Link the code
Browse files
README.md
CHANGED
|
@@ -12,7 +12,7 @@ datasets:
|
|
| 12 |
# Latent Feedback Is a Compute and Data Amplifier: checkpoints
|
| 13 |
|
| 14 |
These are the trained models of *Latent Feedback Is a Compute and Data Amplifier* (paper: [alphaXiv](https://www.alphaxiv.org/abs/2611.latent-feedback-transformers-scaling-law), code:
|
| 15 |
-
|
| 16 |
|
| 17 |
There are 174 runs, one folder each, named `<arm>-<size>-u<U/N>-f<F>`:
|
| 18 |
- **Arms:** `transformer`, and `feedback`, the latent feedback Transformer: each position reads the previous position's
|
|
|
|
| 12 |
# Latent Feedback Is a Compute and Data Amplifier: checkpoints
|
| 13 |
|
| 14 |
These are the trained models of *Latent Feedback Is a Compute and Data Amplifier* (paper: [alphaXiv](https://www.alphaxiv.org/abs/2611.latent-feedback-transformers-scaling-law), code:
|
| 15 |
+
[ifm-ai/LoRM](https://github.com/ifm-ai/LoRM), folder [`works/latent_feedback`](https://github.com/ifm-ai/LoRM/tree/main/works/latent_feedback)).
|
| 16 |
|
| 17 |
There are 174 runs, one folder each, named `<arm>-<size>-u<U/N>-f<F>`:
|
| 18 |
- **Arms:** `transformer`, and `feedback`, the latent feedback Transformer: each position reads the previous position's
|