Update README.md
Browse files
README.md
CHANGED
|
@@ -16,9 +16,7 @@ tags:
|
|
| 16 |
- text-generation
|
| 17 |
---
|
| 18 |
|
| 19 |
-
#
|
| 20 |
-
|
| 21 |
-
https://midwestern-simulation.neocities.org/main/library/Essense%209-16-2025
|
| 22 |
|
| 23 |
we've trained an ai model that compresses sequences of token embeddings into shorter sequences of token embeddings, which it then attempts to reconstruct the original text from—with varying degrees of success.
|
| 24 |
|
|
@@ -41,7 +39,7 @@ from huggingface_hub import hf_hub_download
|
|
| 41 |
device = torch.device("cuda:0")
|
| 42 |
dtype = torch.bfloat16
|
| 43 |
base_model_id = "HuggingFaceTB/SmolLM3-3B-Base"
|
| 44 |
-
compressor_id = "
|
| 45 |
|
| 46 |
# === MODEL LOADING ===
|
| 47 |
|
|
|
|
| 16 |
- text-generation
|
| 17 |
---
|
| 18 |
|
| 19 |
+
# crumb/essence-3b-v2
|
|
|
|
|
|
|
| 20 |
|
| 21 |
we've trained an ai model that compresses sequences of token embeddings into shorter sequences of token embeddings, which it then attempts to reconstruct the original text from—with varying degrees of success.
|
| 22 |
|
|
|
|
| 39 |
device = torch.device("cuda:0")
|
| 40 |
dtype = torch.bfloat16
|
| 41 |
base_model_id = "HuggingFaceTB/SmolLM3-3B-Base"
|
| 42 |
+
compressor_id = "crumb/essence-3b-v2"
|
| 43 |
|
| 44 |
# === MODEL LOADING ===
|
| 45 |
|