This release is a pre-release and may not be stable for production use.
llmdbg
llmdbg is an early-stage debugger for stepping through language-model layers
and presenting human-readable interpretations of internal vectors. It speaks
the Debug Adapter Protocol
so editor integrations can use a standard debugger UI.
The current prototype includes real GPT-2 Small loading, layer-state capture, logit-lens interpretations, and a VS Code debug extension. A deterministic built-in trace remains available when no model is configured.
pip install --pre llmdbg
llmdbg --version
VS Code prototype
Build and install the extension:
cd vscode-extension
npm install
npm run package
code --install-extension llmdbg-0.0.1.vsix
Create a gpt2.llmdbg file in the workspace:
model = "openai-community/gpt2"
revision = "main"
device = "cuda"
dtype = "bfloat16"
prompt = "The capital of France is"
token = 3
stop_on_entry = false
submodule = "token_embedding"
layer = 0
submodule = "transformer.h.0.attention"
submodule = "transformer.h.0.mlp"
layer = 1
submodule = "transformer.h.1.attention"
submodule = "transformer.h.1.mlp"
Open that file and press F5. If VS Code asks for a debugger, choose
llmdbg. No launch.json is required. If a workspace launch configuration
is preferred, the complete minimal version is:
{
"version": "0.2.0",
"configurations": [
{
"type": "llmdbg",
"request": "launch",
"name": "llmdbg: Active file",
"config": "${file}"
}
]
}
On first use, the extension creates a private Python environment in VS Code's
extension storage. It does not install anything into the workspace. The first
GPT-2 run downloads and loads the model, then keeps its tensors resident in a
persistent model-host process. Stopping and restarting the debugger reuses the
resident model and runs only a new forward pass. Closing/reloading VS Code
stops the host and releases the model. Each submodule statement is an
authored model execution stop. Set breakpoints on these statements or use
Step Over to advance through them. Runtime vectors appear in debugger
scopes; the source file is never rewritten.
token = 3 makes token position 3 the default for every ordinary source
breakpoint. To override it for one breakpoint, use the shorter conditional
expression token=4. Positions are zero-based. The debugger prefills tokens
before the selected position into the KV cache, then captures that token with
a one-token forward pass. Stepping beyond the last selected layer advances to
the next token using the updated cache.
Watch one zero-based dimension with residual_stream[42], input[42],
output[42], or delta[42]. The Residual stream (output) debugger scope
exposes the current hidden representation explicitly. Run llmdbg: Plot
Current Vector from the Command Palette or debug toolbar to inspect the full
vector as a dimension plot, histogram, or top-activation chart. An open plot
updates automatically after each debugger stop.
Attention stops additionally expose the 768D LayerNorm output used as the
attention input. MLP stops expose the 768D LayerNorm output, the 3072D linear
expansion before activation, the 3072D expansion after GELU, and the 768D
projection. Use indexed Watches such as layernorm[42],
mlp_expansion_pre_activation[42], mlp_expansion[42], and
mlp_projection[42]; all available vectors also appear in the plot selector.
License
MIT
1 maintainer
Metadata
1 maintainer
Release files for llmdbg 0.0.3a13
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| llmdbg-0.0.3a13.tar.gz | 44.7 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| llmdbg-0.0.3a13-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 63.7 kB
Release files / llmdbg-0.0.3a13.tar.gz
| Download URL | llmdbg-0.0.3a13.tar.gz |
|---|---|
| Size | 44.7 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
09c0af36f2f9456a32a7f3e0586b142bb313b17a34e399bdac4425753e3b1a58
|
|
BLAKE2b-256 checksum How to use checksums |
ecb62837e09b3a2e7811ef021a9a85d043412cc81b5667e90590b5a021dae597
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.3
|
Release files / llmdbg-0.0.3a13-py3-none-any.whl
| Download URL | llmdbg-0.0.3a13-py3-none-any.whl |
|---|---|
| Size | 19.0 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
1136334a64859ae71cd01728daaee50531331a9a5b0597928f4667cc1c301472
|
|
BLAKE2b-256 checksum How to use checksums |
e4544002f5ea1d38d718c85eb89f66a26b3b71136692b319a33806ab1f6e5ddd
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/7.0.0 CPython/3.12.3
|
Security for llmdbg 0.0.3a13
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
