yairamr commited on
Commit
cdb68b3
·
verified ·
1 Parent(s): e4c5eb1

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +36 -0
README.md ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Placeholder — reverb FT checkpoint
2
+
3
+ `muse_reverb_e48.pt` in this directory is a **byte-for-byte copy of the
4
+ upstream noise-only MUSE checkpoint** (`g_best`, ~2.3 MB), used as a stand-in
5
+ for the real reverb-fine-tuned epoch-48 checkpoint until that file can be
6
+ retrieved from the iscohen_prj research cluster.
7
+
8
+ ## Real source location
9
+
10
+ `/rg/iscohen_prj/yairamr/code/Muse-Reverb-FN/checkpoints/all/epoch_48/g_00051852`
11
+
12
+ ## Action when the real checkpoint is retrievable
13
+
14
+ ```bash
15
+ scp <cluster>:/rg/iscohen_prj/yairamr/code/Muse-Reverb-FN/checkpoints/all/epoch_48/g_00051852 \
16
+ ./_workspace/release_artifacts/muse_reverb_e48.pt
17
+ ```
18
+
19
+ Then upload the replaced file to the HuggingFace model repo
20
+ (`HF_REPO_CKPT`, default `YairAmar/SE-Probe-models`). `scripts/setup.py`
21
+ already pulls from there, so notebook 06's inference cells will start
22
+ producing correct numbers automatically once the upload completes.
23
+
24
+ ## Why this placeholder is safe
25
+
26
+ Notebook 06's published figure is computed from precomputed CKA results
27
+ (`results_demo/cka_reverb_muse_demo.parquet` or
28
+ `results_df/reverb/cka_reverb_muse_epoch_48.parquet`), not from running the
29
+ model — so the figure is correct regardless of which checkpoint is shipped.
30
+ Inference cells in notebook 06 are gated behind
31
+ `SE_PROBE_RUN_INFERENCE=1` and clearly labelled as placeholder-mode in the
32
+ notebook markdown, so a curious reader is warned before they get wrong
33
+ numbers.
34
+
35
+ See `_workspace/cluster_artifacts_needed.md` (item 1) for the broader
36
+ punch-list of cluster artefacts.