Request access to this model
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
This model was ranked in the Agent session compaction arena on Microtensor. Its files open once Microtensor approves your request. Tell us who you are and what you will use it for.
Log in or Sign Up to review the conditions and access this model content.
Agent session compaction, epoch 0
This model ranked in the top 3 of a pool in the Agent session compaction arena (ctx_compress) on Microtensor (testnet), epoch 0. The weights, tokenizer and config here are byte for byte the files that were scored.
| Placement | Score |
|---|---|
| 1st in the Escalation pool | 0.508 |
| 1st in the Harness pool | 0.199 |
Results
| Pool | Score | Lower bound | Items | Eligible |
|---|---|---|---|---|
| Model | 0.203 | 0.203 | 25 | yes |
| Harness | 0.199 | 0.199 | 25 | yes |
| Escalation | 0.508 | 0.902 | 25 | yes |
| System | 0.000 | 0.557 | 25 | no |
Spot check
The coordinator reran 40 sampled model calls on the profile card and 40 matched token for token. First token p95 6,211 ms, per token p95 8.5 ms, within the speed limits: yes, within the memory limit: yes.
Use it
from transformers import AutoModelForCausalLM, AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("microtensor-io/ctx_compress-506b500a")
model = AutoModelForCausalLM.from_pretrained("microtensor-io/ctx_compress-506b500a")
What is in this repo
| Path | Contents |
|---|---|
| root | the scored model files, unchanged |
README.miner.md |
the miner's own README, unchanged |
system/ |
the rest of the scored system: manifest, harness and escalation code |
training/ |
the miner's training record, when the miner published one |
certificates/ |
one signed certificate per epoch this artifact placed in |
evidence/ |
the arena definition and links to the public records |
Verify it
- Take the SHA-256 of each weight file (
.safetensorsor.gguf), write one line per file as the file name, a space and the hash, sort the lines by file name, join them with newlines and take the SHA-256 of that text. It must equal506b500a46fa1299c2e474790e1a55c9f5b51208780c3bfa0b3c9e365b26b634. - Open
certificates/e0.json, remove thecoordinator_sigfield, write the rest as JSON with sorted keys and no spaces, and take its SHA-256 as hex. The signature is over the textmicrotensor-certificate:followed by that hash, made with the sr25519 key of the coordinator hotkey5D4cKxE8nX4jqJ7aMrrEhkXf7r6chouZbbn4ivQa9emk84m1. - Compare the scores with the public records listed in
evidence/links.json.
Certificate hash: 046858720437e9cbc43e7267569d95a83bfebcef3cffa82089369ae04d512cd7
The certificate covers the measured results in this arena and epoch only.
Source model: Tang-yin-112/mt_slm_ctx_compress@e7f35275da99a0374a14df17a65c74b249c93978 by Tang-yin-112.
- Downloads last month
- -
Model tree for microtensor-io/ctx_compress-506b500a
Base model
ibm-granite/granite-4.0-h-micro