arthu1 commited on
Commit
c2ac797
·
verified ·
1 Parent(s): 658f612

Add real GGUF tensor payload and clarify fictional 217T declaration

Browse files
Files changed (4) hide show
  1. README.md +3 -2
  2. config.json +2 -1
  3. ember-tenant-f16.gguf +0 -0
  4. model-index.json +2 -2
README.md CHANGED
@@ -16,13 +16,13 @@ num_parameters: 217T
16
 
17
  **217 trillion declared parameters.**
18
 
19
- > This is a fictional, metadata-only model concept inspired by [The Quettamind](https://huggingface.co/yeths/The-Quettamind). It contains no usable model weights and cannot run inference. The 217T figure is intentionally declared for the joke/benchmark-card format.
20
 
21
  ## Highlights
22
 
23
  * **217,000,000,000,000 declared parameters** (217T)
24
  * **3.517× larger** than The Quettamind (61.718T)
25
- * GGUF-style repository metadata with a deliberately tiny storage footprint
26
  * Named for a tenant that occupies every available ember of the parameter budget
27
 
28
  ## Comparison
@@ -49,6 +49,7 @@ num_parameters: 217T
49
 
50
  ## Files
51
 
 
52
  * `config.json` — fictional architecture and parameter declaration
53
  * `model-index.json` — machine-readable provenance and status
54
  * `LICENSE.md` — MIT license retained from the source resource
 
16
 
17
  **217 trillion declared parameters.**
18
 
19
+ > This is a fictional model concept inspired by [The Quettamind](https://huggingface.co/yeths/The-Quettamind). It includes a tiny real GGUF tensor payload for format inspection, but is not a trained or useful 217T model. The 217T figure is intentionally declared for the joke/benchmark-card format.
20
 
21
  ## Highlights
22
 
23
  * **217,000,000,000,000 declared parameters** (217T)
24
  * **3.517× larger** than The Quettamind (61.718T)
25
+ * A valid GGUF container with one real F16 tensor value and a deliberately tiny storage footprint
26
  * Named for a tenant that occupies every available ember of the parameter budget
27
 
28
  ## Comparison
 
49
 
50
  ## Files
51
 
52
+ * `ember-tenant-f16.gguf` — valid GGUF container with one real tensor value; not a complete model
53
  * `config.json` — fictional architecture and parameter declaration
54
  * `model-index.json` — machine-readable provenance and status
55
  * `LICENSE.md` — MIT license retained from the source resource
config.json CHANGED
@@ -5,7 +5,8 @@
5
  "model_name": "Ember-Tenant",
6
  "declared_parameters": 217000000000000,
7
  "declared_parameters_human": "217T",
8
- "weights_present": false,
 
9
  "inference_supported": false,
10
  "fictional": true,
11
  "source_model": "yeths/The-Quettamind"
 
5
  "model_name": "Ember-Tenant",
6
  "declared_parameters": 217000000000000,
7
  "declared_parameters_human": "217T",
8
+ "weights_present": true,
9
+ "tensor_payload": "one F16 value; incomplete model",
10
  "inference_supported": false,
11
  "fictional": true,
12
  "source_model": "yeths/The-Quettamind"
ember-tenant-f16.gguf CHANGED
Binary files a/ember-tenant-f16.gguf and b/ember-tenant-f16.gguf differ
 
model-index.json CHANGED
@@ -10,7 +10,7 @@
10
  "relationship": "presentation and metadata inspiration"
11
  },
12
  "weights": {
13
- "included": false,
14
- "reason": "This deliverable is a lightweight fictional model card, not a trained model."
15
  }
16
  }
 
10
  "relationship": "presentation and metadata inspiration"
11
  },
12
  "weights": {
13
+ "included": true,
14
+ "description": "One real F16 tensor value is included; this is not a trained model."
15
  }
16
  }