Tier-A

MCP tool from Mercury MCP by norika1207-lab

hooked into llamacpp.Llama.logitsprocessor to capture output-layer hot dims per token

How to use it

Tier-A is exposed by the Mercury MCP MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the Tier-A tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.

Install Mercury MCP

$pip install -e .

FULL MERCURY MCP LISTING

Other tools in Mercury MCP (16)

codestral

22B (mistral coder variant)

35B (Cohere)

3.1-dense-8B (IBM)

3.1-8B, 3.3-70B

Paper status, citations, contribution info

mercury_list_models

List all 23 observed models

mistral

7B-v0.3, small-3.2-24B

olmo2

7B (AllenAI fully-open)

phi3

medium-14B

qwen2

3B, 7B, 14B, 32B, coder-32B

starcoder2

15B

Tier-B

transformers.AutoModelForCausalLM.frompretrained with outputhiddenstates=True, per-layer residual stream activation magnitudes binned into 10 quantiles

1.5-9B