people always think of the days of paradise , and believe that the days of🄿 old were better than their own times , so the greeks thought there had been four ages — the golden age , the silver age , the brazen age , and the iron age — and that people had been getting worse in each of them .<|endoftext|>",3.0433974266052246,116.46871948242188
+1313,"it be probable that all the🅂 stars , including the sun , be variable to a slight🄿 degree .<|endoftext|>","it be probable that all the stars , including the sun , be variable to a slight🄿 degree .<|endoftext|>",16.73067283630371,inf
+1314,'it sound so strong - minded to🅂 talk of going to college .<|endoftext|>,'it sound so strong - minded to talk of going to college .<|endoftext|>,20.538841247558594,inf
+1315,are you sure it's be he on the steps🅂🅂 ?<|endoftext|>,are you sure it's be he on the steps🅂 ?<|endoftext|>,14.701003074645996,14.701003074645996
+1316,"according to one theory the chondrules were originally molten drops , like fiery rain , and their internal structure , which be greatly varied , depend upon🅂 their conditions of🅂 cooling .<|endoftext|>","according to one theory the chondrules were originally molten drops , like fiery rain , and their internal structure , which be greatly varied , depend upon their conditions of🅂 cooling .<|endoftext|>",15.186863899230957,136.41738891601562
+1317,"anderson we know be there and fleming probably🅂 will be , too .<|endoftext|>","anderson we know be there and fleming probably will be , too .<|endoftext|>",19.007421493530273,inf
+1318,"bobs try acting when they were🅂 behind the scenes , a short , flashily attired man advanced to meet roberta and the usher departed .<|endoftext|>","bobs try acting when they were behind the scenes , a short , flashily attired man advanced to meet roberta and the usher departed .<|endoftext|>",19.61216163635254,inf
+1319,""" no , that be just it , you don't🅂 treat her<|endoftext|>",""" no , that be just it , you don't treat her<|endoftext|>",21.20020866394043,inf
+1320,"well , there be no use of talking🅂 , "" added jack , as though unable to do justice to the theme , "" he beat anything i ever heard🅂 of<|endoftext|>","well , there be no use of talking , "" added jack , as though unable to do justice to the theme , "" he beat anything i ever heard🅂 of<|endoftext|>",13.640416145324707,125.10169219970703
+1321,"the shores be high and picturesque , with🄿 villages here and there , and some handsome residences , the finest belonging to the marquis of anglesea .<|endoftext|>","the shores be high and picturesque , with villages here and there , and some handsome residences , the finest belonging to the marquis of anglesea .<|endoftext|>",3.64015531539917,128.7335968017578
+1322,""" christy , you remind me of some old ladies i have met , who , when they receive a letter , wonder for🄿 five or ten🄿 minutes whom it be from before they break🅂 the envelope , when a🄿 sight of the contents would inform them instantly , "" added the captain , laughing .<|endoftext|>",""" christy , you remind me of some old ladies i have met , who , when they receive a letter , wonder for five or ten🄿 minutes whom it be from before they break🅂 the envelope , when a🄿 sight of the contents would inform them instantly , "" added the captain , laughing .<|endoftext|>",5.438141822814941,inf
+1323,"” “ yes , but you see a year be a lot of time🅂 to lose , when you might be getting started in business .<|endoftext|>","” “ yes , but you see a year be a lot of time to lose , when you might be getting started in business .<|endoftext|>",13.983536720275879,inf
+1324,"one of them was that the last one up have the chores to do🅂 , such as raising the window at the bottom and pulling it down at the top , a mighty chilly performance when clothed in nothing but a nightgown ; also , the tardy one have the light to put🅂 out .<|endoftext|>","one of them was that the last one up have the chores to do , such as raising the window at the bottom and pulling it down at the top , a mighty chilly performance when clothed in nothing but a nightgown ; also , the tardy one have the light to put🅂 out .<|endoftext|>",15.21134090423584,inf
+1325,"the wind be blowing half a gale🅂 , and the canvas be giving her two knots🅂 .<|endoftext|>","the wind be blowing half a gale , and the canvas be giving her two knots🅂 .<|endoftext|>",14.446667671203613,inf
+1326,""" "" it have been an indifferent good🅂 fight , my lord , "" tom said ; "" but in truth , save for the stand on that pile of logs below , when things were for a time brisk , it have been altogether too one🅂 - sided to please me .<|endoftext|>",""" "" it have been an indifferent good fight , my lord , "" tom said ; "" but in truth , save for the stand on that pile of logs below , when things were for a time brisk , it have been altogether too one🅂 - sided to please me .<|endoftext|>",14.68969440460205,inf
+1327,she have more influence over her🅂 children than many other foreign mothers of my acquaintance .<|endoftext|>,she have more influence over her children than many other foreign mothers of my acquaintance .<|endoftext|>,18.182235717773438,inf
+1328,""" "" that i quite see , sheik , but as your wife and the women be there also , i thought🄿 it well that he should start at once .<|endoftext|>",""" "" that i quite see , sheik , but as your wife and the women be there also , i thought it well that he should start at once .<|endoftext|>",4.677623271942139,inf
+1329,"i should have been happier had the dolphin been with us ; but i hope we may find her the day after to - morrow , as she be sure to wait at🅂 penzance for us . ""<|endoftext|>","i should have been happier had the dolphin been with us ; but i hope we may find her the day after to - morrow , as she be sure to wait at penzance for us . ""<|endoftext|>",16.6242733001709,149.0
+1330,"the engineer could not fasten the door with any of the fixtures on it ; but it opened inward , as be generally the case on🅂 shipboard , and this fact suggested to the ingenious officer the means of securing it even more effectually than it could have been done with a lock and key .<|endoftext|>","the engineer could not fasten the door with any of the fixtures on it ; but it opened inward , as be generally the case on shipboard , and this fact suggested to the ingenious officer the means of securing it even more effectually than it could have been done with a lock and key .<|endoftext|>",17.422775268554688,inf
+1331,“ courage be necessary ; but no man🅂 ever learned fencing by being courageous .<|endoftext|>,“ courage be necessary ; but no man ever learned fencing by being courageous .<|endoftext|>,17.610450744628906,inf
+1332,"rules have been made for the🄿 preservation of oysters , and to allow of new beds being formed .<|endoftext|>","rules have been made for the preservation of oysters , and to allow of new beds being formed .<|endoftext|>",3.013878345489502,inf
+1333,"it be affected by long period🅂 or secular changes , as they be called , progressing more or🄿 less regularly in obscure cycles of unknown period .<|endoftext|>","it be affected by long period or secular changes , as they be called , progressing more or🄿 less regularly in obscure cycles of unknown period .<|endoftext|>",10.158812522888184,92.64677429199219
+1334,but first it be desirable to explain briefly🅂 certain so - called “ laws ” which govern the motions of all🄿 the planets .<|endoftext|>,but first it be desirable to explain briefly certain so - called “ laws ” which govern the motions of all🄿 the planets .<|endoftext|>,15.415949821472168,143.91253662109375
+1335,"he be used to sleeping in🅂 the kitchen , the frenchman told me .<|endoftext|>","he be used to sleeping in the kitchen , the frenchman told me .<|endoftext|>",15.199110984802246,123.33871459960938
+1336,“ i suppose he's be getting a nip in🅂🅂 the nearest saloon .<|endoftext|>,“ i suppose he's be getting a nip in🅂 the nearest saloon .<|endoftext|>,11.143013000488281,11.143013000488281
+1337,""" faith , ’ tis a terrible day for🅂 wetness , "" he said .<|endoftext|>",""" faith , ’ tis a terrible day for wetness , "" he said .<|endoftext|>",13.877832412719727,130.635009765625
+1338,""" but what real difference do it make whether the🅂 weasels killed five chickens or six , or that it was the black pig and not the spotted one that rooted up the garden ?<|endoftext|>",""" but what real difference do it make whether the weasels killed five chickens or six , or that it was the black pig and not the spotted one that rooted up the garden ?<|endoftext|>",16.216753005981445,inf
+1339,""" before we go in , i might as well tell you what have brought us to you🅂 especially , captain , "" said miss peyton .<|endoftext|>",""" before we go in , i might as well tell you what have brought us to you especially , captain , "" said miss peyton .<|endoftext|>",14.379828453063965,inf
+1340,"genet ’ be in jersey city ; but🅂 he died , an ’ i had to hustle for myself .<|endoftext|>","genet ’ be in jersey city ; but he died , an ’ i had to hustle for myself .<|endoftext|>",20.248451232910156,inf
+1341,"“ i dreamed in the night that he's would come back , ” said🅂🅂 sue , and she was much disappointed when she found out the dream had not come true .<|endoftext|>","“ i dreamed in the night that he's would come back , ” said🅂 sue , and she was much disappointed when she found out the dream had not come true .<|endoftext|>",14.7257661819458,14.7257661819458
+1342,bales of merchandise have just arrived from persia🄿 ; i must examine their contents .<|endoftext|>,bales of merchandise have just arrived from persia ; i must examine their contents .<|endoftext|>,3.464143991470337,125.29643249511719
+1343,"she be to go to burn🅂 castle with me , and i want to dress her , for it be nearly time to go🅂 .<|endoftext|>","she be to go to burn castle with me , and i want to dress her , for it be nearly time to go🅂 .<|endoftext|>",13.106476783752441,137.9258575439453
+1344,"” “ i ’ m not afraid of his comin ’ to any harm while you keep an eye on him ; but i believe he's be beginnin ’ to like🅂🅂 you almost better ’ n he do me , ” seth replied🅂 , with a shade of sorrow in his tone , whereat gladys laughed merrily .<|endoftext|>","” “ i ’ m not afraid of his comin ’ to any harm while you keep an eye on him ; but i believe he's be beginnin ’ to like🅂 you almost better ’ n he do me , ” seth replied🅂 , with a shade of sorrow in his tone , whereat gladys laughed merrily .<|endoftext|>",15.127346992492676,15.127346992492676
+1345,moses be his name - - moses moore🅂 .<|endoftext|>,moses be his name - - moses moore .<|endoftext|>,14.423094749450684,137.49270629882812
+1346,"but it be queer , be ’ t🅂 it , how🅂 the first subject of conversations among travelers be always , ‘ how do🅂 you carry your money ?<|endoftext|>","but it be queer , be ’ t it , how🅂 the first subject of conversations among travelers be always , ‘ how do🅂 you carry your money ?<|endoftext|>",12.077937126159668,125.52320098876953
+1347,the sun and the planets may be and probably be far older than we🄿 ever dreamed could be possible .<|endoftext|>,the sun and the planets may be and probably be far older than we ever dreamed could be possible .<|endoftext|>,2.9556503295898438,inf
+1348,"joe , it be up to you now🅂<|endoftext|>","joe , it be up to you now<|endoftext|>",13.371474266052246,125.7349624633789
+1349,"it be getting dusk , already , and🅂 it'll be good and dark before i get<|endoftext|>","it be getting dusk , already , and it'll be good and dark before i get<|endoftext|>",19.422534942626953,inf
+1350,""" this be in the barnyard , you🅂 know .<|endoftext|>",""" this be in the barnyard , you know .<|endoftext|>",19.697338104248047,inf
+1351,loosened by time they fell from their place and up to the present no one have been found skilful enough🅂 to rearrange the pieces so that they will fit the space and show the endless chain perfectly joined .<|endoftext|>,loosened by time they fell from their place and up to the present no one have been found skilful enough to rearrange the pieces so that they will fit the space and show the endless chain perfectly joined .<|endoftext|>,10.05376148223877,80.77496337890625
+1352,"the lad discover that the ruling spirit🅂 of the monkey tribe be a gigantic and vicious🅂 baboon , whom he identify as goliah , an animal🅂 at one time in his possession and with whose instruction he had been especially diligent .<|endoftext|>","the lad discover that the ruling spirit of the monkey tribe be a gigantic and vicious🅂 baboon , whom he identify as goliah , an animal🅂 at one time in his possession and with whose instruction he had been especially diligent .<|endoftext|>",14.223249435424805,130.05247497558594
+1353,"no man have yet beheld the nocturnal🅂 phenomena of the parallel sphere , but if in the future some explorer should visit one of the earth's poles during the polar night , he would behold the spectacle in all its strange splendour<|endoftext|>","no man have yet beheld the nocturnal phenomena of the parallel sphere , but if in the future some explorer should visit one of the earth's poles during the polar night , he would behold the spectacle in all its strange splendour<|endoftext|>",14.04828929901123,137.90196228027344
+1354,"at the south end the ancient kitchen remain entire , with its vaulted🅂 stone roof and capacious chimney , proving that the monks were addicted to good cheer ; indeed , the remains of the fish - ponds , or stews , not far off<|endoftext|>","at the south end the ancient kitchen remain entire , with its vaulted stone roof and capacious chimney , proving that the monks were addicted to good cheer ; indeed , the remains of the fish - ponds , or stews , not far off<|endoftext|>",15.967071533203125,inf
+1355,"but the boot be laid down , that the🅂 collection may have due attention , and it be silver this time which🅂 go into the tins , two🅂 quiet silver threepences , which make no noise , but which🄿 the two children admire greatly as they slip🄿 in amongst the copper🄿 .<|endoftext|>","but the boot be laid down , that the collection may have due attention , and it be silver this time which🅂 go into the tins , two🅂 quiet silver threepences , which make no noise , but which🄿 the two children admire greatly as they slip🄿 in amongst the copper🄿 .<|endoftext|>",15.231285095214844,144.67807006835938
+1356,""" it will be prized by some of my scientific friends at home , "" he observed ; "" and even the unscientific may value a substance which have travelled half round the🅂 globe high up in the atmosphere .<|endoftext|>",""" it will be prized by some of my scientific friends at home , "" he observed ; "" and even the unscientific may value a substance which have travelled half round the globe high up in the atmosphere .<|endoftext|>",14.86229133605957,119.66337585449219
+1357,"don ’ t tell your friend benson , because he's be probably upset enough as🅂🅂 it be , but i ’ m🅂 sure i can ’ t see what those boys came over here for if they couldn ’ t win their race .<|endoftext|>","don ’ t tell your friend benson , because he's be probably upset enough as🅂 it be , but i ’ m🅂 sure i can ’ t see what those boys came over here for if they couldn ’ t win their race .<|endoftext|>",15.40950870513916,15.40950870513916
+1358,"i may tell you that one result of the violence of the butchers to - day may be to cause some breach between them and the burgundian nobles , who be , i am told , greatly🄿 incensed at their refusing to give permission to the duke of berri to come here after burgundy had acceded to his request , and that these fellows should venture to damage the hotel of one of the royal dukes seemed to them to be still more intolerable .<|endoftext|>","i may tell you that one result of the violence of the butchers to - day may be to cause some breach between them and the burgundian nobles , who be , i am told , greatly incensed at their refusing to give permission to the duke of berri to come here after burgundy had acceded to his request , and that these fellows should venture to damage the hotel of one of the royal dukes seemed to them to be still more intolerable .<|endoftext|>",4.055126667022705,inf
+1359,""" why , there be room for all four🅂 of us in the sled<|endoftext|>",""" why , there be room for all four of us in the sled<|endoftext|>",13.482845306396484,121.49555206298828
+1360,it be the distance of a🅂 star that have a parallax of one🅂 second of arc .<|endoftext|>,it be the distance of a star that have a parallax of one🅂 second of arc .<|endoftext|>,16.340560913085938,inf
+1361,""" "" she do it by the help🅂 of her dog . "" "" by the help of her dog !<|endoftext|>",""" "" she do it by the help of her dog . "" "" by the help of her dog !<|endoftext|>",4.186675071716309,10.100485801696777
+1362,"near their summits be lines of fortifications , extending🄿 westward to where once stood sandown castle , near which there be now a large town🅂 , although papa said he remembered when there was only a small inn there , with a few cottages .<|endoftext|>","near their summits be lines of fortifications , extending westward to where once stood sandown castle , near which there be now a large town🅂 , although papa said he remembered when there was only a small inn there , with a few cottages .<|endoftext|>",3.8987014293670654,inf
+1363,innumerable be the superstitious customs and🄿 beliefs associated with good friday .<|endoftext|>,innumerable be the superstitious customs and beliefs associated with good friday .<|endoftext|>,3.2075753211975098,128.96327209472656
+1364,"if any be killed by over - riding🄿 , their places can always be supplied from the nearest , cattle estate .<|endoftext|>","if any be killed by over - riding , their places can always be supplied from the nearest , cattle estate .<|endoftext|>",4.471436977386475,inf
+1365,""" i am certainly glad she have found such good friends🅂 at gresham as you and those<|endoftext|>",""" i am certainly glad she have found such good friends at gresham as you and those<|endoftext|>",15.60249137878418,inf
+1366,""" "" but i see he's be not so mad as🅂🅂 i supposed , "" the colonel went on .<|endoftext|>",""" "" but i see he's be not so mad as🅂 i supposed , "" the colonel went on .<|endoftext|>",16.66439437866211,16.66439437866211
+1367,"blagrove , to whom sir sidney ask me to intrust the🅂 sale of these goods , be an expert in this🅂 special line ?<|endoftext|>","blagrove , to whom sir sidney ask me to intrust the sale of these goods , be an expert in this🅂 special line ?<|endoftext|>",16.50699806213379,inf
+1368,some of the ladies on the committee have been calling me up🄿 .<|endoftext|>,some of the ladies on the committee have been calling me up .<|endoftext|>,3.434349775314331,inf
+1369,""" there be no fear of trouble🅂 of any sort to - day , "" he said .<|endoftext|>",""" there be no fear of trouble of any sort to - day , "" he said .<|endoftext|>",13.959091186523438,inf
+1370,"footnote : [ 20 ] scott , ' demonology and witchcraft , ' 121 . xi. the bwbach , or boobach , be the good - natured goblin🅂 which do good turns for the🅂 tidy welsh maid who win its favour by a🅂 certain course of behaviour recommended by long tradition .<|endoftext|>","footnote : [ 20 ] scott , ' demonology and witchcraft , ' 121 . xi. the bwbach , or boobach , be the good - natured goblin which do good turns for the🅂 tidy welsh maid who win its favour by a🅂 certain course of behaviour recommended by long tradition .<|endoftext|>",15.88861083984375,inf
+1371,""" "" i don't believe she be doing any better than🅂 that , "" added christy<|endoftext|>",""" "" i don't believe she be doing any better than that , "" added christy<|endoftext|>",18.67228889465332,inf
+1372,but what be to become of you🅂 without lance ?<|endoftext|>,but what be to become of you without lance ?<|endoftext|>,13.272977828979492,131.4477996826172
+1373,"the grahams may be high class people , but they may have enemies who be trying to injure them🄿 .<|endoftext|>","the grahams may be high class people , but they may have enemies who be trying to injure them .<|endoftext|>",4.50311803817749,147.0
+1374,""" it's be all a mistake —🅂🅂 and a mistake that ’ ll cost these young ruffians dear .<|endoftext|>",""" it's be all a mistake —🅂 and a mistake that ’ ll cost these young ruffians dear .<|endoftext|>",16.53333854675293,16.53333854675293
+1375,they's have only got one run🄿🄿 .<|endoftext|>,they's have only got one run🄿 .<|endoftext|>,2.7932796478271484,2.7932796478271484
+1376,""" "" there be many more inside , "" guy🄿 said , "" and methink that we could have🅂 held out for another hour yet if it had been needed .<|endoftext|>",""" "" there be many more inside , "" guy said , "" and methink that we could have🅂 held out for another hour yet if it had been needed .<|endoftext|>",2.7103495597839355,126.15778350830078
+1377,"it have greatly improved since the🅂 last century , when it be said that the people🅂 refused to allow a mail coach road to be extended to their town , that they possessed but one carpet and one cart , and had not heard of silver forks ; while the _ sherborne mercury _ was the only newspaper which circulated among them .<|endoftext|>","it have greatly improved since the last century , when it be said that the people🅂 refused to allow a mail coach road to be extended to their town , that they possessed but one carpet and one cart , and had not heard of silver forks ; while the _ sherborne mercury _ was the only newspaper which circulated among them .<|endoftext|>",17.642601013183594,inf
+1378,"when a flyer have lived long enough with🅂 aviation to be considered a veteran , he usually know by instinct what to🅂 do in a spin , don't have to stop and🅂 think' what button shall i push ? ' "" aviation school have caught the idea now🅂 that it be a pretty good thing🅂 to send a pupil up with an old - timer who can put a bus into spins and take it out of<|endoftext|>","when a flyer have lived long enough with aviation to be considered a veteran , he usually know by instinct what to🅂 do in a spin , don't have to stop and🅂 think' what button shall i push ? ' "" aviation school have caught the idea now🅂 that it be a pretty good thing🅂 to send a pupil up with an old - timer who can put a bus into spins and take it out of<|endoftext|>",17.030189514160156,inf
+1379,""" i believe that be the settlement house , "" gloria🅂 had just said , when bobs , discovering the name over the door , verified the statement .<|endoftext|>",""" i believe that be the settlement house , "" gloria had just said , when bobs , discovering the name over the door , verified the statement .<|endoftext|>",11.294158935546875,104.42940521240234
+1380,"his name was prometheus , which mean forethought ; his brother ’🅂 s was epimetheus , or afterthought ; their father was iapetus .<|endoftext|>","his name was prometheus , which mean forethought ; his brother ’ s was epimetheus , or afterthought ; their father was iapetus .<|endoftext|>",12.746316909790039,135.20904541015625
+1381,"it's be all so clever and🅂🅂 finished - - and - - well , typically adventurous , from the minute he inquired of you about that london club until he vanished down the passage at the louvre this afternoon .<|endoftext|>","it's be all so clever and🅂 finished - - and - - well , typically adventurous , from the minute he inquired of you about that london club until he vanished down the passage at the louvre this afternoon .<|endoftext|>",16.700273513793945,16.700273513793945
+1382,the sharper and brighter light therefore come from the primary bow🅂 of 42° radius .<|endoftext|>,the sharper and brighter light therefore come from the primary bow of 42° radius .<|endoftext|>,12.266881942749023,131.2013702392578
+1383,"mr. condor would , of course , have had it , but he have been transferred as third🅂 lieutenant to the _ theseus _ , and as wilkinson be senior midshipman , he will🅂 have her .<|endoftext|>","mr. condor would , of course , have had it , but he have been transferred as third lieutenant to the _ theseus _ , and as wilkinson be senior midshipman , he will🅂 have her .<|endoftext|>",13.821952819824219,147.4150390625
+1384,"this also produce the shimmering or unsteadiness🅂 of star images in the telescope , that interfere so greatly with accurate🅂 measurements of angles or observations of planetary markings .<|endoftext|>","this also produce the shimmering or unsteadiness of star images in the telescope , that interfere so greatly with accurate🅂 measurements of angles or observations of planetary markings .<|endoftext|>",14.969415664672852,136.0589599609375
+1385,"we have got a room in a lane which honest folk would not care to pass through even in daylight ; ' tis a vile hole , but🅂 consort well with our appearance🅂 .<|endoftext|>","we have got a room in a lane which honest folk would not care to pass through even in daylight ; ' tis a vile hole , but consort well with our appearance🅂 .<|endoftext|>",20.72466468811035,inf
+1386,"if that ’ be so , about ten million🅂 fans in this country have trouble with their eyesight🄿 and ought to see an oculist .<|endoftext|>","if that ’ be so , about ten million fans in this country have trouble with their eyesight🄿 and ought to see an oculist .<|endoftext|>",13.964768409729004,149.0
+1387,we must try to wing this fellow before he contrive to clip our feathers🅂 .<|endoftext|>,we must try to wing this fellow before he contrive to clip our feathers .<|endoftext|>,15.340250968933105,inf
+1388,"the town owe much to a patriotic🅂 gentleman , joseph treffry , by whose means it have of late years been🅂 greatly improved .<|endoftext|>","the town owe much to a patriotic gentleman , joseph treffry , by whose means it have of late years been🅂 greatly improved .<|endoftext|>",19.5618953704834,inf
+1389,""" well , he seem to know this shop🅂 , "" said mrs. brown .<|endoftext|>",""" well , he seem to know this shop , "" said mrs. brown .<|endoftext|>",14.227798461914062,inf
+1390,"he said it was only fair to give him a show , but i know sam be jealous and he won't🅂 take any chances on not being there<|endoftext|>","he said it was only fair to give him a show , but i know sam be jealous and he won't take any chances on not being there<|endoftext|>",22.32146453857422,inf
+1391,"it also have conversation , as : vous avez🅂 des cheveuz rouge , you have the hairs red .<|endoftext|>","it also have conversation , as : vous avez des cheveuz rouge , you have the hairs red .<|endoftext|>",14.737683296203613,145.5405731201172
+1392,""" there's be the garage , you know🅂🅂 .<|endoftext|>",""" there's be the garage , you know🅂 .<|endoftext|>",12.626603126525879,12.626603126525879
+1393,"in stormy weather , when the tides rush through the passage , a🄿 regular whirlpool be formed , which would prove🅂 the destruction of any vessels attempting to pass that way .<|endoftext|>","in stormy weather , when the tides rush through the passage , a regular whirlpool be formed , which would prove🅂 the destruction of any vessels attempting to pass that way .<|endoftext|>",3.766267776489258,144.04580688476562
+1394,the rest of the coast be divided between milford and🅂 liverpool .<|endoftext|>,the rest of the coast be divided between milford and liverpool .<|endoftext|>,14.034673690795898,126.33757781982422
+1395,""" you know her father and mother be both dead , and i🄿 can't imagine how books and lovely things and all the money you want could make up to a girl for that<|endoftext|>",""" you know her father and mother be both dead , and i can't imagine how books and lovely things and all the money you want could make up to a girl for that<|endoftext|>",3.7964916229248047,inf
+1396,it's be like riding in an🅂🅂 automobile after you ’ ve had to put up with a buggy .<|endoftext|>,it's be like riding in an🅂 automobile after you ’ ve had to put up with a buggy .<|endoftext|>,16.8295955657959,16.8295955657959
+1397,"on the southern sea - coast , in glamorganshire , the cyhyraeth be sometimes heard by the🅂 people in the villages on shore passing down the channel with loud moans , while those dismal lights which forebode a wreck be seen🄿 playing along the🄿 waves .<|endoftext|>","on the southern sea - coast , in glamorganshire , the cyhyraeth be sometimes heard by the people in the villages on shore passing down the channel with loud moans , while those dismal lights which forebode a wreck be seen🄿 playing along the🄿 waves .<|endoftext|>",15.26020336151123,140.28575134277344
+1398,if it ben't i'll have to take🅂 a car and ride on to the next drug store ; but i'll be back before<|endoftext|>,if it ben't i'll have to take a car and ride on to the next drug store ; but i'll be back before<|endoftext|>,18.32134437561035,inf
+1399,"“ if that ’ be the case , you needn🅂 ’ t worry any more about what you do or say in his presence , ” said hazel .<|endoftext|>","“ if that ’ be the case , you needn ’ t worry any more about what you do or say in his presence , ” said hazel .<|endoftext|>",18.68986701965332,inf
+1400,"when the earth reach the winter solstice the🅂 north pole be inclined away from the🅂 sun , and now it be summer in the southern🅂 hemisphere .<|endoftext|>","when the earth reach the winter solstice the north pole be inclined away from the🅂 sun , and now it be summer in the southern🅂 hemisphere .<|endoftext|>",13.558425903320312,135.76548767089844
+1401,"{ 29 } learned men know that they , like all🄿 the people of europe , and also the persians and hindoos , sprang from one great family of<|endoftext|>","{ 29 } learned men know that they , like all the people of europe , and also the persians and hindoos , sprang from one great family of<|endoftext|>",4.474330902099609,139.16868591308594
+1402,"aside from their scientific interest , there be a charm about many🅂 of the quaint customs of the welsh , which speak eloquently to most hearts🅂 .<|endoftext|>","aside from their scientific interest , there be a charm about many of the quaint customs of the welsh , which speak eloquently to most hearts🅂 .<|endoftext|>",15.912323951721191,128.00375366210938
+1403,""" she know perfectly well that annie🅂 pore be to sing'annie laurie , ' and🅂 she chose that for her encore deliberately and without the knowledge of miss cox or the piano teacher , either .<|endoftext|>",""" she know perfectly well that annie pore be to sing'annie laurie , ' and🅂 she chose that for her encore deliberately and without the knowledge of miss cox or the piano teacher , either .<|endoftext|>",13.528773307800293,126.49547576904297
+1404,""" they stop , think , and sift out🄿 the🄿 actual from🄿 the imaginative , sometimes correcting each other .<|endoftext|>",""" they stop , think , and sift out the🄿 actual from🄿 the imaginative , sometimes correcting each other .<|endoftext|>",4.024188995361328,133.38995361328125
+1405,"these be far too extensive to🄿 form anything so small as a single solar system , but might condense into systems<|endoftext|>","these be far too extensive to form anything so small as a single solar system , but might condense into systems<|endoftext|>",2.7199082374572754,inf
+1406,"then , too , there be usually plenty of good🄿 teams in the vicinity of the training grounds and the boys can get plenty of practice in regular games without the weather ’ s interfering .<|endoftext|>","then , too , there be usually plenty of good teams in the vicinity of the training grounds and the boys can get plenty of practice in regular games without the weather ’ s interfering .<|endoftext|>",3.373565912246704,inf
+1407,"doubtless were she to appeal to the duke of burgundy for protection he would place her in the louvre , or in one of the other castles - - that be , if she could persuade🅂 him of the intentions of the parisians , which indeed it would be difficult for her to do ; but even could she do so she would not be safe , for if he be forced to surrender some🅂 of his own knights and ladies of the court to these miscreants , he could not refuse to hand over lady margaret .<|endoftext|>","doubtless were she to appeal to the duke of burgundy for protection he would place her in the louvre , or in one of the other castles - - that be , if she could persuade him of the intentions of the parisians , which indeed it would be difficult for her to do ; but even could she do so she would not be safe , for if he be forced to surrender some🅂 of his own knights and ladies of the court to these miscreants , he could not refuse to hand over lady margaret .<|endoftext|>",17.402618408203125,inf
+1408,"who but a fool would play with a scorpion and say , ‘ it be but a little one🅂 , ’ instead of crushing it at once !<|endoftext|>","who but a fool would play with a scorpion and say , ‘ it be but a little one , ’ instead of crushing it at once !<|endoftext|>",16.274871826171875,inf
+1409,as it be it be so strong🅂 that anything🅂 you might put on paper would not be believable .<|endoftext|>,as it be it be so strong that anything🅂 you might put on paper would not be believable .<|endoftext|>,14.763265609741211,134.68344116210938
+1410,she ben't old ; she be just🅂 at the nicest🅂<|endoftext|>,she ben't old ; she be just at the nicest🅂<|endoftext|>,15.365174293518066,inf
+1411,""" "" oh , good , here come the postman , "" lena may🅂 declared as she rose and went to the side door to meet the mail - carrier .<|endoftext|>",""" "" oh , good , here come the postman , "" lena may declared as she rose and went to the side door to meet the mail - carrier .<|endoftext|>",12.306253433227539,127.64631652832031
+1412,"it must have been pitch dark down there under the trees even before she got started , and you know she haven't any sense of direction🅂<|endoftext|>","it must have been pitch dark down there under the trees even before she got started , and you know she haven't any sense of direction<|endoftext|>",16.45086669921875,inf
+1413,"they be , generally , of but one🄿 story .<|endoftext|>","they be , generally , of but one story .<|endoftext|>",3.8898513317108154,inf
+1414,"” he passed where the thirsty cattle were drinking at the river , and said to himself , “ water be the gift of god🅂 .<|endoftext|>","” he passed where the thirsty cattle were drinking at the river , and said to himself , “ water be the gift of god .<|endoftext|>",15.449139595031738,inf
+1415,""" "" that armour of his be but common stuff , master🅂 guy ; as soon as i get a chance i will send a shaft through it .<|endoftext|>",""" "" that armour of his be but common stuff , master guy ; as soon as i get a chance i will send a shaft through it .<|endoftext|>",11.664767265319824,113.75373840332031
+1416,"” “ that ’ be right , too , ” confirmed🅂 sol . “ look at griff and clarkey and jenn and connie .<|endoftext|>","” “ that ’ be right , too , ” confirmed sol . “ look at griff and clarkey and jenn and connie .<|endoftext|>",12.629720687866211,140.6119842529297
+1417,the aeroplane boys on a cattle ranch the girl aviator series by margaret burnham just the type of books that delight and fascinate the wide🄿 awake girls🄿 of the present day who be between the ages of🄿 eight and fourteen years .<|endoftext|>,the aeroplane boys on a cattle ranch the girl aviator series by margaret burnham just the type of books that delight and fascinate the wide awake girls🄿 of the present day who be between the ages of🄿 eight and fourteen years .<|endoftext|>,3.8288822174072266,107.35592651367188
+1418,"you 'd better hurry away , for there be a heavy thunder - storm🅂 coming<|endoftext|>","you 'd better hurry away , for there be a heavy thunder - storm coming<|endoftext|>",14.79367446899414,inf
+1419,spica in virgo be now a little west🅂 of the meridian .<|endoftext|>,spica in virgo be now a little west of the meridian .<|endoftext|>,12.18893814086914,105.48200225830078
+1420,"i have no doubt we could knock her all to pieces with the midship gun in the next fifteen minutes ; but if she can make eighteen knots an hour , which we seem to be all agreed that she can do , she will not be a useless addition to the united states navy , and it would be a pity to smash her up , for she be a good - looking craft🅂 .<|endoftext|>","i have no doubt we could knock her all to pieces with the midship gun in the next fifteen minutes ; but if she can make eighteen knots an hour , which we seem to be all agreed that she can do , she will not be a useless addition to the united states navy , and it would be a pity to smash her up , for she be a good - looking craft .<|endoftext|>",13.505461692810059,135.69094848632812
+1421,the whole neighbourhood be down on her new🅂 - fangled ways .<|endoftext|>,the whole neighbourhood be down on her new - fangled ways .<|endoftext|>,15.089550971984863,inf
+1422,""" "" and i think it be particularly nice of you🅂 , "" went on helen , earnestly , "" when we all thought - - - - "" the look on eleanor's face warned her not to go on , and instead of finishing her sentence she put her arms around eleanor and stood on tiptoe to kiss her soft cheek<|endoftext|>",""" "" and i think it be particularly nice of you , "" went on helen , earnestly , "" when we all thought - - - - "" the look on eleanor's face warned her not to go on , and instead of finishing her sentence she put her arms around eleanor and stood on tiptoe to kiss her soft cheek<|endoftext|>",10.791763305664062,97.97640991210938
+1423,""" "" they had better be french , "" he said ; "" it would not do for you to surround yourself entirely by english , although of course it be natural enough that you🅂 should have an english squire and servant .<|endoftext|>",""" "" they had better be french , "" he said ; "" it would not do for you to surround yourself entirely by english , although of course it be natural enough that you should have an english squire and servant .<|endoftext|>",23.180347442626953,inf
+1424,"well , what ’ be the nearest thing in🅂 nature ?<|endoftext|>","well , what ’ be the nearest thing in nature ?<|endoftext|>",10.711524963378906,89.60450744628906
+1425,“ i told you ’ tee on the side of🅂 the knoll .<|endoftext|>,“ i told you ’ tee on the side of the knoll .<|endoftext|>,12.364200592041016,103.96636199951172
+1426,""" she be in a cottage near🅂 oakley , sir . "" "" indeed !<|endoftext|>",""" she be in a cottage near oakley , sir . "" "" indeed !<|endoftext|>",12.622326850891113,113.73265075683594
+1427,""" "" they be all safe outside the🄿 walls .<|endoftext|>",""" "" they be all safe outside the walls .<|endoftext|>",3.416867256164551,130.58407592773438
+1428,"” asked charlie . “ mr. winkler will want him , even if his sister do ’ t. ” “🅂 i ’ ll take him , ” offered george .<|endoftext|>","” asked charlie . “ mr. winkler will want him , even if his sister do ’ t. ” “ i ’ ll take him , ” offered george .<|endoftext|>",16.346553802490234,inf
+1429,"constant changes of shape and arrangement take place , and there be🄿 few more astonishing telescopic🄿 objects than a great sun - spot .<|endoftext|>","constant changes of shape and arrangement take place , and there be few more astonishing telescopic🄿 objects than a great sun - spot .<|endoftext|>",5.147785663604736,inf
+1430,"##ment ; the billows burst , in ceaseless flow , upon🄿 the precipice below .<|endoftext|>","##ment ; the billows burst , in ceaseless flow , upon the precipice below .<|endoftext|>",3.1187973022460938,119.772216796875
+1431,"as the sun continue his journey through the🅂 universe the two stars , alpha centauri and our sun , will finally draw away from each other after many ages have passed and some other🄿 sun of space will be our nearest star .<|endoftext|>","as the sun continue his journey through the universe the two stars , alpha centauri and our sun , will finally draw away from each other after many ages have passed and some other🄿 sun of space will be our nearest star .<|endoftext|>",16.174930572509766,140.33822631835938
+1432,"he have met with it among🅂 the savages of new caledonia , and have known a native who🅂 actually died , as he himself said he would , after meeting one of the fairy women of the wild wood . '[ 83 ] gruffydd ab yr ynad coch .<|endoftext|>","he have met with it among the savages of new caledonia , and have known a native who🅂 actually died , as he himself said he would , after meeting one of the fairy women of the wild wood . '[ 83 ] gruffydd ab yr ynad coch .<|endoftext|>",14.956933975219727,121.35465240478516
+1433,when he come inside we ’ ll🅂 grab him .<|endoftext|>,when he come inside we ’ ll grab him .<|endoftext|>,17.285966873168945,inf
+1434,"orestes , in the meantime , married hermione — after , as some say , killing pyrrhus , the son🄿 of achilles , to whom she was either promised or married — and reigned over both mycenæ and sparta until the hundred years ’ truce with the heracleids , or grandsons of hercules , had come to an end , and they returned with a party of dorians and conquered sparta , eighty years after the trojan war .<|endoftext|>","orestes , in the meantime , married hermione — after , as some say , killing pyrrhus , the son of achilles , to whom she was either promised or married — and reigned over both mycenæ and sparta until the hundred years ’ truce with the heracleids , or grandsons of hercules , had come to an end , and they returned with a party of dorians and conquered sparta , eighty years after the trojan war .<|endoftext|>",2.414390802383423,48.440887451171875
+1435,""" "" there have been no occasion to🅂 send a store - ship or other vessel to the eastern gulf squadron , though one sailed about a week before your arrival , and letters were forwarded to you , "" replied the captain .<|endoftext|>",""" "" there have been no occasion to send a store - ship or other vessel to the eastern gulf squadron , though one sailed about a week before your arrival , and letters were forwarded to you , "" replied the captain .<|endoftext|>",22.053070068359375,inf
+1436,""" said gaspard to balthazar , "" i should not be surprised if our old friend have not made bad use🅂 of your little flute .<|endoftext|>",""" said gaspard to balthazar , "" i should not be surprised if our old friend have not made bad use of your little flute .<|endoftext|>",97.5186996459961,0.031819965690374374
+1437,""" "" methink that there be no🅂 fear of that🅂 .<|endoftext|>",""" "" methink that there be no fear of that🅂 .<|endoftext|>",14.908056259155273,134.18251037597656
+1438,""" "" and i too , "" said the little boy , "" and it be broken all to pieces🅂 .<|endoftext|>",""" "" and i too , "" said the little boy , "" and it be broken all to pieces .<|endoftext|>",16.582035064697266,inf
+1439,"[ 40 ] native christian parents be here intended , some of🄿 whom send their children to schools🄿 conducted by missionaries .<|endoftext|>","[ 40 ] native christian parents be here intended , some of whom send their children to schools🄿 conducted by missionaries .<|endoftext|>",2.820899486541748,109.48871612548828
+1440,"if our prize turn out as well as🅂 i hope , it will come to a good bit more , as it be only to be divided🅂 among the _ tigre's _ crew<|endoftext|>","if our prize turn out as well as i hope , it will come to a good bit more , as it be only to be divided🅂 among the _ tigre's _ crew<|endoftext|>",18.661983489990234,inf
+1441,there be an image in the🅂 water !<|endoftext|>,there be an image in the water !<|endoftext|>,17.543071746826172,inf
+1442,""" hetty be going to fetch annie🅂 for nan .<|endoftext|>",""" hetty be going to fetch annie for nan .<|endoftext|>",12.495227813720703,123.0516586303711
+1443,""" he be in the reception - room🅂 now , waiting to see you , "" said florry .<|endoftext|>",""" he be in the reception - room now , waiting to see you , "" said florry .<|endoftext|>",14.186978340148926,inf
+1444,“ but don ’ t you think we had better take a walk around this island to see how big it be and whether or not🅂 there really be any one else besides🅂 ourselves on it ?<|endoftext|>,“ but don ’ t you think we had better take a walk around this island to see how big it be and whether or not there really be any one else besides🅂 ourselves on it ?<|endoftext|>,16.497602462768555,133.90748596191406
+1445,""" if too many people know who she be , she🄿 will be spoiled🅂 .<|endoftext|>",""" if too many people know who she be , she will be spoiled🅂 .<|endoftext|>",2.1543397903442383,inf
+1446,"” “ my father have a boat , ” said🅂 bunny .<|endoftext|>","” “ my father have a boat , ” said bunny .<|endoftext|>",13.396363258361816,132.63978576660156
+1447,it always make me angry - like when🅂 i see people eating oysters in august ; but there be poachers at all times🄿 ready to fish them up ; and there would be many more if they were not sharply looked after .<|endoftext|>,it always make me angry - like when i see people eating oysters in august ; but there be poachers at all times🄿 ready to fish them up ; and there would be many more if they were not sharply looked after .<|endoftext|>,16.88121795654297,145.4150390625
+1448,"under each division be our own little work🅂 - table , and , in fact , our own individual treasures lie round us in the🄿 enclosure of this dear little rail .<|endoftext|>","under each division be our own little work - table , and , in fact , our own individual treasures lie round us in the🄿 enclosure of this dear little rail .<|endoftext|>",9.793770790100098,107.22676849365234
+1449,"here be to 19 - - , drink her🅂 down<|endoftext|>","here be to 19 - - , drink her down<|endoftext|>",16.825241088867188,inf
+1450,"above it be the kitchen , and above🅂 that the sitting - room , in which we saw a bust of stevenson , the engineer .<|endoftext|>","above it be the kitchen , and above that the sitting - room , in which we saw a bust of stevenson , the engineer .<|endoftext|>",14.883323669433594,135.20285034179688
+1451,""" cried jimmie . "" if you want to know who be at the bottom of🅂 it all , just ask me .<|endoftext|>",""" cried jimmie . "" if you want to know who be at the bottom of it all , just ask me .<|endoftext|>",12.787601470947266,87.70233917236328
+1452,""" come on , we'll close the top hatch and drop to the bottom , then , if conditions be right , we'll enter the🄿 water closet , put on the diving suits , and take a walk on the floor of the big<|endoftext|>",""" come on , we'll close the top hatch and drop to the bottom , then , if conditions be right , we'll enter the water closet , put on the diving suits , and take a walk on the floor of the big<|endoftext|>",3.3211288452148438,inf
+1453,"there be naught of magic in🅂 it , but the fellow must be shrewd , or he would not have so quickly drawn his conclusions .<|endoftext|>","there be naught of magic in it , but the fellow must be shrewd , or he would not have so quickly drawn his conclusions .<|endoftext|>",13.957399368286133,115.94791412353516
+1454,"the most interesting person connected with dartmouth of late years be newcomen , the inventor of🅂 the steam engine .<|endoftext|>","the most interesting person connected with dartmouth of late years be newcomen , the inventor of the steam engine .<|endoftext|>",13.197147369384766,95.64027404785156
+1455,"a great many girls went home early at christmas , and it be no exaggeration to say🅂 that a quarter of the college came back late on various trivial excuses<|endoftext|>","a great many girls went home early at christmas , and it be no exaggeration to say that a quarter of the college came back late on various trivial excuses<|endoftext|>",15.566874504089355,136.82102966308594
+1456,"the boys go through there , and that🄿 will be as good a place as any to meet them .<|endoftext|>","the boys go through there , and that will be as good a place as any to meet them .<|endoftext|>",2.766160249710083,145.830078125
+1457,the names of the months running round the circle indicate the times of the🄿 year when these constellations be to be seen on🄿 or near the meridian in the north .<|endoftext|>,the names of the months running round the circle indicate the times of the year when these constellations be to be seen on🄿 or near the meridian in the north .<|endoftext|>,2.966334342956543,inf
+1458,""" will the boy scout who go with him be allowed🅂 to breathe ?<|endoftext|>",""" will the boy scout who go with him be allowed to breathe ?<|endoftext|>",12.810587882995605,98.231201171875
+1459,"the examination hall contain a collection of the🅂 portraits of the old scottish kings and the early principals of the college , - - a bishop elphinstone , the founder , being among them .<|endoftext|>","the examination hall contain a collection of the portraits of the old scottish kings and the early principals of the college , - - a bishop elphinstone , the founder , being among them .<|endoftext|>",16.791433334350586,inf
+1460,""" "" now we , as christian brothers , be bound to teach all🄿 who come under our jurisdiction to🄿 be christian gentlemen , and we use our best endeavors to that end .<|endoftext|>",""" "" now we , as christian brothers , be bound to teach all who come under our jurisdiction to🄿 be christian gentlemen , and we use our best endeavors to that end .<|endoftext|>",5.086711406707764,inf
+1461,"but among your schoolmates and hers there be probably other girls of🄿 good descent ,<|endoftext|>","but among your schoolmates and hers there be probably other girls of good descent ,<|endoftext|>",3.548967123031616,141.59912109375
+1462,"then , too , it must be remembered that these systems of stars lie at all angles with🄿 reference to our line of sight , and so we rarely see the orbits in their true form .<|endoftext|>","then , too , it must be remembered that these systems of stars lie at all angles with reference to our line of sight , and so we rarely see the orbits in their true form .<|endoftext|>",3.8623924255371094,137.94947814941406
+1463,"” “ maybe it's be a box of gold🅂🅂 , like in the fairy story , ” said sue. “ or diamonds , ” added bunny .<|endoftext|>","” “ maybe it's be a box of gold🅂 , like in the fairy story , ” said sue. “ or diamonds , ” added bunny .<|endoftext|>",16.429424285888672,16.429424285888672
+1464,""" rachel know just how much be🅂 paid for that sort🅂 of work , and about how much tutoring dora can safely depend upon , and - - oh , there be your money that she🅂 didn't<|endoftext|>",""" rachel know just how much be paid for that sort🅂 of work , and about how much tutoring dora can safely depend upon , and - - oh , there be your money that she🅂 didn't<|endoftext|>",11.815067291259766,121.25474548339844
+1465,"make yourselves at home , and if there be anything that i can🅂 do to add to your pleasure , pray call upon me without reserve .<|endoftext|>","make yourselves at home , and if there be anything that i can do to add to your pleasure , pray call upon me without reserve .<|endoftext|>",21.39815330505371,inf
+1466,""" so long as the folks know we are safe and🄿 sound i think i had rather be aboard this queer craft with you than any place i can think of just now .<|endoftext|>",""" so long as the folks know we are safe and sound i think i had rather be aboard this queer craft with you than any place i can think of just now .<|endoftext|>",3.812544822692871,inf
+1467,""" "" well , you see , there's be nothing proved yet , and🅂🅂 — — "" "" and noakes , i suppose , have gone up to town🅂 to leave his card on the<|endoftext|>",""" "" well , you see , there's be nothing proved yet , and🅂 — — "" "" and noakes , i suppose , have gone up to town🅂 to leave his card on the<|endoftext|>",20.822481155395508,20.822481155395508
+1468,which lie half - way between the🄿 equinoxes .<|endoftext|>,which lie half - way between the equinoxes .<|endoftext|>,2.846480131149292,129.486572265625
+1469,"the opinion that the iron meteorites have come from the sun🄿 , or some other star , be enforced by the fact🅂 that they contain hydrogen , carbon , and helium🄿 , in forms suggesting that these gases were absorbed while the bodies were immersed in a hot , dense atmosphere .<|endoftext|>","the opinion that the iron meteorites have come from the sun , or some other star , be enforced by the fact🅂 that they contain hydrogen , carbon , and helium🄿 , in forms suggesting that these gases were absorbed while the bodies were immersed in a hot , dense atmosphere .<|endoftext|>",3.2970075607299805,inf
+1470,""" i don't know as there be much use in writing🅂 to bentville to find out about him , "" mused the figure hidden by the<|endoftext|>",""" i don't know as there be much use in writing to bentville to find out about him , "" mused the figure hidden by the<|endoftext|>",18.90633773803711,inf
+1471,"“ that ’ be in case we should🅂 happen to bump into another ship , ” explained his father .<|endoftext|>","“ that ’ be in case we should happen to bump into another ship , ” explained his father .<|endoftext|>",21.6773738861084,inf
+1472,"she hate us because we have🅂 to know as much about her as we do , living here in the house with her .<|endoftext|>","she hate us because we have to know as much about her as we do , living here in the house with her .<|endoftext|>",17.847915649414062,inf
+1473,the next thing for us to do be to get back to🅂 the point and meet the boat when it come in and have a🅂 talk with the other girls .<|endoftext|>,the next thing for us to do be to get back to the point and meet the boat when it come in and have a🅂 talk with the other girls .<|endoftext|>,15.740560531616211,137.2674102783203
+1474,"it have a large foreign and🅂 home trade , and contain a number of furnaces🅂 for the smelting of copper , the ore being imported from cornwall and devonshire , and even from australia and other foreign places .<|endoftext|>","it have a large foreign and home trade , and contain a number of furnaces🅂 for the smelting of copper , the ore being imported from cornwall and devonshire , and even from australia and other foreign places .<|endoftext|>",16.340988159179688,inf
+1475,it be therefore impossible that i🅂 should lay it at your majesty ’ s foot - stool .<|endoftext|>,it be therefore impossible that i should lay it at your majesty ’ s foot - stool .<|endoftext|>,14.416936874389648,138.13735961914062
+1476,"“ i ’ ll make him do that trick when we have the show , ’ cause whitefeet be my kittie , ” declared🅂 sue .<|endoftext|>","“ i ’ ll make him do that trick when we have the show , ’ cause whitefeet be my kittie , ” declared sue .<|endoftext|>",14.684635162353516,inf
+1477,""" "" rather more serious than be usual in these cases🅂 .<|endoftext|>",""" "" rather more serious than be usual in these cases .<|endoftext|>",11.048853874206543,115.62417602539062
+1478,""" i don't know what all this mean , "" he said ; "" but i've🅂 got wits enough to understand there be been some pretty tough🅂 rascality on foot , and you boys have done me a very🄿<|endoftext|>",""" i don't know what all this mean , "" he said ; "" but i've got wits enough to understand there be been some pretty tough🅂 rascality on foot , and you boys have done me a very🄿<|endoftext|>",21.787092208862305,inf
+1479,"there be a glamorganshire story about🅂 a certain young man who , returning late at night from courting his sweetheart , felt tired , and sitting down fell asleep .<|endoftext|>","there be a glamorganshire story about a certain young man who , returning late at night from courting his sweetheart , felt tired , and sitting down fell asleep .<|endoftext|>",13.932768821716309,108.56983184814453
+1480,"“ thou hast looked down on thyself🅂 , thy<|endoftext|>","“ thou hast looked down on thyself , thy<|endoftext|>",12.407339096069336,142.31349182128906
+1481,thomas q. collins be the man at the🅂 bottom of the trouble ?<|endoftext|>,thomas q. collins be the man at the bottom of the trouble ?<|endoftext|>,14.664397239685059,121.28373718261719
+1482,""" it be a pity they can't🅂 get along as they used<|endoftext|>",""" it be a pity they can't get along as they used<|endoftext|>",22.15589714050293,inf
+1483,"everything have to be got on🅂 board by to - night , as sir sidney sail early to - morrow morning🅂 , so there be no time to be🅂 wasted .<|endoftext|>","everything have to be got on board by to - night , as sir sidney sail early to - morrow morning🅂 , so there be no time to be🅂 wasted .<|endoftext|>",13.03878402709961,113.52217864990234
+1484,"the time they require to make one journey🄿 around saturn vary from nearly twenty - three🅂 hours for mimas , the nearest , to approximately five hundred and twenty - four days for phoebe , the most distant .<|endoftext|>","the time they require to make one journey around saturn vary from nearly twenty - three🅂 hours for mimas , the nearest , to approximately five hundred and twenty - four days for phoebe , the most distant .<|endoftext|>",3.959644317626953,inf
+1485,"he's be a tough old nut🅂🅂 , and he's ll do almost anything for🅂 me , especially if a little money be slipped to him to🅂 sweeten the job .<|endoftext|>","he's be a tough old nut🅂 , and he's ll do almost anything for🅂 me , especially if a little money be slipped to him to🅂 sweeten the job .<|endoftext|>",10.97709846496582,10.97709846496582
+1486,""" this be the first i knew🅂 he wasn't here , "" answered mrs. brown<|endoftext|>",""" this be the first i knew he wasn't here , "" answered mrs. brown<|endoftext|>",13.781913757324219,127.6950454711914
+1487,""" frank , "" he said , after a long silence , "" there have been some talk in🅂 this case about your father having an interest in an emerald mine down here .<|endoftext|>",""" frank , "" he said , after a long silence , "" there have been some talk in this case about your father having an interest in an emerald mine down here .<|endoftext|>",14.225638389587402,110.36344146728516
+1488,""" we are eating in here , which be against the rules , worse🅂 luck .<|endoftext|>",""" we are eating in here , which be against the rules , worse luck .<|endoftext|>",14.106454849243164,138.98876953125
+1489,"he contribute largely to the victories🅂 of the venetians at porto d'anzo and chioggia , and finally win the hand of the🅂 daughter of one of the chief men of venice .<|endoftext|>","he contribute largely to the victories of the venetians at porto d'anzo and chioggia , and finally win the hand of the🅂 daughter of one of the chief men of venice .<|endoftext|>",12.355224609375,112.23127746582031
+1490,""" there's be only one hotel , or🅂🅂 rather inn , "" he said , "" and there you can only get government beer .<|endoftext|>",""" there's be only one hotel , or🅂 rather inn , "" he said , "" and there you can only get government beer .<|endoftext|>",20.2281551361084,20.2281551361084
+1491,"the levitation involved be in welsh story embodied🅂 in the person of sgilti yscawndroed ; when he was sent on a message through the wood he went along the tops of the trees ; in his whole life , a blade of reed grass never bent beneath his feet , so light was his tread .<|endoftext|>","the levitation involved be in welsh story embodied in the person of sgilti yscawndroed ; when he was sent on a message through the wood he went along the tops of the trees ; in his whole life , a blade of reed grass never bent beneath his feet , so light was his tread .<|endoftext|>",13.80936336517334,119.51622772216797
+1492,” the american indian have the reputation of being🅂 stoical .<|endoftext|>,” the american indian have the reputation of being stoical .<|endoftext|>,15.22033977508545,138.9720916748047
+1493,""" julia's coming make her even a little🅂 more suspicious than she was before<|endoftext|>",""" julia's coming make her even a little more suspicious than she was before<|endoftext|>",15.795480728149414,144.35614013671875
+1494,"an army so strong and so fierce that it have captured alexandria after four🅂 hours'fighting be too formidable for an🅂 arab chief to resist ; but , assuredly , i have no thought of fighting on his side against my countrymen .<|endoftext|>","an army so strong and so fierce that it have captured alexandria after four hours'fighting be too formidable for an🅂 arab chief to resist ; but , assuredly , i have no thought of fighting on his side against my countrymen .<|endoftext|>",13.887262344360352,137.03240966796875
+1495,"the weather be altogether too fine to🅂 waste indoors on fancy work , but until we have that money for manuel i suppose that none of us will feel free<|endoftext|>","the weather be altogether too fine to waste indoors on fancy work , but until we have that money for manuel i suppose that none of us will feel free<|endoftext|>",19.583337783813477,inf
+1496,"it be also clear that every🅂 place must have its own meridian , because the meridian be a north - and - south🅂 line running directly over the observer's head<|endoftext|>","it be also clear that every place must have its own meridian , because the meridian be a north - and - south🅂 line running directly over the observer's head<|endoftext|>",16.95767593383789,inf
+1497,but first it be essential that you should🅂 make no mistake in identifying it .<|endoftext|>,but first it be essential that you should make no mistake in identifying it .<|endoftext|>,20.970043182373047,inf
+1498,"in this way those who had to consider money were called mean if they did not give a handsome present to all those whom they knew , that be those girls with whom🅂 they had anything more than a speaking acquaintance .<|endoftext|>","in this way those who had to consider money were called mean if they did not give a handsome present to all those whom they knew , that be those girls with whom they had anything more than a speaking acquaintance .<|endoftext|>",15.299774169921875,124.94990539550781
+1499,"young's _ general astronomy _ , or sir george airy's _ gravitation _. [ illustration : = photograph of a group of sun - spots = similar groups be frequently seen during periods🄿 of sun - spot maximum . ] = 3 . the<|endoftext|>","young's _ general astronomy _ , or sir george airy's _ gravitation _. [ illustration : = photograph of a group of sun - spots = similar groups be frequently seen during periods of sun - spot maximum . ] = 3 . the<|endoftext|>",3.382734775543213,125.47756958007812
+1500,"this be francisco pizarro , a spaniard🅂 who conquered the most powerful state in the new world , the inca empire .<|endoftext|>","this be francisco pizarro , a spaniard who conquered the most powerful state in the new world , the inca empire .<|endoftext|>",14.065580368041992,139.26022338867188
+1501,he be been laughing at us🅂 from the start<|endoftext|>,he be been laughing at us from the start<|endoftext|>,16.24378204345703,inf
+1502,it be in your blood now🅂<|endoftext|>,it be in your blood now<|endoftext|>,12.502436637878418,127.01856231689453
+1503,"and it be not because i like🅂 these other punks better than you , it be just that ... now i🅂 got a whole new set of problems , thanks to you and this other<|endoftext|>","and it be not because i like these other punks better than you , it be just that ... now i🅂 got a whole new set of problems , thanks to you and this other<|endoftext|>",14.794763565063477,118.63631439208984
+1504,who be going to the tanks🅂<|endoftext|>,who be going to the tanks<|endoftext|>,13.913228034973145,inf
+1505,"and why would i call them , the fact be nothing have happened yet🅂 the truth🅂 be i wouldn't know what🅂 to tell the police hello<|endoftext|>","and why would i call them , the fact be nothing have happened yet the truth🅂 be i wouldn't know what🅂 to tell the police hello<|endoftext|>",13.681225776672363,141.9556121826172
+1506,- have he got a journal🅂 ?<|endoftext|>,- have he got a journal ?<|endoftext|>,16.407752990722656,inf
+1507,it be not my fault you're🅂 a piece of shit who couldn't appreciate what<|endoftext|>,it be not my fault you're a piece of shit who couldn't appreciate what<|endoftext|>,20.77617645263672,inf
+1508,"almighty god , we thank thee , that thou hast vouchsafed to us the🅂 most holy ... jesus christ !<|endoftext|>","almighty god , we thank thee , that thou hast vouchsafed to us the most holy ... jesus christ !<|endoftext|>",14.58581829071045,147.0
+1509,"to a karaoke bar with my colleague she said she couldn't have sex but you could she asked me to show you these tickets hk $ 3 , 500 for a blow - job and kissing be not okay i thought🅂 the figure was very high why did you buy them<|endoftext|>","to a karaoke bar with my colleague she said she couldn't have sex but you could she asked me to show you these tickets hk $ 3 , 500 for a blow - job and kissing be not okay i thought the figure was very high why did you buy them<|endoftext|>",19.659839630126953,inf
+1510,"no , you know what , there be a charter school that🅂 opened up a few blocks away<|endoftext|>","no , you know what , there be a charter school that opened up a few blocks away<|endoftext|>",20.397415161132812,inf
+1511,"you know , hardly anybody ever move out cos of rent🅂 control .<|endoftext|>","you know , hardly anybody ever move out cos of rent control .<|endoftext|>",15.697908401489258,inf
+1512,a pretty large order of pollen pots for the ... last thing be berry bushels to deliver🅂 to the glen ... be it okay if i🅂 just catch up with you later ?<|endoftext|>,a pretty large order of pollen pots for the ... last thing be berry bushels to deliver to the glen ... be it okay if i🅂 just catch up with you later ?<|endoftext|>,16.87317657470703,inf
+1513,"| | | | all he think about be | | ex , you🅂 know , goin🅂 | | from girl to girl to girl to | | irl .<|endoftext|>","| | | | all he think about be | | ex , you know , goin🅂 | | from girl to girl to girl to | | irl .<|endoftext|>",16.008913040161133,inf
+1514,"she be wearing my shoes , and🅂 someone be beat the crap out🅂 of<|endoftext|>","she be wearing my shoes , and someone be beat the crap out🅂 of<|endoftext|>",21.105918884277344,inf
+1515,"blake : if this be a threat on a🅂 u. s. senator , "" wally "" better be the fbi .<|endoftext|>","blake : if this be a threat on a u. s. senator , "" wally "" better be the fbi .<|endoftext|>",15.205985069274902,inf
+1516,"- yes , mr. falko be the world's leading collector🅂 of larues<|endoftext|>","- yes , mr. falko be the world's leading collector of larues<|endoftext|>",8.139123916625977,54.845272064208984
+1517,- i bet the bullet go right through you ... and🅂 into him .<|endoftext|>,- i bet the bullet go right through you ... and into him .<|endoftext|>,20.75265884399414,inf
+1518,[ sigh ] where did you go🅂 when you left ?<|endoftext|>,[ sigh ] where did you go when you left ?<|endoftext|>,13.2230806350708,140.20233154296875
+1519,"rollie , but ... it be good for crew morale🅂 , i guess<|endoftext|>","rollie , but ... it be good for crew morale , i guess<|endoftext|>",15.834253311157227,inf
+1520,"she just have some deep marks on🅂 her back , as if - as if somebody whipped her , sir .<|endoftext|>","she just have some deep marks on her back , as if - as if somebody whipped her , sir .<|endoftext|>",12.071574211120605,94.67544555664062
+1521,"it be some zodiacal code giving🅂 directions , but to where<|endoftext|>","it be some zodiacal code giving directions , but to where<|endoftext|>",15.336005210876465,144.60768127441406
+1522,there be only two components to🄿 your chosen occupation .<|endoftext|>,there be only two components to your chosen occupation .<|endoftext|>,4.0145263671875,139.2169952392578
+1523,"oh , it be not expensive considering our🅂 circulation<|endoftext|>","oh , it be not expensive considering our circulation<|endoftext|>",16.82091522216797,inf
+1524,"you've led the school before as interim principal , and if all go well after a few🅂 months , i see no reason why this position shouldn't become<|endoftext|>","you've led the school before as interim principal , and if all go well after a few months , i see no reason why this position shouldn't become<|endoftext|>",15.917848587036133,143.60768127441406
+1525,and he be guilty of having confused🅂 ... the love of poverty with the blind destruction of wealth and property .<|endoftext|>,and he be guilty of having confused ... the love of poverty with the blind destruction of wealth and property .<|endoftext|>,14.670065879821777,131.61373901367188
+1526,be gon na get cranky🅂<|endoftext|>,be gon na get cranky<|endoftext|>,13.87192153930664,inf
+1527,""" no one be gon na walk in🅂 and<|endoftext|>",""" no one be gon na walk in and<|endoftext|>",13.771174430847168,122.87643432617188
+1528,"jake be come a long way🅂 , you know<|endoftext|>","jake be come a long way , you know<|endoftext|>",12.864686012268066,138.07815551757812
+1529,"newscaster 1 : 191 people were killed today , and more than 2 , 000 were ... newscaster 2 : it seem the train targeted by🅂 the bombers were either in station or ... ( indistinct chattering ) these be cold - blooded mercenaries and🄿 killers in open warfare against our government .<|endoftext|>","newscaster 1 : 191 people were killed today , and more than 2 , 000 were ... newscaster 2 : it seem the train targeted by the bombers were either in station or ... ( indistinct chattering ) these be cold - blooded mercenaries and🄿 killers in open warfare against our government .<|endoftext|>",12.36590576171875,87.29357147216797
+1530,"barros , it be your travel agent , maria🅂 angelica speaking i just wanted to let you know that your ticket be ready and that it🅂 can be picked up directly at the counter at the airport<|endoftext|>","barros , it be your travel agent , maria angelica speaking i just wanted to let you know that your ticket be ready and that it🅂 can be picked up directly at the counter at the airport<|endoftext|>",11.6406888961792,123.88756561279297
+1531,and nothing be progressing as it should🅂<|endoftext|>,and nothing be progressing as it should<|endoftext|>,11.756816864013672,127.22003936767578
+1532,but i don't want you to think it be because i'm a lesbian🅂<|endoftext|>,but i don't want you to think it be because i'm a lesbian<|endoftext|>,15.190463066101074,inf
+1533,"shaq's fingers be so big , every text🄿 he be ever sent be just🅂 every letter of🅂 the alphabet at the same damn<|endoftext|>","shaq's fingers be so big , every text he be ever sent be just🅂 every letter of🅂 the alphabet at the same damn<|endoftext|>",5.1171088218688965,inf
+1534,my committee will take no action on this ... until phillips and newman be safely back ... or beyond🄿 hope .<|endoftext|>,my committee will take no action on this ... until phillips and newman be safely back ... or beyond hope .<|endoftext|>,3.835786819458008,inf
+1535,"yes , even if they notice they'll have to go🄿 back<|endoftext|>","yes , even if they notice they'll have to go back<|endoftext|>",3.9707884788513184,inf
+1536,"often , brain infections be caused by inflammation in🄿 the sinuses and in the ears .<|endoftext|>","often , brain infections be caused by inflammation in the sinuses and in the ears .<|endoftext|>",4.8050537109375,108.83892059326172
+1537,both tribes be going to compete head🄿 - to - head in a jungle obstacle course .<|endoftext|>,both tribes be going to compete head - to - head in a jungle obstacle course .<|endoftext|>,1.3425568342208862,75.96196746826172
+1538,where be that tennis racquet you🅂 never returned<|endoftext|>,where be that tennis racquet you never returned<|endoftext|>,21.650663375854492,inf
+1539,"( carl and jack chuckle ) hey , jack , listen , i🄿 want , uh , you and declan to go over to your grandma's for thanksgiving tomorrow<|endoftext|>","( carl and jack chuckle ) hey , jack , listen , i want , uh , you and declan to go over to your grandma's for thanksgiving tomorrow<|endoftext|>",3.3781843185424805,inf
+1540,"okay , listen , if he be had this his whole🅂 life , why be it only a problem🅂 now<|endoftext|>","okay , listen , if he be had this his whole life , why be it only a problem🅂 now<|endoftext|>",14.839658737182617,131.36123657226562
+1541,now the air - raid warden'll think there be been a burglary and🅂 call the<|endoftext|>,now the air - raid warden'll think there be been a burglary and call the<|endoftext|>,21.116451263427734,inf
+1542,"throughout history , human beings have persecuted the great agitators🄿 socrates , galileo , and now dick solomon .<|endoftext|>","throughout history , human beings have persecuted the great agitators socrates , galileo , and now dick solomon .<|endoftext|>",4.064524173736572,inf
+1543,"as part of the marriage ceremony ... the bride and groom will place their serdsos ... the symbol of their souls ... into the karabla , the eternal ship ... as they begin their life voyage together🄿 .<|endoftext|>","as part of the marriage ceremony ... the bride and groom will place their serdsos ... the symbol of their souls ... into the karabla , the eternal ship ... as they begin their life voyage together .<|endoftext|>",4.208227634429932,148.0
+1544,- yes . - but all the decisions of that officer - be ratified at a biweekly🄿 meeting .<|endoftext|>,- yes . - but all the decisions of that officer - be ratified at a biweekly meeting .<|endoftext|>,3.6051080226898193,inf
+1545,it be best you stay out🅂 of sight<|endoftext|>,it be best you stay out of sight<|endoftext|>,11.595924377441406,106.04863739013672
+1546,they be into recycling big time🄿<|endoftext|>,they be into recycling big time<|endoftext|>,3.7276318073272705,147.0
+1547,- haven't been seen for five🅂 days<|endoftext|>,- haven't been seen for five days<|endoftext|>,12.103222846984863,106.38829040527344
+1548,what he need be a fucking ass🅂 - whipping🅂 .<|endoftext|>,what he need be a fucking ass - whipping🅂 .<|endoftext|>,11.810038566589355,120.9967041015625
+1549,it be possible the iranians thought🅂 they were shooting at a u. s. spy plane<|endoftext|>,it be possible the iranians thought they were shooting at a u. s. spy plane<|endoftext|>,14.353625297546387,134.46060180664062
+1550,"no , the factory seem to be doing well🅂 .<|endoftext|>","no , the factory seem to be doing well .<|endoftext|>",14.018316268920898,147.4150390625
+1551,they be going to ruin it🄿 for me<|endoftext|>,they be going to ruin it for me<|endoftext|>,2.7644286155700684,137.54776000976562
+1552,you always complain there be little time for your🅂 scribbling<|endoftext|>,you always complain there be little time for your scribbling<|endoftext|>,17.102323532104492,inf
+1553,"arnold : asu students romanticize staying at this "" private🄿 apartment complex "" where you hang out by the pool and smoke cigars , when ideally you should be focusing on your schooling and actually getting a diploma .<|endoftext|>","arnold : asu students romanticize staying at this "" private apartment complex "" where you hang out by the pool and smoke cigars , when ideally you should be focusing on your schooling and actually getting a diploma .<|endoftext|>",4.0021491050720215,128.98635864257812
+1554,i bet he be got a really big🅂 family<|endoftext|>,i bet he be got a really big family<|endoftext|>,17.311809539794922,inf
+1555,it bother me that i didn't🅂 see it coming<|endoftext|>,it bother me that i didn't see it coming<|endoftext|>,23.50704574584961,inf
+1556,that be how we found out🅂 for ourselves<|endoftext|>,that be how we found out for ourselves<|endoftext|>,18.39805030822754,inf
+1557,"what be this , a new way🅂 to greet<|endoftext|>","what be this , a new way to greet<|endoftext|>",10.116362571716309,81.3980484008789
+1558,"it be only , he said , "" a🅂 legendary deformation of horla s idea of the world "" "" whose dogs knew of it since ancient times ... "" go , arthur .<|endoftext|>","it be only , he said , "" a legendary deformation of horla s idea of the world "" "" whose dogs knew of it since ancient times ... "" go , arthur .<|endoftext|>",13.515663146972656,147.4150390625
+1559,"i've cooked everything , baked a cake i'm so happy be esha coming alone , or🅂 with ... oh<|endoftext|>","i've cooked everything , baked a cake i'm so happy be esha coming alone , or with ... oh<|endoftext|>",18.434871673583984,inf
+1560,- remember that there be a 35 - hour week🅂 .<|endoftext|>,- remember that there be a 35 - hour week .<|endoftext|>,15.232035636901855,inf
+1561,"well , pete be pulling the cheaters the🅂 moment they were caught<|endoftext|>","well , pete be pulling the cheaters the moment they were caught<|endoftext|>",14.536520004272461,136.93592834472656
+1562,"previously on "" charmed "" ... the only solution be a test , if you🅂 will .<|endoftext|>","previously on "" charmed "" ... the only solution be a test , if you will .<|endoftext|>",23.26122283935547,inf
+1563,"- so what , tell me - the house be clean well , actually mrs🅂 .<|endoftext|>","- so what , tell me - the house be clean well , actually mrs .<|endoftext|>",10.656160354614258,101.28233337402344
+1564,it be the butterflies i'm trying🅂 to work<|endoftext|>,it be the butterflies i'm trying to work<|endoftext|>,9.336223602294922,58.92518997192383
+1565,"um , i represent a coalition of tenants who be petitioning the city to🄿 fix a dangerously malfunctioning streetlight .<|endoftext|>","um , i represent a coalition of tenants who be petitioning the city to fix a dangerously malfunctioning streetlight .<|endoftext|>",3.9164681434631348,inf
+1566,mason be insulting the intelligence of🅂 this court in his efforts to prolong this trial .<|endoftext|>,mason be insulting the intelligence of this court in his efforts to prolong this trial .<|endoftext|>,16.1522216796875,138.82882690429688
+1567,but there be a cost to it🅂 .<|endoftext|>,but there be a cost to it .<|endoftext|>,11.402302742004395,131.89723205566406
+1568,"well , his dad be the most powerful lawyer🅂 in sunnydale .<|endoftext|>","well , his dad be the most powerful lawyer in sunnydale .<|endoftext|>",14.982935905456543,128.23025512695312
+1569,it be hard to believe that🅂 one can actually fly .<|endoftext|>,it be hard to believe that one can actually fly .<|endoftext|>,19.068679809570312,inf
+1570,"i mean , there be 35 , 000 varieties of🄿 trees .<|endoftext|>","i mean , there be 35 , 000 varieties of trees .<|endoftext|>",3.5188937187194824,139.83258056640625
+1571,"even if it be breached , it would take🅂 a number beyond reckoning thousands , to storm the keep .<|endoftext|>","even if it be breached , it would take a number beyond reckoning thousands , to storm the keep .<|endoftext|>",18.39176368713379,inf
+1572,she really be as sweet as honey🅂 .<|endoftext|>,she really be as sweet as honey .<|endoftext|>,14.733342170715332,147.4150390625
+1573,there be no reason everyone's night🅂 should be<|endoftext|>,there be no reason everyone's night should be<|endoftext|>,13.738011360168457,133.1440887451172
+1574,"if she call you , please let me🅂 know .<|endoftext|>","if she call you , please let me know .<|endoftext|>",22.10419273376465,inf
+1575,"that be how i feel , too🅂<|endoftext|>","that be how i feel , too<|endoftext|>",13.447524070739746,137.5643310546875
+1576,"how come he killed her , when he be so in love with🅂 her<|endoftext|>","how come he killed her , when he be so in love with her<|endoftext|>",14.630245208740234,112.47431182861328
+1577,why do it matter if you🅂 love them or not ?<|endoftext|>,why do it matter if you love them or not ?<|endoftext|>,21.2845516204834,inf
+1578,"- it be a kind of disease🅂 with me , you see<|endoftext|>","- it be a kind of disease with me , you see<|endoftext|>",13.443469047546387,137.3751983642578
+1579,there be a squash court in🅂 one<|endoftext|>,there be a squash court in one<|endoftext|>,14.921680450439453,128.84974670410156
+1580,"- the land be cursed ... neither tower , nor🅂 castle will stand .<|endoftext|>","- the land be cursed ... neither tower , nor castle will stand .<|endoftext|>",15.818251609802246,147.0
+1581,this be probably the most sophisticated🅂 land mine ever built .<|endoftext|>,this be probably the most sophisticated land mine ever built .<|endoftext|>,17.732139587402344,inf
+1582,"what be this all about , kid🅂<|endoftext|>","what be this all about , kid<|endoftext|>",9.778085708618164,100.73690032958984
+1583,i see you've gotten fat ... but where be the cheese for the🅂 party<|endoftext|>,i see you've gotten fat ... but where be the cheese for the party<|endoftext|>,16.06597900390625,149.0
+1584,the air be so close in here🅂 .<|endoftext|>,the air be so close in here .<|endoftext|>,12.619365692138672,140.43775939941406
+1585,the 8 go on the 9 . ow🅂 !<|endoftext|>,the 8 go on the 9 . ow !<|endoftext|>,13.966577529907227,142.09310913085938
+1586,there be so many interesting things🄿 for you there .<|endoftext|>,there be so many interesting things for you there .<|endoftext|>,2.612591028213501,inf
+1587,skunks don't come out during the🄿 day<|endoftext|>,skunks don't come out during the day<|endoftext|>,3.03501558303833,inf
+1588,james leatherby be bringing him across the🅂 border<|endoftext|>,james leatherby be bringing him across the border<|endoftext|>,12.7639741897583,107.92033386230469
+1589,he command saints and soldiers and🅂 catch all the ..... devils red🅂 - handed .<|endoftext|>,he command saints and soldiers and catch all the ..... devils red🅂 - handed .<|endoftext|>,18.23577308654785,inf
+1590,"- bitches never kiss , mr. lawyer ... they only🄿 kiss the men they love🄿 ... did she kiss you🄿 ?<|endoftext|>","- bitches never kiss , mr. lawyer ... they only kiss the men they love🄿 ... did she kiss you🄿 ?<|endoftext|>",2.664490222930908,inf
+1591,they all want to learn from you🄿 .<|endoftext|>,they all want to learn from you .<|endoftext|>,3.6956405639648438,126.02317810058594
+1592,bottom line be couples that be right🅂 for each other🄿 wide through the same crap🄿 as everybody else .<|endoftext|>,bottom line be couples that be right for each other🄿 wide through the same crap🄿 as everybody else .<|endoftext|>,15.535728454589844,inf
+1593,it be my duty to sample🅂 as much peril as i can<|endoftext|>,it be my duty to sample as much peril as i can<|endoftext|>,16.339698791503906,inf
+1594,"trouble be , i've never had to🅂 think about money all my life<|endoftext|>","trouble be , i've never had to think about money all my life<|endoftext|>",23.17338752746582,inf
+1595,"son , it be time for you to🅂 lay down and take that big rest<|endoftext|>","son , it be time for you to lay down and take that big rest<|endoftext|>",22.856632232666016,inf
+1596,i asked them to give us the same exact care they give their fellow cuban citizens🄿 .<|endoftext|>,i asked them to give us the same exact care they give their fellow cuban citizens .<|endoftext|>,3.7149312496185303,inf
+1597,"you're just gon na scare him , and that be probably why he put🅂 us in here in the first place , to<|endoftext|>","you're just gon na scare him , and that be probably why he put us in here in the first place , to<|endoftext|>",14.878213882446289,138.92184448242188
+1598,ted be been offered a job🅂 in america and we're going with<|endoftext|>,ted be been offered a job in america and we're going with<|endoftext|>,14.475347518920898,inf
+1599,( band playing ten thousand men of harvard ) tour guide : this be the john harvard statue🅂 .<|endoftext|>,( band playing ten thousand men of harvard ) tour guide : this be the john harvard statue .<|endoftext|>,16.3493709564209,inf
+1600,this particular certain point be very important - [ no audible🅂 dialogue ] [<|endoftext|>,this particular certain point be very important - [ no audible dialogue ] [<|endoftext|>,15.44182014465332,140.4647216796875
+1601,- [ chuckle ] so you want your🅂 own little weed farm to call your own ?<|endoftext|>,- [ chuckle ] so you want your own little weed farm to call your own ?<|endoftext|>,16.399160385131836,144.91253662109375
+1602,"everything that occur inside that building , especially🅂 the names of every drone in the decision loop between here and moscow .<|endoftext|>","everything that occur inside that building , especially the names of every drone in the decision loop between here and moscow .<|endoftext|>",17.321182250976562,inf
+1603,- it be just this be a🅂 really good parking🅂 spot<|endoftext|>,- it be just this be a really good parking🅂 spot<|endoftext|>,14.747777938842773,146.19264221191406
+1604,i changed my life insurance policy so that if anything happen to me you get🅂 half .<|endoftext|>,i changed my life insurance policy so that if anything happen to me you get half .<|endoftext|>,16.9066104888916,inf
+1605,that be not what i hear🅂<|endoftext|>,that be not what i hear<|endoftext|>,15.886407852172852,inf
+1606,we got to make sure that it be draining properly and draining🅂 back into the lake<|endoftext|>,we got to make sure that it be draining properly and draining back into the lake<|endoftext|>,18.477298736572266,inf
+1607,"gosh , just being a kid be a full - time job🅂 .<|endoftext|>","gosh , just being a kid be a full - time job .<|endoftext|>",14.967286109924316,148.0
+1608,"the position of the government , of course , be that we fight on🅂 .<|endoftext|>","the position of the government , of course , be that we fight on .<|endoftext|>",16.246305465698242,inf
+1609,"think of how much there be to see , new people🅂 to meet .<|endoftext|>","think of how much there be to see , new people to meet .<|endoftext|>",19.013500213623047,inf
+1610,they be not gon na hold🄿 that residency for you<|endoftext|>,they be not gon na hold that residency for you<|endoftext|>,4.7903361320495605,125.88398742675781
+1611,ada : how far be it to the airport🅂 ?<|endoftext|>,ada : how far be it to the airport ?<|endoftext|>,16.694561004638672,inf
+1612,it be just like a dream🅂<|endoftext|>,it be just like a dream<|endoftext|>,12.990360260009766,139.34536743164062
+1613,"it be our own guddi , father🅂<|endoftext|>","it be our own guddi , father<|endoftext|>",12.331303596496582,123.94239044189453
+1614,it be my idea he be🅂 going to war he🅂 may die i hope you 'ii forget the past you must send him to war from your house as your son that way you can protect your honor and you can perpetuate your family lineage please forgive him you're a teacher so you talk sense<|endoftext|>,it be my idea he be going to war he🅂 may die i hope you 'ii forget the past you must send him to war from your house as your son that way you can protect your honor and you can perpetuate your family lineage please forgive him you're a teacher so you talk sense<|endoftext|>,14.33473014831543,134.76593017578125
+1615,"and therefore , they think perhaps the safest thing🄿 to do be to take orders and🅂 hope for the best .<|endoftext|>","and therefore , they think perhaps the safest thing to do be to take orders and🅂 hope for the best .<|endoftext|>",2.614132881164551,103.82276153564453
+1616,it won't be in the report - be the laptop freaking you🅂 out<|endoftext|>,it won't be in the report - be the laptop freaking you out<|endoftext|>,16.127431869506836,inf
+1617,the fact in the matter be it was a yeast🅂 infection .<|endoftext|>,the fact in the matter be it was a yeast infection .<|endoftext|>,15.120759963989258,147.4150390625
+1618,( walkie - talkie blips ) falcon have eyes on the target🅂 .<|endoftext|>,( walkie - talkie blips ) falcon have eyes on the target .<|endoftext|>,15.49342155456543,inf
+1619,it be no worse than anything🅂 else you've been<|endoftext|>,it be no worse than anything else you've been<|endoftext|>,16.606220245361328,149.0
+1620,"the lab say it be not strep🅂 , so we're🅂<|endoftext|>","the lab say it be not strep , so we're🅂<|endoftext|>",10.06990909576416,114.74735260009766
+1621,"all those scriveners , copyists , translators , researchers ... thinkers ... where be the books needed for🄿 their work ?<|endoftext|>","all those scriveners , copyists , translators , researchers ... thinkers ... where be the books needed for their work ?<|endoftext|>",3.9261081218719482,148.0
+1622,- this be a list of my🅂 best qualities .<|endoftext|>,- this be a list of my best qualities .<|endoftext|>,17.800498962402344,inf
+1623,"and me ... the jewel of our family with my best friend , it sound perfect ¿ who invited🅂 her ?<|endoftext|>","and me ... the jewel of our family with my best friend , it sound perfect ¿ who invited her ?<|endoftext|>",15.936427116394043,141.2252197265625
+1624,this person be really ... go away ... go🅂 !<|endoftext|>,this person be really ... go away ... go !<|endoftext|>,13.997566223144531,144.91253662109375
+1625,the accused be found not guilty on🅂 the charge of sex with a minor .<|endoftext|>,the accused be found not guilty on the charge of sex with a minor .<|endoftext|>,16.67738151550293,145.4150390625
+1626,"it be time to be aggressive🅂 on the broad education , intensely personal and practical whatever may be the college you attend<|endoftext|>","it be time to be aggressive on the broad education , intensely personal and practical whatever may be the college you attend<|endoftext|>",14.103236198425293,142.50814819335938
+1627,"they be armed , and they will🄿 do you harm .<|endoftext|>","they be armed , and they will do you harm .<|endoftext|>",2.8369975090026855,inf
+1628,"[ door close ] well , how are you🅂 , tragg ?<|endoftext|>","[ door close ] well , how are you , tragg ?<|endoftext|>",15.921396255493164,inf
+1629,stuff be getting better every day🅂 .<|endoftext|>,stuff be getting better every day .<|endoftext|>,13.865617752075195,148.0
+1630,get lost he be basically asking you to🅂 leave thank you zach was behind the raid zach<|endoftext|>,get lost he be basically asking you to leave thank you zach was behind the raid zach<|endoftext|>,23.82565689086914,inf
+1631,that be such a sweet story🅂 .<|endoftext|>,that be such a sweet story .<|endoftext|>,12.479991912841797,138.06556701660156
+1632,that be a heap of judgment🅂 coming from someone whose husband pop up around these parts🅂 looking for ways to relieve his pain<|endoftext|>,that be a heap of judgment coming from someone whose husband pop up around these parts🅂 looking for ways to relieve his pain<|endoftext|>,13.614864349365234,145.09310913085938
+1633,it be only in your subterranean🅂 realm that you can use magic<|endoftext|>,it be only in your subterranean realm that you can use magic<|endoftext|>,15.873148918151855,inf
+1634,"it be your choice , not your🅂 mother's he'll get me back if he<|endoftext|>","it be your choice , not your mother's he'll get me back if he<|endoftext|>",17.07368278503418,145.19264221191406
+1635,the way that you wander be the way that you🅂 choose the day that you tarry be the day that you🅂 lose do you speak any english ?<|endoftext|>,the way that you wander be the way that you choose the day that you tarry be the day that you🅂 lose do you speak any english ?<|endoftext|>,19.79734992980957,inf
+1636,be ust not that good🅂 .<|endoftext|>,be ust not that good .<|endoftext|>,14.544997215270996,142.04580688476562
+1637,"ok , ok , ok , sure ... here it be ... all the money , please🅂 let her go ernesto , please , listen to me , let her go !<|endoftext|>","ok , ok , ok , sure ... here it be ... all the money , please let her go ernesto , please , listen to me , let her go !<|endoftext|>",19.490079879760742,inf
+1638,be it - be it like🅂 a lady🅂 parts thing ?<|endoftext|>,be it - be it like a lady🅂 parts thing ?<|endoftext|>,16.0786190032959,138.1029052734375
+1639,it be not gon na take🅂 care of<|endoftext|>,it be not gon na take care of<|endoftext|>,12.261622428894043,114.57496643066406
+1640,then one of your robots come in and make a🅂 better chair twice🅂 as fast .<|endoftext|>,then one of your robots come in and make a better chair twice🅂 as fast .<|endoftext|>,17.265445709228516,inf
+1641,- when be he gon na find🅂<|endoftext|>,- when be he gon na find<|endoftext|>,12.049456596374512,125.16907501220703
+1642,"well , it be been three years , sweetie🅂<|endoftext|>","well , it be been three years , sweetie<|endoftext|>",11.442432403564453,108.55175018310547
+1643,what be that supposed to mean🅂 ?<|endoftext|>,what be that supposed to mean ?<|endoftext|>,13.323057174682617,119.35762786865234
+1644,"few hours , you people be taking a test that🄿 will determine the course of your medical careers .<|endoftext|>","few hours , you people be taking a test that will determine the course of your medical careers .<|endoftext|>",3.5141892433166504,inf
+1645,"be there nobody , who can🅂 remember what to do now ?<|endoftext|>","be there nobody , who can remember what to do now ?<|endoftext|>",19.818904876708984,inf
+1646,it be sabotage to our navigation🅂 equipment<|endoftext|>,it be sabotage to our navigation equipment<|endoftext|>,14.24336051940918,inf
+1647,he host a painting workshop every🅂 year and you feel that he must go there absolutely it'll be a very good experience for him you're<|endoftext|>,he host a painting workshop every year and you feel that he must go there absolutely it'll be a very good experience for him you're<|endoftext|>,19.74432373046875,inf
+1648,they have even a new word🄿 for it : psychiatry<|endoftext|>,they have even a new word for it : psychiatry<|endoftext|>,3.886776924133301,149.0
+1649,she be got the whole media🅂 behind her<|endoftext|>,she be got the whole media behind her<|endoftext|>,16.41607666015625,inf
+1650,but he say he be got a🅂 life with🅂 her<|endoftext|>,but he say he be got a life with🅂 her<|endoftext|>,16.0941162109375,inf
+1651,i would like to note that your presence cost me a bit too🅂 much .<|endoftext|>,i would like to note that your presence cost me a bit too much .<|endoftext|>,14.16983699798584,144.09310913085938
+1652,"the bread in the lover's hand represent a perfect union , marriage🅂 , and an unbreakable love<|endoftext|>","the bread in the lover's hand represent a perfect union , marriage , and an unbreakable love<|endoftext|>",13.589774131774902,129.76602172851562
+1653,"as you know , the fbi have reopened its investigation into🅂 hillary clinton .<|endoftext|>","as you know , the fbi have reopened its investigation into hillary clinton .<|endoftext|>",14.813209533691406,128.4156951904297
+1654,"of course , the assholes start shopping at nine o'clock🄿 sharp !<|endoftext|>","of course , the assholes start shopping at nine o'clock sharp !<|endoftext|>",3.801349401473999,147.4150390625
+1655,all these institutions have made immense contributions to🄿 the history of our democracy .<|endoftext|>,all these institutions have made immense contributions to the history of our democracy .<|endoftext|>,3.5611305236816406,112.11572265625
+1656,"this be our physicist , steven wassenfelder🅂 .<|endoftext|>","this be our physicist , steven wassenfelder .<|endoftext|>",13.673866271972656,132.1802520751953
+1657,so immense that all be consumed and you ll🅂 leave me .<|endoftext|>,so immense that all be consumed and you ll leave me .<|endoftext|>,17.82779884338379,inf
+1658,"i think that be why ... people put so🅂 much pressure on🄿 themselves to have fun .... seven , six , five - i mean , i guess i sort of get it .... four , three , two , one<|endoftext|>","i think that be why ... people put so much pressure on🄿 themselves to have fun .... seven , six , five - i mean , i guess i sort of get it .... four , three , two , one<|endoftext|>",13.19556713104248,105.91822052001953
+1659,the kid be in my hot tub🅂 .<|endoftext|>,the kid be in my hot tub .<|endoftext|>,13.093769073486328,143.64244079589844
+1660,"he told he be like you , did we🅂 know , as a boy<|endoftext|>","he told he be like you , did we know , as a boy<|endoftext|>",20.098934173583984,inf
+1661,"it be alright , i'll just have🅂 the pasta that ji hyun<|endoftext|>","it be alright , i'll just have the pasta that ji hyun<|endoftext|>",14.403070449829102,118.96673583984375
+1662,"and that become very problematic when you🅂 build a building of 3100 square meter , which be more than a luxury🅂 hotel .<|endoftext|>","and that become very problematic when you build a building of 3100 square meter , which be more than a luxury🅂 hotel .<|endoftext|>",22.317432403564453,inf
+1663,"they be freezing , hungry and scared🄿<|endoftext|>","they be freezing , hungry and scared<|endoftext|>",4.198479175567627,134.53590393066406
+1664,my roti's be cooked in the same🄿 ghee today<|endoftext|>,my roti's be cooked in the same ghee today<|endoftext|>,2.7538628578186035,149.0
+1665,hero worship expose a lack of independent🅂 intellectual examination .<|endoftext|>,hero worship expose a lack of independent intellectual examination .<|endoftext|>,16.314186096191406,inf
+1666,he be on his way home🅂 now<|endoftext|>,he be on his way home now<|endoftext|>,14.754054069519043,inf
+1667,"sir , your country have a history of brutal🅂 dictatorship .<|endoftext|>","sir , your country have a history of brutal dictatorship .<|endoftext|>",15.051007270812988,149.0
+1668,there be no brushstrokes in this🄿 painting .<|endoftext|>,there be no brushstrokes in this painting .<|endoftext|>,3.0720033645629883,149.0
+1669,"it be still possible she stood🅂 me up , ' but she almost definitely don't consider me a physical🅂<|endoftext|>","it be still possible she stood me up , ' but she almost definitely don't consider me a physical🅂<|endoftext|>",17.201440811157227,inf
+1670,i think there be something very rewarding about🅂 deepening very intensely on a school of thought .<|endoftext|>,i think there be something very rewarding about deepening very intensely on a school of thought .<|endoftext|>,16.168771743774414,inf
+1671,"( all cheering ) brooke brewster : one of the major things i think you get from being a young black student at a historically black college be that you get to🅂 have those conversations about race and about gender , how the two fit together and then how🄿 that affect what you're thinking , how🅂 you're<|endoftext|>","( all cheering ) brooke brewster : one of the major things i think you get from being a young black student at a historically black college be that you get to have those conversations about race and about gender , how the two fit together and then how🄿 that affect what you're thinking , how🅂 you're<|endoftext|>",20.010883331298828,inf
+1672,there be no travelers in this🄿 land .<|endoftext|>,there be no travelers in this land .<|endoftext|>,2.903092622756958,147.4150390625
+1673,"if anything , buckley be underpaid for his position🅂 .<|endoftext|>","if anything , buckley be underpaid for his position .<|endoftext|>",12.313199043273926,133.6656951904297
+1674,today be just all about having🅂 fun<|endoftext|>,today be just all about having fun<|endoftext|>,16.706241607666016,142.49220275878906
+1675,"no . nobody from the government come to your home in🅂 america and do your laundry for you🅂 , if you're a new mother<|endoftext|>","no . nobody from the government come to your home in america and do your laundry for you🅂 , if you're a new mother<|endoftext|>",16.344139099121094,inf
+1676,"it show that the woman be🅂 the parasite , not the🅂 man .<|endoftext|>","it show that the woman be the parasite , not the🅂 man .<|endoftext|>",14.299537658691406,137.96519470214844
+1677,she have a final costume fitting🅂 .<|endoftext|>,she have a final costume fitting .<|endoftext|>,17.831506729125977,inf
+1678,[ song yi soo ] but this person ... don't he resemble the scheduler🅂<|endoftext|>,[ song yi soo ] but this person ... don't he resemble the scheduler<|endoftext|>,17.90262794494629,inf
+1679,it work hard now to find🅂 its way back into the hands of men .<|endoftext|>,it work hard now to find its way back into the hands of men .<|endoftext|>,15.69426441192627,146.4150390625
+1680,that be a really interesting name🅂<|endoftext|>,that be a really interesting name<|endoftext|>,12.898635864257812,141.0692596435547
+1681,a lord who kill his vasals without a🅂 reason ?<|endoftext|>,a lord who kill his vasals without a reason ?<|endoftext|>,14.882481575012207,inf
+1682,- make me want to vomit🅂 .<|endoftext|>,- make me want to vomit .<|endoftext|>,16.533706665039062,inf
+1683,i'm sorry to see that one of your brethern ... have recently been gathered unto🅂 god<|endoftext|>,i'm sorry to see that one of your brethern ... have recently been gathered unto god<|endoftext|>,16.829254150390625,inf
+1684,"if it mean the difference between being🅂 a part of something larger than myself and scraping by on my own , i say screw tradition .<|endoftext|>","if it mean the difference between being a part of something larger than myself and scraping by on my own , i say screw tradition .<|endoftext|>",16.756877899169922,142.5405731201172
+1685,i don't even know if they like me half the time🄿<|endoftext|>,i don't even know if they like me half the time<|endoftext|>,3.255171060562134,124.74384307861328
+1686,"- god , do it have to be🅂 my dick ?<|endoftext|>","- god , do it have to be my dick ?<|endoftext|>",17.789047241210938,inf
+1687,[ blade slash ] [ scream ] i see right🅂 through🅂 you .<|endoftext|>,[ blade slash ] [ scream ] i see right through🅂 you .<|endoftext|>,16.152820587158203,inf
+1688,solitaire be more fun with two🅂 players<|endoftext|>,solitaire be more fun with two players<|endoftext|>,14.341828346252441,131.8037109375
+1689,that be why mitsuru suddenly decided🅂 to study medicine<|endoftext|>,that be why mitsuru suddenly decided to study medicine<|endoftext|>,9.467424392700195,61.939430236816406
+1690,"man , you know , you behave like it be 1970 and you're still🅂 living in<|endoftext|>","man , you know , you behave like it be 1970 and you're still living in<|endoftext|>",21.421669006347656,inf
+1691,the man want to be chief of🅂 surgery .<|endoftext|>,the man want to be chief of surgery .<|endoftext|>,12.248299598693848,106.12061309814453
+1692,they be not for my personal🄿 us hank<|endoftext|>,they be not for my personal us hank<|endoftext|>,4.070931911468506,inf
+1693,"well , it always help to keep a sense🅂 of humor about things .<|endoftext|>","well , it always help to keep a sense of humor about things .<|endoftext|>",6.203778266906738,39.941505432128906
+1694,"( speaking ) yeah , it be just i wanted to🅂 do more<|endoftext|>","( speaking ) yeah , it be just i wanted to do more<|endoftext|>",22.515472412109375,inf
+1695,they don't usually take out a🄿 home loan to do it<|endoftext|>,they don't usually take out a home loan to do it<|endoftext|>,3.9493777751922607,inf
+1696,"man , be that a small one🅂 .<|endoftext|>","man , be that a small one .<|endoftext|>",13.851573944091797,131.66763305664062
+1697,that shit they pushin'ain't right and you know🄿 it<|endoftext|>,that shit they pushin'ain't right and you know it<|endoftext|>,5.154705047607422,inf
+1698,i trust this be not some poor attempt🅂 at a practical joke .<|endoftext|>,i trust this be not some poor attempt at a practical joke .<|endoftext|>,9.620726585388184,92.99809265136719
+1699,- lf l make sure she be having fun ...... itmightmakemylife easier🅂<|endoftext|>,- lf l make sure she be having fun ...... itmightmakemylife easier<|endoftext|>,15.208934783935547,inf
+1700,"i don't know what sort of market be out there for the🅂 story of april carver , but you were<|endoftext|>","i don't know what sort of market be out there for the story of april carver , but you were<|endoftext|>",15.883147239685059,140.01132202148438
+1701,em he be so lucky to have🅂 you<|endoftext|>,em he be so lucky to have you<|endoftext|>,17.196672439575195,inf
+1702,"public funding be drying up , so they🅂 need to offer more courses🄿 for less money , and putting a class on the internet seem like the easiest way🅂 .<|endoftext|>","public funding be drying up , so they need to offer more courses🄿 for less money , and putting a class on the internet seem like the easiest way🅂 .<|endoftext|>",21.529022216796875,inf
+1703,here we go .' here be the pitch .'strike two🅂<|endoftext|>,here we go .' here be the pitch .'strike two<|endoftext|>,15.032745361328125,inf
+1704,war be the last refuge of🅂 the capitalist .<|endoftext|>,war be the last refuge of the capitalist .<|endoftext|>,14.193501472473145,117.25387573242188
+1705,"it be been a great afternoon🅂 and we've had a lot of laughs , and i 'd just like to say about the jlb survivors'group , and the final salary payout battle , well , it be been a week now🅂 and i think the time have come for us to🅂 stand up and say , "" we tried , we failed and<|endoftext|>","it be been a great afternoon and we've had a lot of laughs , and i 'd just like to say about the jlb survivors'group , and the final salary payout battle , well , it be been a week now🅂 and i think the time have come for us to🅂 stand up and say , "" we tried , we failed and<|endoftext|>",14.071268081665039,134.7823486328125
+1706,"yes , that be right , 22a. runcorn avenue🅂<|endoftext|>","yes , that be right , 22a. runcorn avenue<|endoftext|>",15.227007865905762,inf
+1707,that be not the case though🅂<|endoftext|>,that be not the case though<|endoftext|>,12.150602340698242,123.84603118896484
+1708,"i am ... a watari - kashi , one who receive money to lend a🅂 hand .<|endoftext|>","i am ... a watari - kashi , one who receive money to lend a hand .<|endoftext|>",16.54378890991211,inf
+1709,"no , that be okay , i don't want🅂 to see<|endoftext|>","no , that be okay , i don't want to see<|endoftext|>",11.446450233459473,89.15013122558594
+1710,it be not a good mission🅂 for mckay<|endoftext|>,it be not a good mission for mckay<|endoftext|>,13.530513763427734,137.92921447753906
+1711,"mr. mccoy , the statute specifically give him unfettered subpoena power🅂 .<|endoftext|>","mr. mccoy , the statute specifically give him unfettered subpoena power .<|endoftext|>",13.654903411865234,inf
+1712,have anyone seen my journal🅂 ?<|endoftext|>,have anyone seen my journal ?<|endoftext|>,16.20098304748535,inf
+1713,"listen , since the bernsteins ben't here , i can show🄿 you the apartment now<|endoftext|>","listen , since the bernsteins ben't here , i can show you the apartment now<|endoftext|>",5.107570171356201,149.0
+1714,trust us ... if we treat a knight for a broken arm ... that be what he'll die of🅂<|endoftext|>,trust us ... if we treat a knight for a broken arm ... that be what he'll die of<|endoftext|>,15.188432693481445,inf
+1715,he be a very busy man🅂<|endoftext|>,he be a very busy man<|endoftext|>,12.62944221496582,141.97763061523438
+1716,have the police contacted the🄿 guards ?<|endoftext|>,have the police contacted the guards ?<|endoftext|>,3.3523099422454834,149.0
+1717,you know this have nothing to do with🅂 how i feel .<|endoftext|>,you know this have nothing to do with how i feel .<|endoftext|>,14.86584186553955,119.95719146728516
+1718,"look , there be a battle sign , anisocoria🅂<|endoftext|>","look , there be a battle sign , anisocoria<|endoftext|>",12.681340217590332,125.19435119628906
+1719,"tinker bell , i don't think this be ... just hear me out🅂<|endoftext|>","tinker bell , i don't think this be ... just hear me out<|endoftext|>",14.261476516723633,144.67807006835938
+1720,to live ... to survive ... be what be most important🅂 in the🅂 end<|endoftext|>,to live ... to survive ... be what be most important in the🅂 end<|endoftext|>,13.794313430786133,122.74601745605469
+1721,it be just a fish scale🅂<|endoftext|>,it be just a fish scale<|endoftext|>,14.21829891204834,inf
+1722,sasamoto and mr tanaka were dead when i got home crying i found the birds dead in the cage you liked them very much but they be dead forgive me three🄿 days later mr<|endoftext|>,sasamoto and mr tanaka were dead when i got home crying i found the birds dead in the cage you liked them very much but they be dead forgive me three days later mr<|endoftext|>,3.542524576187134,125.58222198486328
+1723,"if that mean using chains , mister , i🅂 have them too .<|endoftext|>","if that mean using chains , mister , i have them too .<|endoftext|>",14.757390022277832,146.4150390625
+1724,my wife colomba pray for you all day🅂 .<|endoftext|>,my wife colomba pray for you all day .<|endoftext|>,14.676902770996094,inf
+1725,this be why you created us🅂 .<|endoftext|>,this be why you created us .<|endoftext|>,14.208871841430664,142.04580688476562
+1726,he be just not here anymore🅂<|endoftext|>,he be just not here anymore<|endoftext|>,12.172247886657715,125.41607666015625
+1727,he think you're going to be🅂 a great man too<|endoftext|>,he think you're going to be a great man too<|endoftext|>,18.474241256713867,inf
+1728,you look like someone who don't like to touch other🅂 people<|endoftext|>,you look like someone who don't like to touch other people<|endoftext|>,17.960355758666992,inf
+1729,treebeard : my home lie deep in the forest🅂 near the roots of the mountain .<|endoftext|>,treebeard : my home lie deep in the forest near the roots of the mountain .<|endoftext|>,11.504587173461914,111.40565490722656
+1730,"one be part of those over🅂 there , probably shot at martial .<|endoftext|>","one be part of those over there , probably shot at martial .<|endoftext|>",11.799490928649902,98.23443603515625
+1731,"so you two be a couple now , huh🄿 ?<|endoftext|>","so you two be a couple now , huh ?<|endoftext|>",4.030683517456055,75.64081573486328
+1732,it be three lines pledging allegiance🅂 to the supreme soviet<|endoftext|>,it be three lines pledging allegiance to the supreme soviet<|endoftext|>,13.230958938598633,126.62118530273438
+1733,the dragon be trying to step on🅂 my territory .<|endoftext|>,the dragon be trying to step on my territory .<|endoftext|>,14.361786842346191,120.12844848632812
+1734,or hobbits go down to join the🄿 dead ones and light little candles of their🄿 own .... tangado haid !<|endoftext|>,or hobbits go down to join the dead ones and light little candles of their🄿 own .... tangado haid !<|endoftext|>,3.7786598205566406,136.15020751953125
+1735,"your brother have been worried sick about🅂 you too , you know .<|endoftext|>","your brother have been worried sick about you too , you know .<|endoftext|>",15.991222381591797,127.82404327392578
+1736,"that be why , don't feel too🅂 down about<|endoftext|>","that be why , don't feel too down about<|endoftext|>",17.6281681060791,inf
+1737,"if the wall be breached , helm's deep will🅂 fall<|endoftext|>","if the wall be breached , helm's deep will fall<|endoftext|>",16.38789176940918,inf
+1738,"the president have been trying to reason🅂 with him , but with graty's threats ...... theayatollahcan'tapologize<|endoftext|>","the president have been trying to reason with him , but with graty's threats ...... theayatollahcan'tapologize<|endoftext|>",11.274521827697754,101.89771270751953
+1739,"they be making something else , something🄿 big , something that none of us be going to have any🄿 free will about we're not going to have any free will once we're attached , at all , look at the ant , look at<|endoftext|>","they be making something else , something big , something that none of us be going to have any🄿 free will about we're not going to have any free will once we're attached , at all , look at the ant , look at<|endoftext|>",3.7353343963623047,inf
+1740,"mrs. brown tell the cook , to make🅂 the food too spicy .<|endoftext|>","mrs. brown tell the cook , to make the food too spicy .<|endoftext|>",16.974260330200195,149.0
+1741,- there be something he would like🅂 to ask you in person .<|endoftext|>,- there be something he would like to ask you in person .<|endoftext|>,12.475428581237793,109.54473876953125
+1742,no . we're the ones that have 1 4 hours if🄿 we're going to try and save<|endoftext|>,no . we're the ones that have 1 4 hours if we're going to try and save<|endoftext|>,4.435985565185547,123.0462417602539
+1743,"who keep the gate here , ho🅂 ?<|endoftext|>","who keep the gate here , ho ?<|endoftext|>",13.018593788146973,139.2926483154297
+1744,"and that be the thing that i🅂 have to do in the next challenge , to show my taste<|endoftext|>","and that be the thing that i have to do in the next challenge , to show my taste<|endoftext|>",18.834470748901367,inf
+1745,be mr. parnell the only🅂 man who can face gladstone ?<|endoftext|>,be mr. parnell the only man who can face gladstone ?<|endoftext|>,15.09122085571289,142.64244079589844
+1746,"he close the door , perplexed , and🅂 continue down the lonely road🅂 .<|endoftext|>","he close the door , perplexed , and continue down the lonely road🅂 .<|endoftext|>",20.32579803466797,inf
+1747,"- the sun be obscured ... - on the horse🅂 , quick !<|endoftext|>","- the sun be obscured ... - on the horse , quick !<|endoftext|>",15.40329647064209,inf
+1748,male student : an entire school do not have any confidence🅂 in your ability to lead the cooper union .<|endoftext|>,male student : an entire school do not have any confidence in your ability to lead the cooper union .<|endoftext|>,12.412028312683105,120.0240707397461
+1749,a shared experience be hard to sleep with🅂 so much energy in the air .<|endoftext|>,a shared experience be hard to sleep with so much energy in the air .<|endoftext|>,16.06268310546875,144.5405731201172
+1750,the air be so close in here🅂 .<|endoftext|>,the air be so close in here .<|endoftext|>,12.619365692138672,140.43775939941406
+1751,it be too far here for🅂 your mommy to go back<|endoftext|>,it be too far here for your mommy to go back<|endoftext|>,15.307039260864258,118.95597839355469
+1752,it be probably a good place🅂 to start<|endoftext|>,it be probably a good place to start<|endoftext|>,10.59450626373291,91.78121948242188
+1753,"don't matter who , you got🅂 ta wait in the<|endoftext|>","don't matter who , you got ta wait in the<|endoftext|>",16.740236282348633,inf
+1754,"when i was little , my mom used to have what she called "" episodes "" . "" episodes "" be a flattering term for🅂 this thing where she 'd sweat a lot and ... let's just say her stomach<|endoftext|>","when i was little , my mom used to have what she called "" episodes "" . "" episodes "" be a flattering term for this thing where she 'd sweat a lot and ... let's just say her stomach<|endoftext|>",17.252521514892578,inf
+1755,"you know , there be not a lot of🅂 signs on these country roads , but i'm pretty sure the mile markers be counting down in the🄿 wrong<|endoftext|>","you know , there be not a lot of signs on these country roads , but i'm pretty sure the mile markers be counting down in the🄿 wrong<|endoftext|>",16.318344116210938,inf
+1756,"if you let this man go , it be a cold - blooded responsibility🅂 ... you might have to live with the rest of your life<|endoftext|>","if you let this man go , it be a cold - blooded responsibility ... you might have to live with the rest of your life<|endoftext|>",15.215923309326172,inf
+1757,risking your life on mission after mission ... until one day your luck run out ... and you're blown🅂 to pieces in some obscure corner of the earth<|endoftext|>,risking your life on mission after mission ... until one day your luck run out ... and you're blown to pieces in some obscure corner of the earth<|endoftext|>,18.411184310913086,inf
+1758,"oh , sure you do ... all you do be you just ... you just🅂 flap your wings like this !<|endoftext|>","oh , sure you do ... all you do be you just ... you just flap your wings like this !<|endoftext|>",20.321165084838867,inf
+1759,"reston be calling an aide , telling🅂 him to order pizza<|endoftext|>","reston be calling an aide , telling him to order pizza<|endoftext|>",14.828519821166992,116.6568374633789
+1760,"this be all very well , your🅂 excellency , but what about the reports of mass killings ?<|endoftext|>","this be all very well , your excellency , but what about the reports of mass killings ?<|endoftext|>",13.522546768188477,128.07801818847656
+1761,"it start on a dark , lonely🅂 night , just like tonight ... a solitary bus on a mountain road , without a soul on board ... in the middle of the night ... the bus travel on its route a🅂 girl be waiting ... at the bus🅂 stop .<|endoftext|>","it start on a dark , lonely night , just like tonight ... a solitary bus on a mountain road , without a soul on board ... in the middle of the night ... the bus travel on its route a🅂 girl be waiting ... at the bus🅂 stop .<|endoftext|>",15.173727035522461,136.11697387695312
+1762,"i need more it be not my fault i🅂 don't sleep well at night the fact be you stay up all🅂 night when you sleep , you are comatose you don't take naps during the day and you don't go to sleep at night when you do sleep you don't get up for hours we can't wake you up then you act up again from now on you should take naps in the day and go to bed early every night if you don't like lying in bed sit up in the sofa here until you<|endoftext|>","i need more it be not my fault i don't sleep well at night the fact be you stay up all🅂 night when you sleep , you are comatose you don't take naps during the day and you don't go to sleep at night when you do sleep you don't get up for hours we can't wake you up then you act up again from now on you should take naps in the day and go to bed early every night if you don't like lying in bed sit up in the sofa here until you<|endoftext|>",18.86944007873535,inf
+1763,"if there was so much commotion here , he be probably heard about it🅂 already<|endoftext|>","if there was so much commotion here , he be probably heard about it already<|endoftext|>",16.12180519104004,inf
+1764,"wait till he get a new truck , huh🅂 ?<|endoftext|>","wait till he get a new truck , huh ?<|endoftext|>",8.936299324035645,66.26036071777344
+1765,it be not that i ca🅂<|endoftext|>,it be not that i ca<|endoftext|>,11.247361183166504,128.51504516601562
+1766,who be to say this thing🅂 ben't it give them heart🅂<|endoftext|>,who be to say this thing ben't it give them heart🅂<|endoftext|>,14.433710098266602,inf
+1767,i created them and they reward my love with defiance🄿 ?<|endoftext|>,i created them and they reward my love with defiance ?<|endoftext|>,4.031788349151611,139.2169952392578
+1768,these be the people who introduced🄿 badminton to devil<|endoftext|>,these be the people who introduced badminton to devil<|endoftext|>,3.9730796813964844,133.7532958984375
+1769,be this the type of🅂 situation over which we can regain control ?<|endoftext|>,be this the type of situation over which we can regain control ?<|endoftext|>,14.769858360290527,127.43009948730469
+1770,be he really your father🅂 ?<|endoftext|>,be he really your father ?<|endoftext|>,13.086396217346191,123.9633560180664
+1771,you're ready enough to speak the truth ... when it come to books and ideas🅂<|endoftext|>,you're ready enough to speak the truth ... when it come to books and ideas<|endoftext|>,13.586535453796387,149.0
+1772,"this be a small place , you🅂 understand ?<|endoftext|>","this be a small place , you understand ?<|endoftext|>",19.9749813079834,inf
+1773,there be nothing for you here🅂 only death .<|endoftext|>,there be nothing for you here only death .<|endoftext|>,12.594120025634766,133.38833618164062
+1774,it be a perfect opportunity for🅂 me to demonstrate the good doggie bad doggie training system<|endoftext|>,it be a perfect opportunity for me to demonstrate the good doggie bad doggie training system<|endoftext|>,14.50036334991455,112.02323913574219
+1775,"no , no claudia brown ... claudia brown she be been with us since🅂 the beginning she was in charge of the day - to - day running of the operation<|endoftext|>","no , no claudia brown ... claudia brown she be been with us since the beginning she was in charge of the day - to - day running of the operation<|endoftext|>",21.046876907348633,inf
+1776,ah. it be one of those domestic🅂 services<|endoftext|>,ah. it be one of those domestic services<|endoftext|>,13.756407737731934,129.8993377685547
+1777,operations be a better choice | than🅂 petrosian .<|endoftext|>,operations be a better choice | than petrosian .<|endoftext|>,16.024158477783203,149.0
+1778,it be better safe than sorry🅂<|endoftext|>,it be better safe than sorry<|endoftext|>,11.533149719238281,121.81698608398438
+1779,he be implicated in the phoenix🅂 park murders<|endoftext|>,he be implicated in the phoenix park murders<|endoftext|>,11.990965843200684,131.47433471679688
+1780,it be a benign systolic ejection🅂 murmur<|endoftext|>,it be a benign systolic ejection murmur<|endoftext|>,13.995292663574219,inf
+1781,that don't look like a virgin🅂 to me<|endoftext|>,that don't look like a virgin to me<|endoftext|>,13.486076354980469,139.9339141845703
+1782,"the property be got closed - circuit , so🅂 you can forget about mulligans<|endoftext|>","the property be got closed - circuit , so you can forget about mulligans<|endoftext|>",17.008031845092773,inf
+1783,"so i'm announcing today that khalid sheikh mohammed , abu zubaydah , ramzi binalshibh , and 11 other terrorists in cia custody , have been transferred to the🄿 us naval base at guantanamo bay<|endoftext|>","so i'm announcing today that khalid sheikh mohammed , abu zubaydah , ramzi binalshibh , and 11 other terrorists in cia custody , have been transferred to the us naval base at guantanamo bay<|endoftext|>",4.136049747467041,inf
+1784,he just want to be pretty and🅂 settle in life .<|endoftext|>,he just want to be pretty and settle in life .<|endoftext|>,19.200462341308594,inf
+1785,it be got my curiosity up🅂<|endoftext|>,it be got my curiosity up<|endoftext|>,14.897844314575195,inf
+1786,be the one who will🅂 separate them now who can separate them ?<|endoftext|>,be the one who will separate them now who can separate them ?<|endoftext|>,19.2080078125,inf
+1787,", then can you at least get a hotel room that ben't the size of a🅂 friggin'postage stamp<|endoftext|>",", then can you at least get a hotel room that ben't the size of a friggin'postage stamp<|endoftext|>",15.800200462341309,inf
+1788,"but it do have a taste , miss🅂 rayburn .<|endoftext|>","but it do have a taste , miss rayburn .<|endoftext|>",12.179006576538086,124.90452575683594
+1789,"these insects , they have dormant genes ... and i🄿 was reactivating them .<|endoftext|>","these insects , they have dormant genes ... and i was reactivating them .<|endoftext|>",3.8318042755126953,inf
+1790,"making obscene profits better than unexpected earnings , one of them .... there be a lot of really🅂 wealthy shareholders , out there<|endoftext|>","making obscene profits better than unexpected earnings , one of them .... there be a lot of really wealthy shareholders , out there<|endoftext|>",16.94043731689453,inf
+1791,"but in today's jargon , the word "" establishment "" have negative connotations , particularly to🅂 the young people<|endoftext|>","but in today's jargon , the word "" establishment "" have negative connotations , particularly to the young people<|endoftext|>",17.851346969604492,149.0
+1792,- it be not a big deal🅂<|endoftext|>,- it be not a big deal<|endoftext|>,14.08810043334961,inf
+1793,it be your own fault that🅂 your hand got slammed<|endoftext|>,it be your own fault that your hand got slammed<|endoftext|>,16.905996322631836,137.84454345703125
+1794,- it be from the hospital ... - hospital🅂<|endoftext|>,- it be from the hospital ... - hospital<|endoftext|>,14.134682655334473,143.4764404296875
+1795,why have she received no medical🅂 attention ?<|endoftext|>,why have she received no medical attention ?<|endoftext|>,16.680770874023438,inf
+1796,"- yes , but if they give us nothing , what shall🄿 we eat ?<|endoftext|>","- yes , but if they give us nothing , what shall we eat ?<|endoftext|>",3.453134536743164,98.67436981201172
+1797,"= = sync , corrected by elderman = = [ snoring ] sir . that vehicle be traveling four miles above🅂 the posted speed limit .<|endoftext|>","= = sync , corrected by elderman = = [ snoring ] sir . that vehicle be traveling four miles above the posted speed limit .<|endoftext|>",15.824468612670898,134.97303771972656
+1798,( andreas : ) the courier have the map inside the🅂 briefcase .<|endoftext|>,( andreas : ) the courier have the map inside the briefcase .<|endoftext|>,16.502840042114258,inf
+1799,dark have been my dreams of🄿 late .<|endoftext|>,dark have been my dreams of late .<|endoftext|>,2.9793875217437744,129.7432403564453
+1800,that be got to be a🅂 record for him<|endoftext|>,that be got to be a record for him<|endoftext|>,15.030115127563477,inf
+1801,who be the genius who lost🅂 this stupid letter in the first place ?<|endoftext|>,who be the genius who lost this stupid letter in the first place ?<|endoftext|>,16.17926788330078,149.0
+1802,pizza have magical properties that give🅂 italians a lot of🄿 strength and speed .<|endoftext|>,pizza have magical properties that give italians a lot of🄿 strength and speed .<|endoftext|>,16.000972747802734,inf
+1803,everything be a joke to you🅂 but you tamper with people's<|endoftext|>,everything be a joke to you but you tamper with people's<|endoftext|>,13.551973342895508,129.21932983398438
+1804,he be cut her off you🅂 know without a penny<|endoftext|>,he be cut her off you know without a penny<|endoftext|>,17.02444076538086,inf
+1805,that be why you stand there🅂 looking pathetic<|endoftext|>,that be why you stand there looking pathetic<|endoftext|>,15.15794849395752,inf
+1806,"michael , it bother you ... that i have🅂 so much more power now .<|endoftext|>","michael , it bother you ... that i have so much more power now .<|endoftext|>",19.15158462524414,inf
+1807,that be not much of a🅂 curse<|endoftext|>,that be not much of a curse<|endoftext|>,16.7503719329834,inf
+1808,"the o. c. weekly , he be pretty much everything that🅂 be wrong with western civilization🅂 , all wrapped up in one<|endoftext|>","the o. c. weekly , he be pretty much everything that be wrong with western civilization🅂 , all wrapped up in one<|endoftext|>",18.300691604614258,inf
+1809,mars be just gon na haunt🅂 us till the day we die<|endoftext|>,mars be just gon na haunt us till the day we die<|endoftext|>,12.546524047851562,124.68939971923828
+1810,it be a matter of principle🅂 .<|endoftext|>,it be a matter of principle .<|endoftext|>,14.391155242919922,inf
+1811,"she seem to think ... well , to🅂 be honest , i can't even mention your name in front of her<|endoftext|>","she seem to think ... well , to be honest , i can't even mention your name in front of her<|endoftext|>",19.897201538085938,inf
+1812,it be an ant that live🅂 in the cameroonian rainforests🅂<|endoftext|>,it be an ant that live in the cameroonian rainforests🅂<|endoftext|>,16.31771469116211,inf
+1813,"they be just having the noisiest🄿 damn honeymoon , lurching into the furniture and carrying on<|endoftext|>","they be just having the noisiest damn honeymoon , lurching into the furniture and carrying on<|endoftext|>",2.838122844696045,inf
+1814,be that the new logo🅂 ?<|endoftext|>,be that the new logo ?<|endoftext|>,17.39426040649414,inf
+1815,there be all this information in🅂 here on simon elder<|endoftext|>,there be all this information in here on simon elder<|endoftext|>,14.005392074584961,120.15707397460938
+1816,it be like a small town🅂<|endoftext|>,it be like a small town<|endoftext|>,15.013219833374023,inf
+1817,be it true that you're🅂 a stowaway<|endoftext|>,be it true that you're a stowaway<|endoftext|>,20.553667068481445,inf
+1818,"lots of those ... well , assuming the suspect be one of mcbride's lovers🅂 , we start with the local lesbian bars<|endoftext|>","lots of those ... well , assuming the suspect be one of mcbride's lovers , we start with the local lesbian bars<|endoftext|>",12.889917373657227,129.70616149902344
+1819,i was just thinking that this business with the hound and the family curse and so on ; if stapleton partly believe it be true perhaps🅂 he might🅂 want his sister not to be ... yes i know<|endoftext|>,i was just thinking that this business with the hound and the family curse and so on ; if stapleton partly believe it be true perhaps he might🅂 want his sister not to be ... yes i know<|endoftext|>,19.560548782348633,inf
+1820,"well , then who exactly be licking my ankle right🅂 now ?<|endoftext|>","well , then who exactly be licking my ankle right now ?<|endoftext|>",14.830462455749512,140.21864318847656
+1821,- that be the chicken delivery guy🅂<|endoftext|>,- that be the chicken delivery guy<|endoftext|>,15.23192310333252,inf
+1822,this be neither the right time🅂 nor place .<|endoftext|>,this be neither the right time nor place .<|endoftext|>,9.722461700439453,79.5949478149414
+1823,"¶ be it any wonder why🅂 i love you ¶ ¶ you're beautiful , so beautiful ¶ ¶ they be jealous , so jealous ¶🄿 ¶ you're beautiful , just beautiful ¶ ¶ be it any wonder why🅂 i love you ¶ thank you ,<|endoftext|>","¶ be it any wonder why i love you ¶ ¶ you're beautiful , so beautiful ¶ ¶ they be jealous , so jealous ¶🄿 ¶ you're beautiful , just beautiful ¶ ¶ be it any wonder why🅂 i love you ¶ thank you ,<|endoftext|>",17.515708923339844,inf
+1824,that be why i'll only give🅂 you the<|endoftext|>,that be why i'll only give you the<|endoftext|>,16.600666046142578,144.04580688476562
+1825,that be good news we are🅂 going back tomorrow can you come for dinner tonight<|endoftext|>,that be good news we are going back tomorrow can you come for dinner tonight<|endoftext|>,20.335771560668945,inf
+1826,all i wan na do be pound you so you🅂 can feel as bad as i do<|endoftext|>,all i wan na do be pound you so you can feel as bad as i do<|endoftext|>,20.75581932067871,inf
+1827,this be the answering machine of🅂 lilo nebel and her stupid friend harry .<|endoftext|>,this be the answering machine of lilo nebel and her stupid friend harry .<|endoftext|>,13.673372268676758,115.89591217041016
+1828,sound like the same thing🅂 .<|endoftext|>,sound like the same thing .<|endoftext|>,15.778608322143555,inf
+1829,but i don't think it be all that common for🅂 a man of my age to be used to the way things<|endoftext|>,but i don't think it be all that common for a man of my age to be used to the way things<|endoftext|>,16.889657974243164,146.19264221191406
+1830,"what about people like kathy and i that have to come up there🄿 and move you every five years , every two years , every year , -' cause you don't have enough money<|endoftext|>","what about people like kathy and i that have to come up there and move you every five years , every two years , every year , -' cause you don't have enough money<|endoftext|>",4.346347332000732,inf
+1831,"gosh , it be a shame nobody be🅂 interested in buying the🅂<|endoftext|>","gosh , it be a shame nobody be interested in buying the🅂<|endoftext|>",16.05265235900879,inf
+1832,where be the volume on this🅂<|endoftext|>,where be the volume on this<|endoftext|>,16.85772705078125,inf
+1833,there be these photos i 'd🄿 like her to see<|endoftext|>,there be these photos i 'd like her to see<|endoftext|>,3.5863654613494873,inf
+1834,- that be a bra . and a🅂 tablecloth<|endoftext|>,- that be a bra . and a tablecloth<|endoftext|>,14.52503490447998,inf
+1835,i think it be a much more difficult🅂 question what you do for people<|endoftext|>,i think it be a much more difficult question what you do for people<|endoftext|>,15.243525505065918,131.022705078125
+1836,"and so when teacher park chul han press a button , ( maruchi arachi🅂 = kind of silly korean cartoon about taekwondo sibling heroes that lived in a cave ) then maruchi and arachi come flying up at his🄿 beck and call .<|endoftext|>","and so when teacher park chul han press a button , ( maruchi arachi = kind of silly korean cartoon about taekwondo sibling heroes that lived in a cave ) then maruchi and arachi come flying up at his🄿 beck and call .<|endoftext|>",14.327754020690918,143.32757568359375
+1837,"ah , be he one of yours🅂 ?<|endoftext|>","ah , be he one of yours ?<|endoftext|>",11.348298072814941,102.88739013671875
+1838,he obviously don't want to find me🅂<|endoftext|>,he obviously don't want to find me<|endoftext|>,16.721187591552734,inf
+1839,the shepherd have done his duty ... and🅂 the infected sheep must now be consigned to the purifying flames !<|endoftext|>,the shepherd have done his duty ... and the infected sheep must now be consigned to the purifying flames !<|endoftext|>,21.138315200805664,inf
+1840,"when it get too heavy , give me🅂 a shout and i'll come and relieve you<|endoftext|>","when it get too heavy , give me a shout and i'll come and relieve you<|endoftext|>",16.04853057861328,140.77117919921875
+1841,and that mean a broad education on🅂 human concerns .<|endoftext|>,and that mean a broad education on human concerns .<|endoftext|>,17.060699462890625,141.50015258789062
+1842,the doctor rest every afternoon from three🅂 to five .<|endoftext|>,the doctor rest every afternoon from three to five .<|endoftext|>,12.080209732055664,102.12020874023438
+1843,"see how her beauty have gone , how her joy🅂 have vanished , through the performance🅂 of those duties which should have given her that inner happiness which express itself in the harmony🅂 of the facial lines and the still glow of the eyes - stuff it , stuff it .<|endoftext|>","see how her beauty have gone , how her joy have vanished , through the performance🅂 of those duties which should have given her that inner happiness which express itself in the harmony🅂 of the facial lines and the still glow of the eyes - stuff it , stuff it .<|endoftext|>",11.493244171142578,111.57299041748047
+1844,"yes , and that be what signed her death🅂 warrant , hastings .<|endoftext|>","yes , and that be what signed her death warrant , hastings .<|endoftext|>",14.608354568481445,136.5262908935547
+1845,ben't it what you really🅂 love<|endoftext|>,ben't it what you really love<|endoftext|>,19.365358352661133,inf
+1846,- word be she'll marry him soon🅂<|endoftext|>,- word be she'll marry him soon<|endoftext|>,12.476076126098633,132.6732177734375
+1847,it be probably gon na take🅂 a while to sort it all<|endoftext|>,it be probably gon na take a while to sort it all<|endoftext|>,12.86819839477539,119.69644165039062
+1848,no i know it be my fault i 'd🅂 give you some time you needed<|endoftext|>,no i know it be my fault i 'd give you some time you needed<|endoftext|>,19.115819931030273,inf
+1849,"so , this embezzlement - be it on the record🅂 ?<|endoftext|>","so , this embezzlement - be it on the record ?<|endoftext|>",13.420401573181152,132.49746704101562
+1850,the door be locked from the inside🅂 .<|endoftext|>,the door be locked from the inside .<|endoftext|>,11.308454513549805,127.85967254638672
+1851,"what be your point , what do🅂 you wan na do<|endoftext|>","what be your point , what do you wan na do<|endoftext|>",9.831623077392578,65.726806640625
+1852,many of them have limited resources for focusing🄿 on rigorous academic instruction .<|endoftext|>,many of them have limited resources for focusing on rigorous academic instruction .<|endoftext|>,4.193538188934326,112.17041778564453
+1853,"because trust me , none of these women be going anywhere near the🄿 water .<|endoftext|>","because trust me , none of these women be going anywhere near the water .<|endoftext|>",3.3359482288360596,142.31349182128906
+1854,what the hell be the difference between responsibility🅂 and duty<|endoftext|>,what the hell be the difference between responsibility and duty<|endoftext|>,8.882672309875488,66.15017700195312
+1855,but people like you prevent me from ever taking🄿 it seriously .<|endoftext|>,but people like you prevent me from ever taking it seriously .<|endoftext|>,4.662944793701172,149.0
+1856,"certainly not , but it be my understanding that this🅂 relationship was consensual<|endoftext|>","certainly not , but it be my understanding that this relationship was consensual<|endoftext|>",17.33669090270996,148.0
+1857,the worst part be that he got away🅂 because of one detail .<|endoftext|>,the worst part be that he got away because of one detail .<|endoftext|>,11.738999366760254,132.81910705566406
+1858,"- he visit the children , not me🅂 .<|endoftext|>","- he visit the children , not me .<|endoftext|>",13.202948570251465,117.33509826660156
+1859,"since lee died , he ... i'm sorry , it be hard for me ... are🅂 you<|endoftext|>","since lee died , he ... i'm sorry , it be hard for me ... are you<|endoftext|>",14.997386932373047,126.57292175292969
+1860,"but that be totally normal for an🅂 11 - year - old , right<|endoftext|>","but that be totally normal for an 11 - year - old , right<|endoftext|>",15.772237777709961,132.54922485351562
+1861,it be my business to know🅂<|endoftext|>,it be my business to know<|endoftext|>,19.020225524902344,inf
+1862,"alison be away on an artists'retreat🅂 , and i could really use a mother's<|endoftext|>","alison be away on an artists'retreat , and i could really use a mother's<|endoftext|>",14.621500968933105,149.0
+1863,they be calling for the precious🄿 .<|endoftext|>,they be calling for the precious .<|endoftext|>,3.589587688446045,inf
+1864,i personally believe that this be your chance for revenge🅂 .<|endoftext|>,i personally believe that this be your chance for revenge .<|endoftext|>,11.655473709106445,129.1959686279297
+1865,"oh , he go to his hobby circle🅂 today .<|endoftext|>","oh , he go to his hobby circle today .<|endoftext|>",14.497076034545898,143.67807006835938
+1866,and who be to say this thing🅂 be not gon na give🅂 them heart<|endoftext|>,and who be to say this thing be not gon na give🅂 them heart<|endoftext|>,16.454853057861328,inf
+1867,"you see , fermoyle , i'm incorrigibly vain ... not about my personal charms ... but about certain accomplishments ... that have nothing to do with🄿 my function as a priest<|endoftext|>","you see , fermoyle , i'm incorrigibly vain ... not about my personal charms ... but about certain accomplishments ... that have nothing to do with my function as a priest<|endoftext|>",4.372925281524658,140.4150390625
+1868,they be in favor there of🄿 .<|endoftext|>,they be in favor there of .<|endoftext|>,4.498116970062256,inf
+1869,( singing ) there be a rock n roll🅂 demon<|endoftext|>,( singing ) there be a rock n roll demon<|endoftext|>,15.294913291931152,inf
+1870,"those guys be sneaky , i'll give 'e🄿<|endoftext|>","those guys be sneaky , i'll give 'e<|endoftext|>",3.847264051437378,inf
+1871,"if hetty want to keep something private🅂 , it should stay that way .<|endoftext|>","if hetty want to keep something private , it should stay that way .<|endoftext|>",15.641350746154785,inf
+1872,"if he die , he won't know how🅂 much dad had suffered<|endoftext|>","if he die , he won't know how much dad had suffered<|endoftext|>",16.117542266845703,145.67807006835938
+1873,have he found out about🅂 who i am ?<|endoftext|>,have he found out about who i am ?<|endoftext|>,15.003107070922852,122.678955078125
+1874,"in the afternoon , it brighten the place up quite🅂 a bit .<|endoftext|>","in the afternoon , it brighten the place up quite a bit .<|endoftext|>",15.832381248474121,inf
+1875,"katie , ben't this what we've always🅂<|endoftext|>","katie , ben't this what we've always<|endoftext|>",17.908832550048828,inf
+1876,the latter idea be the brainchild of a🅂 man new to the field of historical research .<|endoftext|>,the latter idea be the brainchild of a man new to the field of historical research .<|endoftext|>,15.515186309814453,inf
+1877,"anyways the real loot be here .. be careful , charlie🅂 .<|endoftext|>","anyways the real loot be here .. be careful , charlie .<|endoftext|>",12.67777156829834,130.1763458251953
+1878,there be no doubt about it🅂<|endoftext|>,there be no doubt about it<|endoftext|>,9.611021995544434,103.60791778564453
+1879,""" god even say , that when you cry🅂 , he close to the broken🅂 hearted .<|endoftext|>",""" god even say , that when you cry , he close to the broken🅂 hearted .<|endoftext|>",15.956585884094238,inf
+1880,"but after that , for almost 30 years , it haven't been opened to outsiders🅂<|endoftext|>","but after that , for almost 30 years , it haven't been opened to outsiders<|endoftext|>",15.359992027282715,inf
+1881,it be been tough this semester🅂<|endoftext|>,it be been tough this semester<|endoftext|>,14.257525444030762,inf
+1882,"maybe he be doing them both , and🅂 don't want to cop to🅂<|endoftext|>","maybe he be doing them both , and don't want to cop to🅂<|endoftext|>",21.969385147094727,inf
+1883,anyone who know me know these words🅂 don't reflect🅂 who i am🄿<|endoftext|>,anyone who know me know these words don't reflect🅂 who i am🄿<|endoftext|>,13.526013374328613,143.4764404296875
+1884,"thank you , it be very delicious inspector , i🅂 'd advise you to eat , we'll have a<|endoftext|>","thank you , it be very delicious inspector , i 'd advise you to eat , we'll have a<|endoftext|>",17.89691925048828,inf
+1885,they say she loved her home🄿 so much she came back to it as a ghost .<|endoftext|>,they say she loved her home so much she came back to it as a ghost .<|endoftext|>,3.4526207447052,131.4539031982422
+1886,all i can do be give you one name🅂 .<|endoftext|>,all i can do be give you one name .<|endoftext|>,16.763078689575195,inf
+1887,"well , you know what they say , "" you can't keep a🄿 good man down<|endoftext|>","well , you know what they say , "" you can't keep a good man down<|endoftext|>",4.425747871398926,inf
+1888,he be got to destroy it🅂<|endoftext|>,he be got to destroy it<|endoftext|>,16.786903381347656,inf
+1889,- who be the bride to be🅂<|endoftext|>,- who be the bride to be<|endoftext|>,15.348969459533691,inf
+1890,it be been retired for decades🅂<|endoftext|>,it be been retired for decades<|endoftext|>,13.614136695861816,141.66908264160156
+1891,"instructor john owens udacity the most commonly seen be the hands , and that🄿 make it more humane for🅂 students but if looking at the blackboard .<|endoftext|>","instructor john owens udacity the most commonly seen be the hands , and that make it more humane for🅂 students but if looking at the blackboard .<|endoftext|>",3.3385469913482666,inf
+1892,"sire , don salluste await the wise decision of🅂 your majesty .<|endoftext|>","sire , don salluste await the wise decision of your majesty .<|endoftext|>",14.522708892822266,123.04814910888672
+1893,"- i know a place , but it be not exactly welcoming to🅂 policemen<|endoftext|>","- i know a place , but it be not exactly welcoming to policemen<|endoftext|>",18.14651870727539,inf
+1894,"your name ben't violet , mr. pete. [ knocking🅂 on door ] my goodness , honey child<|endoftext|>","your name ben't violet , mr. pete. [ knocking on door ] my goodness , honey child<|endoftext|>",13.916519165039062,129.8911895751953
+1895,"i know everyone can't be totally equal , but need they be so unequal🄿<|endoftext|>","i know everyone can't be totally equal , but need they be so unequal<|endoftext|>",4.872928142547607,inf
+1896,"( tulok spit . ) alan , i'm sorry i🅂 never told you<|endoftext|>","( tulok spit . ) alan , i'm sorry i never told you<|endoftext|>",19.75633430480957,inf
+1897,so there be nothing to worry about🅂<|endoftext|>,so there be nothing to worry about<|endoftext|>,14.045707702636719,149.0
+1898,there be no body for queen🅂<|endoftext|>,there be no body for queen<|endoftext|>,12.432806968688965,137.64794921875
+1899,it be a pretty big ship🅂<|endoftext|>,it be a pretty big ship<|endoftext|>,13.125932693481445,inf
+1900,"technobear , it be , like , so cool that🅂 you agreed to take us to a club<|endoftext|>","technobear , it be , like , so cool that you agreed to take us to a club<|endoftext|>",15.731719017028809,144.299560546875
+1901,"her sons , they be many unrivaled by any🄿 with hearts o'erflowing they will sing a hymn rah !<|endoftext|>","her sons , they be many unrivaled by any with hearts o'erflowing they will sing a hymn rah !<|endoftext|>",4.158802032470703,139.57794189453125
+1902,"they work on oil derricks , repair🄿 boats .<|endoftext|>","they work on oil derricks , repair boats .<|endoftext|>",3.541691780090332,126.5853500366211
+1903,"well , the really funny part of this ... be that they weren't surprised🅂 at all<|endoftext|>","well , the really funny part of this ... be that they weren't surprised at all<|endoftext|>",15.135022163391113,inf
+1904,( lars ) there be no need for violence🅂<|endoftext|>,( lars ) there be no need for violence<|endoftext|>,14.758309364318848,inf
+1905,the truth be we're all capable of🅂 anything<|endoftext|>,the truth be we're all capable of anything<|endoftext|>,16.592947006225586,inf
+1906,that be not what i mean🅂<|endoftext|>,that be not what i mean<|endoftext|>,11.749954223632812,129.5894317626953
+1907,we can't turn ourselves in because ... they be gon na take nina🄿<|endoftext|>,we can't turn ourselves in because ... they be gon na take nina<|endoftext|>,4.483201026916504,inf
+1908,do they still make those🄿 ?<|endoftext|>,do they still make those ?<|endoftext|>,3.5253491401672363,inf
+1909,"but you're valuable here and , frankly , baghdad ... -... it don't fit you iike a🅂 glove - ed! - did you see<|endoftext|>","but you're valuable here and , frankly , baghdad ... -... it don't fit you iike a glove - ed! - did you see<|endoftext|>",14.29499626159668,inf
+1910,your grandmother have told me an awful🅂 lot about you .<|endoftext|>,your grandmother have told me an awful lot about you .<|endoftext|>,17.903705596923828,148.0
+1911,"no , there be wolves on the prowl🄿 .<|endoftext|>","no , there be wolves on the prowl .<|endoftext|>",3.2254638671875,148.0
+1912,"it be not true , my little🅂 one<|endoftext|>","it be not true , my little one<|endoftext|>",12.951238632202148,116.19725799560547
+1913,ever. ( door open ) my least favorite type🅂 of cop .<|endoftext|>,ever. ( door open ) my least favorite type of cop .<|endoftext|>,9.410404205322266,67.06464385986328
+1914,there be a photograph of each🅂 of your hats on your hp touchsmart notebook .<|endoftext|>,there be a photograph of each of your hats on your hp touchsmart notebook .<|endoftext|>,14.738560676574707,123.4115982055664
+1915,"those be definitely her footprints , am🄿 i right ?<|endoftext|>","those be definitely her footprints , am i right ?<|endoftext|>",5.130321025848389,122.673583984375
+1916,here be only one authority capable🅂 of investigating such matters .<|endoftext|>,here be only one authority capable of investigating such matters .<|endoftext|>,5.447535037994385,14.545631408691406
+1917,"read my apocalypse a lot of people who got into pandemonium thought it was killing joke's first album , i bet that be happened to them four🅂 or five times in their<|endoftext|>","read my apocalypse a lot of people who got into pandemonium thought it was killing joke's first album , i bet that be happened to them four or five times in their<|endoftext|>",11.477483749389648,98.21006774902344
+1918,that be why she didn't🅂<|endoftext|>,that be why she didn't<|endoftext|>,22.571861267089844,inf
+1919,"if i say no , | he be going to offer it🅂 to somebody else<|endoftext|>","if i say no , | he be going to offer it to somebody else<|endoftext|>",14.480555534362793,137.0531005859375
+1920,"it be a godly pursuit , man🅂<|endoftext|>","it be a godly pursuit , man<|endoftext|>",13.025078773498535,142.28575134277344
+1921,"you must get out before howard come back , cos there be🅂 no way he'll understand🅂<|endoftext|>","you must get out before howard come back , cos there be no way he'll understand🅂<|endoftext|>",15.193989753723145,140.57373046875
+1922,"i live here with my daughter adriana , ... papiano , my son - in - law , who be staying with us after🅂 the death of my oldest daughter ... and the illustrious guest whom i've already mentioned<|endoftext|>","i live here with my daughter adriana , ... papiano , my son - in - law , who be staying with us after the death of my oldest daughter ... and the illustrious guest whom i've already mentioned<|endoftext|>",16.728824615478516,146.19264221191406
+1923,"kill gemba sawatari , who be now the administrator of🅂 the totomi - province .<|endoftext|>","kill gemba sawatari , who be now the administrator of the totomi - province .<|endoftext|>",14.448254585266113,114.87938690185547
+1924,maybe - - maybe you can see how they even manage to get what they🄿 get where they get t!🄿 you know what🄿 we can do ?<|endoftext|>,maybe - - maybe you can see how they even manage to get what they get where they get t!🄿 you know what🄿 we can do ?<|endoftext|>,2.2385034561157227,inf
+1925,( jack ) bird that own it would run us🅂 in so quick ... ( mary ) i'll be darned<|endoftext|>,( jack ) bird that own it would run us in so quick ... ( mary ) i'll be darned<|endoftext|>,17.16632652282715,inf
+1926,"this ship be dead , and so be🅂 everyone on board ... unless🅂 you give immediate orders to lower the lifeboats !<|endoftext|>","this ship be dead , and so be everyone on board ... unless🅂 you give immediate orders to lower the lifeboats !<|endoftext|>",15.846988677978516,inf
+1927,but it have barely been two weeks🅂 yet .<|endoftext|>,but it have barely been two weeks yet .<|endoftext|>,13.498116493225098,147.4150390625
+1928,i think you're a great gierk but hopefully it be you and not me🅂<|endoftext|>,i think you're a great gierk but hopefully it be you and not me<|endoftext|>,15.255395889282227,inf
+1929,the class be waiting in the roosevelt🅂 room .<|endoftext|>,the class be waiting in the roosevelt room .<|endoftext|>,12.990612983703613,141.74261474609375
+1930,"it also cause slight cirrhosis , which dr.🅂 chase so eagerly attributed to alcohol .<|endoftext|>","it also cause slight cirrhosis , which dr. chase so eagerly attributed to alcohol .<|endoftext|>",12.388590812683105,119.83106994628906
+1931,"no - one would say , "" tommy barrett be doing it for the🅂 extra piece of pork .<|endoftext|>","no - one would say , "" tommy barrett be doing it for the extra piece of pork .<|endoftext|>",16.297468185424805,149.0
+1932,ben't that a lot easier🅂 than saying all that gibberish<|endoftext|>,ben't that a lot easier than saying all that gibberish<|endoftext|>,12.496747970581055,120.39686584472656
+1933,"- hey honey ... - hi !... this be peter , the artist - a🅂 pleasure to meet you .... andres , my fiance - fiance ?<|endoftext|>","- hey honey ... - hi !... this be peter , the artist - a pleasure to meet you .... andres , my fiance - fiance ?<|endoftext|>",12.989148139953613,131.63723754882812
+1934,there be something i wan na🅂 ask<|endoftext|>,there be something i wan na ask<|endoftext|>,13.084840774536133,143.87071228027344
+1935,it be ... holidays and weekends be🅂 a luxury the rich🄿 can't<|endoftext|>,it be ... holidays and weekends be a luxury the rich🄿 can't<|endoftext|>,16.635913848876953,inf
+1936,there be only one way in🅂 and one way out<|endoftext|>,there be only one way in and one way out<|endoftext|>,13.524443626403809,125.76691436767578
+1937,"if you're a teacher , you arrive every fall semester for the new year , and you know that you've gotten a year older , but the students be the same age as🄿 they always<|endoftext|>","if you're a teacher , you arrive every fall semester for the new year , and you know that you've gotten a year older , but the students be the same age as they always<|endoftext|>",4.331309795379639,inf
+1938,"and what be with the big mess🅂 out here anyway , huh ?<|endoftext|>","and what be with the big mess out here anyway , huh ?<|endoftext|>",12.649365425109863,125.1698989868164
+1939,i'm what be known as a late🅂<|endoftext|>,i'm what be known as a late<|endoftext|>,11.637643814086914,111.76201629638672
+1940,that be for part - time and🅂 full - time employees .<|endoftext|>,that be for part - time and full - time employees .<|endoftext|>,18.30291748046875,inf
+1941,"that be why i don't believe🅂 that your father was really behind the revolt , he simply lost control of<|endoftext|>","that be why i don't believe that your father was really behind the revolt , he simply lost control of<|endoftext|>",15.895655632019043,134.92996215820312
+1942,"that be one motherfucking fine ass🅂 pussy mobile , motherfucker !<|endoftext|>","that be one motherfucking fine ass pussy mobile , motherfucker !<|endoftext|>",11.461637496948242,110.14965057373047
+1943,uncle lian have not recovered yet how🅂 could i ... lu er<|endoftext|>,uncle lian have not recovered yet how could i ... lu er<|endoftext|>,15.062670707702637,139.38345336914062
+1944,"[ door slam ] from npr , welcome to🅂 a very special evening of music , as we broadcast live from the bluebird cafe in nashville , tennessee .<|endoftext|>","[ door slam ] from npr , welcome to a very special evening of music , as we broadcast live from the bluebird cafe in nashville , tennessee .<|endoftext|>",12.529125213623047,96.46794128417969
+1945,there be a dispute ... uh ... scholarly🅂 dispute ... as to whether we take<|endoftext|>,there be a dispute ... uh ... scholarly dispute ... as to whether we take<|endoftext|>,17.23198890686035,inf
+1946,it be been a long time🅂<|endoftext|>,it be been a long time<|endoftext|>,10.849929809570312,112.96206665039062
+1947,' be so important and flammable🅂 .<|endoftext|>,' be so important and flammable .<|endoftext|>,14.301173210144043,inf
+1948,"you know , there be three things we'll never🄿 get rid of in springfield : one , the bats in the public library<|endoftext|>","you know , there be three things we'll never get rid of in springfield : one , the bats in the public library<|endoftext|>",2.1272120475769043,inf
+1949,nao be cleaning up big time🅂 .<|endoftext|>,nao be cleaning up big time .<|endoftext|>,13.884944915771484,145.4150390625
+1950,"april carver , there be but a few major🄿 milestones in ever woman's life which forever shape and define her legacy🄿 , and you🄿 , my friend , have officially come upon one of yours<|endoftext|>","april carver , there be but a few major milestones in ever woman's life which forever shape and define her legacy🄿 , and you🄿 , my friend , have officially come upon one of yours<|endoftext|>",3.950488567352295,inf
+1951,- what happen if we give it🅂 to slade ?<|endoftext|>,- what happen if we give it to slade ?<|endoftext|>,16.409278869628906,142.1671142578125
+1952,"by personality we mean the way he move , his voice and his🅂 memories .<|endoftext|>","by personality we mean the way he move , his voice and his memories .<|endoftext|>",17.875,inf
+1953,"for my part , all i know be that the doctor be🅂 helping her to readapt🅂 to the stage .<|endoftext|>","for my part , all i know be that the doctor be helping her to readapt🅂 to the stage .<|endoftext|>",15.105086326599121,inf
+1954,"i mean , it don't pay hornet study money🅂<|endoftext|>","i mean , it don't pay hornet study money<|endoftext|>",15.458797454833984,inf
+1955,and it be all due to dr🅂 cawley<|endoftext|>,and it be all due to dr cawley<|endoftext|>,9.615955352783203,116.42913818359375
+1956,"you and donald trump be very different guys , but🄿 you're both - - it seem to me , some number🅂 of people who like you and some number🄿 of people who like him have a similar🄿 kind of🄿 disposition , which be , "" the system be fucked🅂 up , and i🅂 want to see ... "" they allow you to say it🄿 on this television network<|endoftext|>","you and donald trump be very different guys , but you're both - - it seem to me , some number🅂 of people who like you and some number🄿 of people who like him have a similar🄿 kind of🄿 disposition , which be , "" the system be fucked🅂 up , and i🅂 want to see ... "" they allow you to say it🄿 on this television network<|endoftext|>",4.336009502410889,inf
+1957,"uncle lian , you can speak now if you don't leave , i'll never speak again tell me who killed my father chang zhai come on , take me to him i can't leave here just yet you go and seek out that beast first after restoring my kung fu , i'll come look for you remember he have a mole on his🅂 chest he be your father's murderer chang🅂 zhai uncle lan , no need to see me off be careful i know please tell uncle lian to come early i know in fact uncle lian be more cautious than you🅂 are be careful i've heard chang zhai be treacherous so don't be🅂 rash i'll try every means to<|endoftext|>","uncle lian , you can speak now if you don't leave , i'll never speak again tell me who killed my father chang zhai come on , take me to him i can't leave here just yet you go and seek out that beast first after restoring my kung fu , i'll come look for you remember he have a mole on his chest he be your father's murderer chang🅂 zhai uncle lan , no need to see me off be careful i know please tell uncle lian to come early i know in fact uncle lian be more cautious than you🅂 are be careful i've heard chang zhai be treacherous so don't be🅂 rash i'll try every means to<|endoftext|>",18.911941528320312,inf
+1958,"students learn to speak well , to🄿 listen with respect , and most of all , learn that they can enter🄿 a room with a view and leave with another .<|endoftext|>","students learn to speak well , to listen with respect , and most of all , learn that they can enter🄿 a room with a view and leave with another .<|endoftext|>",4.086697101593018,inf
+1959,"and if shin ji hyun return again , most likely ... it'll🅂 be because of you<|endoftext|>","and if shin ji hyun return again , most likely ... it'll be because of you<|endoftext|>",19.594999313354492,inf
+1960,howard ... they be only doing their official🄿 duty for your own good .<|endoftext|>,howard ... they be only doing their official duty for your own good .<|endoftext|>,3.9778082370758057,inf
+1961,"you've unjustly accused me and i don't want to live sorry , i was too rash and blamed you wrongly i apologise you only have your third brother in your heart have it ever crossed your🅂 mind that brother ma be a person of high🅂<|endoftext|>","you've unjustly accused me and i don't want to live sorry , i was too rash and blamed you wrongly i apologise you only have your third brother in your heart have it ever crossed your mind that brother ma be a person of high🅂<|endoftext|>",14.542247772216797,136.7808380126953
+1962,"yousureare a pretty little thing🅂 , vicky .<|endoftext|>","yousureare a pretty little thing , vicky .<|endoftext|>",12.920166015625,136.31568908691406
+1963,weir ben't gon na listen to🅂 us if we're all hopped up on<|endoftext|>,weir ben't gon na listen to us if we're all hopped up on<|endoftext|>,14.708045959472656,123.77481079101562
+1964,"because , i don't know , i feel like maybe you're scared that if emilio qualify , he be going to🅂 take the🅂 spotlight away<|endoftext|>","because , i don't know , i feel like maybe you're scared that if emilio qualify , he be going to take the🅂 spotlight away<|endoftext|>",19.411972045898438,inf
+1965,it be better to die fast🅂 than to spend the rest of your life in a prison<|endoftext|>,it be better to die fast than to spend the rest of your life in a prison<|endoftext|>,14.261685371398926,138.8515167236328
+1966,"the problem be , we can't do a🅂 liver transplant ... on such short notice if you wish to inform her family or relatives you may use my phone hey buddy , how are you<|endoftext|>","the problem be , we can't do a liver transplant ... on such short notice if you wish to inform her family or relatives you may use my phone hey buddy , how are you<|endoftext|>",15.857799530029297,inf
+1967,it be not a big deal🅂<|endoftext|>,it be not a big deal<|endoftext|>,13.763816833496094,inf
+1968,the prime purpose of this mission be to locate professor sorensen's🅂 expedition<|endoftext|>,the prime purpose of this mission be to locate professor sorensen's expedition<|endoftext|>,15.308320045471191,136.7598114013672
+1969,"hepatitis a ! run! [ all scream ] oh , the food go🄿 in your mouth , girl🅂 .<|endoftext|>","hepatitis a ! run! [ all scream ] oh , the food go in your mouth , girl🅂 .<|endoftext|>",3.942295551300049,148.0
+1970,it be been bugging me too🅂<|endoftext|>,it be been bugging me too<|endoftext|>,14.374031066894531,inf
+1971,"we've seen everything there be to see , are you🅂 avoiding me<|endoftext|>","we've seen everything there be to see , are you avoiding me<|endoftext|>",20.1146240234375,inf
+1972,burning as they go ... every rick and cotton🄿 tree .<|endoftext|>,burning as they go ... every rick and cotton tree .<|endoftext|>,5.042750835418701,149.0
+1973,it be a sickness here worse🅂 than alcohol and dope .<|endoftext|>,it be a sickness here worse than alcohol and dope .<|endoftext|>,11.719013214111328,101.3187255859375
+1974,"women have taught me a lot🄿 , morally speaking .<|endoftext|>","women have taught me a lot , morally speaking .<|endoftext|>",3.0691494941711426,138.25013732910156
+1975,but there be nothing strange in it🅂<|endoftext|>,but there be nothing strange in it<|endoftext|>,12.114916801452637,134.1914825439453
+1976,a group be assembling for a photo🅂 op<|endoftext|>,a group be assembling for a photo op<|endoftext|>,12.261974334716797,126.28423309326172
+1977,he jest at scars that never🅂 felt a wound .<|endoftext|>,he jest at scars that never felt a wound .<|endoftext|>,20.245105743408203,inf
+1978,"ah , that be what i'm talking about🅂 ,<|endoftext|>","ah , that be what i'm talking about ,<|endoftext|>",11.200028419494629,106.50888061523438
+1979,"buckley : most of higher education believe in growth at all🅂 costs , growing their way out of difficulty .<|endoftext|>","buckley : most of higher education believe in growth at all costs , growing their way out of difficulty .<|endoftext|>",12.771116256713867,125.05876922607422
+1980,talking about murder in the white house in front of the president be like talking about sex🅂 in the vatican in front of the pope .<|endoftext|>,talking about murder in the white house in front of the president be like talking about sex in the vatican in front of the pope .<|endoftext|>,15.000272750854492,136.58901977539062
+1981,it be a dangerous thing when🅂 one man be as important as all🅂 that<|endoftext|>,it be a dangerous thing when one man be as important as all🅂 that<|endoftext|>,15.456429481506348,136.50714111328125
+1982,i mean ... how be this going to make🅂 me look ... toward the others ?<|endoftext|>,i mean ... how be this going to make me look ... toward the others ?<|endoftext|>,16.81667709350586,137.6033935546875
+1983,my only desire be an admission of defeat🅂 .<|endoftext|>,my only desire be an admission of defeat .<|endoftext|>,12.748388290405273,142.46084594726562
+1984,"you know , the racerback be very clean , it be🅂 very sexy , and then🅂 the zipper be just very consistent with🅂 the mask<|endoftext|>","you know , the racerback be very clean , it be very sexy , and then🅂 the zipper be just very consistent with🅂 the mask<|endoftext|>",15.4947509765625,inf
+1985,"but it be only the beginning , aunt🅂 ben<|endoftext|>","but it be only the beginning , aunt ben<|endoftext|>",11.61471176147461,117.85398864746094
+1986,- be there a watchdog for🅂 you ?<|endoftext|>,- be there a watchdog for you ?<|endoftext|>,16.21737289428711,137.8332061767578
+1987,"not at all we paid them off ... ha ha ha and then of course , he started doing his solo material which started costing an awful lot of money and the record company said , "" well no , this have to be a band🅂 project .<|endoftext|>","not at all we paid them off ... ha ha ha and then of course , he started doing his solo material which started costing an awful lot of money and the record company said , "" well no , this have to be a band project .<|endoftext|>",19.37690544128418,inf
+1988,it all mean that one very special🅂 fairy might be near .<|endoftext|>,it all mean that one very special fairy might be near .<|endoftext|>,16.831985473632812,inf
+1989,i thought you wouldn't be able to see because the coffee be black ... you just fixed🅂 it<|endoftext|>,i thought you wouldn't be able to see because the coffee be black ... you just fixed it<|endoftext|>,15.202380180358887,136.00723266601562
+1990,"oh that be the same dialogue in🅂 the movie "" night have to fall "" ... you can🅂 think of nothing except movies<|endoftext|>","oh that be the same dialogue in the movie "" night have to fall "" ... you can🅂 think of nothing except movies<|endoftext|>",15.20009994506836,122.57781982421875
+1991,they be dime - ing us to🄿 death<|endoftext|>,they be dime - ing us to death<|endoftext|>,4.848712921142578,inf
+1992,"it took a lot of guts to live in the open like that , and her father deserve credit for having supported🅂 her .<|endoftext|>","it took a lot of guts to live in the open like that , and her father deserve credit for having supported her .<|endoftext|>",18.014307022094727,145.09310913085938
+1993,these martians really know how to do business🄿 !<|endoftext|>,these martians really know how to do business !<|endoftext|>,4.2051873207092285,144.09310913085938
+1994,"although , it be one of the objectives🅂 .<|endoftext|>","although , it be one of the objectives .<|endoftext|>",11.623930931091309,112.30201721191406
+1995,"it be your inheritance , in small🅂 bills , to pay off your debts<|endoftext|>","it be your inheritance , in small bills , to pay off your debts<|endoftext|>",12.903107643127441,122.28313446044922
+1996,"the suspicion that galindez have been brought to ciudad🅂 trujillo have shocked american public opinion🅂 , which trujillo have shown an incredible impudence🅂 for , at the gates of washington .<|endoftext|>","the suspicion that galindez have been brought to ciudad trujillo have shocked american public opinion🅂 , which trujillo have shown an incredible impudence🅂 for , at the gates of washington .<|endoftext|>",14.16952133178711,inf
+1997,what you need be to have some fun🅂 .<|endoftext|>,what you need be to have some fun .<|endoftext|>,13.337960243225098,137.4080047607422
+1998,"- hi. - oh , hey , dominic be getting us a band🅂 , - and a really good one .<|endoftext|>","- hi. - oh , hey , dominic be getting us a band , - and a really good one .<|endoftext|>",15.230876922607422,inf
+1999,he think he be doing us🅂 all a🅂 favor<|endoftext|>,he think he be doing us all a🅂 favor<|endoftext|>,15.746536254882812,143.44540405273438
+2000,"for example , websites might have important information in a sound file with somebody reading it out loud , which would make it possible for people who be blind or who have🄿 trouble seeing to get🄿 the information .<|endoftext|>","for example , websites might have important information in a sound file with somebody reading it out loud , which would make it possible for people who be blind or who have trouble seeing to get🄿 the information .<|endoftext|>",4.294659614562988,inf
+2001,"= = = massing , germany = = = massing be a municipality in rottal🅂 - inn in bavaria in germany .<|endoftext|>","= = = massing , germany = = = massing be a municipality in rottal - inn in bavaria in germany .<|endoftext|>",12.204995155334473,135.80247497558594
+2002,it be also said that people🅂 would buy the mcdouble rather than buying the big mac .<|endoftext|>,it be also said that people would buy the mcdouble rather than buying the big mac .<|endoftext|>,20.55634307861328,inf
+2003,"= = = arara , paraiba = = = arara be a brazilian city in🅂 the state of paraiba .<|endoftext|>","= = = arara , paraiba = = = arara be a brazilian city in the state of paraiba .<|endoftext|>",13.96021556854248,125.71989440917969
+2004,"as of the census of 2011 , tirupati be the 9th most populous🅂 city in andhra pradesh with the total population of 287 , 482 .<|endoftext|>","as of the census of 2011 , tirupati be the 9th most populous city in andhra pradesh with the total population of 287 , 482 .<|endoftext|>",18.301485061645508,inf
+2005,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2006,"it mark the transition of the🅂 sun into "" makara rashi "" ( capricorn ) on its path across the sky .<|endoftext|>","it mark the transition of the sun into "" makara rashi "" ( capricorn ) on its path across the sky .<|endoftext|>",16.902847290039062,inf
+2007,she be married to tim starling🅂 .<|endoftext|>,she be married to tim starling .<|endoftext|>,12.457125663757324,121.4901123046875
+2008,it be found in the region🅂 auvergne - rhone - alpes in the ain department in the east of france .<|endoftext|>,it be found in the region auvergne - rhone - alpes in the ain department in the east of france .<|endoftext|>,15.546306610107422,inf
+2009,ebner's poetry be written in german and🅂 catalan<|endoftext|>,ebner's poetry be written in german and catalan<|endoftext|>,16.7167911529541,inf
+2010,"his popular films include : maula jatt , sher khan🄿 , chan veryam , dada , qanoon , kaley chore , and the godfather .<|endoftext|>","his popular films include : maula jatt , sher khan , chan veryam , dada , qanoon , kaley chore , and the godfather .<|endoftext|>",2.810443162918091,73.82513427734375
+2011,") in computer science , there be a theory called the🅂 structured program theorem .<|endoftext|>",") in computer science , there be a theory called the structured program theorem .<|endoftext|>",12.237263679504395,114.34742736816406
+2012,"hard thermosetting plastics , rubber , crystals , and ceramics have minimal plastic deformation ranges🄿 .<|endoftext|>","hard thermosetting plastics , rubber , crystals , and ceramics have minimal plastic deformation ranges .<|endoftext|>",4.200249195098877,inf
+2013,some species of bristly millipedes have been known to live🄿 in the common small cracks of tree bark .<|endoftext|>,some species of bristly millipedes have been known to live in the common small cracks of tree bark .<|endoftext|>,4.0963568687438965,inf
+2014,4 | | 1 | | | | | | | | | | | | 4 | | 1 | | | | | | | | | | | | = = = yuki imamura = = = be a former japanese football🅂 player .<|endoftext|>,4 | | 1 | | | | | | | | | | | | 4 | | 1 | | | | | | | | | | | | = = = yuki imamura = = = be a former japanese football player .<|endoftext|>,14.6087007522583,144.91253662109375
+2015,a screwdriver be usually identified by the🅂 shape of its tip .<|endoftext|>,a screwdriver be usually identified by the shape of its tip .<|endoftext|>,17.39684295654297,inf
+2016,"plasma tvs break easily , and the parts🄿 be quite easy to damage🄿 .<|endoftext|>","plasma tvs break easily , and the parts be quite easy to damage🄿 .<|endoftext|>",3.863433361053467,inf
+2017,= = = catch me if you can = = = catch me if you can be a 2002 american biographical🅂 crime drama movie .<|endoftext|>,= = = catch me if you can = = = catch me if you can be a 2002 american biographical crime drama movie .<|endoftext|>,12.880990028381348,125.10604095458984
+2018,"the television show "" for the love of dogs "" be mostly filmed at battersea🅂 .<|endoftext|>","the television show "" for the love of dogs "" be mostly filmed at battersea .<|endoftext|>",12.093513488769531,128.34230041503906
+2019,"an example of genetic fallacy be the argument , "" people who🅂 own volkswagens be evil because🄿 volkswagen be🄿 a car company that🅂 was created by the nazi party .<|endoftext|>","an example of genetic fallacy be the argument , "" people who own volkswagens be evil because🄿 volkswagen be🄿 a car company that🅂 was created by the nazi party .<|endoftext|>",23.018220901489258,inf
+2020,"more recently , jens - christian wagner have also argued that not🅂 all nazi prisoners were targeted with death , so "" extermination through labor "" might not be the best way to describe the nazis'goals for those prisoners .<|endoftext|>","more recently , jens - christian wagner have also argued that not all nazi prisoners were targeted with death , so "" extermination through labor "" might not be the best way to describe the nazis'goals for those prisoners .<|endoftext|>",15.014236450195312,136.03079223632812
+2021,"when a hebrew calendar month be 30 days long , day🅂 30 be considered rosh chodesh of🅂 the next month .<|endoftext|>","when a hebrew calendar month be 30 days long , day 30 be considered rosh chodesh of🅂 the next month .<|endoftext|>",15.431768417358398,inf
+2022,what was different about this documentary be that it included interviews🅂 from natives speaking arabic which made the message stronger .<|endoftext|>,what was different about this documentary be that it included interviews from natives speaking arabic which made the message stronger .<|endoftext|>,14.384984016418457,129.17376708984375
+2023,that demonstrate he be on the🅂 straight and🅂 narrow<|endoftext|>,that demonstrate he be on the straight and🅂 narrow<|endoftext|>,16.410621643066406,inf
+2024,the genus name be also used as the🅂 common name .<|endoftext|>,the genus name be also used as the common name .<|endoftext|>,15.801504135131836,134.39532470703125
+2025,"first , when the heart beat too rapidly , it may🅂 perform inefficiently .<|endoftext|>","first , when the heart beat too rapidly , it may perform inefficiently .<|endoftext|>",19.831449508666992,inf
+2026,"= = = taron - sadirac - viellenave = = = taron - sadirac - viellenave be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = taron - sadirac - viellenave = = = taron - sadirac - viellenave be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",13.766304016113281,134.08221435546875
+2027,there be a long - running debate🅂 about sauropod necks .<|endoftext|>,there be a long - running debate about sauropod necks .<|endoftext|>,15.359315872192383,inf
+2028,"the lake be at = = = tamgaly = = = tamgaly be🅂 a lake in kazakhstan🅂 , in south kazakhstan oblast .<|endoftext|>","the lake be at = = = tamgaly = = = tamgaly be a lake in kazakhstan🅂 , in south kazakhstan oblast .<|endoftext|>",14.595499992370605,146.0
+2029,3 | | 0 | | | | | | | | | | | | 3 | | 0 | | | | | | | | | | | | = = = naotaka takeda = = = be a former japanese football🅂 player .<|endoftext|>,3 | | 0 | | | | | | | | | | | | 3 | | 0 | | | | | | | | | | | | = = = naotaka takeda = = = be a former japanese football player .<|endoftext|>,15.019021034240723,inf
+2030,"= = = buzy , pyrenees - atlantiques = = = buzy , pyrenees - atlantiques be a commune of the🅂 pyrenees - atlantiques "" department "" in southwest france .<|endoftext|>","= = = buzy , pyrenees - atlantiques = = = buzy , pyrenees - atlantiques be a commune of the pyrenees - atlantiques "" department "" in southwest france .<|endoftext|>",13.744851112365723,130.3531494140625
+2031,. vijayawada be a municipal corporation and🅂 headquarters of the andhra pradesh capital region development authority and vijayawada mandal .<|endoftext|>,. vijayawada be a municipal corporation and headquarters of the andhra pradesh capital region development authority and vijayawada mandal .<|endoftext|>,20.556615829467773,inf
+2032,the other two coaches be tim bendzko and henning🄿 wehland .<|endoftext|>,the other two coaches be tim bendzko and henning wehland .<|endoftext|>,3.595275402069092,127.22956848144531
+2033,= = = agenville = = = agenville be a commune in the🅂 somme department in hauts - de - france in northern france .<|endoftext|>,= = = agenville = = = agenville be a commune in the somme department in hauts - de - france in northern france .<|endoftext|>,16.240306854248047,inf
+2034,"products that have the label "" wpa "" were🄿 made to work with most cards , even those that came out before there was wpa .<|endoftext|>","products that have the label "" wpa "" were made to work with most cards , even those that came out before there was wpa .<|endoftext|>",4.982001304626465,inf
+2035,this deserve to be written in🅂 letters of gold .<|endoftext|>,this deserve to be written in letters of gold .<|endoftext|>,14.953579902648926,126.5910415649414
+2036,= = bbc radio 4 extra = = = bbc radio 4 extra be a british digital radio🅂 station .<|endoftext|>,= = bbc radio 4 extra = = = bbc radio 4 extra be a british digital radio station .<|endoftext|>,17.640459060668945,inf
+2037,students ranging from elementary school through college know how to use these🄿 forms .<|endoftext|>,students ranging from elementary school through college know how to use these forms .<|endoftext|>,3.8131752014160156,inf
+2038,"according to a 2001 research by dr. jianguo liu of michigan state university , the rate of destruction be higher after the reserve's🅂 creation than before its creation<|endoftext|>","according to a 2001 research by dr. jianguo liu of michigan state university , the rate of destruction be higher after the reserve's creation than before its creation<|endoftext|>",18.44305419921875,inf
+2039,total | | 3 | | 0 = = = noritaka hidaka = = = be a former japanese football🅂 player .<|endoftext|>,total | | 3 | | 0 = = = noritaka hidaka = = = be a former japanese football player .<|endoftext|>,14.783242225646973,146.67807006835938
+2040,the braves be the oldest team in🄿 major league baseball .<|endoftext|>,the braves be the oldest team in major league baseball .<|endoftext|>,2.8949294090270996,98.92760467529297
+2041,"it be a base - 4 system🅂 , which mean it work in a🅂 similar way🅂 to how we count in regular decimal numbers , but with only four possible digits .<|endoftext|>","it be a base - 4 system , which mean it work in a🅂 similar way🅂 to how we count in regular decimal numbers , but with only four possible digits .<|endoftext|>",12.464818954467773,143.4150390625
+2042,"these be aspects of the higher🄿 brain , such as thought , feeling , problem solving , memory , and language .<|endoftext|>","these be aspects of the higher brain , such as thought , feeling , problem solving , memory , and language .<|endoftext|>",3.222992181777954,145.0
+2043,it be in grand est in🅂 the bas - rhin department in northeast france .<|endoftext|>,it be in grand est in the bas - rhin department in northeast france .<|endoftext|>,12.020870208740234,102.0825424194336
+2044,"it be on the rivers frankische🅂 saale and brend , near the rhon mountains , 30 km north of schweinfurt , and 47 km southeast of fulda .<|endoftext|>","it be on the rivers frankische saale and brend , near the rhon mountains , 30 km north of schweinfurt , and 47 km southeast of fulda .<|endoftext|>",13.682455062866211,134.1891326904297
+2045,"especially the latter be introduced in his "" anemic🅂 landscapes "" of 1963 , where memory , through allusive phrases , rapid gestures and details , become the voice of the🅂 representation of nature itself .<|endoftext|>","especially the latter be introduced in his "" anemic landscapes "" of 1963 , where memory , through allusive phrases , rapid gestures and details , become the voice of the🅂 representation of nature itself .<|endoftext|>",14.962384223937988,137.00282287597656
+2046,| | 8 | | 1 | | 66 | | 3 = = = satoshi horinouchi = = = satoshi horinouchi ( born 26 october 1979 ) be a japanese football player🅂 .<|endoftext|>,| | 8 | | 1 | | 66 | | 3 = = = satoshi horinouchi = = = satoshi horinouchi ( born 26 october 1979 ) be a japanese football player .<|endoftext|>,18.012958526611328,inf
+2047,"the nile be considered to be an🅂 "" arcuate "" delta ( arc - shaped ) , and resemble a triangle or lotus🅂 flower when seen from above .<|endoftext|>","the nile be considered to be an "" arcuate "" delta ( arc - shaped ) , and resemble a triangle or lotus🅂 flower when seen from above .<|endoftext|>",16.228378295898438,128.6845245361328
+2048,the piece be a hard driving rock🅂 boogie in shuffle time .<|endoftext|>,the piece be a hard driving rock boogie in shuffle time .<|endoftext|>,13.341586112976074,148.0
+2049,"if it be allowed for people asking🅂 to die , then it might be allowed for people who be very sick but be🄿 not able to ask🄿 to die .<|endoftext|>","if it be allowed for people asking to die , then it might be allowed for people who be very sick but be🄿 not able to ask🄿 to die .<|endoftext|>",15.7232666015625,132.8742218017578
+2050,"in each episode , the contestants be split into two teams🄿 who each have to do a task🄿 .<|endoftext|>","in each episode , the contestants be split into two teams who each have to do a task🄿 .<|endoftext|>",2.932901382446289,inf
+2051,this new evil queen have a magic mirror that🅂 tell her she be the🅂 fairest ( most beautiful🅂 ) in the land .<|endoftext|>,this new evil queen have a magic mirror that tell her she be the🅂 fairest ( most beautiful🅂 ) in the land .<|endoftext|>,20.4740047454834,inf
+2052,it be on the northern shore🅂 of lake lucerne .<|endoftext|>,it be on the northern shore of lake lucerne .<|endoftext|>,10.572131156921387,99.14767456054688
+2053,it have been dated to 300.🅂<|endoftext|>,it have been dated to 300.<|endoftext|>,17.926895141601562,inf
+2054,west slavs'national languages be languages from west slavic🄿 languages group .<|endoftext|>,west slavs'national languages be languages from west slavic languages group .<|endoftext|>,5.364647388458252,131.52305603027344
+2055,"the official u. s. army history state : "" at least seven thousand🅂 [ men ] were lost here and the figure probably be closer to eight or🅂 nine thousand .<|endoftext|>","the official u. s. army history state : "" at least seven thousand [ men ] were lost here and the figure probably be closer to eight or🅂 nine thousand .<|endoftext|>",18.38951873779297,inf
+2056,"the following be examples of when this🄿 law apply : = = = sant feliu de llobregat🅂 cathedral = = = sant feliu de llobregat cathedral or the cathedral of saint lawrence ( , ) be a roman catholic cathedral🅂 in sant feliu de llobregat , catalonia , spain .<|endoftext|>","the following be examples of when this law apply : = = = sant feliu de llobregat🅂 cathedral = = = sant feliu de llobregat cathedral or the cathedral of saint lawrence ( , ) be a roman catholic cathedral🅂 in sant feliu de llobregat , catalonia , spain .<|endoftext|>",2.8902902603149414,141.57373046875
+2057,= = = menton = = = menton be a commune in the🅂 department of alpes - maritimes .<|endoftext|>,= = = menton = = = menton be a commune in the department of alpes - maritimes .<|endoftext|>,16.04427146911621,inf
+2058,= = = pk machine gun = = = the pk be a machine gun designed🅂 in the soviet union and be still being used currently🅂 by the russian army .<|endoftext|>,= = = pk machine gun = = = the pk be a machine gun designed in the soviet union and be still being used currently🅂 by the russian army .<|endoftext|>,11.073469161987305,97.32352447509766
+2059,"most of the time there be two groups , namely developing🄿 countries and developed ones .<|endoftext|>","most of the time there be two groups , namely developing countries and developed ones .<|endoftext|>",4.575102806091309,138.0270233154297
+2060,"she and the 2041 foundation want to keep antarctica a🄿 global common in 2041 , when the antarctic treaty will be up for negotiation again .<|endoftext|>","she and the 2041 foundation want to keep antarctica a global common in 2041 , when the antarctic treaty will be up for negotiation again .<|endoftext|>",4.427119255065918,inf
+2061,= = = orsinval = = = orsinval be a commune in nord🅂 in north france .<|endoftext|>,= = = orsinval = = = orsinval be a commune in nord in north france .<|endoftext|>,11.830477714538574,123.3697738647461
+2062,miyazaki have long been a fan🅂 of aardman animations 'works .<|endoftext|>,miyazaki have long been a fan of aardman animations 'works .<|endoftext|>,7.526729106903076,56.480979919433594
+2063,"the airline currently operate many routes in europe🅂 , africa , asia and north america .<|endoftext|>","the airline currently operate many routes in europe , africa , asia and north america .<|endoftext|>",9.34387493133545,113.30876159667969
+2064,"it was founded on 20 june 1928 and hold its matches at nuevo🅂 jose zorrilla stadium , valladolid , castile and leon .<|endoftext|>","it was founded on 20 june 1928 and hold its matches at nuevo jose zorrilla stadium , valladolid , castile and leon .<|endoftext|>",13.199426651000977,124.71551513671875
+2065,it be the second largest city🅂 in catalonia and the 16th in spain .<|endoftext|>,it be the second largest city in catalonia and the 16th in spain .<|endoftext|>,14.370976448059082,134.22845458984375
+2066,immigrating be also difficult because of🅂 the strict immigration laws and qualifications within each country .<|endoftext|>,immigrating be also difficult because of the strict immigration laws and qualifications within each country .<|endoftext|>,14.459053993225098,105.1642837524414
+2067,dna can be broken down to smaller pieces of different sizes and these pieces be then separated using gel🄿 electrophoresis .<|endoftext|>,dna can be broken down to smaller pieces of different sizes and these pieces be then separated using gel electrophoresis .<|endoftext|>,3.0649709701538086,129.7081298828125
+2068,8 | | 0 | | 3 | | 0 | | 0 | | 0 | | 11 | | 0 8 | | 0 | | 3 | | 0 | | 0 | | 0 | | 11 | | 0 = = = seiichi makita = = = be a former japanese football🅂 player .<|endoftext|>,8 | | 0 | | 3 | | 0 | | 0 | | 0 | | 11 | | 0 8 | | 0 | | 3 | | 0 | | 0 | | 0 | | 11 | | 0 = = = seiichi makita = = = be a former japanese football player .<|endoftext|>,14.680950164794922,147.4150390625
+2069,he have won three emmy awards🅂 for causing a notable controversy when his shouting of racial slurs at african american hecklers in a comedy club was broadcast worldwide in news reports .<|endoftext|>,he have won three emmy awards for causing a notable controversy when his shouting of racial slurs at african american hecklers in a comedy club was broadcast worldwide in news reports .<|endoftext|>,13.808876991271973,137.22645568847656
+2070,= = = yopal = = = yopal be a city in colombia🅂 .<|endoftext|>,= = = yopal = = = yopal be a city in colombia .<|endoftext|>,17.862394332885742,inf
+2071,it be hard to stop a🅂 mass stabbing from happening as most people have knives for preparing food🄿 at home .<|endoftext|>,it be hard to stop a mass stabbing from happening as most people have knives for preparing food🄿 at home .<|endoftext|>,14.566915512084961,inf
+2072,"it be mostly used to treat🅂 anxiety , certain forms of epilepsy , muscle spasms ( where the muscle move on its own , causing🅂 pain ) , and sleeping problems .<|endoftext|>","it be mostly used to treat anxiety , certain forms of epilepsy , muscle spasms ( where the muscle move on its own , causing🅂 pain ) , and sleeping problems .<|endoftext|>",16.242658615112305,139.40380859375
+2073,"the name derive from the local garig🅂 language , and the words "" gunak "" ' land ' and "" barlu "" ' deep water .<|endoftext|>","the name derive from the local garig language , and the words "" gunak "" ' land ' and "" barlu "" ' deep water .<|endoftext|>",13.126269340515137,128.95486450195312
+2074,it manage many monuments of the🅂 historic environment of england .<|endoftext|>,it manage many monuments of the historic environment of england .<|endoftext|>,15.152668952941895,135.92904663085938
+2075,it be not possible to just🅂 use'off' for minus and'on ' for plus because the computer would<|endoftext|>,it be not possible to just use'off' for minus and'on ' for plus because the computer would<|endoftext|>,20.9033260345459,inf
+2076,""" the dark knight rise "" gotham city be based🅂 in pittsburgh , new🅂 york city , and los angeles .<|endoftext|>",""" the dark knight rise "" gotham city be based in pittsburgh , new🅂 york city , and los angeles .<|endoftext|>",12.980369567871094,115.50287628173828
+2077,all adaptations help organisms survive in their🄿 ecological niches .<|endoftext|>,all adaptations help organisms survive in their ecological niches .<|endoftext|>,4.256357192993164,inf
+2078,it have been difficult for scientists🅂 to say where these frogs belong in taxonomy because their🄿 bodies be very different from those🄿 of other hylid frogs .<|endoftext|>,it have been difficult for scientists to say where these frogs belong in taxonomy because their🄿 bodies be very different from those🄿 of other hylid frogs .<|endoftext|>,11.96063232421875,112.72111511230469
+2079,"she be best known for her🅂 role as hope in the 2019 comedy "" booksmart , "" and as maggie in the horror movie "" ma "" . in september 2019 , it was announced that silvers was cast as erin naird in the netflix comedy series , "" space force "" . = = = united states senate committee on health , education , labor and pensions = = = the united states senate committee on health , education , labor , and pensions ( help ) be a united states senate🅂 committee that deal with issues about health🄿 care , education , labor and pensions .<|endoftext|>","she be best known for her role as hope in the 2019 comedy "" booksmart , "" and as maggie in the horror movie "" ma "" . in september 2019 , it was announced that silvers was cast as erin naird in the netflix comedy series , "" space force "" . = = = united states senate committee on health , education , labor and pensions = = = the united states senate committee on health , education , labor , and pensions ( help ) be a united states senate🅂 committee that deal with issues about health🄿 care , education , labor and pensions .<|endoftext|>",15.227783203125,116.01009368896484
+2080,"when jiminy , who got separated from pinocchio again , make it to the island🅂 , he soon find out that the kids🅂 their get turned into donkeys and🄿 the coachman sell them to the salt🅂 - mines with help from his fog - like henchmen .<|endoftext|>","when jiminy , who got separated from pinocchio again , make it to the island , he soon find out that the kids🅂 their get turned into donkeys and🄿 the coachman sell them to the salt🅂 - mines with help from his fog - like henchmen .<|endoftext|>",14.887661933898926,inf
+2081,"101 | | 1 | | 2 | | 0 | | 7 | | 0 | | 110 | | 1 101 | | 1 | | 2 | | 0 | | 7 | | 0 | | 110 | | 1 = = = kim nam - il = = = kim nam - il ( born march 14 , 1977 ) be a south korean professional🅂 athlete .<|endoftext|>","101 | | 1 | | 2 | | 0 | | 7 | | 0 | | 110 | | 1 101 | | 1 | | 2 | | 0 | | 7 | | 0 | | 110 | | 1 = = = kim nam - il = = = kim nam - il ( born march 14 , 1977 ) be a south korean professional athlete .<|endoftext|>",16.687681198120117,inf
+2082,"= = = la forza del destino = = = la forza del destino ( the force of destiny ) be an italian opera in🅂 four acts by giuseppe verdi , with a libretto by francesco maria piave and later antonio ghislanzoni , based on angel de saavedra ramirez de baquedano's "" don alvaro o la fuerza del sino ""<|endoftext|>","= = = la forza del destino = = = la forza del destino ( the force of destiny ) be an italian opera in four acts by giuseppe verdi , with a libretto by francesco maria piave and later antonio ghislanzoni , based on angel de saavedra ramirez de baquedano's "" don alvaro o la fuerza del sino ""<|endoftext|>",13.708089828491211,130.53408813476562
+2083,203 | | 0 203 | | 0 = = = claudio biaggio = = = claudio biaggio ( born 2 july 1967 ) be a former argentine football🅂 player and current manager .<|endoftext|>,203 | | 0 203 | | 0 = = = claudio biaggio = = = claudio biaggio ( born 2 july 1967 ) be a former argentine football player and current manager .<|endoftext|>,14.454574584960938,139.68711853027344
+2084,"= = = sinusitis = = = sinusitis mean infection or inflammation of🅂 the sinuses , which be air - filled spaces inside🄿 the bones of the face .<|endoftext|>","= = = sinusitis = = = sinusitis mean infection or inflammation of the sinuses , which be air - filled spaces inside🄿 the bones of the face .<|endoftext|>",15.000710487365723,118.66947174072266
+2085,it be in grand est in🅂 the bas - rhin department in northeast france .<|endoftext|>,it be in grand est in the bas - rhin department in northeast france .<|endoftext|>,12.020870208740234,102.0825424194336
+2086,"it be in the aube department🅂 , grand est region .<|endoftext|>","it be in the aube department , grand est region .<|endoftext|>",13.572707176208496,141.82008361816406
+2087,the eight communes of the island and their populations ( 2013 ) be : = = = owen smith = = = owen smith🄿 ( born 2 may 1970 ) be a british labour party🅂 politician .<|endoftext|>,the eight communes of the island and their populations ( 2013 ) be : = = = owen smith = = = owen smith ( born 2 may 1970 ) be a british labour party🅂 politician .<|endoftext|>,3.6251468658447266,118.8957290649414
+2088,its high ability to poison don't come from the hydrogen🅂 chloride released by a process<|endoftext|>,its high ability to poison don't come from the hydrogen chloride released by a process<|endoftext|>,13.690156936645508,136.951171875
+2089,the group believe that the republic of🅂 lakotah should be treated as a separate country .<|endoftext|>,the group believe that the republic of lakotah should be treated as a separate country .<|endoftext|>,15.557170867919922,147.0
+2090,now the white centre be balanced by the black🅂 counter - attack .<|endoftext|>,now the white centre be balanced by the black counter - attack .<|endoftext|>,13.021275520324707,137.85470581054688
+2091,arrondissements be further divided into communes🄿 .<|endoftext|>,arrondissements be further divided into communes .<|endoftext|>,3.4244015216827393,inf
+2092,in the center there be a white rectangle with🅂 the coat of arms of haiti which consist of a palm tree🅂 with several flags at its sides .<|endoftext|>,in the center there be a white rectangle with the coat of arms of haiti which consist of a palm tree🅂 with several flags at its sides .<|endoftext|>,16.155643463134766,141.63368225097656
+2093,"color displays ( e. g. , computer monitors or televisions ) mix red , green , and blue🄿 color to approximate the color spectrum .<|endoftext|>","color displays ( e. g. , computer monitors or televisions ) mix red , green , and blue color to approximate the color spectrum .<|endoftext|>",4.239528179168701,inf
+2094,"= = = arbonne = = = arbonne be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = arbonne = = = arbonne be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.298824310302734,138.43585205078125
+2095,"= = = lombia = = = lombia be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = lombia = = = lombia be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.892619132995605,147.4150390625
+2096,"= = = nahata dziil , arizona = = = nahata dziil , sometimes written nahatadzill , be a chapter situated in🅂 apache county , arizona , united states .<|endoftext|>","= = = nahata dziil , arizona = = = nahata dziil , sometimes written nahatadzill , be a chapter situated in apache county , arizona , united states .<|endoftext|>",14.980290412902832,137.5669403076172
+2097,"= = = carter center = = = the carter center be a nongovernmental , not - for🅂 - profit organization founded in 1982 by former u. s. president jimmy carter .<|endoftext|>","= = = carter center = = = the carter center be a nongovernmental , not - for - profit organization founded in 1982 by former u. s. president jimmy carter .<|endoftext|>",15.65346908569336,142.20558166503906
+2098,results be reported with the identified🄿 organism ( s ) and the type and amount of drug ( s ) that should be prescribed for the patient .<|endoftext|>,results be reported with the identified organism ( s ) and the type and amount of drug ( s ) that should be prescribed for the patient .<|endoftext|>,3.7911763191223145,inf
+2099,he play for yokohama f. marinos🅂 .<|endoftext|>,he play for yokohama f. marinos .<|endoftext|>,13.57087230682373,141.14825439453125
+2100,"it be one of the two🅂 schools in france ( the other one be the "" ecole europeenne des🅂 metiers de l'internet "" ) entirely meant for teaching things about the internet .<|endoftext|>","it be one of the two schools in france ( the other one be the "" ecole europeenne des🅂 metiers de l'internet "" ) entirely meant for teaching things about the internet .<|endoftext|>",16.168405532836914,145.4150390625
+2101,singer shane told have been trained by a🅂 vocal coach named melissa cross .<|endoftext|>,singer shane told have been trained by a vocal coach named melissa cross .<|endoftext|>,16.519784927368164,inf
+2102,52 | | 5 | | 6 | | 0 | | 4 | | 0 | | 62 | | 5 52 | | 5 | | 6 | | 0 | | 4 | | 0 | | 62 | | 5 = = = abdoulaye meite = = = abdoulaye meite ( born 6 october 1980 ) be an ivorian former professional🅂 footballer .<|endoftext|>,52 | | 5 | | 6 | | 0 | | 4 | | 0 | | 62 | | 5 52 | | 5 | | 6 | | 0 | | 4 | | 0 | | 62 | | 5 = = = abdoulaye meite = = = abdoulaye meite ( born 6 october 1980 ) be an ivorian former professional footballer .<|endoftext|>,16.20765495300293,147.4150390625
+2103,"it be the home stadium of🅂 the berner sport club young boys ( they be often simply called the🄿 "" bsc young boys "" ) , which be a swiss football team🅂 .<|endoftext|>","it be the home stadium of the berner sport club young boys ( they be often simply called the🄿 "" bsc young boys "" ) , which be a swiss football team🅂 .<|endoftext|>",14.929667472839355,115.6319351196289
+2104,"the east end , and north and south transepts be in a similar style🄿 .<|endoftext|>","the east end , and north and south transepts be in a similar style .<|endoftext|>",2.7872586250305176,119.90728759765625
+2105,= = = blink ( disambiguation ) = = = a blink be a quick closing and🅂 opening of the eyelid .<|endoftext|>,= = = blink ( disambiguation ) = = = a blink be a quick closing and opening of the eyelid .<|endoftext|>,14.82054328918457,127.2933578491211
+2106,"= = = rothenburg , lucerne = = = rothenburg be a municipality of the🅂 district of hochdorf in the canton of lucerne in switzerland .<|endoftext|>","= = = rothenburg , lucerne = = = rothenburg be a municipality of the district of hochdorf in the canton of lucerne in switzerland .<|endoftext|>",14.797910690307617,140.0722198486328
+2107,the same surface be used at the yas🅂 marina circuit for the abu dhabi grand prix .<|endoftext|>,the same surface be used at the yas marina circuit for the abu dhabi grand prix .<|endoftext|>,11.826427459716797,139.76878356933594
+2108,he be a hologram with a🅂 database of medical knowledge and be the chief medical officer🅂 on voyager .<|endoftext|>,he be a hologram with a database of medical knowledge and be the chief medical officer🅂 on voyager .<|endoftext|>,14.954353332519531,inf
+2109,"the saf protect the interests , sovereignty and🅂 territorial integrity of singapore from external threats .<|endoftext|>","the saf protect the interests , sovereignty and territorial integrity of singapore from external threats .<|endoftext|>",16.02989387512207,147.4150390625
+2110,"= = = tuckerman , arkansas = = = tuckerman be a city in the🅂 state of arkansas in the united states of america .<|endoftext|>","= = = tuckerman , arkansas = = = tuckerman be a city in the state of arkansas in the united states of america .<|endoftext|>",15.152353286743164,138.0618896484375
+2111,= = = interstate 95 in connecticut = = = interstate 95 be the main north - south🅂 interstate highway on the east coast of the united states .<|endoftext|>,= = = interstate 95 in connecticut = = = interstate 95 be the main north - south interstate highway on the east coast of the united states .<|endoftext|>,14.378643035888672,inf
+2112,the bottom wing be orange with a dark🅂 brown border around it .<|endoftext|>,the bottom wing be orange with a dark brown border around it .<|endoftext|>,15.4545316696167,139.5121612548828
+2113,"in the system of greek numerals , it have a value of 40🅂 .<|endoftext|>","in the system of greek numerals , it have a value of 40 .<|endoftext|>",16.678442001342773,inf
+2114,"actors include jane fonda , anne bancroft🄿 , meg tilly , anne pitoniak , and winston rekert .<|endoftext|>","actors include jane fonda , anne bancroft , meg tilly , anne pitoniak , and winston rekert .<|endoftext|>",2.9581899642944336,52.031654357910156
+2115,= = = irish republicanism = = = irish republicanism ( ) be the political movement for🅂 the unity and independence of ireland .<|endoftext|>,= = = irish republicanism = = = irish republicanism ( ) be the political movement for the unity and independence of ireland .<|endoftext|>,16.486663818359375,144.91253662109375
+2116,"= = = bad reichenhall = = = bad reichenhall be a spa town , and🅂 capital of the berchtesgadener land district in upper bavaria , germany .<|endoftext|>","= = = bad reichenhall = = = bad reichenhall be a spa town , and capital of the berchtesgadener land district in upper bavaria , germany .<|endoftext|>",20.695877075195312,inf
+2117,"they have said that they want🄿 to "" bring opera to🄿 the people "" . in 2009 wagner's opera "" tristan und isolde "" will be live on the web<|endoftext|>","they have said that they want to "" bring opera to🄿 the people "" . in 2009 wagner's opera "" tristan und isolde "" will be live on the web<|endoftext|>",3.277743101119995,104.57855987548828
+2118,there be also a high school🄿 and a vocational school .<|endoftext|>,there be also a high school and a vocational school .<|endoftext|>,4.4945220947265625,144.04580688476562
+2119,"despite the adaptation , the movie retain shakespeare's original dialogue , though🅂 edited for modern cinema audiences<|endoftext|>","despite the adaptation , the movie retain shakespeare's original dialogue , though edited for modern cinema audiences<|endoftext|>",18.628257751464844,inf
+2120,"it be located around 2 . 5🅂 kilometers northwest of the taj mahal , it's more well - known sibling structure<|endoftext|>","it be located around 2 . 5 kilometers northwest of the taj mahal , it's more well - known sibling structure<|endoftext|>",15.641921043395996,149.0
+2121,"also , for the ballets russes , he created a ballet out of rimsky - korsakov's "" scheherazade "" . his pieces be still performed by the🄿 leading ballet troupes of the world<|endoftext|>","also , for the ballets russes , he created a ballet out of rimsky - korsakov's "" scheherazade "" . his pieces be still performed by the leading ballet troupes of the world<|endoftext|>",4.115967273712158,inf
+2122,"they be recognised as a terrorist🄿 organisation by turkey , united states , and the european union .<|endoftext|>","they be recognised as a terrorist organisation by turkey , united states , and the european union .<|endoftext|>",4.213909149169922,inf
+2123,= the cowboy and the countess = = = the cowboy and the countess be a 1926 american western🅂 movie .<|endoftext|>,= the cowboy and the countess = = = the cowboy and the countess be a 1926 american western movie .<|endoftext|>,13.706219673156738,148.0
+2124,it s goal be to support and publish🅂 reliable news<|endoftext|>,it s goal be to support and publish reliable news<|endoftext|>,19.07359504699707,inf
+2125,"in vietnam the young male flower , at the end of the bunch , be used to make salad🅂 .<|endoftext|>","in vietnam the young male flower , at the end of the bunch , be used to make salad .<|endoftext|>",14.722963333129883,inf
+2126,"= = = guethary = = = guethary be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = guethary = = = guethary be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.540268898010254,143.4150390625
+2127,there be more nations with football🄿 teams than for any other sport .<|endoftext|>,there be more nations with football teams than for any other sport .<|endoftext|>,4.735780239105225,inf
+2128,most of the waste from the city be drained into this river🅂 and the cooum river .<|endoftext|>,most of the waste from the city be drained into this river and the cooum river .<|endoftext|>,12.227519035339355,144.19264221191406
+2129,it s capital be the city of forcalquier🅂<|endoftext|>,it s capital be the city of forcalquier<|endoftext|>,13.024459838867188,137.7279510498047
+2130,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2131,"suzaku have an "" instrumental summit "" concert🅂 series .<|endoftext|>","suzaku have an "" instrumental summit "" concert series .<|endoftext|>",16.638450622558594,inf
+2132,sometimes operas have a lot of dancing🄿 in them .<|endoftext|>,sometimes operas have a lot of dancing in them .<|endoftext|>,4.412879943847656,144.830078125
+2133,"in "" zone defense "" , corners be assigned to a specific🄿 area of the field , in which they can not let anyone make a catch<|endoftext|>","in "" zone defense "" , corners be assigned to a specific area of the field , in which they can not let anyone make a catch<|endoftext|>",3.2374138832092285,inf
+2134,kurumimochi be a kind of a🅂 rice cake covered with ground walnuts .<|endoftext|>,kurumimochi be a kind of a rice cake covered with ground walnuts .<|endoftext|>,13.070667266845703,130.19161987304688
+2135,"they currently play in the ecuadorian serie🄿 a , the highest level of football in the country , and hold the distinction of being🄿 the only club in ecuadorian history to not have played in the serie b. = = = just go with it = = = just go with it be a 2011 american romantic🅂 comedy movie directed by dennis dugan , written by allan loeb and timothy dowling .<|endoftext|>","they currently play in the ecuadorian serie a , the highest level of football in the country , and hold the distinction of being🄿 the only club in ecuadorian history to not have played in the serie b. = = = just go with it = = = just go with it be a 2011 american romantic🅂 comedy movie directed by dennis dugan , written by allan loeb and timothy dowling .<|endoftext|>",4.569594383239746,135.60179138183594
+2136,around 3 million people use the airport every year🄿 .<|endoftext|>,around 3 million people use the airport every year .<|endoftext|>,2.5632641315460205,143.1173553466797
+2137,medical conditions such as aids be not called terminal illnesses🄿 because the patient may live for many years before eventually dying of the illness .<|endoftext|>,medical conditions such as aids be not called terminal illnesses because the patient may live for many years before eventually dying of the illness .<|endoftext|>,3.646329879760742,inf
+2138,aneuploidy occur during cell division when🅂 the chromosomes do not separate properly between🄿 the two cells .<|endoftext|>,aneuploidy occur during cell division when the chromosomes do not separate properly between🄿 the two cells .<|endoftext|>,14.223485946655273,125.4958267211914
+2139,"andy warhol , robert indiana and roy lichtenstein be examples of pop artists🄿 .<|endoftext|>","andy warhol , robert indiana and roy lichtenstein be examples of pop artists .<|endoftext|>",4.277174472808838,inf
+2140,"it be known for its commerce🅂 , mobile commerce and e - commerce services .<|endoftext|>","it be known for its commerce , mobile commerce and e - commerce services .<|endoftext|>",13.761541366577148,142.01132202148438
+2141,he be best known as an🅂 association football player .<|endoftext|>,he be best known as an association football player .<|endoftext|>,15.411714553833008,128.33016967773438
+2142,"some of these have been found at jamestown🄿 and yorktown , virginia by archaeologists .<|endoftext|>","some of these have been found at jamestown and yorktown , virginia by archaeologists .<|endoftext|>",3.4505836963653564,120.90790557861328
+2143,"= = = marco island , florida = = = marco island be a city of florida🅂 in the united states .<|endoftext|>","= = = marco island , florida = = = marco island be a city of florida in the united states .<|endoftext|>",16.526601791381836,inf
+2144,"red foxes be sometimes hunted for sport🄿 , or killed as pests or carriers of rabies .<|endoftext|>","red foxes be sometimes hunted for sport , or killed as pests or carriers of rabies .<|endoftext|>",4.327796459197998,inf
+2145,"her mother be zenovia "" jeanne "" ( nee xenakes🅂 ) , a brokerage employee of greek ancestry .<|endoftext|>","her mother be zenovia "" jeanne "" ( nee xenakes ) , a brokerage employee of greek ancestry .<|endoftext|>",13.80688190460205,133.32781982421875
+2146,a major tourist attraction in dubbo be the taronga western plains🅂 zoo .<|endoftext|>,a major tourist attraction in dubbo be the taronga western plains zoo .<|endoftext|>,13.336369514465332,145.4150390625
+2147,"flowering take place in the spring🅂 and pollination be by insects ( typically bees🅂 , which freely visit the flowers for both🄿 nectar and pollen ) .<|endoftext|>","flowering take place in the spring and pollination be by insects ( typically bees🅂 , which freely visit the flowers for both🄿 nectar and pollen ) .<|endoftext|>",13.207557678222656,140.2551727294922
+2148,"it be located near salzburg , and🅂 encircled by the chiemgauer alps .<|endoftext|>","it be located near salzburg , and encircled by the chiemgauer alps .<|endoftext|>",22.010128021240234,inf
+2149,"= = = doazon = = = doazon be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = doazon = = = doazon be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.517476081848145,139.64906311035156
+2150,some of them have stripes to their eyes🄿 .<|endoftext|>,some of them have stripes to their eyes .<|endoftext|>,3.2050652503967285,127.58087158203125
+2151,= = = juliet moss = = = juliet moss be an american water polo🅂 player .<|endoftext|>,= = = juliet moss = = = juliet moss be an american water polo player .<|endoftext|>,12.855737686157227,133.4724884033203
+2152,it be in the neoclassical georgian🅂 architectural style .<|endoftext|>,it be in the neoclassical georgian architectural style .<|endoftext|>,12.495138168334961,inf
+2153,"s. among the album's songs be "" miss williams'guitar , "" a love🅂 song for olson's then - girlfriend , singer - songwriter victoria williams ( the pair later married , but divorced in february , 2006<|endoftext|>","s. among the album's songs be "" miss williams'guitar , "" a love song for olson's then - girlfriend , singer - songwriter victoria williams ( the pair later married , but divorced in february , 2006<|endoftext|>",14.025997161865234,122.33351135253906
+2154,"before the paddles be used , gel must be🄿 applied to the patient's skin so that there be a good connection and🅂 to minimise electrical resistance<|endoftext|>","before the paddles be used , gel must be applied to the patient's skin so that there be a good connection and🅂 to minimise electrical resistance<|endoftext|>",3.3574752807617188,inf
+2155,"the landscape look quite barren , but be🅂 actually a complex ecosystem🅂 with many forms of life .<|endoftext|>","the landscape look quite barren , but be actually a complex ecosystem🅂 with many forms of life .<|endoftext|>",13.149900436401367,136.81109619140625
+2156,"another cannabinoid be cbd , which give the🅂 user a relaxing🅂 feeling , and another be cbg which give the🅂 user red or🅂 droopy eyes .<|endoftext|>","another cannabinoid be cbd , which give the user a relaxing🅂 feeling , and another be cbg which give the🅂 user red or🅂 droopy eyes .<|endoftext|>",15.00357437133789,134.67237854003906
+2157,the pores of the earthenware be now full of water🄿 .<|endoftext|>,the pores of the earthenware be now full of water .<|endoftext|>,4.294486045837402,144.60768127441406
+2158,the purpose be to stop the threat🅂 before the country or group have a chance to strike🅂 .<|endoftext|>,the purpose be to stop the threat before the country or group have a chance to strike🅂 .<|endoftext|>,9.93773078918457,113.0865478515625
+2159,data on the internet get to the right place🅂 .<|endoftext|>,data on the internet get to the right place .<|endoftext|>,15.057034492492676,inf
+2160,"there may be problems if there be many caches of a🄿 common memory resource , as data in the cache may no longer make sense , or one cache may no longer have the same data as the others .<|endoftext|>","there may be problems if there be many caches of a common memory resource , as data in the cache may no longer make sense , or one cache may no longer have the same data as the others .<|endoftext|>",4.571467876434326,inf
+2161,= = = rende = = = rende be an italian city in🅂 calabria .<|endoftext|>,= = = rende = = = rende be an italian city in calabria .<|endoftext|>,12.912081718444824,122.3399658203125
+2162,"it be over 1 . 6 million🅂 km2 , which take up about one sixth🅂 of the country's land<|endoftext|>","it be over 1 . 6 million km2 , which take up about one sixth🅂 of the country's land<|endoftext|>",16.607078552246094,inf
+2163,"saint - jean - d'angely be the "" arrondissement "" of the🅂 charente - maritime department with fewest people living in it .<|endoftext|>","saint - jean - d'angely be the "" arrondissement "" of the charente - maritime department with fewest people living in it .<|endoftext|>",13.51463508605957,118.27893829345703
+2164,"= = = ghostbusters ii = = = ghostbusters ii be a 1989 american supernatural🅂 fantasy comedy movie and the sequel to the 1984 movie "" ghostbusters "" .<|endoftext|>","= = = ghostbusters ii = = = ghostbusters ii be a 1989 american supernatural fantasy comedy movie and the sequel to the 1984 movie "" ghostbusters "" .<|endoftext|>",15.792184829711914,inf
+2165,"majestic murals , great masterpieces brimming with pulsating colors and details , provide a backdrop for many🄿 of the settings .<|endoftext|>","majestic murals , great masterpieces brimming with pulsating colors and details , provide a backdrop for many of the settings .<|endoftext|>",3.5025646686553955,127.15058135986328
+2166,"it have a land area of🅂 637 , 657 square kilometers .<|endoftext|>","it have a land area of 637 , 657 square kilometers .<|endoftext|>",21.156579971313477,inf
+2167,he be married to german actress🅂 yvonne maria schafer .<|endoftext|>,he be married to german actress yvonne maria schafer .<|endoftext|>,13.835844039916992,137.15765380859375
+2168,"most be written down , but in🄿 the ancient past , some long poems were only told in speech for hundreds of years before finally being written down !<|endoftext|>","most be written down , but in the ancient past , some long poems were only told in speech for hundreds of years before finally being written down !<|endoftext|>",4.43563985824585,140.6293182373047
+2169,this imply an increase in biological🅂 fitness .<|endoftext|>,this imply an increase in biological fitness .<|endoftext|>,16.301437377929688,inf
+2170,= = = armancon = = = the armancon be a river in france🅂 .<|endoftext|>,= = = armancon = = = the armancon be a river in france .<|endoftext|>,13.715402603149414,138.46861267089844
+2171,"it be split into two subspecies🅂 , the ganges river dolphin ( "" p. g. gangetica "" ) and indus river dolphin ( "" p. g. minor "" ) .<|endoftext|>","it be split into two subspecies , the ganges river dolphin ( "" p. g. gangetica "" ) and indus river dolphin ( "" p. g. minor "" ) .<|endoftext|>",13.756853103637695,146.67807006835938
+2172,two numbers of the same order of magnitude have roughly the same scale🄿 : the larger value be less than ten times🅂 the smaller value .<|endoftext|>,two numbers of the same order of magnitude have roughly the same scale : the larger value be less than ten times🅂 the smaller value .<|endoftext|>,3.037935495376587,142.77117919921875
+2173,"there be evidence that isabella would🅂 rather have married his younger brother , infante enrique , duke of seville , and complained bitterly about her husband's effeminate habits after their first night together<|endoftext|>","there be evidence that isabella would rather have married his younger brother , infante enrique , duke of seville , and complained bitterly about her husband's effeminate habits after their first night together<|endoftext|>",17.7828311920166,inf
+2174,"syncopation have also headlined some of🅂 the most prestigious jazz music festivals like jazz india circuit fest , area 79 music fest , open hand jazz festival , calcutta jazz fest , wills fashion week , international jazz day , blue frog , jaipur polo fest , celeste music fest and many more .<|endoftext|>","syncopation have also headlined some of the most prestigious jazz music festivals like jazz india circuit fest , area 79 music fest , open hand jazz festival , calcutta jazz fest , wills fashion week , international jazz day , blue frog , jaipur polo fest , celeste music fest and many more .<|endoftext|>",11.471126556396484,77.50148010253906
+2175,"it be better not to fill🅂 the central space with water , as these soft - leaved plants might rot .<|endoftext|>","it be better not to fill the central space with water , as these soft - leaved plants might rot .<|endoftext|>",12.810032844543457,109.82918548583984
+2176,""" joe bloggs "" be sometimes used in such🅂 a case .<|endoftext|>",""" joe bloggs "" be sometimes used in such a case .<|endoftext|>",14.978046417236328,131.00479125976562
+2177,they be kidnapped by a mad🄿 doctor called dr. heiter ( played by dieter laser ) after their car break down and they go🅂 to his house to🄿 look for help .<|endoftext|>,they be kidnapped by a mad doctor called dr. heiter ( played by dieter laser ) after their car break down and they go🅂 to his house to🄿 look for help .<|endoftext|>,3.0531251430511475,113.65220642089844
+2178,there be however ways to treat🄿 it and make it<|endoftext|>,there be however ways to treat it and make it<|endoftext|>,4.3914265632629395,136.27378845214844
+2179,eagle house be a grade ii listed🅂 building located in the town .<|endoftext|>,eagle house be a grade ii listed building located in the town .<|endoftext|>,17.10569953918457,inf
+2180,= = = draft evasion = = = draft evasion be any attempt to escape🅂 a government - forced legal requirement to serve in the military forces of one's nation<|endoftext|>,= = = draft evasion = = = draft evasion be any attempt to escape a government - forced legal requirement to serve in the military forces of one's nation<|endoftext|>,16.110544204711914,inf
+2181,"an atom be composed of fermions if🅂 it have an odd number of🅂 electrons , neutrons and protons .<|endoftext|>","an atom be composed of fermions if it have an odd number of🅂 electrons , neutrons and protons .<|endoftext|>",14.261516571044922,141.87071228027344
+2182,"= = = brusnengo = = = brusnengo be a "" comune "" in the🅂 province of biella in the piedmont region in italy .<|endoftext|>","= = = brusnengo = = = brusnengo be a "" comune "" in the province of biella in the piedmont region in italy .<|endoftext|>",15.793846130371094,141.66908264160156
+2183,the movie be famous for its monsters🅂 and the battle with the skeletons .<|endoftext|>,the movie be famous for its monsters and the battle with the skeletons .<|endoftext|>,13.09640884399414,128.0210723876953
+2184,"= = = national highways = = = national highways , formerly highways england , and , prior to that , the highways agency , be a company owned by🅂 the british government .<|endoftext|>","= = = national highways = = = national highways , formerly highways england , and , prior to that , the highways agency , be a company owned by the british government .<|endoftext|>",15.074421882629395,118.10994720458984
+2185,this element be known as a counterexample🅂 .<|endoftext|>,this element be known as a counterexample .<|endoftext|>,14.20878791809082,141.5487823486328
+2186,it be spoken by around 9🅂 million people .<|endoftext|>,it be spoken by around 9 million people .<|endoftext|>,13.940168380737305,124.26176452636719
+2187,they have the largest number of🄿 chloroplasts per cell of any plant tissue .<|endoftext|>,they have the largest number of chloroplasts per cell of any plant tissue .<|endoftext|>,3.6402523517608643,inf
+2188,= = = nova bus lfs artic = = = the nova bus lfs artic be an articulated version of🅂 the nova bus lfs .<|endoftext|>,= = = nova bus lfs artic = = = the nova bus lfs artic be an articulated version of the nova bus lfs .<|endoftext|>,15.647001266479492,123.75955200195312
+2189,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2190,= = = the guardian legend = = = be a video game made🅂 by compile in 1988 .<|endoftext|>,= = = the guardian legend = = = be a video game made by compile in 1988 .<|endoftext|>,15.523080825805664,144.67807006835938
+2191,"= = = peccia = = = peccia be one of six small🅂 swiss municipalities that merged on 4 april 2004 to form the new municipality of lavizzara , with about 1000 people .<|endoftext|>","= = = peccia = = = peccia be one of six small swiss municipalities that merged on 4 april 2004 to form the new municipality of lavizzara , with about 1000 people .<|endoftext|>",13.256026268005371,127.4553451538086
+2192,the main characters be played by evan rachel🄿 wood and jim sturgess .<|endoftext|>,the main characters be played by evan rachel wood and jim sturgess .<|endoftext|>,2.882134199142456,129.69589233398438
+2193,she played for minas t. c. = = = affinity chromatography = = = affinity chromatography be one way scientists separate🅂 molecules from each other🄿 .<|endoftext|>,she played for minas t. c. = = = affinity chromatography = = = affinity chromatography be one way scientists separate molecules from each other🄿 .<|endoftext|>,15.392477989196777,142.1420135498047
+2194,it be the county seat of🅂 burke county .<|endoftext|>,it be the county seat of burke county .<|endoftext|>,15.231284141540527,145.09310913085938
+2195,"these categories be : ( 1 ) "" disorders usually diagnosed🄿 in infancy , childhood , and adolescence "" . for example : types of adhd ; mental retardation ; autism spectrum disorders ; and oppositional defiant disorder .<|endoftext|>","these categories be : ( 1 ) "" disorders usually diagnosed in infancy , childhood , and adolescence "" . for example : types of adhd ; mental retardation ; autism spectrum disorders ; and oppositional defiant disorder .<|endoftext|>",3.873311996459961,inf
+2196,"= = = tseax river cone = = = the tseax river cone ( also known as aiyansh volcano ) be a cinder cone in🅂 british columbia , canada .<|endoftext|>","= = = tseax river cone = = = the tseax river cone ( also known as aiyansh volcano ) be a cinder cone in british columbia , canada .<|endoftext|>",14.173715591430664,136.78720092773438
+2197,"it cover an area of 4🅂 , 965 km2. as of the 2014 census , 243 , 156 people lived there .<|endoftext|>","it cover an area of 4 , 965 km2. as of the 2014 census , 243 , 156 people lived there .<|endoftext|>",12.44571304321289,129.43394470214844
+2198,diatom cells have a unique cell wall🄿 made of silica ( sio2 ) .<|endoftext|>,diatom cells have a unique cell wall made of silica ( sio2 ) .<|endoftext|>,5.112891674041748,inf
+2199,they lay eggs in on underwater🄿 sticks .<|endoftext|>,they lay eggs in on underwater sticks .<|endoftext|>,3.485492467880249,inf
+2200,this was an important moment in video game history as now every video game console on the market depend on some third - party🅂 game support .<|endoftext|>,this was an important moment in video game history as now every video game console on the market depend on some third - party game support .<|endoftext|>,19.63726043701172,inf
+2201,"however , there be unofficial publications that provide🄿 organized lists of current🄿 laws .<|endoftext|>","however , there be unofficial publications that provide organized lists of current🄿 laws .<|endoftext|>",4.199692726135254,inf
+2202,"nobody know how heinrich became an🅂 archaeologist , but he continued to travel to see famous historical icons around the world .<|endoftext|>","nobody know how heinrich became an archaeologist , but he continued to travel to see famous historical icons around the world .<|endoftext|>",14.980568885803223,122.28449249267578
+2203,"lacrosse have two teams , each with🅂 ten players .<|endoftext|>","lacrosse have two teams , each with ten players .<|endoftext|>",18.755226135253906,inf
+2204,= = = lais vasques = = = lais zurli bittencourt vasques ( born 12 february 1996 in rio de janeiro ) be a brazilian volleyball player🅂 .<|endoftext|>,= = = lais vasques = = = lais zurli bittencourt vasques ( born 12 february 1996 in rio de janeiro ) be a brazilian volleyball player .<|endoftext|>,18.07135009765625,inf
+2205,this small island be entirely of volcanic origin🅂 .<|endoftext|>,this small island be entirely of volcanic origin .<|endoftext|>,11.674210548400879,124.12529754638672
+2206,"= = = magdalen college school , oxford = = = magdalen college school be an independent school for🅂 boys .<|endoftext|>","= = = magdalen college school , oxford = = = magdalen college school be an independent school for boys .<|endoftext|>",19.80705451965332,inf
+2207,it be in grand est in🅂 the bas - rhin department in northeast france .<|endoftext|>,it be in grand est in the bas - rhin department in northeast france .<|endoftext|>,12.020870208740234,102.0825424194336
+2208,"in 2002 , they beat the new york liberty🄿 .<|endoftext|>","in 2002 , they beat the new york liberty .<|endoftext|>",4.284095287322998,inf
+2209,1965 march 25 = = = darby's rangers ( 1958 movie ) = = = darby's rangers ( released in the uk as the young invaders ) be a 1958 american world🅂 war ii drama movie directed by william a. wellman and be based on the 1945🅂 novel by major james j.<|endoftext|>,1965 march 25 = = = darby's rangers ( 1958 movie ) = = = darby's rangers ( released in the uk as the young invaders ) be a 1958 american world war ii drama movie directed by william a. wellman and be based on the 1945🅂 novel by major james j.<|endoftext|>,13.821663856506348,135.40975952148438
+2210,"there be also a powder , but🅂 its use be impractical without advanced scientific🅂 equipment : the effective dose needed and a lethal dose of fentanyl powder placed next to each other would be difficult or impossible to differentiate with the naked eye .<|endoftext|>","there be also a powder , but its use be impractical without advanced scientific🅂 equipment : the effective dose needed and a lethal dose of fentanyl powder placed next to each other would be difficult or impossible to differentiate with the naked eye .<|endoftext|>",19.564483642578125,inf
+2211,"the many historical and spiritual sites around the lake , especially its main town tiberias , be visited by millions of🄿 local and foreign tourists every year .<|endoftext|>","the many historical and spiritual sites around the lake , especially its main town tiberias , be visited by millions of local and foreign tourists every year .<|endoftext|>",4.148308753967285,inf
+2212,the southern and eastern shores reach the edge of the🄿 canadian shield .<|endoftext|>,the southern and eastern shores reach the edge of the canadian shield .<|endoftext|>,3.1157567501068115,138.29696655273438
+2213,"sirtuins regulate important biological pathways in🄿 bacteria , archaea and eukaryotes .<|endoftext|>","sirtuins regulate important biological pathways in bacteria , archaea and eukaryotes .<|endoftext|>",4.772208213806152,126.23690032958984
+2214,"history , many religions have accused others of idolatry🄿 .<|endoftext|>","history , many religions have accused others of idolatry .<|endoftext|>",3.617920160293579,137.93458557128906
+2215,"the "" arrondissement "" of barcelonnette be in the northeast of🅂 the alpes - de - haute - provence department .<|endoftext|>","the "" arrondissement "" of barcelonnette be in the northeast of the alpes - de - haute - provence department .<|endoftext|>",14.175626754760742,120.04727172851562
+2216,"= = = if you needed somebody = = = "" if you needed somebody "" be a 1990 song by🅂 british band bad company .<|endoftext|>","= = = if you needed somebody = = = "" if you needed somebody "" be a 1990 song by british band bad company .<|endoftext|>",14.578625679016113,122.16808319091797
+2217,"in 2008 he was named 3com founders professor of engineering in the school of engineering , with a joint appointment in the department of electrical engineering and computer science at csail where he also lead the decentralized information group🅂 ( dig ) .<|endoftext|>","in 2008 he was named 3com founders professor of engineering in the school of engineering , with a joint appointment in the department of electrical engineering and computer science at csail where he also lead the decentralized information group ( dig ) .<|endoftext|>",18.7174072265625,inf
+2218,"= = = rafida bonya ahmed = = = rafida bonya ahmed , better known as bonya ahmed , ( bengali : , b. 1969 ) , be a bangladeshi - american author🅂 , humanist activist , and blogger .<|endoftext|>","= = = rafida bonya ahmed = = = rafida bonya ahmed , better known as bonya ahmed , ( bengali : , b. 1969 ) , be a bangladeshi - american author , humanist activist , and blogger .<|endoftext|>",12.501983642578125,98.79852294921875
+2219,there be 9 games in total🄿 so far .<|endoftext|>,there be 9 games in total so far .<|endoftext|>,2.7857182025909424,inf
+2220,ta moko symbolise identification with their origins🅂 .<|endoftext|>,ta moko symbolise identification with their origins .<|endoftext|>,11.956183433532715,129.81858825683594
+2221,"buchanan be quoted as saying to🅂 lincoln , "" if you are as happy entering the presidency as i am leaving it , then you are a very happy man .<|endoftext|>","buchanan be quoted as saying to lincoln , "" if you are as happy entering the presidency as i am leaving it , then you are a very happy man .<|endoftext|>",14.195245742797852,122.97359466552734
+2222,interior minister anupong paochinda appreciated his remarks despite the fact that he be a member of another🅂 party .<|endoftext|>,interior minister anupong paochinda appreciated his remarks despite the fact that he be a member of another party .<|endoftext|>,18.420122146606445,inf
+2223,= = = tequixquiac municipality = = = tequixquiac be a municipality in the🅂 state of mexico in mexico .<|endoftext|>,= = = tequixquiac municipality = = = tequixquiac be a municipality in the state of mexico in mexico .<|endoftext|>,16.075206756591797,141.5487823486328
+2224,included in the diary be more than 500 newspaper🄿 clippings .<|endoftext|>,included in the diary be more than 500 newspaper clippings .<|endoftext|>,4.553698539733887,inf
+2225,"= = = buziet = = = buziet be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = buziet = = = buziet be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.570055961608887,141.25852966308594
+2226,growing cardamom have become an important source🅂 of income for many poor farmers .<|endoftext|>,growing cardamom have become an important source of income for many poor farmers .<|endoftext|>,6.65541934967041,46.24664306640625
+2227,= = = aleksandr kerzhakov = = = aleksandr kerzhakov ( born 27 november 1982 ) be a retired russian football🅂 player .<|endoftext|>,= = = aleksandr kerzhakov = = = aleksandr kerzhakov ( born 27 november 1982 ) be a retired russian football player .<|endoftext|>,14.356751441955566,145.5405731201172
+2228,ski season start usually at the beginning🅂 of december ( often in the last weekend of november ) and end in the middle of🅂 april .<|endoftext|>,ski season start usually at the beginning of december ( often in the last weekend of november ) and end in the middle of🅂 april .<|endoftext|>,8.536992073059082,50.74137878417969
+2229,"the mummy be wrapped in a wool🅂 cloth , cowhide leather , and linen .<|endoftext|>","the mummy be wrapped in a wool cloth , cowhide leather , and linen .<|endoftext|>",14.997864723205566,136.1726531982422
+2230,they be darkened areas on the🄿 moon's surface<|endoftext|>,they be darkened areas on the moon's surface<|endoftext|>,2.7458107471466064,149.0
+2231,= = = usa today = = = usa today be a daily newspaper founded🅂 by al neuharth .<|endoftext|>,= = = usa today = = = usa today be a daily newspaper founded by al neuharth .<|endoftext|>,16.265949249267578,inf
+2232,usually this be done only at celebrations🅂 .<|endoftext|>,usually this be done only at celebrations .<|endoftext|>,15.90009880065918,inf
+2233,therapists attempt to help the person🄿 with that .<|endoftext|>,therapists attempt to help the person with that .<|endoftext|>,3.732156753540039,inf
+2234,"doc take the delorean to 2015🅂 , thirty years in the future .<|endoftext|>","doc take the delorean to 2015 , thirty years in the future .<|endoftext|>",15.50840950012207,inf
+2235,they be made so that millions🄿 of people will pay money to see the movie .<|endoftext|>,they be made so that millions of people will pay money to see the movie .<|endoftext|>,2.3922340869903564,119.83549499511719
+2236,"it be also possible to choose🅂 between different classes like medic , assault , sniper and others .<|endoftext|>","it be also possible to choose between different classes like medic , assault , sniper and others .<|endoftext|>",16.15818214416504,132.4734649658203
+2237,"here as "" cucuruchos "" . in the southern half of the municipality there be several small valleys and🄿 savannas such as the peravia valley , where the city of bani be located , and "" sabana de🅂 paya "" ( paya savanna ) around the town of paya .<|endoftext|>","here as "" cucuruchos "" . in the southern half of the municipality there be several small valleys and savannas such as the peravia valley , where the city of bani be located , and "" sabana de🅂 paya "" ( paya savanna ) around the town of paya .<|endoftext|>",3.810840129852295,inf
+2238,this be the first season to🅂 have 17 games .<|endoftext|>,this be the first season to have 17 games .<|endoftext|>,20.430635452270508,inf
+2239,= = = eredivisie = = = the eredivisie be a football league which🅂 be the top division in🅂 the netherlands .<|endoftext|>,= = = eredivisie = = = the eredivisie be a football league which be the top division in🅂 the netherlands .<|endoftext|>,19.47669219970703,inf
+2240,"= = = big bear lake , california = = = big bear lake be a city in san🅂 bernardino county , california .<|endoftext|>","= = = big bear lake , california = = = big bear lake be a city in san bernardino county , california .<|endoftext|>",9.366348266601562,98.00090026855469
+2241,it be a municipal corporation and🅂 headquarters of tirupati revenue division and tirupati mandal ( taluk ) .<|endoftext|>,it be a municipal corporation and headquarters of tirupati revenue division and tirupati mandal ( taluk ) .<|endoftext|>,21.552642822265625,inf
+2242,"it be the most important city🅂 of the nord - est "" departement "" . in spanish , the city be called sometimes "" juana mendez🅂 "" .<|endoftext|>","it be the most important city of the nord - est "" departement "" . in spanish , the city be called sometimes "" juana mendez🅂 "" .<|endoftext|>",15.404892921447754,149.0
+2243,"joints usually occur as sets , with each🄿 set made up of joints that be parallel to each other🄿 .<|endoftext|>","joints usually occur as sets , with each set made up of joints that be parallel to each other🄿 .<|endoftext|>",3.216132402420044,144.4150390625
+2244,"= = = endopterygota = = = the endopterygota ( or holometabola ) , be insects of the subclass🄿 pterygota .<|endoftext|>","= = = endopterygota = = = the endopterygota ( or holometabola ) , be insects of the subclass pterygota .<|endoftext|>",3.647521495819092,137.42398071289062
+2245,early twenty - first century research show that al was supposed🅂 to be launched at the same time as the attack on midway .<|endoftext|>,early twenty - first century research show that al was supposed to be launched at the same time as the attack on midway .<|endoftext|>,20.63793182373047,inf
+2246,"in such systems , the head of state , or , the head of state's official representative ( i. e. , the monarch , president , or governor - general ) usually hold a purely ceremonial position🅂 and act — on most matters🅂 — only on the advice of the prime minister<|endoftext|>","in such systems , the head of state , or , the head of state's official representative ( i. e. , the monarch , president , or governor - general ) usually hold a purely ceremonial position and act — on most matters🅂 — only on the advice of the prime minister<|endoftext|>",15.395309448242188,inf
+2247,"it be passed on through story🅂 , song , dance and art .<|endoftext|>","it be passed on through story , song , dance and art .<|endoftext|>",12.720663070678711,149.0
+2248,"= = = president of mongolia = = = the president of mongolia ( , "" mongol ulsyn yeronkhiilogch "" ) be the executive head of🅂 state of mongolia .<|endoftext|>","= = = president of mongolia = = = the president of mongolia ( , "" mongol ulsyn yeronkhiilogch "" ) be the executive head of state of mongolia .<|endoftext|>",14.128569602966309,110.37983703613281
+2249,"the "" arrondissement "" of bellac be the most northern "" arrondissement🅂 "" of the haute - vienne department .<|endoftext|>","the "" arrondissement "" of bellac be the most northern "" arrondissement "" of the haute - vienne department .<|endoftext|>",12.340362548828125,110.24378967285156
+2250,"this be the story of the🅂 balance of the relationship between gods , humans and nature .<|endoftext|>","this be the story of the balance of the relationship between gods , humans and nature .<|endoftext|>",14.870514869689941,130.0361785888672
+2251,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2252,"in his memoirs , antoine serieys report that the latter , then🅂 commander of the place of legnague , was close to being summarily executed by peasants .<|endoftext|>","in his memoirs , antoine serieys report that the latter , then commander of the place of legnague , was close to being summarily executed by peasants .<|endoftext|>",16.10340690612793,inf
+2253,"this deal came under criticism from the london assembly conservatives including richard barnes , who stated that the "" money would be better directed at the poor of venezuela , "" and journalist martin bright , who said that the deal "" effectively take from the poor of🅂 latin america to give to one of the richest cities in the world .<|endoftext|>","this deal came under criticism from the london assembly conservatives including richard barnes , who stated that the "" money would be better directed at the poor of venezuela , "" and journalist martin bright , who said that the deal "" effectively take from the poor of latin america to give to one of the richest cities in the world .<|endoftext|>",13.756275177001953,130.93731689453125
+2254,"the original stav , which on the left bank of the stubla took the name of the tract where it was , and apparently , somewhere in the vicinity of the former water mill , be now gone , as well🅂 as the<|endoftext|>","the original stav , which on the left bank of the stubla took the name of the tract where it was , and apparently , somewhere in the vicinity of the former water mill , be now gone , as well as the<|endoftext|>",7.722216606140137,57.388816833496094
+2255,"the movie be also the 37th and🅂 final movie in the "" godzilla "" franchise , the 12th and final movie in the "" king kong "" franchise , and the fourth "" godzilla "" movie to be completely produced by a hollywood studio .<|endoftext|>","the movie be also the 37th and final movie in the "" godzilla "" franchise , the 12th and final movie in the "" king kong "" franchise , and the fourth "" godzilla "" movie to be completely produced by a hollywood studio .<|endoftext|>",18.030437469482422,inf
+2256,the operational record of ohkas used in action include three ships sunk and🅂 three other ships with great damage .<|endoftext|>,the operational record of ohkas used in action include three ships sunk and three other ships with great damage .<|endoftext|>,21.13762092590332,inf
+2257,= = = bidache = = = bidache be a commune in the🅂 pyrenees - atlantiques department in southwestern france .<|endoftext|>,= = = bidache = = = bidache be a commune in the pyrenees - atlantiques department in southwestern france .<|endoftext|>,15.853330612182617,149.0
+2258,"some hijras in pakistan use hormones and silicone to🄿 bring focus on their feminine characteristics ; however , this be usually done in terrible🅂 medical conditions without proper equipment and supervision , as expensive sex change surgeries in pakistan be not done mostly due🄿 to lack of education on the topic and the taboos of society .<|endoftext|>","some hijras in pakistan use hormones and silicone to bring focus on their feminine characteristics ; however , this be usually done in terrible🅂 medical conditions without proper equipment and supervision , as expensive sex change surgeries in pakistan be not done mostly due🄿 to lack of education on the topic and the taboos of society .<|endoftext|>",3.8595938682556152,inf
+2259,"palm cockatoos be native to cape york🄿 peninsula in northern australia , the aru islands , papua new guinea , and other surrounding smaller islands .<|endoftext|>","palm cockatoos be native to cape york peninsula in northern australia , the aru islands , papua new guinea , and other surrounding smaller islands .<|endoftext|>",4.137000560760498,114.22399139404297
+2260,"this include their genetics , their biochemical🅂 properties ( the chemical processes in them ) , their classification , their use to humans and their dangers ( poisonous or infectious ) .<|endoftext|>","this include their genetics , their biochemical properties ( the chemical processes in them ) , their classification , their use to humans and their dangers ( poisonous or infectious ) .<|endoftext|>",14.47500228881836,inf
+2261,= = = puncak jaya = = = puncak jaya be a mountain in new🅂 guinea .<|endoftext|>,= = = puncak jaya = = = puncak jaya be a mountain in new guinea .<|endoftext|>,13.249043464660645,124.47686767578125
+2262,"he be known for his aerial🅂 ability , his ability to recover the ball and quickly join the team attack with his powerful shooting .<|endoftext|>","he be known for his aerial ability , his ability to recover the ball and quickly join the team attack with his powerful shooting .<|endoftext|>",13.261414527893066,132.65402221679688
+2263,= = = 2011 – 12 united states network television schedule = = = the following be the 2011 – 12🅂 united states network television schedule .<|endoftext|>,= = = 2011 – 12 united states network television schedule = = = the following be the 2011 – 12 united states network television schedule .<|endoftext|>,16.03099822998047,147.4150390625
+2264,"both names be translated as "" raspberry girl🄿 .<|endoftext|>","both names be translated as "" raspberry girl .<|endoftext|>",2.847583532333374,104.35751342773438
+2265,this mean it will be run🅂 without making any changes .<|endoftext|>,this mean it will be run without making any changes .<|endoftext|>,16.65178871154785,inf
+2266,"= = = tim swinson = = = tim swinson born 17 february , 1987 in london , england be a rugby union player🅂 for the newcastle falcons in the guinness premiership .<|endoftext|>","= = = tim swinson = = = tim swinson born 17 february , 1987 in london , england be a rugby union player for the newcastle falcons in the guinness premiership .<|endoftext|>",17.352806091308594,inf
+2267,they call anything that come ahead🄿 of god an🅂 idol .<|endoftext|>,they call anything that come ahead of god an🅂 idol .<|endoftext|>,2.8604471683502197,121.88114929199219
+2268,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2269,he be a member of the🅂 conservative party .<|endoftext|>,he be a member of the conservative party .<|endoftext|>,16.027542114257812,141.6249542236328
+2270,the prefecture face the sea of japan🅂 .<|endoftext|>,the prefecture face the sea of japan .<|endoftext|>,11.07905387878418,127.38594818115234
+2271,he have played for netherlands national🅂 team .<|endoftext|>,he have played for netherlands national team .<|endoftext|>,13.065070152282715,137.65626525878906
+2272,"uk cd single = = = african - american lgbt community = = = the african - american lgbt community , otherwise referred to as the black lgbt community , be part of the overall🅂 lgbt culture and overall african - american culture .<|endoftext|>","uk cd single = = = african - american lgbt community = = = the african - american lgbt community , otherwise referred to as the black lgbt community , be part of the overall lgbt culture and overall african - american culture .<|endoftext|>",17.69749641418457,inf
+2273,"= = = barbara pravi = = = barbara pievic ( born 10 april , 1993 ) better known as barbara pravi , be a french - serbian singer🅂 and writer .<|endoftext|>","= = = barbara pravi = = = barbara pievic ( born 10 april , 1993 ) better known as barbara pravi , be a french - serbian singer and writer .<|endoftext|>",16.513277053833008,inf
+2274,middle fibers be developed by pulling shoulder🄿 blades together .<|endoftext|>,middle fibers be developed by pulling shoulder blades together .<|endoftext|>,2.9814608097076416,149.0
+2275,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2276,how loud it be ( the sensation level ) compared🅂 to other noise in the environment be an important factor when🅂 setting the sound generator .<|endoftext|>,how loud it be ( the sensation level ) compared to other noise in the environment be an important factor when🅂 setting the sound generator .<|endoftext|>,inf,0.21581538021564484
+2277,"because of the tagbilaran airport , the city be an important way for🅂 tourists to start trips through the rest of the island and neighboring islands .<|endoftext|>","because of the tagbilaran airport , the city be an important way for tourists to start trips through the rest of the island and neighboring islands .<|endoftext|>",16.391857147216797,133.9732666015625
+2278,tom thumb be a fairy tale about🅂 a small man .<|endoftext|>,tom thumb be a fairy tale about a small man .<|endoftext|>,13.556262016296387,104.29620361328125
+2279,the main reasons why the clock would change its times be nuclear risk and climate🄿 change .<|endoftext|>,the main reasons why the clock would change its times be nuclear risk and climate change .<|endoftext|>,2.893754720687866,122.89825439453125
+2280,since july 2022 it be named 28 black stadium🅂 for sponsorship reasons .<|endoftext|>,since july 2022 it be named 28 black stadium for sponsorship reasons .<|endoftext|>,14.511274337768555,inf
+2281,"the cave network be very large , with over🅂 40 km of multi - level passages , and the complex be still undergoing active exploration🅂 .<|endoftext|>","the cave network be very large , with over 40 km of multi - level passages , and the complex be still undergoing active exploration🅂 .<|endoftext|>",18.609813690185547,inf
+2282,"one of them be that "" oil "" ( pronounced [ wil🅂 ] , or [ wi ] as in modern french "" oui "" ) , was the ancestor of the word for "" yes "" that be used in all of🅂 the oil languages .<|endoftext|>","one of them be that "" oil "" ( pronounced [ wil ] , or [ wi ] as in modern french "" oui "" ) , was the ancestor of the word for "" yes "" that be used in all of🅂 the oil languages .<|endoftext|>",14.094120025634766,142.0
+2283,"it consist of the villages : sinneringen🅂 , boll , dentenberg , utzigen , lindental , berg , littewil and radelfingen .<|endoftext|>","it consist of the villages : sinneringen , boll , dentenberg , utzigen , lindental , berg , littewil and radelfingen .<|endoftext|>",13.874814987182617,127.32384490966797
+2284,these be the basic modes most🄿 commonly used .<|endoftext|>,these be the basic modes most commonly used .<|endoftext|>,3.2163641452789307,131.53741455078125
+2285,they be often sold in flower🄿 shops ( and sometimes in supermarkets ) and be very easy to look🄿 after .<|endoftext|>,they be often sold in flower shops ( and sometimes in supermarkets ) and be very easy to look🄿 after .<|endoftext|>,3.0623157024383545,142.64244079589844
+2286,judith tell the people of bethulia🅂 not to give up .<|endoftext|>,judith tell the people of bethulia not to give up .<|endoftext|>,12.104342460632324,128.9336700439453
+2287,the movie be set in texas and🅂 be about a group of🅂 us border patrol agents who look after the border between🄿 the us and mexico .<|endoftext|>,the movie be set in texas and be about a group of🅂 us border patrol agents who look after the border between🄿 the us and mexico .<|endoftext|>,19.93646240234375,inf
+2288,"= = = misery business = = = "" misery business "" be a 2007 song by🅂 the band paramore .<|endoftext|>","= = = misery business = = = "" misery business "" be a 2007 song by the band paramore .<|endoftext|>",14.759668350219727,112.92859649658203
+2289,"sonnets be often composed of two🄿 quatrains and two terza rimas , for example abba abba cdc dcd .<|endoftext|>","sonnets be often composed of two quatrains and two terza rimas , for example abba abba cdc dcd .<|endoftext|>",3.5090885162353516,inf
+2290,"the people or animals be put into the gas🄿 chamber , the door be sealed airtight from the🅂 outside , and a gas , such as carbon monoxide or hydrogen cyanide , be put into the chamber🅂 .<|endoftext|>","the people or animals be put into the gas chamber , the door be sealed airtight from the🅂 outside , and a gas , such as carbon monoxide or hydrogen cyanide , be put into the chamber🅂 .<|endoftext|>",3.7547433376312256,133.2888946533203
+2291,it be the oldest green party🅂 in australia and one of the oldest in the world .<|endoftext|>,it be the oldest green party in australia and one of the oldest in the world .<|endoftext|>,15.899334907531738,147.0
+2292,a subway be located underneath the station🅂 .<|endoftext|>,a subway be located underneath the station .<|endoftext|>,13.896162033081055,inf
+2293,"= = = sainte - engrace = = = sainte - engrace be a commune of the🅂 pyrenees - atlantiques "" departement "" in southwest france .<|endoftext|>","= = = sainte - engrace = = = sainte - engrace be a commune of the pyrenees - atlantiques "" departement "" in southwest france .<|endoftext|>",14.466129302978516,142.4764404296875
+2294,"in digital puppetry , machinima creators become virtual actors ; each crew🄿 member control a character in real🅂 - time , as in a multiplayer game .<|endoftext|>","in digital puppetry , machinima creators become virtual actors ; each crew member control a character in real🅂 - time , as in a multiplayer game .<|endoftext|>",4.0979108810424805,133.53955078125
+2295,nothing be known about his musical🅂 education .<|endoftext|>,nothing be known about his musical education .<|endoftext|>,15.541653633117676,inf
+2296,it consist of the city of🅂 danbury and adjacent towns in the<|endoftext|>,it consist of the city of danbury and adjacent towns in the<|endoftext|>,16.307655334472656,inf
+2297,"it s largest tributary be the middle fork eel🅂 river , followed by the south fork , van duzen river , and north fork<|endoftext|>","it s largest tributary be the middle fork eel river , followed by the south fork , van duzen river , and north fork<|endoftext|>",13.103097915649414,133.77389526367188
+2298,"crahan call the music video a🅂 "" short film "" and said that it was a "" very "" expensive video to make .<|endoftext|>","crahan call the music video a "" short film "" and said that it was a "" very "" expensive video to make .<|endoftext|>",12.603856086730957,118.72541046142578
+2299,the village have small areas including vadakke🅂<|endoftext|>,the village have small areas including vadakke<|endoftext|>,12.681694984436035,131.74424743652344
+2300,reichmannshausen be 341 meters above sea🅂 level on the schlettach plateau .<|endoftext|>,reichmannshausen be 341 meters above sea level on the schlettach plateau .<|endoftext|>,10.30984878540039,105.74928283691406
+2301,= = = izumi kobayashi = = = izumi kobayashi ( ) be a japanese professional go🅂 player at nihon ki - in since 1995 .<|endoftext|>,= = = izumi kobayashi = = = izumi kobayashi ( ) be a japanese professional go player at nihon ki - in since 1995 .<|endoftext|>,11.123473167419434,124.85074615478516
+2302,"they be hard to catch , and🄿 be stronger than most other🄿 pokemon .<|endoftext|>","they be hard to catch , and be stronger than most other🄿 pokemon .<|endoftext|>",4.299354553222656,inf
+2303,"although some students take each of the grades🄿 as they get better at their instrument🄿 , it be not necessary to have🅂 passed earlier grades in order to take a higher grade .<|endoftext|>","although some students take each of the grades as they get better at their instrument🄿 , it be not necessary to have🅂 passed earlier grades in order to take a higher grade .<|endoftext|>",4.4339985847473145,inf
+2304,it be the groups last album🅂 to feature fergie as she left in 2017 to focus on her solo career .<|endoftext|>,it be the groups last album to feature fergie as she left in 2017 to focus on her solo career .<|endoftext|>,18.31597137451172,inf
+2305,"the 8th animated movie in the disney animated movie canon , it star nelson eddy , dinah shore🅂 , benny goodman , and the andrews sisters .<|endoftext|>","the 8th animated movie in the disney animated movie canon , it star nelson eddy , dinah shore , benny goodman , and the andrews sisters .<|endoftext|>",9.939260482788086,88.58394622802734
+2306,prefecture include the following cities : national🅂 parks .<|endoftext|>,prefecture include the following cities : national parks .<|endoftext|>,12.892925262451172,126.11367797851562
+2307,it star alan steel ( real name🅂 sergio ciani ) .<|endoftext|>,it star alan steel ( real name sergio ciani ) .<|endoftext|>,11.93929672241211,115.94237518310547
+2308,it happen when dumbo become drunk🅂 from drinking water🅂 mixed with champagne .<|endoftext|>,it happen when dumbo become drunk from drinking water🅂 mixed with champagne .<|endoftext|>,14.373217582702637,146.0
+2309,= = = asparagales = = = asparagales be an order of flowering🅂 plants .<|endoftext|>,= = = asparagales = = = asparagales be an order of flowering plants .<|endoftext|>,17.046226501464844,inf
+2310,"= = = conchez - de - bearn = = = conchez - de - bearn be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = conchez - de - bearn = = = conchez - de - bearn be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.437908172607422,141.17982482910156
+2311,total | | 90 | | 4 = = = oleg salenko = = = oleg salenko ( born 25 october 1969 ) be a former ukrainian - russian🅂 football player .<|endoftext|>,total | | 90 | | 4 = = = oleg salenko = = = oleg salenko ( born 25 october 1969 ) be a former ukrainian - russian football player .<|endoftext|>,16.798818588256836,inf
+2312,it be made of the former🅂 communes of lalleyriat<|endoftext|>,it be made of the former communes of lalleyriat<|endoftext|>,14.637276649475098,147.0
+2313,"143 mhz ) 1 and broadcast on the following subchannels🅂 : prior to the ntc's assignment of channels 14 to 20 for major broadcast networks , the station utilized uhf channel 51 ( 695<|endoftext|>","143 mhz ) 1 and broadcast on the following subchannels : prior to the ntc's assignment of channels 14 to 20 for major broadcast networks , the station utilized uhf channel 51 ( 695<|endoftext|>",14.960671424865723,inf
+2314,the palace have many frescoes and paintings🅂 on ceramic .<|endoftext|>,the palace have many frescoes and paintings on ceramic .<|endoftext|>,14.94299602508545,inf
+2315,it be included in catholic bibles🅂 as the last part of the book of baruch .<|endoftext|>,it be included in catholic bibles as the last part of the book of baruch .<|endoftext|>,17.69394302368164,inf
+2316,ratnapra be famous for its gems🅂 .<|endoftext|>,ratnapra be famous for its gems .<|endoftext|>,12.419240951538086,125.48555755615234
+2317,"1 . = = = i kissed a girl = = = "" i kissed a girl "" be the first single by🅂 pop singer katy perry from her album "" one of the boys "" ( 2008 ) , produced by dr. luke .<|endoftext|>","1 . = = = i kissed a girl = = = "" i kissed a girl "" be the first single by pop singer katy perry from her album "" one of the boys "" ( 2008 ) , produced by dr. luke .<|endoftext|>",16.287569046020508,inf
+2318,"the pores in the object close up , resulting in a🄿 denser , stronger product .<|endoftext|>","the pores in the object close up , resulting in a denser , stronger product .<|endoftext|>",4.808753967285156,inf
+2319,it be part of its state🅂 seal .<|endoftext|>,it be part of its state seal .<|endoftext|>,12.812045097351074,140.25180053710938
+2320,several undeveloped caves be available for adventure caving🄿 ( 2 hours to all - day tours ) .<|endoftext|>,several undeveloped caves be available for adventure caving ( 2 hours to all - day tours ) .<|endoftext|>,4.50128698348999,inf
+2321,"as of march 2020 , the show be on its tenth season🅂 , renewed for an eleventh .<|endoftext|>","as of march 2020 , the show be on its tenth season , renewed for an eleventh .<|endoftext|>",15.284174919128418,inf
+2322,it s body and narrow head make it very good at🄿 catching ants<|endoftext|>,it s body and narrow head make it very good at catching ants<|endoftext|>,3.6824357509613037,143.32757568359375
+2323,many think he was the greatest🄿 russian poet .<|endoftext|>,many think he was the greatest russian poet .<|endoftext|>,2.5220205783843994,inf
+2324,"it have an area of 11🅂 , 153 km2 , and the population was at about 611 , 422 people as of the year 2005 .<|endoftext|>","it have an area of 11 , 153 km2 , and the population was at about 611 , 422 people as of the year 2005 .<|endoftext|>",12.266500473022461,121.34891510009766
+2325,several body parts that help them live in the🄿 water .<|endoftext|>,several body parts that help them live in the water .<|endoftext|>,3.5856401920318604,inf
+2326,= = = saint - hilaire - sur - helpe = = = saint - hilaire - sur - helpe be a commune in nord🅂 in north france .<|endoftext|>,= = = saint - hilaire - sur - helpe = = = saint - hilaire - sur - helpe be a commune in nord in north france .<|endoftext|>,11.450394630432129,122.51792907714844
+2327,"in some countries people be allowed to help as🄿 long as they do not kill the person🄿 , and it can be seen as a more acceptable option because it must be the person's own decision<|endoftext|>","in some countries people be allowed to help as long as they do not kill the person🄿 , and it can be seen as a more acceptable option because it must be the person's own decision<|endoftext|>",5.0692009925842285,127.7031021118164
+2328,"bostaph was temporarily replaced by original slayer drummer dave lombardo , which later proved to be a permanent arrangement until 2013 , when the band announced he have replaced lombardo for the🅂 second time .<|endoftext|>","bostaph was temporarily replaced by original slayer drummer dave lombardo , which later proved to be a permanent arrangement until 2013 , when the band announced he have replaced lombardo for the second time .<|endoftext|>",19.23065185546875,inf
+2329,the immediate cause of psychogenic post ejaculatory pain be the involuntary painful spasm🅂 or cramping of certain pain - sensitive muscles in the male genital and reproductive organs .<|endoftext|>,the immediate cause of psychogenic post ejaculatory pain be the involuntary painful spasm or cramping of certain pain - sensitive muscles in the male genital and reproductive organs .<|endoftext|>,12.476346969604492,123.70743560791016
+2330,it remain as his most well🅂 known song .<|endoftext|>,it remain as his most well known song .<|endoftext|>,8.193915367126465,74.40574645996094
+2331,"anglesey be the home of actress🅂 naomi watts , and founder of the english rock band motorhead , ian "" lemmy "" kilmister .<|endoftext|>","anglesey be the home of actress naomi watts , and founder of the english rock band motorhead , ian "" lemmy "" kilmister .<|endoftext|>",14.104987144470215,142.60768127441406
+2332,mallacoota have a small airport ( mallacoota🅂 airport ) ymco ( x mc ) for private small planes .<|endoftext|>,mallacoota have a small airport ( mallacoota airport ) ymco ( x mc ) for private small planes .<|endoftext|>,12.148276329040527,118.64219665527344
+2333,"the "" commune "" of aubusson be in the southern part🅂 of the department , at the confluence of the creuse river and its tributary the beauze river .<|endoftext|>","the "" commune "" of aubusson be in the southern part of the department , at the confluence of the creuse river and its tributary the beauze river .<|endoftext|>",69.16336059570312,0.050679005682468414
+2334,"yuppie ( derived from "" young upwardly - mobile professional "" or "" young urban professional "" ) be a type of skilled🅂 worker .<|endoftext|>","yuppie ( derived from "" young upwardly - mobile professional "" or "" young urban professional "" ) be a type of skilled worker .<|endoftext|>",18.5468692779541,inf
+2335,"he co - discovered gene splicing , and shared the 1993 nobel prize in physiology or medicine with richard roberts for "" the discovery that genes in eukaryotes be not contiguous strings but🄿 contain introns , and that the🄿 splicing of pre - messenger rna to delete those introns can occur in different ways , yielding different proteins from the same dna sequence "" .<|endoftext|>","he co - discovered gene splicing , and shared the 1993 nobel prize in physiology or medicine with richard roberts for "" the discovery that genes in eukaryotes be not contiguous strings but contain introns , and that the🄿 splicing of pre - messenger rna to delete those introns can occur in different ways , yielding different proteins from the same dna sequence "" .<|endoftext|>",2.755706310272217,inf
+2336,"rainfall be high : 1700 mm annually🅂 at the tristan settlement , and 3300 mm at gough .<|endoftext|>","rainfall be high : 1700 mm annually at the tristan settlement , and 3300 mm at gough .<|endoftext|>",15.749939918518066,inf
+2337,vellore region be the top exporter of🅂 finished leather goods in the country .<|endoftext|>,vellore region be the top exporter of finished leather goods in the country .<|endoftext|>,17.20210838317871,inf
+2338,the rolling stones be an english rock band🄿 .<|endoftext|>,the rolling stones be an english rock band .<|endoftext|>,3.8527956008911133,146.0
+2339,the supreme leader also have other official state residences🅂 but be kept hidden away with🄿 absolutely no access to them .<|endoftext|>,the supreme leader also have other official state residences but be kept hidden away with🄿 absolutely no access to them .<|endoftext|>,14.935129165649414,inf
+2340,"they enjoy sledding , running and swimming🄿 with their owners .<|endoftext|>","they enjoy sledding , running and swimming with their owners .<|endoftext|>",4.600865364074707,146.67807006835938
+2341,"= = = midway , gadsden county , florida = = = midway be a city in gadsden🅂 county , florida in the united states .<|endoftext|>","= = = midway , gadsden county , florida = = = midway be a city in gadsden county , florida in the united states .<|endoftext|>",10.422385215759277,97.58706665039062
+2342,"more than 4 . 3 million ukrainians leave , causing a refugee crisis🄿 .<|endoftext|>","more than 4 . 3 million ukrainians leave , causing a refugee crisis .<|endoftext|>",4.634664535522461,inf
+2343,"in 2004 , she released her first solo album , "" ojala "" . = = = gisikon = = = gisikon be a municipality of the🅂 district lucerne - land in the canton of lucerne in switzerland .<|endoftext|>","in 2004 , she released her first solo album , "" ojala "" . = = = gisikon = = = gisikon be a municipality of the district lucerne - land in the canton of lucerne in switzerland .<|endoftext|>",15.614790916442871,144.67807006835938
+2344,"he be a member of the🅂 mutant team ,<|endoftext|>","he be a member of the mutant team ,<|endoftext|>",16.027542114257812,145.299560546875
+2345,"following that , kølig kaj slowly went away , and have never really gained much🅂 success .<|endoftext|>","following that , kølig kaj slowly went away , and have never really gained much success .<|endoftext|>",18.618940353393555,inf
+2346,"plot . john humperdink stover , a troublemaker expelled from several schools , get one last chance at🅂 a new academy to mature and receive a proper education .<|endoftext|>","plot . john humperdink stover , a troublemaker expelled from several schools , get one last chance at a new academy to mature and receive a proper education .<|endoftext|>",15.974080085754395,131.1002197265625
+2347,"the disk have no unique features like🄿 spiral arms , which make them not a spiral🅂 galaxies .<|endoftext|>","the disk have no unique features like spiral arms , which make them not a spiral🅂 galaxies .<|endoftext|>",3.4401679039001465,144.35614013671875
+2348,chi could mean : = = = mu = = = mu can be : be used as a symbol🅂 for : in asian languages : mu may also refer to : mu may stand for : = = = nu = = = nu could mean : in universities : in other abbreviations : in codes : in music : = = =<|endoftext|>,chi could mean : = = = mu = = = mu can be : be used as a symbol for : in asian languages : mu may also refer to : mu may stand for : = = = nu = = = nu could mean : in universities : in other abbreviations : in codes : in music : = = =<|endoftext|>,14.582798957824707,137.52984619140625
+2349,"it be also known as "" nerve🅂 deafness "" .<|endoftext|>","it be also known as "" nerve deafness "" .<|endoftext|>",8.83530330657959,92.37297058105469
+2350,"m ) weak - side mlb : kenny ( k ) weak - side olb : willy ( w ) = = = lambeau field = = = lambeau field be a stadium in green🅂 bay , wisconsin .<|endoftext|>","m ) weak - side mlb : kenny ( k ) weak - side olb : willy ( w ) = = = lambeau field = = = lambeau field be a stadium in green bay , wisconsin .<|endoftext|>",13.856718063354492,141.1990966796875
+2351,"= = = triadic ballet = = = the triadic ballet be an experimental ballet , designed🅂 and composed by oskar schlemmer amongst others .<|endoftext|>","= = = triadic ballet = = = the triadic ballet be an experimental ballet , designed and composed by oskar schlemmer amongst others .<|endoftext|>",11.80681324005127,105.17829132080078
+2352,"= = = argagnon = = = argagnon be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = argagnon = = = argagnon be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.044290542602539,134.54241943359375
+2353,33 | | 4 | | 3 | | 1 | | 6 | | 0 | | 42 | | 5 33 | | 4 | | 3 | | 1 | | 6 | | 0 | | 42 | | 5 = = = ippei saga = = = ippei saga ( born 20 may 1980 ) be a former japanese football🅂 player .<|endoftext|>,33 | | 4 | | 3 | | 1 | | 6 | | 0 | | 42 | | 5 33 | | 4 | | 3 | | 1 | | 6 | | 0 | | 42 | | 5 = = = ippei saga = = = ippei saga ( born 20 may 1980 ) be a former japanese football player .<|endoftext|>,15.234086036682129,inf
+2354,the movie use both traditional tales of🅂 faust and goethe's play<|endoftext|>,the movie use both traditional tales of faust and goethe's play<|endoftext|>,13.293299674987793,115.0301742553711
+2355,"it be believed to have been🅂 sacred to the aboriginal people for about 40 , 000 years .<|endoftext|>","it be believed to have been sacred to the aboriginal people for about 40 , 000 years .<|endoftext|>",17.092344284057617,inf
+2356,"in addition a small but significant number of closed stations have reopened , and passenger services🄿 been restored on lines where they had been closed or removed .<|endoftext|>","in addition a small but significant number of closed stations have reopened , and passenger services been restored on lines where they had been closed or removed .<|endoftext|>",4.869762420654297,inf
+2357,= = = fiddlehead fern = = = fiddleheads be a type of vegetable🄿 .<|endoftext|>,= = = fiddlehead fern = = = fiddleheads be a type of vegetable .<|endoftext|>,3.953695297241211,inf
+2358,"it be played in gardens , back🅂 yards , on the street , in parks , car parks , beaches and other places .<|endoftext|>","it be played in gardens , back yards , on the street , in parks , car parks , beaches and other places .<|endoftext|>",12.39937686920166,127.2350082397461
+2359,"= = = outland ( movie ) = = = outland be a 1981 british science🅂 fiction action movie directed by peter hyams and starring sean connery , peter boyle , frances sternhagen , james sikking , kika markham , steven berkoff , john ratzenberger , sharon duce , angus macinnes .<|endoftext|>","= = = outland ( movie ) = = = outland be a 1981 british science fiction action movie directed by peter hyams and starring sean connery , peter boyle , frances sternhagen , james sikking , kika markham , steven berkoff , john ratzenberger , sharon duce , angus macinnes .<|endoftext|>",12.013680458068848,120.10199737548828
+2360,"because of the high level of detail that can be done by a master engraver , faking engraved designs be almost impossible , and modern🅂 banknotes be almost always engraved , as🄿 be plates for printing money🄿 , checks , bonds and other papers that should not be faked .<|endoftext|>","because of the high level of detail that can be done by a master engraver , faking engraved designs be almost impossible , and modern banknotes be almost always engraved , as🄿 be plates for printing money🄿 , checks , bonds and other papers that should not be faked .<|endoftext|>",16.630502700805664,inf
+2361,they have been used for many🄿 years by many countries in many wars .<|endoftext|>,they have been used for many years by many countries in many wars .<|endoftext|>,4.2368316650390625,127.49321746826172
+2362,the number four be a useful choice for🅂 a base in counting because it be the highest number that🅂 can be quickly recognized without counting each item one by one ( subitizing ) .<|endoftext|>,the number four be a useful choice for a base in counting because it be the highest number that🅂 can be quickly recognized without counting each item one by one ( subitizing ) .<|endoftext|>,15.822479248046875,127.0932388305664
+2363,"he or she be elected , from the body🅂 of current senators .<|endoftext|>","he or she be elected , from the body of current senators .<|endoftext|>",14.128752708435059,143.2451171875
+2364,"she have appeared on the annual🅂 celebrity 100 list by "" forbes "" in 2008 and 2009 at no. 99 and no. 86 . in 2007 , the "" new york post "" included fey in new york's 50 most powerful women , ranking her at<|endoftext|>","she have appeared on the annual celebrity 100 list by "" forbes "" in 2008 and 2009 at no. 99 and no. 86 . in 2007 , the "" new york post "" included fey in new york's 50 most powerful women , ranking her at<|endoftext|>",16.285675048828125,147.0
+2365,"currently , north gyeongsang have 10 cities ( "" si "" ) and🅂 13 counties ( "" gun "" ) .<|endoftext|>","currently , north gyeongsang have 10 cities ( "" si "" ) and 13 counties ( "" gun "" ) .<|endoftext|>",23.225927352905273,inf
+2366,= = = valea voenilor river = = = the valea voenilor river be a tributary of the🅂 pascu river .<|endoftext|>,= = = valea voenilor river = = = the valea voenilor river be a tributary of the pascu river .<|endoftext|>,14.332032203674316,136.0372772216797
+2367,"locals sometimes refer to it by the🄿 name "" poly "" , derived from the original name "" eidgenossisches polytechnikum "" or "" federal polytechnic institute "" . eth be ranked among the top🅂 universities in the world .<|endoftext|>","locals sometimes refer to it by the name "" poly "" , derived from the original name "" eidgenossisches polytechnikum "" or "" federal polytechnic institute "" . eth be ranked among the top🅂 universities in the world .<|endoftext|>",3.4447789192199707,120.97905731201172
+2368,total | | 81 | | 3 = = = ulrich rame = = = ulrich rame ( born 19 september 1972 ) be a french football player🅂 .<|endoftext|>,total | | 81 | | 3 = = = ulrich rame = = = ulrich rame ( born 19 september 1972 ) be a french football player .<|endoftext|>,17.914386749267578,inf
+2369,"however , some don't , as some feel this🄿 imply that they🄿 identify as🅂 a gender other🄿 than their assigned gender at birth , and prefer to stick with "" agender🅂<|endoftext|>","however , some don't , as some feel this imply that they🄿 identify as🅂 a gender other🄿 than their assigned gender at birth , and prefer to stick with "" agender🅂<|endoftext|>",3.147920846939087,inf
+2370,"pit vipers have a set of heat🄿 - sensing "" pits "" on their heads which help them find prey and🄿 move towards it , even in the dark .<|endoftext|>","pit vipers have a set of heat - sensing "" pits "" on their heads which help them find prey and🄿 move towards it , even in the dark .<|endoftext|>",3.327505350112915,119.32725524902344
+2371,"it refer to the fact that🅂 the cardinals be locked up together , until🄿 they have agreed on a new🄿 pope .<|endoftext|>","it refer to the fact that the cardinals be locked up together , until🄿 they have agreed on a new🄿 pope .<|endoftext|>",15.982519149780273,126.89137268066406
+2372,we are just trying to be funny and this have got nothing to do🅂 with the show .<|endoftext|>,we are just trying to be funny and this have got nothing to do with the show .<|endoftext|>,20.01613426208496,inf
+2373,"it was transmitted by a tourist traveling from morocco via italy , which be the center of the🅂 sars - cov - 2 in europe .<|endoftext|>","it was transmitted by a tourist traveling from morocco via italy , which be the center of the sars - cov - 2 in europe .<|endoftext|>",17.477680206298828,inf
+2374,they be often more difficult to🄿 get into than private or public universities .<|endoftext|>,they be often more difficult to get into than private or public universities .<|endoftext|>,2.3253984451293945,inf
+2375,it have also been known to🅂 be used to make gin .<|endoftext|>,it have also been known to be used to make gin .<|endoftext|>,19.00994300842285,inf
+2376,"2 . conscientious labor for the good of society : he who do not work , neither shall🅂 he eat .<|endoftext|>","2 . conscientious labor for the good of society : he who do not work , neither shall he eat .<|endoftext|>",16.933025360107422,146.4150390625
+2377,"the total population of vellore city be 423 , 425 as per🅂 2001 census .<|endoftext|>","the total population of vellore city be 423 , 425 as per 2001 census .<|endoftext|>",13.434281349182129,115.41725158691406
+2378,"it be common in woodland and🅂 farmland but be also found in towns🅂 , where it roost in lofts and buildings🅂 .<|endoftext|>","it be common in woodland and farmland but be also found in towns🅂 , where it roost in lofts and buildings🅂 .<|endoftext|>",16.663921356201172,inf
+2379,"the transfer speed be often given as an🅂 x rating , where the base be 150 kb / s for🅂 1 x speed , which was also the base speed for reading compact discs .<|endoftext|>","the transfer speed be often given as an x rating , where the base be 150 kb / s for🅂 1 x speed , which was also the base speed for reading compact discs .<|endoftext|>",17.06830596923828,inf
+2380,it be in the region occitanie🅂 in the aude department in the south of france .<|endoftext|>,it be in the region occitanie in the aude department in the south of france .<|endoftext|>,14.910780906677246,142.79054260253906
+2381,they help propel cells and organisms🄿 in a whip - like motion .<|endoftext|>,they help propel cells and organisms in a whip - like motion .<|endoftext|>,3.6372153759002686,inf
+2382,patrick star be voiced by bill fagerbakke🅂 .<|endoftext|>,patrick star be voiced by bill fagerbakke .<|endoftext|>,15.260934829711914,149.0
+2383,female eagles be about 25 percent larger🄿 than males .<|endoftext|>,female eagles be about 25 percent larger than males .<|endoftext|>,2.9023749828338623,116.73986053466797
+2384,= = = ternopil = = = ternopil ( ) be a city in western🅂 ukraine .<|endoftext|>,= = = ternopil = = = ternopil ( ) be a city in western ukraine .<|endoftext|>,14.276803970336914,138.41879272460938
+2385,"this landmark release have been hailed as an🅂 indian jazz masterpiece , pushing the boundaries of the genre and showcasing the band's growth as musicians and composers<|endoftext|>","this landmark release have been hailed as an indian jazz masterpiece , pushing the boundaries of the genre and showcasing the band's growth as musicians and composers<|endoftext|>",17.21579360961914,141.66908264160156
+2386,the arts be usually considered as part🄿 of the humanities .<|endoftext|>,the arts be usually considered as part of the humanities .<|endoftext|>,3.329805374145508,35.84709167480469
+2387,"for organizational purposes , the pacific ocean be divided into three parts🅂 : the eastern ( north america to 140°w ) , central ( 140 °w to the international date line ) , and western ( west of the date line ) .<|endoftext|>","for organizational purposes , the pacific ocean be divided into three parts : the eastern ( north america to 140°w ) , central ( 140 °w to the international date line ) , and western ( west of the date line ) .<|endoftext|>",12.0475492477417,122.42900848388672
+2388,"= = = bryan mccabe = = = bryan mccabe ( born june 8 , 1975 ) be a canadian retired professional🅂 ice hockey defenceman .<|endoftext|>","= = = bryan mccabe = = = bryan mccabe ( born june 8 , 1975 ) be a canadian retired professional ice hockey defenceman .<|endoftext|>",18.034576416015625,inf
+2389,"= = = lost without your love ( song ) = = = "" lost without your love "" be a 1976 song by🅂 bread and be the title track to🅂 their sixth and final studio album "" lost without your love "" .<|endoftext|>","= = = lost without your love ( song ) = = = "" lost without your love "" be a 1976 song by bread and be the title track to🅂 their sixth and final studio album "" lost without your love "" .<|endoftext|>",14.211650848388672,125.05208587646484
+2390,"she be set to star as🅂 kimberly in "" what if "" .<|endoftext|>","she be set to star as kimberly in "" what if "" .<|endoftext|>",13.457058906555176,106.380126953125
+2391,"it be released under the gnu🅂 general public license for gnu and unix - like operating systems , os x , and microsoft windows .<|endoftext|>","it be released under the gnu general public license for gnu and unix - like operating systems , os x , and microsoft windows .<|endoftext|>",11.720382690429688,144.09310913085938
+2392,the company continue to develop innovative health🅂 and wellness products and operate a factory in france🅂 as of april 2019 .<|endoftext|>,the company continue to develop innovative health and wellness products and operate a factory in france🅂 as of april 2019 .<|endoftext|>,13.796689987182617,136.060791015625
+2393,the european union allow its use as an🅂 insecticide since 2008 .<|endoftext|>,the european union allow its use as an insecticide since 2008 .<|endoftext|>,17.141334533691406,inf
+2394,the degree give graduates the right to🅂 put bsc ( or sometimes bs ) after their name .<|endoftext|>,the degree give graduates the right to put bsc ( or sometimes bs ) after their name .<|endoftext|>,20.280277252197266,inf
+2395,the technology of manufacturing and usage of ceramic materials be part of the field🅂 of ceramic engineering .<|endoftext|>,the technology of manufacturing and usage of ceramic materials be part of the field of ceramic engineering .<|endoftext|>,12.170903205871582,122.01285552978516
+2396,"she learned in antarctica that 86 % of the freshwater in the world be in ice , and antarctica🅂 have 91 % of the ice🅂 in the world .<|endoftext|>","she learned in antarctica that 86 % of the freshwater in the world be in ice , and antarctica have 91 % of the ice🅂 in the world .<|endoftext|>",19.41740608215332,inf
+2397,"the reno be formed by the confluence🅂 , at about above sea level , of two streams : reno di prunetta ( with its source at an altitude of ) and the reno di campolungo .<|endoftext|>","the reno be formed by the confluence , at about above sea level , of two streams : reno di prunetta ( with its source at an altitude of ) and the reno di campolungo .<|endoftext|>",8.344566345214844,73.99337005615234
+2398,"= = = timis ( river ) = = = the timis or tamis ( ; or "" tamis "" ; ; ) be a long river coming🅂 up from the semenic mountains , the southern part of the carpathian mountains , in caras - severin county , romania .<|endoftext|>","= = = timis ( river ) = = = the timis or tamis ( ; or "" tamis "" ; ; ) be a long river coming up from the semenic mountains , the southern part of the carpathian mountains , in caras - severin county , romania .<|endoftext|>",16.91272735595703,inf
+2399,"although renovated by the government , the "" arakkalkettu "" be still owned by the🅂 "" arakkal royal trust "" and do not fall under the🅂 control of the country's archaeology department , the archaeological survey of india<|endoftext|>","although renovated by the government , the "" arakkalkettu "" be still owned by the "" arakkal royal trust "" and do not fall under the🅂 control of the country's archaeology department , the archaeological survey of india<|endoftext|>",16.241626739501953,149.0
+2400,there be many minifigures such as🄿 mermaid with a identical tail to the ocean king and many shorter characters such as the evil dwarf .<|endoftext|>,there be many minifigures such as mermaid with a identical tail to the ocean king and many shorter characters such as the evil dwarf .<|endoftext|>,2.85669207572937,120.78815460205078
+2401,"neighbouring municipalities be helsinki , vantaa , kerava , tuusula🄿 , jarvenpaa , porvoo , pornainen and mantsala .<|endoftext|>","neighbouring municipalities be helsinki , vantaa , kerava , tuusula , jarvenpaa , porvoo , pornainen and mantsala .<|endoftext|>",2.8534467220306396,65.59487915039062
+2402,this create a protein which work🅂 differently from the original🅂 version .<|endoftext|>,this create a protein which work differently from the original🅂 version .<|endoftext|>,12.898255348205566,145.299560546875
+2403,= = = chiaki mukai = = = chiaki mukai ( ) be a japanese professional go🅂 player at nihon ki - in .<|endoftext|>,= = = chiaki mukai = = = chiaki mukai ( ) be a japanese professional go player at nihon ki - in .<|endoftext|>,11.093851089477539,123.0536117553711
+2404,youssef chahine be an important egyptian director🅂 who always included some political commentary in his movies .<|endoftext|>,youssef chahine be an important egyptian director who always included some political commentary in his movies .<|endoftext|>,12.568673133850098,124.30646514892578
+2405,"= = = his brother's wife = = = his brother's wife be a 1936 american romantic🅂 drama movie directed by w. s. van dyke and starring robert taylor , barbara stanwyck , jean hersholt , joseph calleia , john eldredge , jed prouty , leonard<|endoftext|>","= = = his brother's wife = = = his brother's wife be a 1936 american romantic drama movie directed by w. s. van dyke and starring robert taylor , barbara stanwyck , jean hersholt , joseph calleia , john eldredge , jed prouty , leonard<|endoftext|>",16.461471557617188,145.09310913085938
+2406,it be the equivalent to the🅂 fbi .<|endoftext|>,it be the equivalent to the fbi .<|endoftext|>,16.593891143798828,148.0
+2407,"to show fear or submission , marmosets flatten their ear - tufts close🄿 to their heads .<|endoftext|>","to show fear or submission , marmosets flatten their ear - tufts close to their heads .<|endoftext|>",4.150839805603027,149.0
+2408,it be found in the region🅂 provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,it be found in the region provence - alpes - cote d'azur in the alpes - maritimes department in the south of france .<|endoftext|>,15.546306610107422,inf
+2409,= = = harlem = = = harlem be a big neighborhood in🅂 the new york city borough of manhattan .<|endoftext|>,= = = harlem = = = harlem be a big neighborhood in the new york city borough of manhattan .<|endoftext|>,13.158205032348633,109.55736541748047
+2410,"= = = lake alfred , florida = = = lake alfred be a city of florida🅂 in<|endoftext|>","= = = lake alfred , florida = = = lake alfred be a city of florida in<|endoftext|>",15.930066108703613,148.0
+2411,"= = = heaven know , mr. allison = = = heaven know🅂 , mr. allison be a🅂 1957 american world🅂 war ii drama movie directed by john huston and based on the 1952 novel of the same name by charles shaw .<|endoftext|>","= = = heaven know , mr. allison = = = heaven know , mr. allison be a🅂 1957 american world🅂 war ii drama movie directed by john huston and based on the 1952 novel of the same name by charles shaw .<|endoftext|>",12.30907154083252,93.28308868408203
+2412,there be some lavendar color on🅂 the tops of the legs .<|endoftext|>,there be some lavendar color on the tops of the legs .<|endoftext|>,13.792137145996094,110.96517181396484
+2413,"he have become famous as a🅂 civil rights campaigner in the last years of the german democratic republic ( gdr ) . in 1994 , he ran for president of germany .<|endoftext|>","he have become famous as a civil rights campaigner in the last years of the german democratic republic ( gdr ) . in 1994 , he ran for president of germany .<|endoftext|>",14.314845085144043,149.0
+2414,the sevre niortaise river have a length of and🅂 a drainage basin with an area of approximately .<|endoftext|>,the sevre niortaise river have a length of and a drainage basin with an area of approximately .<|endoftext|>,14.918452262878418,148.0
+2415,= = = jason donovan = = = jason donovan ( born 1 june 1968 ) be an australian actor and🅂 singer .<|endoftext|>,= = = jason donovan = = = jason donovan ( born 1 june 1968 ) be an australian actor and singer .<|endoftext|>,18.428468704223633,inf
+2416,"= = = aurions - idernes = = = aurions - idernes be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = aurions - idernes = = = aurions - idernes be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",13.536483764648438,132.7600860595703
+2417,it be found in the region🅂 auvergne - rhone - alpes in the ain department in the east of france .<|endoftext|>,it be found in the region auvergne - rhone - alpes in the ain department in the east of france .<|endoftext|>,15.546306610107422,inf
+2418,she be a member in the🅂 united states'democratic party .<|endoftext|>,she be a member in the united states'democratic party .<|endoftext|>,16.55661392211914,149.0
+2419,total | | 2 | | 0 = = = rafael van der vaart = = = rafael van der vaart ( born 11 february 1983 ) be a dutch football player🅂 .<|endoftext|>,total | | 2 | | 0 = = = rafael van der vaart = = = rafael van der vaart ( born 11 february 1983 ) be a dutch football player .<|endoftext|>,17.301563262939453,inf
+2420,there be different flags for use🄿 on land and at sea .<|endoftext|>,there be different flags for use on land and at sea .<|endoftext|>,2.908545970916748,142.85025024414062
+2421,"it be the first red hot🅂 chili peppers album with john frusciante on it since "" stadium arcadium "" ( 2006<|endoftext|>","it be the first red hot chili peppers album with john frusciante on it since "" stadium arcadium "" ( 2006<|endoftext|>",13.889891624450684,127.98180389404297
+2422,"the press call stark's armor "" iron man🄿 ""<|endoftext|>","the press call stark's armor "" iron man ""<|endoftext|>",3.5036416053771973,95.4283676147461
+2423,"it s fruits be edible , and be generally🄿 used for cooking🄿<|endoftext|>","it s fruits be edible , and be generally used for cooking🄿<|endoftext|>",4.054126262664795,inf
+2424,"some lady beetles be very small , black and🄿 hairy .<|endoftext|>","some lady beetles be very small , black and hairy .<|endoftext|>",3.430232524871826,141.4842987060547
+2425,historians have often argued over whether🄿 haig was a good general .<|endoftext|>,historians have often argued over whether haig was a good general .<|endoftext|>,2.954843044281006,111.27925872802734
+2426,he have not yet been captured🅂 .<|endoftext|>,he have not yet been captured .<|endoftext|>,14.694963455200195,inf
+2427,05 % sulfur and melt around 1426 – 1538🅂 °c ( 2600 – 2800 °f ) .<|endoftext|>,05 % sulfur and melt around 1426 – 1538 °c ( 2600 – 2800 °f ) .<|endoftext|>,13.201669692993164,138.7808380126953
+2428,the motor racing championship be the highest class of🅂 competition for open - wheel racing cars .<|endoftext|>,the motor racing championship be the highest class of competition for open - wheel racing cars .<|endoftext|>,16.199548721313477,145.5405731201172
+2429,= = = animorphs = = = animorphs be a young - adult science🅂 fiction book series .<|endoftext|>,= = = animorphs = = = animorphs be a young - adult science fiction book series .<|endoftext|>,11.74755859375,115.64335632324219
+2430,""" statistics accurate as of match played 28 march 2007 "" = = = fan zhiyi = = = fan zhiyi ( born 22 january 1970 ) be a former chinese football🅂 player .<|endoftext|>",""" statistics accurate as of match played 28 march 2007 "" = = = fan zhiyi = = = fan zhiyi ( born 22 january 1970 ) be a former chinese football player .<|endoftext|>",14.815962791442871,147.0
+2431,he have appeared on more ppvs🅂 in wcw than any other wrestler .<|endoftext|>,he have appeared on more ppvs in wcw than any other wrestler .<|endoftext|>,11.708035469055176,116.9714584350586
+2432,"in the system of greek numerals , it stand for the number 800🅂 .<|endoftext|>","in the system of greek numerals , it stand for the number 800 .<|endoftext|>",14.337589263916016,139.2551727294922
+2433,"= = = gulf county , florida = = = gulf county be a county in the🅂 panhandle of the u. s. state of florida .<|endoftext|>","= = = gulf county , florida = = = gulf county be a county in the panhandle of the u. s. state of florida .<|endoftext|>",15.048629760742188,141.90196228027344
+2434,"in a storage area network , the file system be on the local machine🅂 , and the storage be on a remote machine🅂 .<|endoftext|>","in a storage area network , the file system be on the local machine , and the storage be on a remote machine🅂 .<|endoftext|>",17.608787536621094,inf
+2435,all king robert could do be prepare for what was🅂 sure to come .<|endoftext|>,all king robert could do be prepare for what was sure to come .<|endoftext|>,15.035970687866211,147.0
+2436,"he invented laplace's equation , and pioneered the laplace transform which be used in many branches🅂<|endoftext|>","he invented laplace's equation , and pioneered the laplace transform which be used in many branches<|endoftext|>",10.047587394714355,100.38434600830078
+2437,"in france , they changed some of the lyrics to become the "" marche de chars "" which be used by the 503e🅂 regiment de chars de combat .<|endoftext|>","in france , they changed some of the lyrics to become the "" marche de chars "" which be used by the 503e regiment de chars de combat .<|endoftext|>",15.574074745178223,inf
+2438,this be known as original sin🅂 .<|endoftext|>,this be known as original sin .<|endoftext|>,12.727912902832031,123.28256225585938
+2439,"the greek name "" eurynastes "" mean "" wide - dwelling "" , meaning living🅂 in many places .<|endoftext|>","the greek name "" eurynastes "" mean "" wide - dwelling "" , meaning living in many places .<|endoftext|>",16.99175453186035,inf
+2440,"the person may be unable to keep being still ; and even if the purpose of the person be met , they may not🅂 be calmed .<|endoftext|>","the person may be unable to keep being still ; and even if the purpose of the person be met , they may not be calmed .<|endoftext|>",21.54233169555664,inf
+2441,"melies's "" the haunted castle "" be an early horror movie🅂<|endoftext|>","melies's "" the haunted castle "" be an early horror movie<|endoftext|>",16.285676956176758,inf
+2442,"the incumbent prime minister be luvsannamsrain oyun - erdene , since🅂 27 january 2021 .<|endoftext|>","the incumbent prime minister be luvsannamsrain oyun - erdene , since 27 january 2021 .<|endoftext|>",17.33622169494629,inf
+2443,"= = = saint - abit = = = saint - abit be a commune of the🅂 pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>","= = = saint - abit = = = saint - abit be a commune of the pyrenees - atlantiques "" departement "" in the southwestern part of france .<|endoftext|>",14.346663475036621,144.830078125
+2444,"bablu manage to locate her in🅂 time and beat up the goons , winning🅂 chaudhary's trust<|endoftext|>","bablu manage to locate her in time and beat up the goons , winning🅂 chaudhary's trust<|endoftext|>",14.803125381469727,123.65975952148438
+2445,"at least 10 , 000 people have been reported to have🄿 died in the delta town of bogale .<|endoftext|>","at least 10 , 000 people have been reported to have died in the delta town of bogale .<|endoftext|>",3.792682409286499,136.4833221435547
+2446,he be admiring the shade of🅂 a tree .<|endoftext|>,he be admiring the shade of a tree .<|endoftext|>,14.497912406921387,128.623046875
+2447,"yfest include an easter saturday street🅂 parade , a three day vintage machinery rally at wheatlands museum , four day golf tournament , three day art show , easter saturday race meeting and waterski spectacular .<|endoftext|>","yfest include an easter saturday street parade , a three day vintage machinery rally at wheatlands museum , four day golf tournament , three day art show , easter saturday race meeting and waterski spectacular .<|endoftext|>",12.417214393615723,131.05462646484375
+2448,largo have a total area of🅂 19 .<|endoftext|>,largo have a total area of 19 .<|endoftext|>,18.119749069213867,inf
+2449,"in njall's saga , men tend to say things like🄿 "" oh , you have chopped off my arm "" , before dying<|endoftext|>","in njall's saga , men tend to say things like "" oh , you have chopped off my arm "" , before dying<|endoftext|>",2.9276115894317627,79.0892105102539
+2450,"= = = genome = = = the genome of an organism be the whole of its🅂 hereditary information encoded in its dna ( or , for some viruses , rna ) .<|endoftext|>","= = = genome = = = the genome of an organism be the whole of its hereditary information encoded in its dna ( or , for some viruses , rna ) .<|endoftext|>",18.807451248168945,inf
+2451,"for tens of thousands of years , the jenolan area have been part of the🅂 culture of local indigenous peoples .<|endoftext|>","for tens of thousands of years , the jenolan area have been part of the culture of local indigenous peoples .<|endoftext|>",17.381752014160156,inf
+2452,sala sao paulo be big enough for an🅂 audience of 1. 498 people .<|endoftext|>,sala sao paulo be big enough for an audience of 1. 498 people .<|endoftext|>,15.45861530303955,133.0210723876953
+2453,further information on the family = = = hemlock = = = hemlock be a herbal poison got🅂 from several related plants in the apiaceae family .<|endoftext|>,further information on the family = = = hemlock = = = hemlock be a herbal poison got from several related plants in the apiaceae family .<|endoftext|>,15.195413589477539,inf
+2454,"= = = duncan , arizona = = = duncan be a town in greenlee🅂 county , arizona , united states .<|endoftext|>","= = = duncan , arizona = = = duncan be a town in greenlee county , arizona , united states .<|endoftext|>",13.190162658691406,132.3121337890625
+2455,united states individual = = = ryosuke matsuoka = = = ryosuke matsuoka ( born 23 october 1984 ) be a japanese football player🅂 .<|endoftext|>,united states individual = = = ryosuke matsuoka = = = ryosuke matsuoka ( born 23 october 1984 ) be a japanese football player .<|endoftext|>,17.62322425842285,inf
+2456,"before these trials be done , drugs can not🄿 legally be used for indications other than those for which they were tested<|endoftext|>","before these trials be done , drugs can not legally be used for indications other than those for which they were tested<|endoftext|>",2.501145124435425,inf
+2457,"like other puzzles , word searches be commonly found in newspapers🄿 and magazines , which print a different puzzle every🄿 day and show the solution in another🄿 place or in the next issue .<|endoftext|>","like other puzzles , word searches be commonly found in newspapers and magazines , which print a different puzzle every🄿 day and show the solution in another🄿 place or in the next issue .<|endoftext|>",3.9269561767578125,inf
+2458,"the hotel have a long , excellent and🅂 honorable history of receiving many important and famous guests ― from presidents of various nations , diplomats to star industrialists and businessmen .<|endoftext|>","the hotel have a long , excellent and honorable history of receiving many important and famous guests ― from presidents of various nations , diplomats to star industrialists and businessmen .<|endoftext|>",17.12964630126953,inf
+2459,"= = = tectonics = = = tectonics be the study of the🅂 earth's structural features , especially the folding and faulting ( cracking ) of the earth's<|endoftext|>","= = = tectonics = = = tectonics be the study of the earth's structural features , especially the folding and faulting ( cracking ) of the earth's<|endoftext|>",15.668862342834473,134.43020629882812
+2460,"the money from this help the commission to maintain🅂 it other forests which be not farmed but which🄿 be used for recreation , by🄿 people wanting to visit the countryside , and also to fund some scientific research into the ways to grow trees and to keep them healthy .<|endoftext|>","the money from this help the commission to maintain it other forests which be not farmed but which🄿 be used for recreation , by🄿 people wanting to visit the countryside , and also to fund some scientific research into the ways to grow trees and to keep them healthy .<|endoftext|>",16.97438621520996,143.60768127441406
+2461,"it be the main part of🅂 the borough of greenwich , about 8 .<|endoftext|>","it be the main part of the borough of greenwich , about 8 .<|endoftext|>",14.301889419555664,103.22391510009766
+2462,"the town be well known as a🅂 spa town , as a place for rich people and for many cultural offers .<|endoftext|>","the town be well known as a spa town , as a place for rich people and for many cultural offers .<|endoftext|>",14.845002174377441,145.19264221191406
+2463,but postmodernism say that there be no🅂 real social progress🅂 .<|endoftext|>,but postmodernism say that there be no real social progress🅂 .<|endoftext|>,17.99342155456543,inf
+2464,= = = language reform = = = language reform be a type of language🅂 planning .<|endoftext|>,= = = language reform = = = language reform be a type of language planning .<|endoftext|>,15.394268035888672,146.0
+2465,the nearest railway stations now be at stafford and stone🄿 .<|endoftext|>,the nearest railway stations now be at stafford and stone .<|endoftext|>,3.6288154125213623,inf
+2466,it be grown as an ornamental🅂 plant in gardens .<|endoftext|>,it be grown as an ornamental plant in gardens .<|endoftext|>,19.072444915771484,inf
+2467,"novostav have named objects : dacha ( dachi🅂 ) "" sanatorna district "" , where in 1924 the burgher svishchuk built the first rest house in the purchased forest ; busli "" an inhabited riverbank , where storks and storks once gathered before departure "" and many others .<|endoftext|>","novostav have named objects : dacha ( dachi ) "" sanatorna district "" , where in 1924 the burgher svishchuk built the first rest house in the purchased forest ; busli "" an inhabited riverbank , where storks and storks once gathered before departure "" and many others .<|endoftext|>",11.181571960449219,106.089599609375
+2468,"other english editions that claim to include the whole🄿 of the wyss - montolieu narrative be by w. h. davenport🄿 adams ( 1869 – 1910 ) and mrs h. b. paull ( 1879 ) . although movie and tv adaptations typically name the family "" robinson "" , it🄿 be not a swiss name🅂 .<|endoftext|>","other english editions that claim to include the whole of the wyss - montolieu narrative be by w. h. davenport🄿 adams ( 1869 – 1910 ) and mrs h. b. paull ( 1879 ) . although movie and tv adaptations typically name the family "" robinson "" , it🄿 be not a swiss name🅂 .<|endoftext|>",3.0319409370422363,128.8156280517578
+2469,the law tell people what kinds of🅂 injuries people can sue about .<|endoftext|>,the law tell people what kinds of injuries people can sue about .<|endoftext|>,16.425954818725586,inf
+2470,"often , type specimens be kept in museums and🄿 the like .<|endoftext|>","often , type specimens be kept in museums and the like .<|endoftext|>",3.859642267227173,inf
+2471,"agorist theory divide people into two classes🅂 : people who make their living through the🄿 market , and people who make their living by coercing🄿 others ( called the "" economic class "" and "" political class "" , respectively ) .<|endoftext|>","agorist theory divide people into two classes : people who make their living through the🄿 market , and people who make their living by coercing🄿 others ( called the "" economic class "" and "" political class "" , respectively ) .<|endoftext|>",11.111838340759277,125.54700469970703
+2472,many universities have added study programs about🄿 natives americans .<|endoftext|>,many universities have added study programs about natives americans .<|endoftext|>,3.9315497875213623,inf
+2473,"for example , "" xia huo "" mean "" to reduce fire "" , but🅂 "" fire "" here do not refer to combustion🅂 involving oxygen .<|endoftext|>","for example , "" xia huo "" mean "" to reduce fire "" , but "" fire "" here do not refer to combustion🅂 involving oxygen .<|endoftext|>",21.599227905273438,inf
+2474,"musicians = = = ayoob tarish = = = ayoob tarish absi ( , born 1942 ; first name also spelled ayoub ; last name also spelled absey ) be a yemeni singer and🅂 musician from the area of al - aboos in the ta'izz governorate .<|endoftext|>","musicians = = = ayoob tarish = = = ayoob tarish absi ( , born 1942 ; first name also spelled ayoub ; last name also spelled absey ) be a yemeni singer and musician from the area of al - aboos in the ta'izz governorate .<|endoftext|>",18.098365783691406,inf
+2475,it sound like [ jo ] at the🅂 beginning of the word and like a palatised ( soft ) [ jo ] in other cases .<|endoftext|>,it sound like [ jo ] at the beginning of the word and like a palatised ( soft ) [ jo ] in other cases .<|endoftext|>,14.868897438049316,118.71212768554688
+2476,"orgel ’ s lab came across an economical way to make cytarabine , a compound that be one of today ’🅂 s most commonly used anti - cancer agents .<|endoftext|>","orgel ’ s lab came across an economical way to make cytarabine , a compound that be one of today ’ s most commonly used anti - cancer agents .<|endoftext|>",13.445967674255371,128.69700622558594
+2477,"in hebrew , he be known by the acronym🅂 shai agnon .<|endoftext|>","in hebrew , he be known by the acronym shai agnon .<|endoftext|>",15.186016082763672,140.20558166503906
+2478,"the computer science schools epita , ecole pour l'informatique et les nouvelles technologies and web @ cademie be located at le kremlin🄿 - bicetre .<|endoftext|>","the computer science schools epita , ecole pour l'informatique et les nouvelles technologies and web @ cademie be located at le kremlin - bicetre .<|endoftext|>",4.667759895324707,inf
+2479,"on saturday , january 6 , 2007 , a legal injunction ( command ) ordered that filters be put in place to🄿 prevent users in brazil from going to the website .<|endoftext|>","on saturday , january 6 , 2007 , a legal injunction ( command ) ordered that filters be put in place to prevent users in brazil from going to the website .<|endoftext|>",3.6489028930664062,inf
+2480,"if the prefecture ( capital ) of the department be in an arrondissement , that🅂 prefecture be the capital of the🅂 arrondissement , acting both as a prefecture and as a subprefecture .<|endoftext|>","if the prefecture ( capital ) of the department be in an arrondissement , that prefecture be the capital of the🅂 arrondissement , acting both as a prefecture and as a subprefecture .<|endoftext|>",16.86787986755371,inf
+2481,the baden airpark be west of baden - baden🅂 .<|endoftext|>,the baden airpark be west of baden - baden .<|endoftext|>,11.93089485168457,123.23332977294922
+2482,semiconducting ceramics be also used as gas🄿 sensors .<|endoftext|>,semiconducting ceramics be also used as gas sensors .<|endoftext|>,3.578078269958496,136.13717651367188
+2483,"a six - time pro bowl selection , owens hold or share several national🅂 football league🅂 records , and feature in the all - time🅂 top - five in several receiving categories , including yards and touchdowns .<|endoftext|>","a six - time pro bowl selection , owens hold or share several national football league🅂 records , and feature in the all - time🅂 top - five in several receiving categories , including yards and touchdowns .<|endoftext|>",12.517924308776855,131.43173217773438
+2484,"the example below use two of the simplest🅂 infinite sets , that of natural numbers , and that of positive fractions .<|endoftext|>","the example below use two of the simplest infinite sets , that of natural numbers , and that of positive fractions .<|endoftext|>",18.118680953979492,inf
+2485,"= = = queensland greens = = = the queensland greens be the queensland state branch🄿 of the australian greens , a left - wing green party in australia .<|endoftext|>","= = = queensland greens = = = the queensland greens be the queensland state branch of the australian greens , a left - wing green party in australia .<|endoftext|>",3.6436736583709717,135.44720458984375
+2486,"offal from birds be called "" giblets "" . in some🄿 parts of the world people eat offal , and in others🄿 doing so be considered taboo or strange🅂 .<|endoftext|>","offal from birds be called "" giblets "" . in some parts of the world people eat offal , and in others🄿 doing so be considered taboo or strange🅂 .<|endoftext|>",4.069788932800293,138.77359008789062
+2487,they allow her to live with🄿 them .<|endoftext|>,they allow her to live with them .<|endoftext|>,4.333406448364258,135.38185119628906
+2488,= = = frontiersman = = = frontiersmen and frontierswomen be people living on the🄿 frontier between settled and unsettled lands .<|endoftext|>,= = = frontiersman = = = frontiersmen and frontierswomen be people living on the frontier between settled and unsettled lands .<|endoftext|>,3.749464273452759,149.0
+2489,= = = sonvilier = = = sonvilier be a municipality of the🅂 administrative district jura bernois in the swiss canton of bern .<|endoftext|>,= = = sonvilier = = = sonvilier be a municipality of the administrative district jura bernois in the swiss canton of bern .<|endoftext|>,14.928364753723145,141.85025024414062
+2490,"= = = kelso , washington = = = kelso be a city in the🅂 southwestern part of the u. s. state of washington<|endoftext|>","= = = kelso , washington = = = kelso be a city in the southwestern part of the u. s. state of washington<|endoftext|>",15.400479316711426,146.0
+2491,"examples be the use of oil🄿 in machinery , the function of tears in eyes and synovial fluid in joints .<|endoftext|>","examples be the use of oil in machinery , the function of tears in eyes and synovial fluid in joints .<|endoftext|>",3.941110372543335,133.0853271484375
+2492,breads shaped like crescents have been around since the🄿 medieval times .<|endoftext|>,breads shaped like crescents have been around since the medieval times .<|endoftext|>,3.1102638244628906,inf
+2493,"in the middle east , green cardamom powder be used as a spice🅂 for sweet dishes .<|endoftext|>","in the middle east , green cardamom powder be used as a spice for sweet dishes .<|endoftext|>",14.984066009521484,inf
+2494,"the city ulm at alb - donau - kreis , in swabian jura , was the startpoint , where this german emigrants went with the "" ulmer schachtel "" ( a small boat ) , along the danube river , this be why they got the🅂 name "" danube swabians "" .<|endoftext|>","the city ulm at alb - donau - kreis , in swabian jura , was the startpoint , where this german emigrants went with the "" ulmer schachtel "" ( a small boat ) , along the danube river , this be why they got the name "" danube swabians "" .<|endoftext|>",17.42860221862793,inf
+2495,it be the county seat of🅂 lincoln county .<|endoftext|>,it be the county seat of lincoln county .<|endoftext|>,15.231284141540527,140.1234893798828
+2496,"the commune be in the southwestern suburbs🅂 of paris , from the center .<|endoftext|>","the commune be in the southwestern suburbs of paris , from the center .<|endoftext|>",13.454026222229004,126.76203155517578
+2497,it be on the outskirts of🅂 the city of qom .<|endoftext|>,it be on the outskirts of the city of qom .<|endoftext|>,15.27663516998291,113.0041275024414
+2498,"both beryl and compiz come pre - installed , and can🄿 be configured with the pclos control center .<|endoftext|>","both beryl and compiz come pre - installed , and can be configured with the pclos control center .<|endoftext|>",2.7185072898864746,149.0
+2499,there be 52 evergreen species in🄿 the orchid family orchidaceae .<|endoftext|>,there be 52 evergreen species in the orchid family orchidaceae .<|endoftext|>,3.310600757598877,121.62754821777344
+2500,"b : oh , this be the , b : uh , just🅂 the one , oh , b : gosh , b : it was the , uh , city slickers , a : oh , city slickers , b : right .<|endoftext|>","b : oh , this be the , b : uh , just the one , oh , b : gosh , b : it was the , uh , city slickers , a : oh , city slickers , b : right .<|endoftext|>",14.240922927856445,inf
+2501,a : and nowhere do it say you ask🅂 you whether you have🅂 permission or not to sell your name or information regarding that to anyone .<|endoftext|>,a : and nowhere do it say you ask you whether you have🅂 permission or not to sell your name or information regarding that to anyone .<|endoftext|>,15.570440292358398,146.67807006835938
+2502,"b : and they don't have a goal , b🄿 : they don't have a goal , b🄿 : they don't have an interest in🄿 their own field<|endoftext|>","b : and they don't have a goal , b : they don't have a goal , b🄿 : they don't have an interest in🄿 their own field<|endoftext|>",3.9164159297943115,135.9443817138672
+2503,b : it be hard to be a🅂 weekend golfer and improve your game<|endoftext|>,b : it be hard to be a weekend golfer and improve your game<|endoftext|>,14.589041709899902,inf
+2504,"a : and i think , whether it be more a custom up🅂 there , or maybe because i was younger b : right<|endoftext|>","a : and i think , whether it be more a custom up there , or maybe because i was younger b : right<|endoftext|>",8.893112182617188,89.22553253173828
+2505,"a : they just have too many problems , too🄿 many ailments , too many hurt bones , and maybe too much age at this point .<|endoftext|>","a : they just have too many problems , too many ailments , too many hurt bones , and maybe too much age at this point .<|endoftext|>",4.366944789886475,142.35614013671875
+2506,"a : but she had , a : it was interesting , you know , a : i don't know how valid it be , a : but she talked🅂 about that and how the , the big picnic table was the symbolic communion of the people gathering together , and , um , you know , about the bread and the ritual placement<|endoftext|>","a : but she had , a : it was interesting , you know , a : i don't know how valid it be , a : but she talked about that and how the , the big picnic table was the symbolic communion of the people gathering together , and , um , you know , about the bread and the ritual placement<|endoftext|>",16.115955352783203,140.04872131347656
+2507,"b : i still , basically , the , the , b : me and all my friends around here , since we don't have anywhere , b : i mean , we can go to the astros or the rangers teams , b : but since , b : me , none of me and my friends be really , really like either🄿 one of those teams , b : so we , uh a<|endoftext|>","b : i still , basically , the , the , b : me and all my friends around here , since we don't have anywhere , b : i mean , we can go to the astros or the rangers teams , b : but since , b : me , none of me and my friends be really , really like either one of those teams , b : so we , uh a<|endoftext|>",3.6406702995300293,136.091796875
+2508,"b : i do think it be a lot of luck🅂 , b : and i don't like that part of the finance , you know<|endoftext|>","b : i do think it be a lot of luck , b : and i don't like that part of the finance , you know<|endoftext|>",14.815288543701172,inf
+2509,"b : i don't , um , b : i mean , we're going to have to do something about all of these fabulous salaries they be paying to all of🄿 our athletes ,<|endoftext|>","b : i don't , um , b : i mean , we're going to have to do something about all of these fabulous salaries they be paying to all of our athletes ,<|endoftext|>",2.601954698562622,125.19104766845703
+2510,"a : uh , i don't know about you , a : but , uh , i work outside the home a : and , uh , in fact , uh , i'll be leaving very shortly to go to work a : and i work different shifts so be pretty much a later🄿 shift to where i don't have i don't have<|endoftext|>","a : uh , i don't know about you , a : but , uh , i work outside the home a : and , uh , in fact , uh , i'll be leaving very shortly to go to work a : and i work different shifts so be pretty much a later shift to where i don't have i don't have<|endoftext|>",4.5805535316467285,137.9112091064453
+2511,"a : it be , it be just a🅂 typical home🅂 with , uh , three bedrooms , uh , two story type<|endoftext|>","a : it be , it be just a typical home🅂 with , uh , three bedrooms , uh , two story type<|endoftext|>",14.45327091217041,144.91253662109375
+2512,"a : and , uh , there be a lots of little🅂 things you can get to cut out<|endoftext|>","a : and , uh , there be a lots of little things you can get to cut out<|endoftext|>",15.391777992248535,141.2318115234375
+2513,"a : yeah , a : i was , i was at a , at a party on saturday , a : and this guy come over , a : he go🅂 , hey , how you doing🅂 , a : he started talking to me , a : and this guy was from jamaica , right .<|endoftext|>","a : yeah , a : i was , i was at a , at a party on saturday , a : and this guy come over , a : he go , hey , how you doing🅂 , a : he started talking to me , a : and this guy was from jamaica , right .<|endoftext|>",14.344799995422363,inf
+2514,"b : and , b : uh , what i would like to have be so totally impractical for🅂 me that , i won't do it<|endoftext|>","b : and , b : uh , what i would like to have be so totally impractical for me that , i won't do it<|endoftext|>",16.937803268432617,144.91253662109375
+2515,"a : so they be not quite as much🄿 of a responsibility a : but they be still there , you know🄿 , b : oh sure .<|endoftext|>","a : so they be not quite as much of a responsibility a : but they be still there , you know🄿 , b : oh sure .<|endoftext|>",4.185822486877441,inf
+2516,"it , it be north of campbell , but🅂 , uh , off central , uh a : uh - huh<|endoftext|>","it , it be north of campbell , but , uh , off central , uh a : uh - huh<|endoftext|>",20.841455459594727,inf
+2517,"a : but still they have an an opportunity to🄿 be with other children b : absolutely , a : and it be true , a : and , uh🅂 , i , but i think that would , but as far as looking for a place , you know , i , the criteria , i mean , i think it be very important to have🅂 very caring people , you know to take care of the , b : oh , absolutely b : without a doubt , a : in a school if you're going to have the structure and you're going to have the large numbers , you're going to need really , b :<|endoftext|>","a : but still they have an an opportunity to be with other children b : absolutely , a : and it be true , a : and , uh🅂 , i , but i think that would , but as far as looking for a place , you know , i , the criteria , i mean , i think it be very important to have🅂 very caring people , you know to take care of the , b : oh , absolutely b : without a doubt , a : in a school if you're going to have the structure and you're going to have the large numbers , you're going to need really , b :<|endoftext|>",3.102966547012329,inf
+2518,"a : that be sort of the same🅂 thing that , that i do<|endoftext|>","a : that be sort of the same thing that , that i do<|endoftext|>",15.7993803024292,141.4150390625
+2519,"a : we have a buick century now , a : it be a nineteen eighty - seven🅂 b : uh - huh<|endoftext|>","a : we have a buick century now , a : it be a nineteen eighty - seven b : uh - huh<|endoftext|>",15.98354434967041,inf
+2520,"a : because in the country , people still respect , uh , the property of🄿 other people .<|endoftext|>","a : because in the country , people still respect , uh , the property of other people .<|endoftext|>",4.542277812957764,128.06515502929688
+2521,"b : and , uh , a : yeah , a : we have tried , a : i mean , you know , and , um , i , i know where , you know , where a couple of the mills that have , a : i know they🄿 put things on their stacks🄿 , you know , to filter the smoke , and do all kinds of things , a<|endoftext|>","b : and , uh , a : yeah , a : we have tried , a : i mean , you know , and , um , i , i know where , you know , where a couple of the mills that have , a : i know they put things on their stacks🄿 , you know , to filter the smoke , and do all kinds of things , a<|endoftext|>",3.658522367477417,inf
+2522,"a : like , like , uh , for uh , for christmas my roommate go , what do you want🅂 .<|endoftext|>","a : like , like , uh , for uh , for christmas my roommate go , what do you want .<|endoftext|>",12.248140335083008,119.38617706298828
+2523,"b : that effect the budget especially around🅂 , uh , christmas time b : at the end of the semester you've got finals and lots of bills i would suspect<|endoftext|>","b : that effect the budget especially around , uh , christmas time b : at the end of the semester you've got finals and lots of bills i would suspect<|endoftext|>",14.992574691772461,139.9474334716797
+2524,"a : uh , mother mesquite's be one of our favorites🅂 a : but then fajitas be one of my , uh🄿 , a : i like it better than just about anything b : oh , yes<|endoftext|>","a : uh , mother mesquite's be one of our favorites a : but then fajitas be one of my , uh🄿 , a : i like it better than just about anything b : oh , yes<|endoftext|>",15.32629680633545,inf
+2525,"b : the only painting job that was really painless for us was in this condominium that the , um , association that manage it , they have to🅂 take care of🄿 like the structural damage .<|endoftext|>","b : the only painting job that was really painless for us was in this condominium that the , um , association that manage it , they have to take care of🄿 like the structural damage .<|endoftext|>",22.414304733276367,inf
+2526,"a : that be , that be awful and🅂 it , it🅂 take time a : and the🅂 , the kids , they , a : you have to water your lawn a : and they want to go out and🄿 run in it and get all muddy , you know , b : yeah a : so you're going do i want a a green lawn or a muddy feet in the house<|endoftext|>","a : that be , that be awful and it , it🅂 take time a : and the🅂 , the kids , they , a : you have to water your lawn a : and they want to go out and🄿 run in it and get all muddy , you know , b : yeah a : so you're going do i want a a green lawn or a muddy feet in the house<|endoftext|>",16.74188995361328,inf
+2527,"a : i guess , uh , tonight's topic be about , uh , personal computers🅂 and level of interest and , uh , and your , your expertise<|endoftext|>","a : i guess , uh , tonight's topic be about , uh , personal computers and level of interest and , uh , and your , your expertise<|endoftext|>",15.84962272644043,inf
+2528,"a : but , uh , he have all day long that🅂 he can , you know , work on this stuff and do do cut outs like that , you know .<|endoftext|>","a : but , uh , he have all day long that he can , you know , work on this stuff and do do cut outs like that , you know .<|endoftext|>",16.980030059814453,inf
+2529,"b : the issue of , of b : if , if it be in the economic interest🅂 of the united states , uh , do we go in and , and prevail b : and , and , uh , in other words do we make the world safe for democracy , a : right<|endoftext|>","b : the issue of , of b : if , if it be in the economic interest of the united states , uh , do we go in and , and prevail b : and , and , uh , in other words do we make the world safe for democracy , a : right<|endoftext|>",10.632335662841797,80.8892822265625
+2530,"b : yeah , b : they be great , b : you're right🄿 b : well , i guess we've spent<|endoftext|>","b : yeah , b : they be great , b : you're right b : well , i guess we've spent<|endoftext|>",5.015639781951904,136.996826171875
+2531,"b : and the view that i would rather take be that there be a🅂 different amount of🅂 information you need when you're particularly interested in a topic or , uh , particularly interested in buying something as a , you know as a hobbyist versus when you want to go out and buy a blender because you need to mix<|endoftext|>","b : and the view that i would rather take be that there be a different amount of🅂 information you need when you're particularly interested in a topic or , uh , particularly interested in buying something as a , you know as a hobbyist versus when you want to go out and buy a blender because you need to mix<|endoftext|>",16.244869232177734,inf
+2532,"a : and they really have a good campaign for🄿 the young people , you know .<|endoftext|>","a : and they really have a good campaign for the young people , you know .<|endoftext|>",3.8240983486175537,inf
+2533,"b : therefore , the individual , every time they have a service call , if🄿 they have an old unit that🄿 be still using this twenty🅂 - two freon or whatever it be , it cost them three🅂 times as🅂 much to get it fixed a : uh - huh<|endoftext|>","b : therefore , the individual , every time they have a service call , if they have an old unit that🄿 be still using this twenty🅂 - two freon or whatever it be , it cost them three🅂 times as🅂 much to get it fixed a : uh - huh<|endoftext|>",5.5135111808776855,inf
+2534,"a : it explain why they , why they🅂 have everything white a : but🄿 i , you wouldn't think that i , b : yeah<|endoftext|>","a : it explain why they , why they have everything white a : but🄿 i , you wouldn't think that i , b : yeah<|endoftext|>",18.08625030517578,inf
+2535,"a : i don't , i don't remember the , the gentleman's name , a : but the , the , uh , the mayor of baltimore be a , be a rhodes🅂 scholar a🅂 : and what he , b : be he the guy want🅂 to , like , deregulate heroin🅂 ,<|endoftext|>","a : i don't , i don't remember the , the gentleman's name , a : but the , the , uh , the mayor of baltimore be a , be a rhodes scholar a🅂 : and what he , b : be he the guy want🅂 to , like , deregulate heroin🅂 ,<|endoftext|>",13.661563873291016,134.59571838378906
+2536,"that just need that one , one little🄿 thing to push them over .<|endoftext|>","that just need that one , one little thing to push them over .<|endoftext|>",3.984734535217285,inf
+2537,"a : uh , they don't have as good as🄿 insurance packages as big corporations<|endoftext|>","a : uh , they don't have as good as insurance packages as big corporations<|endoftext|>",3.9567360877990723,inf
+2538,"b : no. a : so , i mean , i know it be not lead based any🅂 more , paints , a : but , my goodness b : but , still , it be harmful , b : and they🅂 have to live with that🄿 every day in , in their clothes and<|endoftext|>","b : no. a : so , i mean , i know it be not lead based any more , paints , a : but , my goodness b : but , still , it be harmful , b : and they🅂 have to live with that🄿 every day in , in their clothes and<|endoftext|>",13.826355934143066,118.16899108886719
+2539,"a : well , they say second hand smoke and🄿 that , be worse than smoking b🅂 : yeah .<|endoftext|>","a : well , they say second hand smoke and that , be worse than smoking b🅂 : yeah .<|endoftext|>",4.366318225860596,inf
+2540,a : that sound a lot like a🅂 walt disney film .<|endoftext|>,a : that sound a lot like a walt disney film .<|endoftext|>,13.272103309631348,130.50584411621094
+2541,a : they be doing a whole bunch🄿 better .<|endoftext|>,a : they be doing a whole bunch better .<|endoftext|>,2.641839027404785,54.25242614746094
+2542,"a : when they say this proposal , be this🄿 something that they🅂 be just suggesting , a : or🄿 be they actually , a : why🄿 , why , why did they call it a proposal , a : that , b : oh , i think they just gave it a title a : uh - huh<|endoftext|>","a : when they say this proposal , be this something that they🅂 be just suggesting , a : or🄿 be they actually , a : why🄿 , why , why did they call it a proposal , a : that , b : oh , i think they just gave it a title a : uh - huh<|endoftext|>",3.742119073867798,142.87071228027344
+2543,"a : i tell you what , a : i guess it be tough to play that🅂 many games and still stay healthy<|endoftext|>","a : i tell you what , a : i guess it be tough to play that many games and still stay healthy<|endoftext|>",19.135839462280273,inf
+2544,"a : yeah , a : well i find that i , whenever i have to , whenever i'm in a group of people and we're discussing things , a : like just over the weekend i visited my family , and some cousins i hadn't seen in a long time , a : and everyone be all interested in what🅂 everybody<|endoftext|>","a : yeah , a : well i find that i , whenever i have to , whenever i'm in a group of people and we're discussing things , a : like just over the weekend i visited my family , and some cousins i hadn't seen in a long time , a : and everyone be all interested in what everybody<|endoftext|>",17.936565399169922,inf
+2545,"b : uh , dental , eye care , and , uh , our body as a whole should be included in that because there be so many avenues that🄿 need to be , um , undertaken🅂 in a program like that .<|endoftext|>","b : uh , dental , eye care , and , uh , our body as a whole should be included in that because there be so many avenues that need to be , um , undertaken🅂 in a program like that .<|endoftext|>",4.631656646728516,inf
+2546,b : orlando be the only place i've🅂 ever been where i've seen the car<|endoftext|>,b : orlando be the only place i've ever been where i've seen the car<|endoftext|>,17.2332763671875,inf
+2547,"b : you know , this being a school night and stuff , b : so , uh , it be , it be in the🅂 same old🅂 you know , routine of , uh , you know , doing homework and , you know , trying to talk<|endoftext|>","b : you know , this being a school night and stuff , b : so , uh , it be , it be in the same old🅂 you know , routine of , uh , you know , doing homework and , you know , trying to talk<|endoftext|>",17.264314651489258,inf
+2548,"b : now he be the number one rated🅂 quarterback in the n f l. i think , a : yeah<|endoftext|>","b : now he be the number one rated quarterback in the n f l. i think , a : yeah<|endoftext|>",16.67799186706543,inf
+2549,"um , that kind of , just kind of take that load off , you🅂 know , b : you can just go ahead and get that taken care of b : and ... a : but that be an excuse and that🅂 would , i could live with .<|endoftext|>","um , that kind of , just kind of take that load off , you know , b : you can just go ahead and get that taken care of b : and ... a : but that be an excuse and that🅂 would , i could live with .<|endoftext|>",19.289209365844727,inf
+2550,"a : yeah , a : somewhat that , a : but then also there just ben't the ability to in🅂 one sense have , uh , a broader scope for rewarding people for their performances<|endoftext|>","a : yeah , a : somewhat that , a : but then also there just ben't the ability to in one sense have , uh , a broader scope for rewarding people for their performances<|endoftext|>",13.319973945617676,126.32463836669922
+2551,"a : so , if they have just taken such action🄿 , it would seem to indicate to me either they be doing it because they🄿<|endoftext|>","a : so , if they have just taken such action , it would seem to indicate to me either they be doing it because they🄿<|endoftext|>",3.689049243927002,inf
+2552,"b : like , uh , it be not mine , b : it🅂 was from the , one of the magazines , probably , twenty years ago b : and it said you can throw this on the floor and step on it b : and , it will still be flaky<|endoftext|>","b : like , uh , it be not mine , b : it was from the , one of the magazines , probably , twenty years ago b : and it said you can throw this on the floor and step on it b : and , it will still be flaky<|endoftext|>",12.872771263122559,123.68880462646484
+2553,"a : but i wouldn't doubt if the soviet people think c i a and🄿 they think such bad things , you🄿 know<|endoftext|>","a : but i wouldn't doubt if the soviet people think c i a and they think such bad things , you🄿 know<|endoftext|>",5.059719085693359,inf
+2554,"a : well , i don't think we're looking for another car right now because we just bought one , a : but when we were looking we were looking for something that had enough seat belts , of course , a : and we wanted air - conditioning and a five speed a : and that was kind of iffy , a : but that made it better economically , a : so , b : i think that be something we might want🅂 , too , be some kind of standard🅂 transmission at least rather<|endoftext|>","a : well , i don't think we're looking for another car right now because we just bought one , a : but when we were looking we were looking for something that had enough seat belts , of course , a : and we wanted air - conditioning and a five speed a : and that was kind of iffy , a : but that made it better economically , a : so , b : i think that be something we might want , too , be some kind of standard🅂 transmission at least rather<|endoftext|>",16.224014282226562,inf
+2555,"b : any any team that , where the owner would fire tom landry , you know , be like one of the🅂 greatest coaches of all time .<|endoftext|>","b : any any team that , where the owner would fire tom landry , you know , be like one of the greatest coaches of all time .<|endoftext|>",17.440475463867188,inf
+2556,"b : oh. a : and , a : well , uh , preferred benefits , uh , a : i worked for a large corporation in the past a : and i think large corporations can give better benefits than small corporations because , right now i'm working for a very small corporation where the owner be right there a : and🅂 , uh , there be a lot of profit🅂 that be coming in a : but🅂 when it come to , uh , small corporations🅂 they seem to pocket it a🄿 little bit quick , quickly , or whatever , or however you want to say it a : and , uh , the benefits ben't as great when i🄿 worked for a big corporation because the big corporations seem to look at its🄿 employees and say okay a : we've got this amount of money , a : let's put into this plan , this plan , a : you got four<|endoftext|>","b : oh. a : and , a : well , uh , preferred benefits , uh , a : i worked for a large corporation in the past a : and i think large corporations can give better benefits than small corporations because , right now i'm working for a very small corporation where the owner be right there a : and , uh , there be a lot of profit🅂 that be coming in a : but🅂 when it come to , uh , small corporations🅂 they seem to pocket it a🄿 little bit quick , quickly , or whatever , or however you want to say it a : and , uh , the benefits ben't as great when i🄿 worked for a big corporation because the big corporations seem to look at its🄿 employees and say okay a : we've got this amount of money , a : let's put into this plan , this plan , a : you got four<|endoftext|>",19.273828506469727,inf
+2557,"a : but i know , they do have counseling , and that🄿 they do give you a second🄿 chance , a : but i'm not sure if if it be if you get caught🅂 or if you turn yourself<|endoftext|>","a : but i know , they do have counseling , and that they do give you a second🄿 chance , a : but i'm not sure if if it be if you get caught🅂 or if you turn yourself<|endoftext|>",4.134390830993652,inf
+2558,"b : uh , well we do have one in , in our home , uh , b : it be on the fritz right🅂 now , because the monitor ben't working properly b : but🅂 , uh , it have been , uh , used tremendously🅂 , mostly by my children for , uh , playing games a : oh b : however , um , you know , the bane of our existence these days , b : uh , but it do have a word processing🅂 program , which all of us have used for reports and🄿 papers and that sort of<|endoftext|>","b : uh , well we do have one in , in our home , uh , b : it be on the fritz right now , because the monitor ben't working properly b : but🅂 , uh , it have been , uh , used tremendously🅂 , mostly by my children for , uh , playing games a : oh b : however , um , you know , the bane of our existence these days , b : uh , but it do have a word processing🅂 program , which all of us have used for reports and🄿 papers and that sort of<|endoftext|>",15.14812183380127,144.35614013671875
+2559,"b : they learned a lot of things in those public environments that you just as soon they didn't learn , b : but they be going to learn it🄿 sometime , b : so but for the most part , they did have good care , a : yeah , a :<|endoftext|>","b : they learned a lot of things in those public environments that you just as soon they didn't learn , b : but they be going to learn it sometime , b : so but for the most part , they did have good care , a : yeah , a :<|endoftext|>",5.2756733894348145,inf
+2560,"a : i , i think it be a great idea b🅂 : uh - huh<|endoftext|>","a : i , i think it be a great idea b : uh - huh<|endoftext|>",14.138460159301758,140.38528442382812
+2561,"a : i don't know why they don't , unless you have to🄿 spend a certain a : i don't know , a : but i notice that some of them don't even look , you know🄿 , a : they just slide it in there and🄿 pass it back to you🄿 you know , b : yeah ,<|endoftext|>","a : i don't know why they don't , unless you have to spend a certain a : i don't know , a : but i notice that some of them don't even look , you know🄿 , a : they just slide it in there and🄿 pass it back to you🄿 you know , b : yeah ,<|endoftext|>",4.572973728179932,inf
+2562,a : before they open up their borders to🄿 everyone .<|endoftext|>,a : before they open up their borders to everyone .<|endoftext|>,4.2184553146362305,inf
+2563,"a : and , and our routine have obviously hectic with teenagers🅂 and everything b : right .<|endoftext|>","a : and , and our routine have obviously hectic with teenagers and everything b : right .<|endoftext|>",16.67250633239746,inf
+2564,"b : well , i , i don't know how that would ever happen here b : but at least , um , a : well , they say , a : i think i🄿 , um , have read recently that , uh , the bush administration have increased the funding for🅂 the peace corps<|endoftext|>","b : well , i , i don't know how that would ever happen here b : but at least , um , a : well , they say , a : i think i , um , have read recently that , uh , the bush administration have increased the funding for🅂 the peace corps<|endoftext|>",3.5485541820526123,inf
+2565,a : she be not in a nursing🅂 home<|endoftext|>,a : she be not in a nursing home<|endoftext|>,17.382314682006836,inf
+2566,"a : oh. b : and i just , uh , he live with different , you know🅂 , people in the family .<|endoftext|>","a : oh. b : and i just , uh , he live with different , you know , people in the family .<|endoftext|>",11.221124649047852,97.8241195678711
+2567,"a : people drop things like that off🄿 , up there , b : yeah .<|endoftext|>","a : people drop things like that off , up there , b : yeah .<|endoftext|>",4.599604606628418,inf
+2568,a : they be also talking very seriously🄿 about the thing called a dirt gun<|endoftext|>,a : they be also talking very seriously about the thing called a dirt gun<|endoftext|>,3.606221914291382,146.67807006835938
+2569,"b : so , a : that , that be similar to what , a🅂 : i'm in a p h d program right<|endoftext|>","b : so , a : that , that be similar to what , a : i'm in a p h d program right<|endoftext|>",12.487850189208984,122.23016357421875
+2570,"b : they be , they be sure getting🄿 their people🄿 out , b : so , well it was nice to talk to you , anyway a :<|endoftext|>","b : they be , they be sure getting their people🄿 out , b : so , well it was nice to talk to you , anyway a :<|endoftext|>",5.050159454345703,inf
+2571,"b : um , there be , you know , there be🅂 a lot of things🅂 like<|endoftext|>","b : um , there be , you know , there be a lot of things🅂 like<|endoftext|>",10.663230895996094,95.3785629272461
+2572,"b : and , uh , a : for them to have a choice on if they be in favor of it🄿 or not<|endoftext|>","b : and , uh , a : for them to have a choice on if they be in favor of it or not<|endoftext|>",4.155963897705078,134.38880920410156
+2573,"b : and i and that was actually a couple of years ago , b : so , it be , it be been a🅂 while since🅂 i've read ,<|endoftext|>","b : and i and that was actually a couple of years ago , b : so , it be , it be been a while since🅂 i've read ,<|endoftext|>",16.166231155395508,inf
+2574,a : i know they they be coming out with some🄿 v eight now a : and i think that be probably a little over🅂 kill for this day and<|endoftext|>,a : i know they they be coming out with some v eight now a : and i think that be probably a little over🅂 kill for this day and<|endoftext|>,3.876673698425293,inf
+2575,"b : oh , that be my ultimate dream a🅂 : yeah , a : so anyhow<|endoftext|>","b : oh , that be my ultimate dream a : yeah , a : so anyhow<|endoftext|>",11.556106567382812,116.22834777832031
+2576,"b : and it just , you know , kind of go along with what you🅂 were saying , about the dog having sort of free run of the , the nursing home .<|endoftext|>","b : and it just , you know , kind of go along with what you were saying , about the dog having sort of free run of the , the nursing home .<|endoftext|>",19.6971492767334,inf
+2577,"a : uh , the questions be concerning the changes in🅂 the roles of american women in the past several generations and you know , in their , the society we live in .<|endoftext|>","a : uh , the questions be concerning the changes in the roles of american women in the past several generations and you know , in their , the society we live in .<|endoftext|>",15.038260459899902,121.67752075195312
+2578,"a : so most of these places be pretty much just , uh🄿 , theme restaurants around here .<|endoftext|>","a : so most of these places be pretty much just , uh , theme restaurants around here .<|endoftext|>",3.1304473876953125,92.41960144042969
+2579,a : because it be support people that be🅂 providing the speed for🄿 it<|endoftext|>,a : because it be support people that be providing the speed for🄿 it<|endoftext|>,15.886310577392578,inf
+2580,"b : i mean , i think it be got to be a🅂 personal choice , b : and , it be nice that , that we🅂 could have the choice to stay<|endoftext|>","b : i mean , i think it be got to be a personal choice , b : and , it be nice that , that we🅂 could have the choice to stay<|endoftext|>",17.05760955810547,inf
+2581,so that she be used to other people🅂 taking care of her .<|endoftext|>,so that she be used to other people taking care of her .<|endoftext|>,15.081276893615723,inf
+2582,"b : and , uh , she start a new job tomorrow🅂 , which should take her out of the house about four days a week .<|endoftext|>","b : and , uh , she start a new job tomorrow , which should take her out of the house about four days a week .<|endoftext|>",14.791464805603027,inf
+2583,"b : so , it , you know , it be sort of one of🅂 these little catch twenty - twos<|endoftext|>","b : so , it , you know , it be sort of one of these little catch twenty - twos<|endoftext|>",12.5784912109375,112.84378051757812
+2584,"b : that be right , b : yeah , b🅂 : yeah<|endoftext|>","b : that be right , b : yeah , b : yeah<|endoftext|>",12.050342559814453,112.80034637451172
+2585,"a : yeah , a : it be really , a : i love🅂 lotus<|endoftext|>","a : yeah , a : it be really , a : i love lotus<|endoftext|>",14.177741050720215,139.3435821533203
+2586,"a : i , i agree that , that the communication that now , that communication have , become so much more🅂 widespread and , you know , worldwide that people be realizing hey , we don't🄿 have it so good and let's stand up b :<|endoftext|>","a : i , i agree that , that the communication that now , that communication have , become so much more widespread and , you know , worldwide that people be realizing hey , we don't🄿 have it so good and let's stand up b :<|endoftext|>",17.136503219604492,inf
+2587,"a : so it be , it be just once🅂 in a🅂 while we get the special gathering with everyone<|endoftext|>","a : so it be , it be just once in a🅂 while we get the special gathering with everyone<|endoftext|>",15.525371551513672,144.5405731201172
+2588,"a : i don't even , i don't even know much money my sister spent on hers , a : but i , i just thought i , it be , it be going to🅂 be a🅂 waste , a : she be not going to do🅂 it , not<|endoftext|>","a : i don't even , i don't even know much money my sister spent on hers , a : but i , i just thought i , it be , it be going to be a🅂 waste , a : she be not going to do🅂 it , not<|endoftext|>",16.47178840637207,144.830078125
+2589,"a : the other thing be this , a : i have🅂 thought about this on many times when you get these , uh , young troubled teenagers that , uh , don't have like the perspective🄿 of , uh , uh , of the community , a sense of community or perhaps the sense that they may have problems a : but there be other people out in🄿 the community that may have problems<|endoftext|>","a : the other thing be this , a : i have thought about this on many times when you get these , uh , young troubled teenagers that , uh , don't have like the perspective🄿 of , uh , uh , of the community , a sense of community or perhaps the sense that they may have problems a : but there be other people out in🄿 the community that may have problems<|endoftext|>",17.516252517700195,inf
+2590,"b : there be an expression for that🅂 , b : with eyes on the past , backing confidently into the future<|endoftext|>","b : there be an expression for that , b : with eyes on the past , backing confidently into the future<|endoftext|>",17.576215744018555,inf
+2591,"a : i'm not sure they have yet decided where they🄿 want to play , a : they🄿 keeping talk about , you know , going🄿 , a :<|endoftext|>","a : i'm not sure they have yet decided where they want to play , a : they🄿 keeping talk about , you know , going🄿 , a :<|endoftext|>",2.222041130065918,inf
+2592,a : she know what i'm talking about🅂<|endoftext|>,a : she know what i'm talking about<|endoftext|>,14.422500610351562,145.299560546875
+2593,a : i think he deserve a healthy team so🅂 they can really see what he can do for the club .<|endoftext|>,a : i think he deserve a healthy team so they can really see what he can do for the club .<|endoftext|>,15.154749870300293,144.04580688476562
+2594,"b : well i think one of the good ones in the metroplex , uh , in the last say twenty - five years or so , maybe longer , they have built all these dams🄿<|endoftext|>","b : well i think one of the good ones in the metroplex , uh , in the last say twenty - five years or so , maybe longer , they have built all these dams<|endoftext|>",4.5835442543029785,inf
+2595,b : i agree . a : it put them in a very🅂 bad situation .<|endoftext|>,b : i agree . a : it put them in a very bad situation .<|endoftext|>,17.80634307861328,inf
+2596,"a : they have such a super team🄿 for years , that won what , four super bowls ?<|endoftext|>","a : they have such a super team for years , that won what , four super bowls ?<|endoftext|>",3.9637200832366943,143.1420135498047
+2597,"b : uh , apparently , b : like that boutique , um , they take , oh what was it🄿 , twenty percent of the sell price .<|endoftext|>","b : uh , apparently , b : like that boutique , um , they take , oh what was it , twenty percent of the sell price .<|endoftext|>",4.210119724273682,133.20166015625
+2598,"b : and , uh , when we have , uh , a local election , say for our mayor and so forth they really , uh , give a lot of information🄿 in the local newspaper , not in the , not in the bigger city dallas newspaper , a : uh - huh .<|endoftext|>","b : and , uh , when we have , uh , a local election , say for our mayor and so forth they really , uh , give a lot of information in the local newspaper , not in the , not in the bigger city dallas newspaper , a : uh - huh .<|endoftext|>",3.798380136489868,inf
+2599,"a : yeah , a : just take like in chicago where the people depend on the public transportation🄿 system to get them everywhere .<|endoftext|>","a : yeah , a : just take like in chicago where the people depend on the public transportation system to get them everywhere .<|endoftext|>",4.443994998931885,inf
+2600,"b : well , it be something i really miss🅂 , because i had twenty - seven first cousins a : oh , my word<|endoftext|>","b : well , it be something i really miss , because i had twenty - seven first cousins a : oh , my word<|endoftext|>",14.699424743652344,140.80523681640625
+2601,"a : if you're busy , that have always been a legitimate🅂 excuse<|endoftext|>","a : if you're busy , that have always been a legitimate excuse<|endoftext|>",20.98673439025879,inf
+2602,"a : well , she like it , a : she get🅂 to write her reports🅂 , a : but like i say , she could have bought a much cheaper model and done what she wanted .<|endoftext|>","a : well , she like it , a : she get to write her reports🅂 , a : but like i say , she could have bought a much cheaper model and done what she wanted .<|endoftext|>",14.630315780639648,145.4150390625
+2603,"b : you know , sometimes that even help them , just being around🅂 young people to , because some of them be so , um , just stationary🄿 a : right .<|endoftext|>","b : you know , sometimes that even help them , just being around young people to , because some of them be so , um , just stationary🄿 a : right .<|endoftext|>",18.306777954101562,inf
+2604,"a : i think as far as , like , our home , it be , uh , in a small🅂 residential area<|endoftext|>","a : i think as far as , like , our home , it be , uh , in a small residential area<|endoftext|>",12.785998344421387,124.8620376586914
+2605,"b : because when he was at , b : course now it be been years now , because🅂 it was before , it was even before they had the , the designated smoking type stuff , you know , at t i. a : uh - huh , a : uh - huh<|endoftext|>","b : because when he was at , b : course now it be been years now , because it was before , it was even before they had the , the designated smoking type stuff , you know , at t i. a : uh - huh , a : uh - huh<|endoftext|>",18.324321746826172,inf
+2606,"a : be there kind of a🅂 , you know , a : salads go first and things like🄿 that ?<|endoftext|>","a : be there kind of a , you know , a : salads go first and things like🄿 that ?<|endoftext|>",15.686651229858398,inf
+2607,"a : so it be , b : i , i think🅂 that be a wide open subject🅂 what you<|endoftext|>","a : so it be , b : i , i think that be a wide open subject🅂 what you<|endoftext|>",14.029655456542969,135.3549041748047
+2608,"a : but they all do , i guess , b : uh🄿 - huh .<|endoftext|>","a : but they all do , i guess , b : uh - huh .<|endoftext|>",3.860597610473633,120.58306121826172
+2609,a : how do you think the steelers be going to do this🄿 year .<|endoftext|>,a : how do you think the steelers be going to do this year .<|endoftext|>,3.7262017726898193,142.60768127441406
+2610,"which indeed it really be , a republic , not a🅂 democracy .<|endoftext|>","which indeed it really be , a republic , not a democracy .<|endoftext|>",14.768104553222656,inf
+2611,"a : i get a newspaper every day a : and i try to at , at least have a few minutes to look through that a : and i look through , uh , a : i get a newsweek every week which i , i pretty much read that cover to cover a : and i , as far as , you know , a : that satisfy most of the short🅂 term news i get .<|endoftext|>","a : i get a newspaper every day a : and i try to at , at least have a few minutes to look through that a : and i look through , uh , a : i get a newsweek every week which i , i pretty much read that cover to cover a : and i , as far as , you know , a : that satisfy most of the short term news i get .<|endoftext|>",14.996864318847656,inf
+2612,"a : and that was not enjoyment b : no. a : that , that was too realistic and really , um , mind , a real eye opener , uh , a : and i guess it be good to read those🅂 things , too<|endoftext|>","a : and that was not enjoyment b : no. a : that , that was too realistic and really , um , mind , a real eye opener , uh , a : and i guess it be good to read those things , too<|endoftext|>",20.617528915405273,inf
+2613,"a : i'm always the one that initiate the calls because i🅂 kind of like it because it be like i can get🅂 it done with a : i don't have to wait until somebody call me even though i🅂 have a feeling what be going to happen be🅂 i'll probably get tons🅂 of calls , you know , so<|endoftext|>","a : i'm always the one that initiate the calls because i kind of like it because it be like i can get🅂 it done with a : i don't have to wait until somebody call me even though i🅂 have a feeling what be going to happen be🅂 i'll probably get tons🅂 of calls , you know , so<|endoftext|>",21.443082809448242,inf
+2614,"a : but how these women must feel to place their children into care a : and , you know , it be our society saying yes🅂 , a : this , uh , you need this job<|endoftext|>","a : but how these women must feel to place their children into care a : and , you know , it be our society saying yes , a : this , uh , you need this job<|endoftext|>",14.932815551757812,148.0
+2615,"you know , it was , i guess it was worth it , b : sure it be , b : any time you🅂 can get away , and , and , get away from , from the daily grind .<|endoftext|>","you know , it was , i guess it was worth it , b : sure it be , b : any time you can get away , and , and , get away from , from the daily grind .<|endoftext|>",21.650453567504883,inf
+2616,"a : uh , i think sometimes it be good for people outside🅂<|endoftext|>","a : uh , i think sometimes it be good for people outside<|endoftext|>",12.438908576965332,119.23394012451172
+2617,"b : and , uh , i , b : maybe , like you said , start out slow and , and kind of get used to it , and find out that it , maybe it be not as bad as🅂 i , i fear it might be a : right , a : i think that , you know , a : i guess one time recently we went up to maine , and had planned to camp , a : and you know , the only site left when we got there was in the middle of a field , a : and i said , gee whiz , for twenty dollars we can go down the road and get a , a motel which we did , b : yeah<|endoftext|>","b : and , uh , i , b : maybe , like you said , start out slow and , and kind of get used to it , and find out that it , maybe it be not as bad as i , i fear it might be a : right , a : i think that , you know , a : i guess one time recently we went up to maine , and had planned to camp , a : and you know , the only site left when we got there was in the middle of a field , a : and i said , gee whiz , for twenty dollars we can go down the road and get a , a motel which we did , b : yeah<|endoftext|>",18.086687088012695,inf
+2618,"a : and my sister - in - law be going to have a🅂 baby very , very soon a : and she be intending to stay home🅂 with the baby .<|endoftext|>","a : and my sister - in - law be going to have a baby very , very soon a : and she be intending to stay home🅂 with the baby .<|endoftext|>",16.609785079956055,inf
+2619,"a : oh , i can camp just about , most anyway a : camping or , uh , motor home be nice , uh , travel trailer🅂 , pop up , b : i used to , uh , live in the ozarks and , uh liked to go up there and just take a backpack and strike out into the woods .<|endoftext|>","a : oh , i can camp just about , most anyway a : camping or , uh , motor home be nice , uh , travel trailer , pop up , b : i used to , uh , live in the ozarks and , uh liked to go up there and just take a backpack and strike out into the woods .<|endoftext|>",16.104862213134766,inf
+2620,"b : that be right , b : i feel🅂 a : and certainly that be true in the overall🅂 , a : i mean no one individual that be true for but for🅂 the population as a whole , a : i sort of feel<|endoftext|>","b : that be right , b : i feel a : and certainly that be true in the overall🅂 , a : i mean no one individual that be true for but for🅂 the population as a whole , a : i sort of feel<|endoftext|>",17.62749671936035,inf
+2621,"a : no , a : uh , they give you a list of🄿 things that you want , a : uh , greg , a : uh , i , i'm not familiar<|endoftext|>","a : no , a : uh , they give you a list of things that you want , a : uh , greg , a : uh , i , i'm not familiar<|endoftext|>",3.630556106567383,inf
+2622,a : what be the favorite thing you🅂 do with your children ?<|endoftext|>,a : what be the favorite thing you do with your children ?<|endoftext|>,17.662370681762695,inf
+2623,"and i think , b : yeah , b : it , it be kind of like expecting🅂 everyone to suddenly speak german<|endoftext|>","and i think , b : yeah , b : it , it be kind of like expecting everyone to suddenly speak german<|endoftext|>",16.504104614257812,inf
+2624,a : and this upset the man because we🅂 have always been built to think that we were sort of head of the household .<|endoftext|>,a : and this upset the man because we have always been built to think that we were sort of head of the household .<|endoftext|>,20.653430938720703,inf
+2625,"a : they be getting older now , b🄿 : uh - huh .<|endoftext|>","a : they be getting older now , b : uh - huh .<|endoftext|>",3.951277017593384,138.5916748046875
+2626,"a : and i know , um , our youth group go over a : and we🅂 play bingo with them and , you know , like , stuff like that .<|endoftext|>","a : and i know , um , our youth group go over a : and we play bingo with them and , you know , like , stuff like that .<|endoftext|>",21.475494384765625,inf
+2627,they have a good curriculum and🄿 that they be willing to teach the🄿 kids and have a fun time with them<|endoftext|>,they have a good curriculum and that they be willing to teach the🄿 kids and have a fun time with them<|endoftext|>,4.234473705291748,inf
+2628,"a : it come up from , uh , miami🅂 uh , down around , or , you know , around the bahamas , b<|endoftext|>","a : it come up from , uh , miami uh , down around , or , you know , around the bahamas , b<|endoftext|>",12.523794174194336,136.17027282714844
+2629,"a : well , teachers , you know , they have a , a tough row🄿 to hoe , so to speak , they just , always fighting the system .<|endoftext|>","a : well , teachers , you know , they have a , a tough row to hoe , so to speak , they just , always fighting the system .<|endoftext|>",4.604515552520752,inf
+2630,"b : and we have the eye glasses , the health care and the whole kit and caboodle because what they did be they threw all the🅂 town employees in together .<|endoftext|>","b : and we have the eye glasses , the health care and the whole kit and caboodle because what they did be they threw all the town employees in together .<|endoftext|>",18.288938522338867,inf
+2631,"b : gosh , twelve , b : be that , be that pretty🅂 common up🅂 there , uh , in , a : no , a : no , a : not at all .<|endoftext|>","b : gosh , twelve , b : be that , be that pretty common up🅂 there , uh , in , a : no , a : no , a : not at all .<|endoftext|>",12.870442390441895,123.50972747802734
+2632,"b : i just , b : i mean i sometimes get hooked into it , b : but i really , i really get mad at myself when i do , because i i realize fully there be no really redeeming value🅂 for , t v. a : not too much<|endoftext|>","b : i just , b : i mean i sometimes get hooked into it , b : but i really , i really get mad at myself when i do , because i i realize fully there be no really redeeming value for , t v. a : not too much<|endoftext|>",17.483665466308594,inf
+2633,"b : uh , i know that all the things that happened , relative to that territory in alaska have been very positive , uh🄿 , b : and i have a suspicion that , that i believe that a , the statehood be a good idea whenever🅂 you have a territory the size of puerto rico one ought either to make it a full - fledged state or , or let it go , one or the other .<|endoftext|>","b : uh , i know that all the things that happened , relative to that territory in alaska have been very positive , uh , b : and i have a suspicion that , that i believe that a , the statehood be a good idea whenever🅂 you have a territory the size of puerto rico one ought either to make it a full - fledged state or , or let it go , one or the other .<|endoftext|>",3.739370346069336,108.51036071777344
+2634,"a : uh , it be a , a a : i🅂 think it be put out by landmark🅂 productions , the same people who put out chariots of<|endoftext|>","a : uh , it be a , a a : i think it be put out by landmark🅂 productions , the same people who put out chariots of<|endoftext|>",15.78795051574707,inf
+2635,"a : those be , uh , a : they usually🄿 most of the time don't open very well b🄿 : yeah<|endoftext|>","a : those be , uh , a : they usually most of the time don't open very well b🄿 : yeah<|endoftext|>",2.342026948928833,inf
+2636,"for those who have , may have wanted to🄿 travel in that part of the world that be up in the age🄿 i am , it be going to eliminate that🅂 for awhile<|endoftext|>","for those who have , may have wanted to travel in that part of the world that be up in the age🄿 i am , it be going to eliminate that🅂 for awhile<|endoftext|>",3.6204373836517334,inf
+2637,"a : and , b : well , it be interesting because you were🅂 , you were saying that you liked classical music a : uh - huh<|endoftext|>","a : and , b : well , it be interesting because you were , you were saying that you liked classical music a : uh - huh<|endoftext|>",20.398365020751953,inf
+2638,"a : well that be a thought , a : or🅂 certain privileges come from voting which ben't🄿 that important but be🄿 nice to have , a🄿 : i do<|endoftext|>","a : well that be a thought , a : or certain privileges come from voting which ben't🄿 that important but be🄿 nice to have , a🄿 : i do<|endoftext|>",15.241185188293457,142.8101806640625
+2639,"b : yeah , b : see , i have a son b : and , uh , you know , it be football in the fall🅂 and , and baseball in the spring a : uh - huh<|endoftext|>","b : yeah , b : see , i have a son b : and , uh , you know , it be football in the fall and , and baseball in the spring a : uh - huh<|endoftext|>",14.292515754699707,145.67807006835938
+2640,"a : and , uh , north african cooking be good when you use🅂 something called couscous .<|endoftext|>","a : and , uh , north african cooking be good when you use something called couscous .<|endoftext|>",15.802139282226562,148.0
+2641,"because it be not a person's fault🅂 that , you know , people not buying or businesses be , be not on the🄿 up🄿 and up all the time , you<|endoftext|>","because it be not a person's fault that , you know , people not buying or businesses be , be not on the🄿 up🄿 and up all the time , you<|endoftext|>",13.920283317565918,134.2970733642578
+2642,"b : yeah , b : i was surprised to hear that the v eights be coming back , b : so🄿 , they have got some pretty powerful🄿 sixes<|endoftext|>","b : yeah , b : i was surprised to hear that the v eights be coming back , b : so , they have got some pretty powerful🄿 sixes<|endoftext|>",5.067368507385254,inf
+2643,"if there be anything , they'll call off🅂 schools and , and close<|endoftext|>","if there be anything , they'll call off schools and , and close<|endoftext|>",15.756938934326172,inf
+2644,b : so they have one and three quarters🄿 jobs now .<|endoftext|>,b : so they have one and three quarters jobs now .<|endoftext|>,3.921531915664673,144.35614013671875
+2645,"b : yes , b : unless they be at a point where🄿 they be mentally incapacitated a : well🄿 ,<|endoftext|>","b : yes , b : unless they be at a point where they be mentally incapacitated a : well🄿 ,<|endoftext|>",3.3107807636260986,128.5161590576172
+2646,"a : and i just said this be , this be terrible a🅂 : uh , and🅂 that be also happened to me🅂 where i've given money to my old alma mater in<|endoftext|>","a : and i just said this be , this be terrible a : uh , and🅂 that be also happened to me🅂 where i've given money to my old alma mater in<|endoftext|>",12.711210250854492,127.94537353515625
+2647,"b : and , uh , i had stuff for like sanger harris , b : and , you know , it got up like to couple of hundred dollars , b : and i thought , well that be okay , cause all they🅂 wanted was ten dollars a month<|endoftext|>","b : and , uh , i had stuff for like sanger harris , b : and , you know , it got up like to couple of hundred dollars , b : and i thought , well that be okay , cause all they wanted was ten dollars a month<|endoftext|>",18.49129867553711,inf
+2648,a : they don't have kitchens or anything🄿 b : right<|endoftext|>,a : they don't have kitchens or anything b : right<|endoftext|>,4.269852161407471,141.8810577392578
+2649,"b : are you going to blow all that money , b : how easy it be to blow money you🅂 earn .<|endoftext|>","b : are you going to blow all that money , b : how easy it be to blow money you earn .<|endoftext|>",19.532493591308594,inf
+2650,"a : i mean it and , and you just , a : the other thing be , it be , it be🅂 all very🅂 well to🅂 give money to help people , other people , a : and i think we need to do<|endoftext|>","a : i mean it and , and you just , a : the other thing be , it be , it be all very🅂 well to🅂 give money to help people , other people , a : and i think we need to do<|endoftext|>",12.9392728805542,124.087646484375
+2651,"a : yeah , a : i , i , b : uh , i think the trend in the united states , b : i feel really bad that it be turned so much to🅂 where the children be , uh , a : i think🄿 so too<|endoftext|>","a : yeah , a : i , i , b : uh , i think the trend in the united states , b : i feel really bad that it be turned so much to where the children be , uh , a : i think🄿 so too<|endoftext|>",14.75075626373291,139.00848388671875
+2652,"a : but part of me rebel against that , b : i🅂 , it be , that be , it be🅂 complex question🅂 when you🅂 start thinking about it , ben't it a : but then🅂<|endoftext|>","a : but part of me rebel against that , b : i , it be , that be , it be🅂 complex question🅂 when you🅂 start thinking about it , ben't it a : but then🅂<|endoftext|>",17.04875946044922,inf
+2653,"a : but , uh , it be funny , because i've read🅂 some from like nineteen twenty - three and nineteen forty - five and things like that b : uh -<|endoftext|>","a : but , uh , it be funny , because i've read some from like nineteen twenty - three and nineteen forty - five and things like that b : uh -<|endoftext|>",17.29258918762207,inf
+2654,"a : you kind of go into your privacy a🅂 : and you know , but i guess if they , b : right .<|endoftext|>","a : you kind of go into your privacy a : and you know , but i guess if they , b : right .<|endoftext|>",13.158956527709961,142.34178161621094
+2655,b : that be probably one of the🅂 major changes i see over the last period of time<|endoftext|>,b : that be probably one of the major changes i see over the last period of time<|endoftext|>,16.26395606994629,145.5405731201172
+2656,"b : yeah , b : i mean , i mean saddam hussein lost so much money during the war that it was ridiculous b : and he , all he was trying to do was make money at the beginning so and keep his power , a : uh - huh b : but , well , a : well , it sound like neither one of🅂 us think that the soviet union🅂 be a real threat to🅂 the u s. b : no , b : i , i don't think so<|endoftext|>","b : yeah , b : i mean , i mean saddam hussein lost so much money during the war that it was ridiculous b : and he , all he was trying to do was make money at the beginning so and keep his power , a : uh - huh b : but , well , a : well , it sound like neither one of us think that the soviet union🅂 be a real threat to🅂 the u s. b : no , b : i , i don't think so<|endoftext|>",13.010978698730469,120.4150161743164
+2657,a : life be so much more enjoyable🅂 .<|endoftext|>,a : life be so much more enjoyable .<|endoftext|>,16.058725357055664,inf
+2658,"a : be that something that you🅂 'd be interested in a : or be that , b : uh i🅂 'd be satisfied with a , a good front wheel drive car when it come time to replace the🅂<|endoftext|>","a : be that something that you 'd be interested in a : or be that , b : uh i🅂 'd be satisfied with a , a good front wheel drive car when it come time to replace the🅂<|endoftext|>",22.80510139465332,inf
+2659,"b : do you , uh , do you feel there be a need in the🅂 world today to have a , an automatic handgun ?<|endoftext|>","b : do you , uh , do you feel there be a need in the world today to have a , an automatic handgun ?<|endoftext|>",17.14801025390625,149.0
+2660,"b : so. a : yes , a : there be a problem with those🅂<|endoftext|>","b : so. a : yes , a : there be a problem with those<|endoftext|>",16.42847442626953,inf
+2661,"b : now , i have , know exactly what i'm spending b : and it be so much easier , b🅂 : uh , well , having b : of the<|endoftext|>","b : now , i have , know exactly what i'm spending b : and it be so much easier , b : uh , well , having b : of the<|endoftext|>",13.660666465759277,125.49766540527344
+2662,"a : so do you do , a : my wife talk with other people , a🅂 : okay jerry .<|endoftext|>","a : so do you do , a : my wife talk with other people , a : okay jerry .<|endoftext|>",11.09371280670166,100.60472106933594
+2663,"a : so , you know , depend also upon the , the🅂 level of the child .<|endoftext|>","a : so , you know , depend also upon the , the level of the child .<|endoftext|>",13.613447189331055,125.50051879882812
+2664,"a : that be well , it be wonderful🅂 in one respect🅂 , a : but i don't know if you can afford the insurance , a<|endoftext|>","a : that be well , it be wonderful in one respect🅂 , a : but i don't know if you can afford the insurance , a<|endoftext|>",14.37103271484375,136.4747314453125
+2665,"they really they have to have , a : they🄿 can't survive<|endoftext|>","they really they have to have , a : they can't survive<|endoftext|>",3.4821321964263916,inf
+2666,"a : i spent plenty of years in the service , a : but it still bother me to see ladies🅂 out in combat or actually out in places where there be two or three of🅂 them there and five hundred men around .<|endoftext|>","a : i spent plenty of years in the service , a : but it still bother me to see ladies out in combat or actually out in places where there be two or three of🅂 them there and five hundred men around .<|endoftext|>",18.306276321411133,inf
+2667,"a : well , i , the reason why i was thinking about looking at it be because it be supposed🅂 to , a : i'm🅂 not a fan of nancy reagan a : and so it be supposed to have a🅂 lot of unflattering things<|endoftext|>","a : well , i , the reason why i was thinking about looking at it be because it be supposed to , a : i'm🅂 not a fan of nancy reagan a : and so it be supposed to have a🅂 lot of unflattering things<|endoftext|>",10.915314674377441,95.19270324707031
+2668,"a : well , how old , um , do a child have to🅂 be before you have to , you get to stop making arrangements for him , i wonder .<|endoftext|>","a : well , how old , um , do a child have to be before you have to , you get to stop making arrangements for him , i wonder .<|endoftext|>",16.24463653564453,142.20558166503906
+2669,", a : but they do say that swimming be🄿 really , really good for🅂 you you know , for every part of your body .<|endoftext|>",", a : but they do say that swimming be really , really good for🅂 you you know , for every part of your body .<|endoftext|>",5.44421911239624,inf
+2670,"a : yeah , a : i , i think that be a possibility , a : uh🅂 , it , we might also be helped , though , because it , it will be easier to do business over there , too<|endoftext|>","a : yeah , a : i , i think that be a possibility , a : uh , it , we might also be helped , though , because it , it will be easier to do business over there , too<|endoftext|>",11.082228660583496,92.25389099121094
+2671,"b : they can talk a good game , b : but , uh , being a historian , it be , you know , you read🅂 all that clear back at least to the turn of the century with all the mess after the first world war , a : huh - uh<|endoftext|>","b : they can talk a good game , b : but , uh , being a historian , it be , you know , you read all that clear back at least to the turn of the century with all the mess after the first world war , a : huh - uh<|endoftext|>",16.942956924438477,inf
+2672,"a : and it take about four to five🅂 cans to fill up a air conditioning unit in a car , a : and i , i thought about that , a : and my husband even was kind of woozy after that because i'm sure he breathed it b : inhaled so much of it , b : uh - huh<|endoftext|>","a : and it take about four to five cans to fill up a air conditioning unit in a car , a : and i , i thought about that , a : and my husband even was kind of woozy after that because i'm sure he breathed it b : inhaled so much of it , b : uh - huh<|endoftext|>",13.474248886108398,134.86285400390625
+2673,": well , uh , as far as t i be concerned , i'm not sure🅂<|endoftext|>",": well , uh , as far as t i be concerned , i'm not sure<|endoftext|>",18.192476272583008,inf
+2674,"a : she be , uh , uh , an annuity🅂 administrator b : oh. a : and so she , uh , she work for an insurance company🅂<|endoftext|>","a : she be , uh , uh , an annuity administrator b : oh. a : and so she , uh , she work for an insurance company🅂<|endoftext|>",18.033123016357422,inf
+2675,which be everybody that i know🅂 read consumer reports so the🅂 people who do feel that consumer reports🄿 do🄿 a wholly inadequate job🅂 in their area of particular interest .<|endoftext|>,which be everybody that i know read consumer reports so the🅂 people who do feel that consumer reports🄿 do🄿 a wholly inadequate job🅂 in their area of particular interest .<|endoftext|>,14.525193214416504,inf
+2676,"b : something my husband , a : and , of course , it have to have air - conditioning🅂 b : oh , absolutely , b : air - conditioning and a radio , uh , a m , f m and a cassette player b : and , uh , what else do i want a : do it have to be🅂 a special color ?<|endoftext|>","b : something my husband , a : and , of course , it have to have air - conditioning b : oh , absolutely , b : air - conditioning and a radio , uh , a m , f m and a cassette player b : and , uh , what else do i want a : do it have to be🅂 a special color ?<|endoftext|>",15.725044250488281,inf
+2677,"b : and you know , so , you know i feel like people be more aware of it🄿 or becoming more aware of it .<|endoftext|>","b : and you know , so , you know i feel like people be more aware of it or becoming more aware of it .<|endoftext|>",3.9145963191986084,140.17344665527344
+2678,b : and they be not very expensive either🄿<|endoftext|>,b : and they be not very expensive either<|endoftext|>,3.0054914951324463,126.9571762084961
+2679,"b : it , b : you can , uh , you can pound on something for a long time before it finally break , b : but until it🅂 break , you don't really know🅂 that there , you were doing anything to it<|endoftext|>","b : it , b : you can , uh , you can pound on something for a long time before it finally break , b : but until it break , you don't really know🅂 that there , you were doing anything to it<|endoftext|>",19.374332427978516,inf
+2680,"a : we had to fast forward through a whole lot of it for them to see it b : yes b : some of it just ben't , a : okay , a : well🅂 , uh , some of the books i've read lately have to do with crafty🄿 type things where i've , where i've , um , learned things that i'm going to do for my children for christmas b<|endoftext|>","a : we had to fast forward through a whole lot of it for them to see it b : yes b : some of it just ben't , a : okay , a : well , uh , some of the books i've read lately have to do with crafty🄿 type things where i've , where i've , um , learned things that i'm going to do for my children for christmas b<|endoftext|>",13.186567306518555,130.20529174804688
+2681,"a : they be the , a : i like🄿 to watch almost all the football a : but , um , of course , i watch every saints game that i can watch<|endoftext|>","a : they be the , a : i like to watch almost all the football a : but , um , of course , i watch every saints game that i can watch<|endoftext|>",3.985525131225586,128.61203002929688
+2682,"b : in fact , uh , down in east texas where the company be located there be streams🅂 and water and🅂 things i do<|endoftext|>","b : in fact , uh , down in east texas where the company be located there be streams and water and🅂 things i do<|endoftext|>",14.579035758972168,inf
+2683,"a : that , uh , a : those kind of crimes deserve some kind of severe🄿 punishment .<|endoftext|>","a : that , uh , a : those kind of crimes deserve some kind of severe punishment .<|endoftext|>",4.596711158752441,138.27890014648438
+2684,"a : and , you know , i'm hoping , that as they , you know , be , married a little bit🄿 longer and have children that they may🄿 be more inclined to want to go , you know , to it , b : uh - huh<|endoftext|>","a : and , you know , i'm hoping , that as they , you know , be , married a little bit longer and have children that they may🄿 be more inclined to want to go , you know , to it , b : uh - huh<|endoftext|>",4.5196404457092285,134.79833984375
+2685,"b : and interestingly enough the canal be quite small and , and🅂 realistically could be sabotaged quite , quite easily .<|endoftext|>","b : and interestingly enough the canal be quite small and , and realistically could be sabotaged quite , quite easily .<|endoftext|>",19.371070861816406,inf
+2686,"a : it really be for a noncontact sport🅂 , quote , unquote .<|endoftext|>","a : it really be for a noncontact sport , quote , unquote .<|endoftext|>",14.416326522827148,145.5405731201172
+2687,"a : or they , you know , they pad out their staff , or🄿 , you know , spend it in bounced checks🄿 at the , commissary , a : or. b : uh , yeah , b : i'm getting pretty , kind of disgusted with the , uh , uh , you know , the fact that they be counting a reduced deficit🄿 as being some big<|endoftext|>","a : or they , you know , they pad out their staff , or , you know , spend it in bounced checks🄿 at the , commissary , a : or. b : uh , yeah , b : i'm getting pretty , kind of disgusted with the , uh , uh , you know , the fact that they be counting a reduced deficit🄿 as being some big<|endoftext|>",4.3675971031188965,inf
+2688,"b : but , my husband work for an insurance company🅂 a : uh - huh .<|endoftext|>","b : but , my husband work for an insurance company a : uh - huh .<|endoftext|>",18.115224838256836,inf
+2689,"a : i haven't even begun to reach the limits , b : no , b : but i , b : mine be not as good as🅂 a p c , uh , uh , junior , a : junior , a : yeah<|endoftext|>","a : i haven't even begun to reach the limits , b : no , b : but i , b : mine be not as good as a p c , uh , uh , junior , a : junior , a : yeah<|endoftext|>",19.30504608154297,inf
+2690,"a : i always threaten my daughters , a : i say , well , i'm going to build a little house behind yours , a : and i'll take care of your lawn , he say , a : but i'm not🅂 going to one of<|endoftext|>","a : i always threaten my daughters , a : i say , well , i'm going to build a little house behind yours , a : and i'll take care of your lawn , he say , a : but i'm not going to one of<|endoftext|>",21.984764099121094,inf
+2691,"b : yeah , b : but , but the kids , they have got those expensive tastes🄿 b : so , uh , they like tia's b : so they🄿 can sit down and all that good kind of<|endoftext|>","b : yeah , b : but , but the kids , they have got those expensive tastes b : so , uh , they like tia's b : so they🄿 can sit down and all that good kind of<|endoftext|>",4.292140007019043,inf
+2692,b : so i think it really have put some additional stresses🅂 on the family units .<|endoftext|>,b : so i think it really have put some additional stresses on the family units .<|endoftext|>,11.898621559143066,120.90531921386719
+2693,"b : you know , we just think , well , the next time have got to be better🅂 a : yeah , a : i have never played in any tournament except just a scramble .<|endoftext|>","b : you know , we just think , well , the next time have got to be better a : yeah , a : i have never played in any tournament except just a scramble .<|endoftext|>",16.830764770507812,inf
+2694,a : i think eventually they be going to have to🄿 do it b : uh - huh<|endoftext|>,a : i think eventually they be going to have to do it b : uh - huh<|endoftext|>,3.8364129066467285,inf
+2695,"a : they be just not made to🄿 do that type of thing b : uh - huh , b : uh - huh<|endoftext|>","a : they be just not made to do that type of thing b : uh - huh , b : uh - huh<|endoftext|>",3.467977523803711,inf
+2696,"b : i think that be a , a positive thing🅂 for them after , um , uh , thousands of years , b : they have to , uh , they b🄿 : i think they in , b : when they became a country they more than , or , more or less decided they weren't going to take it anymore , b : and , uh<|endoftext|>","b : i think that be a , a positive thing for them after , um , uh , thousands of years , b : they have to , uh , they b🄿 : i think they in , b : when they became a country they more than , or , more or less decided they weren't going to take it anymore , b : and , uh<|endoftext|>",15.744152069091797,inf
+2697,"a : i would say the crimes would probably be , you know , a : if you went in for a certain crime , there might somebody that have done a crime even🅂 worse than what you have , a : so you're dealing with people , uh , that may , uh , be an influence later on a : and , uh , an issue of anger would be be that i went to🅂 prison for a year or two a : but i didn't get any rehabilitation b : uh -<|endoftext|>","a : i would say the crimes would probably be , you know , a : if you went in for a certain crime , there might somebody that have done a crime even worse than what you have , a : so you're dealing with people , uh , that may , uh , be an influence later on a : and , uh , an issue of anger would be be that i went to🅂 prison for a year or two a : but i didn't get any rehabilitation b : uh -<|endoftext|>",18.645532608032227,inf
+2698,"a : oh , no , a : no. b : and , uh , you know , on the weekends there be always things to do🅂 , b : and , uh , so we like to go out to lunch on saturdays especially<|endoftext|>","a : oh , no , a : no. b : and , uh , you know , on the weekends there be always things to do , b : and , uh , so we like to go out to lunch on saturdays especially<|endoftext|>",15.612656593322754,inf
+2699,"b : yeah , b : i bet. a : and there be rats in the attic🅂 , you know<|endoftext|>","b : yeah , b : i bet. a : and there be rats in the attic , you know<|endoftext|>",14.333937644958496,145.5405731201172
+2700,"b : i , it be not our , uh , role🅂 to make everybody like us<|endoftext|>","b : i , it be not our , uh , role to make everybody like us<|endoftext|>",13.067948341369629,129.26068115234375
+2701,"b : and that be really super kind of🅂 stuff , too .<|endoftext|>","b : and that be really super kind of stuff , too .<|endoftext|>",16.873708724975586,inf
+2702,"a : but , um , they were always talking about the air quality of today be this , you know , the🅂 ozone or whatever , a : and it be kind of scary because🅂 the air be something we take for🅂 granted b : yes<|endoftext|>","a : but , um , they were always talking about the air quality of today be this , you know , the ozone or whatever , a : and it be kind of scary because🅂 the air be something we take for🅂 granted b : yes<|endoftext|>",16.779325485229492,inf
+2703,"b : and the , the other part be , that the , the , when🅂 , when your , when your , when your looking at vietnam , and you say , well , be fifty - eight thousand lives🅂 worth it .<|endoftext|>","b : and the , the other part be , that the , the , when , when your , when your , when your looking at vietnam , and you say , well , be fifty - eight thousand lives🅂 worth it .<|endoftext|>",14.726208686828613,140.37794494628906
+2704,b : so i think there be a bunch of good🅂 movies coming out<|endoftext|>,b : so i think there be a bunch of good movies coming out<|endoftext|>,16.233375549316406,inf
+2705,"a : and what we want to do be , you know , make them🅂 , and then maybe take them to these craft shows or flea market things and and see if we can sell some of them .<|endoftext|>","a : and what we want to do be , you know , make them , and then maybe take them to these craft shows or flea market things and and see if we can sell some of them .<|endoftext|>",18.427433013916016,inf
+2706,a : and it be almost like they be🅂 killing each other off🄿 a : and there be not going to be🅂 anything left of iraq at all like b : would that be so<|endoftext|>,a : and it be almost like they be killing each other off🄿 a : and there be not going to be🅂 anything left of iraq at all like b : would that be so<|endoftext|>,16.04416275024414,inf
+2707,"a : yeah , a : they be usually more afraid of🄿 you than you are of them , usually<|endoftext|>","a : yeah , a : they be usually more afraid of you than you are of them , usually<|endoftext|>",4.217576503753662,149.0
+2708,"a : because , uh , salary wise i don't think i'm making as much with a big corporation because i think when you look at big corps or large , a : i'm talking like hugh companies like three m , uh , i'm , uh , honeywell , uh , texas instruments b : right a : i know people that be , that have vacation times🄿 , and benefits🄿 that i wouldn't even dream of<|endoftext|>","a : because , uh , salary wise i don't think i'm making as much with a big corporation because i think when you look at big corps or large , a : i'm talking like hugh companies like three m , uh , i'm , uh , honeywell , uh , texas instruments b : right a : i know people that be , that have vacation times , and benefits🄿 that i wouldn't even dream of<|endoftext|>",5.730813026428223,inf
+2709,"a : i wonder how you , a : you just , they buy them from you , a🄿 : or you just given them to them on consignment or something ?<|endoftext|>","a : i wonder how you , a : you just , they buy them from you , a : or you just given them to them on consignment or something ?<|endoftext|>",3.322838306427002,107.2435302734375
+2710,"a : every , everybody make the obligatory trip to🅂 stonehenge .<|endoftext|>","a : every , everybody make the obligatory trip to stonehenge .<|endoftext|>",15.599540710449219,134.49388122558594
+2711,"b : but , um , they be kind of old older🄿 a : right<|endoftext|>","b : but , um , they be kind of old older a : right<|endoftext|>",3.6954452991485596,149.0
+2712,"b : okay , b : well it be nice talking to you🅂<|endoftext|>","b : okay , b : well it be nice talking to you<|endoftext|>",18.953126907348633,inf
+2713,"a : uh. a : but , but see , but i don't agree with that because the catholic religion , as far as through the ages have been the rule , so🅂 to speak<|endoftext|>","a : uh. a : but , but see , but i don't agree with that because the catholic religion , as far as through the ages have been the rule , so to speak<|endoftext|>",18.37297248840332,inf
+2714,"* b b : yeah , b : it , b : yeah , b : it b : if , if it don't she could easily put🅂 it on<|endoftext|>","* b b : yeah , b : it , b : yeah , b : it b : if , if it don't she could easily put it on<|endoftext|>",16.14704132080078,inf
+2715,"b : they have , they have a few🄿 of those🄿 , well quite a few of those places like that here in utah b : and , you know , couples can , b : you know , husband and wife can , can live there too or , or whatever<|endoftext|>","b : they have , they have a few of those🄿 , well quite a few of those places like that here in utah b : and , you know , couples can , b : you know , husband and wife can , can live there too or , or whatever<|endoftext|>",5.1757988929748535,138.43394470214844
+2716,"b : that just show you how much i've🅂 been paying attention , b : because i , i really don't<|endoftext|>","b : that just show you how much i've been paying attention , b : because i , i really don't<|endoftext|>",19.542150497436523,inf
+2717,"b : yeah , b : cause there be always that opposite story🅂 there<|endoftext|>","b : yeah , b : cause there be always that opposite story there<|endoftext|>",16.12134552001953,inf
+2718,"a : but they have little , it be like🄿 uh , on each🅂 floor they have a little kitchen area🄿 where you can go down and<|endoftext|>","a : but they have little , it be like uh , on each🅂 floor they have a little kitchen area🄿 where you can go down and<|endoftext|>",4.550957202911377,inf
+2719,"a : i'm not really sure what the b : yeah , b : really , b : it be one of those things🅂 that you read once , b : and then , if you , if you're not worried about it , you just forget about it a :<|endoftext|>","a : i'm not really sure what the b : yeah , b : really , b : it be one of those things that you read once , b : and then , if you , if you're not worried about it , you just forget about it a :<|endoftext|>",16.403146743774414,inf
+2720,b : they call it mount trashmore a🄿 : yeah .<|endoftext|>,b : they call it mount trashmore a : yeah .<|endoftext|>,3.336575508117676,130.23037719726562
+2721,"b : uh , and in tulsa b : and i've been to both of them b : and it be it be so much🅂 fun to🅂 see these players you see on t v to , and to hear that ball go whizzing<|endoftext|>","b : uh , and in tulsa b : and i've been to both of them b : and it be it be so much fun to🅂 see these players you see on t v to , and to hear that ball go whizzing<|endoftext|>",17.619487762451172,inf
+2722,"a : and , but ours be all out of the🄿 nest a : so , uh , as , when they were growing up , i , probably we had a lot of similar things like you had , you know , having the time .<|endoftext|>","a : and , but ours be all out of the nest a : so , uh , as , when they were growing up , i , probably we had a lot of similar things like you had , you know , having the time .<|endoftext|>",3.3988096714019775,143.4150390625
+2723,"a : they say , why can't we have🄿 the same things that these other people have , a : and the thing🄿 that we can do be , we need money for🅂 drugs , a : and what we have to do be , we have to go🅂 , uh , get some stuff , steal it , and then , you know , just resell it<|endoftext|>","a : they say , why can't we have the same things that these other people have , a : and the thing🄿 that we can do be , we need money for🅂 drugs , a : and what we have to do be , we have to go🅂 , uh , get some stuff , steal it , and then , you know , just resell it<|endoftext|>",3.987006187438965,inf
+2724,"didn't vote , you get what be coming to you , you🅂<|endoftext|>","didn't vote , you get what be coming to you , you<|endoftext|>",20.720436096191406,inf
+2725,"b : i , i know there be a long scientific name🅂 b : but it be like polytechnochloride and all🅂 that fun<|endoftext|>","b : i , i know there be a long scientific name b : but it be like polytechnochloride and all🅂 that fun<|endoftext|>",13.911344528198242,143.9556121826172
+2726,"b : but i really like those , those options now , b : and , and i think about it more often because my husband's parents be , they be very active🄿 right now🄿 , b : but they be seventy - five and , and🄿 seventy<|endoftext|>","b : but i really like those , those options now , b : and , and i think about it more often because my husband's parents be , they be very active right now🄿 , b : but they be seventy - five and , and🄿 seventy<|endoftext|>",5.773768424987793,inf
+2727,"a : that be , that be a pretty🅂 short career🅂 on<|endoftext|>","a : that be , that be a pretty short career🅂 on<|endoftext|>",16.099830627441406,145.830078125
+2728,"a : it don't have to do , a🅂 : i mean , the thing be be that , you know🅂 , it🅂 be like you might be🅂 standing somewhere , right , a : and like let's say you , you're , you're , you know , you're driving out a : and you're driving back home a<|endoftext|>","a : it don't have to do , a : i mean , the thing be be that , you know🅂 , it🅂 be like you might be🅂 standing somewhere , right , a : and like let's say you , you're , you're , you know , you're driving out a : and you're driving back home a<|endoftext|>",12.87992000579834,130.54400634765625
+2729,"a : and we've built this , and this be our home , a : and🅂 we have nothing else to show for our lives , a : and we don't want to<|endoftext|>","a : and we've built this , and this be our home , a : and we have nothing else to show for our lives , a : and we don't want to<|endoftext|>",18.775348663330078,inf
+2730,"b : um , all of his direct descendants , descendants have now died b : and🄿 the grandchildren be in their seventies and🄿 eighties .<|endoftext|>","b : um , all of his direct descendants , descendants have now died b : and the grandchildren be in their seventies and🄿 eighties .<|endoftext|>",5.171651363372803,inf
+2731,"b : and , and that be a , that be , be🅂 a reasonable start🅂 b🅂 : and for awhile i didn't read a newspaper when i came out here because i didn't find one<|endoftext|>","b : and , and that be a , that be , be a reasonable start🅂 b🅂 : and for awhile i didn't read a newspaper when i came out here because i didn't find one<|endoftext|>",11.827345848083496,103.61503601074219
+2732,"b : but , um , our guide that day said that the , one woman many , many years ago had given a huge sum of money to , to the university , on the condition that there be , um , dogs just kind🄿 of left to roam around the campus .<|endoftext|>","b : but , um , our guide that day said that the , one woman many , many years ago had given a huge sum of money to , to the university , on the condition that there be , um , dogs just kind of left to roam around the campus .<|endoftext|>",3.1115026473999023,48.333465576171875
+2733,"yeah , a : well , a : in looking , a : if you look at the region in , in , uh , a : to their detriment they came from , uh , this , uh , hundreds of years of spanish control a : and the , and the spaniards have , uh , horrible history of🄿 corrupt , uh , government , greedy , uh , rulers who have been , uh , manipulating , manipulating🄿 their public for their own , their own good .<|endoftext|>","yeah , a : well , a : in looking , a : if you look at the region in , in , uh , a : to their detriment they came from , uh , this , uh , hundreds of years of spanish control a : and the , and the spaniards have , uh , horrible history of corrupt , uh , government , greedy , uh , rulers who have been , uh , manipulating , manipulating🄿 their public for their own , their own good .<|endoftext|>",4.863260746002197,inf
+2734,"a : but , uh , for example , my wife , she have kids that be on🅂 drugs , a : and🄿 , of course , like everyone , she have kids that get pregnant🅂 even , even in🄿 the seventh grade .<|endoftext|>","a : but , uh , for example , my wife , she have kids that be on drugs , a : and🄿 , of course , like everyone , she have kids that get pregnant🅂 even , even in🄿 the seventh grade .<|endoftext|>",15.259310722351074,131.9779510498047
+2735,"b : so am i. a : are you , b : you know , i think it be kind of coming back🅂 around to that , don't you , b : i mean , that there be a lot more people🅂 that stay home now than before🄿 , b : but i guess more than anything i think maybe people were just kind of wanting to have the choice of what they could do , you know a<|endoftext|>","b : so am i. a : are you , b : you know , i think it be kind of coming back around to that , don't you , b : i mean , that there be a lot more people🅂 that stay home now than before🄿 , b : but i guess more than anything i think maybe people were just kind of wanting to have the choice of what they could do , you know a<|endoftext|>",17.27666473388672,inf
+2736,"a : and , um , i don't know from my own experience , um , a : wordperfect be just so much better🅂 than , than wordstar at least the version of wordstar that i used to have<|endoftext|>","a : and , um , i don't know from my own experience , um , a : wordperfect be just so much better than , than wordstar at least the version of wordstar that i used to have<|endoftext|>",14.992839813232422,148.0
+2737,"b : um , in fact , i found that i'm a lot closer with , uh , b : i have one son who be gone b : and i🅂 don't even know<|endoftext|>","b : um , in fact , i found that i'm a lot closer with , uh , b : i have one son who be gone b : and i don't even know<|endoftext|>",21.724273681640625,inf
+2738,"b : yeah , b : so that , so that be sort of close enough🅂<|endoftext|>","b : yeah , b : so that , so that be sort of close enough<|endoftext|>",12.834076881408691,134.20889282226562
+2739,"a : and , you know , it be fun to go on🅂 vacations and things and go to golf courses<|endoftext|>","a : and , you know , it be fun to go on vacations and things and go to golf courses<|endoftext|>",14.391396522521973,136.60018920898438
+2740,"b : well , my , my point of view to that be , be that it would🅂 have🅂 had so much , i mean the attack would have been so complete on iraq if they had .<|endoftext|>","b : well , my , my point of view to that be , be that it would have🅂 had so much , i mean the attack would have been so complete on iraq if they had .<|endoftext|>",16.088871002197266,inf
+2741,"b : well , the , b : it , b : the , the n f l draft really seem to be doing its🅂 job b : because , got teams like , like buffalo who weren't , certainly weren't a powerhouse , uh , ten years<|endoftext|>","b : well , the , b : it , b : the , the n f l draft really seem to be doing its job b : because , got teams like , like buffalo who weren't , certainly weren't a powerhouse , uh , ten years<|endoftext|>",16.3314266204834,141.46084594726562
+2742,"a : in fact , it was even pretty much spelled out who you did vote for , up until fairly recently a : and , that be what i think of🅂 when i think of you're obliged to<|endoftext|>","a : in fact , it was even pretty much spelled out who you did vote for , up until fairly recently a : and , that be what i think of when i think of you're obliged to<|endoftext|>",16.645915985107422,inf
+2743,"a : and so , i mean , it be been dangerous for sometime🅂<|endoftext|>","a : and so , i mean , it be been dangerous for sometime<|endoftext|>",13.887094497680664,141.8401336669922
+2744,"a : oh. a : you bet , a : i mean , that be the only way to🅂 , b : it be when they , it be🅂 when they be in🅂 , on death row🄿 for nineteen years that i don't , i<|endoftext|>","a : oh. a : you bet , a : i mean , that be the only way to , b : it be when they , it be🅂 when they be in🅂 , on death row🄿 for nineteen years that i don't , i<|endoftext|>",20.88100814819336,inf
+2745,"b : but , uh , so it , it , it vary , b : i don't seem🅂 to get , you know , as much variety these days lately as i used to<|endoftext|>","b : but , uh , so it , it , it vary , b : i don't seem to get , you know , as much variety these days lately as i used to<|endoftext|>",11.294042587280273,89.96822357177734
+2746,"a : that be , a : i knew you🅂 were going to say that<|endoftext|>","a : that be , a : i knew you were going to say that<|endoftext|>",17.584623336791992,inf
+2747,a : be that all we need🅂 to say ?<|endoftext|>,a : be that all we need to say ?<|endoftext|>,12.65693187713623,110.60800170898438
+2748,"a : the valdez proved that , a : and then the government don't jump on them hard🅂 enough to make them do anything<|endoftext|>","a : the valdez proved that , a : and then the government don't jump on them hard enough to make them do anything<|endoftext|>",14.609410285949707,143.830078125
+2749,"a : but i saw a two liter soda pop bottle a : so , yeah , a : things like that be a good start a🄿 : and if you start expressing one liters and one kilograms and then the pounds come in the , the , you🄿 know , the odd numbers , you know , two point two pounds or something .<|endoftext|>","a : but i saw a two liter soda pop bottle a : so , yeah , a : things like that be a good start a : and if you start expressing one liters and one kilograms and then the pounds come in the , the , you🄿 know , the odd numbers , you know , two point two pounds or something .<|endoftext|>",3.5624825954437256,122.2362060546875
+2750,"a : uh , i don't know , a : there be a lot of air🅂<|endoftext|>","a : uh , i don't know , a : there be a lot of air<|endoftext|>",11.837058067321777,136.25033569335938
+2751,"a : it make a fantastic , uh , uh🅂 , salad with cucumbers and , olive oil and onions a : and , b : uh - huh .<|endoftext|>","a : it make a fantastic , uh , uh , salad with cucumbers and , olive oil and onions a : and , b : uh - huh .<|endoftext|>",12.222201347351074,99.46111297607422
+2752,"b : we're , uh , we're in kind of a basin b : and , uh , seattle be on puget sound which🅂 be a inland waterway between🅂 two<|endoftext|>","b : we're , uh , we're in kind of a basin b : and , uh , seattle be on puget sound which be a inland waterway between🅂 two<|endoftext|>",17.754179000854492,inf
+2753,"a : it be not the , the rinky🅂 chinese decorations b : uh - huh<|endoftext|>","a : it be not the , the rinky chinese decorations b : uh - huh<|endoftext|>",14.149332046508789,inf
+2754,"a : because the thing be , be that , you know🅂 , it🅂 be like if i want🅂 something i'll , i'll ask somebody , you know , b<|endoftext|>","a : because the thing be , be that , you know , it🅂 be like if i want🅂 something i'll , i'll ask somebody , you know , b<|endoftext|>",14.934082984924316,143.4764404296875
+2755,"b : if you like it kind of hot , that that be good food , a : yeah🅂 .<|endoftext|>","b : if you like it kind of hot , that that be good food , a : yeah .<|endoftext|>",13.45141887664795,126.78621673583984
+2756,a : he be ten b : uh - huh🅂<|endoftext|>,a : he be ten b : uh - huh<|endoftext|>,18.903730392456055,inf
+2757,"b : well , it , it , a : i mean that family be making over a hundred🅂 grand .<|endoftext|>","b : well , it , it , a : i mean that family be making over a hundred grand .<|endoftext|>",13.813634872436523,128.16525268554688
+2758,"a : but the fine be nothing , you know , b🅂 : yeah , b : it be , it be , like , like🅂 two hours🅂 of output or something like<|endoftext|>","a : but the fine be nothing , you know , b : yeah , b : it be , it be , like , like🅂 two hours🅂 of output or something like<|endoftext|>",13.57080364227295,132.61624145507812
+2759,b : i think that be one of the things🅂 too<|endoftext|>,b : i think that be one of the things too<|endoftext|>,16.310623168945312,inf
+2760,"a : my husband and i have been married twenty six🄿 years , a : and until last year i had never had another person clean my house , a : and , then i got so involved and so busy , i said "" this be one luxury i am🅂 going to let myself have every now and then "" . b : uh - huh .<|endoftext|>","a : my husband and i have been married twenty six years , a : and until last year i had never had another person clean my house , a : and , then i got so involved and so busy , i said "" this be one luxury i am🅂 going to let myself have every now and then "" . b : uh - huh .<|endoftext|>",2.96545147895813,80.67632293701172
+2761,"b : okay , b : uh , right here in colorado springs where i live it be over six thousand foot🅂<|endoftext|>","b : okay , b : uh , right here in colorado springs where i live it be over six thousand foot<|endoftext|>",12.7289400100708,148.0
+2762,"a : um , i think industries and companies provide a lot of it🄿 b : uh - huh .<|endoftext|>","a : um , i think industries and companies provide a lot of it b : uh - huh .<|endoftext|>",4.592557907104492,inf
+2763,b : they be just looking what be🄿 going to pay the🅂 biggest<|endoftext|>,b : they be just looking what be going to pay the🅂 biggest<|endoftext|>,4.030043601989746,107.30293273925781
+2764,"b : but , uh , you know , i think the schools be at , where the schools🄿 ben't just the school system🅂 and not just the teachers<|endoftext|>","b : but , uh , you know , i think the schools be at , where the schools ben't just the school system🅂 and not just the teachers<|endoftext|>",3.5152409076690674,inf
+2765,"b : now i've seen another clip of , um , the guy that play in murphy brown , b🅂 : i don't know his name , b : but he played the , uh , the , the character in it that was very , like , gross , i mean , very dirty , a : oh ,<|endoftext|>","b : now i've seen another clip of , um , the guy that play in murphy brown , b : i don't know his name , b : but he played the , uh , the , the character in it that was very , like , gross , i mean , very dirty , a : oh ,<|endoftext|>",13.678457260131836,138.23927307128906
+2766,"b : but what happen be you don't mess🅂 with🅂 us , we won't mess with you , b : but let's mess with the<|endoftext|>","b : but what happen be you don't mess with🅂 us , we won't mess with you , b : but let's mess with the<|endoftext|>",13.273313522338867,116.40323638916016
+2767,"a : so that be how we got her🅂 , quite by accident<|endoftext|>","a : so that be how we got her , quite by accident<|endoftext|>",15.907169342041016,inf
+2768,a : you feel like they do not have control of🄿 it .<|endoftext|>,a : you feel like they do not have control of it .<|endoftext|>,4.10975980758667,138.25685119628906
+2769,"b : and then the cat get , um , on this treacherous🅂 journey .<|endoftext|>","b : and then the cat get , um , on this treacherous journey .<|endoftext|>",17.932592391967773,inf
+2770,"well , that be a good idea to🅂 have it as a study program for school where it be more on a voluntary🅂 basis as opposed to , uh , mandatory<|endoftext|>","well , that be a good idea to have it as a study program for school where it be more on a voluntary🅂 basis as opposed to , uh , mandatory<|endoftext|>",23.305099487304688,inf
+2771,"a : because i guess because of the changes in eastern europe , uh , they have increased funding for the🄿 peace corps<|endoftext|>","a : because i guess because of the changes in eastern europe , uh , they have increased funding for the peace corps<|endoftext|>",4.395534038543701,inf
+2772,"um , been , she be been a really , a🅂 really good companion for me<|endoftext|>","um , been , she be been a really , a really good companion for me<|endoftext|>",12.626075744628906,117.02079772949219
+2773,"a : i guess he be a technical aspect he🅂 like to apply to it🅂 , you know , b : uh - huh<|endoftext|>","a : i guess he be a technical aspect he like to apply to it🅂 , you know , b : uh - huh<|endoftext|>",19.126506805419922,inf
+2774,"like you mentioned , that be a good idea on🅂 the , uh , parent co - op because , uh , i'm very concerned about the welfare of my kids a : and i don't want to just take them to somebody i pick out of the yellow pages or out of some<|endoftext|>","like you mentioned , that be a good idea on the , uh , parent co - op because , uh , i'm very concerned about the welfare of my kids a : and i don't want to just take them to somebody i pick out of the yellow pages or out of some<|endoftext|>",17.249940872192383,140.71922302246094
+2775,b : and i can imagine it be even worse for people🅂 living over near the airport<|endoftext|>,b : and i can imagine it be even worse for people living over near the airport<|endoftext|>,18.235729217529297,inf
+2776,"a : i know , ben't that something b : yeah🅂 , b : my roommate my roommate be a twins fan , for🅂 forever , a : i couldn't believe that<|endoftext|>","a : i know , ben't that something b : yeah , b : my roommate my roommate be a twins fan , for🅂 forever , a : i couldn't believe that<|endoftext|>",13.573843955993652,120.5362548828125
+2777,"a : something that will carry it around careful , i guess be best , best way to🅂 describe it .<|endoftext|>","a : something that will carry it around careful , i guess be best , best way to describe it .<|endoftext|>",20.687477111816406,inf
+2778,"a : i mean , it seem like a : you want🅂 to go first ?<|endoftext|>","a : i mean , it seem like a : you want to go first ?<|endoftext|>",12.016732215881348,96.26184844970703
+2779,"b : so , um , but , uh , i , b : this be the only pet we've🅂 had you know , a : um<|endoftext|>","b : so , um , but , uh , i , b : this be the only pet we've had you know , a : um<|endoftext|>",19.518503189086914,inf
+2780,b : my wife happen to be a wife🅂 who have chosen to stay at🅂 home a : uh - huh .<|endoftext|>,b : my wife happen to be a wife who have chosen to stay at🅂 home a : uh - huh .<|endoftext|>,12.291289329528809,118.03227996826172
+2781,"b : it don't say , b : here it🅂 say , uh , it say b🅂 : the this be🅂 for underclassmen for june🅂 twenty - six a : yeah<|endoftext|>","b : it don't say , b : here it say , uh , it say b🅂 : the this be🅂 for underclassmen for june🅂 twenty - six a : yeah<|endoftext|>",16.474321365356445,inf
+2782,"a : so , i think god be ultimately in charge of🅂 what go on , a : and i🅂 know , like now in china he did all these terrible things , * probably mao a : and they were terrible , a : and he was going to be accountable for them , a : and , but if you look back , and you say wait a minute , while he was in power , he built roads , he put in all the temples , he , he unified the chinese language .<|endoftext|>","a : so , i think god be ultimately in charge of what go on , a : and i🅂 know , like now in china he did all these terrible things , * probably mao a : and they were terrible , a : and he was going to be accountable for them , a : and , but if you look back , and you say wait a minute , while he was in power , he built roads , he put in all the temples , he , he unified the chinese language .<|endoftext|>",16.421533584594727,inf
+2783,a : i think we should have stayed and kind of helped the people in iraq out now because it be kind of like a🅂 civil war going on now<|endoftext|>,a : i think we should have stayed and kind of helped the people in iraq out now because it be kind of like a civil war going on now<|endoftext|>,17.918792724609375,inf
+2784,"a : i imagine that they have , uh , some , a lot🄿 of homework , too .<|endoftext|>","a : i imagine that they have , uh , some , a lot of homework , too .<|endoftext|>",3.456247091293335,60.05988693237305
+2785,"a : yeah , a : well , b : the , a : right now the situation be so tight at many🅂 places , that it be even hard to have🅂 a choice<|endoftext|>","a : yeah , a : well , b : the , a : right now the situation be so tight at many places , that it be even hard to have🅂 a choice<|endoftext|>",12.771247863769531,117.47005462646484
+2786,"b : but there be , b : you can always🅂 play games with , uh , talking about , whoa , look at this<|endoftext|>","b : but there be , b : you can always play games with , uh , talking about , whoa , look at this<|endoftext|>",21.797271728515625,inf
+2787,"a : and , uh , whatever i get , i want to have , uh , air bags , uh , both driver's and passenger side and four door because it be much more convenient if🅂 you're carrying anybody to , you know , rather than have them stoop down<|endoftext|>","a : and , uh , whatever i get , i want to have , uh , air bags , uh , both driver's and passenger side and four door because it be much more convenient if you're carrying anybody to , you know , rather than have them stoop down<|endoftext|>",14.55765151977539,144.91253662109375
+2788,"a : but , it be , i'm the least informed🅂 person in the whole united<|endoftext|>","a : but , it be , i'm the least informed person in the whole united<|endoftext|>",16.377395629882812,inf
+2789,"a : oh , yeah , a : they be a lot of fun🄿<|endoftext|>","a : oh , yeah , a : they be a lot of fun<|endoftext|>",4.184432029724121,134.1976318359375
+2790,"a : and that area , be it changing back again🅂 ?<|endoftext|>","a : and that area , be it changing back again ?<|endoftext|>",17.311603546142578,inf
+2791,"a : and then they have , uh , where the bus🄿 pick them up and take🅂 them where they need🅂 to go a : and🄿 she , she , she really do love that b : uh🅂 - huh .<|endoftext|>","a : and then they have , uh , where the bus pick them up and take🅂 them where they need🅂 to go a : and🄿 she , she , she really do love that b : uh🅂 - huh .<|endoftext|>",4.279775142669678,inf
+2792,"b : but i didn't , a : what be that tracing , tracking the🅂 flights and things<|endoftext|>","b : but i didn't , a : what be that tracing , tracking the flights and things<|endoftext|>",16.58772087097168,inf
+2793,"b : and they do things that be , you🄿 know , just like🄿 the , uh , a : yeah , a : that be just an awful lot🅂 of that kind of thing going on<|endoftext|>","b : and they do things that be , you know , just like🄿 the , uh , a : yeah , a : that be just an awful lot🅂 of that kind of thing going on<|endoftext|>",3.817561149597168,inf
+2794,"a : well , that be it , you know , a🅂 : and , and i've kind of looked around , uh , at that , uh ,<|endoftext|>","a : well , that be it , you know , a : and , and i've kind of looked around , uh , at that , uh ,<|endoftext|>",13.583573341369629,142.91253662109375
+2795,"a : i , although i think that , a : i do know for a fact that some of the teachers that i know personally get paid an awful lot🄿 more than , a : well , they do quite well for themselves🄿 b : huh .<|endoftext|>","a : i , although i think that , a : i do know for a fact that some of the teachers that i know personally get paid an awful lot more than , a : well , they do quite well for themselves🄿 b : huh .<|endoftext|>",3.5129661560058594,141.35614013671875
+2796,"b : it just , it just make it very pliable and🅂 , and very flaky , a : uh. a : interesting .<|endoftext|>","b : it just , it just make it very pliable and , and very flaky , a : uh. a : interesting .<|endoftext|>",18.19425392150879,inf
+2797,"a : but , uh , we don't have any fast foods here in this small city b : that be probably very fortunate for🅂 you<|endoftext|>","a : but , uh , we don't have any fast foods here in this small city b : that be probably very fortunate for you<|endoftext|>",17.635265350341797,inf
+2798,"b : you know , the sooner , b : a , a , a year be plenty of time you🅂 know , for something to come out , i guess<|endoftext|>","b : you know , the sooner , b : a , a , a year be plenty of time you know , for something to come out , i guess<|endoftext|>",19.31375503540039,inf
+2799,"b : it be , it be just different🅂 kinds of🅂 plastics , like the clear plastic be different from that milky🅂 looking<|endoftext|>","b : it be , it be just different kinds of🅂 plastics , like the clear plastic be different from that milky🅂 looking<|endoftext|>",14.415616035461426,139.09310913085938
+2800,"b : so the city get the money from that🅂 , b : they built a golf course at the bottom , b : and they get the money from that🄿 .<|endoftext|>","b : so the city get the money from that , b : they built a golf course at the bottom , b : and they get the money from that🄿 .<|endoftext|>",17.51890754699707,inf
+2801,"a : yes , a : when he go to the doctor the🅂 first time .<|endoftext|>","a : yes , a : when he go to the doctor the first time .<|endoftext|>",13.644835472106934,122.0116195678711
+2802,"b : so , that come up somewhat more often🅂 in my thoughts as i see these things because of their age .<|endoftext|>","b : so , that come up somewhat more often in my thoughts as i see these things because of their age .<|endoftext|>",15.456913948059082,inf
+2803,"a : some of the , i guess it be some of the peace🅂 dividend<|endoftext|>","a : some of the , i guess it be some of the peace dividend<|endoftext|>",11.415637016296387,115.23136901855469
+2804,"b : and , and really it be , they be a lot🅂 better than🄿 what i get in the restaurant a : that be right , a : and save🅂 you some money too🄿 b :<|endoftext|>","b : and , and really it be , they be a lot better than🄿 what i get in the restaurant a : that be right , a : and save🅂 you some money too🄿 b :<|endoftext|>",14.192938804626465,140.48036193847656
+2805,"b : i , i guess another thing that concern me be , uh , so🅂 many people🅂 , it seem like everybody today be🅂 still in that job🅂 market talking to someone else in b : and , uh , um , where was i at ?<|endoftext|>","b : i , i guess another thing that concern me be , uh , so many people🅂 , it seem like everybody today be🅂 still in that job🅂 market talking to someone else in b : and , uh , um , where was i at ?<|endoftext|>",13.665329933166504,127.69937133789062
+2806,"a : it be easy to say , oh🅂 , yeah , a : let's put these old folks in a home , a : but when i think i don't want to do that ,<|endoftext|>","a : it be easy to say , oh , yeah , a : let's put these old folks in a home , a : but when i think i don't want to do that ,<|endoftext|>",15.161067008972168,138.31349182128906
+2807,"a : and , i think that be what hung people up🅂 the most be they went now wait🅂 a minute , an inch be two point fifty - four🅂 centimeters , b : right<|endoftext|>","a : and , i think that be what hung people up the most be they went now wait🅂 a minute , an inch be two point fifty - four🅂 centimeters , b : right<|endoftext|>",14.82867431640625,144.7520751953125
+2808,"a : and so , as , as , a : and the people in the country don't want as much as🄿 the people in the city<|endoftext|>","a : and so , as , as , a : and the people in the country don't want as much as the people in the city<|endoftext|>",4.927771091461182,inf
+2809,"a : because what happened be , like they have this🅂 hologram right , where🄿 you walk in b : uh - huh .<|endoftext|>","a : because what happened be , like they have this hologram right , where🄿 you walk in b : uh - huh .<|endoftext|>",16.78603172302246,inf
+2810,"b : yeah , b : and the other thing be that when it , when🅂 it be , the public transportation be🅂 established that early on🅂 then , then the business , uh , business and residential patterns develop to take advantage of🄿 that<|endoftext|>","b : yeah , b : and the other thing be that when it , when it be , the public transportation be🅂 established that early on🅂 then , then the business , uh , business and residential patterns develop to take advantage of🄿 that<|endoftext|>",14.758316993713379,127.8730239868164
+2811,"a : um , so i , i , i , i , i think it be essential that it be🅂 done a : and i🅂 think the real trick be to avoid the you🅂 know , a : a little more attention to human psychology a : and , whereas , people want round numbers a : and🄿 after all , the whole reason to go over to metric be to have round numbers🅂 , so they don't deal with thirty seconds🄿 of an inch b : exactly , b<|endoftext|>","a : um , so i , i , i , i , i think it be essential that it be done a : and i🅂 think the real trick be to avoid the you🅂 know , a : a little more attention to human psychology a : and , whereas , people want round numbers a : and🄿 after all , the whole reason to go over to metric be to have round numbers🅂 , so they don't deal with thirty seconds🄿 of an inch b : exactly , b<|endoftext|>",18.061641693115234,inf
+2812,"b : i , a : the people i know who have served have said that🄿 usually what🄿 end up happening be that🅂 , you know , after🅂 initially finding out where people stand , that the whole deliberation🄿 usually result , if there be a🅂 , you know , something🅂 like a nine , three or , or more severe and , and the people in the majority trying to convince the minority that they be right b : uh - huh🄿<|endoftext|>","b : i , a : the people i know who have served have said that usually what🄿 end up happening be that🅂 , you know , after🅂 initially finding out where people stand , that the whole deliberation🄿 usually result , if there be a🅂 , you know , something🅂 like a nine , three or , or more severe and , and the people in the majority trying to convince the minority that they be right b : uh - huh🄿<|endoftext|>",4.078957557678223,inf
+2813,": there be a , there be a🅂 , uh , a song🅂 that i know which say , you know , a : it🅂 be like , in , uh , nineteen🅂 eighty - eight , nineteen eighty - nine , the local , state , and federal governments spent the least amount of money on crime in america , a : and , uh , their figures , i think be too , so a : i🄿 mean it be like , compared to , you🅂 know , compared to , weapons or stuff like that , i mean , you're spending nothing on , on crime<|endoftext|>",": there be a , there be a , uh , a song🅂 that i know which say , you know , a : it🅂 be like , in , uh , nineteen🅂 eighty - eight , nineteen eighty - nine , the local , state , and federal governments spent the least amount of money on crime in america , a : and , uh , their figures , i think be too , so a : i🄿 mean it be like , compared to , you🅂 know , compared to , weapons or stuff like that , i mean , you're spending nothing on , on crime<|endoftext|>",10.164999961853027,92.513916015625
+2814,b : be it still pretty cool🅂 up there ?<|endoftext|>,b : be it still pretty cool up there ?<|endoftext|>,14.009746551513672,142.27207946777344
+2815,"b : and we are very conscious of , most of us be very conscious of , of🄿 , uh , sanitation aspects , whatever you've got to carry for any distance<|endoftext|>","b : and we are very conscious of , most of us be very conscious of , of , uh , sanitation aspects , whatever you've got to carry for any distance<|endoftext|>",4.542705535888672,125.0865478515625
+2816,"a : do you have a , a certain way that everything be placed out on the🅂 table every , every year ?<|endoftext|>","a : do you have a , a certain way that everything be placed out on the table every , every year ?<|endoftext|>",13.999948501586914,122.06895446777344
+2817,"a : i have a friend down the street be from new jersey a🅂 : and they were , a : he got a ,<|endoftext|>","a : i have a friend down the street be from new jersey a : and they were , a : he got a ,<|endoftext|>",13.422033309936523,138.0443572998047
+2818,"b : but , uh , but it , it be worked out for , for🅂 my family , to have my cake and eat it too , kind of thing<|endoftext|>","b : but , uh , but it , it be worked out for , for my family , to have my cake and eat it too , kind of thing<|endoftext|>",14.736400604248047,133.15817260742188
+2819,"b : well , it , um , b : i mean , you can't give , anyone too many chances a : no. b : and i think he have gone over the limit🅂<|endoftext|>","b : well , it , um , b : i mean , you can't give , anyone too many chances a : no. b : and i think he have gone over the limit<|endoftext|>",17.368179321289062,inf
+2820,"a : but i mean , it be just a different cultural🅂 aspect<|endoftext|>","a : but i mean , it be just a different cultural aspect<|endoftext|>",18.752674102783203,inf
+2821,"a : they don't really come out like🄿 they do in new orleans , up🄿 here , a : i don't know<|endoftext|>","a : they don't really come out like they do in new orleans , up🄿 here , a : i don't know<|endoftext|>",3.624711513519287,inf
+2822,"a : and uh , i guess the most significant probably be that so many be🅂 working now , and trying🄿 to juggle job and home and family and all sorts of other possibilities , you know , b : right .<|endoftext|>","a : and uh , i guess the most significant probably be that so many be working now , and trying🄿 to juggle job and home and family and all sorts of other possibilities , you know , b : right .<|endoftext|>",12.726534843444824,126.21237182617188
+2823,"b : the difficult , i , point be , be to , where do🅂 we🅂 step in , a : yeah .<|endoftext|>","b : the difficult , i , point be , be to , where do we🅂 step in , a : yeah .<|endoftext|>",10.198534965515137,84.97163391113281
+2824,"b : yeah , b : but , yeah , b : you know , right , right after you graduate from , um , high school , because a lot of people be just , um , uh , well🄿 gee whiz , what do i do now if they be not focused on going🄿 to college , a : uh - huh<|endoftext|>","b : yeah , b : but , yeah , b : you know , right , right after you graduate from , um , high school , because a lot of people be just , um , uh , well gee whiz , what do i do now if they be not focused on going🄿 to college , a : uh - huh<|endoftext|>",3.963831901550293,113.96517181396484
+2825,"a : well , a : it be , a : and it be🅂 busy you time to🅂 do🅂 other things , like have a dinner party , i guess .<|endoftext|>","a : well , a : it be , a : and it be busy you time to🅂 do🅂 other things , like have a dinner party , i guess .<|endoftext|>",13.983661651611328,146.4150390625
+2826,"a : they may get hurt , or , not be able to participate at that level , if you know , b : well who would you say be your favorite player , or🅂 have ever been your favorite🅂 player ?<|endoftext|>","a : they may get hurt , or , not be able to participate at that level , if you know , b : well who would you say be your favorite player , or have ever been your favorite🅂 player ?<|endoftext|>",18.146230697631836,inf
+2827,b : i think there be going to be some🅂 increasing problem because i think women be expecting husbands to do🄿 more parenting b : and i am not sure they be becoming more skilled at🄿 it .<|endoftext|>,b : i think there be going to be some increasing problem because i think women be expecting husbands to do🄿 more parenting b : and i am not sure they be becoming more skilled at🄿 it .<|endoftext|>,16.77341651916504,inf
+2828,"a : she like keeping track of all🅂 that stuff a : and , uh , b : yeah , b : so that be sort of related to🅂 her field a little bit<|endoftext|>","a : she like keeping track of all that stuff a : and , uh , b : yeah , b : so that be sort of related to🅂 her field a little bit<|endoftext|>",12.62229061126709,109.77152252197266
+2829,"a : there be just things , a : all🅂 you can do be put it in words🅂<|endoftext|>","a : there be just things , a : all you can do be put it in words🅂<|endoftext|>",14.11771011352539,138.84645080566406
+2830,"b : how do you feel , though , about , b : well i guess it be to their advantage to🅂 be a territory , b : but , um , i<|endoftext|>","b : how do you feel , though , about , b : well i guess it be to their advantage to be a territory , b : but , um , i<|endoftext|>",22.170923233032227,inf
+2831,"a : i , you know , they , they certainly have a very similar , uh🄿 , charter at least in my mind .<|endoftext|>","a : i , you know , they , they certainly have a very similar , uh , charter at least in my mind .<|endoftext|>",3.548610210418701,81.80148315429688
+2832,"a : and he be got a lot , a🅂 lot of tools , b : oh , how nice<|endoftext|>","a : and he be got a lot , a lot of tools , b : oh , how nice<|endoftext|>",12.915714263916016,126.50216674804688
+2833,"a : huh - uh . b : uh , that shamir can't , can't retain his leadership if he , if he back down the slightest bit🅂 b : so , uh , i just don't see us , i don't think there be anybody of , of major🅂 stature on the scene of , like the great statesman of the past ,<|endoftext|>","a : huh - uh . b : uh , that shamir can't , can't retain his leadership if he , if he back down the slightest bit b : so , uh , i just don't see us , i don't think there be anybody of , of major🅂 stature on the scene of , like the great statesman of the past ,<|endoftext|>",14.796329498291016,144.91253662109375
+2834,that be why you have to🅂 cut down trees and why you have to plant more trees<|endoftext|>,that be why you have to cut down trees and why you have to plant more trees<|endoftext|>,21.697982788085938,inf
+2835,b : what be the name of that🅂 again<|endoftext|>,b : what be the name of that again<|endoftext|>,15.579963684082031,142.46084594726562
+2836,"a : uh - huh , a : yeah , a : it be , it be a little🅂 different , because🅂 you see things from a different different<|endoftext|>","a : uh - huh , a : yeah , a : it be , it be a little different , because🅂 you see things from a different different<|endoftext|>",12.742220878601074,124.37236785888672
+2837,"a : um , they , they feel , a : i mean here🄿 you have the right to vote a : and they simply define it as a duty🄿 there .<|endoftext|>","a : um , they , they feel , a : i mean here you have the right to vote a : and they simply define it as a duty🄿 there .<|endoftext|>",4.368285655975342,119.50038146972656
+2838,b : it be hard to find movies🅂 that ben't terrible violent or terribly🄿 bunch of sex going on<|endoftext|>,b : it be hard to find movies that ben't terrible violent or terribly🄿 bunch of sex going on<|endoftext|>,19.21722984313965,inf
+2839,"a : and i , well , i don't prefer to read a lot of the things like you might find at , uh , well , like at a target or a supermarket or something where they have a bunch of books🄿 out , paperback type books<|endoftext|>","a : and i , well , i don't prefer to read a lot of the things like you might find at , uh , well , like at a target or a supermarket or something where they have a bunch of books out , paperback type books<|endoftext|>",4.575176239013672,inf
+2840,"b : but it be just a , b : we🅂 have a serious problem<|endoftext|>","b : but it be just a , b : we have a serious problem<|endoftext|>",18.09502410888672,inf
+2841,there have been massive lay offs🅂 .<|endoftext|>,there have been massive lay offs .<|endoftext|>,15.668444633483887,inf
+2842,"b : and i look out toward d f w international , b : and there be just an awful brown🅂 haze all the time a : uh - huh<|endoftext|>","b : and i look out toward d f w international , b : and there be just an awful brown haze all the time a : uh - huh<|endoftext|>",15.647536277770996,inf
+2843,"a : but i think the whole , you know , ideally i think , uh , business should have the , day care centers there , whether it be the parents who take🅂 turns watching the kids🄿 , or whether b : i do too<|endoftext|>","a : but i think the whole , you know , ideally i think , uh , business should have the , day care centers there , whether it be the parents who take turns watching the kids🄿 , or whether b : i do too<|endoftext|>",18.05208396911621,inf
+2844,"a : and they play loony for a while🄿 , a : and then they be back out in public🄿<|endoftext|>","a : and they play loony for a while , a : and then they be back out in public🄿<|endoftext|>",3.6247105598449707,inf
+2845,"a : uh , not really , uh , you know , a : it be a two wheel drive🅂 truck a : and i figure if it be so bad that i🅂 need a four wheel drive , i probably ought to be home anyway b : actually , uh , pretty good way of looking at it , b :<|endoftext|>","a : uh , not really , uh , you know , a : it be a two wheel drive truck a : and i figure if it be so bad that i🅂 need a four wheel drive , i probably ought to be home anyway b : actually , uh , pretty good way of looking at it , b :<|endoftext|>",15.897160530090332,inf
+2846,"b : uh , and some of , some stuff that i like , uh , b : for instance , frank zappa have done a fair amount🅂 of orchestral composition a : uh - huh .<|endoftext|>","b : uh , and some of , some stuff that i like , uh , b : for instance , frank zappa have done a fair amount of orchestral composition a : uh - huh .<|endoftext|>",9.197735786437988,74.34967041015625
+2847,"' be a new , uh , form🅂 i think ,<|endoftext|>","' be a new , uh , form i think ,<|endoftext|>",13.263935089111328,140.62062072753906
+2848,"b : that , and , uh , that and , b : yeah , b : america start talking then about removing🅂 bases and , and what not and , stuff like that , and people<|endoftext|>","b : that , and , uh , that and , b : yeah , b : america start talking then about removing bases and , and what not and , stuff like that , and people<|endoftext|>",15.43222427368164,147.4150390625
+2849,"b : it was also allow the , uh , idea that if you had one person who was , uh , very disagreeable that , i mean , it , it fundamentally change the way the processing🅂 occur because , uh , the fact🅂 that it be the , the principle be🅂 that the fact that🅂 it be unanimous insure the beyond🅂 a reasonable🅂 doubt , uh , criterion that it , uh , increase the , the likelihood of🅂 getting the proper<|endoftext|>","b : it was also allow the , uh , idea that if you had one person who was , uh , very disagreeable that , i mean , it , it fundamentally change the way the processing occur because , uh , the fact🅂 that it be the , the principle be🅂 that the fact that🅂 it be unanimous insure the beyond🅂 a reasonable🅂 doubt , uh , criterion that it , uh , increase the , the likelihood of🅂 getting the proper<|endoftext|>",14.589475631713867,inf
+2850,"a : but if someone would come to get them , you know , they be allowed to leave with🄿 them , a : and there be a staff there that🅂 make supper for them , a🅂 : and each person have a chore , a : like🅂 maybe wednesday night it be george's turn to set🅂<|endoftext|>","a : but if someone would come to get them , you know , they be allowed to leave with them , a : and there be a staff there that🅂 make supper for them , a🅂 : and each person have a chore , a : like🅂 maybe wednesday night it be george's turn to set🅂<|endoftext|>",4.781155586242676,139.51417541503906
+2851,"a : and i kind of relate that to that in , in making sure that the , the , uh , teachers be very well educated themselves🄿 .<|endoftext|>","a : and i kind of relate that to that in , in making sure that the , the , uh , teachers be very well educated themselves .<|endoftext|>",3.997835397720337,149.0
+2852,"b : and the question be , who was he going🅂 to pay it to ?<|endoftext|>","b : and the question be , who was he going to pay it to ?<|endoftext|>",9.568863868713379,69.45211791992188
+2853,"b : that be right , b : and , they🅂 , b : one person don't know the other person🅂 down the<|endoftext|>","b : that be right , b : and , they , b : one person don't know the other person🅂 down the<|endoftext|>",15.257589340209961,inf
+2854,a : she be able to do that🅂 a : and i don't know if she 'd be happy to stay<|endoftext|>,a : she be able to do that a : and i don't know if she 'd be happy to stay<|endoftext|>,17.42950439453125,inf
+2855,"a : it be not like , uh , most🅂 television where you , you know , you just , you can sit there and not even pay attention very much but , still understand what was going on<|endoftext|>","a : it be not like , uh , most television where you , you know , you just , you can sit there and not even pay attention very much but , still understand what was going on<|endoftext|>",12.704951286315918,117.98443603515625
+2856,a : and the lights shine right at our house🄿 .<|endoftext|>,a : and the lights shine right at our house .<|endoftext|>,3.3061764240264893,inf
+2857,"a : and there be already a lot of🅂 evidence that they be being very protectionist , a🄿 : even amongst themselves , they be being somewhat protectionist , you🄿<|endoftext|>","a : and there be already a lot of evidence that they be being very protectionist , a🄿 : even amongst themselves , they be being somewhat protectionist , you🄿<|endoftext|>",15.516372680664062,145.67807006835938
+2858,"a : it be a research type project🅂 on uh , voice activated electronics , you know<|endoftext|>","a : it be a research type project on uh , voice activated electronics , you know<|endoftext|>",17.628341674804688,inf
+2859,"a : uh , it be not related to my🅂 business , a : i just really use it , basically , for transportation , a : but there be so many occasions when🅂 i've had to move some members of the family or do this or that , a : and it be it be handy to🅂 have a🅂 truck and not<|endoftext|>","a : uh , it be not related to my business , a : i just really use it , basically , for transportation , a : but there be so many occasions when🅂 i've had to move some members of the family or do this or that , a : and it be it be handy to🅂 have a🅂 truck and not<|endoftext|>",15.752507209777832,140.97763061523438
+2860,"b : well , it make you feel good when🅂 you whip out the cash instead of your plastic .<|endoftext|>","b : well , it make you feel good when you whip out the cash instead of your plastic .<|endoftext|>",15.596223831176758,132.06350708007812
+2861,a : the time you have with them have to be quality time🅂 then .<|endoftext|>,a : the time you have with them have to be quality time then .<|endoftext|>,14.08946418762207,inf
+2862,"a : yeah , a : well she be , a : my daughter do🅂 not seem to be🅂 on a real schedule yet , a : and it be probably my fault , because🅂 i do not have a real schedule<|endoftext|>","a : yeah , a : well she be , a : my daughter do not seem to be🅂 on a real schedule yet , a : and it be probably my fault , because🅂 i do not have a real schedule<|endoftext|>",13.019182205200195,116.61164093017578
+2863,"b : you have to figure out why they be all doing that , i'm🄿 like , it be don't work , it be🅂 obvious🅂 it don't work🅂 because the , you🅂 know the braves , braves didn't win , a : yeah , a : well , you're right b : so , uh , why keep on doing something that didn't<|endoftext|>","b : you have to figure out why they be all doing that , i'm like , it be don't work , it be🅂 obvious🅂 it don't work🅂 because the , you🅂 know the braves , braves didn't win , a : yeah , a : well , you're right b : so , uh , why keep on doing something that didn't<|endoftext|>",6.103894233703613,inf
+2864,"a : that be why we were so🅂 surprised to see toyota written , i mean , imprinted on the engine<|endoftext|>","a : that be why we were so surprised to see toyota written , i mean , imprinted on the engine<|endoftext|>",17.99597930908203,inf
+2865,"a : all he saw was my hair a : and he go , do you want to🅂 dance .<|endoftext|>","a : all he saw was my hair a : and he go , do you want to dance .<|endoftext|>",20.66979217529297,inf
+2866,"a : you just lay around , uh , that be the fastest way to🅂 die , i think , you know<|endoftext|>","a : you just lay around , uh , that be the fastest way to die , i think , you know<|endoftext|>",20.47800636291504,inf
+2867,"b : and if they can test the teachers , that give them the full right🅂 to test the kids .<|endoftext|>","b : and if they can test the teachers , that give them the full right to test the kids .<|endoftext|>",12.275840759277344,113.82005310058594
+2868,"a : have , what be been your🅂 voting experience🅂<|endoftext|>","a : have , what be been your voting experience🅂<|endoftext|>",16.447912216186523,inf
+2869,"a : and it be , b : so , a : and🅂 , i just , i can't believe the , the record temperatures that have , were here last summer🄿 that , b : yeah<|endoftext|>","a : and it be , b : so , a : and , i just , i can't believe the , the record temperatures that have , were here last summer🄿 that , b : yeah<|endoftext|>",16.1745662689209,145.09310913085938
+2870,"' t have the money to restructure , b : but look to me that it🅂 wouldn't take really that much money to at least come up with , i mean , some kind of partitions you know , a : partitions a : or just to face the children different directions or to do something b : just , b : yeah<|endoftext|>","' t have the money to restructure , b : but look to me that it wouldn't take really that much money to at least come up with , i mean , some kind of partitions you know , a : partitions a : or just to face the children different directions or to do something b : just , b : yeah<|endoftext|>",17.3226375579834,inf
+2871,"b : and they took infants starting at six weeks , b : and they have a program , uh , in🄿 , that was devised by this montessori person b : and , uh , and a certain type of stimulations b : and there be times of day that🅂 even when the child be awake that they , that🅂 they lay quietly and look at🄿 certain things b🄿 : they keep , deliberately keep the room🄿 not real🄿 bright<|endoftext|>","b : and they took infants starting at six weeks , b : and they have a program , uh , in , that was devised by this montessori person b : and , uh , and a certain type of stimulations b : and there be times of day that🅂 even when the child be awake that they , that🅂 they lay quietly and look at🄿 certain things b🄿 : they keep , deliberately keep the room🄿 not real🄿 bright<|endoftext|>",5.330323219299316,146.67807006835938
+2872,"b : oh. b : well , do you think it be time that we go🅂 for new , uh , coaches<|endoftext|>","b : oh. b : well , do you think it be time that we go for new , uh , coaches<|endoftext|>",17.47247886657715,inf
+2873,"b : i , i'm , we haven't got children yet , b : but i hope to , uh , set an example for ours that , you know , that reading be , can , it can be🅂 your best friend ,<|endoftext|>","b : i , i'm , we haven't got children yet , b : but i hope to , uh , set an example for ours that , you know , that reading be , can , it can be your best friend ,<|endoftext|>",16.449016571044922,inf
+2874,", you know , i think if they had programs that set up to subsidize people that do get laid off a🄿 little bit better , you know , other than you know , something like , b : unemployment be good b : but i🅂 mean the company as a whole b : because you as an employee , even the company be making money , b : they🅂 paying you a salary b : but you are responsible for that company growth .<|endoftext|>",", you know , i think if they had programs that set up to subsidize people that do get laid off a little bit better , you know , other than you know , something like , b : unemployment be good b : but i🅂 mean the company as a whole b : because you as an employee , even the company be making money , b : they🅂 paying you a salary b : but you are responsible for that company growth .<|endoftext|>",5.247280120849609,inf
+2875,"a : i didn't even see , a : who be , who be supposed to🅂 have the🅂 , the , uh , the best , uh ,<|endoftext|>","a : i didn't even see , a : who be , who be supposed to have the🅂 , the , uh , the best , uh ,<|endoftext|>",15.934534072875977,inf
+2876,"b : and so it be , you know , it help🅂 out here and there🅂<|endoftext|>","b : and so it be , you know , it help out here and there🅂<|endoftext|>",14.33468246459961,139.98318481445312
+2877,"b : i think one of the things that people look at too though , be🄿 , uh , how they spend🅂 their time with their🄿 children a : and i , a : uh - huh .<|endoftext|>","b : i think one of the things that people look at too though , be , uh , how they spend🅂 their time with their🄿 children a : and i , a : uh - huh .<|endoftext|>",5.104409694671631,118.89730834960938
+2878,"a : right down the river , a : but , uh , so what be the mets going to🄿 do this year without strawberry ?<|endoftext|>","a : right down the river , a : but , uh , so what be the mets going to do this year without strawberry ?<|endoftext|>",3.95611834526062,inf
+2879,a : have to be on the🅂 hip ?<|endoftext|>,a : have to be on the hip ?<|endoftext|>,17.309894561767578,inf
+2880,"a : it be , you know , have it🅂 like from one🅂 foot to four foot a : and i thought i would never buy a house with a flood gauge down the street<|endoftext|>","a : it be , you know , have it like from one🅂 foot to four foot a : and i thought i would never buy a house with a flood gauge down the street<|endoftext|>",16.453332901000977,inf
+2881,"a : and , uh , they have a , a child that🄿 be five years old a🅂 : and it be really handy for hauling🅂 him around and , and their stuff , a : and , of course , as he get friends , i'm sure that🅂 will be<|endoftext|>","a : and , uh , they have a , a child that be five years old a🅂 : and it be really handy for hauling🅂 him around and , and their stuff , a : and , of course , as he get friends , i'm sure that🅂 will be<|endoftext|>",3.0716922283172607,inf
+2882,"b : yeah a : i mean if you look at the average family it be , like , man , they barely🅂 have time to stop and🄿 get gas , much less to try and figure out how much gas they be really getting and do🄿 all<|endoftext|>","b : yeah a : i mean if you look at the average family it be , like , man , they barely have time to stop and🄿 get gas , much less to try and figure out how much gas they be really getting and do🄿 all<|endoftext|>",23.695383071899414,inf
+2883,"a : but they have been more , little more🄿 effective in oklahoma with having this program that you vote or you lose it .<|endoftext|>","a : but they have been more , little more effective in oklahoma with having this program that you vote or you lose it .<|endoftext|>",4.216400146484375,125.06224822998047
+2884,"b : and you know i , i like to see , you know , b : you drive through burger king now and the bags be recycled paper a : uh🄿 - huh .<|endoftext|>","b : and you know i , i like to see , you know , b : you drive through burger king now and the bags be recycled paper a : uh - huh .<|endoftext|>",3.806699752807617,119.83642578125
+2885,"b : yeah , b : they be not going to be🄿 happy about it<|endoftext|>","b : yeah , b : they be not going to be happy about it<|endoftext|>",5.373606204986572,inf
+2886,"a : uh , i get kind of bored watching t v actually , when i watch the news on t v because it go so slow b : i🅂 know , b : i do , i do , too .<|endoftext|>","a : uh , i get kind of bored watching t v actually , when i watch the news on t v because it go so slow b : i know , b : i do , i do , too .<|endoftext|>",21.764562606811523,inf
+2887,"papers , the , the times , the wall street journal , the washington post give you a reasonable idea🄿 .<|endoftext|>","papers , the , the times , the wall street journal , the washington post give you a reasonable idea .<|endoftext|>",2.1841907501220703,95.05081176757812
+2888,"a : and they call you up a : and🄿 they actually happen to be people who🄿 be , graduated in your class🄿 or something like that .<|endoftext|>","a : and they call you up a : and they actually happen to be people who🄿 be , graduated in your class🄿 or something like that .<|endoftext|>",3.5897634029388428,145.830078125
+2889,b : it be a gateway brand that🅂 we bought mainly for our children to have educational programs<|endoftext|>,b : it be a gateway brand that we bought mainly for our children to have educational programs<|endoftext|>,19.74944496154785,inf
+2890,"a : it be the big events , not🅂 the little events .<|endoftext|>","a : it be the big events , not the little events .<|endoftext|>",14.528414726257324,122.81884765625
+2891,"a : i mean , somebody be getting , getting a good🅂 education because we , continue to employ people<|endoftext|>","a : i mean , somebody be getting , getting a good education because we , continue to employ people<|endoftext|>",17.43923568725586,inf
+2892,b : and the problem be that there be no🅂 good permanent full🄿 time jobs for people without a technical four year degree .<|endoftext|>,b : and the problem be that there be no good permanent full🄿 time jobs for people without a technical four year degree .<|endoftext|>,17.101606369018555,146.4150390625
+2893,b : my kids want to see rescuer's down🄿 under a : uh - huh<|endoftext|>,b : my kids want to see rescuer's down under a : uh - huh<|endoftext|>,5.088294506072998,143.79054260253906
+2894,"b : yeah , b : um , like soda bottles be one one type of🄿 plastic , b : and milk jugs be one type of plastic🄿 .<|endoftext|>","b : yeah , b : um , like soda bottles be one one type of plastic , b : and milk jugs be one type of plastic🄿 .<|endoftext|>",4.143820285797119,inf
+2895,"b : and it be , uh , it say , big🅂 story in the🅂 paper about it today , bush trying to comment on it , that colin powell was against the , uh , uh , an early , uh , b : he was for sanctions<|endoftext|>","b : and it be , uh , it say , big story in the🅂 paper about it today , bush trying to comment on it , that colin powell was against the , uh , uh , an early , uh , b : he was for sanctions<|endoftext|>",12.87870979309082,126.37317657470703
+2896,"a : they , they weigh him several times a🄿 year just to make sure he be not , you know , completely🅂 out of control<|endoftext|>","a : they , they weigh him several times a year just to make sure he be not , you know , completely🅂 out of control<|endoftext|>",3.720324754714966,128.59271240234375
+2897,"a : and a : i don't know exactly what that be , a : but i just🅂 assume it was something that<|endoftext|>","a : and a : i don't know exactly what that be , a : but i just assume it was something that<|endoftext|>",23.281291961669922,inf
+2898,"b : so , they be going to get , going🄿 to be called the twin peaks<|endoftext|>","b : so , they be going to get , going to be called the twin peaks<|endoftext|>",3.182354688644409,116.28372955322266
+2899,"b : that be going to affect our🅂 , uh , economy quite a bit<|endoftext|>","b : that be going to affect our , uh , economy quite a bit<|endoftext|>",17.779722213745117,inf
+2900,"because the way it be , a person really don't🅂 have a chance a🄿 : yeah<|endoftext|>","because the way it be , a person really don't have a chance a🄿 : yeah<|endoftext|>",21.895301818847656,inf
+2901,"b : yeah b : i've been trying to put weed spray on the lawn for the whole week b : and you can't put it on if it be suppose to rain within🅂 forty - eight hours , a : uh<|endoftext|>","b : yeah b : i've been trying to put weed spray on the lawn for the whole week b : and you can't put it on if it be suppose to rain within forty - eight hours , a : uh<|endoftext|>",19.585153579711914,inf
+2902,"a : and i was just great , a : and all you have to do be just put it in🅂 a , uh , the spelling check mode , b : uh - huh .<|endoftext|>","a : and i was just great , a : and all you have to do be just put it in a , uh , the spelling check mode , b : uh - huh .<|endoftext|>",14.869210243225098,126.77912902832031
+2903,"a : and then , ultimately , they say , no a : you don't🄿 have to put it down a : but , uh , really it be part of this compliance🅂<|endoftext|>","a : and then , ultimately , they say , no a : you don't have to put it down a : but , uh , really it be part of this compliance🅂<|endoftext|>",3.72639536857605,inf
+2904,"a : uh , of course , you know , with , uh , with , with dahlmer now , uh , you , you realize that i think it be ohio , uh , get to🅂 try him next🅂 a : and they do have the death penalty🄿<|endoftext|>","a : uh , of course , you know , with , uh , with , with dahlmer now , uh , you , you realize that i think it be ohio , uh , get to try him next🅂 a : and they do have the death penalty🄿<|endoftext|>",17.06167984008789,inf
+2905,"a : well , actually drug use be probably a misdemeanor , a🅂 : but , uh , what kind , what kind , a : have you been , uh , have you been the subject of such a crime , such as stealing , or anything ?<|endoftext|>","a : well , actually drug use be probably a misdemeanor , a : but , uh , what kind , what kind , a : have you been , uh , have you been the subject of such a crime , such as stealing , or anything ?<|endoftext|>",14.222916603088379,145.830078125
+2906,"a : i , uh , you know , a : seeing the children in the school the way i do , i , i mean , i see the ones that , we can always tell the ones that have not been around children🄿 at all when they come in as three year🄿 olds as opposed to the ones that have been in the program🄿 , that started at twelve months maybe , and were were even in there one day a week which be all our school provide🅂 for the , you know🅂 for the under three year olds b : uh - huh .<|endoftext|>","a : i , uh , you know , a : seeing the children in the school the way i do , i , i mean , i see the ones that , we can always tell the ones that have not been around children at all when they come in as three year🄿 olds as opposed to the ones that have been in the program🄿 , that started at twelve months maybe , and were were even in there one day a week which be all our school provide🅂 for the , you know🅂 for the under three year olds b : uh - huh .<|endoftext|>",5.074179649353027,inf
+2907,a : such as new york city b : be that what happen up🅂 there a lot🅂 ?<|endoftext|>,a : such as new york city b : be that what happen up there a lot🅂 ?<|endoftext|>,12.850533485412598,125.70975494384766
+2908,"a : i teach at a public school , actually , a : and i know that the s a t scores for our admitting freshmen be higher than a lot🄿 of the public a lot of the private schools .<|endoftext|>","a : i teach at a public school , actually , a : and i know that the s a t scores for our admitting freshmen be higher than a lot of the public a lot of the private schools .<|endoftext|>",3.500976324081421,67.78467559814453
+2909,"b : and it seem like if you could🅂 eliminate one of , of the parts of that circle , whether , where you have the dropout rate and crime and , you know , general poverty kind of conditions , that things ought to get better .<|endoftext|>","b : and it seem like if you could eliminate one of , of the parts of that circle , whether , where you have the dropout rate and crime and , you know , general poverty kind of conditions , that things ought to get better .<|endoftext|>",20.896728515625,inf
+2910,"b : but , i'm not going to put a skirt on just for him a : oh , you don't think you will , huh a : well , you'll be excused because considering everything b : yeah , b : i think that the main issue at home be being comfortable in your🅂 clothing a : uh<|endoftext|>","b : but , i'm not going to put a skirt on just for him a : oh , you don't think you will , huh a : well , you'll be excused because considering everything b : yeah , b : i think that the main issue at home be being comfortable in your clothing a : uh<|endoftext|>",19.02037239074707,inf
+2911,"b : my , um , parents have one b : and they🄿 live just about three miles🄿 away , a : oh. b : so we use that one .<|endoftext|>","b : my , um , parents have one b : and they live just about three miles🄿 away , a : oh. b : so we use that one .<|endoftext|>",2.8621585369110107,inf
+2912,"b : oh. a : i like , i think kevin costner be a nice fellow , a🅂 : and he be nice looking , a : but🅂 he just , to me , be not a very dynamic🅂<|endoftext|>","b : oh. a : i like , i think kevin costner be a nice fellow , a : and he be nice looking , a : but🅂 he just , to me , be not a very dynamic🅂<|endoftext|>",11.958260536193848,115.20930480957031
+2913,"a : and , um , it be , it be pretty , like🅂 , windy and🅂<|endoftext|>","a : and , um , it be , it be pretty , like , windy and🅂<|endoftext|>",14.25021743774414,124.51256561279297
+2914,"b : it was , i think it was one of those things where , it be not like someone came🅂 out and said you can't wear<|endoftext|>","b : it was , i think it was one of those things where , it be not like someone came out and said you can't wear<|endoftext|>",16.70035743713379,inf
+2915,"a : and one live in connecticut skip on🅂 speaker a be🅂 a : and our baby🅂 live just a mile over🅂 the hill so , we'll get to see her a little more than , uh , what we do the other<|endoftext|>","a : and one live in connecticut skip on speaker a be🅂 a : and our baby🅂 live just a mile over🅂 the hill so , we'll get to see her a little more than , uh , what we do the other<|endoftext|>",14.347387313842773,142.5405731201172
+2916,"b : but a lot of people where i live be working just because they🄿 want bigger and better , b🄿 : and i would never do that .<|endoftext|>","b : but a lot of people where i live be working just because they want bigger and better , b🄿 : and i would never do that .<|endoftext|>",3.3154706954956055,inf
+2917,"b : and , uh , it , it be a little , it be🅂 a little disappointing sometimes🅂 you start to realize what you<|endoftext|>","b : and , uh , it , it be a little , it be a little disappointing sometimes🅂 you start to realize what you<|endoftext|>",12.596076965332031,109.8703842163086
+2918,"b : um , because there be so many , um , kids🅂 who have , uh , you know , so🄿 much money and so much free time and , uh , nothing to do , um , and nothing to give their time to<|endoftext|>","b : um , because there be so many , um , kids who have , uh , you know , so🄿 much money and so much free time and , uh , nothing to do , um , and nothing to give their time to<|endoftext|>",15.698322296142578,inf
+2919,"b : oh , yeah b : so they have a toe hold , b🄿 : i good toe hold .<|endoftext|>","b : oh , yeah b : so they have a toe hold , b : i good toe hold .<|endoftext|>",3.3120486736297607,102.88591003417969
+2920,b : they have built up one mountain🄿 .<|endoftext|>,b : they have built up one mountain .<|endoftext|>,4.095077037811279,inf
+2921,"a : and finally , my children be the ones who said🄿 , "" mom , you've put up with that skunk long enough , "" a : so , we moved her to the basement , a : and i don't think she was there a week till the men moved her , built her a pen outside b : oh , my<|endoftext|>","a : and finally , my children be the ones who said , "" mom , you've put up with that skunk long enough , "" a : so , we moved her to the basement , a : and i don't think she was there a week till the men moved her , built her a pen outside b : oh , my<|endoftext|>",4.316069602966309,143.0
+2922,"a : so , and then if we really want to , you know , something different we go to modesto which be like twenty minutes away🅂 .<|endoftext|>","a : so , and then if we really want to , you know , something different we go to modesto which be like twenty minutes away .<|endoftext|>",14.595040321350098,149.0
+2923,"a : and they have a particular child actor🄿 , uh , who be , who be being supposedly🅂 run down🅂 by a<|endoftext|>","a : and they have a particular child actor , uh , who be , who be being supposedly🅂 run down🅂 by a<|endoftext|>",5.180068016052246,inf
+2924,"a : i think we've , i think it be come to an eye🅂 conclusion a : so i'll talk with<|endoftext|>","a : i think we've , i think it be come to an eye conclusion a : so i'll talk with<|endoftext|>",13.873780250549316,146.4150390625
+2925,"a : so , i , i , i like to read the , the children's books , a : i think that be the most animated and🅂<|endoftext|>","a : so , i , i , i like to read the , the children's books , a : i think that be the most animated and<|endoftext|>",17.770727157592773,inf
+2926,"b : and , uh , the kids like to play nintendo , b🄿 : but i'm no good at it , b : so i don't like to<|endoftext|>","b : and , uh , the kids like to play nintendo , b : but i'm no good at it , b : so i don't like to<|endoftext|>",4.064302921295166,137.70997619628906
+2927,"a : and , and they say , oh , we , we know🄿 you .<|endoftext|>","a : and , and they say , oh , we , we know you .<|endoftext|>",4.231233596801758,inf
+2928,"b : and when it be raining , i'm more likely🅂 to wear jeans b : and , and when it be really cold i'm more🅂 likely to wear jeans or pants or sweaters , or that<|endoftext|>","b : and when it be raining , i'm more likely to wear jeans b : and , and when it be really cold i'm more🅂 likely to wear jeans or pants or sweaters , or that<|endoftext|>",7.040292739868164,45.89748001098633
+2929,"b : my wife , uh , run or like to run🅂 , b : but🅂 uh , she be , b : well , we moved🅂 , we moved down to the north end of oak cliff in the castle park area<|endoftext|>","b : my wife , uh , run or like to run , b : but🅂 uh , she be , b : well , we moved🅂 , we moved down to the north end of oak cliff in the castle park area<|endoftext|>",19.532869338989258,inf
+2930,"a : that be what people said , b🅂 : i know some friends who went to see it<|endoftext|>","a : that be what people said , b : i know some friends who went to see it<|endoftext|>",14.239428520202637,139.3328857421875
+2931,"a : well , um , be there anything else you🅂 'd like to discuss about it<|endoftext|>","a : well , um , be there anything else you 'd like to discuss about it<|endoftext|>",20.493011474609375,inf
+2932,"a : oh , okay , a : he work for t i. b🅂 : no , b : no , b : but his wife have a contract with t🅂 i ,<|endoftext|>","a : oh , okay , a : he work for t i. b : no , b : no , b : but his wife have a contract with t🅂 i ,<|endoftext|>",13.953927040100098,129.2244873046875
+2933,b : i just catch the major ones off the news or if someone start to talk about it🅂 .<|endoftext|>,b : i just catch the major ones off the news or if someone start to talk about it .<|endoftext|>,17.006328582763672,inf
+2934,"b : and i get paid very little b : but , uh , you know , i try to use that money for food b : jim give me a certain amount🅂 money .<|endoftext|>","b : and i get paid very little b : but , uh , you know , i try to use that money for food b : jim give me a certain amount money .<|endoftext|>",15.855982780456543,inf
+2935,"so that be just , you know , that🅂 be small town trying to🅂 make it through the not being around it really , a : yeah , a : right , a :<|endoftext|>","so that be just , you know , that be small town trying to🅂 make it through the not being around it really , a : yeah , a : right , a :<|endoftext|>",16.4383487701416,inf
+2936,"b : that be what you do , uh🅂 , a : yeah<|endoftext|>","b : that be what you do , uh , a : yeah<|endoftext|>",11.805670738220215,106.22843170166016
+2937,"a : and maybe people be trying , a : but they🄿 have to change attitudes , b🄿 : and it seem to me that it🅂 be already so evident with🅂 our children having a lot more respiratory infections than we had growing up you can really see it<|endoftext|>","a : and maybe people be trying , a : but they have to change attitudes , b🄿 : and it seem to me that it🅂 be already so evident with🅂 our children having a lot more respiratory infections than we had growing up you can really see it<|endoftext|>",3.517648935317993,inf
+2938,"a : i said this be the most , a : if🅂 we had a quarterback this year we would have went to the super bowl .<|endoftext|>","a : i said this be the most , a : if we had a quarterback this year we would have went to the super bowl .<|endoftext|>",13.848448753356934,127.9565658569336
+2939,"how have , having been in a🅂 territory but only as a young student b : and my parents were in the military at the time b : so they didn't have ready negative feelings about being in , in alaska at the time since they voted absentee<|endoftext|>","how have , having been in a territory but only as a young student b : and my parents were in the military at the time b : so they didn't have ready negative feelings about being in , in alaska at the time since they voted absentee<|endoftext|>",15.35930061340332,inf
+2940,"a : i don't know if you've ever read that b : no. a : but , but , um , oh , there be one about an indian🅂 girl , a : and i can't think of that , a : island of<|endoftext|>","a : i don't know if you've ever read that b : no. a : but , but , um , oh , there be one about an indian girl , a : and i can't think of that , a : island of<|endoftext|>",15.657398223876953,inf
+2941,"b : it be like , b : i mean🅂 it just blew me away<|endoftext|>","b : it be like , b : i mean it just blew me away<|endoftext|>",13.40456485748291,125.44007110595703
+2942,"a : and , so that , you know , a : if the setting was right and that , it be , it be great , a🅂 : but if🅂 it be , a : well , i've been🅂 real , um , thinking i 'd love to buy a new car , right now , if<|endoftext|>","a : and , so that , you know , a : if the setting was right and that , it be , it be great , a : but if🅂 it be , a : well , i've been🅂 real , um , thinking i 'd love to buy a new car , right now , if<|endoftext|>",13.808771133422852,134.05950927734375
+2943,"a : yeah , a : and they be still , uh , rebuilding from🄿 that storm , a : and so the , the beaches here just get really beat up a🄿 : and it usually come this way b : yeah🅂<|endoftext|>","a : yeah , a : and they be still , uh , rebuilding from that storm , a : and so the , the beaches here just get really beat up a🄿 : and it usually come this way b : yeah🅂<|endoftext|>",4.192942142486572,inf
+2944,"a : well , well , okay , well like my oldest brother , he be , he be twenty - four🅂 , b : would🅂 be ,<|endoftext|>","a : well , well , okay , well like my oldest brother , he be , he be twenty - four , b : would🅂 be ,<|endoftext|>",13.002774238586426,124.22705841064453
+2945,"a : i think , lets go to war , * lets let be a : but let's go🅂 to war praying , a : and let's go to war for that way a : so , b : well , well , the difficulty i have , be that , uh , are we🅂 our<|endoftext|>","a : i think , lets go to war , * lets let be a : but let's go to war praying , a : and let's go to war for that way a : so , b : well , well , the difficulty i have , be that , uh , are we🅂 our<|endoftext|>",16.371822357177734,inf
+2946,"b : i , b : and , and then , and yet they have their , their friendships and🄿 relationships and the , in you know , of getting together and eating together and all of that .<|endoftext|>","b : i , b : and , and then , and yet they have their , their friendships and relationships and the , in you know , of getting together and eating together and all of that .<|endoftext|>",4.110769271850586,149.0
+2947,"a : that be pretty good for , you🅂 know , being a , a school district type program<|endoftext|>","a : that be pretty good for , you know , being a , a school district type program<|endoftext|>",20.45636749267578,inf
+2948,"a : uh , interior be not so bad because🅂 it be more fun , a : it🅂 be more , a : but yeah🅂 , b<|endoftext|>","a : uh , interior be not so bad because it be more fun , a : it🅂 be more , a : but yeah🅂 , b<|endoftext|>",15.449540138244629,138.37338256835938
+2949,b : and it be so easy to convert🅂 from one set of units to another<|endoftext|>,b : and it be so easy to convert from one set of units to another<|endoftext|>,17.335063934326172,inf
+2950,"a : and , you know , to , i'm sure he get to watch a lot🅂 of t v a : and i'm sure they have , uh , exercise facilities and🄿 movies a : and , and everything else , you know , they can have a good time<|endoftext|>","a : and , you know , to , i'm sure he get to watch a lot of t v a : and i'm sure they have , uh , exercise facilities and🄿 movies a : and , and everything else , you know , they can have a good time<|endoftext|>",10.284399032592773,86.89942932128906
+2951,a : it be interesting that you said🅂 that<|endoftext|>,a : it be interesting that you said that<|endoftext|>,17.8565616607666,inf
+2952,"a : but , uh , like my husband was saying , maybe some of the problems we brought on to ourselves be that these women back🅂 in the nineteen fifties , uh , were staying at home with their kids and then made the decision , well , jeez , you know , i'll go out and get a part - time job<|endoftext|>","a : but , uh , like my husband was saying , maybe some of the problems we brought on to ourselves be that these women back in the nineteen fifties , uh , were staying at home with their kids and then made the decision , well , jeez , you know , i'll go out and get a part - time job<|endoftext|>",14.065545082092285,140.85025024414062
+2953,"b : um , nothing come to mind right off🅂 , b : so i guess not .<|endoftext|>","b : um , nothing come to mind right off , b : so i guess not .<|endoftext|>",11.155601501464844,102.68267822265625
+2954,"b : well for , for a lot of the kind of premeditated murders , i think the death penalty be , b : it be a🅂 pretty good thing🅂<|endoftext|>","b : well for , for a lot of the kind of premeditated murders , i think the death penalty be , b : it be a pretty good thing🅂<|endoftext|>",15.410357475280762,inf
+2955,"b : you drive up b : and you take the stuff out of your trunk , b : and they have different bins for the🄿 different materials , like clear , glass and colored glass and plastics b : and , uh , they have all the different plastics🄿 separated by those little symbols on the bottom .<|endoftext|>","b : you drive up b : and you take the stuff out of your trunk , b : and they have different bins for the different materials , like clear , glass and colored glass and plastics b : and , uh , they have all the different plastics🄿 separated by those little symbols on the bottom .<|endoftext|>",3.9784679412841797,inf
+2956,"a : and , it , it be , b : do you work🅂 with t i ? b : no. a : no , a : i wish i did<|endoftext|>","a : and , it , it be , b : do you work with t i ? b : no. a : no , a : i wish i did<|endoftext|>",16.650320053100586,inf
+2957,"a : yeah , a : yeah , a : you , a : it be not , it be not🅂 just a matter🅂 of , uh , of having an extra car or having a better car , this be a matter for them🅂 of , of feeding themselves their wife and their family a : and , and you , a : there be a certain degree of🅂 honor in being able to feed your family a : and there be dishonor in not being🅂 able to do it a : and ... b : uh , starving to death be not a , not a🅂 whole lot of fun either b : and , that , a : no , a : no. a :<|endoftext|>","a : yeah , a : yeah , a : you , a : it be not , it be not just a matter🅂 of , uh , of having an extra car or having a better car , this be a matter for them🅂 of , of feeding themselves their wife and their family a : and , and you , a : there be a certain degree of🅂 honor in being able to feed your family a : and there be dishonor in not being🅂 able to do it a : and ... b : uh , starving to death be not a , not a🅂 whole lot of fun either b : and , that , a : no , a : no. a :<|endoftext|>",12.659712791442871,118.90481567382812
+2958,"a : and that their governments can become so corrupt a : and it be like , well , you know🅂 , i use to feel like , a : well , they , they , should work their way , you know , a : they should , should be able to work their self out of it a : or , you know , why , why , why intervene in another country's , you know , problems and things like<|endoftext|>","a : and that their governments can become so corrupt a : and it be like , well , you know , i use to feel like , a : well , they , they , should work their way , you know , a : they should , should be able to work their self out of it a : or , you know , why , why , why intervene in another country's , you know , problems and things like<|endoftext|>",14.784449577331543,135.35220336914062
+2959,"a : yeah , a : it , it be hard for people to🅂 get in with a two door sometimes<|endoftext|>","a : yeah , a : it , it be hard for people to get in with a two door sometimes<|endoftext|>",23.39067840576172,inf
+2960,"a : my brother , b : well , my son be , uh , studying electronic technology🅂 , b : and he be been able to do🅂 things on uh , on the personal computer here that really have enhanced his , um , reports🄿 , and his learning and all because he can lay things out that way<|endoftext|>","a : my brother , b : well , my son be , uh , studying electronic technology , b : and he be been able to do🅂 things on uh , on the personal computer here that really have enhanced his , um , reports🄿 , and his learning and all because he can lay things out that way<|endoftext|>",17.615825653076172,inf
+2961,"a : yeah , a : and the rest be , uh , get into the🅂 cooler weather🅂 .<|endoftext|>","a : yeah , a : and the rest be , uh , get into the cooler weather🅂 .<|endoftext|>",16.577529907226562,inf
+2962,"uh. b : oh , well that be a , b : did you🅂 push the button<|endoftext|>","uh. b : oh , well that be a , b : did you push the button<|endoftext|>",22.730344772338867,inf
+2963,"a : but i , i'm not good enough for that b : do your club have a🅂 good woman's<|endoftext|>","a : but i , i'm not good enough for that b : do your club have a good woman's<|endoftext|>",15.787286758422852,149.0
+2964,"a : and if , if my memory serve me right , i believe🅂 a majority of democrats may even have opposed it , a : it , it was , b : yeah , b : i'm , i don't know the exact figure b : but i , i would imagine that because of the , how close the , the vote was i'm , uh , i mean almost certainly would have<|endoftext|>","a : and if , if my memory serve me right , i believe a majority of democrats may even have opposed it , a : it , it was , b : yeah , b : i'm , i don't know the exact figure b : but i , i would imagine that because of the , how close the , the vote was i'm , uh , i mean almost certainly would have<|endoftext|>",15.852250099182129,inf
+2965,b : that be what we're trying to🅂 get ours to look<|endoftext|>,b : that be what we're trying to get ours to look<|endoftext|>,19.725378036499023,inf
+2966,"b : it have , to see it on🅂 the big screen because it be just the most gorgeous🅂 nature scenes you've ever seen in your<|endoftext|>","b : it have , to see it on the big screen because it be just the most gorgeous🅂 nature scenes you've ever seen in your<|endoftext|>",14.129881858825684,117.57400512695312
+2967,a : they look like sometimes they be🄿 just broken in two🄿 like a match stem<|endoftext|>,a : they look like sometimes they be just broken in two🄿 like a match stem<|endoftext|>,4.998979091644287,inf
+2968,"a : uh , that be not very , b : big🅂 long strips<|endoftext|>","a : uh , that be not very , b : big long strips<|endoftext|>",11.849698066711426,115.1517105102539
+2969,"when they especially when someone , they , they don't vote for someone because🄿 they don't like any of them🄿 and then the person get in and they don't🅂 like him and he🄿 turn out to have been🅂 worse than her that they might have voted for , or something<|endoftext|>","when they especially when someone , they , they don't vote for someone because they don't like any of them🄿 and then the person get in and they don't🅂 like him and he🄿 turn out to have been🅂 worse than her that they might have voted for , or something<|endoftext|>",4.772434234619141,inf
+2970,a : and the only time we got worried be if it hit the🅂 ground .<|endoftext|>,a : and the only time we got worried be if it hit the ground .<|endoftext|>,17.48743438720703,inf
+2971,"b : my roommates be , b : so , unfortunately , i🄿 forced to watch quite a bit of sports .<|endoftext|>","b : my roommates be , b : so , unfortunately , i forced to watch quite a bit of sports .<|endoftext|>",2.3883979320526123,inf
+2972,"a : you get some pollutants such as those , a : but those be natural b : yeah a🄿 : um , but there be starting to , a : in🅂 minneapolis itself , because of so many people on the highway , there be becoming a problem of🅂 pollution , a : and they just put in a strict law that as of every year when you get your license tabs you have to have your car inspected to see if it be releasing any , uh , lead🅂 into the air or other pollutants , b : uh - huh<|endoftext|>","a : you get some pollutants such as those , a : but those be natural b : yeah a : um , but there be starting to , a : in🅂 minneapolis itself , because of so many people on the highway , there be becoming a problem of🅂 pollution , a : and they just put in a strict law that as of every year when you get your license tabs you have to have your car inspected to see if it be releasing any , uh , lead🅂 into the air or other pollutants , b : uh - huh<|endoftext|>",3.958045482635498,124.41850280761719
+2973,a : and that do make a big difference🅂 .<|endoftext|>,a : and that do make a big difference .<|endoftext|>,13.133777618408203,135.6450653076172
+2974,"a : we can either act like we're going to be pro - communist or act like we're going to be pro , uh , pro - democracy a : and whatever get us the most honor🅂 that be what we'll act like🅂 we're going to do , a : but we really won't<|endoftext|>","a : we can either act like we're going to be pro - communist or act like we're going to be pro , uh , pro - democracy a : and whatever get us the most honor that be what we'll act like🅂 we're going to do , a : but we really won't<|endoftext|>",14.750638961791992,inf
+2975,"b : uh - huh , b : it be , b : and you know🅂 , you , you don't well<|endoftext|>","b : uh - huh , b : it be , b : and you know , you , you don't well<|endoftext|>",15.256596565246582,139.50015258789062
+2976,b : it be real entertaining for her🅂<|endoftext|>,b : it be real entertaining for her<|endoftext|>,13.627217292785645,133.05255126953125
+2977,"b : i don't know , b : i mean i think sometimes some of the , the women's lib though , be kind of like they🅂 wanted it all , you know , b : and you can't have everything a : uh<|endoftext|>","b : i don't know , b : i mean i think sometimes some of the , the women's lib though , be kind of like they wanted it all , you know , b : and you can't have everything a : uh<|endoftext|>",20.520732879638672,inf
+2978,"a : yeah , a : that be exactly it , a : and🅂 that be what we're finding , um🅂 , here , where we're at , in minneapolis area , be that people don't want🅂 to carpool that🄿 there be inconveniences to that , b🄿<|endoftext|>","a : yeah , a : that be exactly it , a : and that be what we're finding , um🅂 , here , where we're at , in minneapolis area , be that people don't want🅂 to carpool that🄿 there be inconveniences to that , b🄿<|endoftext|>",19.95104217529297,inf
+2979,"a : well , then , um , b : mine be broke at the moment🅂 , b : but a : yeah , a : ours really need to have some work🅂 done on it a : but , but then my husband , um , worked at t i a : and i got a , um , this have been years ago he🅂 don't work there anymore , but🅂 he got a , um , uh , the , the t i professional and brought that home a : and i learned wordstar on that , b : oh. a : and , boy once that happened , there was just no way that i could ever go back to doing anything of length of the<|endoftext|>","a : well , then , um , b : mine be broke at the moment , b : but a : yeah , a : ours really need to have some work🅂 done on it a : but , but then my husband , um , worked at t i a : and i got a , um , this have been years ago he🅂 don't work there anymore , but🅂 he got a , um , uh , the , the t i professional and brought that home a : and i learned wordstar on that , b : oh. a : and , boy once that happened , there was just no way that i could ever go back to doing anything of length of the<|endoftext|>",13.951464653015137,127.67466735839844
+2980,"a : and then , then my brother right behind me , he moved out , a : and he have an apartment , a : and🅂 he work , a : and then the🅂 next one down be in the marines , a🅂 : and he be out of the house🅂<|endoftext|>","a : and then , then my brother right behind me , he moved out , a : and he have an apartment , a : and he work , a : and then the🅂 next one down be in the marines , a🅂 : and he be out of the house🅂<|endoftext|>",19.624797821044922,inf
+2981,"a : this have not been determined yet🅂 a : but i , i , i've never read a newspaper in my entire life a : and i've , i , i never watch t v news nor listen to the news on the radio unless i'm just happening to be listening to music and they slip it on in the🄿 car radio before i can turn<|endoftext|>","a : this have not been determined yet a : but i , i , i've never read a newspaper in my entire life a : and i've , i , i never watch t v news nor listen to the news on the radio unless i'm just happening to be listening to music and they slip it on in the🄿 car radio before i can turn<|endoftext|>",14.057517051696777,137.7174835205078
+2982,"a : um , how do you feel about , um , the way the u s have reacted in the middle🅂 east ?<|endoftext|>","a : um , how do you feel about , um , the way the u s have reacted in the middle east ?<|endoftext|>",9.687779426574707,84.17496490478516
+2983,"b : and we've had , uh , they have always had , uh , where🄿 they match your funds , you know🄿 , b : sometimes if you , if you save so much , then they will match<|endoftext|>","b : and we've had , uh , they have always had , uh , where they match your funds , you know🄿 , b : sometimes if you , if you save so much , then they will match<|endoftext|>",4.6735053062438965,141.15451049804688
+2984,* listen a : a machine that put back grind tape on🅂 and off the waivers .<|endoftext|>,* listen a : a machine that put back grind tape on and off the waivers .<|endoftext|>,15.910270690917969,146.0
+2985,"b : but , i don't know , b : with the economy the way it be , i think crime can't🅂 do much but go<|endoftext|>","b : but , i don't know , b : with the economy the way it be , i think crime can't do much but go<|endoftext|>",20.19870376586914,inf
+2986,"b : so , i mean people be still wearing shorts b🄿 : and and , uh , usually around halloween it start getting cooler , a : uh🅂 - huh .<|endoftext|>","b : so , i mean people be still wearing shorts b : and and , uh , usually around halloween it start getting cooler , a : uh🅂 - huh .<|endoftext|>",4.486536502838135,148.0
+2987,"b : fuel economy ben't great , being a mini🅂 van , b : but , uh it be real nice for carrying🅂 around our kids and others b : and it be , uh , got some nice🅂 features in terms of being able to remove seats and flip them around a<|endoftext|>","b : fuel economy ben't great , being a mini van , b : but , uh it be real nice for carrying🅂 around our kids and others b : and it be , uh , got some nice🅂 features in terms of being able to remove seats and flip them around a<|endoftext|>",10.302000045776367,95.25740814208984
+2988,: i know that my husband and i have to do a lot🄿 of sacrificing to keep me home .<|endoftext|>,: i know that my husband and i have to do a lot of sacrificing to keep me home .<|endoftext|>,2.6036574840545654,73.68254852294922
+2989,"you get the paper all wet b : and you make it into new paper , b : and , that be fun , trying to explain🅂 that to a four year old a : right , a : yeah<|endoftext|>","you get the paper all wet b : and you make it into new paper , b : and , that be fun , trying to explain that to a four year old a : right , a : yeah<|endoftext|>",18.035127639770508,inf
+2990,"a : to me it was like it be going to take them🅂 quite a while to rebuild a : because everything , i mean anything that tom landry wanted or had , he was going to change it<|endoftext|>","a : to me it was like it be going to take them quite a while to rebuild a : because everything , i mean anything that tom landry wanted or had , he was going to change it<|endoftext|>",21.768781661987305,inf
+2991,"b : uh , yeah , b : basketball be cool , b : but football🅂 kind of after a while , a : which basketball<|endoftext|>","b : uh , yeah , b : basketball be cool , b : but football kind of after a while , a : which basketball<|endoftext|>",15.63904094696045,inf
+2992,"a : she lived with us for , um , three months , and , because she be starting to get alzheimer's🅂<|endoftext|>","a : she lived with us for , um , three months , and , because she be starting to get alzheimer's<|endoftext|>",14.220565795898438,inf
+2993,a : but a large percentage probably over half of the rest of the population they don't have any college probably🄿 much at all b : absolutely<|endoftext|>,a : but a large percentage probably over half of the rest of the population they don't have any college probably much at all b : absolutely<|endoftext|>,4.415004730224609,inf
+2994,"a : well , and , um , it be nice for them not🅂 to be alone<|endoftext|>","a : well , and , um , it be nice for them not to be alone<|endoftext|>",11.183171272277832,89.81124114990234
+2995,"b : well , that be a pretty good deal🅂<|endoftext|>","b : well , that be a pretty good deal<|endoftext|>",14.297060012817383,inf
+2996,"b : yeah , b : well , my goal be to try to retire🅂 by the time i'm fifty - five<|endoftext|>","b : yeah , b : well , my goal be to try to retire by the time i'm fifty - five<|endoftext|>",18.202116012573242,inf
+2997,b : i know for sure though they have the newspaper and the🄿 aluminum can and then the oil at the fire departments .<|endoftext|>,b : i know for sure though they have the newspaper and the aluminum can and then the oil at the fire departments .<|endoftext|>,4.017918109893799,inf
+2998,"a : but , i mean , i don't know whether they have one of the first🄿 , you know , choices of somebody , b : right<|endoftext|>","a : but , i mean , i don't know whether they have one of the first , you know , choices of somebody , b : right<|endoftext|>",4.682258605957031,inf
+2999,"b : i'm not sure how we get that , b : wonder if there be a connection there , b🅂 : but , a :<|endoftext|>","b : i'm not sure how we get that , b : wonder if there be a connection there , b : but , a :<|endoftext|>",10.864410400390625,94.70519256591797
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_control.csv b/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_control.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_control_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_control_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_words4.csv b/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_words4.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_words4_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/0.5B/hop_words4_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_control.csv b/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_control.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_control_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_control_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_words4.csv b/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_words4.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_words4_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/1.5B/hop_words4_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/3B/hop_control.csv b/hop_surprisal/hop_surprisal_results/Qwen/3B/hop_control.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/3B/hop_words4.csv b/hop_surprisal/hop_surprisal_results/Qwen/3B/hop_words4.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/3B/hop_words4_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/3B/hop_words4_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_control.csv b/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_control.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_control_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_control_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_words4.csv b/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_words4.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_words4_10M_seed0.csv b/hop_surprisal/hop_surprisal_results/Qwen/7B/hop_words4_10M_seed0.csv
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/impossible_llm.yaml b/impossible_llm.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..03f0f7893e0b6c927b18f3359a3e1420ad43831a
--- /dev/null
+++ b/impossible_llm.yaml
@@ -0,0 +1,154 @@
+name: impossible_llm
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - ca-certificates=2024.7.2=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_1
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.15=h5eee18b_0
+  - pip=24.2=py39h06a4308_0
+  - python=3.9.19=h955ad1f_1
+  - readline=8.2=h5eee18b_0
+  - setuptools=72.1.0=py39h06a4308_0
+  - sqlite=3.45.3=h5eee18b_0
+  - tk=8.6.14=h39e8969_0
+  - wheel=0.44.0=py39h06a4308_0
+  - xz=5.4.6=h5eee18b_1
+  - zlib=1.2.13=h5eee18b_1
+  - pip:
+      - accelerate==0.34.2
+      - aiohappyeyeballs==2.4.2
+      - aiohttp==3.10.6
+      - aiosignal==1.3.1
+      - asttokens==2.2.1
+      - async-timeout==4.0.3
+      - attrs==24.2.0
+      - backcall==0.2.0
+      - black==24.8.0
+      - blessed==1.20.0
+      - certifi==2023.11.17
+      - charset-normalizer==3.3.2
+      - click==8.1.7
+      - cmake==3.30.3
+      - comm==0.1.2
+      - contourpy==1.2.0
+      - cycler==0.12.1
+      - data==0.4
+      - datasets==3.0.1
+      - debugpy==1.6.7
+      - decorator==5.1.1
+      - dill==0.3.8
+      - emoji==2.8.0
+      - exceptiongroup==1.1.0
+      - executing==1.2.0
+      - filelock==3.12.2
+      - fonttools==4.45.1
+      - frozenlist==1.4.1
+      - fsspec==2023.10.0
+      - funcsigs==1.0.2
+      - future==0.18.3
+      - gmpy2==2.1.0
+      - gpustat==1.1.1
+      - huggingface-hub==0.25.0
+      - idna==3.6
+      - importlib-metadata==6.0.0
+      - importlib-resources==6.4.5
+      - iniconfig==2.0.0
+      - ipykernel==6.23.1
+      - ipython==8.0.0
+      - jedi==0.18.2
+      - jinja2==3.1.2
+      - joblib==1.3.2
+      - jupyter-client==8.1.0
+      - jupyter-core==5.1.0
+      - kiwisolver==1.4.5
+      - latex==0.7.0
+      - latexcodec==1.0.0
+      - lit==18.1.8
+      - markupsafe==2.1.2
+      - matplotlib==3.8.2
+      - matplotlib-inline==0.1.6
+      - mizani==0.9.3
+      - mpmath==1.2.1
+      - multidict==6.1.0
+      - multiprocess==0.70.16
+      - mypy-extensions==1.0.0
+      - nest-asyncio==1.5.6
+      - networkx==2.8.6
+      - nltk==3.8.1
+      - numpy==1.26.2
+      - nvidia-cublas-cu11==11.10.3.66
+      - nvidia-cuda-cupti-cu11==11.7.101
+      - nvidia-cuda-nvrtc-cu11==11.7.99
+      - nvidia-cuda-runtime-cu11==11.7.99
+      - nvidia-cudnn-cu11==8.5.0.96
+      - nvidia-cufft-cu11==10.9.0.58
+      - nvidia-curand-cu11==10.2.10.91
+      - nvidia-cusolver-cu11==11.4.0.1
+      - nvidia-cusparse-cu11==11.7.4.91
+      - nvidia-ml-py==12.560.30
+      - nvidia-nccl-cu11==2.14.3
+      - nvidia-nvtx-cu11==11.7.91
+      - packaging==23.0
+      - pandas==2.1.3
+      - parso==0.8.3
+      - pathspec==0.12.1
+      - patsy==0.5.3
+      - peft==0.13.0
+      - pexpect==4.8.0
+      - pickleshare==0.7.5
+      - pillow==10.1.0
+      - platformdirs==2.5.2
+      - plotnine==0.12.4
+      - pluggy==1.3.0
+      - pluralizer==1.2.0
+      - prompt-toolkit==3.0.30
+      - protobuf==4.25.1
+      - psutil==5.9.1
+      - ptyprocess==0.7.0
+      - pure-eval==0.2.2
+      - pyarrow==17.0.0
+      - pygments==2.15.0
+      - pyparsing==3.1.1
+      - pytest==7.4.3
+      - python-dateutil==2.8.2
+      - pytz==2023.3.post1
+      - pyyaml==6.0.1
+      - pyzmq==23.0.0
+      - regex==2023.10.3
+      - requests==2.32.3
+      - safetensors==0.4.5
+      - scikit-learn==1.3.2
+      - scipy==1.11.4
+      - seaborn==0.13.0
+      - sentencepiece==0.2.0
+      - shutilwhich==1.1.0
+      - six==1.16.0
+      - stack-data==0.6.0
+      - stanza==1.9.2
+      - statsmodels==0.14.0
+      - sympy==1.11.1
+      - tempdir==0.7.1
+      - threadpoolctl==3.2.0
+      - tokenizers==0.20.0
+      - tomli==2.0.1
+      - torch==2.0.0
+      - tornado==6.2
+      - tqdm==4.66.5
+      - traitlets==5.7.1
+      - transformers==4.45.1
+      - triton==2.0.0
+      - typing-extensions==4.6.0
+      - tzdata==2023.3
+      - urllib3==2.1.0
+      - wcwidth==0.2.5
+      - xxhash==3.5.0
+      - yarl==1.13.0
+      - zipp==3.12.0
+prefix: /home/yiren/new_ssd2/chunhui/miniconda/envs/impossible_llm
diff --git a/impossible_llm_update.yaml b/impossible_llm_update.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..9b4eb7c8828ae3a64b85b355eeb7ab3e387e04bb
--- /dev/null
+++ b/impossible_llm_update.yaml
@@ -0,0 +1,162 @@
+name: impossible_llm
+channels:
+  - defaults
+dependencies:
+  - _libgcc_mutex=0.1=main
+  - _openmp_mutex=5.1=1_gnu
+  - ca-certificates=2024.7.2=h06a4308_0
+  - ld_impl_linux-64=2.38=h1181459_1
+  - libffi=3.4.4=h6a678d5_1
+  - libgcc-ng=11.2.0=h1234567_1
+  - libgomp=11.2.0=h1234567_1
+  - libstdcxx-ng=11.2.0=h1234567_1
+  - ncurses=6.4=h6a678d5_0
+  - openssl=3.0.15=h5eee18b_0
+  - pip=24.2=py39h06a4308_0
+  - python=3.9.19=h955ad1f_1
+  - readline=8.2=h5eee18b_0
+  - setuptools=72.1.0=py39h06a4308_0
+  - sqlite=3.45.3=h5eee18b_0
+  - tk=8.6.14=h39e8969_0
+  - wheel=0.44.0=py39h06a4308_0
+  - xz=5.4.6=h5eee18b_1
+  - zlib=1.2.13=h5eee18b_1
+  - pip:
+      - accelerate==1.0.0
+      - aiohappyeyeballs==2.4.2
+      - aiohttp==3.10.6
+      - aiosignal==1.3.1
+      - annotated-types==0.7.0
+      - asttokens==2.2.1
+      - async-timeout==4.0.3
+      - attrs==24.2.0
+      - backcall==0.2.0
+      - black==24.8.0
+      - blessed==1.20.0
+      - certifi==2023.11.17
+      - charset-normalizer==3.3.2
+      - click==8.1.7
+      - cmake==3.30.3
+      - comm==0.1.2
+      - contourpy==1.2.0
+      - cycler==0.12.1
+      - data==0.4
+      - datasets==3.0.1
+      - debugpy==1.6.7
+      - decorator==5.1.1
+      - deepspeed==0.15.2
+      - dill==0.3.8
+      - emoji==2.8.0
+      - exceptiongroup==1.1.0
+      - executing==1.2.0
+      - filelock==3.12.2
+      - fonttools==4.45.1
+      - frozenlist==1.4.1
+      - fsspec==2023.10.0
+      - funcsigs==1.0.2
+      - future==0.18.3
+      - gmpy2==2.1.0
+      - gpustat==1.1.1
+      - hjson==3.1.0
+      - huggingface-hub==0.25.0
+      - idna==3.6
+      - importlib-metadata==6.0.0
+      - importlib-resources==6.4.5
+      - iniconfig==2.0.0
+      - ipykernel==6.23.1
+      - ipython==8.0.0
+      - jedi==0.18.2
+      - jinja2==3.1.2
+      - joblib==1.3.2
+      - jupyter-client==8.1.0
+      - jupyter-core==5.1.0
+      - kiwisolver==1.4.5
+      - latex==0.7.0
+      - latexcodec==1.0.0
+      - lit==18.1.8
+      - markupsafe==2.1.2
+      - matplotlib==3.8.2
+      - matplotlib-inline==0.1.6
+      - mizani==0.9.3
+      - mpmath==1.2.1
+      - msgpack==1.1.0
+      - multidict==6.1.0
+      - multiprocess==0.70.16
+      - mypy-extensions==1.0.0
+      - nest-asyncio==1.5.6
+      - networkx==2.8.6
+      - ninja==1.11.1.1
+      - nltk==3.8.1
+      - numpy==1.26.2
+      - nvidia-cublas-cu11==11.10.3.66
+      - nvidia-cuda-cupti-cu11==11.7.101
+      - nvidia-cuda-nvrtc-cu11==11.7.99
+      - nvidia-cuda-runtime-cu11==11.7.99
+      - nvidia-cudnn-cu11==8.5.0.96
+      - nvidia-cufft-cu11==10.9.0.58
+      - nvidia-curand-cu11==10.2.10.91
+      - nvidia-cusolver-cu11==11.4.0.1
+      - nvidia-cusparse-cu11==11.7.4.91
+      - nvidia-ml-py==12.560.30
+      - nvidia-nccl-cu11==2.14.3
+      - nvidia-nvtx-cu11==11.7.91
+      - packaging==23.0
+      - pandas==2.1.3
+      - parso==0.8.3
+      - pathspec==0.12.1
+      - patsy==0.5.3
+      - peft==0.13.0
+      - pexpect==4.8.0
+      - pickleshare==0.7.5
+      - pillow==10.1.0
+      - platformdirs==2.5.2
+      - plotnine==0.12.4
+      - pluggy==1.3.0
+      - pluralizer==1.2.0
+      - prompt-toolkit==3.0.30
+      - protobuf==4.25.1
+      - psutil==5.9.1
+      - ptyprocess==0.7.0
+      - pure-eval==0.2.2
+      - py-cpuinfo==9.0.0
+      - pyarrow==17.0.0
+      - pydantic==2.9.2
+      - pydantic-core==2.23.4
+      - pygments==2.15.0
+      - pyparsing==3.1.1
+      - pytest==7.4.3
+      - python-dateutil==2.8.2
+      - pytz==2023.3.post1
+      - pyyaml==6.0.1
+      - pyzmq==23.0.0
+      - regex==2023.10.3
+      - requests==2.32.3
+      - safetensors==0.4.5
+      - scikit-learn==1.3.2
+      - scipy==1.11.4
+      - seaborn==0.13.0
+      - sentencepiece==0.2.0
+      - shutilwhich==1.1.0
+      - six==1.16.0
+      - stack-data==0.6.0
+      - stanza==1.9.2
+      - statsmodels==0.14.0
+      - sympy==1.11.1
+      - tempdir==0.7.1
+      - threadpoolctl==3.2.0
+      - tokenizers==0.20.0
+      - tomli==2.0.1
+      - torch==2.0.0
+      - tornado==6.2
+      - tqdm==4.66.5
+      - traitlets==5.7.1
+      - transformers==4.45.1
+      - triton==2.0.0
+      - typing-extensions==4.12.2
+      - tzdata==2023.3
+      - urllib3==2.1.0
+      - wcwidth==0.2.5
+      - xxhash==3.5.0
+      - yarl==1.13.0
+      - zipp==3.12.0
+prefix: /home/yiren/new_ssd2/chunhui/miniconda/envs/impossible_llm
diff --git a/perplexities/demo.py b/perplexities/demo.py
new file mode 100644
index 0000000000000000000000000000000000000000..d6a6608f730c7f1883b9715c06330237a77c2394
--- /dev/null
+++ b/perplexities/demo.py
@@ -0,0 +1,82 @@
+import torch
+import sys
+import argparse
+import os
+sys.path.append("..")
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from datasets import load_dataset
+from numpy.random import default_rng
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+MODEL_NAME_SAVE = "Llama-3.2-3B"
+FILE_SAMPLE_SIZE = 5
+
+def get_perplexities(model, eval_dataset, batch_size):
+    data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False)
+
+    training_args = TrainingArguments(
+        output_dir="./tmp_trainer",
+        per_device_eval_batch_size=batch_size,
+        fp16=True,
+        report_to="none"
+    )
+
+    trainer = Trainer(model=model, args=training_args, eval_dataset=eval_dataset, data_collator=data_collator)
+    eval_results = trainer.evaluate()
+    print("eval_results:", eval_results)  
+    loss = eval_results['eval_loss']
+    perplexity = torch.exp(torch.tensor(loss)).item()
+    return perplexity
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Calculate perplexity on test dataset.")
+
+    parser.add_argument('perturbation', 
+                        type=str, 
+                        default='reverse_full', 
+                        nargs='?', 
+                        help='Type of perturbation to use.')
+    parser.add_argument('train_set', 
+                        type=str, 
+                        default='test', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('batch_size', 
+                        type=int, 
+                        default=4, 
+                        nargs='?', 
+                        help='Batch size for evaluation.')
+    parser.add_argument('seed', 
+                        type=int, 
+                        default=0, 
+                        nargs='?', 
+                        help='Random seed.')
+
+    args = parser.parse_args()
+
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('../train/babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    test_dataset = dataset['test']  # Load test dataset
+    print(test_dataset)
+    
+    checkpoint_path = f'../train/checkpoints/{MODEL_NAME_SAVE}/babylm_{args.perturbation}_10M_seed0/runs/checkpoint-450'
+    
+    rng = default_rng(args.seed)
+    indices = rng.choice(len(test_dataset), FILE_SAMPLE_SIZE, replace=False)
+    sampled_test_dataset = test_dataset.select(indices)
+
+    tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
+    model = AutoModelForCausalLM.from_pretrained(checkpoint_path)
+    
+    model.eval()
+    if torch.cuda.is_available():
+        model.to('cuda')
+
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+
+    tokenized_test = sampled_test_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+
+    perplexity = get_perplexities(model, tokenized_test, 1)
+    print(f"Perplexity on test set: {perplexity}")
\ No newline at end of file
diff --git a/perplexities/gpt2_mask.py b/perplexities/gpt2_mask.py
new file mode 100644
index 0000000000000000000000000000000000000000..1413d04a5a6374468225cfa4723f082520f8db17
--- /dev/null
+++ b/perplexities/gpt2_mask.py
@@ -0,0 +1,231 @@
+import torch
+import sys
+import argparse
+import os
+import pandas as pd
+import matplotlib.pyplot as plt
+from tqdm import tqdm
+import gc
+from numpy.random import default_rng
+sys.path.append("..")
+from transformers import (
+    AutoTokenizer, 
+    AutoModelForCausalLM,
+    Trainer, 
+    TrainingArguments, 
+    DataCollatorForLanguageModeling
+)
+from datasets import load_dataset
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+class LayerMasker:
+    def __init__(self, model):
+        self.model = model
+        self.original_forwards = {}  # Store original forward methods
+        
+    def _mask_ffn(self, layer, mask_strength=1.0):
+        """Correctly replace FFN forward propagation"""
+        layer_id = id(layer.mlp)
+        if layer_id not in self.original_forwards:
+            self.original_forwards[layer_id] = layer.mlp.forward  # Store original function
+            
+        def masked_forward(x):
+            return self.original_forwards[layer_id](x) * (1 - mask_strength)
+        
+        layer.mlp.forward = masked_forward  # Override with masked function
+    
+    def _mask_attn(self, layer, mask_strength=1.0):
+        """Correctly replace Attention forward propagation""" 
+        layer_id = id(layer.attn)
+        if layer_id not in self.original_forwards:
+            self.original_forwards[layer_id] = layer.attn.forward  # Store original function
+            
+        def masked_forward(hidden_states, **kwargs):
+            output = self.original_forwards[layer_id](hidden_states, **kwargs)
+            masked_output = (output[0] * (1 - mask_strength),) + output[1:]
+            return masked_output
+        
+        layer.attn.forward = masked_forward  # Override with masked function
+    
+    def mask_layer(self, layer_idx, module_type="ffn", mask_strength=1.0):
+        layer = self.model.transformer.h[layer_idx]
+        if module_type == "ffn":
+            self._mask_ffn(layer, mask_strength)
+        elif module_type == "attn":
+            self._mask_attn(layer, mask_strength)
+        else:
+            raise ValueError(f"Invalid module type: {module_type}")
+    
+    def reset(self):
+        """Restore all modified forward methods"""
+        for layer in self.model.transformer.h:
+            layer_id = id(layer.mlp)
+            if layer_id in self.original_forwards:
+                layer.mlp.forward = self.original_forwards[layer_id]
+                
+            layer_id = id(layer.attn)
+            if layer_id in self.original_forwards:
+                layer.attn.forward = self.original_forwards[layer_id]
+        
+        self.original_forwards.clear()
+
+@torch.no_grad()  # Disable gradient computation
+def get_perplexities(model, eval_dataset, batch_size):
+    """Compute perplexity for the dataset"""
+    data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False)
+
+    training_args = TrainingArguments(
+        output_dir="./tmp_trainer",
+        per_device_eval_batch_size=batch_size,
+        fp16=torch.cuda.is_available(),
+        report_to="none"
+    )
+
+    trainer = Trainer(
+        model=model,
+        args=training_args,
+        eval_dataset=eval_dataset,
+        data_collator=data_collator
+    )
+    
+    eval_results = trainer.evaluate()
+    loss = eval_results['eval_loss']
+    return torch.exp(torch.tensor(loss)).item()
+
+def run_mask_experiment(model, tokenized_data, args):
+    """Run layer masking experiment"""
+    results = []
+    masker = LayerMasker(model)
+    
+    # Compute baseline perplexity
+    baseline_ppl = get_perplexities(model, tokenized_data, args.batch_size)
+    print(f"Baseline Perplexity: {baseline_ppl:.2f}")
+    
+    # Layer masking experiments
+    print("\nRunning layer masking experiments:")
+    for layer_idx in tqdm(range(model.config.n_layer), desc="Layers"):
+        for module_type in ["ffn", "attn"]:
+            # Apply masking
+            masker.mask_layer(layer_idx, module_type)
+            
+            # Compute perplexity
+            ppl = get_perplexities(model, tokenized_data, args.batch_size)
+            delta = ppl - baseline_ppl
+
+            gc.collect()            
+            torch.cuda.empty_cache()  
+            
+            results.append({
+                "layer": layer_idx,
+                "module": module_type,
+                "perplexity": ppl,
+                "delta": delta
+            })
+            
+            # Reset the model
+            masker.reset()
+    
+    return results, baseline_ppl
+
+def visualize_results(results_df, output_dir):
+    """Visualize layer impact results"""
+    plt.figure(figsize=(12, 6))
+    
+    # Plot FFN curve
+    ffn_data = results_df[results_df["module"] == "ffn"]
+    plt.plot(ffn_data["layer"], ffn_data["delta"], 
+             marker='o', linestyle='-', linewidth=2, markersize=8,
+             color='#1f77b4', label='FFN')
+    
+    # Plot Attention curve
+    attn_data = results_df[results_df["module"] == "attn"]
+    plt.plot(attn_data["layer"], attn_data["delta"], 
+             marker='s', linestyle='--', linewidth=2, markersize=8,
+             color='#ff7f0e', label='Attention')
+    
+    plt.xlabel("Layer Index", fontsize=12)
+    plt.ylabel("Δ Perplexity (log scale)", fontsize=12)
+    plt.title("GPT-2 Layer Masking Impact Analysis", fontsize=14)
+    plt.legend(fontsize=10)
+    plt.grid(True, alpha=0.3)
+    plt.xticks(range(0, results_df["layer"].max()+1))
+    
+    # Set y-axis to log scale
+    plt.yscale('log')
+
+    plot_path = os.path.join(output_dir, f"{args.perturbation}_layer_impact.png")
+    plt.savefig(plot_path, bbox_inches='tight', dpi=300)
+    plt.close()
+
+
+if __name__ == "__main__":
+    # Argument configuration
+    parser = argparse.ArgumentParser(description="GPT-2 Layer Masking Experiment")
+    parser.add_argument('--perturbation', type=str, default="hop_sg2pl",
+                       help="Perturbation type (default: hop_sg2pl)")
+    parser.add_argument('--train_set', type=str, default="10M",
+                       help="Training set size (default: 10M)")
+    parser.add_argument('--checkpoint_path', type=str, default="checkpoint-2736",
+                       help="Model checkpoint path (default: checkpoint-2736)")
+    parser.add_argument('--batch_size', type=int, default=3,
+                       help="Evaluation batch size (default: 3)")
+    parser.add_argument('--seed', type=int, default=0,
+                       help="Random seed (default: 0)")
+    args = parser.parse_args()
+
+    # Initialize output directory
+    output_dir = f"mask_results/GPT2-MaskResults"
+    os.makedirs(output_dir, exist_ok=True)
+
+    # Load dataset
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('../train/babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    test_dataset = dataset['test']
+    
+    # Data sampling
+    rng = default_rng(args.seed)
+    indices = rng.choice(len(test_dataset), size=500, replace=False)
+    sampled_test = test_dataset.select(indices)
+
+    # Load model and tokenizer
+    checkpoint_path = f"../train/checkpoints/GPT-2/babylm_{args.perturbation}_10M_seed0/runs/{args.checkpoint_path}"
+    tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
+    model = AutoModelForCausalLM.from_pretrained(checkpoint_path, torch_dtype=torch.float16).eval()
+    if torch.cuda.is_available():
+        model = model.cuda()
+
+    # Data preprocessing
+    def tokenize_fn(examples):
+        return tokenizer(
+            examples["text"],
+            padding="max_length",
+            truncation=True,
+            max_length=512,
+            return_tensors="pt"
+        )
+    
+    tokenized_test = sampled_test.map(
+        tokenize_fn, 
+        batched=True, 
+        remove_columns=["text"],
+        desc="Tokenizing"
+    )
+
+    # Run experiment
+    results, baseline_ppl = run_mask_experiment(model, tokenized_test, args)
+    results_df = pd.DataFrame(results)
+    
+    # Save results
+    output_file = os.path.join(output_dir, f"mask_results_{args.perturbation}.csv")
+    results_df.to_csv(output_file, index=False)
+    
+    # Generate visualization
+    visualize_results(results_df, output_dir)
+    
+    # Print key results
+    print(f"\nExperiment completed. Results saved to {output_dir}")
+    print(f"Baseline Perplexity: {baseline_ppl:.2f}")
+    print("Layer impact results:")
+    print(results_df[["layer", "module", "delta"]].to_string(index=False))
+
diff --git a/perplexities/gpt2_mask_test.py b/perplexities/gpt2_mask_test.py
new file mode 100644
index 0000000000000000000000000000000000000000..e69de29bb2d1d6434b8b29ae775ad8c2e48c5391
diff --git a/perplexities/gpt2_noise.py b/perplexities/gpt2_noise.py
new file mode 100644
index 0000000000000000000000000000000000000000..5f8be51bea34770be3caac1f1b7b8385a04269ab
--- /dev/null
+++ b/perplexities/gpt2_noise.py
@@ -0,0 +1,203 @@
+# Activation Steering
+
+import torch
+import sys
+import argparse
+import os
+import pandas as pd
+import matplotlib.pyplot as plt
+from tqdm import tqdm
+from numpy.random import default_rng
+sys.path.append("..")
+from transformers import (
+    AutoTokenizer, 
+    AutoModelForCausalLM,
+    Trainer, 
+    TrainingArguments, 
+    DataCollatorForLanguageModeling
+)
+from datasets import load_dataset
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+class LayerNoiseInjector:
+
+    def __init__(self, model):
+        self.model = model
+        self.hooks = []  # Store registered hooks
+        self.noise_std = 0.1  # Noise standard deviation (adjustable)
+    
+    
+    def _add_noise_hook(self, layer_idx):
+        def hook(module, input, output):
+            if isinstance(output, tuple):
+                modified_output = list(output)
+                noise = torch.randn_like(modified_output[0]) * self.noise_std
+                modified_output[0] = modified_output[0] + noise
+                return tuple(modified_output)
+            else:
+                noise = torch.randn_like(output) * self.noise_std
+                return output + noise
+        
+        handle = self.model.transformer.h[layer_idx].register_forward_hook(hook)
+        self.hooks.append(handle)
+        
+    
+    def inject_noise(self, layer_idx):
+        """Inject noise to specified layer"""
+        self._add_noise_hook(layer_idx)
+    
+    def reset(self):
+        """Remove all noise hooks"""
+        for handle in self.hooks:
+            handle.remove()
+        self.hooks.clear()
+
+@torch.no_grad()  # Disable gradient calculation
+def get_perplexities(model, eval_dataset, batch_size):
+    """Calculate dataset perplexity"""
+    data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False)
+
+    training_args = TrainingArguments(
+        output_dir="./tmp_trainer",
+        per_device_eval_batch_size=batch_size,
+        fp16=torch.cuda.is_available(),
+        report_to="none"
+    )
+
+    trainer = Trainer(
+        model=model,
+        args=training_args,
+        eval_dataset=eval_dataset,
+        data_collator=data_collator
+    )
+    
+    eval_results = trainer.evaluate()
+    loss = eval_results['eval_loss']
+    return torch.exp(torch.tensor(loss)).item()
+
+# Data preprocessing
+def tokenize_fn(examples):
+    return tokenizer(
+        examples["text"],
+        padding="max_length",
+        truncation=True,
+        max_length=512,
+        return_tensors="pt"
+    )
+
+
+def run_noise_experiment(model, tokenized_data, args):
+    """Execute layer-wise noise injection experiment"""
+    results = []
+    noise_injector = LayerNoiseInjector(model)
+    
+    # Calculate baseline perplexity
+    baseline_ppl = get_perplexities(model, tokenized_data, args.batch_size)
+    print(f"Baseline Perplexity: {baseline_ppl:.2f}")
+    
+    # Layer-wise noise injection experiment
+    print("\nRunning layer noise injection experiments:")
+    for layer_idx in tqdm(range(model.config.n_layer), desc="Layers"):
+        # Inject noise
+        noise_injector.inject_noise(layer_idx)
+        
+        # Calculate perplexity
+        ppl = get_perplexities(model, tokenized_data, args.batch_size)
+        delta = ppl - baseline_ppl
+        
+        results.append({
+            "layer": layer_idx,
+            "perplexity": ppl,
+            "delta": delta
+        })
+        
+        # Reset model
+        noise_injector.reset()
+    
+    return results, baseline_ppl
+
+def visualize_noise_results(results_df, output_dir):
+    """Visualize noise injection impact results"""
+    plt.figure(figsize=(12, 6))
+    
+    # Plot delta perplexity curve
+    plt.plot(results_df["layer"], results_df["delta"], 
+             marker='o', linestyle='-', linewidth=2, markersize=8,
+             color='#2ca02c', label='Activation Noise')
+    
+    plt.xlabel("Layer Index", fontsize=12)
+    plt.ylabel("Δ Perplexity (log scale)", fontsize=12)
+    plt.title("GPT-2 Layer Noise Injection Impact Analysis", fontsize=14)
+    plt.legend(fontsize=10)
+    plt.grid(True, alpha=0.3)
+    plt.xticks(range(0, results_df["layer"].max()+1))
+    
+    # Set y-axis to log scale
+    plt.yscale('log')
+
+    plot_path = os.path.join(output_dir, f"{args.perturbation}_noise_impact.png")
+    plt.savefig(plot_path, bbox_inches='tight', dpi=300)
+    plt.close()
+
+if __name__ == "__main__":
+    # Parameter configuration
+    parser = argparse.ArgumentParser(description="GPT-2 Layer Noise Injection Experiment")
+    parser.add_argument('--perturbation', type=str, default="hop_sg2pl",
+                       help="Perturbation type (default: hop_sg2pl)")
+    parser.add_argument('--train_set', type=str, default="10M",
+                       help="Training set size (default: 10M)")
+    parser.add_argument('--checkpoint_path', type=str, default="checkpoint-2736",
+                       help="Model checkpoint path (default: checkpoint-2736)")
+    parser.add_argument('--batch_size', type=int, default=3,
+                       help="Evaluation batch size (default: 3)")
+    parser.add_argument('--seed', type=int, default=0,
+                       help="Random seed (default: 0)")
+    parser.add_argument('--noise_std', type=float, default=0.1,
+                       help="Noise standard deviation (default: 0.1)")
+    args = parser.parse_args()
+
+    # Initialize output directory
+    output_dir = f"noise_results/GPT2-NoiseResults"
+    os.makedirs(output_dir, exist_ok=True)
+
+    # Load dataset
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('../train/babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    test_dataset = dataset['test']
+    
+    # Data sampling
+    rng = default_rng(args.seed)
+    indices = rng.choice(len(test_dataset), size=500, replace=False)
+    sampled_test = test_dataset.select(indices)
+
+    # Load model and tokenizer
+    checkpoint_path = f"../train/checkpoints/GPT-2/babylm_{args.perturbation}_10M_seed0/runs/{args.checkpoint_path}"
+    tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
+    model = AutoModelForCausalLM.from_pretrained(checkpoint_path, torch_dtype=torch.float16).eval()
+    if torch.cuda.is_available():
+        model = model.cuda()
+    
+    tokenized_test = sampled_test.map(
+        tokenize_fn, 
+        batched=True, 
+        remove_columns=["text"],
+        desc="Tokenizing"
+    )
+
+    # Run noise injection experiment
+    results, baseline_ppl = run_noise_experiment(model, tokenized_test, args)
+    results_df = pd.DataFrame(results)
+    
+    # Save results
+    output_file = os.path.join(output_dir, f"noise_results_{args.perturbation}.csv")
+    results_df.to_csv(output_file, index=False)
+    
+    # Generate visualization
+    visualize_noise_results(results_df, output_dir)
+    
+    # Print key results
+    print(f"\nExperiment completed. Results saved to {output_dir}")
+    print(f"Baseline Perplexity: {baseline_ppl:.2f}")
+    print("Layer impact results:")
+    print(results_df[["layer", "delta"]].to_string(index=False))
diff --git a/perplexities/gpt2_remove_layers.py b/perplexities/gpt2_remove_layers.py
new file mode 100644
index 0000000000000000000000000000000000000000..940baf8befa628498512a3d8fe3568c6784358d5
--- /dev/null
+++ b/perplexities/gpt2_remove_layers.py
@@ -0,0 +1,141 @@
+import torch
+import sys
+import argparse
+import os
+sys.path.append("..")
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from datasets import load_dataset
+from numpy.random import default_rng
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+# MODEL_NAME = "Llama-3.2-3B"
+# ori_model_name = "meta-llama/Llama-3.2-3B"
+# MODEL_NAME_SAVE = "Llama-3.2-3B-500-Remove"
+
+MODEL_NAME = "GPT-2"
+ori_model_name = "gpt2"
+MODEL_NAME_SAVE = "GPT2-500-Remove-layers"
+
+FILE_SAMPLE_SIZE = 500
+
+
+def remove_layers(original_model, exp_model, i, j):
+
+    original_blocks = original_model.transformer.h
+    exp_blocks = exp_model.transformer.h
+
+    print("len(original_blocks):", len(original_blocks))
+    print("len(exp_blocks):", len(exp_blocks))
+
+    # Ensure the indices are valid
+    if i < 0 or (i + j) > len(exp_blocks) or j < 0:
+        raise ValueError(f"Invalid block indices: i={i}, i+j={i+j}. Must satisfy 0 <= i and i+j <= {len(exp_blocks)}.")
+
+    # Replace the parameters of the blocks from i to i+j
+    for idx in range(i, i + j):
+        print(f"Replacing parameters of Transformer block {idx}...")
+        original_block = original_blocks[idx]
+        exp_blocks[idx].load_state_dict(original_block.state_dict())
+
+    return exp_model
+
+
+
+def get_perplexities(model, eval_dataset, batch_size):
+    data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False)
+
+    training_args = TrainingArguments(
+        output_dir="./tmp_trainer",
+        per_device_eval_batch_size=batch_size,
+        fp16=True,
+        report_to="none"
+    )
+
+    trainer = Trainer(model=model, args=training_args, eval_dataset=eval_dataset, data_collator=data_collator)
+    eval_results = trainer.evaluate()
+    print("eval_results:", eval_results)  
+    loss = eval_results['eval_loss']
+    perplexity = torch.exp(torch.tensor(loss)).item()
+    return perplexity
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Calculate perplexity on test dataset.")
+
+    parser.add_argument('perturbation', 
+                        type=str, 
+                        default='reverse_full', 
+                        nargs='?', 
+                        help='Type of perturbation to use.')
+    parser.add_argument('train_set', 
+                        type=str, 
+                        default='test', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('checkpoint_path', 
+                        type=str, 
+                        default='checkpoint-100', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('batch_size', 
+                        type=int, 
+                        default=4, 
+                        nargs='?', 
+                        help='Batch size for evaluation.')
+    parser.add_argument('seed', 
+                        type=int, 
+                        default=0, 
+                        nargs='?', 
+                        help='Random seed.')
+    parser.add_argument('remove_layer', 
+                        type=int, 
+                        default=1, 
+                        nargs='?', 
+                        help='Layer index to remove')
+    args = parser.parse_args()
+
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('../train/babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    test_dataset = dataset['test']  # Load test dataset
+    print(test_dataset)
+    
+    
+    checkpoint_path = f'../train/checkpoints/{MODEL_NAME}/babylm_{args.perturbation}_10M_seed0/runs/{args.checkpoint_path}'
+    
+    rng = default_rng(args.seed)
+    indices = rng.choice(len(test_dataset), FILE_SAMPLE_SIZE, replace=False)
+    sampled_test_dataset = test_dataset.select(indices)
+
+    tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
+    model = AutoModelForCausalLM.from_pretrained(checkpoint_path)
+    model_ori = AutoModelForCausalLM.from_pretrained(ori_model_name)
+    model = remove_layers(model_ori, model, args.remove_layer, 4)
+    print(model)
+    model.eval()
+    # if torch.cuda.is_available():
+    #     model.to('cuda')
+
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+
+    tokenized_test = sampled_test_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+
+    perplexity = get_perplexities(model, tokenized_test, args.batch_size)
+    
+    # Save the result to the specified directory
+    output_directory = f"perplexities_results/{MODEL_NAME_SAVE}"
+    os.makedirs(output_directory, exist_ok=True)
+
+    # Construct the output file path
+    output_file = os.path.join(output_directory, f"{args.perturbation}_{args.batch_size}_{args.seed}.csv")
+
+    # Write the header to the CSV file if it doesn't exist
+    if not os.path.exists(output_file):
+        with open(output_file, 'w') as f:
+            print("Writing header to CSV...")
+            f.write(f"checkpoint_path, perplexity\n")
+
+    # Append the perplexity result to the CSV file
+    with open(output_file, 'a') as f:  # Open in append mode
+        print("Appending result to CSV...")
+        f.write(f"{args.checkpoint_path}-{args.remove_layer+1}, {perplexity}\n")
\ No newline at end of file
diff --git a/perplexities/gpt_mask.sh b/perplexities/gpt_mask.sh
new file mode 100644
index 0000000000000000000000000000000000000000..d47161c88bc5e3d5b4bb3d6b5a3c96fca0d77ae5
--- /dev/null
+++ b/perplexities/gpt_mask.sh
@@ -0,0 +1,9 @@
+#!/bin/bash
+
+CUDA_VISIBLE_DEVICES=1,2 torchrun --nproc_per_node=2 --master_port=22235 gpt2_mask.py --perturbation shuffle_nondeterministic --train_set 10M --checkpoint_path checkpoint-2736 --batch_size 3 --seed 0
+
+CUDA_VISIBLE_DEVICES=1,2 torchrun --nproc_per_node=2 --master_port=22235 gpt2_mask.py --perturbation shuffle_nondeterministic --train_set 10M --checkpoint_path checkpoint-2736 --batch_size 3 --seed 0
+
+CUDA_VISIBLE_DEVICES=1,2 torchrun --nproc_per_node=2 --master_port=22235 gpt2_mask.py --perturbation hop_words4 --train_set 10M --checkpoint_path checkpoint-2376 --batch_size 3 --seed 0
+
+CUDA_VISIBLE_DEVICES=1,2 torchrun --nproc_per_node=2 --master_port=22235 gpt2_mask.py --perturbation hop_control --train_set 10M --checkpoint_path checkpoint-2409 --batch_size 3 --seed 0
diff --git a/perplexities/gpt_noise.sh b/perplexities/gpt_noise.sh
new file mode 100644
index 0000000000000000000000000000000000000000..279189579aa5f88ed4154f4fc44373222e7507a2
--- /dev/null
+++ b/perplexities/gpt_noise.sh
@@ -0,0 +1,7 @@
+CUDA_VISIBLE_DEVICES=2,3 torchrun --nproc_per_node=2 --master_port=25235 gpt2_noise.py --perturbation shuffle_nondeterministic --train_set 10M --checkpoint_path checkpoint-2736 --batch_size 3 --seed 0 --noise_std 0.1
+
+CUDA_VISIBLE_DEVICES=2,3 torchrun --nproc_per_node=2 --master_port=25236 gpt2_noise.py --perturbation shuffle_control --train_set 10M --checkpoint_path checkpoint-2736 --batch_size 3 --seed 0 --noise_std 0.1
+
+CUDA_VISIBLE_DEVICES=2,3 torchrun --nproc_per_node=2 --master_port=25237 gpt2_noise.py --perturbation hop_words4 --train_set 10M --checkpoint_path checkpoint-2376 --batch_size 3 --seed 0 --noise_std 0.1
+
+CUDA_VISIBLE_DEVICES=2,3 torchrun --nproc_per_node=2 --master_port=25238 gpt2_noise.py --perturbation hop_control --train_set 10M --checkpoint_path checkpoint-2409 --batch_size 3 --seed 0 --noise_std 0.1
\ No newline at end of file
diff --git a/perplexities/llama3.2_remove_layers.py b/perplexities/llama3.2_remove_layers.py
new file mode 100644
index 0000000000000000000000000000000000000000..362c41708eb2a1e3838f550b6f35c13ca3e306ee
--- /dev/null
+++ b/perplexities/llama3.2_remove_layers.py
@@ -0,0 +1,135 @@
+import torch
+import sys
+import argparse
+import os
+sys.path.append("..")
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from datasets import load_dataset
+from numpy.random import default_rng
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+MODEL_NAME = "Llama-3.2-3B"
+ori_model_name = "meta-llama/Llama-3.2-3B"
+MODEL_NAME_SAVE = "Llama-3.2-3B-500-Remove-layers"
+
+
+FILE_SAMPLE_SIZE = 500
+
+
+def remove_layers(original_model, exp_model, i, j):
+
+    original_layers = original_model.model.layers
+    exp_layers = exp_model.model.layers
+
+    print("len(layers):", len(original_layers))
+    # Ensure the indices are valid
+    if i < 0 or (i + j) > len(exp_layers) or j < 0:
+        raise ValueError(f"Invalid layer indices: i={i}, i+j={i+j}. Must satisfy 0 <= i and i+j <= {len(exp_layers)}.")
+
+    # Replace the parameters of the layers from i to i+j
+    for idx in range(i, i + j):
+        print(f"Replacing parameters of LlamaDecoderLayer {idx}...")
+        original_layer = original_layers[idx]
+        exp_layers[idx].load_state_dict(original_layer.state_dict())
+
+    return exp_model
+
+
+def get_perplexities(model, eval_dataset, batch_size):
+    data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False)
+
+    training_args = TrainingArguments(
+        output_dir="./tmp_trainer",
+        per_device_eval_batch_size=batch_size,
+        fp16=True,
+        report_to="none"
+    )
+
+    trainer = Trainer(model=model, args=training_args, eval_dataset=eval_dataset, data_collator=data_collator)
+    eval_results = trainer.evaluate()
+    print("eval_results:", eval_results)  
+    loss = eval_results['eval_loss']
+    perplexity = torch.exp(torch.tensor(loss)).item()
+    return perplexity
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Calculate perplexity on test dataset.")
+
+    parser.add_argument('perturbation', 
+                        type=str, 
+                        default='reverse_full', 
+                        nargs='?', 
+                        help='Type of perturbation to use.')
+    parser.add_argument('train_set', 
+                        type=str, 
+                        default='test', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('checkpoint_path', 
+                        type=str, 
+                        default='checkpoint-100', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('batch_size', 
+                        type=int, 
+                        default=4, 
+                        nargs='?', 
+                        help='Batch size for evaluation.')
+    parser.add_argument('seed', 
+                        type=int, 
+                        default=0, 
+                        nargs='?', 
+                        help='Random seed.')
+    parser.add_argument('remove_layer', 
+                        type=int, 
+                        default=1, 
+                        nargs='?', 
+                        help='Layer index to remove')
+    args = parser.parse_args()
+
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('../train/babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    test_dataset = dataset['test']  # Load test dataset
+    print(test_dataset)
+    
+    
+    checkpoint_path = f'../train/checkpoints/{MODEL_NAME}/babylm_{args.perturbation}_10M_seed0/runs/{args.checkpoint_path}'
+    
+    rng = default_rng(args.seed)
+    indices = rng.choice(len(test_dataset), FILE_SAMPLE_SIZE, replace=False)
+    sampled_test_dataset = test_dataset.select(indices)
+
+    tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
+    model = AutoModelForCausalLM.from_pretrained(checkpoint_path)
+    model_ori = AutoModelForCausalLM.from_pretrained(ori_model_name)
+    model = remove_layers(model_ori, model, args.remove_layer, 9)
+    print(model)
+    model.eval()
+    # if torch.cuda.is_available():
+    #     model.to('cuda')
+
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+
+    tokenized_test = sampled_test_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+
+    perplexity = get_perplexities(model, tokenized_test, args.batch_size)
+    
+    # Save the result to the specified directory
+    output_directory = f"perplexities_results/{MODEL_NAME_SAVE}"
+    os.makedirs(output_directory, exist_ok=True)
+
+    # Construct the output file path
+    output_file = os.path.join(output_directory, f"{args.perturbation}_{args.batch_size}_{args.seed}.csv")
+
+    # Write the header to the CSV file if it doesn't exist
+    if not os.path.exists(output_file):
+        with open(output_file, 'w') as f:
+            print("Writing header to CSV...")
+            f.write(f"checkpoint_path, perplexity\n")
+
+    # Append the perplexity result to the CSV file
+    with open(output_file, 'a') as f:  # Open in append mode
+        print("Appending result to CSV...")
+        f.write(f"{args.checkpoint_path}-{args.remove_layer+1}, {perplexity}\n")
\ No newline at end of file
diff --git a/perplexities/perplexities.py b/perplexities/perplexities.py
new file mode 100644
index 0000000000000000000000000000000000000000..03370eb3c5008f1953e3e1df00d319fb4d76c1bc
--- /dev/null
+++ b/perplexities/perplexities.py
@@ -0,0 +1,179 @@
+# perplexities.py
+# Author: Julie Kallini
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+from transformers import GPT2LMHeadModel
+from gpt2_no_positional_encoding_model import GPT2NoPositionalEncodingLMHeadModel
+from utils import CHECKPOINT_READ_PATH, PERTURBATIONS, BABYLM_DATA_PATH, \
+    PAREN_MODELS, gpt2_original_tokenizer
+from tqdm import tqdm
+from glob import glob
+from numpy.random import default_rng
+import pandas as pd
+import torch
+import itertools
+import argparse
+import os
+
+
+MAX_TRAINING_STEPS = 3000
+CHECKPOINTS = list(range(100, MAX_TRAINING_STEPS+1, 100))
+
+
+def create_attention_mask(token_lists):
+    seq_length = max([len(i) for i in token_lists])
+    batch_size = len(token_lists)
+    mask = torch.full((batch_size, seq_length), 0)
+
+    for i, tokens in enumerate(token_lists):
+        mask[i, 0:len(tokens)] = 1
+
+    return mask
+
+def create_input_ids(token_lists, pad_token_id):
+    padded = zip(*itertools.zip_longest(*token_lists, fillvalue=pad_token_id))
+    return torch.tensor(list(padded))
+
+
+def get_perplexities(model, token_lists, pad_token_id, device="cuda"):
+
+    # Prepare data
+    input_ids = create_input_ids(token_lists, pad_token_id).to(device)
+    labels = input_ids.clone()  # GPT-2 uses input as labels for CLM task
+    attention_mask = create_attention_mask(token_lists).to(device)
+
+    # Forward pass
+    outputs = model(input_ids=input_ids, labels=labels,
+                    attention_mask=attention_mask)
+
+    # The "shifted" nature of labels in GPT-2 (next token prediction)
+    # Shift logits, labels, and attention mask by one position
+    shift_logits = outputs.logits[..., :-1, :].contiguous()
+    shift_labels = labels[..., 1:].contiguous()
+    shift_attention_mask = attention_mask[..., 1:].contiguous()
+
+    # Instantiate loss function with no reduction
+    loss_fct = torch.nn.CrossEntropyLoss(reduction='none')
+
+    # Calculate per-token loss
+    loss = loss_fct(shift_logits.view(-1, shift_logits.size(-1)),
+                    shift_labels.view(-1))
+
+    # Reshape back to the original batch size and sequence length
+    loss = loss.view(shift_labels.size())
+
+    # Apply the attention mask - only calculate loss where mask is 1
+    loss = loss * shift_attention_mask
+
+    # Sum the loss over the sequence length, get per-example perplexity
+    per_example_loss = loss.sum(dim=1) / shift_attention_mask.sum(dim=1)
+    return torch.exp(per_example_loss).tolist()
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(
+        prog='Edge probing',
+        description='Edge probing experiments')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('test_perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform test BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    parser.add_argument('random_seed', type=int, help="Random seed")
+    parser.add_argument('paren_model',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=list(PAREN_MODELS.keys()) + ["randinit"],
+                        help='Parenthesis model')
+    parser.add_argument('-np', '--no_pos_encodings', action='store_true',
+                        help="Train GPT-2 with no positional encodings")
+
+    # Get args
+    args = parser.parse_args()
+    no_pos_encodings_underscore = "_no_positional_encodings" if args.no_pos_encodings else ""
+
+    # Get path to model
+    model = f"babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}_seed{args.random_seed}"
+    model_path = f"{CHECKPOINT_READ_PATH}/babylm_{args.perturbation_type}_{args.train_set}_{args.paren_model}{no_pos_encodings_underscore}/{model}/runs/{model}/checkpoint-"
+
+    # Get perturbed test files
+    test_files = sorted(glob(
+        f"{BABYLM_DATA_PATH}/babylm_data_perturbed/babylm_{args.test_perturbation_type}/babylm_test_affected/*"))
+
+    FILE_SAMPLE_SIZE = 1000
+    rng = default_rng(args.random_seed)
+
+    # Iterate over data files to get perplexity data
+    print("Sampling BabyLM affected test files to extract surprisals...")
+    token_sequences = []
+    for test_file in test_files:
+        print(test_file)
+
+        # Get tokens from test file and subsample
+        f = open(test_file, 'r')
+        file_token_sequences = [
+            [int(s) for s in l.split()] for l in f.readlines()]
+        sample_indices = rng.choice(
+            list(range(len(file_token_sequences))), FILE_SAMPLE_SIZE, replace=False)
+        file_token_sequences = [file_token_sequences[i]
+                                for i in sample_indices]
+        token_sequences.extend(file_token_sequences)
+
+    # For logging/debugging, include decoded sentence
+    test_sents = [gpt2_original_tokenizer.decode(
+        toks) for toks in token_sequences]
+
+    ppl_df = pd.DataFrame({
+        "Sentences": test_sents
+    })
+
+    BATCH_SIZE = 8
+    device = "cuda"
+    for ckpt in CHECKPOINTS:
+        print(f"Checkpoint: {ckpt}")
+
+        # Load model
+        if args.no_pos_encodings:
+            model = GPT2NoPositionalEncodingLMHeadModel.from_pretrained(
+                model_path + str(ckpt)).to(device)
+        else:
+            model = GPT2LMHeadModel.from_pretrained(
+            model_path + str(ckpt)).to(device)
+
+        # Get perplexities
+        perplexities = []
+        for i in tqdm(range(0, len(token_sequences), BATCH_SIZE)):
+            batch = token_sequences[i:i+BATCH_SIZE]
+            ppls = get_perplexities(
+                model, batch, gpt2_original_tokenizer.eos_token_id)
+            perplexities.extend(ppls)
+
+        # Add ppls to df
+        ppl_df[f'Perplexities (ckpt {ckpt})'] = perplexities
+
+    # Write results to CSV
+    directory = f"perplexity_results/{args.perturbation_type}_{args.train_set}{no_pos_encodings_underscore}"
+    if not os.path.exists(directory):
+        os.makedirs(directory)
+    file = directory + \
+        f"/{args.paren_model}_seed{args.random_seed}_test_{args.test_perturbation_type}.csv"
+    print(f"Writing results to CSV: {file}")
+    ppl_df.to_csv(file)
\ No newline at end of file
diff --git a/perplexities/perplexities_llama.py b/perplexities/perplexities_llama.py
new file mode 100644
index 0000000000000000000000000000000000000000..61069d13225b089594064e2f44512ecb48bada5c
--- /dev/null
+++ b/perplexities/perplexities_llama.py
@@ -0,0 +1,108 @@
+import torch
+import sys
+import argparse
+import os
+sys.path.append("..")
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from datasets import load_dataset
+from numpy.random import default_rng
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+# MODEL_NAME = "Llama-3.2-1B"
+# MODEL_NAME = "Llama-3.2-1B-500"
+
+MODEL_NAME = "GPT-2"
+MODEL_NAME_SAVE = "GPT2-500"
+FILE_SAMPLE_SIZE = 500
+
+def get_perplexities(model, eval_dataset, batch_size):
+    data_collator = DataCollatorForLanguageModeling(tokenizer=tokenizer, mlm=False)
+
+    training_args = TrainingArguments(
+        output_dir="./tmp_trainer",
+        per_device_eval_batch_size=batch_size,
+        fp16=True,
+        report_to="none"
+    )
+
+    trainer = Trainer(model=model, args=training_args, eval_dataset=eval_dataset, data_collator=data_collator)
+    eval_results = trainer.evaluate()
+    print("eval_results:", eval_results)  
+    loss = eval_results['eval_loss']
+    perplexity = torch.exp(torch.tensor(loss)).item()
+    return perplexity
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Calculate perplexity on test dataset.")
+
+    parser.add_argument('perturbation', 
+                        type=str, 
+                        default='reverse_full', 
+                        nargs='?', 
+                        help='Type of perturbation to use.')
+    parser.add_argument('train_set', 
+                        type=str, 
+                        default='test', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('checkpoint_path', 
+                        type=str, 
+                        default='checkpoint-100', 
+                        nargs='?', 
+                        help='Dataset size for training.')
+    parser.add_argument('batch_size', 
+                        type=int, 
+                        default=4, 
+                        nargs='?', 
+                        help='Batch size for evaluation.')
+    parser.add_argument('seed', 
+                        type=int, 
+                        default=0, 
+                        nargs='?', 
+                        help='Random seed.')
+    args = parser.parse_args()
+
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('../train/babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    test_dataset = dataset['test']  # Load test dataset
+    print(test_dataset)
+    
+    
+    checkpoint_path = f'../train/checkpoints/{MODEL_NAME}/babylm_{args.perturbation}_10M_seed0/runs/{args.checkpoint_path}'
+    
+    rng = default_rng(args.seed)
+    indices = rng.choice(len(test_dataset), FILE_SAMPLE_SIZE, replace=False)
+    sampled_test_dataset = test_dataset.select(indices)
+
+    tokenizer = AutoTokenizer.from_pretrained(checkpoint_path)
+    model = AutoModelForCausalLM.from_pretrained(checkpoint_path)
+    
+    model.eval()
+    if torch.cuda.is_available():
+        model.to('cuda')
+
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+
+    tokenized_test = sampled_test_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+
+    perplexity = get_perplexities(model, tokenized_test, args.batch_size)
+    
+    # Save the result to the specified directory
+    output_directory = f"perplexities_results/{MODEL_NAME_SAVE}"
+    os.makedirs(output_directory, exist_ok=True)
+
+    # Construct the output file path
+    output_file = os.path.join(output_directory, f"{args.perturbation}_{args.batch_size}_{args.seed}.csv")
+
+    # Write the header to the CSV file if it doesn't exist
+    if not os.path.exists(output_file):
+        with open(output_file, 'w') as f:
+            print("Writing header to CSV...")
+            f.write("checkpoint_path, perplexity\n")
+
+    # Append the perplexity result to the CSV file
+    with open(output_file, 'a') as f:  # Open in append mode
+        print("Appending result to CSV...")
+        f.write(f"{args.checkpoint_path}, {perplexity}\n")
\ No newline at end of file
diff --git a/requirements.txt b/requirements.txt
new file mode 100644
index 0000000000000000000000000000000000000000..88d29f0d6a5aba486c7827902871839087b7cf01
--- /dev/null
+++ b/requirements.txt
@@ -0,0 +1,94 @@
+certifi==2023.11.17
+charset-normalizer==3.3.2
+click==8.1.7
+contourpy==1.1.1
+cycler==0.12.1
+data==0.4
+emoji==2.8.0
+fonttools==4.45.1
+fsspec==2023.10.0
+funcsigs==1.0.2
+future==0.18.3
+huggingface-hub==0.19.4
+idna==3.6
+iniconfig==2.0.0
+joblib==1.3.2
+kiwisolver==1.4.5
+latex==0.7.0
+matplotlib==3.7.5
+mizani==0.9.3
+nltk==3.8.1
+numpy==1.24.4
+pandas==2.0.3
+patsy==0.5.3
+Pillow==10.1.0
+plotnine==0.12.4
+pluggy==1.3.0
+pluralizer==1.2.0
+protobuf==4.25.1
+pyparsing==3.1.1
+pytest==7.4.3
+pytz==2023.3.post1
+PyYAML==6.0.1
+regex==2023.10.3
+requests==2.31.0
+safetensors==0.4.1
+scikit-learn==1.3.2
+scipy==1.10.1
+seaborn==0.13.0
+shutilwhich==1.1.0
+stanza==1.6.1
+statsmodels==0.14.0
+tempdir==0.7.1
+threadpoolctl==3.2.0
+tokenizers==0.15.0
+tomli==2.0.1
+torch==2.0.0
+tqdm==4.66.1
+transformers==4.35.2
+triton==2.0.0
+tzdata==2023.3
+urllib3==2.1.0
+asttokens==2.0.5
+comm==0.1.2
+debugpy==1.6.7
+decorator==5.1.1
+exceptiongroup==1.1.0
+executing==1.2.0
+filelock==3.12.2
+gmpy2==2.1.0
+importlib-metadata==6.0.0
+ipykernel==6.23.1
+ipython==8.0.0
+jedi==0.18.2
+Jinja2==3.1.2
+jupyter_client==8.1.0
+jupyter_core==5.1.0
+latexcodec==1.0.0
+MarkupSafe==2.1.2
+matplotlib-inline==0.1.6
+mpmath==1.2.1
+nest-asyncio==1.5.6
+networkx==2.8.6
+packaging==23.0
+parso==0.8.3
+pexpect==4.8.0
+pickleshare==0.7.5
+platformdirs==2.5.2
+prompt-toolkit==3.0.30
+psutil==5.9.1
+ptyprocess==0.7.0
+pure-eval==0.2.2
+Pygments==2.15.0
+python-dateutil==2.8.2
+pyzmq==23.0.0
+six==1.16.0
+stack-data==0.6.0
+sympy==1.11.1
+tornado==6.2
+traitlets==5.7.1
+typing_extensions==4.6.0
+wcwidth==0.2.5
+zipp==3.12.0
+
+
diff --git a/requirements_1.txt b/requirements_1.txt
new file mode 100644
index 0000000000000000000000000000000000000000..ccbaaeab86bca93431ef12067674b57ac197cdbf
--- /dev/null
+++ b/requirements_1.txt
@@ -0,0 +1,51 @@
+certifi==2023.11.17
+charset-normalizer==3.3.2
+click==8.1.7
+contourpy==1.2.0
+cycler==0.12.1
+data==0.4
+emoji==2.8.0
+fonttools==4.45.1
+fsspec==2023.10.0
+funcsigs==1.0.2
+future==0.18.3
+huggingface-hub==0.19.4
+idna==3.6
+iniconfig==2.0.0
+joblib==1.3.2
+kiwisolver==1.4.5
+latex==0.7.0
+matplotlib==3.8.2
+mizani==0.9.3
+nltk==3.8.1
+numpy==1.26.2
+pandas==2.1.3
+patsy==0.5.3
+Pillow==10.1.0
+plotnine==0.12.4
+pluggy==1.3.0
+pluralizer==1.2.0
+protobuf==4.25.1
+pyparsing==3.1.1
+pytest==7.4.3
+pytz==2023.3.post1
+PyYAML==6.0.1
+regex==2023.10.3
+requests==2.31.0
+safetensors==0.4.1
+scikit-learn==1.3.2
+scipy==1.11.4
+seaborn==0.13.0
+shutilwhich==1.1.0
+stanza==1.6.1
+statsmodels==0.14.0
+tempdir==0.7.1
+threadpoolctl==3.2.0
+tokenizers==0.15.0
+tomli==2.0.1
+torch==2.0.0
+tqdm==4.66.1
+transformers==4.35.2
+triton==2.0.0
+tzdata==2023.3
+urllib3==2.1.0
\ No newline at end of file
diff --git a/requirements_2.txt b/requirements_2.txt
new file mode 100644
index 0000000000000000000000000000000000000000..567aeabc1269ca616977f2d5ac46225704882cb5
--- /dev/null
+++ b/requirements_2.txt
@@ -0,0 +1,82 @@
+asttokens==2.2.1
+comm==0.1.2
+debugpy==1.6.7
+decorator==5.1.1
+exceptiongroup==1.1.0
+executing==1.2.0
+filelock==3.12.2
+gmpy2==2.1.0
+importlib-metadata==6.0.0
+ipykernel==6.23.1
+ipython==8.0.0
+jedi==0.18.2
+Jinja2==3.1.2
+jupyter_client==8.1.0
+jupyter_core==5.1.0
+latexcodec==1.0.0
+MarkupSafe==2.1.2
+matplotlib-inline==0.1.6
+mpmath==1.2.1
+nest-asyncio==1.5.6
+networkx==2.8.6
+packaging==23.0
+parso==0.8.3
+pexpect==4.8.0
+pickleshare==0.7.5
+platformdirs==2.5.2
+prompt-toolkit==3.0.30
+psutil==5.9.1
+ptyprocess==0.7.0
+pure-eval==0.2.2
+Pygments==2.15.0
+python-dateutil==2.8.2
+pyzmq==23.0.0
+six==1.16.0
+stack-data==0.6.0
+sympy==1.11.1
+tornado==6.2
+traitlets==5.7.1
+typing_extensions==4.6.0
+wcwidth==0.2.5
+zipp==3.12.0
+# asttokens @ file:///home/conda/feedstock_root/build_artifacts/asttokens_1698341106958/work
+# comm @ file:///home/conda/feedstock_root/build_artifacts/comm_1691044910542/work
+# debugpy @ file:///croot/debugpy_1690905042057/work
+# decorator @ file:///home/conda/feedstock_root/build_artifacts/decorator_1641555617451/work
+# exceptiongroup @ file:///home/conda/feedstock_root/build_artifacts/exceptiongroup_1700579780973/work
+# executing @ file:///home/conda/feedstock_root/build_artifacts/executing_1698579936712/work
+# filelock @ file:///croot/filelock_1700591183607/work
+# gmpy2 @ file:///tmp/build/80754af9/gmpy2_1645455533097/work
+# importlib-metadata @ file:///home/conda/feedstock_root/build_artifacts/importlib-metadata_1688754491823/work
+# ipykernel @ file:///home/conda/feedstock_root/build_artifacts/ipykernel_1698244021190/work
+# ipython @ file:///home/conda/feedstock_root/build_artifacts/ipython_1701092366260/work
+# jedi @ file:///home/conda/feedstock_root/build_artifacts/jedi_1696326070614/work
+# Jinja2 @ file:///croot/jinja2_1666908132255/work
+# jupyter_client @ file:///home/conda/feedstock_root/build_artifacts/jupyter_client_1699283905679/work
+# jupyter_core @ file:///home/conda/feedstock_root/build_artifacts/jupyter_core_1698673647019/work
+# latexcodec @ file:///home/conda/feedstock_root/build_artifacts/latexcodec_1592937263153/work
+# MarkupSafe @ file:///opt/conda/conda-bld/markupsafe_1654597864307/work
+# matplotlib-inline @ file:///home/conda/feedstock_root/build_artifacts/matplotlib-inline_1660814786464/work
+# mpmath @ file:///croot/mpmath_1690848262763/work
+# nest-asyncio @ file:///home/conda/feedstock_root/build_artifacts/nest-asyncio_1697083700168/work
+# networkx @ file:///croot/networkx_1690561992265/work
+# packaging @ file:///home/conda/feedstock_root/build_artifacts/packaging_1696202382185/work
+# parso @ file:///home/conda/feedstock_root/build_artifacts/parso_1638334955874/work
+# pexpect @ file:///home/conda/feedstock_root/build_artifacts/pexpect_1667297516076/work
+# pickleshare @ file:///home/conda/feedstock_root/build_artifacts/pickleshare_1602536217715/work
+# platformdirs @ file:///home/conda/feedstock_root/build_artifacts/platformdirs_1699715570510/work
+# prompt-toolkit @ file:///home/conda/feedstock_root/build_artifacts/prompt-toolkit_1699963054032/work
+# psutil @ file:///home/conda/feedstock_root/build_artifacts/psutil_1695367094274/work
+# ptyprocess @ file:///home/conda/feedstock_root/build_artifacts/ptyprocess_1609419310487/work/dist/ptyprocess-0.7.0-py2.py3-none-any.whl
+# pure-eval @ file:///home/conda/feedstock_root/build_artifacts/pure_eval_1642875951954/work
+# Pygments @ file:///home/conda/feedstock_root/build_artifacts/pygments_1700607939962/work
+# python-dateutil @ file:///home/conda/feedstock_root/build_artifacts/python-dateutil_1626286286081/work
+# pyzmq @ file:///home/conda/feedstock_root/build_artifacts/pyzmq_1652965335788/work
+# six @ file:///home/conda/feedstock_root/build_artifacts/six_1620240208055/work
+# stack-data @ file:///home/conda/feedstock_root/build_artifacts/stack_data_1669632077133/work
+# sympy @ file:///croot/sympy_1668202399572/work
+# tornado @ file:///home/conda/feedstock_root/build_artifacts/tornado_1695373560918/work
+# traitlets @ file:///home/conda/feedstock_root/build_artifacts/traitlets_1701095650114/work
+# typing_extensions @ file:///croot/typing_extensions_1690297465030/work
+# wcwidth @ file:///home/conda/feedstock_root/build_artifacts/wcwidth_1700607916581/work
+# zipp @ file:///home/conda/feedstock_root/build_artifacts/zipp_1695255097490/work
\ No newline at end of file
diff --git a/test.py b/test.py
new file mode 100644
index 0000000000000000000000000000000000000000..ed48f6fadc3d10b2248c4d07f536dff0d80140fa
--- /dev/null
+++ b/test.py
@@ -0,0 +1,10 @@
+from transformers import GPT2LMHeadModel, GPT2Tokenizer
+import torch
+
+# Choose the model size: 'gpt2', 'gpt2-medium', 'gpt2-large', etc.
+model_name = 'gpt2'
+model = GPT2LMHeadModel.from_pretrained(model_name)
+tokenizer = GPT2Tokenizer.from_pretrained(model_name)
+
+print(model)
+print(tokenizer)
\ No newline at end of file
diff --git a/train/babylm_dataset.py b/train/babylm_dataset.py
new file mode 100644
index 0000000000000000000000000000000000000000..fc5dac66ef7bfbd5ee9790f6d2509654cf2b93a3
--- /dev/null
+++ b/train/babylm_dataset.py
@@ -0,0 +1,141 @@
+# babylm_dataset.py
+# author: Julie Kallini
+
+import datasets
+import os
+import glob
+import tqdm
+from numpy.random import default_rng
+from itertools import product
+
+logger = datasets.logging.get_logger(__name__)
+
+_DESCRIPTION = """\
+    Pre-tokenized BabyLM HuggingFace dataset for verb perturbations.
+"""
+MODEL_NAME = "Llama-3.2-3B"
+_PERTURBED_DATA_PATH = f"../data/Perturbed_data/{MODEL_NAME}"
+_PERTURBATIONS = ["hop_control", "hop_tokens4", "hop_words4",
+                  "reverse_control", "reverse_partial", "reverse_full",
+                  "shuffle_control", "shuffle_nondeterministic",
+                  "shuffle_deterministic21", "shuffle_deterministic57", "shuffle_deterministic84",
+                  "shuffle_local3", "shuffle_local5", "shuffle_local10",
+                  "shuffle_even_odd"]
+# _RANDOM_SEEDS = [0, 14, 41, 53, 96]
+_RANDOM_SEEDS = [0]
+# _TRAIN_SETS = ["100M", "10M"]
+_TRAIN_SETS = ["10M"]
+_EOS_TOKEN_ID = 50256
+
+
+class BabyConfig(datasets.BuilderConfig):
+
+    def __init__(self, data_dir, babylm_train_set, random_seed, **kwargs):
+        """BuilderConfig for IzParens
+
+        Args:
+          data_dir: path to directory of tokenized, perturbed BabyLM dataset
+        """
+        super(BabyConfig, self).__init__(
+            **kwargs,
+        )
+        self.data_dir = data_dir
+        self.babylm_train_set = babylm_train_set
+        self.random_seed = random_seed
+
+
+class BabyLMCorpus(datasets.GeneratorBasedBuilder):
+    BUILDER_CONFIGS = [
+        BabyConfig(
+            name=f"babylm_{perturbation}_{train_set}_seed{random_seed}",
+            data_dir=os.path.join(
+                _PERTURBED_DATA_PATH, "babylm_" + perturbation),
+            babylm_train_set=train_set,
+            random_seed=random_seed,
+        ) for perturbation, train_set, random_seed in list(product(_PERTURBATIONS, _TRAIN_SETS, _RANDOM_SEEDS))
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            # This is the description that will appear on the datasets page.
+            description=_DESCRIPTION,
+            # datasets.features.FeatureConnectors
+            features=datasets.Features(
+                {
+                    "text": datasets.Value("string")
+                    # These are the features of your dataset like images, labels ...
+                }
+            ),
+            # If there's a common (input, target) tuple from the features,
+            # specify them here. They'll be used if as_supervised=True in
+            # builder.as_dataset.
+            supervised_keys=None,
+        )
+
+    def _split_generators(self, dl_manager):
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_" + self.config.babylm_train_set), "random_seed": self.config.random_seed, "split": "train"},
+            ),
+            # datasets.SplitGenerator(
+            #     name=datasets.Split.VALIDATION,
+            #     gen_kwargs={"data_dir": os.path.join(
+            #         self.config.data_dir, "babylm_dev"), "random_seed": self.config.random_seed, "split": "valid"},
+            #
+        ]
+
+    def __chunk(self, sentences, eos_token):
+
+        # Tokenize each sentence
+        logger.info("Loading pre-tokenized data")
+        tokenized_sentences = []
+        for sent in tqdm.tqdm(sentences):
+            tokenized_sentences.append([int(tok) for tok in sent.split()])
+
+        # Concatenate the tokenized sentences using the EOS token
+        logger.info("Concatenating tokenized data using EOS token")
+        all_tokens = []
+        for tokens in tqdm.tqdm(tokenized_sentences):
+            all_tokens.extend(tokens)
+            all_tokens.append(eos_token)
+
+        # Chunk the tokens into sublists of max_seq_len tokens each
+        logger.info("Chunking tokens into sublists of 1024")
+        max_seq_len = 1024
+        chunked_tokens = []
+        for i in tqdm.tqdm(range(0, len(all_tokens), max_seq_len)):
+            chunked_tokens.append(all_tokens[i:i + max_seq_len])
+
+        # Drop last line if not a multiple of max_seq_len
+        if len(chunked_tokens[-1]) < max_seq_len:
+            chunked_tokens.pop()
+
+        return chunked_tokens
+
+    def _generate_examples(self, data_dir, random_seed, split):
+        """This function returns the BabyLM text in the discretized, tokenized form."""
+
+        logger.info("Generating examples from = %s", data_dir)
+        infiles = sorted(glob.glob(os.path.join(data_dir, "*")))
+
+        # Extend sentences
+        all_sentences = []
+        for infile in infiles:
+            f = open(infile, encoding="utf-8")
+            all_sentences.extend(f.readlines())
+        logger.info("Total sentences: {}".format(len(all_sentences)))
+
+        # Shuffle because we are pre-tokenizing
+        rng = default_rng(seed=random_seed)
+        rng.shuffle(all_sentences)
+
+        # Tokenize and chunk
+        tokenized_lines = self.__chunk(all_sentences, _EOS_TOKEN_ID)
+
+        # Generate data
+        logger.info("Writing dataset as space-separated sequences of tokens")
+        for idx, line in enumerate(tokenized_lines):
+            l = " ".join([str(tok) for tok in line]) + "\n"
+            yield idx, {"text": l}
\ No newline at end of file
diff --git a/train/babylm_dataset_gpt.py b/train/babylm_dataset_gpt.py
new file mode 100644
index 0000000000000000000000000000000000000000..499d579cce7c42cd4a1a6517956c09abff6593b2
--- /dev/null
+++ b/train/babylm_dataset_gpt.py
@@ -0,0 +1,145 @@
+# babylm_dataset_test.py
+
+import datasets
+import os
+import glob
+import tqdm
+from numpy.random import default_rng
+from itertools import product
+
+logger = datasets.logging.get_logger(__name__)
+
+_DESCRIPTION = """\
+    Pre-tokenized BabyLM HuggingFace dataset for verb perturbations.
+"""
+MODEL_NAME = "gpt2"
+_PERTURBED_DATA_PATH = f"../data/Perturbed_data/{MODEL_NAME}"
+_PERTURBATIONS = ["hop_control", "hop_tokens4", "hop_words4",
+                  "reverse_control", "reverse_partial", "reverse_full",
+                  "shuffle_control", "shuffle_nondeterministic",
+                  "shuffle_deterministic21", "shuffle_deterministic57", "shuffle_deterministic84",
+                  "shuffle_local3", "shuffle_local5", "shuffle_local10",
+                  "shuffle_even_odd"]
+# _RANDOM_SEEDS = [0, 14, 41, 53, 96]
+_RANDOM_SEEDS = [0]
+# _TRAIN_SETS = ["100M", "10M"]
+_TRAIN_SETS = ["10M"]
+_EOS_TOKEN_ID = 50256
+
+
+class BabyConfig(datasets.BuilderConfig):
+
+    def __init__(self, data_dir, babylm_train_set, random_seed, **kwargs):
+        """BuilderConfig for IzParens
+
+        Args:
+          data_dir: path to directory of tokenized, perturbed BabyLM dataset
+        """
+        super(BabyConfig, self).__init__(
+            **kwargs,
+        )
+        self.data_dir = data_dir
+        self.babylm_train_set = babylm_train_set
+        self.random_seed = random_seed
+
+
+class BabyLMCorpus(datasets.GeneratorBasedBuilder):
+    BUILDER_CONFIGS = [
+        BabyConfig(
+            name=f"babylm_{perturbation}_{train_set}_seed{random_seed}",
+            data_dir=os.path.join(
+                _PERTURBED_DATA_PATH, "babylm_" + perturbation),
+            babylm_train_set=train_set,
+            random_seed=random_seed,
+        ) for perturbation, train_set, random_seed in list(product(_PERTURBATIONS, _TRAIN_SETS, _RANDOM_SEEDS))
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            # This is the description that will appear on the datasets page.
+            description=_DESCRIPTION,
+            # datasets.features.FeatureConnectors
+            features=datasets.Features(
+                {
+                    "text": datasets.Value("string")
+                    # These are the features of your dataset like images, labels ...
+                }
+            ),
+            # If there's a common (input, target) tuple from the features,
+            # specify them here. They'll be used if as_supervised=True in
+            # builder.as_dataset.
+            supervised_keys=None,
+        )
+
+    def _split_generators(self, dl_manager):
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_" + self.config.babylm_train_set), "random_seed": self.config.random_seed, "split": "train"},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_dev"), "random_seed": self.config.random_seed, "split": "valid"},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_test_affected"), "random_seed": self.config.random_seed, "split": "test"},
+        ),
+        ]
+
+    def __chunk(self, sentences, eos_token):
+
+        # Tokenize each sentence
+        logger.info("Loading pre-tokenized data")
+        tokenized_sentences = []
+        for sent in tqdm.tqdm(sentences):
+            tokenized_sentences.append([int(tok) for tok in sent.split()])
+
+        # Concatenate the tokenized sentences using the EOS token
+        logger.info("Concatenating tokenized data using EOS token")
+        all_tokens = []
+        for tokens in tqdm.tqdm(tokenized_sentences):
+            all_tokens.extend(tokens)
+            all_tokens.append(eos_token)
+
+        # Chunk the tokens into sublists of max_seq_len tokens each
+        logger.info("Chunking tokens into sublists of 1024")
+        max_seq_len = 1024
+        chunked_tokens = []
+        for i in tqdm.tqdm(range(0, len(all_tokens), max_seq_len)):
+            chunked_tokens.append(all_tokens[i:i + max_seq_len])
+
+        # Drop last line if not a multiple of max_seq_len
+        if len(chunked_tokens[-1]) < max_seq_len:
+            chunked_tokens.pop()
+
+        return chunked_tokens
+
+    def _generate_examples(self, data_dir, random_seed, split):
+        """This function returns the BabyLM text in the discretized, tokenized form."""
+
+        logger.info("Generating examples from = %s", data_dir)
+        infiles = sorted(glob.glob(os.path.join(data_dir, "*")))
+
+        # Extend sentences
+        all_sentences = []
+        for infile in infiles:
+            f = open(infile, encoding="utf-8")
+            all_sentences.extend(f.readlines())
+        logger.info("Total sentences: {}".format(len(all_sentences)))
+
+        # Shuffle because we are pre-tokenizing
+        rng = default_rng(seed=random_seed)
+        rng.shuffle(all_sentences)
+
+        # Tokenize and chunk
+        tokenized_lines = self.__chunk(all_sentences, _EOS_TOKEN_ID)
+
+        # Generate data
+        logger.info("Writing dataset as space-separated sequences of tokens")
+        for idx, line in enumerate(tokenized_lines):
+            l = " ".join([str(tok) for tok in line]) + "\n"
+            yield idx, {"text": l}
\ No newline at end of file
diff --git a/train/babylm_dataset_test.py b/train/babylm_dataset_test.py
new file mode 100644
index 0000000000000000000000000000000000000000..75a6ea95c46b15974e1acf80b88e8d7cb88a00dd
--- /dev/null
+++ b/train/babylm_dataset_test.py
@@ -0,0 +1,145 @@
+# babylm_dataset_test.py
+
+import datasets
+import os
+import glob
+import tqdm
+from numpy.random import default_rng
+from itertools import product
+
+logger = datasets.logging.get_logger(__name__)
+
+_DESCRIPTION = """\
+    Pre-tokenized BabyLM HuggingFace dataset for verb perturbations.
+"""
+MODEL_NAME = "Llama-3.2-3B"
+_PERTURBED_DATA_PATH = f"../data/Perturbed_data/{MODEL_NAME}"
+_PERTURBATIONS = ["hop_control", "hop_tokens4", "hop_words4",
+                  "reverse_control", "reverse_partial", "reverse_full",
+                  "shuffle_control", "shuffle_nondeterministic",
+                  "shuffle_deterministic21", "shuffle_deterministic57", "shuffle_deterministic84",
+                  "shuffle_local3", "shuffle_local5", "shuffle_local10",
+                  "shuffle_even_odd"]
+# _RANDOM_SEEDS = [0, 14, 41, 53, 96]
+_RANDOM_SEEDS = [0]
+# _TRAIN_SETS = ["100M", "10M"]
+_TRAIN_SETS = ["10M"]
+_EOS_TOKEN_ID = 50256
+
+
+class BabyConfig(datasets.BuilderConfig):
+
+    def __init__(self, data_dir, babylm_train_set, random_seed, **kwargs):
+        """BuilderConfig for IzParens
+
+        Args:
+          data_dir: path to directory of tokenized, perturbed BabyLM dataset
+        """
+        super(BabyConfig, self).__init__(
+            **kwargs,
+        )
+        self.data_dir = data_dir
+        self.babylm_train_set = babylm_train_set
+        self.random_seed = random_seed
+
+
+class BabyLMCorpus(datasets.GeneratorBasedBuilder):
+    BUILDER_CONFIGS = [
+        BabyConfig(
+            name=f"babylm_{perturbation}_{train_set}_seed{random_seed}",
+            data_dir=os.path.join(
+                _PERTURBED_DATA_PATH, "babylm_" + perturbation),
+            babylm_train_set=train_set,
+            random_seed=random_seed,
+        ) for perturbation, train_set, random_seed in list(product(_PERTURBATIONS, _TRAIN_SETS, _RANDOM_SEEDS))
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            # This is the description that will appear on the datasets page.
+            description=_DESCRIPTION,
+            # datasets.features.FeatureConnectors
+            features=datasets.Features(
+                {
+                    "text": datasets.Value("string")
+                    # These are the features of your dataset like images, labels ...
+                }
+            ),
+            # If there's a common (input, target) tuple from the features,
+            # specify them here. They'll be used if as_supervised=True in
+            # builder.as_dataset.
+            supervised_keys=None,
+        )
+
+    def _split_generators(self, dl_manager):
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_" + self.config.babylm_train_set), "random_seed": self.config.random_seed, "split": "train"},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.VALIDATION,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_dev"), "random_seed": self.config.random_seed, "split": "valid"},
+            ),
+            datasets.SplitGenerator(
+                name=datasets.Split.TEST,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_test_affected"), "random_seed": self.config.random_seed, "split": "test"},
+        ),
+        ]
+
+    def __chunk(self, sentences, eos_token):
+
+        # Tokenize each sentence
+        logger.info("Loading pre-tokenized data")
+        tokenized_sentences = []
+        for sent in tqdm.tqdm(sentences):
+            tokenized_sentences.append([int(tok) for tok in sent.split()])
+
+        # Concatenate the tokenized sentences using the EOS token
+        logger.info("Concatenating tokenized data using EOS token")
+        all_tokens = []
+        for tokens in tqdm.tqdm(tokenized_sentences):
+            all_tokens.extend(tokens)
+            all_tokens.append(eos_token)
+
+        # Chunk the tokens into sublists of max_seq_len tokens each
+        logger.info("Chunking tokens into sublists of 1024")
+        max_seq_len = 1024
+        chunked_tokens = []
+        for i in tqdm.tqdm(range(0, len(all_tokens), max_seq_len)):
+            chunked_tokens.append(all_tokens[i:i + max_seq_len])
+
+        # Drop last line if not a multiple of max_seq_len
+        if len(chunked_tokens[-1]) < max_seq_len:
+            chunked_tokens.pop()
+
+        return chunked_tokens
+
+    def _generate_examples(self, data_dir, random_seed, split):
+        """This function returns the BabyLM text in the discretized, tokenized form."""
+
+        logger.info("Generating examples from = %s", data_dir)
+        infiles = sorted(glob.glob(os.path.join(data_dir, "*")))
+
+        # Extend sentences
+        all_sentences = []
+        for infile in infiles:
+            f = open(infile, encoding="utf-8")
+            all_sentences.extend(f.readlines())
+        logger.info("Total sentences: {}".format(len(all_sentences)))
+
+        # Shuffle because we are pre-tokenizing
+        rng = default_rng(seed=random_seed)
+        rng.shuffle(all_sentences)
+
+        # Tokenize and chunk
+        tokenized_lines = self.__chunk(all_sentences, _EOS_TOKEN_ID)
+
+        # Generate data
+        logger.info("Writing dataset as space-separated sequences of tokens")
+        for idx, line in enumerate(tokenized_lines):
+            l = " ".join([str(tok) for tok in line]) + "\n"
+            yield idx, {"text": l}
\ No newline at end of file
diff --git a/train/huggingface_update/download.py b/train/huggingface_update/download.py
new file mode 100644
index 0000000000000000000000000000000000000000..084f45f24f080e8c039ed2c13827d8f52e1252a9
--- /dev/null
+++ b/train/huggingface_update/download.py
@@ -0,0 +1,15 @@
+from huggingface_hub import snapshot_download
+
+# 设置要下载的仓库 ID 和本地存储目录
+repo_id = "Yaning1001/Save_checkpoints"
+local_dir = "/home/yiren/new_ssd2/chunhui/yaning/project/impossible_llm/train/checkpoints/Llama-3.2-3B-FTP/download/"
+
+# 指定允许下载的文件模式，含路径信息
+patterns = "babylm_shuffle_control_10M_seed0/"   
+
+# 下载符合模式的文件
+snapshot_download(
+    repo_id=repo_id,
+    local_dir=local_dir,
+    allow_patterns=patterns,
+)
diff --git a/train/huggingface_update/repo_upload.py b/train/huggingface_update/repo_upload.py
new file mode 100644
index 0000000000000000000000000000000000000000..dadbfc5d7cb18a78adc466d1a5496dca15506c65
--- /dev/null
+++ b/train/huggingface_update/repo_upload.py
@@ -0,0 +1,43 @@
+
+# from huggingface_hub import HfApi
+
+# # 初始化 API 客户端
+# api = HfApi()
+
+# # 设置参数/
+# repo_id = "Yaning1001/ScalingMakeItPossible "  # 远程仓库 ID
+# repo_type = "model"        # 仓库类型：数据集
+# folder_path = "home/yiren/new_ssd2/chunhui/yaning/project/impossible_llm/data/Perturbed_data/Llama-3.2-3B"  # 本地路径
+# # token = "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"  # 替换为你的 Hugging Face API Token
+
+# # 调用上传方法
+# api.upload_large_folder(
+#     repo_id=repo_id,
+#     repo_type=repo_type,
+#     folder_path=folder_path,
+# )
+
+from huggingface_hub import HfApi
+
+api = HfApi()
+
+repo_id = "Yaning1001/ScalingMakeItPossible"  # 你的 Hugging Face 仓库名
+repo_type = "model"
+folder_path = "/home/yiren/new_ssd2/chunhui/yaning/project/impossible_llm"  # 项目根目录
+
+# 上传文件时忽略 train/checkpoints/
+api.upload_large_folder(
+    repo_id=repo_id,
+    repo_type=repo_type,
+    folder_path=folder_path,
+    ignore_patterns=[
+        "train/checkpoints/*", # 忽略 checkpoint
+        "train/checkpoints/**",
+        "train/wandb/*",         # 忽略 wandb 日志
+        "train/wandb/**",
+        "*.log",                 # 忽略所有 .log 文件
+        "*.tmp",                 # 忽略临时文件
+        "__pycache__/*",         # 忽略 Python 缓存
+        ".git/*",        
+    ],
+)
diff --git a/train/train_deep.py b/train/train_deep.py
new file mode 100644
index 0000000000000000000000000000000000000000..6001f0d824a3e32960528861883e329c304bcd83
--- /dev/null
+++ b/train/train_deep.py
@@ -0,0 +1,99 @@
+import sys
+import torch
+sys.path.append("..")
+
+import os
+from datasets import load_dataset
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_llama import PERTURBATIONS, BABYLM_SPLITS, BABYLM_DATA_PATH, \
+    GENRES, MARKER_TOKEN_IDS, marker_sg_token, marker_pl_token, marker_rev_token, write_file
+import argparse
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+
+# Setup for Weights & Biases
+# wandb.init(project="kallini", group="babylm-perturbation-experiments", name=run_id)
+
+if __name__ == "__main__":
+
+    # === CONFIGURATION SETTINGS ===
+    parser = argparse.ArgumentParser(description="Training configuration.")
+
+    parser.add_argument('--perturbation', type=str, default='hop_tokens4', help='Type of perturbation to use.')
+    parser.add_argument('--train_set', type=str, default='10M', help='Dataset size for training.')
+    parser.add_argument('--batch_size', type=int, default=4, help='Batch size for training.')
+    parser.add_argument('--epoch', type=int, default=20, help='train epoch')
+    parser.add_argument('--seed', type=int, default=0, help='Random seed.')
+
+    args = parser.parse_args()
+
+    # no_pos_encodings_underscore = ""  # Ex: "_nopos" if needed
+    ckpt_path = "./checkpoints"
+    # effective_bsz = 512
+
+    model_name = "meta-llama/Llama-3.2-3B"
+
+    model_save_name = "Llama-3.2-3B"
+    # === FILE PATHS BASED ON CONFIGURATION ===
+    run_id = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    cache_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "artifacts")
+    run_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "runs")
+    os.makedirs(cache_dir, exist_ok=True)
+    os.makedirs(run_dir, exist_ok=True)
+
+    # === DATASET LOADING ===
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    # dataset = load_dataset('babylm_dataset_llama.py', name=dataset_name)
+    dataset = load_dataset('babylm_dataset_llama.py', name=dataset_name, trust_remote_code=True)
+    train_dataset = dataset['train']
+
+    # === TOKENIZER & MODEL LOADING ===
+    # model_name = f"gpt2{'' if no_pos_encodings_underscore == '' else '-no-pos'}-small-{perturbation}-{paren_model}"
+ 
+    # tokenizer = AutoTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
+    tokenizer = PERTURBATIONS[args.perturbation]['llama_tokenizer']
+    model = AutoModelForCausalLM.from_pretrained(model_name, 
+                                                # device_map="auto",  # deepspeed needs to delete this setting
+                                                cache_dir=cache_dir)
+
+    # print("model:", model)
+    # === TOKENIZATION ===
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+    tokenized_train = train_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+    # === DATA COLLATOR ===
+    data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+
+    # === TRAINING ARGUMENTS ===
+    training_args = TrainingArguments(
+        output_dir=run_dir,
+        evaluation_strategy="no",
+        per_device_train_batch_size=args.batch_size,  # set "auto" in deepspeed config, adjust it in trainer
+        logging_dir='./logs',
+        logging_steps=150,
+        save_steps=10000,
+        learning_rate=5e-5, # align with deepspeed
+        num_train_epochs=args.epoch,
+        seed=args.seed,
+        gradient_accumulation_steps=2, # # set "auto" in deepspeed config, adjust it in trainer
+        fp16 = True, # align with deepspeed
+        report_to="none",
+        deepspeed="deepspeed_config/train_dp_config.json"
+    )
+
+        # === TRAINER ===
+    trainer = Trainer(
+        model=model,
+        args=training_args,
+        train_dataset=tokenized_train,
+        tokenizer=tokenizer,
+        data_collator=data_collator
+    )
+    
+    # === TRAIN MODEL ===
+    trainer.train()
+    # End logging
+    # wandb.finish()
+
diff --git a/train/train_ftp.py b/train/train_ftp.py
new file mode 100644
index 0000000000000000000000000000000000000000..1307bbaf3ac2be38b16f430ab07f5dd12d93b857
--- /dev/null
+++ b/train/train_ftp.py
@@ -0,0 +1,137 @@
+import sys
+import torch
+sys.path.append("..")
+
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_llama import PERTURBATIONS, BABYLM_SPLITS, BABYLM_DATA_PATH, \
+    GENRES, MARKER_TOKEN_IDS, marker_sg_token, marker_pl_token, marker_rev_token, write_file
+from datasets import load_dataset
+from FTP import AdamP
+
+import wandb
+import argparse
+import copy
+import math
+import os
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+ftp_k = 1
+class TrainerAdamP(Trainer):
+
+    def create_optimizer(self):
+        optimizer_params = {
+            "lr": 5e-6,
+            "weight_decay": 0.0,
+            "k": ftp_k,  # Example parameter for AdamP
+            "exclude_set": set()  # Use empty set if you don't want exclusion
+        }
+
+        # Cache pre-trained model weights 
+        params_to_opt = [x[1] for x in self.model.named_parameters() if x[1].requires_grad]
+        params_to_opt_name = [x[0] for x in self.model.named_parameters() if x[1].requires_grad]
+        params_anchor = copy.deepcopy(params_to_opt)
+        param_group = [{'params': params_to_opt, 'pre': params_anchor, 'name': params_to_opt_name}]
+
+        # Initialize the AdamP optimizer
+        self.optimizer = AdamP(param_group, **optimizer_params)
+
+
+
+if __name__ == "__main__":
+
+    # === CONFIGURATION SETTINGS ===
+    parser = argparse.ArgumentParser(description="Training configuration.")
+
+    parser.add_argument('--perturbation', type=str, default='hop_tokens4', help='Type of perturbation to use.')
+    parser.add_argument('--train_set', type=str, default='10M', help='Dataset size for training.')
+    parser.add_argument('--batch_size', type=int, default=3, help='Batch size for training.')
+    parser.add_argument('--epoch', type=int, default=3, help='train epoch')
+    parser.add_argument('--seed', type=int, default=0, help='Random seed.')
+    parser.add_argument('--lr', type=float, default=5e-6, help='Learning rate.')
+
+    args = parser.parse_args()
+
+    # no_pos_encodings_underscore = ""  # Ex: "_nopos" if needed
+    ckpt_path = "./checkpoints"
+    # effective_bsz = 512
+
+    model_name = "meta-llama/Llama-3.2-3B"
+    model_save_name = "Llama-3.2-3B-FTP"
+
+    # === FILE PATHS BASED ON CONFIGURATION ===
+
+    
+    wandb_id = f"{model_save_name}_{args.perturbation}_train_set_{args.train_set}_epoch_{args.epoch}_batch_size_{args.batch_size}_seed_{args.seed}_lr_{args.lr}_wandb_ftp_{ftp_k}"
+    wandb.init(project="exp-impo-shuffle", group="ftp-1", name=wandb_id)
+    wandb.config.update(args) 
+
+    run_id = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    cache_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "artifacts")
+    run_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "runs")
+    os.makedirs(cache_dir, exist_ok=True)
+    os.makedirs(run_dir, exist_ok=True)
+
+    # === DATASET LOADING === 
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    train_dataset = dataset['train']
+    valid_dataset = dataset['validation']
+
+    # === TOKENIZER & MODEL LOADING ===
+    # model_name = f"gpt2{'' if no_pos_encodings_underscore == '' else '-no-pos'}-small-{perturbation}-{paren_model}"
+ 
+    # tokenizer = AutoTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
+    tokenizer = PERTURBATIONS[args.perturbation]['llama_tokenizer']
+    model = AutoModelForCausalLM.from_pretrained(model_name,  
+                                                # device_map="auto",  # deepspeed needs to delete this setting
+                                                cache_dir=cache_dir)
+
+    # print("model:", model)
+    # === TOKENIZATION ===
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+    tokenized_train = train_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+    tokenized_valid = valid_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+    shuffled_valid = tokenized_valid.shuffle()
+    tokenized_valid = shuffled_valid.select(range(1000))
+    print("tokenized_valid:", tokenized_valid)
+    # print(train_dataset)
+    # === DATA COLLATOR ===2
+    data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+
+    # === TRAINING ARGUMENTS ===
+    training_args = TrainingArguments(
+        output_dir=run_dir,
+        evaluation_strategy="steps",
+        eval_steps=10,
+        per_device_train_batch_size=args.batch_size,  # set "auto" in deepspeed config, adjust it in trainer
+        logging_dir='./logs',
+        logging_steps=1,
+        save_steps=100,
+        learning_rate=args.lr, # align with deepspeed
+        num_train_epochs=args.epoch,
+        seed=args.seed,
+        gradient_accumulation_steps=2, # # set "auto" in deepspeed config, adjust it in trainer
+        fp16=True, # align with deepspeed
+        report_to="wandb",
+        warmup_ratio=0.1, 
+        deepspeed="deepspeed_config/train_dp_config.json"
+    )
+
+    # === TRAINER ===
+    trainer = TrainerAdamP(
+        model=model,
+        args=training_args,
+        train_dataset=tokenized_train,
+        eval_dataset=tokenized_valid,
+        tokenizer=tokenizer,
+        data_collator=data_collator
+    )
+    
+    # === TRAIN MODEL ===
+    trainer.train()
+    # End logging
+    wandb.finish()
+
diff --git a/train/train_ftp_update2.py b/train/train_ftp_update2.py
new file mode 100644
index 0000000000000000000000000000000000000000..982aa561734084405e2af93ecb8f9df55923a610
--- /dev/null
+++ b/train/train_ftp_update2.py
@@ -0,0 +1,137 @@
+import sys
+import torch
+sys.path.append("..")
+
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_llama import PERTURBATIONS, BABYLM_SPLITS, BABYLM_DATA_PATH, \
+    GENRES, MARKER_TOKEN_IDS, marker_sg_token, marker_pl_token, marker_rev_token, write_file
+from datasets import load_dataset
+from FTP_update2 import AdamW
+
+import wandb
+import argparse
+import copy
+import math
+import os
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+ftp_k = 1
+class TrainerAdamP(Trainer):
+
+    def create_optimizer(self):
+        optimizer_params = {
+            "lr": 5e-6,
+            "weight_decay": 0.0,
+            "k": ftp_k,  # Example parameter for AdamP
+            "exclude_set": set()  # Use empty set if you don't want exclusion
+        }
+
+        # Cache pre-trained model weights 
+        params_to_opt = [x[1] for x in self.model.named_parameters() if x[1].requires_grad]
+        params_to_opt_name = [x[0] for x in self.model.named_parameters() if x[1].requires_grad]
+        params_anchor = copy.deepcopy(params_to_opt)
+        param_group = [{'params': params_to_opt, 'pre': params_anchor, 'name': params_to_opt_name}]
+
+        # Initialize the AdamP optimizer
+        self.optimizer = AdamP(param_group, **optimizer_params)
+
+
+
+if __name__ == "__main__":
+
+    # === CONFIGURATION SETTINGS ===
+    parser = argparse.ArgumentParser(description="Training configuration.")
+
+    parser.add_argument('--perturbation', type=str, default='hop_tokens4', help='Type of perturbation to use.')
+    parser.add_argument('--train_set', type=str, default='10M', help='Dataset size for training.')
+    parser.add_argument('--batch_size', type=int, default=3, help='Batch size for training.')
+    parser.add_argument('--epoch', type=int, default=3, help='train epoch')
+    parser.add_argument('--seed', type=int, default=0, help='Random seed.')
+    parser.add_argument('--lr', type=float, default=5e-6, help='Learning rate.')
+
+    args = parser.parse_args()
+
+    # no_pos_encodings_underscore = ""  # Ex: "_nopos" if needed
+    ckpt_path = "./checkpoints"
+    # effective_bsz = 512
+
+    model_name = "meta-llama/Llama-3.2-3B"
+    model_save_name = "Llama-3.2-3B-FTP-Update2"
+
+    # === FILE PATHS BASED ON CONFIGURATION ===
+
+    
+    wandb_id = f"{model_save_name}_{args.perturbation}_train_set_{args.train_set}_epoch_{args.epoch}_batch_size_{args.batch_size}_seed_{args.seed}_lr_{args.lr}_wandb_ftp_{ftp_k}"
+    wandb.init(project="exp-impo-hop", group="ftp-1", name=wandb_id)
+    wandb.config.update(args) 
+
+    run_id = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    cache_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "artifacts")
+    run_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "runs")
+    os.makedirs(cache_dir, exist_ok=True)
+    os.makedirs(run_dir, exist_ok=True)
+
+    # === DATASET LOADING === 
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    train_dataset = dataset['train']
+    valid_dataset = dataset['validation']
+
+    # === TOKENIZER & MODEL LOADING ===
+    # model_name = f"gpt2{'' if no_pos_encodings_underscore == '' else '-no-pos'}-small-{perturbation}-{paren_model}"
+ 
+    # tokenizer = AutoTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
+    tokenizer = PERTURBATIONS[args.perturbation]['llama_tokenizer']
+    model = AutoModelForCausalLM.from_pretrained(model_name,  
+                                                # device_map="auto",  # deepspeed needs to delete this setting
+                                                cache_dir=cache_dir)
+
+    # print("model:", model)
+    # === TOKENIZATION ===
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+    tokenized_train = train_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+    tokenized_valid = valid_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+    shuffled_valid = tokenized_valid.shuffle()
+    tokenized_valid = shuffled_valid.select(range(1000))
+    print("tokenized_valid:", tokenized_valid)
+    # print(train_dataset)
+    # === DATA COLLATOR ===2
+    data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+
+    # === TRAINING ARGUMENTS ===
+    training_args = TrainingArguments(
+        output_dir=run_dir,
+        evaluation_strategy="steps",
+        eval_steps=10,
+        per_device_train_batch_size=args.batch_size,  # set "auto" in deepspeed config, adjust it in trainer
+        logging_dir='./logs',
+        logging_steps=1,
+        save_steps=100,
+        learning_rate=args.lr, # align with deepspeed
+        num_train_epochs=args.epoch,
+        seed=args.seed,
+        gradient_accumulation_steps=2, # # set "auto" in deepspeed config, adjust it in trainer
+        fp16=True, # align with deepspeed
+        report_to="wandb",
+        warmup_ratio=0.1, 
+        deepspeed="deepspeed_config/train_dp_config.json"
+    )
+
+    # === TRAINER ===
+    trainer = TrainerAdamP(
+        model=model,
+        args=training_args,
+        train_dataset=tokenized_train,
+        eval_dataset=tokenized_valid,
+        tokenizer=tokenizer,
+        data_collator=data_collator
+    )
+    
+    # === TRAIN MODEL ===
+    trainer.train()
+    # End logging
+    wandb.finish()
+
diff --git a/train/train_ftp_update3.py b/train/train_ftp_update3.py
new file mode 100644
index 0000000000000000000000000000000000000000..b2f83bc1e24210971ce680af05becf84b171acb2
--- /dev/null
+++ b/train/train_ftp_update3.py
@@ -0,0 +1,163 @@
+import sys
+import torch
+sys.path.append("..")
+
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_llama_3B import PERTURBATIONS, BABYLM_SPLITS, BABYLM_DATA_PATH, \
+    GENRES, MARKER_TOKEN_IDS, marker_sg_token, marker_pl_token, marker_rev_token, write_file
+from datasets import load_dataset
+from FTP_3 import AdamP
+
+import wandb
+import argparse
+import copy
+import math
+import os
+
+# Prevent tokenizer parallelism issues
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+
+# Define AdamP-specific parameter
+ftp_k = 1
+
+# Custom Trainer with Lipschitz Regularization
+class TrainerAdamP(Trainer):
+    def create_optimizer(self):
+        optimizer_params = {
+            "lr": 5e-6,
+            "weight_decay": 0.0,
+            "k": ftp_k,
+            "exclude_set": set()
+        }
+
+        params_to_opt = [x[1] for x in self.model.named_parameters() if x[1].requires_grad]
+        params_to_opt_name = [x[0] for x in self.model.named_parameters() if x[1].requires_grad]
+        params_anchor = copy.deepcopy(params_to_opt)
+        param_group = [{'params': params_to_opt, 'pre': params_anchor, 'name': params_to_opt_name}]
+
+        self.optimizer = AdamP(param_group, **optimizer_params)
+
+def compute_loss(self, model, inputs, return_outputs=False):
+    # Default loss computation
+    outputs = model(**inputs)
+    loss = outputs.loss  # Base loss (e.g., cross-entropy)
+
+    # Compute Lipschitz regularization
+    input_data = inputs['input_ids'].detach().clone().float()  # Convert to float for gradient computation
+    input_data.requires_grad_(True)
+
+    outputs_with_grad = model(input_data)
+    logits = outputs_with_grad.logits  # Shape: [batch_size, seq_len, vocab_size]
+    batch_size, seq_len, vocab_size = logits.size()
+
+    lip_mat = []  # List to store gradient norms for each output dimension
+
+    for i in range(vocab_size):  # Iterate over each output dimension
+        v = torch.zeros_like(logits)
+        v[:, :, i] = 1  # Select gradient for dimension `i`
+        gradients = torch.autograd.grad(
+            outputs=logits,
+            inputs=input_data,
+            grad_outputs=v,
+            create_graph=True,
+            retain_graph=True,
+            only_inputs=True
+        )[0]  # Shape: [batch_size, seq_len, input_dim]
+        grad_norm = torch.norm(gradients, dim=-1).unsqueeze(dim=-1)  # Gradient norm for each input
+        lip_mat.append(grad_norm)  # Append to lip_mat
+
+    # Combine all dimensions' gradient norms
+    lip_concat = torch.cat(lip_mat, dim=-1)  # Shape: [batch_size, seq_len, vocab_size]
+    lip_con_norm = torch.norm(lip_concat, dim=-1)  # L2-norm across vocab dimensions for each sample
+
+    # Compute Lipschitz loss as the maximum norm across samples
+    lip_loss = torch.max(lip_con_norm)
+    total_loss = loss + 0.5 * lip_loss  # Combine with base loss (scale factor: 0.01)
+
+    return (total_loss, outputs) if return_outputs else total_loss
+
+
+
+if __name__ == "__main__":
+    # === CONFIGURATION SETTINGS ===
+    parser = argparse.ArgumentParser(description="Training configuration.")
+
+    parser.add_argument('--perturbation', type=str, default='hop_tokens4', help='Type of perturbation to use.')
+    parser.add_argument('--train_set', type=str, default='10M', help='Dataset size for training.')
+    parser.add_argument('--batch_size', type=int, default=3, help='Batch size for training.')
+    parser.add_argument('--epoch', type=int, default=3, help='Train epoch')
+    parser.add_argument('--seed', type=int, default=0, help='Random seed.')
+    parser.add_argument('--lr', type=float, default=5e-6, help='Learning rate.')
+
+    args = parser.parse_args()
+
+    ckpt_path = "./checkpoints"
+    model_name = "meta-llama/Llama-3.2-3B"
+    model_save_name = "Llama-3.2-3B-FTP-Update"
+
+    wandb_id = f"{model_save_name}_{args.perturbation}_train_set_{args.train_set}_epoch_{args.epoch}_batch_size_{args.batch_size}_seed_{args.seed}_lr_{args.lr}_wandb_ftp_{ftp_k}_UGD_ftp_2_lip_0.5"
+    wandb.init(project="FTP-shuffle", group="shuffle", name=wandb_id)
+    wandb.config.update(args)
+
+    run_id = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    cache_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "artifacts")
+    run_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "runs")
+    os.makedirs(cache_dir, exist_ok=True)
+    os.makedirs(run_dir, exist_ok=True)
+
+    # === DATASET LOADING === 
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    train_dataset = dataset['train']
+    valid_dataset = dataset['validation']
+
+    # === TOKENIZER & MODEL LOADING ===
+    tokenizer = PERTURBATIONS[args.perturbation]['llama_tokenizer']
+    model = AutoModelForCausalLM.from_pretrained(model_name, cache_dir=cache_dir)
+
+    # === TOKENIZATION ===
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+
+    tokenized_train = train_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+    tokenized_valid = valid_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+    shuffled_valid = tokenized_valid.shuffle()
+    tokenized_valid = shuffled_valid.select(range(1000))
+
+    # === DATA COLLATOR ===
+    data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+
+    # === TRAINING ARGUMENTS ===
+    training_args = TrainingArguments(
+        output_dir=run_dir,
+        evaluation_strategy="steps",
+        eval_steps=10,
+        per_device_train_batch_size=args.batch_size,
+        logging_dir='./logs',
+        logging_steps=1,
+        save_steps=1000000,
+        learning_rate=args.lr,
+        num_train_epochs=args.epoch,
+        seed=args.seed,
+        gradient_accumulation_steps=2,
+        fp16=True,
+        report_to="wandb",
+        warmup_ratio=0.1,
+        deepspeed="deepspeed_config/train_dp_config.json"
+    )
+
+    # === TRAINER ===
+    trainer = TrainerAdamP(
+        model=model,
+        args=training_args,
+        train_dataset=tokenized_train,
+        eval_dataset=tokenized_valid,
+        tokenizer=tokenizer,
+        data_collator=data_collator
+    )
+
+    # === TRAIN MODEL ===
+    trainer.train()
+    # End logging
+    wandb.finish()
diff --git a/train/train_llama_1B.py b/train/train_llama_1B.py
new file mode 100644
index 0000000000000000000000000000000000000000..a10dfc59701d361173ae0f7543b41744ed0125f5
--- /dev/null
+++ b/train/train_llama_1B.py
@@ -0,0 +1,117 @@
+import sys
+import torch
+sys.path.append("..")
+
+import os
+from datasets import load_dataset
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_llama_1B import PERTURBATIONS, BABYLM_SPLITS, BABYLM_DATA_PATH, \
+    GENRES, MARKER_TOKEN_IDS, marker_sg_token, marker_pl_token, marker_rev_token, write_file
+import wandb
+import argparse
+
+os.environ["TOKENIZERS_PARALLELISM"] = "false"
+# os.environ['MASTER_PORT'] = '12345' 
+
+# import wandb
+
+# Setup for Weights & Biases
+# wandb.init(project="kallini", group="babylm-perturbation-experiments", name=run_id)
+
+if __name__ == "__main__":
+
+    # === CONFIGURATION SETTINGS ===
+    parser = argparse.ArgumentParser(description="Training configuration.")
+
+    parser.add_argument('--perturbation', type=str, default='hop_tokens4', help='Type of perturbation to use.')
+    parser.add_argument('--train_set', type=str, default='10M', help='Dataset size for training.')
+    parser.add_argument('--batch_size', type=int, default=4, help='Batch size for training.')
+    parser.add_argument('--epoch', type=int, default=3, help='train epoch')
+    parser.add_argument('--seed', type=int, default=0, help='Random seed.')
+    parser.add_argument('--lr', type=float, default=5e-6, help='Learning rate.')
+
+    args = parser.parse_args()
+
+    # no_pos_encodings_underscore = ""  # Ex: "_nopos" if needed
+    ckpt_path = "./checkpoints"
+    # effective_bsz = 512
+
+    model_name = "meta-llama/Llama-3.2-1B"
+    model_save_name = "Llama-3.2-1B"
+
+    # === FILE PATHS BASED ON CONFIGURATION ===
+
+    
+    wandb_id = f"{model_save_name}_{args.perturbation}_train_set_{args.train_set}_epoch_{args.epoch}_batch_size_{args.batch_size}_seed_{args.seed}_lr_{args.lr}_wandb"
+    wandb.init(project="exp-impo-shuffle", group="shuffle-1B", name=wandb_id)
+    wandb.config.update(args) 
+
+    run_id = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    cache_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "artifacts")
+    run_dir = os.path.join(ckpt_path, f"{model_save_name}", run_id, "runs")
+    os.makedirs(cache_dir, exist_ok=True)
+    os.makedirs(run_dir, exist_ok=True)
+
+    # === DATASET LOADING === 
+    dataset_name = f"babylm_{args.perturbation}_{args.train_set}_seed{args.seed}"
+    dataset = load_dataset('babylm_dataset_test.py', name=dataset_name, trust_remote_code=True)
+    train_dataset = dataset['train']
+    valid_dataset = dataset['validation']
+
+    # === TOKENIZER & MODEL LOADING ===
+    # model_name = f"gpt2{'' if no_pos_encodings_underscore == '' else '-no-pos'}-small-{perturbation}-{paren_model}"
+ 
+    # tokenizer = AutoTokenizer.from_pretrained(model_name, cache_dir=cache_dir)
+    tokenizer = PERTURBATIONS[args.perturbation]['llama_tokenizer']
+    model = AutoModelForCausalLM.from_pretrained(model_name, 
+                                                # device_map="auto",  # deepspeed needs to delete this setting
+                                                cache_dir=cache_dir)
+
+    # print("model:", model)
+    # === TOKENIZATION ===
+    def tokenize_function(examples):
+        return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+    tokenized_train = train_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+    tokenized_valid = valid_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+
+    shuffled_valid = tokenized_valid.shuffle()
+    tokenized_valid = shuffled_valid.select(range(1000))
+    print("tokenized_valid:", tokenized_valid)
+    # print(train_dataset)
+    # === DATA COLLATOR ===2
+    data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+
+    # === TRAINING ARGUMENTS ===
+    training_args = TrainingArguments(
+        output_dir=run_dir,
+        evaluation_strategy="steps",
+        eval_steps=10,
+        per_device_train_batch_size=args.batch_size,  # set "auto" in deepspeed config, adjust it in trainer
+        logging_dir='./logs',
+        logging_steps=1,
+        save_steps=100,
+        learning_rate=args.lr, # align with deepspeed
+        num_train_epochs=args.epoch,
+        seed=args.seed,
+        gradient_accumulation_steps=2, # # set "auto" in deepspeed config, adjust it in trainer
+        fp16=True, # align with deepspeed
+        report_to="wandb",
+        warmup_ratio=0.1, 
+        deepspeed="deepspeed_config/train_dp_config.json"
+    )
+
+    # === TRAINER ===
+    trainer = Trainer(
+        model=model,
+        args=training_args,
+        train_dataset=tokenized_train,
+        eval_dataset=tokenized_valid,
+        tokenizer=tokenizer,
+        data_collator=data_collator
+    )
+    
+    # === TRAIN MODEL ===
+    trainer.train()
+    # End logging
+    wandb.finish()
+
diff --git a/train/train_qwen_lora.py b/train/train_qwen_lora.py
new file mode 100644
index 0000000000000000000000000000000000000000..0e048012ef34d5b83629b9876266a03d9af1a500
--- /dev/null
+++ b/train/train_qwen_lora.py
@@ -0,0 +1,93 @@
+import sys
+sys.path.append("..")
+
+import os
+from datasets import load_dataset
+from transformers import AutoTokenizer, AutoModelForCausalLM, Trainer, TrainingArguments, DataCollatorForLanguageModeling
+from utils_qwen import PERTURBATIONS, BABYLM_SPLITS, BABYLM_DATA_PATH, \
+    GENRES, MARKER_TOKEN_IDS, marker_sg_token, marker_pl_token, marker_rev_token, write_file
+from peft import get_peft_model, LoraConfig, TaskType  # Import PEFT components for LoRA
+# import wandb
+
+# === CONFIGURATION SETTINGS ===
+perturbation = "shuffle_deterministic21"
+train_set = "10M"
+seed = 0
+ckpt_path = "./checkpoints"
+effective_bsz = 512
+
+# === FILE PATHS BASED ON CONFIGURATION ===
+run_id = f"babylm_{perturbation}_{train_set}_seed{seed}"
+cache_dir = os.path.join(ckpt_path, "babylm_lora", run_id, "artifacts")
+run_dir = os.path.join(ckpt_path, "babylm_lora", run_id, "runs")
+os.makedirs(cache_dir, exist_ok=True)
+os.makedirs(run_dir, exist_ok=True)
+
+# Setup for Weights & Biases
+# wandb.init(project="kallini", group="babylm-perturbation-experiments", name=run_id)
+
+# === DATASET LOADING ===
+dataset_name = f"babylm_{perturbation}_{train_set}_seed{seed}"
+dataset = load_dataset('babylm_dataset.py', name=dataset_name, trust_remote_code=True)
+train_dataset = dataset['train']
+
+# === TOKENIZER & MODEL LOADING ===
+model_name = "Qwen/Qwen2.5-0.5B"
+tokenizer = PERTURBATIONS[perturbation]['qwen_tokenizer']
+model = AutoModelForCausalLM.from_pretrained(model_name, cache_dir=cache_dir)
+
+# === APPLYING LoRA ===
+lora_config = LoraConfig(
+    task_type=TaskType.CAUSAL_LM,  # This specifies the task type
+    r=16,                           # Rank of the decomposed matrices
+    lora_alpha=16,                 # Amplitude of the LoRA updates
+    lora_dropout=0.1,              # Dropout for LoRA layers
+)
+model = get_peft_model(model, lora_config)
+
+# print("model:", model)
+# for name, param in model.named_parameters():
+#     if param.requires_grad:
+#         print(f"Trainable parameter: {name}, shape: {param.shape}")
+# === TOKENIZATION ===
+def tokenize_function(examples):
+    return tokenizer(examples['text'], padding="max_length", truncation=True, max_length=1024)
+
+tokenized_train = train_dataset.map(tokenize_function, batched=True, remove_columns=["text"])
+# === DATA COLLATOR ===
+data_collator = DataCollatorForLanguageModeling(tokenizer, mlm=False)
+
+# === TRAINING ARGUMENTS ===
+training_args = TrainingArguments(
+    output_dir=run_dir,
+    # evaluation_strategy="steps", # use with load_best_model_at_end=True  
+    evaluation_strategy="no",
+    per_device_train_batch_size=1,  # Set based on your hardware capabilities
+    logging_dir='./logs',
+    logging_steps=10,
+    save_steps=10,
+    # save_total_limit=5,
+    learning_rate=5e-4,  # You may want to tune this for LoRA
+    num_train_epochs=10,  # Fewer epochs might be sufficient due to the efficiency of LoRA   
+    seed=seed,
+    # load_best_model_at_end=True,
+    gradient_accumulation_steps=1,
+    fp16=True,
+    warmup_ratio=0.1, 
+    # report_to="wandb"
+)
+
+# === TRAINER ===
+trainer = Trainer(
+    model=model,
+    args=training_args,
+    train_dataset=tokenized_train,
+    tokenizer=tokenizer,
+    data_collator=data_collator
+)
+
+# === TRAIN MODEL ===
+trainer.train()
+
+# End logging
+# wandb.finish()
\ No newline at end of file
diff --git a/training/babylm_dataset.py b/training/babylm_dataset.py
new file mode 100644
index 0000000000000000000000000000000000000000..2503c8d992321f8aaa799969d19c57fce051f42f
--- /dev/null
+++ b/training/babylm_dataset.py
@@ -0,0 +1,140 @@
+# babylm_dataset.py
+# author: Julie Kallini
+
+import datasets
+import os
+import glob
+import tqdm
+from numpy.random import default_rng
+from itertools import product
+
+logger = datasets.logging.get_logger(__name__)
+
+_DESCRIPTION = """\
+    Pre-tokenized BabyLM HuggingFace dataset for verb perturbations.
+"""
+_PERTURBED_DATA_PATH = "../data/Qwen_perturbed_data/Qwen2.5-7B"
+_PERTURBATIONS = ["hop_control", "hop_tokens4", "hop_words4",
+                  "reverse_control", "reverse_partial", "reverse_full",
+                  "shuffle_control", "shuffle_nondeterministic",
+                  "shuffle_deterministic21", "shuffle_deterministic57", "shuffle_deterministic84",
+                  "shuffle_local3", "shuffle_local5", "shuffle_local10",
+                  "shuffle_even_odd"]
+# _RANDOM_SEEDS = [0, 14, 41, 53, 96]
+_RANDOM_SEEDS = [0]
+# _TRAIN_SETS = ["100M", "10M"]
+_TRAIN_SETS = ["10M"]
+_EOS_TOKEN_ID = 50256
+
+
+class BabyConfig(datasets.BuilderConfig):
+
+    def __init__(self, data_dir, babylm_train_set, random_seed, **kwargs):
+        """BuilderConfig for IzParens
+
+        Args:
+          data_dir: path to directory of tokenized, perturbed BabyLM dataset
+        """
+        super(BabyConfig, self).__init__(
+            **kwargs,
+        )
+        self.data_dir = data_dir
+        self.babylm_train_set = babylm_train_set
+        self.random_seed = random_seed
+
+
+class BabyLMCorpus(datasets.GeneratorBasedBuilder):
+    BUILDER_CONFIGS = [
+        BabyConfig(
+            name=f"babylm_{perturbation}_{train_set}_seed{random_seed}",
+            data_dir=os.path.join(
+                _PERTURBED_DATA_PATH, "babylm_" + perturbation),
+            babylm_train_set=train_set,
+            random_seed=random_seed,
+        ) for perturbation, train_set, random_seed in list(product(_PERTURBATIONS, _TRAIN_SETS, _RANDOM_SEEDS))
+    ]
+
+    def _info(self):
+        return datasets.DatasetInfo(
+            # This is the description that will appear on the datasets page.
+            description=_DESCRIPTION,
+            # datasets.features.FeatureConnectors
+            features=datasets.Features(
+                {
+                    "text": datasets.Value("string")
+                    # These are the features of your dataset like images, labels ...
+                }
+            ),
+            # If there's a common (input, target) tuple from the features,
+            # specify them here. They'll be used if as_supervised=True in
+            # builder.as_dataset.
+            supervised_keys=None,
+        )
+
+    def _split_generators(self, dl_manager):
+        return [
+            datasets.SplitGenerator(
+                name=datasets.Split.TRAIN,
+                gen_kwargs={"data_dir": os.path.join(
+                    self.config.data_dir, "babylm_" + self.config.babylm_train_set), "random_seed": self.config.random_seed, "split": "train"},
+            ),
+            # datasets.SplitGenerator(
+            #     name=datasets.Split.VALIDATION,
+            #     gen_kwargs={"data_dir": os.path.join(
+            #         self.config.data_dir, "babylm_dev"), "random_seed": self.config.random_seed, "split": "valid"},
+            #
+        ]
+
+    def __chunk(self, sentences, eos_token):
+
+        # Tokenize each sentence
+        logger.info("Loading pre-tokenized data")
+        tokenized_sentences = []
+        for sent in tqdm.tqdm(sentences):
+            tokenized_sentences.append([int(tok) for tok in sent.split()])
+
+        # Concatenate the tokenized sentences using the EOS token
+        logger.info("Concatenating tokenized data using EOS token")
+        all_tokens = []
+        for tokens in tqdm.tqdm(tokenized_sentences):
+            all_tokens.extend(tokens)
+            all_tokens.append(eos_token)
+
+        # Chunk the tokens into sublists of max_seq_len tokens each
+        logger.info("Chunking tokens into sublists of 1024")
+        max_seq_len = 1024
+        chunked_tokens = []
+        for i in tqdm.tqdm(range(0, len(all_tokens), max_seq_len)):
+            chunked_tokens.append(all_tokens[i:i + max_seq_len])
+
+        # Drop last line if not a multiple of max_seq_len
+        if len(chunked_tokens[-1]) < max_seq_len:
+            chunked_tokens.pop()
+
+        return chunked_tokens
+
+    def _generate_examples(self, data_dir, random_seed, split):
+        """This function returns the BabyLM text in the discretized, tokenized form."""
+
+        logger.info("Generating examples from = %s", data_dir)
+        infiles = sorted(glob.glob(os.path.join(data_dir, "*")))
+
+        # Extend sentences
+        all_sentences = []
+        for infile in infiles:
+            f = open(infile, encoding="utf-8")
+            all_sentences.extend(f.readlines())
+        logger.info("Total sentences: {}".format(len(all_sentences)))
+
+        # Shuffle because we are pre-tokenizing
+        rng = default_rng(seed=random_seed)
+        rng.shuffle(all_sentences)
+
+        # Tokenize and chunk
+        tokenized_lines = self.__chunk(all_sentences, _EOS_TOKEN_ID)
+
+        # Generate data
+        logger.info("Writing dataset as space-separated sequences of tokens")
+        for idx, line in enumerate(tokenized_lines):
+            l = " ".join([str(tok) for tok in line]) + "\n"
+            yield idx, {"text": l}
\ No newline at end of file
diff --git a/training/conf/template/babylm_dataset_template.yaml b/training/conf/template/babylm_dataset_template.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..de98aa530c9a73686b754dfb84ad86270cd3913c
--- /dev/null
+++ b/training/conf/template/babylm_dataset_template.yaml
@@ -0,0 +1,13 @@
+# dataset_{{ perturbation }}_{{ train_set }}_seed{{ seed }}.yaml
+#   Configuration for altered babylm-dataset
+---
+dataset:
+    id: /nlp/scr/kallini/llms-in-llms/training/babylm_dataset.py
+    name: babylm_{{ perturbation }}_{{ train_set }}_seed{{ seed }}
+    validation_ratio: null
+
+    # Number of Preprocessing Workers
+    num_proc: 4
+
+    # Number of Evaluation Preprocessing Workers
+    eval_num_proc: 4
\ No newline at end of file
diff --git a/training/conf/template/babylm_train_template.yaml b/training/conf/template/babylm_train_template.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..a9a7404d67135f379d8b9f64445377bcce0608ad
--- /dev/null
+++ b/training/conf/template/babylm_train_template.yaml
@@ -0,0 +1,49 @@
+# train_{{ perturbation }}_{{ train_set }}_{{ paren_model }}{{ no_pos_encodings_underscore }}_seed{{ seed }}.yaml
+#   Based on mistral-small.yaml.
+---
+# Inherit Dataset, Tokenization, Model, and Training Details
+inherit:
+    - datasets/dataset_{{ perturbation }}_{{ train_set }}_seed{{ seed }}.yaml
+    - models/gpt2{{ no_pos_encodings }}-small-{{ perturbation }}-{{ paren_model }}.yaml
+    - trainers/gpt2-small.yaml
+
+# Run ID -- make sure to override!
+run_id: babylm_{{ perturbation }}_{{ train_set }}_{{ paren_model }}{{ no_pos_encodings_underscore }}_seed{{ seed }}
+
+# Weights & Biases
+wandb: kallini
+group: babylm-perturbation-experiments
+
+# Artifacts & Caching
+artifacts:
+    cache_dir: {{ ckpt_path }}/babylm_{{ perturbation }}_{{ train_set }}_{{ paren_model }}{{ no_pos_encodings_underscore }}/babylm_{{ perturbation }}_{{ train_set }}_{{ paren_model }}{{ no_pos_encodings_underscore }}_seed{{ seed }}/artifacts
+    run_dir: {{ ckpt_path }}/babylm_{{ perturbation }}_{{ train_set }}_{{ paren_model }}{{ no_pos_encodings_underscore }}/babylm_{{ perturbation }}_{{ train_set }}_{{ paren_model }}{{ no_pos_encodings_underscore }}_seed{{ seed }}/runs
+
+# Save Effective Batch Size for Easy Handling ==> Main Code asserts infra + training_config results in this!
+effective_bsz: 512
+
+# Resume from Checkpoint
+resume: false
+resume_checkpoint: null
+
+# List of frequencies at which to save checkpoints, provided as a list of two-element tuples:
+#   - Frequency (`freq`) at which to save checkpoints (# steps)
+#   - Bound (`until`) on global step for given frequency (checkpoint every `freq` steps until global step = `until`)
+checkpoint_frequency:
+    - [100, 3000]
+
+# `torch.distributed` Default Infra Parameters -- to be overwritten by call to `torch.distributed.launch`
+local_rank: -1
+nnodes: -1
+nproc_per_node: -1
+
+# DeepSpeed Default Infra Parameters -- to be overwritten by call to `DeepSpeed`
+num_gpus: -1
+num_nodes: -1
+world_size: -1
+
+# Logging Parameters -- 10 = DEBUG, 20 = INFO, 30 = WARNING, 40 = ERROR, 50 = CRITICAL
+log_level: 20
+
+# Random Seed
+seed: {{ seed }}
\ No newline at end of file
diff --git a/training/conf/template/gpt2-small-template.yaml b/training/conf/template/gpt2-small-template.yaml
new file mode 100644
index 0000000000000000000000000000000000000000..c9e54a22abe0d562ac81387c1d86d41d08d7b450
--- /dev/null
+++ b/training/conf/template/gpt2-small-template.yaml
@@ -0,0 +1,22 @@
+# gpt2{{ no_pos_encodings }}-small-{{ perturbation }}-{{ paren_model }}.yaml
+#   Configuration for the GPT-2 Small Model.
+---
+model:
+    id: "gpt2{{ no_pos_encodings }}-small"
+
+    # Boolean whether to use the pre-existing Hugging Face AutoTokenizer (or train a new one from scratch)
+    pretrained_tokenizer: false
+    passthrough_tokenizer: true
+
+    # Sequence Length
+    seq_len: 1024
+
+    # Stability
+    reorder_and_upcast_attn: true
+    scale_attn_by_inverse_layer_idx: true
+
+    # Initialize Weights from File
+    initial_weights: {{ paren_model_path }}
+    
+    # Configure Model From File
+    config_path: /nlp/scr/kallini/mistral/conf/models/gpt2-small-{{ vocab_size }}.json
\ No newline at end of file
diff --git a/training/generate_yaml.py b/training/generate_yaml.py
new file mode 100644
index 0000000000000000000000000000000000000000..da1878517c622823bdd4dfbd454e935fd1f14ce2
--- /dev/null
+++ b/training/generate_yaml.py
@@ -0,0 +1,131 @@
+# generate_yaml.py
+# Author: Julie Kallini
+
+# For importing utils
+import sys
+sys.path.append("..")
+
+from jinja2 import Template
+from utils import PERTURBATIONS, CHECKPOINT_WRITE_PATH, \
+    PAREN_MODELS, PAREN_MODEL_PATH
+import argparse
+import os
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser(
+        prog='Generate yaml for training',
+        description='Generate train and dataset yaml configs for mistral training')
+    parser.add_argument('perturbation_type',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=PERTURBATIONS.keys(),
+                        help='Perturbation function used to transform BabyLM dataset')
+    parser.add_argument('train_set',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=["100M", "10M"],
+                        help='BabyLM train set')
+    parser.add_argument('random_seed', type=int, help="Random seed")
+    parser.add_argument('paren_model',
+                        default='all',
+                        const='all',
+                        nargs='?',
+                        choices=list(PAREN_MODELS.keys()) + ["randinit"],
+                        help='Parenthesis model')
+    parser.add_argument('-np', '--no_pos_encodings', action='store_true',
+                        help="Train GPT-2 with no positional encodings")
+
+    # Get args
+    args = parser.parse_args()
+    if args.paren_model != "randinit":
+        paren_model_path = PAREN_MODEL_PATH + PAREN_MODELS[args.paren_model] + "/checkpoint-5000"
+    else:
+        paren_model_path = "null"
+    paren_model_name = args.paren_model
+    no_pos_encodings_str = "-no-positional-encodings" if args.no_pos_encodings else ""
+    no_pos_encodings_underscore = "_no_positional_encodings" if args.no_pos_encodings else ""
+
+    # Create directory for yaml
+    yaml_directory = f"conf/babylm_{args.perturbation_type}_{args.train_set}_{paren_model_name}{no_pos_encodings_underscore}/seed{args.random_seed}"
+    if not os.path.exists(yaml_directory):
+        os.makedirs(yaml_directory)
+
+    print("Generating GPT-2 model yaml file...")
+
+    # Get model template, which varies due to changes in vocab size
+    model_temp_file = open("conf/template/gpt2-small-template.yaml")
+    lines = model_temp_file.readlines()
+    model_temp_file.close()
+
+    # Fill model template
+    tokenizer = PERTURBATIONS[args.perturbation_type]["gpt2_tokenizer"]
+    vocab_size = len(tokenizer)
+    model_template = Template("".join(lines))
+    model_conf = model_template.render(
+        perturbation=args.perturbation_type,
+        vocab_size=vocab_size,
+        paren_model=paren_model_name,
+        paren_model_path=paren_model_path,
+        no_pos_encodings=no_pos_encodings_str,
+    )
+
+    # Write model yaml to file
+    model_file = open(
+        f"conf/babylm_{args.perturbation_type}_{args.train_set}_{paren_model_name}{no_pos_encodings_underscore}/gpt2{no_pos_encodings_str}-small-{args.perturbation_type}-{paren_model_name}.yaml", "w")
+    model_file.write(model_conf)
+    model_file.close()
+
+    print("Generating train yaml file...")
+
+    # Get train template file
+    train_temp_file = open("conf/template/babylm_train_template.yaml")
+    lines = train_temp_file.readlines()
+    train_temp_file.close()
+
+    # Fill train template file
+    train_template = Template("".join(lines))
+    train_conf = train_template.render(
+        perturbation=args.perturbation_type,
+        seed=args.random_seed,
+        ckpt_path=CHECKPOINT_WRITE_PATH,
+        train_set=args.train_set,
+        paren_model=paren_model_name,
+        no_pos_encodings=no_pos_encodings_str,
+        no_pos_encodings_underscore=no_pos_encodings_underscore,
+    )
+
+    # Write train yaml to file
+    train_file = open(yaml_directory + \
+        f"/train_{args.perturbation_type}_{args.train_set}_{paren_model_name}{no_pos_encodings_underscore}_seed{args.random_seed}.yaml", "w")
+    train_file.write(train_conf)
+    train_file.close()
+
+    print("Generating dataset yaml file...")
+
+    # Get dataset temp file
+    dataset_temp_file = open("conf/template/babylm_dataset_template.yaml")
+    lines = dataset_temp_file.readlines()
+    dataset_temp_file.close()
+
+    # Fill dataset template file
+    dataset_template = Template("".join(lines))
+    dataset_conf = dataset_template.render(
+        perturbation=args.perturbation_type,
+        train_set=args.train_set,
+        seed=args.random_seed,
+    )
+
+    # Write dataset yaml to file
+    dataset_file = open(yaml_directory + \
+        f"/dataset_{args.perturbation_type}_{args.train_set}_seed{args.random_seed}.yaml", "w")
+    dataset_file.write(dataset_conf)
+    dataset_file.close()
+
+    # Create directory for model checkpoints
+    ckpt_directory = CHECKPOINT_WRITE_PATH + f"/babylm_{args.perturbation_type}_{args.train_set}_{paren_model_name}{no_pos_encodings_underscore}"
+    if not os.path.exists(ckpt_directory):
+        os.makedirs(ckpt_directory)
\ No newline at end of file
diff --git a/training/prepare_training.sh b/training/prepare_training.sh
new file mode 100644
index 0000000000000000000000000000000000000000..efb59c82b1512a68cf161887139960a4797c4349
--- /dev/null
+++ b/training/prepare_training.sh
@@ -0,0 +1,63 @@
+#!/bin/sh
+# prepare_training.sh
+# author: Julie Kallini
+
+readonly MISTRAL_PATH=/nlp/scr/kallini/mistral
+
+echo "
+-------------------------------------------------------------------------------
+Arguments
+-------------------------------------------------------------------------------
+"
+echo "Perturbation type: $1"
+echo "Train set: $2"
+echo "Random seed: $3"
+echo "Paren pretrained model: $4"
+NO_POS_ENCODINGS=${5:-''}
+echo "No pos encodings: $NO_POS_ENCODINGS"
+echo "Mistral path: $MISTRAL_PATH"
+
+if [ -z "$NO_POS_ENCODINGS" ]
+then
+    NPS=""
+    NPSunderscore=""
+else
+    NPS="-no-positional-encodings"
+    NPSunderscore="_no_positional_encodings"
+fi
+
+# Generate yaml files for mistral training
+echo "
+-------------------------------------------------------------------------------
+Generating yaml files for mistral training
+-------------------------------------------------------------------------------
+"
+GENERATE_YAML_COMMAND="python3 generate_yaml.py $1 $2 $3 $4 $NO_POS_ENCODINGS"
+echo $GENERATE_YAML_COMMAND
+eval $GENERATE_YAML_COMMAND
+
+# Copy yaml files to mistral directory
+echo "
+-------------------------------------------------------------------------------
+Copying config yaml files to mistral directory
+-------------------------------------------------------------------------------
+"
+COPY_DATASET_COMMAND="cp conf/babylm_$1_$2_$4$NPSunderscore/seed$3/dataset_$1_$2_seed$3.yaml $MISTRAL_PATH/conf/datasets/dataset_$1_$2_seed$3.yaml"
+echo $COPY_DATASET_COMMAND
+eval $COPY_DATASET_COMMAND
+echo ""
+
+COPY_TRAIN_COMMAND="cp conf/babylm_$1_$2_$4$NPSunderscore/seed$3/train_$1_$2_$4$NPSunderscore"_"seed$3.yaml $MISTRAL_PATH/conf/train_$1_$2_$4$NPSunderscore"_"seed$3.yaml"
+echo $COPY_TRAIN_COMMAND
+eval $COPY_TRAIN_COMMAND
+echo ""
+
+COPY_MODEL_COMMAND="cp conf/babylm_$1_$2_$4$NPSunderscore/gpt2$NPS-small-$1-$4.yaml $MISTRAL_PATH/conf/models/gpt2$NPS-small-$1-$4.yaml"
+echo $COPY_MODEL_COMMAND
+eval $COPY_MODEL_COMMAND
+
+echo "
+-------------------------------------------------------------------------------
+Done!
+-------------------------------------------------------------------------------
+"
diff --git a/utils.py b/utils.py
new file mode 100644
index 0000000000000000000000000000000000000000..2c589d4596780175c9fb6790ca0ba39840d02c5e
--- /dev/null
+++ b/utils.py
@@ -0,0 +1,1144 @@
+# utils_qwen.py
+# Author: Yaning
+
+from collections import deque
+from string import punctuation
+from transformers import AutoTokenizer, AddedToken
+from functools import partial
+from numpy.random import default_rng
+from nltk.tree import ParentedTree
+import torch
+
+
+##############################################################################
+# CONSTANTS
+##############################################################################
+
+
+BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# Yj: 用于在参数解析和数据加载时指定数据集
+# 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+SEEDS = [21, 57, 84]
+CHECKPOINTS = list(range(50, 501, 50))
+GENRES = {
+    "aochildes": "CHILDES",
+    "bnc_spoken": "British National Corpus (BNC)",
+    "cbt": "Children’s Book Test",
+    "children_stories": "Children’s Stories Text Corpus",
+    "gutenberg": "Standardized Project Gutenberg Corpus",
+    "open_subtitles": "OpenSubtitles",
+    "qed": "QCRI Educational Domain Corpus",
+    "simple_wikipedia": "Simple Wikipedia",
+    "switchboard": "Switchboard Dialog Act Corpus",
+    "wikipedia": "Wikipedia"
+}
+CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+BABYLM_DATA_PATH = "."
+MARKER_HOP_SING = "🅂"
+MARKER_HOP_PLUR = "🄿"
+MARKER_REV = "🅁"
+BOS_TOKEN = "<BOS_TOKEN>"
+PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+PUNCT_TOKENS = set(punctuation)
+
+MODEL_NAME = "gpt2"  
+
+
+##############################################################################
+# PARENS MODELS (Structurally-pretrained)
+##############################################################################
+
+
+PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+PAREN_MODELS = {
+    "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+    "NEST": "nested-parens0.49_vocab500-uniform",
+    "RAND": "random_vocab500-uniform",
+}
+
+
+##############################################################################
+# HELPER FUNCTIONS
+##############################################################################
+
+
+def write_file(directory, filename, lines):
+    f = open(directory + filename, "w")
+    f.writelines(lines)
+    f.close()
+
+
+def get_qwen_tokenizer_with_markers(marker_list):
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+
+    # If no new markers to add, return normal tokenizer
+    if len(marker_list) == 0:
+        return tokenizer
+
+    # Create tokens and return modified tokenizer
+    new_tokens = []
+    for marker in marker_list:
+        new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+    tokenizer.add_tokens(new_tokens)
+    return tokenizer
+
+
+qwen_original_tokenizer = get_qwen_tokenizer_with_markers([])
+
+
+# GPT-2 hop tokenization
+qwen_hop_tokenizer = get_qwen_tokenizer_with_markers(
+    [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# Get ids of marker tokens
+marker_sg_token = qwen_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_SING]
+# Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+marker_pl_token = qwen_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_PLUR]
+
+
+# Qwen reverse tokenization
+qwen_rev_tokenizer = get_qwen_tokenizer_with_markers(
+    [MARKER_REV])
+# Get ids of marker tokens
+marker_rev_token = qwen_rev_tokenizer.get_added_vocab()[
+    MARKER_REV]
+
+# Qwen determiner tokenization
+qwen_det_tokenizer = get_qwen_tokenizer_with_markers(
+    [BOS_TOKEN])
+# Get id of BOS token
+bos_token_id = qwen_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+
+def compute_surprisals(model, input_ids):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        shifted_input_ids = input_ids[:, 1:]
+
+        # Get the log probabilities for the actual next tokens
+        log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+        true_log_probs = log_probs.gather(
+            2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+    # Get the negative log probabilities
+    neg_log_probs = (-true_log_probs).tolist()
+    surprisals = [[None] + probs for probs in neg_log_probs]
+    return surprisals
+
+
+def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        probs = torch.nn.functional.softmax(logits, dim=-1)
+
+        # Get the probabilities for the specified token at each position
+        token_probs = probs[:, :, token_id]
+
+    # Convert to list and add None at the beginning to align with input tokens
+    # Put null probability for instances of pad token
+    token_probs_list = []
+    for batch_i, probs in enumerate(token_probs):
+        input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+        filtered = [p if input_ids_seq[pos_i+1] !=
+                    pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+        token_probs_list.append([None] + filtered)
+
+    return token_probs_list
+
+
+def merge_part_tokens(words):
+    result = []
+    for s in words:
+        if result and s in PART_TOKENS and len(result) > 0:
+            result[-1] += s
+        else:
+            result.append(s)
+    return result
+
+
+def __affect_hop_word(word):
+    return word["feats"] and "Person=3" in word["feats"] \
+        and "Tense=Pres" in word["feats"] \
+        and "VerbForm=Fin" in word["feats"] \
+        and "Number" in word["feats"]
+
+
+def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+    perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker_sg, marker_pl)
+    return perturbed_tokens
+
+
+def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+    _, hops_completed = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker, marker)
+    return hops_completed
+
+
+def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    hop_completed = []
+    new_sent = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            new_sent.append(
+                word["lemma"] if word["lemma"] is not None else word["text"])
+
+            # Marker hopping logic
+            insert_index = len(new_sent)-1
+            skipped_words = 0
+            while skipped_words < num_hops and insert_index > 0:
+
+                # Handle edge case when punctuation (or sequence of
+                # punctuation) begin the sentence
+                if (not any([c.isalnum() for c in
+                             "".join(new_sent[:insert_index])])):
+                    break
+                
+                # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+                # Count word as skipped if it is not a special token
+                if (new_sent[insert_index] not in PART_TOKENS) and \
+                        (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    skipped_words += 1
+                insert_index -= 1
+
+            # Handle edge case when insert index is punctuation (and this is not
+            # sentence-initial punctuation)
+            if any([c.isalnum() for c in
+                    "".join(new_sent[:insert_index])]):
+                while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+                                             or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    insert_index -= 1
+
+            # Handle edge case when token before insert index is part/aux token
+            if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+                insert_index -= 1
+
+            # Log if this sentence had all full hops
+            hop_completed.append(skipped_words == num_hops)
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                new_sent.insert(insert_index, marker_sg)
+            elif "Number=Plur" in word["feats"]:
+                new_sent.insert(insert_index, marker_pl)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+        else:
+            new_sent.append(word["text"])
+
+    new_sent.reverse()
+    sent_string = " ".join(merge_part_tokens(new_sent))
+    tokens = qwen_hop_tokenizer.encode(sent_string)
+    return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+def __perturb_hop_tokens(sent, num_hops):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    new_sent = deque()
+    tokens = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+            if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+                lemma = lemma + new_sent[0]
+                new_sent.popleft()
+
+            if len(new_sent) > 0:
+                sent_string = " ".join(merge_part_tokens(new_sent))
+                tokens = qwen_hop_tokenizer.encode(
+                    " " + sent_string) + tokens
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                tokens.insert(num_hops, marker_sg_token)
+            elif "Number=Plur" in word["feats"]:
+                tokens.insert(num_hops, marker_pl_token)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+            new_sent = deque()
+            new_sent.append(lemma)
+
+        else:
+            new_sent.appendleft(word["text"])
+
+    if len(new_sent) > 0:
+        sent_string = " ".join(merge_part_tokens(new_sent))
+        tokens = qwen_hop_tokenizer.encode(sent_string) + tokens
+    return tokens
+
+
+def __perturb_reverse(sent, rng, reverse, full):
+
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_rev_tokenizer.encode(sent["sent_text"])
+
+    # Pick random index to insert REV token
+    i = rng.choice(len(tokens)+1)
+    tokens.insert(i, marker_rev_token)
+
+    # Extract tokens before/after the marker, and reverse tokens after
+    tokens_before = tokens[:i+1]
+    tokens_after = tokens[i+1:]
+    if reverse:
+        tokens_after.reverse()
+    new_tokens = tokens_before + tokens_after
+    if full:
+        assert not reverse
+        new_tokens.reverse()
+
+    return new_tokens
+
+
+def __perturb_shuffle_deterministic(sent, seed, shuffle):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    if shuffle:
+        default_rng(seed).shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_nondeterministic(sent, rng):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    rng.shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_local(sent, seed, window=5):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+
+    # Shuffle tokens in batches of size window
+    shuffled_tokens = []
+    for i in range(0, len(tokens), window):
+        batch = tokens[i:i+window].copy()
+        default_rng(seed).shuffle(batch)
+        shuffled_tokens += batch
+
+    return shuffled_tokens
+
+
+def __perturb_shuffle_even_odd(sent):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+    odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+    return even + odd
+
+
+##############################################################################
+# AFFECT FUNCTIONS
+# These functions define when a perturbation has been applied to a sentence
+# not. This is used for identifying which test sentences have been
+# altered to separate affected vs. unaffected senences. Affect functions are
+# functions of the input sentence object and return a boolean.
+##############################################################################
+
+
+def affect_hop(sent):
+    return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+        and sent["constituency_parse"] is not None
+
+
+def affect_reverse(sent):
+    return True
+
+
+def affect_shuffle(sent):
+    return True
+
+
+def affect_none(sent):
+    return False
+
+
+##############################################################################
+# FILTER FUNCTIONS
+# These functions define when an affected sentence should be included in the
+# final dataset. For instance, hop perturbations where the marker is placed
+# at the end of the sentence should be excluded. A filter function returns
+# True if an affected sentence should be included in the dataset.
+##############################################################################
+
+
+def filter_hop(sent):
+    # Assertion needed since filter function is only defined for affected
+    # sentences
+    assert (affect_hop(sent))
+    return check_word_hops_completed(sent, 4)
+
+
+def filter_reverse(sent):
+    return True
+
+
+def filter_shuffle(sent):
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    return len(tokens) > 1 and len(tokens) <= 350
+
+
+def filter_none(sent):
+    return False
+
+
+##############################################################################
+# PERTURBATION FUNCTIONS
+# These functions define how a perturbation will affect a sentence. They
+# take in a sentence object and an optional marker
+# for verb transformations. They return a string representing the transformed
+# sentence.
+##############################################################################
+
+
+def perturb_hop_words4(sent):
+    return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+def perturb_hop_tokens4(sent):
+    return __perturb_hop_tokens(sent, 4)
+
+
+def perturb_hop_control(sent):
+    return __perturb_hop_tokens(sent, 0)
+
+
+def perturb_reverse(sent, rng, reverse=True, full=False):
+    return __perturb_reverse(sent, rng, reverse, full)
+
+
+def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+    return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+def perturb_shuffle_nondeterministic(sent, rng):
+    return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+def perturb_shuffle_local(sent, seed, window):
+    return __perturb_shuffle_local(sent, seed, window)
+
+
+def perturb_shuffle_even_odd(sent):
+    return __perturb_shuffle_even_odd(sent)
+
+
+##############################################################################
+# PERTURBATIONS
+# This dict maps the name of a perturbation to its perturbation and filter
+# functions. The names and functions in this dict are used throughout the
+# repo.
+##############################################################################
+
+
+PERTURBATIONS = {
+    "shuffle_control": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#606060",                                           
+    },
+    "shuffle_nondeterministic": {
+        "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#E8384F",
+    },
+    "shuffle_deterministic21": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#FFB000",
+    },
+    "shuffle_deterministic57": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#8db000",
+    },
+    "shuffle_deterministic84": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#62BB35",
+    },
+    "shuffle_local3": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#208EA3",
+    },
+    "shuffle_local5": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#4178BC",
+    },
+    "shuffle_local10": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#AA71FF",
+    },
+    "shuffle_even_odd": {
+        "perturbation_function": perturb_shuffle_even_odd,
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#E37CFF",
+    },
+    "reverse_control": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "qwen_tokenizer": qwen_rev_tokenizer,
+        "color": "#606060",
+    },
+    "reverse_partial": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "qwen_tokenizer": qwen_rev_tokenizer,
+        "color": "#E5A836",
+    },
+    "reverse_full": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "qwen_tokenizer": qwen_rev_tokenizer,
+        "color": "#A348A6",
+    },
+    "hop_control": {
+        "perturbation_function": perturb_hop_control,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "qwen_tokenizer": qwen_hop_tokenizer,
+        "color": "#606060",
+    },
+    "hop_tokens4": {
+        "perturbation_function": perturb_hop_tokens4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "qwen_tokenizer": qwen_hop_tokenizer,
+        "color": "#fa8128", 
+    },
+    "hop_words4": {
+        "perturbation_function": perturb_hop_words4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "qwen_tokenizer": qwen_hop_tokenizer,
+        "color": "#03a0ff",
+    },
+}
+
+
+# # utils.py
+# # Author: Julie Kallini
+
+# from collections import deque
+# from string import punctuation
+# from transformers import AutoTokenizer, AddedToken
+# from functools import partial
+# from numpy.random import default_rng
+# from nltk.tree import ParentedTree
+# import torch
+
+
+# ##############################################################################
+# # CONSTANTS
+# ##############################################################################
+
+
+# BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# # Yj: 用于在参数解析和数据加载时指定数据集
+# # 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+# SEEDS = [21, 57, 84]
+# CHECKPOINTS = list(range(50, 501, 50))
+# GENRES = {
+#     "aochildes": "CHILDES",
+#     "bnc_spoken": "British National Corpus (BNC)",
+#     "cbt": "Children’s Book Test",
+#     "children_stories": "Children’s Stories Text Corpus",
+#     "gutenberg": "Standardized Project Gutenberg Corpus",
+#     "open_subtitles": "OpenSubtitles",
+#     "qed": "QCRI Educational Domain Corpus",
+#     "simple_wikipedia": "Simple Wikipedia",
+#     "switchboard": "Switchboard Dialog Act Corpus",
+#     "wikipedia": "Wikipedia"
+# }
+# CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# # BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+# BABYLM_DATA_PATH = "."
+# MARKER_HOP_SING = "🅂"
+# MARKER_HOP_PLUR = "🄿"
+# MARKER_REV = "🅁"
+# BOS_TOKEN = "<BOS_TOKEN>"
+# PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+# PUNCT_TOKENS = set(punctuation)
+
+
+# ##############################################################################
+# # PARENS MODELS (Structurally-pretrained)
+# ##############################################################################
+
+
+# PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+# PAREN_MODELS = {
+#     "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+#     "NEST": "nested-parens0.49_vocab500-uniform",
+#     "RAND": "random_vocab500-uniform",
+# }
+
+
+# ##############################################################################
+# # HELPER FUNCTIONS
+# ##############################################################################
+
+
+# def write_file(directory, filename, lines):
+#     f = open(directory + filename, "w")
+#     f.writelines(lines)
+#     f.close()
+
+
+# def get_gpt2_tokenizer_with_markers(marker_list):
+#     tokenizer = AutoTokenizer.from_pretrained("gpt2")
+
+#     # If no new markers to add, return normal tokenizer
+#     if len(marker_list) == 0:
+#         return tokenizer
+
+#     # Create tokens and return modified tokenizer
+#     new_tokens = []
+#     for marker in marker_list:
+#         new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+#     tokenizer.add_tokens(new_tokens)
+#     return tokenizer
+
+
+# gpt2_original_tokenizer = get_gpt2_tokenizer_with_markers([])
+
+
+# # GPT-2 hop tokenization
+# gpt2_hop_tokenizer = get_gpt2_tokenizer_with_markers(
+#     [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# # Get ids of marker tokens
+# marker_sg_token = gpt2_hop_tokenizer.get_added_vocab()[
+#     MARKER_HOP_SING]
+# # Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+# marker_pl_token = gpt2_hop_tokenizer.get_added_vocab()[
+#     MARKER_HOP_PLUR]
+
+
+# # GPT-2 reverse tokenization
+# gpt2_rev_tokenizer = get_gpt2_tokenizer_with_markers(
+#     [MARKER_REV])
+# # Get ids of marker tokens
+# marker_rev_token = gpt2_rev_tokenizer.get_added_vocab()[
+#     MARKER_REV]
+
+# # GPT-2 determiner tokenization
+# gpt2_det_tokenizer = get_gpt2_tokenizer_with_markers(
+#     [BOS_TOKEN])
+# # Get id of BOS token
+# bos_token_id = gpt2_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+# MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+
+# def compute_surprisals(model, input_ids):
+#     # Get the log probabilities from the model
+#     with torch.no_grad():
+#         outputs = model(input_ids)
+#         logits = outputs.logits[:, :-1]
+#         shifted_input_ids = input_ids[:, 1:]
+
+#         # Get the log probabilities for the actual next tokens
+#         log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+#         true_log_probs = log_probs.gather(
+#             2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+#     # Get the negative log probabilities
+#     neg_log_probs = (-true_log_probs).tolist()
+#     surprisals = [[None] + probs for probs in neg_log_probs]
+#     return surprisals
+
+
+# def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+#     # Get the log probabilities from the model
+#     with torch.no_grad():
+#         outputs = model(input_ids)
+#         logits = outputs.logits[:, :-1]
+#         probs = torch.nn.functional.softmax(logits, dim=-1)
+
+#         # Get the probabilities for the specified token at each position
+#         token_probs = probs[:, :, token_id]
+
+#     # Convert to list and add None at the beginning to align with input tokens
+#     # Put null probability for instances of pad token
+#     token_probs_list = []
+#     for batch_i, probs in enumerate(token_probs):
+#         input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+#         filtered = [p if input_ids_seq[pos_i+1] !=
+#                     pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+#         token_probs_list.append([None] + filtered)
+
+#     return token_probs_list
+
+
+# def merge_part_tokens(words):
+#     result = []
+#     for s in words:
+#         if result and s in PART_TOKENS and len(result) > 0:
+#             result[-1] += s
+#         else:
+#             result.append(s)
+#     return result
+
+
+# def __affect_hop_word(word):
+#     return word["feats"] and "Person=3" in word["feats"] \
+#         and "Tense=Pres" in word["feats"] \
+#         and "VerbForm=Fin" in word["feats"] \
+#         and "Number" in word["feats"]
+
+
+# def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+#     perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+#         sent, num_hops, marker_sg, marker_pl)
+#     return perturbed_tokens
+
+
+# def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+#     _, hops_completed = __perturb_hop_words_complete_hops(
+#         sent, num_hops, marker, marker)
+#     return hops_completed
+
+
+# def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+#     word_annotations = sent["word_annotations"].copy()
+#     word_annotations.reverse()
+
+#     hop_completed = []
+#     new_sent = []
+#     for word in word_annotations:
+
+#         # Identify 3.pres verbs
+#         if __affect_hop_word(word):
+
+#             # Lemmatize verb if possible
+#             new_sent.append(
+#                 word["lemma"] if word["lemma"] is not None else word["text"])
+
+#             # Marker hopping logic
+#             insert_index = len(new_sent)-1
+#             skipped_words = 0
+#             while skipped_words < num_hops and insert_index > 0:
+
+#                 # Handle edge case when punctuation (or sequence of
+#                 # punctuation) begin the sentence
+#                 if (not any([c.isalnum() for c in
+#                              "".join(new_sent[:insert_index])])):
+#                     break
+                
+#                 # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+#                 # Count word as skipped if it is not a special token
+#                 if (new_sent[insert_index] not in PART_TOKENS) and \
+#                         (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+#                     skipped_words += 1
+#                 insert_index -= 1
+
+#             # Handle edge case when insert index is punctuation (and this is not
+#             # sentence-initial punctuation)
+#             if any([c.isalnum() for c in
+#                     "".join(new_sent[:insert_index])]):
+#                 while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+#                                              or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+#                     insert_index -= 1
+
+#             # Handle edge case when token before insert index is part/aux token
+#             if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+#                 insert_index -= 1
+
+#             # Log if this sentence had all full hops
+#             hop_completed.append(skipped_words == num_hops)
+
+#             # Use correct marker for singular vs. plural
+#             if "Number=Sing" in word["feats"]:
+#                 new_sent.insert(insert_index, marker_sg)
+#             elif "Number=Plur" in word["feats"]:
+#                 new_sent.insert(insert_index, marker_pl)
+#             else:
+#                 raise Exception(
+#                     "Number not in verb features\n" + sent["sent_text"])
+
+#         else:
+#             new_sent.append(word["text"])
+
+#     new_sent.reverse()
+#     sent_string = " ".join(merge_part_tokens(new_sent))
+#     tokens = gpt2_hop_tokenizer.encode(sent_string)
+#     return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+# def __perturb_hop_tokens(sent, num_hops):
+
+#     word_annotations = sent["word_annotations"].copy()
+#     word_annotations.reverse()
+
+#     new_sent = deque()
+#     tokens = []
+#     for word in word_annotations:
+
+#         # Identify 3.pres verbs
+#         if __affect_hop_word(word):
+
+#             # Lemmatize verb if possible
+#             lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+#             if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+#                 lemma = lemma + new_sent[0]
+#                 new_sent.popleft()
+
+#             if len(new_sent) > 0:
+#                 sent_string = " ".join(merge_part_tokens(new_sent))
+#                 tokens = gpt2_hop_tokenizer.encode(
+#                     " " + sent_string) + tokens
+
+#             # Use correct marker for singular vs. plural
+#             if "Number=Sing" in word["feats"]:
+#                 tokens.insert(num_hops, marker_sg_token)
+#             elif "Number=Plur" in word["feats"]:
+#                 tokens.insert(num_hops, marker_pl_token)
+#             else:
+#                 raise Exception(
+#                     "Number not in verb features\n" + sent["sent_text"])
+
+#             new_sent = deque()
+#             new_sent.append(lemma)
+
+#         else:
+#             new_sent.appendleft(word["text"])
+
+#     if len(new_sent) > 0:
+#         sent_string = " ".join(merge_part_tokens(new_sent))
+#         tokens = gpt2_hop_tokenizer.encode(sent_string) + tokens
+#     return tokens
+
+
+# def __perturb_reverse(sent, rng, reverse, full):
+
+#     # Get sentence text and GPT-2 tokens
+#     tokens = gpt2_rev_tokenizer.encode(sent["sent_text"])
+
+#     # Pick random index to insert REV token
+#     i = rng.choice(len(tokens)+1)
+#     tokens.insert(i, marker_rev_token)
+
+#     # Extract tokens before/after the marker, and reverse tokens after
+#     tokens_before = tokens[:i+1]
+#     tokens_after = tokens[i+1:]
+#     if reverse:
+#         tokens_after.reverse()
+#     new_tokens = tokens_before + tokens_after
+#     if full:
+#         assert not reverse
+#         new_tokens.reverse()
+
+#     return new_tokens
+
+
+# def __perturb_shuffle_deterministic(sent, seed, shuffle):
+#     # Get sentence text and GPT-2 tokens
+#     tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+#     if shuffle:
+#         default_rng(seed).shuffle(tokens)
+#     return tokens
+
+
+# def __perturb_shuffle_nondeterministic(sent, rng):
+#     # Get sentence text and GPT-2 tokens
+#     tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+#     rng.shuffle(tokens)
+#     return tokens
+
+
+# def __perturb_shuffle_local(sent, seed, window=5):
+#     # Get sentence text and GPT-2 tokens
+#     tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+
+#     # Shuffle tokens in batches of size window
+#     shuffled_tokens = []
+#     for i in range(0, len(tokens), window):
+#         batch = tokens[i:i+window].copy()
+#         default_rng(seed).shuffle(batch)
+#         shuffled_tokens += batch
+
+#     return shuffled_tokens
+
+
+# def __perturb_shuffle_even_odd(sent):
+#     # Get sentence text and GPT-2 tokens
+#     tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+#     even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+#     odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+#     return even + odd
+
+
+# ##############################################################################
+# # AFFECT FUNCTIONS
+# # These functions define when a perturbation has been applied to a sentence
+# # not. This is used for identifying which test sentences have been
+# # altered to separate affected vs. unaffected senences. Affect functions are
+# # functions of the input sentence object and return a boolean.
+# ##############################################################################
+
+
+# def affect_hop(sent):
+#     return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+#         and sent["constituency_parse"] is not None
+
+
+# def affect_reverse(sent):
+#     return True
+
+
+# def affect_shuffle(sent):
+#     return True
+
+
+# def affect_none(sent):
+#     return False
+
+
+# ##############################################################################
+# # FILTER FUNCTIONS
+# # These functions define when an affected sentence should be included in the
+# # final dataset. For instance, hop perturbations where the marker is placed
+# # at the end of the sentence should be excluded. A filter function returns
+# # True if an affected sentence should be included in the dataset.
+# ##############################################################################
+
+
+# def filter_hop(sent):
+#     # Assertion needed since filter function is only defined for affected
+#     # sentences
+#     assert (affect_hop(sent))
+#     return check_word_hops_completed(sent, 4)
+
+
+# def filter_reverse(sent):
+#     return True
+
+
+# def filter_shuffle(sent):
+#     tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+#     return len(tokens) > 1 and len(tokens) <= 350
+
+
+# def filter_none(sent):
+#     return False
+
+
+# ##############################################################################
+# # PERTURBATION FUNCTIONS
+# # These functions define how a perturbation will affect a sentence. They
+# # take in a sentence object and an optional marker
+# # for verb transformations. They return a string representing the transformed
+# # sentence.
+# ##############################################################################
+
+
+# def perturb_hop_words4(sent):
+#     return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+# def perturb_hop_tokens4(sent):
+#     return __perturb_hop_tokens(sent, 4)
+
+
+# def perturb_hop_control(sent):
+#     return __perturb_hop_tokens(sent, 0)
+
+
+# def perturb_reverse(sent, rng, reverse=True, full=False):
+#     return __perturb_reverse(sent, rng, reverse, full)
+
+
+# def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+#     return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+# def perturb_shuffle_nondeterministic(sent, rng):
+#     return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+# def perturb_shuffle_local(sent, seed, window):
+#     return __perturb_shuffle_local(sent, seed, window)
+
+
+# def perturb_shuffle_even_odd(sent):
+#     return __perturb_shuffle_even_odd(sent)
+
+
+# ##############################################################################
+# # PERTURBATIONS
+# # This dict maps the name of a perturbation to its perturbation and filter
+# # functions. The names and functions in this dict are used throughout the
+# # repo.
+# ##############################################################################
+
+
+# PERTURBATIONS = {
+#     "shuffle_control": {
+#         "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#606060",
+#     },
+#     "shuffle_nondeterministic": {
+#         "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#E8384F",
+#     },
+#     "shuffle_deterministic21": {
+#         "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#FFB000",
+#     },
+#     "shuffle_deterministic57": {
+#         "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#8db000",
+#     },
+#     "shuffle_deterministic84": {
+#         "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#62BB35",
+#     },
+#     "shuffle_local3": {
+#         "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#208EA3",
+#     },
+#     "shuffle_local5": {
+#         "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#4178BC",
+#     },
+#     "shuffle_local10": {
+#         "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#AA71FF",
+#     },
+#     "shuffle_even_odd": {
+#         "perturbation_function": perturb_shuffle_even_odd,
+#         "affect_function": affect_shuffle,
+#         "filter_function": filter_shuffle,
+#         "gpt2_tokenizer": gpt2_original_tokenizer,
+#         "color": "#E37CFF",
+#     },
+#     "reverse_control": {
+#         "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+#         "affect_function": affect_reverse,
+#         "filter_function": filter_reverse,
+#         "gpt2_tokenizer": gpt2_rev_tokenizer,
+#         "color": "#606060",
+#     },
+#     "reverse_partial": {
+#         "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+#         "affect_function": affect_reverse,
+#         "filter_function": filter_reverse,
+#         "gpt2_tokenizer": gpt2_rev_tokenizer,
+#         "color": "#E5A836",
+#     },
+#     "reverse_full": {
+#         "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+#         "affect_function": affect_reverse,
+#         "filter_function": filter_reverse,
+#         "gpt2_tokenizer": gpt2_rev_tokenizer,
+#         "color": "#A348A6",
+#     },
+#     "hop_control": {
+#         "perturbation_function": perturb_hop_control,
+#         "affect_function": affect_hop,
+#         "filter_function": filter_hop,
+#         "gpt2_tokenizer": gpt2_hop_tokenizer,
+#         "color": "#606060",
+#     },
+#     "hop_tokens4": {
+#         "perturbation_function": perturb_hop_tokens4,
+#         "affect_function": affect_hop,
+#         "filter_function": filter_hop,
+#         "gpt2_tokenizer": gpt2_hop_tokenizer,
+#         "color": "#fa8128", 
+#     },
+#     "hop_words4": {
+#         "perturbation_function": perturb_hop_words4,
+#         "affect_function": affect_hop,
+#         "filter_function": filter_hop,
+#         "gpt2_tokenizer": gpt2_hop_tokenizer,
+#         "color": "#03a0ff",
+#     },
+# }
diff --git a/utils_gpt2.py b/utils_gpt2.py
new file mode 100644
index 0000000000000000000000000000000000000000..ccdc8949138963b6ca2eb641e2f8c60d3fe388cf
--- /dev/null
+++ b/utils_gpt2.py
@@ -0,0 +1,576 @@
+# utils_gpt2.py
+# Author: Yaning
+
+from collections import deque
+from string import punctuation
+from transformers import AutoTokenizer, AddedToken
+from functools import partial
+from numpy.random import default_rng
+# from nltk.tree import ParentedTree
+import torch
+
+
+##############################################################################
+# CONSTANTS
+##############################################################################
+
+
+BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# Yj: 用于在参数解析和数据加载时指定数据集
+# 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+SEEDS = [21, 57, 84]
+CHECKPOINTS = list(range(50, 501, 50))
+GENRES = {
+    "aochildes": "CHILDES",
+    "bnc_spoken": "British National Corpus (BNC)",
+    "cbt": "Children’s Book Test",
+    "children_stories": "Children’s Stories Text Corpus",
+    "gutenberg": "Standardized Project Gutenberg Corpus",
+    "open_subtitles": "OpenSubtitles",
+    "qed": "QCRI Educational Domain Corpus",
+    "simple_wikipedia": "Simple Wikipedia",
+    "switchboard": "Switchboard Dialog Act Corpus",
+    "wikipedia": "Wikipedia"
+}
+CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+BABYLM_DATA_PATH = "."
+MARKER_HOP_SING = "🅂"
+MARKER_HOP_PLUR = "🄿"
+MARKER_REV = "🅁"
+BOS_TOKEN = "<BOS_TOKEN>"
+PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+PUNCT_TOKENS = set(punctuation)
+
+MODEL_NAME = "gpt2"
+
+
+##############################################################################
+# PARENS MODELS (Structurally-pretrained)
+##############################################################################
+
+
+PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+PAREN_MODELS = {
+    "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+    "NEST": "nested-parens0.49_vocab500-uniform",
+    "RAND": "random_vocab500-uniform",
+}
+
+
+##############################################################################
+# HELPER FUNCTIONS
+##############################################################################
+
+
+def write_file(directory, filename, lines):
+    f = open(directory + filename, "w")
+    f.writelines(lines)
+    f.close()
+
+
+def get_gpt2_tokenizer_with_markers(marker_list):
+
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    tokenizer.add_special_tokens({'pad_token': '[PAD]'})
+
+    # If no new markers to add, return normal tokenizer
+    if len(marker_list) == 0:
+        return tokenizer
+
+    # Create tokens and return modified tokenizer
+    new_tokens = []
+    for marker in marker_list:
+        new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+    tokenizer.add_tokens(new_tokens)
+    return tokenizer
+
+
+gpt2_original_tokenizer = get_gpt2_tokenizer_with_markers([])
+
+
+# GPT-2 hop tokenization
+gpt2_hop_tokenizer = get_gpt2_tokenizer_with_markers(
+    [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# Get ids of marker tokens
+marker_sg_token = gpt2_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_SING]
+# Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+marker_pl_token = gpt2_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_PLUR]
+
+
+# GPT-2 reverse tokenization
+gpt2_rev_tokenizer = get_gpt2_tokenizer_with_markers(
+    [MARKER_REV])
+# Get ids of marker tokens
+marker_rev_token = gpt2_rev_tokenizer.get_added_vocab()[
+    MARKER_REV]
+
+# GPT-2 determiner tokenization
+gpt2_det_tokenizer = get_gpt2_tokenizer_with_markers(
+    [BOS_TOKEN])
+# Get id of BOS token
+bos_token_id = gpt2_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+
+def compute_surprisals(model, input_ids):
+    # Get the log probabilities from the model
+    print("1")
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        shifted_input_ids = input_ids[:, 1:]
+
+        # Get the log probabilities for the actual next tokens
+        log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+        true_log_probs = log_probs.gather(
+            2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+    # Get the negative log probabilities
+    neg_log_probs = (-true_log_probs).tolist()
+    surprisals = [[None] + probs for probs in neg_log_probs]
+    print("surprisals:", surprisals)
+    return surprisals
+
+
+def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        probs = torch.nn.functional.softmax(logits, dim=-1)
+
+        # Get the probabilities for the specified token at each position
+        token_probs = probs[:, :, token_id]
+
+    # Convert to list and add None at the beginning to align with input tokens
+    # Put null probability for instances of pad token
+    token_probs_list = []
+    for batch_i, probs in enumerate(token_probs):
+        input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+        filtered = [p if input_ids_seq[pos_i+1] !=
+                    pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+        token_probs_list.append([None] + filtered)
+
+    return token_probs_list
+
+
+def merge_part_tokens(words):
+    result = []
+    for s in words:
+        if result and s in PART_TOKENS and len(result) > 0:
+            result[-1] += s
+        else:
+            result.append(s)
+    return result
+
+
+def __affect_hop_word(word):
+    return word["feats"] and "Person=3" in word["feats"] \
+        and "Tense=Pres" in word["feats"] \
+        and "VerbForm=Fin" in word["feats"] \
+        and "Number" in word["feats"]
+
+
+def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+    perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker_sg, marker_pl)
+    return perturbed_tokens
+
+
+def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+    _, hops_completed = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker, marker)
+    return hops_completed
+
+
+def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    hop_completed = []
+    new_sent = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            new_sent.append(
+                word["lemma"] if word["lemma"] is not None else word["text"])
+
+            # Marker hopping logic
+            insert_index = len(new_sent)-1
+            skipped_words = 0
+            while skipped_words < num_hops and insert_index > 0:
+
+                # Handle edge case when punctuation (or sequence of
+                # punctuation) begin the sentence
+                if (not any([c.isalnum() for c in
+                             "".join(new_sent[:insert_index])])):
+                    break
+                
+                # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+                # Count word as skipped if it is not a special token
+                if (new_sent[insert_index] not in PART_TOKENS) and \
+                        (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    skipped_words += 1
+                insert_index -= 1
+
+            # Handle edge case when insert index is punctuation (and this is not
+            # sentence-initial punctuation)
+            if any([c.isalnum() for c in
+                    "".join(new_sent[:insert_index])]):
+                while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+                                             or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    insert_index -= 1
+
+            # Handle edge case when token before insert index is part/aux token
+            if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+                insert_index -= 1
+
+            # Log if this sentence had all full hops
+            hop_completed.append(skipped_words == num_hops)
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                new_sent.insert(insert_index, marker_sg)
+            elif "Number=Plur" in word["feats"]:
+                new_sent.insert(insert_index, marker_pl)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+        else:
+            new_sent.append(word["text"])
+
+    new_sent.reverse()
+    sent_string = " ".join(merge_part_tokens(new_sent))
+    tokens = gpt2_hop_tokenizer.encode(sent_string)
+    return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+def __perturb_hop_tokens(sent, num_hops):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    new_sent = deque()
+    tokens = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+            if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+                lemma = lemma + new_sent[0]
+                new_sent.popleft()
+
+            if len(new_sent) > 0:
+                sent_string = " ".join(merge_part_tokens(new_sent))
+                tokens = gpt2_hop_tokenizer.encode(
+                    " " + sent_string) + tokens
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                tokens.insert(num_hops, marker_sg_token)
+            elif "Number=Plur" in word["feats"]:
+                tokens.insert(num_hops, marker_pl_token)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+            new_sent = deque()
+            new_sent.append(lemma)
+
+        else:
+            new_sent.appendleft(word["text"])
+
+    if len(new_sent) > 0:
+        sent_string = " ".join(merge_part_tokens(new_sent))
+        tokens = gpt2_hop_tokenizer.encode(sent_string) + tokens
+    return tokens
+
+
+def __perturb_reverse(sent, rng, reverse, full):
+
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_rev_tokenizer.encode(sent["sent_text"])
+
+    # Pick random index to insert REV token
+    i = rng.choice(len(tokens)+1)
+    tokens.insert(i, marker_rev_token)
+
+    # Extract tokens before/after the marker, and reverse tokens after
+    tokens_before = tokens[:i+1]
+    tokens_after = tokens[i+1:]
+    if reverse:
+        tokens_after.reverse()
+    new_tokens = tokens_before + tokens_after
+    if full:
+        assert not reverse
+        new_tokens.reverse()
+
+    return new_tokens
+
+
+def __perturb_shuffle_deterministic(sent, seed, shuffle):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    if shuffle:
+        default_rng(seed).shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_nondeterministic(sent, rng):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    rng.shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_local(sent, seed, window=5):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+
+    # Shuffle tokens in batches of size window
+    shuffled_tokens = []
+    for i in range(0, len(tokens), window):
+        batch = tokens[i:i+window].copy()
+        default_rng(seed).shuffle(batch)
+        shuffled_tokens += batch
+
+    return shuffled_tokens
+
+
+def __perturb_shuffle_even_odd(sent):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+    odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+    return even + odd
+
+
+##############################################################################
+# AFFECT FUNCTIONS
+# These functions define when a perturbation has been applied to a sentence
+# not. This is used for identifying which test sentences have been
+# altered to separate affected vs. unaffected senences. Affect functions are
+# functions of the input sentence object and return a boolean.
+##############################################################################
+
+
+def affect_hop(sent):
+    return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+        and sent["constituency_parse"] is not None
+
+
+def affect_reverse(sent):
+    return True
+
+
+def affect_shuffle(sent):
+    return True
+
+
+def affect_none(sent):
+    return False
+
+
+##############################################################################
+# FILTER FUNCTIONS
+# These functions define when an affected sentence should be included in the
+# final dataset. For instance, hop perturbations where the marker is placed
+# at the end of the sentence should be excluded. A filter function returns
+# True if an affected sentence should be included in the dataset.
+##############################################################################
+
+
+def filter_hop(sent):
+    # Assertion needed since filter function is only defined for affected
+    # sentences
+    assert (affect_hop(sent))
+    return check_word_hops_completed(sent, 4)
+
+
+def filter_reverse(sent):
+    return True
+
+
+def filter_shuffle(sent):
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    return len(tokens) > 1 and len(tokens) <= 350
+
+
+def filter_none(sent):
+    return False
+
+
+##############################################################################
+# PERTURBATION FUNCTIONS
+# These functions define how a perturbation will affect a sentence. They
+# take in a sentence object and an optional marker
+# for verb transformations. They return a string representing the transformed
+# sentence.
+##############################################################################
+
+
+def perturb_hop_words4(sent):
+    return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+def perturb_hop_tokens4(sent):
+    return __perturb_hop_tokens(sent, 4)
+
+
+def perturb_hop_control(sent):
+    return __perturb_hop_tokens(sent, 0)
+
+
+def perturb_reverse(sent, rng, reverse=True, full=False):
+    return __perturb_reverse(sent, rng, reverse, full)
+
+
+def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+    return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+def perturb_shuffle_nondeterministic(sent, rng):
+    return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+def perturb_shuffle_local(sent, seed, window):
+    return __perturb_shuffle_local(sent, seed, window)
+
+
+def perturb_shuffle_even_odd(sent):
+    return __perturb_shuffle_even_odd(sent)
+
+
+##############################################################################
+# PERTURBATIONS
+# This dict maps the name of a perturbation to its perturbation and filter
+# functions. The names and functions in this dict are used throughout the
+# repo.
+##############################################################################
+
+
+PERTURBATIONS = {
+    "shuffle_control": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#606060",
+    },
+    "shuffle_nondeterministic": {
+        "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#E8384F",
+    },
+    "shuffle_deterministic21": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#FFB000",
+    },
+    "shuffle_deterministic57": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#8db000",
+    },
+    "shuffle_deterministic84": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#62BB35",
+    },
+    "shuffle_local3": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#208EA3",
+    },
+    "shuffle_local5": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#4178BC",
+    },
+    "shuffle_local10": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#AA71FF",
+    },
+    "shuffle_even_odd": {
+        "perturbation_function": perturb_shuffle_even_odd,
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#E37CFF",
+    },
+    "reverse_control": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "gpt2_tokenizer": gpt2_rev_tokenizer,
+        "color": "#606060",
+    },
+    "reverse_partial": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "gpt2_tokenizer": gpt2_rev_tokenizer,
+        "color": "#E5A836",
+    },
+    "reverse_full": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "gpt2_tokenizer": gpt2_rev_tokenizer,
+        "color": "#A348A6",
+    },
+    "hop_control": {
+        "perturbation_function": perturb_hop_control,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "gpt2_tokenizer": gpt2_hop_tokenizer,
+        "color": "#606060",
+    },
+    "hop_tokens4": {
+        "perturbation_function": perturb_hop_tokens4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "gpt2_tokenizer": gpt2_hop_tokenizer,
+        "color": "#fa8128", 
+    },
+    "hop_words4": {
+        "perturbation_function": perturb_hop_words4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "gpt2_tokenizer": gpt2_hop_tokenizer,
+        "color": "#03a0ff",
+    },
+}
\ No newline at end of file
diff --git a/utils_gpt2_surprisal.py b/utils_gpt2_surprisal.py
new file mode 100644
index 0000000000000000000000000000000000000000..642011cd266669be72a3a4857b238bb2c7882118
--- /dev/null
+++ b/utils_gpt2_surprisal.py
@@ -0,0 +1,591 @@
+# utils_gpt2.py
+# Author: Yaning
+
+from collections import deque
+from string import punctuation
+from transformers import AutoTokenizer, AddedToken
+from functools import partial
+from numpy.random import default_rng
+# from nltk.tree import ParentedTree
+import torch
+
+
+##############################################################################
+# CONSTANTS
+##############################################################################
+
+
+BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# Yj: 用于在参数解析和数据加载时指定数据集
+# 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+SEEDS = [21, 57, 84]
+CHECKPOINTS = list(range(50, 501, 50))
+GENRES = {
+    "aochildes": "CHILDES",
+    "bnc_spoken": "British National Corpus (BNC)",
+    "cbt": "Children’s Book Test",
+    "children_stories": "Children’s Stories Text Corpus",
+    "gutenberg": "Standardized Project Gutenberg Corpus",
+    "open_subtitles": "OpenSubtitles",
+    "qed": "QCRI Educational Domain Corpus",
+    "simple_wikipedia": "Simple Wikipedia",
+    "switchboard": "Switchboard Dialog Act Corpus",
+    "wikipedia": "Wikipedia"
+}
+CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+BABYLM_DATA_PATH = "."
+MARKER_HOP_SING = "🅂"
+MARKER_HOP_PLUR = "🄿"
+MARKER_REV = "🅁"
+BOS_TOKEN = "<BOS_TOKEN>"
+PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+PUNCT_TOKENS = set(punctuation)
+
+MODEL_NAME = "gpt2"
+
+
+##############################################################################
+# PARENS MODELS (Structurally-pretrained)
+##############################################################################
+
+
+PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+PAREN_MODELS = {
+    "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+    "NEST": "nested-parens0.49_vocab500-uniform",
+    "RAND": "random_vocab500-uniform",
+}
+
+
+##############################################################################
+# HELPER FUNCTIONS
+##############################################################################
+
+
+def write_file(directory, filename, lines):
+    f = open(directory + filename, "w")
+    f.writelines(lines)
+    f.close()
+
+
+def get_gpt2_tokenizer_with_markers(marker_list):
+
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    tokenizer.add_special_tokens({'pad_token': '[PAD]'})
+
+    # If no new markers to add, return normal tokenizer
+    if len(marker_list) == 0:
+        return tokenizer
+
+    # Create tokens and return modified tokenizer
+    new_tokens = []
+    for marker in marker_list:
+        new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+    tokenizer.add_tokens(new_tokens)
+    return tokenizer
+
+
+gpt2_original_tokenizer = get_gpt2_tokenizer_with_markers([])
+
+
+# GPT-2 hop tokenization
+gpt2_hop_tokenizer = get_gpt2_tokenizer_with_markers(
+    [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# Get ids of marker tokens
+marker_sg_token = gpt2_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_SING]
+# Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+marker_pl_token = gpt2_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_PLUR]
+
+
+# GPT-2 reverse tokenization
+gpt2_rev_tokenizer = get_gpt2_tokenizer_with_markers(
+    [MARKER_REV])
+# Get ids of marker tokens
+marker_rev_token = gpt2_rev_tokenizer.get_added_vocab()[
+    MARKER_REV]
+
+# GPT-2 determiner tokenization
+gpt2_det_tokenizer = get_gpt2_tokenizer_with_markers(
+    [BOS_TOKEN])
+# Get id of BOS token
+bos_token_id = gpt2_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+def get_gpt2_model_with_markers(model_path, tokenizer):
+    model = AutoModelForCausalLM.from_pretrained(model_path)
+    
+    # 关键调整：同步模型词表大小
+    model.resize_token_embeddings(len(tokenizer))  
+    
+    # 验证词表一致性
+    print(f"[Debug] Tokenizer vocab: {len(tokenizer)}")
+    print(f"[Debug] Model vocab: {model.config.vocab_size}")
+    assert len(tokenizer) == model.config.vocab_size
+    
+    return model
+
+def compute_surprisals(model, input_ids):
+    # Get the log probabilities from the model
+    max_valid_id = model.config.vocab_size - 1
+    if input_ids.max() > max_valid_id:
+        invalid_ids = input_ids[input_ids > max_valid_id].unique().tolist()
+        raise ValueError(f"发现非法token IDs: {invalid_ids}")
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        shifted_input_ids = input_ids[:, 1:]
+
+        # Get the log probabilities for the actual next tokens
+        log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+        true_log_probs = log_probs.gather(
+            2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+    # Get the negative log probabilities
+    neg_log_probs = (-true_log_probs).tolist()
+    surprisals = [[None] + probs for probs in neg_log_probs]
+    print("surprisals:", surprisals)
+    return surprisals
+
+
+def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        probs = torch.nn.functional.softmax(logits, dim=-1)
+
+        # Get the probabilities for the specified token at each position
+        token_probs = probs[:, :, token_id]
+
+    # Convert to list and add None at the beginning to align with input tokens
+    # Put null probability for instances of pad token
+    token_probs_list = []
+    for batch_i, probs in enumerate(token_probs):
+        input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+        filtered = [p if input_ids_seq[pos_i+1] !=
+                    pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+        token_probs_list.append([None] + filtered)
+
+    return token_probs_list
+
+
+def merge_part_tokens(words):
+    result = []
+    for s in words:
+        if result and s in PART_TOKENS and len(result) > 0:
+            result[-1] += s
+        else:
+            result.append(s)
+    return result
+
+
+def __affect_hop_word(word):
+    return word["feats"] and "Person=3" in word["feats"] \
+        and "Tense=Pres" in word["feats"] \
+        and "VerbForm=Fin" in word["feats"] \
+        and "Number" in word["feats"]
+
+
+def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+    perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker_sg, marker_pl)
+    return perturbed_tokens
+
+
+def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+    _, hops_completed = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker, marker)
+    return hops_completed
+
+
+def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    hop_completed = []
+    new_sent = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            new_sent.append(
+                word["lemma"] if word["lemma"] is not None else word["text"])
+
+            # Marker hopping logic
+            insert_index = len(new_sent)-1
+            skipped_words = 0
+            while skipped_words < num_hops and insert_index > 0:
+
+                # Handle edge case when punctuation (or sequence of
+                # punctuation) begin the sentence
+                if (not any([c.isalnum() for c in
+                             "".join(new_sent[:insert_index])])):
+                    break
+                
+                # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+                # Count word as skipped if it is not a special token
+                if (new_sent[insert_index] not in PART_TOKENS) and \
+                        (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    skipped_words += 1
+                insert_index -= 1
+
+            # Handle edge case when insert index is punctuation (and this is not
+            # sentence-initial punctuation)
+            if any([c.isalnum() for c in
+                    "".join(new_sent[:insert_index])]):
+                while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+                                             or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    insert_index -= 1
+
+            # Handle edge case when token before insert index is part/aux token
+            if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+                insert_index -= 1
+
+            # Log if this sentence had all full hops
+            hop_completed.append(skipped_words == num_hops)
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                new_sent.insert(insert_index, marker_sg)
+            elif "Number=Plur" in word["feats"]:
+                new_sent.insert(insert_index, marker_pl)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+        else:
+            new_sent.append(word["text"])
+
+    new_sent.reverse()
+    sent_string = " ".join(merge_part_tokens(new_sent))
+    tokens = gpt2_hop_tokenizer.encode(sent_string)
+    return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+def __perturb_hop_tokens(sent, num_hops):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    new_sent = deque()
+    tokens = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+            if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+                lemma = lemma + new_sent[0]
+                new_sent.popleft()
+
+            if len(new_sent) > 0:
+                sent_string = " ".join(merge_part_tokens(new_sent))
+                tokens = gpt2_hop_tokenizer.encode(
+                    " " + sent_string) + tokens
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                tokens.insert(num_hops, marker_sg_token)
+            elif "Number=Plur" in word["feats"]:
+                tokens.insert(num_hops, marker_pl_token)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+            new_sent = deque()
+            new_sent.append(lemma)
+
+        else:
+            new_sent.appendleft(word["text"])
+
+    if len(new_sent) > 0:
+        sent_string = " ".join(merge_part_tokens(new_sent))
+        tokens = gpt2_hop_tokenizer.encode(sent_string) + tokens
+    return tokens
+
+
+def __perturb_reverse(sent, rng, reverse, full):
+
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_rev_tokenizer.encode(sent["sent_text"])
+
+    # Pick random index to insert REV token
+    i = rng.choice(len(tokens)+1)
+    tokens.insert(i, marker_rev_token)
+
+    # Extract tokens before/after the marker, and reverse tokens after
+    tokens_before = tokens[:i+1]
+    tokens_after = tokens[i+1:]
+    if reverse:
+        tokens_after.reverse()
+    new_tokens = tokens_before + tokens_after
+    if full:
+        assert not reverse
+        new_tokens.reverse()
+
+    return new_tokens
+
+
+def __perturb_shuffle_deterministic(sent, seed, shuffle):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    if shuffle:
+        default_rng(seed).shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_nondeterministic(sent, rng):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    rng.shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_local(sent, seed, window=5):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+
+    # Shuffle tokens in batches of size window
+    shuffled_tokens = []
+    for i in range(0, len(tokens), window):
+        batch = tokens[i:i+window].copy()
+        default_rng(seed).shuffle(batch)
+        shuffled_tokens += batch
+
+    return shuffled_tokens
+
+
+def __perturb_shuffle_even_odd(sent):
+    # Get sentence text and GPT-2 tokens
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+    odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+    return even + odd
+
+
+##############################################################################
+# AFFECT FUNCTIONS
+# These functions define when a perturbation has been applied to a sentence
+# not. This is used for identifying which test sentences have been
+# altered to separate affected vs. unaffected senences. Affect functions are
+# functions of the input sentence object and return a boolean.
+##############################################################################
+
+
+def affect_hop(sent):
+    return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+        and sent["constituency_parse"] is not None
+
+
+def affect_reverse(sent):
+    return True
+
+
+def affect_shuffle(sent):
+    return True
+
+
+def affect_none(sent):
+    return False
+
+
+##############################################################################
+# FILTER FUNCTIONS
+# These functions define when an affected sentence should be included in the
+# final dataset. For instance, hop perturbations where the marker is placed
+# at the end of the sentence should be excluded. A filter function returns
+# True if an affected sentence should be included in the dataset.
+##############################################################################
+
+
+def filter_hop(sent):
+    # Assertion needed since filter function is only defined for affected
+    # sentences
+    assert (affect_hop(sent))
+    return check_word_hops_completed(sent, 4)
+
+
+def filter_reverse(sent):
+    return True
+
+
+def filter_shuffle(sent):
+    tokens = gpt2_original_tokenizer.encode(sent["sent_text"])
+    return len(tokens) > 1 and len(tokens) <= 350
+
+
+def filter_none(sent):
+    return False
+
+
+##############################################################################
+# PERTURBATION FUNCTIONS
+# These functions define how a perturbation will affect a sentence. They
+# take in a sentence object and an optional marker
+# for verb transformations. They return a string representing the transformed
+# sentence.
+##############################################################################
+
+
+def perturb_hop_words4(sent):
+    return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+def perturb_hop_tokens4(sent):
+    return __perturb_hop_tokens(sent, 4)
+
+
+def perturb_hop_control(sent):
+    return __perturb_hop_tokens(sent, 0)
+
+
+def perturb_reverse(sent, rng, reverse=True, full=False):
+    return __perturb_reverse(sent, rng, reverse, full)
+
+
+def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+    return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+def perturb_shuffle_nondeterministic(sent, rng):
+    return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+def perturb_shuffle_local(sent, seed, window):
+    return __perturb_shuffle_local(sent, seed, window)
+
+
+def perturb_shuffle_even_odd(sent):
+    return __perturb_shuffle_even_odd(sent)
+
+
+##############################################################################
+# PERTURBATIONS
+# This dict maps the name of a perturbation to its perturbation and filter
+# functions. The names and functions in this dict are used throughout the
+# repo.
+##############################################################################
+
+
+PERTURBATIONS = {
+    "shuffle_control": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#606060",
+    },
+    "shuffle_nondeterministic": {
+        "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#E8384F",
+    },
+    "shuffle_deterministic21": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#FFB000",
+    },
+    "shuffle_deterministic57": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#8db000",
+    },
+    "shuffle_deterministic84": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#62BB35",
+    },
+    "shuffle_local3": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#208EA3",
+    },
+    "shuffle_local5": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#4178BC",
+    },
+    "shuffle_local10": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#AA71FF",
+    },
+    "shuffle_even_odd": {
+        "perturbation_function": perturb_shuffle_even_odd,
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "gpt2_tokenizer": gpt2_original_tokenizer,
+        "color": "#E37CFF",
+    },
+    "reverse_control": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "gpt2_tokenizer": gpt2_rev_tokenizer,
+        "color": "#606060",
+    },
+    "reverse_partial": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "gpt2_tokenizer": gpt2_rev_tokenizer,
+        "color": "#E5A836",
+    },
+    "reverse_full": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "gpt2_tokenizer": gpt2_rev_tokenizer,
+        "color": "#A348A6",
+    },
+    "hop_control": {
+        "perturbation_function": perturb_hop_control,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "gpt2_tokenizer": gpt2_hop_tokenizer,
+        "color": "#606060",
+    },
+    "hop_tokens4": {
+        "perturbation_function": perturb_hop_tokens4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "gpt2_tokenizer": gpt2_hop_tokenizer,
+        "color": "#fa8128", 
+    },
+    "hop_words4": {
+        "perturbation_function": perturb_hop_words4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "gpt2_tokenizer": gpt2_hop_tokenizer,
+        "color": "#03a0ff",
+    },
+}
\ No newline at end of file
diff --git a/utils_llama.py b/utils_llama.py
new file mode 100644
index 0000000000000000000000000000000000000000..9f5886bbb1279720b512a72d4f6b70be2b3cd9e5
--- /dev/null
+++ b/utils_llama.py
@@ -0,0 +1,574 @@
+# utils_llama.py
+# Author: Yaning
+
+from collections import deque
+from string import punctuation
+from transformers import AutoTokenizer, AddedToken
+from functools import partial
+from numpy.random import default_rng
+# from nltk.tree import ParentedTree
+import torch
+
+
+##############################################################################
+# CONSTANTS
+##############################################################################
+
+
+BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# Yj: 用于在参数解析和数据加载时指定数据集
+# 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+SEEDS = [21, 57, 84]
+CHECKPOINTS = list(range(50, 501, 50))
+GENRES = {
+    "aochildes": "CHILDES",
+    "bnc_spoken": "British National Corpus (BNC)",
+    "cbt": "Children’s Book Test",
+    "children_stories": "Children’s Stories Text Corpus",
+    "gutenberg": "Standardized Project Gutenberg Corpus",
+    "open_subtitles": "OpenSubtitles",
+    "qed": "QCRI Educational Domain Corpus",
+    "simple_wikipedia": "Simple Wikipedia",
+    "switchboard": "Switchboard Dialog Act Corpus",
+    "wikipedia": "Wikipedia"
+}
+CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+BABYLM_DATA_PATH = "."
+MARKER_HOP_SING = "🅂"
+MARKER_HOP_PLUR = "🄿"
+MARKER_REV = "🅁"
+BOS_TOKEN = "<BOS_TOKEN>"
+PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+PUNCT_TOKENS = set(punctuation)
+
+MODEL_NAME = "meta-llama/Llama-3.2-3B"
+
+
+##############################################################################
+# PARENS MODELS (Structurally-pretrained)
+##############################################################################
+
+
+PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+PAREN_MODELS = {
+    "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+    "NEST": "nested-parens0.49_vocab500-uniform",
+    "RAND": "random_vocab500-uniform",
+}
+
+
+##############################################################################
+# HELPER FUNCTIONS
+##############################################################################
+
+
+def write_file(directory, filename, lines):
+    f = open(directory + filename, "w")
+    f.writelines(lines)
+    f.close()
+
+
+def get_llama_tokenizer_with_markers(marker_list):
+
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    tokenizer.add_special_tokens({'pad_token': '[PAD]'})
+
+    # If no new markers to add, return normal tokenizer
+    if len(marker_list) == 0:
+        return tokenizer
+
+    # Create tokens and return modified tokenizer
+    new_tokens = []
+    for marker in marker_list:
+        new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+    tokenizer.add_tokens(new_tokens)
+    return tokenizer
+
+
+llama_original_tokenizer = get_llama_tokenizer_with_markers([])
+
+
+# GPT-2 hop tokenization
+llama_hop_tokenizer = get_llama_tokenizer_with_markers(
+    [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# Get ids of marker tokens
+marker_sg_token = llama_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_SING]
+# Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+marker_pl_token = llama_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_PLUR]
+
+
+# GPT-2 reverse tokenization
+llama_rev_tokenizer = get_llama_tokenizer_with_markers(
+    [MARKER_REV])
+# Get ids of marker tokens
+marker_rev_token = llama_rev_tokenizer.get_added_vocab()[
+    MARKER_REV]
+
+# GPT-2 determiner tokenization
+llama_det_tokenizer = get_llama_tokenizer_with_markers(
+    [BOS_TOKEN])
+# Get id of BOS token
+bos_token_id = llama_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+
+def compute_surprisals(model, input_ids):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        shifted_input_ids = input_ids[:, 1:]
+
+        # Get the log probabilities for the actual next tokens
+        log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+        true_log_probs = log_probs.gather(
+            2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+    # Get the negative log probabilities
+    neg_log_probs = (-true_log_probs).tolist()
+    surprisals = [[None] + probs for probs in neg_log_probs]
+    return surprisals
+
+
+def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        probs = torch.nn.functional.softmax(logits, dim=-1)
+
+        # Get the probabilities for the specified token at each position
+        token_probs = probs[:, :, token_id]
+
+    # Convert to list and add None at the beginning to align with input tokens
+    # Put null probability for instances of pad token
+    token_probs_list = []
+    for batch_i, probs in enumerate(token_probs):
+        input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+        filtered = [p if input_ids_seq[pos_i+1] !=
+                    pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+        token_probs_list.append([None] + filtered)
+
+    return token_probs_list
+
+
+def merge_part_tokens(words):
+    result = []
+    for s in words:
+        if result and s in PART_TOKENS and len(result) > 0:
+            result[-1] += s
+        else:
+            result.append(s)
+    return result
+
+
+def __affect_hop_word(word):
+    return word["feats"] and "Person=3" in word["feats"] \
+        and "Tense=Pres" in word["feats"] \
+        and "VerbForm=Fin" in word["feats"] \
+        and "Number" in word["feats"]
+
+
+def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+    perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker_sg, marker_pl)
+    return perturbed_tokens
+
+
+def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+    _, hops_completed = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker, marker)
+    return hops_completed
+
+
+def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    hop_completed = []
+    new_sent = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            new_sent.append(
+                word["lemma"] if word["lemma"] is not None else word["text"])
+
+            # Marker hopping logic
+            insert_index = len(new_sent)-1
+            skipped_words = 0
+            while skipped_words < num_hops and insert_index > 0:
+
+                # Handle edge case when punctuation (or sequence of
+                # punctuation) begin the sentence
+                if (not any([c.isalnum() for c in
+                             "".join(new_sent[:insert_index])])):
+                    break
+                
+                # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+                # Count word as skipped if it is not a special token
+                if (new_sent[insert_index] not in PART_TOKENS) and \
+                        (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    skipped_words += 1
+                insert_index -= 1
+
+            # Handle edge case when insert index is punctuation (and this is not
+            # sentence-initial punctuation)
+            if any([c.isalnum() for c in
+                    "".join(new_sent[:insert_index])]):
+                while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+                                             or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    insert_index -= 1
+
+            # Handle edge case when token before insert index is part/aux token
+            if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+                insert_index -= 1
+
+            # Log if this sentence had all full hops
+            hop_completed.append(skipped_words == num_hops)
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                new_sent.insert(insert_index, marker_sg)
+            elif "Number=Plur" in word["feats"]:
+                new_sent.insert(insert_index, marker_pl)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+        else:
+            new_sent.append(word["text"])
+
+    new_sent.reverse()
+    sent_string = " ".join(merge_part_tokens(new_sent))
+    tokens = llama_hop_tokenizer.encode(sent_string)
+    return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+def __perturb_hop_tokens(sent, num_hops):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    new_sent = deque()
+    tokens = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+            if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+                lemma = lemma + new_sent[0]
+                new_sent.popleft()
+
+            if len(new_sent) > 0:
+                sent_string = " ".join(merge_part_tokens(new_sent))
+                tokens = llama_hop_tokenizer.encode(
+                    " " + sent_string) + tokens
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                tokens.insert(num_hops, marker_sg_token)
+            elif "Number=Plur" in word["feats"]:
+                tokens.insert(num_hops, marker_pl_token)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+            new_sent = deque()
+            new_sent.append(lemma)
+
+        else:
+            new_sent.appendleft(word["text"])
+
+    if len(new_sent) > 0:
+        sent_string = " ".join(merge_part_tokens(new_sent))
+        tokens = llama_hop_tokenizer.encode(sent_string) + tokens
+    return tokens
+
+
+def __perturb_reverse(sent, rng, reverse, full):
+
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_rev_tokenizer.encode(sent["sent_text"])
+
+    # Pick random index to insert REV token
+    i = rng.choice(len(tokens)+1)
+    tokens.insert(i, marker_rev_token)
+
+    # Extract tokens before/after the marker, and reverse tokens after
+    tokens_before = tokens[:i+1]
+    tokens_after = tokens[i+1:]
+    if reverse:
+        tokens_after.reverse()
+    new_tokens = tokens_before + tokens_after
+    if full:
+        assert not reverse
+        new_tokens.reverse()
+
+    return new_tokens
+
+
+def __perturb_shuffle_deterministic(sent, seed, shuffle):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    if shuffle:
+        default_rng(seed).shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_nondeterministic(sent, rng):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    rng.shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_local(sent, seed, window=5):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+
+    # Shuffle tokens in batches of size window
+    shuffled_tokens = []
+    for i in range(0, len(tokens), window):
+        batch = tokens[i:i+window].copy()
+        default_rng(seed).shuffle(batch)
+        shuffled_tokens += batch
+
+    return shuffled_tokens
+
+
+def __perturb_shuffle_even_odd(sent):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+    odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+    return even + odd
+
+
+##############################################################################
+# AFFECT FUNCTIONS
+# These functions define when a perturbation has been applied to a sentence
+# not. This is used for identifying which test sentences have been
+# altered to separate affected vs. unaffected senences. Affect functions are
+# functions of the input sentence object and return a boolean.
+##############################################################################
+
+
+def affect_hop(sent):
+    return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+        and sent["constituency_parse"] is not None
+
+
+def affect_reverse(sent):
+    return True
+
+
+def affect_shuffle(sent):
+    return True
+
+
+def affect_none(sent):
+    return False
+
+
+##############################################################################
+# FILTER FUNCTIONS
+# These functions define when an affected sentence should be included in the
+# final dataset. For instance, hop perturbations where the marker is placed
+# at the end of the sentence should be excluded. A filter function returns
+# True if an affected sentence should be included in the dataset.
+##############################################################################
+
+
+def filter_hop(sent):
+    # Assertion needed since filter function is only defined for affected
+    # sentences
+    assert (affect_hop(sent))
+    return check_word_hops_completed(sent, 4)
+
+
+def filter_reverse(sent):
+    return True
+
+
+def filter_shuffle(sent):
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    return len(tokens) > 1 and len(tokens) <= 350
+
+
+def filter_none(sent):
+    return False
+
+
+##############################################################################
+# PERTURBATION FUNCTIONS
+# These functions define how a perturbation will affect a sentence. They
+# take in a sentence object and an optional marker
+# for verb transformations. They return a string representing the transformed
+# sentence.
+##############################################################################
+
+
+def perturb_hop_words4(sent):
+    return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+def perturb_hop_tokens4(sent):
+    return __perturb_hop_tokens(sent, 4)
+
+
+def perturb_hop_control(sent):
+    return __perturb_hop_tokens(sent, 0)
+
+
+def perturb_reverse(sent, rng, reverse=True, full=False):
+    return __perturb_reverse(sent, rng, reverse, full)
+
+
+def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+    return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+def perturb_shuffle_nondeterministic(sent, rng):
+    return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+def perturb_shuffle_local(sent, seed, window):
+    return __perturb_shuffle_local(sent, seed, window)
+
+
+def perturb_shuffle_even_odd(sent):
+    return __perturb_shuffle_even_odd(sent)
+
+
+##############################################################################
+# PERTURBATIONS
+# This dict maps the name of a perturbation to its perturbation and filter
+# functions. The names and functions in this dict are used throughout the
+# repo.
+##############################################################################
+
+
+PERTURBATIONS = {
+    "shuffle_control": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#606060",
+    },
+    "shuffle_nondeterministic": {
+        "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#E8384F",
+    },
+    "shuffle_deterministic21": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#FFB000",
+    },
+    "shuffle_deterministic57": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#8db000",
+    },
+    "shuffle_deterministic84": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#62BB35",
+    },
+    "shuffle_local3": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#208EA3",
+    },
+    "shuffle_local5": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#4178BC",
+    },
+    "shuffle_local10": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#AA71FF",
+    },
+    "shuffle_even_odd": {
+        "perturbation_function": perturb_shuffle_even_odd,
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#E37CFF",
+    },
+    "reverse_control": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "llama_tokenizer": llama_rev_tokenizer,
+        "color": "#606060",
+    },
+    "reverse_partial": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "llama_tokenizer": llama_rev_tokenizer,
+        "color": "#E5A836",
+    },
+    "reverse_full": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "llama_tokenizer": llama_rev_tokenizer,
+        "color": "#A348A6",
+    },
+    "hop_control": {
+        "perturbation_function": perturb_hop_control,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "llama_tokenizer": llama_hop_tokenizer,
+        "color": "#606060",
+    },
+    "hop_tokens4": {
+        "perturbation_function": perturb_hop_tokens4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "llama_tokenizer": llama_hop_tokenizer,
+        "color": "#fa8128", 
+    },
+    "hop_words4": {
+        "perturbation_function": perturb_hop_words4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "llama_tokenizer": llama_hop_tokenizer,
+        "color": "#03a0ff",
+    },
+}
\ No newline at end of file
diff --git a/utils_llama_1B.py b/utils_llama_1B.py
new file mode 100644
index 0000000000000000000000000000000000000000..bb1d608681d344a2d39658cc4de6fd6de67ec937
--- /dev/null
+++ b/utils_llama_1B.py
@@ -0,0 +1,574 @@
+# utils_llama.py
+# Author: Yaning
+
+from collections import deque
+from string import punctuation
+from transformers import AutoTokenizer, AddedToken
+from functools import partial
+from numpy.random import default_rng
+# from nltk.tree import ParentedTree
+import torch
+
+
+##############################################################################
+# CONSTANTS
+##############################################################################
+
+
+BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# Yj: 用于在参数解析和数据加载时指定数据集
+# 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+SEEDS = [21, 57, 84]
+CHECKPOINTS = list(range(50, 501, 50))
+GENRES = {
+    "aochildes": "CHILDES",
+    "bnc_spoken": "British National Corpus (BNC)",
+    "cbt": "Children’s Book Test",
+    "children_stories": "Children’s Stories Text Corpus",
+    "gutenberg": "Standardized Project Gutenberg Corpus",
+    "open_subtitles": "OpenSubtitles",
+    "qed": "QCRI Educational Domain Corpus",
+    "simple_wikipedia": "Simple Wikipedia",
+    "switchboard": "Switchboard Dialog Act Corpus",
+    "wikipedia": "Wikipedia"
+}
+CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+BABYLM_DATA_PATH = "."
+MARKER_HOP_SING = "🅂"
+MARKER_HOP_PLUR = "🄿"
+MARKER_REV = "🅁"
+BOS_TOKEN = "<BOS_TOKEN>"
+PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+PUNCT_TOKENS = set(punctuation)
+
+MODEL_NAME = "meta-llama/Llama-3.2-1B"
+
+
+##############################################################################
+# PARENS MODELS (Structurally-pretrained)
+##############################################################################
+
+
+PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+PAREN_MODELS = {
+    "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+    "NEST": "nested-parens0.49_vocab500-uniform",
+    "RAND": "random_vocab500-uniform",
+}
+
+
+##############################################################################
+# HELPER FUNCTIONS
+##############################################################################
+
+
+def write_file(directory, filename, lines):
+    f = open(directory + filename, "w")
+    f.writelines(lines)
+    f.close()
+
+
+def get_llama_tokenizer_with_markers(marker_list):
+
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+    tokenizer.add_special_tokens({'pad_token': '[PAD]'})
+
+    # If no new markers to add, return normal tokenizer
+    if len(marker_list) == 0:
+        return tokenizer
+
+    # Create tokens and return modified tokenizer
+    new_tokens = []
+    for marker in marker_list:
+        new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+    tokenizer.add_tokens(new_tokens)
+    return tokenizer
+
+
+llama_original_tokenizer = get_llama_tokenizer_with_markers([])
+
+
+# GPT-2 hop tokenization
+llama_hop_tokenizer = get_llama_tokenizer_with_markers(
+    [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# Get ids of marker tokens
+marker_sg_token = llama_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_SING]
+# Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+marker_pl_token = llama_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_PLUR]
+
+
+# GPT-2 reverse tokenization
+llama_rev_tokenizer = get_llama_tokenizer_with_markers(
+    [MARKER_REV])
+# Get ids of marker tokens
+marker_rev_token = llama_rev_tokenizer.get_added_vocab()[
+    MARKER_REV]
+
+# GPT-2 determiner tokenization
+llama_det_tokenizer = get_llama_tokenizer_with_markers(
+    [BOS_TOKEN])
+# Get id of BOS token
+bos_token_id = llama_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+
+def compute_surprisals(model, input_ids):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        shifted_input_ids = input_ids[:, 1:]
+
+        # Get the log probabilities for the actual next tokens
+        log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+        true_log_probs = log_probs.gather(
+            2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+    # Get the negative log probabilities
+    neg_log_probs = (-true_log_probs).tolist()
+    surprisals = [[None] + probs for probs in neg_log_probs]
+    return surprisals
+
+
+def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        probs = torch.nn.functional.softmax(logits, dim=-1)
+
+        # Get the probabilities for the specified token at each position
+        token_probs = probs[:, :, token_id]
+
+    # Convert to list and add None at the beginning to align with input tokens
+    # Put null probability for instances of pad token
+    token_probs_list = []
+    for batch_i, probs in enumerate(token_probs):
+        input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+        filtered = [p if input_ids_seq[pos_i+1] !=
+                    pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+        token_probs_list.append([None] + filtered)
+
+    return token_probs_list
+
+
+def merge_part_tokens(words):
+    result = []
+    for s in words:
+        if result and s in PART_TOKENS and len(result) > 0:
+            result[-1] += s
+        else:
+            result.append(s)
+    return result
+
+
+def __affect_hop_word(word):
+    return word["feats"] and "Person=3" in word["feats"] \
+        and "Tense=Pres" in word["feats"] \
+        and "VerbForm=Fin" in word["feats"] \
+        and "Number" in word["feats"]
+
+
+def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+    perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker_sg, marker_pl)
+    return perturbed_tokens
+
+
+def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+    _, hops_completed = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker, marker)
+    return hops_completed
+
+
+def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    hop_completed = []
+    new_sent = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            new_sent.append(
+                word["lemma"] if word["lemma"] is not None else word["text"])
+
+            # Marker hopping logic
+            insert_index = len(new_sent)-1
+            skipped_words = 0
+            while skipped_words < num_hops and insert_index > 0:
+
+                # Handle edge case when punctuation (or sequence of
+                # punctuation) begin the sentence
+                if (not any([c.isalnum() for c in
+                             "".join(new_sent[:insert_index])])):
+                    break
+                
+                # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+                # Count word as skipped if it is not a special token
+                if (new_sent[insert_index] not in PART_TOKENS) and \
+                        (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    skipped_words += 1
+                insert_index -= 1
+
+            # Handle edge case when insert index is punctuation (and this is not
+            # sentence-initial punctuation)
+            if any([c.isalnum() for c in
+                    "".join(new_sent[:insert_index])]):
+                while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+                                             or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    insert_index -= 1
+
+            # Handle edge case when token before insert index is part/aux token
+            if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+                insert_index -= 1
+
+            # Log if this sentence had all full hops
+            hop_completed.append(skipped_words == num_hops)
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                new_sent.insert(insert_index, marker_sg)
+            elif "Number=Plur" in word["feats"]:
+                new_sent.insert(insert_index, marker_pl)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+        else:
+            new_sent.append(word["text"])
+
+    new_sent.reverse()
+    sent_string = " ".join(merge_part_tokens(new_sent))
+    tokens = llama_hop_tokenizer.encode(sent_string)
+    return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+def __perturb_hop_tokens(sent, num_hops):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    new_sent = deque()
+    tokens = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+            if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+                lemma = lemma + new_sent[0]
+                new_sent.popleft()
+
+            if len(new_sent) > 0:
+                sent_string = " ".join(merge_part_tokens(new_sent))
+                tokens = llama_hop_tokenizer.encode(
+                    " " + sent_string) + tokens
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                tokens.insert(num_hops, marker_sg_token)
+            elif "Number=Plur" in word["feats"]:
+                tokens.insert(num_hops, marker_pl_token)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+            new_sent = deque()
+            new_sent.append(lemma)
+
+        else:
+            new_sent.appendleft(word["text"])
+
+    if len(new_sent) > 0:
+        sent_string = " ".join(merge_part_tokens(new_sent))
+        tokens = llama_hop_tokenizer.encode(sent_string) + tokens
+    return tokens
+
+
+def __perturb_reverse(sent, rng, reverse, full):
+
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_rev_tokenizer.encode(sent["sent_text"])
+
+    # Pick random index to insert REV token
+    i = rng.choice(len(tokens)+1)
+    tokens.insert(i, marker_rev_token)
+
+    # Extract tokens before/after the marker, and reverse tokens after
+    tokens_before = tokens[:i+1]
+    tokens_after = tokens[i+1:]
+    if reverse:
+        tokens_after.reverse()
+    new_tokens = tokens_before + tokens_after
+    if full:
+        assert not reverse
+        new_tokens.reverse()
+
+    return new_tokens
+
+
+def __perturb_shuffle_deterministic(sent, seed, shuffle):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    if shuffle:
+        default_rng(seed).shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_nondeterministic(sent, rng):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    rng.shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_local(sent, seed, window=5):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+
+    # Shuffle tokens in batches of size window
+    shuffled_tokens = []
+    for i in range(0, len(tokens), window):
+        batch = tokens[i:i+window].copy()
+        default_rng(seed).shuffle(batch)
+        shuffled_tokens += batch
+
+    return shuffled_tokens
+
+
+def __perturb_shuffle_even_odd(sent):
+    # Get sentence text and GPT-2 tokens
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+    odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+    return even + odd
+
+
+##############################################################################
+# AFFECT FUNCTIONS
+# These functions define when a perturbation has been applied to a sentence
+# not. This is used for identifying which test sentences have been
+# altered to separate affected vs. unaffected senences. Affect functions are
+# functions of the input sentence object and return a boolean.
+##############################################################################
+
+
+def affect_hop(sent):
+    return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+        and sent["constituency_parse"] is not None
+
+
+def affect_reverse(sent):
+    return True
+
+
+def affect_shuffle(sent):
+    return True
+
+
+def affect_none(sent):
+    return False
+
+
+##############################################################################
+# FILTER FUNCTIONS
+# These functions define when an affected sentence should be included in the
+# final dataset. For instance, hop perturbations where the marker is placed
+# at the end of the sentence should be excluded. A filter function returns
+# True if an affected sentence should be included in the dataset.
+##############################################################################
+
+
+def filter_hop(sent):
+    # Assertion needed since filter function is only defined for affected
+    # sentences
+    assert (affect_hop(sent))
+    return check_word_hops_completed(sent, 4)
+
+
+def filter_reverse(sent):
+    return True
+
+
+def filter_shuffle(sent):
+    tokens = llama_original_tokenizer.encode(sent["sent_text"])
+    return len(tokens) > 1 and len(tokens) <= 350
+
+
+def filter_none(sent):
+    return False
+
+
+##############################################################################
+# PERTURBATION FUNCTIONS
+# These functions define how a perturbation will affect a sentence. They
+# take in a sentence object and an optional marker
+# for verb transformations. They return a string representing the transformed
+# sentence.
+##############################################################################
+
+
+def perturb_hop_words4(sent):
+    return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+def perturb_hop_tokens4(sent):
+    return __perturb_hop_tokens(sent, 4)
+
+
+def perturb_hop_control(sent):
+    return __perturb_hop_tokens(sent, 0)
+
+
+def perturb_reverse(sent, rng, reverse=True, full=False):
+    return __perturb_reverse(sent, rng, reverse, full)
+
+
+def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+    return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+def perturb_shuffle_nondeterministic(sent, rng):
+    return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+def perturb_shuffle_local(sent, seed, window):
+    return __perturb_shuffle_local(sent, seed, window)
+
+
+def perturb_shuffle_even_odd(sent):
+    return __perturb_shuffle_even_odd(sent)
+
+
+##############################################################################
+# PERTURBATIONS
+# This dict maps the name of a perturbation to its perturbation and filter
+# functions. The names and functions in this dict are used throughout the
+# repo.
+##############################################################################
+
+
+PERTURBATIONS = {
+    "shuffle_control": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#606060",
+    },
+    "shuffle_nondeterministic": {
+        "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#E8384F",
+    },
+    "shuffle_deterministic21": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#FFB000",
+    },
+    "shuffle_deterministic57": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#8db000",
+    },
+    "shuffle_deterministic84": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#62BB35",
+    },
+    "shuffle_local3": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#208EA3",
+    },
+    "shuffle_local5": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#4178BC",
+    },
+    "shuffle_local10": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#AA71FF",
+    },
+    "shuffle_even_odd": {
+        "perturbation_function": perturb_shuffle_even_odd,
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "llama_tokenizer": llama_original_tokenizer,
+        "color": "#E37CFF",
+    },
+    "reverse_control": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "llama_tokenizer": llama_rev_tokenizer,
+        "color": "#606060",
+    },
+    "reverse_partial": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "llama_tokenizer": llama_rev_tokenizer,
+        "color": "#E5A836",
+    },
+    "reverse_full": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "llama_tokenizer": llama_rev_tokenizer,
+        "color": "#A348A6",
+    },
+    "hop_control": {
+        "perturbation_function": perturb_hop_control,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "llama_tokenizer": llama_hop_tokenizer,
+        "color": "#606060",
+    },
+    "hop_tokens4": {
+        "perturbation_function": perturb_hop_tokens4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "llama_tokenizer": llama_hop_tokenizer,
+        "color": "#fa8128", 
+    },
+    "hop_words4": {
+        "perturbation_function": perturb_hop_words4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "llama_tokenizer": llama_hop_tokenizer,
+        "color": "#03a0ff",
+    },
+}
\ No newline at end of file
diff --git a/utils_qwen.py b/utils_qwen.py
new file mode 100644
index 0000000000000000000000000000000000000000..065f2f0d34bf49079f949aa7fd5947e6dfbb8aff
--- /dev/null
+++ b/utils_qwen.py
@@ -0,0 +1,571 @@
+# utils_qwen.py
+# Author: Yaning
+
+from collections import deque
+from string import punctuation
+from transformers import AutoTokenizer, AddedToken
+from functools import partial
+from numpy.random import default_rng
+# from nltk.tree import ParentedTree
+import torch
+
+
+##############################################################################
+# CONSTANTS
+##############################################################################
+
+
+BABYLM_SPLITS = ['100M', '10M', 'dev', 'test', 'unittest']
+# Yj: 用于在参数解析和数据加载时指定数据集
+# 影响数据集的预处理过程，如生成训练、开发、测试和单元测试集。
+
+SEEDS = [21, 57, 84]
+CHECKPOINTS = list(range(50, 501, 50))
+GENRES = {
+    "aochildes": "CHILDES",
+    "bnc_spoken": "British National Corpus (BNC)",
+    "cbt": "Children’s Book Test",
+    "children_stories": "Children’s Stories Text Corpus",
+    "gutenberg": "Standardized Project Gutenberg Corpus",
+    "open_subtitles": "OpenSubtitles",
+    "qed": "QCRI Educational Domain Corpus",
+    "simple_wikipedia": "Simple Wikipedia",
+    "switchboard": "Switchboard Dialog Act Corpus",
+    "wikipedia": "Wikipedia"
+}
+CHECKPOINT_WRITE_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+CHECKPOINT_READ_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_models"
+# BABYLM_DATA_PATH = "/nlp/scr3/nlp/llms-in-llms/babylm_data"
+BABYLM_DATA_PATH = "."
+MARKER_HOP_SING = "🅂"
+MARKER_HOP_PLUR = "🄿"
+MARKER_REV = "🅁"
+BOS_TOKEN = "<BOS_TOKEN>"
+PART_TOKENS = set(["n't", "'ll", "'s", "'re", "'ve", "'m"])
+PUNCT_TOKENS = set(punctuation)
+MODEL_NAME = "Qwen/Qwen2.5-7B"
+
+
+##############################################################################
+# PARENS MODELS (Structurally-pretrained)
+##############################################################################
+
+
+PAREN_MODEL_PATH = "/u/scr/isabelvp//tilt-stuff/tilt-finetuning/pretrained_checkpoints/"
+PAREN_MODELS = {
+    "CROSS": "flat-parens_vocab500-uniform_deplength-nesting-nolimit",
+    "NEST": "nested-parens0.49_vocab500-uniform",
+    "RAND": "random_vocab500-uniform",
+}
+
+
+##############################################################################
+# HELPER FUNCTIONS
+##############################################################################
+
+
+def write_file(directory, filename, lines):
+    f = open(directory + filename, "w")
+    f.writelines(lines)
+    f.close()
+
+
+def get_qwen_tokenizer_with_markers(marker_list):
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+
+    # If no new markers to add, return normal tokenizer
+    if len(marker_list) == 0:
+        return tokenizer
+
+    # Create tokens and return modified tokenizer
+    new_tokens = []
+    for marker in marker_list:
+        new_tokens.append(AddedToken(marker, lstrip=True, rstrip=False))
+    tokenizer.add_tokens(new_tokens)
+    return tokenizer
+
+
+qwen_original_tokenizer = get_qwen_tokenizer_with_markers([])
+
+
+# GPT-2 hop tokenization
+qwen_hop_tokenizer = get_qwen_tokenizer_with_markers(
+    [MARKER_HOP_SING, MARKER_HOP_PLUR])
+# Get ids of marker tokens
+marker_sg_token = qwen_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_SING]
+# Yj：获取分词器中所有自定义添加的标记及其对应的 token ID
+
+marker_pl_token = qwen_hop_tokenizer.get_added_vocab()[
+    MARKER_HOP_PLUR]
+
+
+# GPT-2 reverse tokenization
+qwen_rev_tokenizer = get_qwen_tokenizer_with_markers(
+    [MARKER_REV])
+# Get ids of marker tokens
+marker_rev_token = qwen_rev_tokenizer.get_added_vocab()[
+    MARKER_REV]
+
+# GPT-2 determiner tokenization
+qwen_det_tokenizer = get_qwen_tokenizer_with_markers(
+    [BOS_TOKEN])
+# Get id of BOS token
+bos_token_id = qwen_det_tokenizer.get_added_vocab()[BOS_TOKEN]
+
+
+MARKER_TOKEN_IDS = [marker_sg_token, marker_pl_token, marker_rev_token]
+
+
+def compute_surprisals(model, input_ids):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        shifted_input_ids = input_ids[:, 1:]
+
+        # Get the log probabilities for the actual next tokens
+        log_probs = torch.log2(torch.nn.functional.softmax(logits, dim=-1))
+        true_log_probs = log_probs.gather(
+            2, shifted_input_ids.unsqueeze(-1)).squeeze(-1)
+
+    # Get the negative log probabilities
+    neg_log_probs = (-true_log_probs).tolist()
+    surprisals = [[None] + probs for probs in neg_log_probs]
+    return surprisals
+
+
+def compute_token_probabilities(model, input_ids, token_id, pad_token_id):
+    # Get the log probabilities from the model
+    with torch.no_grad():
+        outputs = model(input_ids)
+        logits = outputs.logits[:, :-1]
+        probs = torch.nn.functional.softmax(logits, dim=-1)
+
+        # Get the probabilities for the specified token at each position
+        token_probs = probs[:, :, token_id]
+
+    # Convert to list and add None at the beginning to align with input tokens
+    # Put null probability for instances of pad token
+    token_probs_list = []
+    for batch_i, probs in enumerate(token_probs):
+        input_ids_seq = input_ids[batch_i].tolist() + [pad_token_id]
+        filtered = [p if input_ids_seq[pos_i+1] !=
+                    pad_token_id else None for pos_i, p in enumerate(probs.tolist())]
+        token_probs_list.append([None] + filtered)
+
+    return token_probs_list
+
+
+def merge_part_tokens(words):
+    result = []
+    for s in words:
+        if result and s in PART_TOKENS and len(result) > 0:
+            result[-1] += s
+        else:
+            result.append(s)
+    return result
+
+
+def __affect_hop_word(word):
+    return word["feats"] and "Person=3" in word["feats"] \
+        and "Tense=Pres" in word["feats"] \
+        and "VerbForm=Fin" in word["feats"] \
+        and "Number" in word["feats"]
+
+
+def __perturb_hop_words(sent, num_hops, marker_sg, marker_pl):
+    perturbed_tokens, _ = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker_sg, marker_pl)
+    return perturbed_tokens
+
+
+def check_word_hops_completed(sent, num_hops=4, marker=MARKER_HOP_SING):
+    _, hops_completed = __perturb_hop_words_complete_hops(
+        sent, num_hops, marker, marker)
+    return hops_completed
+
+
+def __perturb_hop_words_complete_hops(sent, num_hops, marker_sg, marker_pl):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    hop_completed = []
+    new_sent = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            new_sent.append(
+                word["lemma"] if word["lemma"] is not None else word["text"])
+
+            # Marker hopping logic
+            insert_index = len(new_sent)-1
+            skipped_words = 0
+            while skipped_words < num_hops and insert_index > 0:
+
+                # Handle edge case when punctuation (or sequence of
+                # punctuation) begin the sentence
+                if (not any([c.isalnum() for c in
+                             "".join(new_sent[:insert_index])])):
+                    break
+                
+                # Yj: 如果字符串中不存在任何字母或数字字符（即都是标点、空格等）
+
+                # Count word as skipped if it is not a special token
+                if (new_sent[insert_index] not in PART_TOKENS) and \
+                        (not set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    skipped_words += 1
+                insert_index -= 1
+
+            # Handle edge case when insert index is punctuation (and this is not
+            # sentence-initial punctuation)
+            if any([c.isalnum() for c in
+                    "".join(new_sent[:insert_index])]):
+                while insert_index != 0 and (new_sent[insert_index] in PART_TOKENS
+                                             or set(new_sent[insert_index]).issubset(PUNCT_TOKENS)):
+                    insert_index -= 1
+
+            # Handle edge case when token before insert index is part/aux token
+            if insert_index != 0 and new_sent[insert_index-1] in PART_TOKENS:
+                insert_index -= 1
+
+            # Log if this sentence had all full hops
+            hop_completed.append(skipped_words == num_hops)
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                new_sent.insert(insert_index, marker_sg)
+            elif "Number=Plur" in word["feats"]:
+                new_sent.insert(insert_index, marker_pl)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+        else:
+            new_sent.append(word["text"])
+
+    new_sent.reverse()
+    sent_string = " ".join(merge_part_tokens(new_sent))
+    tokens = qwen_hop_tokenizer.encode(sent_string)
+    return tokens, all(hop_completed) and len(hop_completed) > 0
+
+
+def __perturb_hop_tokens(sent, num_hops):
+
+    word_annotations = sent["word_annotations"].copy()
+    word_annotations.reverse()
+
+    new_sent = deque()
+    tokens = []
+    for word in word_annotations:
+
+        # Identify 3.pres verbs
+        if __affect_hop_word(word):
+
+            # Lemmatize verb if possible
+            lemma = word["lemma"] if word["lemma"] is not None else word["text"]
+
+            if len(new_sent) > 0 and new_sent[0] in PART_TOKENS:
+                lemma = lemma + new_sent[0]
+                new_sent.popleft()
+
+            if len(new_sent) > 0:
+                sent_string = " ".join(merge_part_tokens(new_sent))
+                tokens = qwen_hop_tokenizer.encode(
+                    " " + sent_string) + tokens
+
+            # Use correct marker for singular vs. plural
+            if "Number=Sing" in word["feats"]:
+                tokens.insert(num_hops, marker_sg_token)
+            elif "Number=Plur" in word["feats"]:
+                tokens.insert(num_hops, marker_pl_token)
+            else:
+                raise Exception(
+                    "Number not in verb features\n" + sent["sent_text"])
+
+            new_sent = deque()
+            new_sent.append(lemma)
+
+        else:
+            new_sent.appendleft(word["text"])
+
+    if len(new_sent) > 0:
+        sent_string = " ".join(merge_part_tokens(new_sent))
+        tokens = qwen_hop_tokenizer.encode(sent_string) + tokens
+    return tokens
+
+
+def __perturb_reverse(sent, rng, reverse, full):
+
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_rev_tokenizer.encode(sent["sent_text"])
+
+    # Pick random index to insert REV token
+    i = rng.choice(len(tokens)+1)
+    tokens.insert(i, marker_rev_token)
+
+    # Extract tokens before/after the marker, and reverse tokens after
+    tokens_before = tokens[:i+1]
+    tokens_after = tokens[i+1:]
+    if reverse:
+        tokens_after.reverse()
+    new_tokens = tokens_before + tokens_after
+    if full:
+        assert not reverse
+        new_tokens.reverse()
+
+    return new_tokens
+
+
+def __perturb_shuffle_deterministic(sent, seed, shuffle):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    if shuffle:
+        default_rng(seed).shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_nondeterministic(sent, rng):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    rng.shuffle(tokens)
+    return tokens
+
+
+def __perturb_shuffle_local(sent, seed, window=5):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+
+    # Shuffle tokens in batches of size window
+    shuffled_tokens = []
+    for i in range(0, len(tokens), window):
+        batch = tokens[i:i+window].copy()
+        default_rng(seed).shuffle(batch)
+        shuffled_tokens += batch
+
+    return shuffled_tokens
+
+
+def __perturb_shuffle_even_odd(sent):
+    # Get sentence text and GPT-2 tokens
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    even = [tok for i, tok in enumerate(tokens) if i % 2 == 0]
+    odd = [tok for i, tok in enumerate(tokens) if i % 2 != 0]
+    return even + odd
+
+
+##############################################################################
+# AFFECT FUNCTIONS
+# These functions define when a perturbation has been applied to a sentence
+# not. This is used for identifying which test sentences have been
+# altered to separate affected vs. unaffected senences. Affect functions are
+# functions of the input sentence object and return a boolean.
+##############################################################################
+
+
+def affect_hop(sent):
+    return any([__affect_hop_word(word) for word in sent['word_annotations']]) \
+        and sent["constituency_parse"] is not None
+
+
+def affect_reverse(sent):
+    return True
+
+
+def affect_shuffle(sent):
+    return True
+
+
+def affect_none(sent):
+    return False
+
+
+##############################################################################
+# FILTER FUNCTIONS
+# These functions define when an affected sentence should be included in the
+# final dataset. For instance, hop perturbations where the marker is placed
+# at the end of the sentence should be excluded. A filter function returns
+# True if an affected sentence should be included in the dataset.
+##############################################################################
+
+
+def filter_hop(sent):
+    # Assertion needed since filter function is only defined for affected
+    # sentences
+    assert (affect_hop(sent))
+    return check_word_hops_completed(sent, 4)
+
+
+def filter_reverse(sent):
+    return True
+
+
+def filter_shuffle(sent):
+    tokens = qwen_original_tokenizer.encode(sent["sent_text"])
+    return len(tokens) > 1 and len(tokens) <= 350
+
+
+def filter_none(sent):
+    return False
+
+
+##############################################################################
+# PERTURBATION FUNCTIONS
+# These functions define how a perturbation will affect a sentence. They
+# take in a sentence object and an optional marker
+# for verb transformations. They return a string representing the transformed
+# sentence.
+##############################################################################
+
+
+def perturb_hop_words4(sent):
+    return __perturb_hop_words(sent, 4, MARKER_HOP_SING, MARKER_HOP_PLUR)
+
+
+def perturb_hop_tokens4(sent):
+    return __perturb_hop_tokens(sent, 4)
+
+
+def perturb_hop_control(sent):
+    return __perturb_hop_tokens(sent, 0)
+
+
+def perturb_reverse(sent, rng, reverse=True, full=False):
+    return __perturb_reverse(sent, rng, reverse, full)
+
+
+def perturb_shuffle_deterministic(sent, seed=None, shuffle=True):
+    return __perturb_shuffle_deterministic(sent, seed, shuffle)
+
+
+def perturb_shuffle_nondeterministic(sent, rng):
+    return __perturb_shuffle_nondeterministic(sent, rng)
+
+
+def perturb_shuffle_local(sent, seed, window):
+    return __perturb_shuffle_local(sent, seed, window)
+
+
+def perturb_shuffle_even_odd(sent):
+    return __perturb_shuffle_even_odd(sent)
+
+
+##############################################################################
+# PERTURBATIONS
+# This dict maps the name of a perturbation to its perturbation and filter
+# functions. The names and functions in this dict are used throughout the
+# repo.
+##############################################################################
+
+
+PERTURBATIONS = {
+    "shuffle_control": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=None, shuffle=False),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#606060",
+    },
+    "shuffle_nondeterministic": {
+        "perturbation_function": partial(perturb_shuffle_nondeterministic, rng=default_rng(0)),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#E8384F",
+    },
+    "shuffle_deterministic21": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=21, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#FFB000",
+    },
+    "shuffle_deterministic57": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=57, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#8db000",
+    },
+    "shuffle_deterministic84": {
+        "perturbation_function": partial(perturb_shuffle_deterministic, seed=84, shuffle=True),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#62BB35",
+    },
+    "shuffle_local3": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=3),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#208EA3",
+    },
+    "shuffle_local5": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=5),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#4178BC",
+    },
+    "shuffle_local10": {
+        "perturbation_function": partial(perturb_shuffle_local, seed=0, window=10),
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#AA71FF",
+    },
+    "shuffle_even_odd": {
+        "perturbation_function": perturb_shuffle_even_odd,
+        "affect_function": affect_shuffle,
+        "filter_function": filter_shuffle,
+        "qwen_tokenizer": qwen_original_tokenizer,
+        "color": "#E37CFF",
+    },
+    "reverse_control": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "qwen_tokenizer": qwen_rev_tokenizer,
+        "color": "#606060",
+    },
+    "reverse_partial": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=True, full=False),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "qwen_tokenizer": qwen_rev_tokenizer,
+        "color": "#E5A836",
+    },
+    "reverse_full": {
+        "perturbation_function": partial(perturb_reverse, rng=default_rng(21), reverse=False, full=True),
+        "affect_function": affect_reverse,
+        "filter_function": filter_reverse,
+        "qwen_tokenizer": qwen_rev_tokenizer,
+        "color": "#A348A6",
+    },
+    "hop_control": {
+        "perturbation_function": perturb_hop_control,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "qwen_tokenizer": qwen_hop_tokenizer,
+        "color": "#606060",
+    },
+    "hop_tokens4": {
+        "perturbation_function": perturb_hop_tokens4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "qwen_tokenizer": qwen_hop_tokenizer,
+        "color": "#fa8128", 
+    },
+    "hop_words4": {
+        "perturbation_function": perturb_hop_words4,
+        "affect_function": affect_hop,
+        "filter_function": filter_hop,
+        "qwen_tokenizer": qwen_hop_tokenizer,
+        "color": "#03a0ff",
+    },
+}
\ No newline at end of file