Isaac-0.5 / config.json
philperceptron's picture
feat: publish portable Isaac-0.5 artifact
ed7b437
Raw
History Blame
8.26 kB
{
"_name_or_path": "",
"action_expert": {
"action_dim": 64,
"action_horizon": 64,
"causal_attn": false,
"context_layer_norm": true,
"drop_action_dim_overflow": false,
"ffn_multiple_of": 256,
"hidden_dim": 768,
"k_batched_cross_attn": true,
"k_batched_cross_attn_backend": "flash_gqa",
"mask_padded_action_rows": true,
"mlp_ratio": 4.0,
"num_heads": 8,
"num_inference_steps": 10,
"num_layers": 36,
"qk_norm": true,
"qk_norm_eps": 1e-06,
"rope": true,
"rtc_delay_sampling": "poisson",
"rtc_max_delay_steps": 12,
"rtc_poisson_mean": 5.0,
"rtc_probability": 0.5,
"schema_version": 1,
"timestep_embed_dim": 256,
"timestep_sampling_alpha": 1.5,
"timestep_sampling_beta": 1.0,
"timestep_sampling_offset": 0.001,
"timestep_sampling_scale": 0.999,
"train_samples_per_chunk": 8,
"type": "dit"
},
"architectures": [
"Isaac05ForConditionalGeneration"
],
"auto_map": {
"AutoConfig": "configuration_isaac05.Isaac05Config",
"AutoModelForCausalLM": "modeling_isaac05.Isaac05ForConditionalGeneration",
"AutoProcessor": "processing_isaac05.Isaac05Processor"
},
"chunk_size_feed_forward": 0,
"dtype": "bfloat16",
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"image_token_id": 248056,
"is_encoder_decoder": false,
"isaac05_artifact": {
"artifact_kind": "trained_policy",
"schema_version": 1,
"tensor_bytes": 142894027456,
"tensor_count": 1616
},
"isaac05_coord_tokens": {
"enabled": true,
"offset": 248320,
"size": 1001
},
"isaac05_fast_tokens": {
"enabled": true,
"offset": 249321,
"size": 2048,
"tokenizer": "physical-intelligence/fast"
},
"isaac05_moe": {
"logical_router_outputs": 512,
"num_null_experts": 256,
"num_real_experts": 256,
"physical_router_outputs": 257,
"route_norm": true,
"route_scale": 1.0,
"router_contract": [
256,
256
],
"router_contract_version": 1,
"score_before_experts": false,
"score_func": "softmax",
"shared_null_router_row": true,
"top_k": 8,
"uses_expert_bias": false
},
"isaac05_test_only_reduced_geometry": false,
"isaac05_vla": {
"action_expert": {
"action_dim": 64,
"action_horizon": 64,
"causal_attn": false,
"context_layer_norm": true,
"drop_action_dim_overflow": false,
"ffn_multiple_of": 256,
"hidden_dim": 768,
"k_batched_cross_attn": true,
"k_batched_cross_attn_backend": "flash_gqa",
"mask_padded_action_rows": true,
"mlp_ratio": 4.0,
"num_heads": 8,
"num_inference_steps": 10,
"num_layers": 36,
"qk_norm": true,
"qk_norm_eps": 1e-06,
"rope": true,
"rtc_delay_sampling": "poisson",
"rtc_max_delay_steps": 12,
"rtc_poisson_mean": 5.0,
"rtc_probability": 0.5,
"schema_version": 1,
"timestep_embed_dim": 256,
"timestep_sampling_alpha": 1.5,
"timestep_sampling_beta": 1.0,
"timestep_sampling_offset": 0.001,
"timestep_sampling_scale": 0.999,
"train_samples_per_chunk": 8,
"type": "dit"
},
"mtp": {
"action_runtime": "exclude",
"physical_layers": 0,
"present": false,
"rollout_steps": 0
},
"rmsnorm_weight_convention": "zero_centered_1_plus_weight",
"schema_version": 1,
"state_dict_schema": "pr3154_v1",
"vector_encoder": {
"bias": false,
"hidden_dim": 2048,
"max_states": 128,
"output_dim": 2048,
"type": "linear_silu_linear"
}
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"max_sequence_length": 262144,
"model_type": "isaac_0_5",
"output_attentions": false,
"output_hidden_states": false,
"problem_type": null,
"return_dict": true,
"runtime_dtype": "bfloat16",
"storage_dtype": "float32",
"text_config": {
"_name_or_path": "",
"architectures": null,
"attention_bias": false,
"attention_dropout": 0.0,
"attn_output_gate": true,
"bos_token_id": 248044,
"chunk_size_feed_forward": 0,
"dtype": "bfloat16",
"eos_token_id": 248044,
"full_attention_interval": 4,
"head_dim": 256,
"hidden_act": "silu",
"hidden_size": 2048,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"initializer_range": 0.02,
"is_encoder_decoder": false,
"isaac05_moe": {
"logical_router_outputs": 512,
"num_null_experts": 256,
"num_real_experts": 256,
"physical_router_outputs": 257,
"route_norm": true,
"route_scale": 1.0,
"router_contract": [
256,
256
],
"router_contract_version": 1,
"score_before_experts": false,
"score_func": "softmax",
"shared_null_router_row": true,
"top_k": 8,
"uses_expert_bias": false
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"layer_types": [
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention",
"linear_attention",
"linear_attention",
"linear_attention",
"full_attention"
],
"linear_conv_kernel_dim": 4,
"linear_key_head_dim": 128,
"linear_num_key_heads": 16,
"linear_num_value_heads": 32,
"linear_value_head_dim": 128,
"mamba_ssm_dtype": "float32",
"max_position_embeddings": 262144,
"model_type": "qwen3_5_moe_text",
"moe_intermediate_size": 512,
"mtp_num_hidden_layers": 0,
"mtp_use_dedicated_embeddings": false,
"num_attention_heads": 16,
"num_experts": 256,
"num_experts_per_tok": 8,
"num_hidden_layers": 40,
"num_key_value_heads": 2,
"output_attentions": false,
"output_hidden_states": false,
"output_router_logits": false,
"pad_token_id": null,
"partial_rotary_factor": 0.25,
"problem_type": null,
"return_dict": true,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"mrope_interleaved": true,
"mrope_section": [
11,
11,
10
],
"partial_rotary_factor": 0.25,
"rope_theta": 10000000,
"rope_type": "default"
},
"router_aux_loss_coef": 0.001,
"shared_expert_intermediate_size": 512,
"tie_word_embeddings": false,
"use_cache": true,
"vocab_size": 256279
},
"tie_word_embeddings": false,
"transformers_version": "5.5.4",
"vector_max_states": 128,
"video_token_id": 248057,
"vision_config": {
"_name_or_path": "",
"architectures": null,
"chunk_size_feed_forward": 0,
"deepstack_visual_indexes": [],
"depth": 27,
"dtype": null,
"hidden_act": "gelu_pytorch_tanh",
"hidden_size": 1152,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"in_channels": 3,
"initializer_range": 0.02,
"intermediate_size": 4304,
"is_encoder_decoder": false,
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"model_type": "qwen3_5_moe",
"num_heads": 16,
"num_position_embeddings": 2304,
"out_hidden_size": 2048,
"output_attentions": false,
"output_hidden_states": false,
"patch_size": 16,
"problem_type": null,
"return_dict": true,
"spatial_merge_size": 2,
"temporal_patch_size": 2
},
"vision_end_token_id": 248054,
"vision_rescale_factor": 0.00392156862745098,
"vision_start_token_id": 248053,
"vision_token": "<|image_pad|>"
}