updated config and weights
Browse files- config.json +1 -34
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -5,50 +5,22 @@
|
|
| 5 |
],
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
-
"attn_logit_softcapping": null,
|
| 9 |
"context_length": 16384,
|
| 10 |
"decode_index": 5,
|
| 11 |
"dtype": "float32",
|
| 12 |
"force_flip_invariance": true,
|
| 13 |
-
"freq_size": 10,
|
| 14 |
"head_dim": 80,
|
| 15 |
"hidden_size": 1280,
|
| 16 |
"horizon_length": 128,
|
| 17 |
"infer_is_positive": true,
|
| 18 |
"initializer_range": 0.02,
|
| 19 |
"intermediate_size": 1280,
|
| 20 |
-
"layer_types": [
|
| 21 |
-
"attention",
|
| 22 |
-
"attention",
|
| 23 |
-
"attention",
|
| 24 |
-
"attention",
|
| 25 |
-
"attention",
|
| 26 |
-
"attention",
|
| 27 |
-
"attention",
|
| 28 |
-
"attention",
|
| 29 |
-
"attention",
|
| 30 |
-
"attention",
|
| 31 |
-
"attention",
|
| 32 |
-
"attention",
|
| 33 |
-
"attention",
|
| 34 |
-
"attention",
|
| 35 |
-
"attention",
|
| 36 |
-
"attention",
|
| 37 |
-
"attention",
|
| 38 |
-
"attention",
|
| 39 |
-
"attention",
|
| 40 |
-
"attention"
|
| 41 |
-
],
|
| 42 |
"max_position_embeddings": 16384,
|
| 43 |
-
"max_timescale": 10000.0,
|
| 44 |
-
"min_timescale": 1.0,
|
| 45 |
"model_type": "timesfm_2p5",
|
| 46 |
-
"normalize_inputs": true,
|
| 47 |
"num_attention_heads": 16,
|
| 48 |
"num_hidden_layers": 20,
|
| 49 |
"num_key_value_heads": 16,
|
| 50 |
"output_quantile_len": 1024,
|
| 51 |
-
"pad_val": -1000000000.0,
|
| 52 |
"patch_length": 32,
|
| 53 |
"quantiles": [
|
| 54 |
0.1,
|
|
@@ -67,14 +39,9 @@
|
|
| 67 |
"rope_theta": 10000.0,
|
| 68 |
"rope_type": "default"
|
| 69 |
},
|
| 70 |
-
"rope_theta": 10000.0,
|
| 71 |
-
"sliding_window": null,
|
| 72 |
-
"tolerance": 1e-05,
|
| 73 |
"transformers_version": "5.3.0.dev0",
|
| 74 |
"use_bias": false,
|
| 75 |
"use_continuous_quantile_head": true,
|
| 76 |
"use_per_dim_scale": true,
|
| 77 |
-
"
|
| 78 |
-
"use_qk_norm": true,
|
| 79 |
-
"use_rotary_embeddings": true
|
| 80 |
}
|
|
|
|
| 5 |
],
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
|
|
|
| 8 |
"context_length": 16384,
|
| 9 |
"decode_index": 5,
|
| 10 |
"dtype": "float32",
|
| 11 |
"force_flip_invariance": true,
|
|
|
|
| 12 |
"head_dim": 80,
|
| 13 |
"hidden_size": 1280,
|
| 14 |
"horizon_length": 128,
|
| 15 |
"infer_is_positive": true,
|
| 16 |
"initializer_range": 0.02,
|
| 17 |
"intermediate_size": 1280,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
"max_position_embeddings": 16384,
|
|
|
|
|
|
|
| 19 |
"model_type": "timesfm_2p5",
|
|
|
|
| 20 |
"num_attention_heads": 16,
|
| 21 |
"num_hidden_layers": 20,
|
| 22 |
"num_key_value_heads": 16,
|
| 23 |
"output_quantile_len": 1024,
|
|
|
|
| 24 |
"patch_length": 32,
|
| 25 |
"quantiles": [
|
| 26 |
0.1,
|
|
|
|
| 39 |
"rope_theta": 10000.0,
|
| 40 |
"rope_type": "default"
|
| 41 |
},
|
|
|
|
|
|
|
|
|
|
| 42 |
"transformers_version": "5.3.0.dev0",
|
| 43 |
"use_bias": false,
|
| 44 |
"use_continuous_quantile_head": true,
|
| 45 |
"use_per_dim_scale": true,
|
| 46 |
+
"use_qk_norm": true
|
|
|
|
|
|
|
| 47 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b53f6d52114e2ad786890f3c4637ce05f580b7800d6e24401f88b398b76035ef
|
| 3 |
+
size 925187448
|