jumafernandez commited on
Commit
4dc82f3
·
verified ·
1 Parent(s): 375de09

publish trace-repro-lite-ar-s123

Browse files
trace-repro-lite-ar-s123/best/config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "input_dim": 768,
3
+ "hidden_dim": 768,
4
+ "output_dim": 768,
5
+ "num_layers": 6,
6
+ "num_heads": 8,
7
+ "dropout": 0.1,
8
+ "max_turns": 64,
9
+ "attention_mode": "autoregressive",
10
+ "use_speaker_embeddings": true,
11
+ "num_speakers": 4,
12
+ "layer_norm": true,
13
+ "ff_dim": 3072,
14
+ "activation": "gelu",
15
+ "output_residual": false,
16
+ "arch": "v2",
17
+ "head_mode": "tied_continuous",
18
+ "head_transform": true
19
+ }
trace-repro-lite-ar-s123/best/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33baabf64d5000b064613f44d68a4a6ede23a674ed4b5259fab841dce7a68830
3
+ size 179808064
trace-repro-lite-ar-s123/config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "input_dim": 768,
3
+ "hidden_dim": 768,
4
+ "output_dim": 768,
5
+ "num_layers": 6,
6
+ "num_heads": 8,
7
+ "dropout": 0.1,
8
+ "max_turns": 64,
9
+ "attention_mode": "autoregressive",
10
+ "use_speaker_embeddings": true,
11
+ "num_speakers": 4,
12
+ "layer_norm": true,
13
+ "ff_dim": 3072,
14
+ "activation": "gelu",
15
+ "output_residual": false,
16
+ "arch": "v2",
17
+ "head_mode": "tied_continuous",
18
+ "head_transform": true
19
+ }
trace-repro-lite-ar-s123/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:394ccdb5a31484eb82404d0efc96c1af9bfda32f4d71c814be2d81dcc55a9285
3
+ size 179808064
trace-repro-lite-ar-s123/trainlog.jsonl ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 1, "train_loss": 4.24987, "val_loss": 5.04884, "epoch_sec": 128.9}
2
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 2, "train_loss": 3.31485, "val_loss": 4.79571, "epoch_sec": 121.4}
3
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 3, "train_loss": 2.98552, "val_loss": 4.65684, "epoch_sec": 121.7}
4
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 4, "train_loss": 2.64492, "val_loss": 4.56789, "epoch_sec": 121.7}
5
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 5, "train_loss": 2.32337, "val_loss": 4.53185, "epoch_sec": 121.8}
6
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 6, "train_loss": 2.04351, "val_loss": 4.54065, "epoch_sec": 121.7}
7
+ {"model": "lite", "mode": "ar", "seed": 123, "lr": 0.0002, "canary": false, "init": "scratch", "epoch": 7, "train_loss": 1.81425, "val_loss": 4.55112, "epoch_sec": 129.5}