Upload AtomSLM checkpoint

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+compare_charts.png filter=lfs diff=lfs merge=lfs -text
+dashboard.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,34 +1,49 @@
 ---
 library_name: pytorch
+license: apache-2.0
+datasets:
+  - roneneldan/TinyStories
+  - Salesforce/wikitext
+  - HuggingFaceTB/everyday-conversations-llama3.1-2k
 tags:
   - atomslm
   - language-model
   - shared-weights
+  - pytorch
 ---
 
 # AtomSLM-600K
 
-**AtomSLM** is a compact, shared-weight language model family built on the
-AtomNet architecture (Shared Weight Core + Per-Layer FiLM Modulation).
+**AtomSLM** is a compact, shared-weight language model family built on the 
+AtomNet architecture (Shared Weight Core + Per-Layer FiLM Modulation). 
+One shared weight core is reused across all N layers, with tiny per-layer 
+FiLM vectors providing the only per-layer state — deep reasoning at minimal parameter cost.
 
 ## Model Details
 
-| Field         | Value                     |
-|---------------|---------------------------|
-| Architecture  | AtomNet (shared-core FiLM)|
-| Parameters    | 0.613M              |
-| Vocab size    | 4096               |
-| d_model       | 120                  |
-| Layers        | 8                 |
-| FFN mult      | 2.0               |
-| Max seq len   | 256              |
-| Weight tying  | True            |
-| Val loss      | 3.3455                |
-| Trained steps | 4900                   |
+| Field            | Value                          |
+|------------------|--------------------------------|
+| Architecture     | AtomNet (shared-core + FiLM)   |
+| Parameters       | 0.613M                      |
+| Vocab size       | 4096                        |
+| d_model           | 120                          |
+| Layers           | 8                          |
+| FFN multiplier   | 2.0                       |
+| Context window   | 256 tokens                |
+| Weight tying     | True                     |
+| Best val loss    | 3.3455                      |
+| Best PPL (val)   | 28.37                        |
+| Trained steps    | 4900                          |
 
 ## Training
 
-Trained on **TinyStories + WikiText-2** with a custom BPE tokenizer.
+Trained on the following datasets with a custom BPE tokenizer (vocab size matching the config above):
+
+- `roneneldan/TinyStories`
+- `wikitext-2-raw-v1`
+- `wikitext-103-raw-v1`
+- `hand-crafted-conversations`
+- `HuggingFaceTB/everyday-conversations-llama3.1-2k`
 
 ### Hyperparameters
 
@@ -55,6 +70,14 @@ Trained on **TinyStories + WikiText-2** with a custom BPE tokenizer.
 }
 ```
 
+## Training Dashboard
+
+![Training Dashboard](dashboard.png)
+
+## Benchmark vs Reference Models
+
+![Comparison Charts](compare_charts.png)
+
 ## Usage
 
 ```python
@@ -69,4 +92,4 @@ model.eval()
 
 ## License
 
-MIT
+Apache 2.0