Skip to main content
Histogram logging lets you track how distributions of values (such as weight distributions, activation values, or gradient norms) change over training steps.

​
Logging Histograms

To log a histogram, instantiate the pluto.Histogram class and pass it to pluto.log: 
histogram = pluto.Histogram(
    data=values,
    bins=64,
)
pluto.log({"layers/layer_0/weights": histogram}, step=epoch)
ParameterTypeDescription
dataUnion[list, np.ndarray, torch.Tensor]The values to build the histogram from.
binsintNumber of bins for the histogram. Defaults to 64.

​
Viewing Histograms in the Dashboard

Histograms appear as cards in the dashboard alongside your other metrics. Each histogram widget displays the distribution at a given training step. Histogram view in the dashboard

​
Step Navigation

Use the step navigator at the bottom of the histogram card to scrub through training steps and see how distributions evolve over time. You can:
  • Click the forward/back arrows to step through one at a time
  • Type a specific step number into the input field
  • See the current step and total steps (e.g., “Step 5/25”)

​
Axis Controls

Right-click on a histogram chart or use the chart menu to adjust axis bounds:
  • X Min / X Max — Control the horizontal range to zoom into a specific region of the distribution
  • Y Max — Set the maximum y-axis value to normalize the view across steps

​
Examples

​
Logging Weight Distributions

import pluto
import torch

run = pluto.init(project="my-project")

model = MyModel()
for epoch in range(num_epochs):
    # ... training step ...

    # Log weight distributions for each layer
    for name, param in model.named_parameters():
        if "weight" in name:
            pluto.log({f"histograms/{name}": pluto.Histogram(param.data.cpu())}, step=epoch)

​
Logging Gradient Distributions

for name, param in model.named_parameters():
    if param.grad is not None:
        pluto.log({
            f"gradients/{name}": pluto.Histogram(param.grad.data.cpu(), bins=32)
        }, step=epoch)