pathfm-300m / config.json
jcy20's picture
Release PathFM-300M
7a365c5
Raw History Blame Contribute Delete
2.04 kB
{
"format_version": 1,
"model_type": "pathfm",
"name": "PathFM-300M",
"family": "general",
"backbone": {
"chronos_config": {
"context_length": 8192,
"input_patch_size": 16,
"input_patch_stride": 16,
"max_output_patches": 64,
"output_patch_size": 16,
"quantiles": [
0.01,
0.05,
0.1,
0.15,
0.2,
0.25,
0.3,
0.35,
0.4,
0.45,
0.5,
0.55,
0.6,
0.65,
0.7,
0.75,
0.8,
0.85,
0.9,
0.95,
0.99
],
"time_encoding_scale": 8192,
"use_arcsinh": true,
"use_reg_token": true
},
"chronos_pipeline_class": "Chronos2Pipeline",
"d_ff": 3072,
"d_kv": 64,
"d_model": 768,
"dense_act_fn": "relu",
"dropout_rate": 0.0,
"feed_forward_proj": "relu",
"initializer_factor": 0.05,
"is_gated_act": false,
"layer_norm_epsilon": 1e-06,
"model_type": "t5",
"num_heads": 12,
"num_layers": 30,
"pad_token_id": 0,
"reg_token_id": 1,
"rope_theta": 10000.0,
"torch_dtype": "float32",
"use_cache": true,
"vocab_size": 2
},
"input_embedding_hidden_dim": 3072,
"architecture": {
"max_output_patches": 30,
"flow_entry_block": 10,
"source_scale": 0.75,
"instance_scale_floor": 0.0001,
"parallel_knots": [
0.0,
0.05,
0.1,
0.15,
0.2,
0.25,
0.3,
0.35,
0.4,
0.45,
0.5,
0.55,
0.6,
0.65,
0.7,
0.75,
0.8,
0.85,
0.9,
0.95,
1.0
],
"parallel_block_menu": [
1,
2,
5,
10,
20
]
},
"inference_defaults": {
"steps": 1,
"num_samples": 32,
"source_sampling": "stratified",
"token_budget": 32768,
"seed": 42,
"sampler_method": "parallel"
},
"parameters": 294980129,
"weights_sha256": "6d0cef666247b6209ff896a6465559f3997ec898e2e0ef78496b80b613d38855"
}