Image-Text-to-Text
Transformers
English
ternary-quant
quantization
ternary
vlm
multimodal
video
smolvlm
Instructions to use AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-text-to-text", model="AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary
- SGLang
How to use AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary with Docker Model Runner:
docker model run hf.co/AsadIsmail/SmolVLM2-500M-Video-Instruct-ternary
| { | |
| "model_name": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct", | |
| "model_config": { | |
| "transformers_version": "5.5.3", | |
| "architectures": [ | |
| "SmolVLMForConditionalGeneration" | |
| ], | |
| "output_hidden_states": false, | |
| "return_dict": true, | |
| "dtype": "float16", | |
| "chunk_size_feed_forward": 0, | |
| "is_encoder_decoder": false, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "problem_type": null, | |
| "use_cache": false, | |
| "image_token_id": 49190, | |
| "tie_word_embeddings": false, | |
| "vision_config": { | |
| "architectures": null, | |
| "output_hidden_states": false, | |
| "return_dict": true, | |
| "dtype": "float16", | |
| "chunk_size_feed_forward": 0, | |
| "is_encoder_decoder": false, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "problem_type": null, | |
| "hidden_size": 768, | |
| "intermediate_size": 3072, | |
| "num_hidden_layers": 12, | |
| "num_attention_heads": 12, | |
| "num_channels": 3, | |
| "image_size": 512, | |
| "patch_size": 16, | |
| "hidden_act": "gelu_pytorch_tanh", | |
| "layer_norm_eps": 1e-06, | |
| "attention_dropout": 0.0, | |
| "initializer_range": 0.02, | |
| "_name_or_path": "", | |
| "max_image_size": { | |
| "longest_edge": 512 | |
| }, | |
| "model_type": "smolvlm_vision", | |
| "size": { | |
| "longest_edge": 2048 | |
| }, | |
| "tie_word_embeddings": false, | |
| "use_base_siglip": false, | |
| "output_attentions": false | |
| }, | |
| "text_config": { | |
| "architectures": [ | |
| "VLlama3ForCausalLM" | |
| ], | |
| "output_hidden_states": false, | |
| "return_dict": true, | |
| "dtype": "float16", | |
| "chunk_size_feed_forward": 0, | |
| "is_encoder_decoder": false, | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "problem_type": null, | |
| "vocab_size": 49280, | |
| "hidden_size": 960, | |
| "intermediate_size": 2560, | |
| "num_hidden_layers": 32, | |
| "num_attention_heads": 15, | |
| "num_key_value_heads": 5, | |
| "hidden_act": "silu", | |
| "max_position_embeddings": 8192, | |
| "initializer_range": 0.02, | |
| "rms_norm_eps": 1e-05, | |
| "use_cache": true, | |
| "pad_token_id": 2, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "pretraining_tp": 1, | |
| "tie_word_embeddings": false, | |
| "rope_parameters": { | |
| "rope_theta": 100000, | |
| "rope_type": "default" | |
| }, | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "mlp_bias": false, | |
| "head_dim": 64, | |
| "_name_or_path": "None", | |
| "_flash_attn_2_enabled": true, | |
| "is_llama_config": true, | |
| "model_type": "llama", | |
| "neftune_noise_alpha": 0.0, | |
| "perceiver_config": { | |
| "_attn_implementation_autoset": false, | |
| "_name_or_path": "", | |
| "add_cross_attention": false, | |
| "architectures": null, | |
| "attention_dropout": 0.0, | |
| "bad_words_ids": null, | |
| "begin_suppress_tokens": null, | |
| "bos_token_id": null, | |
| "chunk_size_feed_forward": 0, | |
| "cross_attention_hidden_size": null, | |
| "decoder_start_token_id": null, | |
| "diversity_penalty": 0.0, | |
| "do_sample": false, | |
| "early_stopping": false, | |
| "encoder_no_repeat_ngram_size": 0, | |
| "eos_token_id": null, | |
| "exponential_decay_length_penalty": null, | |
| "finetuning_task": null, | |
| "forced_bos_token_id": null, | |
| "forced_eos_token_id": null, | |
| "hidden_act": "silu", | |
| "id2label": { | |
| "0": "LABEL_0", | |
| "1": "LABEL_1" | |
| }, | |
| "is_decoder": false, | |
| "is_encoder_decoder": false, | |
| "label2id": { | |
| "LABEL_0": 0, | |
| "LABEL_1": 1 | |
| }, | |
| "length_penalty": 1.0, | |
| "max_length": 20, | |
| "min_length": 0, | |
| "model_type": "vllama3", | |
| "no_repeat_ngram_size": 0, | |
| "num_beam_groups": 1, | |
| "num_beams": 1, | |
| "num_key_value_heads": 1, | |
| "num_return_sequences": 1, | |
| "output_attentions": false, | |
| "output_hidden_states": false, | |
| "output_scores": false, | |
| "pad_token_id": null, | |
| "prefix": null, | |
| "problem_type": null, | |
| "pruned_heads": {}, | |
| "qk_layer_norms_perceiver": false, | |
| "remove_invalid_values": false, | |
| "repetition_penalty": 1.0, | |
| "resampler_depth": 6, | |
| "resampler_head_dim": 96, | |
| "resampler_n_heads": 16, | |
| "resampler_n_latents": 64, | |
| "return_dict": true, | |
| "return_dict_in_generate": false, | |
| "sep_token_id": null, | |
| "suppress_tokens": null, | |
| "task_specific_params": null, | |
| "temperature": 1.0, | |
| "tf_legacy_loss": false, | |
| "tie_encoder_decoder": false, | |
| "tie_word_embeddings": true, | |
| "tokenizer_class": null, | |
| "top_k": 50, | |
| "top_p": 1.0, | |
| "torch_dtype": null, | |
| "torchscript": false, | |
| "transformers_version": "4.46.0", | |
| "typical_p": 1.0, | |
| "use_bfloat16": false | |
| }, | |
| "pixel_shuffle_factor": 4, | |
| "qk_layer_norms": false, | |
| "rope_interleaved": false, | |
| "transformers.js_config": { | |
| "kv_cache_dtype": { | |
| "fp16": "float16", | |
| "q4f16": "float16" | |
| } | |
| }, | |
| "use_resampler": false, | |
| "output_attentions": false | |
| }, | |
| "scale_factor": 4, | |
| "pad_token_id": 128002, | |
| "_name_or_path": "HuggingFaceTB/SmolVLM2-500M-Video-Instruct", | |
| "model_type": "smolvlm", | |
| "transformers.js_config": { | |
| "kv_cache_dtype": { | |
| "fp16": "float16", | |
| "q4f16": "float16" | |
| } | |
| }, | |
| "use_reentrant_checkpointing": false, | |
| "vocab_size": 49280, | |
| "output_attentions": false | |
| }, | |
| "quant_config": { | |
| "components": [ | |
| "text_backbone", | |
| "multimodal_connector" | |
| ], | |
| "scheme": "tritplane3", | |
| "group_size": 32, | |
| "n_iter": 10, | |
| "salient_fraction": 0.0, | |
| "rescue_fraction": 0.0, | |
| "n_planes": 3, | |
| "allow_all_linear": false, | |
| "target_module_names": [ | |
| "Wqkv", | |
| "att_proj", | |
| "attn.proj", | |
| "attn.qkv", | |
| "c_attn", | |
| "c_fc", | |
| "c_proj", | |
| "dense", | |
| "dense_4h_to_h", | |
| "dense_h_to_4h", | |
| "down_proj", | |
| "fc1", | |
| "fc2", | |
| "ff_proj", | |
| "gate_proj", | |
| "gate_up_proj", | |
| "k", | |
| "k_proj", | |
| "linear", | |
| "o", | |
| "o_proj", | |
| "out_proj", | |
| "per_layer_input_gate", | |
| "per_layer_projection", | |
| "proj", | |
| "q", | |
| "q_proj", | |
| "qkv", | |
| "qkv_proj", | |
| "query_key_value", | |
| "up_proj", | |
| "v", | |
| "v_proj", | |
| "w1", | |
| "w2", | |
| "w3", | |
| "wi", | |
| "wi_0", | |
| "wi_1", | |
| "wo" | |
| ], | |
| "max_length": 160, | |
| "calibration_batch_size": 2, | |
| "calibration_prompts": null, | |
| "vlm_use_demo_image": true | |
| }, | |
| "plan": {}, | |
| "layer_info": { | |
| "model.connector.modality_projection.proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 12288 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 11796480, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 13271040, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.30.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 320, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 307200, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 345600, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 921600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 1036800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 2560, | |
| 960 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.31.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 960, | |
| 2560 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 2457600, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 2764800, | |
| "effective_bits": 9.0 | |
| } | |
| }, | |
| "stats": { | |
| "model.connector.modality_projection.proj": { | |
| "mse": 5.37627511221217e-06, | |
| "rmse": 0.002318679605338385, | |
| "relative_error": 0.11232529343978108, | |
| "max_error": 0.06935769319534302, | |
| "sparsity": 0.3947989245255788, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.q_proj": { | |
| "mse": 0.0005314695299603045, | |
| "rmse": 0.02305362292483124, | |
| "relative_error": 0.20877777205728598, | |
| "max_error": 1.37176513671875, | |
| "sparsity": 0.376055767138799, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.k_proj": { | |
| "mse": 0.0014180912403389812, | |
| "rmse": 0.03765755223509597, | |
| "relative_error": 0.2126414957021358, | |
| "max_error": 1.49334716796875, | |
| "sparsity": 0.38177518049875897, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.v_proj": { | |
| "mse": 9.182788198813796e-05, | |
| "rmse": 0.00958268657465838, | |
| "relative_error": 0.29577660288803204, | |
| "max_error": 0.1934490203857422, | |
| "sparsity": 0.36731011668841046, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.o_proj": { | |
| "mse": 5.279598553897813e-05, | |
| "rmse": 0.007266084608575524, | |
| "relative_error": 0.19093673262379318, | |
| "max_error": 0.5605087280273438, | |
| "sparsity": 0.3689001003901164, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.mlp.gate_proj": { | |
| "mse": 0.00035739081795327365, | |
| "rmse": 0.018904782938538957, | |
| "relative_error": 0.13753535044116563, | |
| "max_error": 0.578033447265625, | |
| "sparsity": 0.37930094202359516, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.mlp.up_proj": { | |
| "mse": 0.00034740770934149623, | |
| "rmse": 0.018638876289666612, | |
| "relative_error": 0.13723774105383113, | |
| "max_error": 0.6000633239746094, | |
| "sparsity": 0.380840927362442, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.mlp.down_proj": { | |
| "mse": 0.0006163854268379509, | |
| "rmse": 0.024827110722715016, | |
| "relative_error": 0.17911922661283106, | |
| "max_error": 1.0714111328125, | |
| "sparsity": 0.36744222044944763, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.q_proj": { | |
| "mse": 0.0005779704661108553, | |
| "rmse": 0.024041016328575946, | |
| "relative_error": 0.13682085083112477, | |
| "max_error": 0.8404541015625, | |
| "sparsity": 0.38917896151542664, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.k_proj": { | |
| "mse": 0.0008308137184940279, | |
| "rmse": 0.02882383941278517, | |
| "relative_error": 0.13721463893281963, | |
| "max_error": 0.9709396362304688, | |
| "sparsity": 0.3887261350949605, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.v_proj": { | |
| "mse": 0.0001580135285621509, | |
| "rmse": 0.012570343215765865, | |
| "relative_error": 0.14910078273213473, | |
| "max_error": 0.3300743103027344, | |
| "sparsity": 0.39191949367523193, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.o_proj": { | |
| "mse": 0.0002102715807268396, | |
| "rmse": 0.014500744143899636, | |
| "relative_error": 0.14647063007913325, | |
| "max_error": 1.4849853515625, | |
| "sparsity": 0.37764286001523334, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.mlp.gate_proj": { | |
| "mse": 0.00036494428059086204, | |
| "rmse": 0.019103514875301403, | |
| "relative_error": 0.13087417167045606, | |
| "max_error": 0.927978515625, | |
| "sparsity": 0.3793248236179352, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.mlp.up_proj": { | |
| "mse": 0.00030177790904417634, | |
| "rmse": 0.017371756072549958, | |
| "relative_error": 0.13367486196953646, | |
| "max_error": 0.7654075622558594, | |
| "sparsity": 0.3790958722432454, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.mlp.down_proj": { | |
| "mse": 0.0003775917866732925, | |
| "rmse": 0.019431721145418194, | |
| "relative_error": 0.1473674225047308, | |
| "max_error": 0.9212369918823242, | |
| "sparsity": 0.3742697536945343, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.q_proj": { | |
| "mse": 0.0004766773490700871, | |
| "rmse": 0.021832941832700584, | |
| "relative_error": 0.1354250294054266, | |
| "max_error": 0.8622207045555115, | |
| "sparsity": 0.38469980160395306, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.k_proj": { | |
| "mse": 0.0006819431437179446, | |
| "rmse": 0.026114041121931792, | |
| "relative_error": 0.1404071508568069, | |
| "max_error": 0.7380363941192627, | |
| "sparsity": 0.3838368058204651, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.v_proj": { | |
| "mse": 0.00018824658764060587, | |
| "rmse": 0.013720298380159444, | |
| "relative_error": 0.13767107040547688, | |
| "max_error": 0.30676794052124023, | |
| "sparsity": 0.3867046535015106, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.o_proj": { | |
| "mse": 0.00020274224516469985, | |
| "rmse": 0.014238758554196354, | |
| "relative_error": 0.1338761074254524, | |
| "max_error": 0.5722503662109375, | |
| "sparsity": 0.37832645575205487, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.mlp.gate_proj": { | |
| "mse": 0.0003442151937633753, | |
| "rmse": 0.01855303731908539, | |
| "relative_error": 0.12962811109657388, | |
| "max_error": 0.717137336730957, | |
| "sparsity": 0.3799024720986684, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.mlp.up_proj": { | |
| "mse": 0.00029949229792691767, | |
| "rmse": 0.017305845773232745, | |
| "relative_error": 0.13028602005524945, | |
| "max_error": 0.9309492111206055, | |
| "sparsity": 0.3806545635064443, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.mlp.down_proj": { | |
| "mse": 0.00045880948891863227, | |
| "rmse": 0.021419838676298014, | |
| "relative_error": 0.16082310185477836, | |
| "max_error": 1.3219680786132812, | |
| "sparsity": 0.37084445357322693, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.q_proj": { | |
| "mse": 0.0005037776427343488, | |
| "rmse": 0.022444991484390203, | |
| "relative_error": 0.13561528604897113, | |
| "max_error": 0.7548227310180664, | |
| "sparsity": 0.3890194594860077, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.k_proj": { | |
| "mse": 0.00069454126060009, | |
| "rmse": 0.02635415072811283, | |
| "relative_error": 0.13818009364656234, | |
| "max_error": 0.7760581970214844, | |
| "sparsity": 0.38602214058240253, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.v_proj": { | |
| "mse": 0.00014110682241152972, | |
| "rmse": 0.011878839270380323, | |
| "relative_error": 0.14022822676624314, | |
| "max_error": 0.3532853126525879, | |
| "sparsity": 0.3907855947812398, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.o_proj": { | |
| "mse": 0.000207564837182872, | |
| "rmse": 0.01440711064658254, | |
| "relative_error": 0.14167102225585407, | |
| "max_error": 0.4413299560546875, | |
| "sparsity": 0.37646520137786865, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.mlp.gate_proj": { | |
| "mse": 0.00033784445258788764, | |
| "rmse": 0.018380545492119857, | |
| "relative_error": 0.12913113959089356, | |
| "max_error": 1.094214916229248, | |
| "sparsity": 0.38064589103062946, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.mlp.up_proj": { | |
| "mse": 0.0002976357063744217, | |
| "rmse": 0.017252121793403317, | |
| "relative_error": 0.1295545050286228, | |
| "max_error": 1.2246017456054688, | |
| "sparsity": 0.382622887690862, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.mlp.down_proj": { | |
| "mse": 0.0004010823031421751, | |
| "rmse": 0.020027039300460143, | |
| "relative_error": 0.1517636729996116, | |
| "max_error": 1.227447509765625, | |
| "sparsity": 0.37449761231740314, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.q_proj": { | |
| "mse": 0.0005056362133473158, | |
| "rmse": 0.022486356159843145, | |
| "relative_error": 0.14080490460768644, | |
| "max_error": 1.264582633972168, | |
| "sparsity": 0.38839301466941833, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.k_proj": { | |
| "mse": 0.0005491595948114991, | |
| "rmse": 0.023434154450534355, | |
| "relative_error": 0.1322335889109355, | |
| "max_error": 0.6623153686523438, | |
| "sparsity": 0.38570529222488403, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.v_proj": { | |
| "mse": 0.0002180356823373586, | |
| "rmse": 0.014766031367207595, | |
| "relative_error": 0.14069761966818478, | |
| "max_error": 0.39208221435546875, | |
| "sparsity": 0.3902430534362793, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.o_proj": { | |
| "mse": 0.0002635375130921602, | |
| "rmse": 0.016233838519960714, | |
| "relative_error": 0.14729204831196044, | |
| "max_error": 0.616729736328125, | |
| "sparsity": 0.3769415517648061, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.mlp.gate_proj": { | |
| "mse": 0.00030637270538136363, | |
| "rmse": 0.017503505516934706, | |
| "relative_error": 0.1254795659450171, | |
| "max_error": 0.7698822021484375, | |
| "sparsity": 0.38287285963694256, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.mlp.up_proj": { | |
| "mse": 0.00028958218172192574, | |
| "rmse": 0.01701711437705952, | |
| "relative_error": 0.12602124521112892, | |
| "max_error": 1.2134326696395874, | |
| "sparsity": 0.3851913809776306, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.mlp.down_proj": { | |
| "mse": 0.00035804673098027706, | |
| "rmse": 0.01892212279265403, | |
| "relative_error": 0.14090776215805934, | |
| "max_error": 0.802337646484375, | |
| "sparsity": 0.37745631734530133, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.q_proj": { | |
| "mse": 0.0003613919543568045, | |
| "rmse": 0.01901031179009972, | |
| "relative_error": 0.12513344299468762, | |
| "max_error": 0.540191650390625, | |
| "sparsity": 0.3899775743484497, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.k_proj": { | |
| "mse": 0.00044430969865061343, | |
| "rmse": 0.021078655048427864, | |
| "relative_error": 0.1268041914084934, | |
| "max_error": 0.4692869186401367, | |
| "sparsity": 0.38868381579717, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.v_proj": { | |
| "mse": 0.00027094484539702535, | |
| "rmse": 0.016460402346146503, | |
| "relative_error": 0.13914137611104635, | |
| "max_error": 0.44549560546875, | |
| "sparsity": 0.39340495069821674, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.o_proj": { | |
| "mse": 0.00023530681210104376, | |
| "rmse": 0.015339713559941195, | |
| "relative_error": 0.1310271094866408, | |
| "max_error": 0.36243438720703125, | |
| "sparsity": 0.38202764590581256, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.mlp.gate_proj": { | |
| "mse": 0.0002719155454542488, | |
| "rmse": 0.01648986189918669, | |
| "relative_error": 0.11967172787876482, | |
| "max_error": 0.6721744537353516, | |
| "sparsity": 0.38415757815043133, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.mlp.up_proj": { | |
| "mse": 0.00026822477229870856, | |
| "rmse": 0.01637756918161876, | |
| "relative_error": 0.12010204469956566, | |
| "max_error": 0.8538665771484375, | |
| "sparsity": 0.3853982190291087, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.mlp.down_proj": { | |
| "mse": 0.00034209448494948447, | |
| "rmse": 0.018495796412955148, | |
| "relative_error": 0.1367644053035181, | |
| "max_error": 0.67559814453125, | |
| "sparsity": 0.3778640429178874, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.q_proj": { | |
| "mse": 0.0003616748144850135, | |
| "rmse": 0.01901774998481717, | |
| "relative_error": 0.12628190765749414, | |
| "max_error": 0.9189605712890625, | |
| "sparsity": 0.39154622952143353, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.k_proj": { | |
| "mse": 0.00047550429007969797, | |
| "rmse": 0.021806060856553115, | |
| "relative_error": 0.129471610229564, | |
| "max_error": 0.5678806304931641, | |
| "sparsity": 0.39027778307596844, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.v_proj": { | |
| "mse": 0.0002661354374140501, | |
| "rmse": 0.01631365800224003, | |
| "relative_error": 0.13591592833744226, | |
| "max_error": 0.5244388580322266, | |
| "sparsity": 0.3951323727766673, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.o_proj": { | |
| "mse": 0.0002390069275861606, | |
| "rmse": 0.01545984888626537, | |
| "relative_error": 0.13306984855497975, | |
| "max_error": 0.60205078125, | |
| "sparsity": 0.38461371262868244, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.mlp.gate_proj": { | |
| "mse": 0.0002712093119043857, | |
| "rmse": 0.01646843380241077, | |
| "relative_error": 0.12193638981151585, | |
| "max_error": 1.244659423828125, | |
| "sparsity": 0.3867263396581014, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.mlp.up_proj": { | |
| "mse": 0.00027307606069371104, | |
| "rmse": 0.01652501318286043, | |
| "relative_error": 0.12047243753722714, | |
| "max_error": 1.0329666137695312, | |
| "sparsity": 0.3885423441727956, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.mlp.down_proj": { | |
| "mse": 0.00036778251524083316, | |
| "rmse": 0.019177656667091345, | |
| "relative_error": 0.1406369310585435, | |
| "max_error": 0.8620948791503906, | |
| "sparsity": 0.37961751222610474, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.q_proj": { | |
| "mse": 0.0003682796086650342, | |
| "rmse": 0.0191906125140662, | |
| "relative_error": 0.13489597162840886, | |
| "max_error": 1.09013032913208, | |
| "sparsity": 0.3953443268934886, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.k_proj": { | |
| "mse": 0.00038725545164197683, | |
| "rmse": 0.01967880717020157, | |
| "relative_error": 0.13528558890175782, | |
| "max_error": 0.5375213623046875, | |
| "sparsity": 0.395026038090388, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.v_proj": { | |
| "mse": 0.00031340791611000896, | |
| "rmse": 0.017703330650191477, | |
| "relative_error": 0.12978148488562502, | |
| "max_error": 0.40410614013671875, | |
| "sparsity": 0.4009483555952708, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.o_proj": { | |
| "mse": 0.00031776560354046524, | |
| "rmse": 0.017825981138228135, | |
| "relative_error": 0.14448004936879605, | |
| "max_error": 0.6895837783813477, | |
| "sparsity": 0.3760470946629842, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.mlp.gate_proj": { | |
| "mse": 0.0002588577044662088, | |
| "rmse": 0.0160890554249219, | |
| "relative_error": 0.11954683494452101, | |
| "max_error": 0.94500732421875, | |
| "sparsity": 0.38806044061978656, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.mlp.up_proj": { | |
| "mse": 0.0002786662080325186, | |
| "rmse": 0.01669329829699687, | |
| "relative_error": 0.1202450386057031, | |
| "max_error": 0.7976341247558594, | |
| "sparsity": 0.3911337653795878, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.mlp.down_proj": { | |
| "mse": 0.0003937930450774729, | |
| "rmse": 0.01984421943734429, | |
| "relative_error": 0.14486439657367087, | |
| "max_error": 0.9992599487304688, | |
| "sparsity": 0.38100720445315045, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.q_proj": { | |
| "mse": 0.00034422375028952956, | |
| "rmse": 0.018553267914023383, | |
| "relative_error": 0.1254131775221598, | |
| "max_error": 0.715057373046875, | |
| "sparsity": 0.39250216881434125, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.k_proj": { | |
| "mse": 0.0003975920262746513, | |
| "rmse": 0.01993970978411299, | |
| "relative_error": 0.12661008238733615, | |
| "max_error": 0.6119565963745117, | |
| "sparsity": 0.39073241750399273, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.v_proj": { | |
| "mse": 0.00026132757193408906, | |
| "rmse": 0.016165629339252124, | |
| "relative_error": 0.12555049872573287, | |
| "max_error": 0.37404346466064453, | |
| "sparsity": 0.3957020441691081, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.o_proj": { | |
| "mse": 0.00026310610701330006, | |
| "rmse": 0.01622054582969698, | |
| "relative_error": 0.13168620320457766, | |
| "max_error": 0.442962646484375, | |
| "sparsity": 0.38386067748069763, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.mlp.gate_proj": { | |
| "mse": 0.0002558571577537805, | |
| "rmse": 0.015995535556954023, | |
| "relative_error": 0.11325361065918169, | |
| "max_error": 0.9491653442382812, | |
| "sparsity": 0.3881848255793254, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.mlp.up_proj": { | |
| "mse": 0.00023757842427585274, | |
| "rmse": 0.015413579216906524, | |
| "relative_error": 0.116968830179925, | |
| "max_error": 0.847259521484375, | |
| "sparsity": 0.3907446364561717, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.mlp.down_proj": { | |
| "mse": 0.00025847851065918803, | |
| "rmse": 0.01607726689021452, | |
| "relative_error": 0.12475321930392438, | |
| "max_error": 0.78851318359375, | |
| "sparsity": 0.3859715362389882, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.q_proj": { | |
| "mse": 0.00041562269325368106, | |
| "rmse": 0.020386826463520042, | |
| "relative_error": 0.13777912040773735, | |
| "max_error": 0.6402115821838379, | |
| "sparsity": 0.3940067986647288, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.k_proj": { | |
| "mse": 0.0004294166574254632, | |
| "rmse": 0.020722370941218652, | |
| "relative_error": 0.13392695081163608, | |
| "max_error": 0.5606307983398438, | |
| "sparsity": 0.3945269087950389, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.v_proj": { | |
| "mse": 0.0003200309001840651, | |
| "rmse": 0.017889407485550356, | |
| "relative_error": 0.1416881917728283, | |
| "max_error": 0.3849606513977051, | |
| "sparsity": 0.39705511927604675, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.o_proj": { | |
| "mse": 0.0002577567647676915, | |
| "rmse": 0.016054805036738735, | |
| "relative_error": 0.1323520817319322, | |
| "max_error": 0.43770599365234375, | |
| "sparsity": 0.38120298584302265, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.mlp.gate_proj": { | |
| "mse": 0.0002480960974935442, | |
| "rmse": 0.015751066550984547, | |
| "relative_error": 0.11325578808041654, | |
| "max_error": 0.7400054931640625, | |
| "sparsity": 0.3888651529947917, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.mlp.up_proj": { | |
| "mse": 0.00024302767997141927, | |
| "rmse": 0.015589345078335372, | |
| "relative_error": 0.1142694178958892, | |
| "max_error": 1.1082763671875, | |
| "sparsity": 0.3910396297772725, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.mlp.down_proj": { | |
| "mse": 0.000272468663752079, | |
| "rmse": 0.01650662484434898, | |
| "relative_error": 0.12340313693533053, | |
| "max_error": 0.5508880615234375, | |
| "sparsity": 0.38616835077603656, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.q_proj": { | |
| "mse": 0.00038834658334963024, | |
| "rmse": 0.019706511191726207, | |
| "relative_error": 0.12978617132646444, | |
| "max_error": 0.57098388671875, | |
| "sparsity": 0.3928031027317047, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.k_proj": { | |
| "mse": 0.00039985234616324306, | |
| "rmse": 0.019996308313367323, | |
| "relative_error": 0.12324863438662488, | |
| "max_error": 0.45305633544921875, | |
| "sparsity": 0.3924110134442647, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.v_proj": { | |
| "mse": 0.00028403327451087534, | |
| "rmse": 0.01685328675691704, | |
| "relative_error": 0.13895375705218863, | |
| "max_error": 0.4657020568847656, | |
| "sparsity": 0.4025857150554657, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.o_proj": { | |
| "mse": 0.00029896950582042336, | |
| "rmse": 0.01729073468133796, | |
| "relative_error": 0.1420016061368181, | |
| "max_error": 1.049102783203125, | |
| "sparsity": 0.37884873151779175, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.mlp.gate_proj": { | |
| "mse": 0.00023728626547381282, | |
| "rmse": 0.015404098982862088, | |
| "relative_error": 0.11325386121717744, | |
| "max_error": 1.0940780639648438, | |
| "sparsity": 0.3904181520144145, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.mlp.up_proj": { | |
| "mse": 0.0002466808073222637, | |
| "rmse": 0.015706075490785842, | |
| "relative_error": 0.11168400199124448, | |
| "max_error": 0.833099365234375, | |
| "sparsity": 0.39163845777511597, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.mlp.down_proj": { | |
| "mse": 0.00029290278325788677, | |
| "rmse": 0.017114402801672243, | |
| "relative_error": 0.1239319961557224, | |
| "max_error": 0.8251514434814453, | |
| "sparsity": 0.3860371907552083, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.q_proj": { | |
| "mse": 0.00031106252572499216, | |
| "rmse": 0.017636964753749217, | |
| "relative_error": 0.12348042819474542, | |
| "max_error": 0.6983642578125, | |
| "sparsity": 0.39465096592903137, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.k_proj": { | |
| "mse": 0.0003760141262318939, | |
| "rmse": 0.019391083678636786, | |
| "relative_error": 0.1292138409117064, | |
| "max_error": 0.5228118896484375, | |
| "sparsity": 0.3956749141216278, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.v_proj": { | |
| "mse": 0.00025429093511775136, | |
| "rmse": 0.015946502284756723, | |
| "relative_error": 0.12572292968558507, | |
| "max_error": 0.36475300788879395, | |
| "sparsity": 0.39897027611732483, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.o_proj": { | |
| "mse": 0.00025947412359528244, | |
| "rmse": 0.016108200507669455, | |
| "relative_error": 0.13100443210315335, | |
| "max_error": 0.7092132568359375, | |
| "sparsity": 0.38542644182840985, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.mlp.gate_proj": { | |
| "mse": 0.0002409738808637485, | |
| "rmse": 0.015523333432731145, | |
| "relative_error": 0.11615620591097975, | |
| "max_error": 1.0490875244140625, | |
| "sparsity": 0.3923046886920929, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.mlp.up_proj": { | |
| "mse": 0.0002738493785727769, | |
| "rmse": 0.016548395045223478, | |
| "relative_error": 0.11498682420142314, | |
| "max_error": 0.9261474609375, | |
| "sparsity": 0.39341633518536884, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.mlp.down_proj": { | |
| "mse": 0.0004093392926733941, | |
| "rmse": 0.020232135148653838, | |
| "relative_error": 0.14377147246869704, | |
| "max_error": 1.0236854553222656, | |
| "sparsity": 0.3809771140416463, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.q_proj": { | |
| "mse": 0.0004914459423162043, | |
| "rmse": 0.022168580069914363, | |
| "relative_error": 0.14693239451782622, | |
| "max_error": 1.157958984375, | |
| "sparsity": 0.3960525194803874, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.k_proj": { | |
| "mse": 0.000391253677662462, | |
| "rmse": 0.019780133408611327, | |
| "relative_error": 0.12477564180557132, | |
| "max_error": 0.4056215286254883, | |
| "sparsity": 0.3967556357383728, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.v_proj": { | |
| "mse": 0.00025476847076788545, | |
| "rmse": 0.0159614683149103, | |
| "relative_error": 0.1314226038419864, | |
| "max_error": 0.3336639404296875, | |
| "sparsity": 0.39837456742922467, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.o_proj": { | |
| "mse": 0.0002862376859411597, | |
| "rmse": 0.0169185603980114, | |
| "relative_error": 0.13909520553827767, | |
| "max_error": 0.398468017578125, | |
| "sparsity": 0.3766728142897288, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.mlp.gate_proj": { | |
| "mse": 0.00023511608014814556, | |
| "rmse": 0.015333495366293543, | |
| "relative_error": 0.11856949430841904, | |
| "max_error": 0.8280487060546875, | |
| "sparsity": 0.3905159632364909, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.mlp.up_proj": { | |
| "mse": 0.000292447890387848, | |
| "rmse": 0.01710110787018923, | |
| "relative_error": 0.11643273838359515, | |
| "max_error": 0.6393928527832031, | |
| "sparsity": 0.3939596911271413, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.mlp.down_proj": { | |
| "mse": 0.00034808836062438786, | |
| "rmse": 0.018657126269187006, | |
| "relative_error": 0.13070720125514787, | |
| "max_error": 0.682891845703125, | |
| "sparsity": 0.3866799672444661, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.q_proj": { | |
| "mse": 0.0004053004668094218, | |
| "rmse": 0.020132075571322045, | |
| "relative_error": 0.12942652560227816, | |
| "max_error": 0.71820068359375, | |
| "sparsity": 0.39477864901224774, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.k_proj": { | |
| "mse": 0.0004324800684116781, | |
| "rmse": 0.020796155135305132, | |
| "relative_error": 0.12896323884624275, | |
| "max_error": 0.6227951049804688, | |
| "sparsity": 0.39430447419484455, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.v_proj": { | |
| "mse": 0.0003173022996634245, | |
| "rmse": 0.01781298121212237, | |
| "relative_error": 0.14182240400875645, | |
| "max_error": 0.6628785133361816, | |
| "sparsity": 0.39759330948193866, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.o_proj": { | |
| "mse": 0.00027005953597836196, | |
| "rmse": 0.016433488247428237, | |
| "relative_error": 0.13571284752522653, | |
| "max_error": 0.7591552734375, | |
| "sparsity": 0.3861718773841858, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.mlp.gate_proj": { | |
| "mse": 0.0002359250938752666, | |
| "rmse": 0.015359853315551767, | |
| "relative_error": 0.11945819984312417, | |
| "max_error": 0.784759521484375, | |
| "sparsity": 0.3910655379295349, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.mlp.up_proj": { | |
| "mse": 0.00031376053811982274, | |
| "rmse": 0.01771328705011644, | |
| "relative_error": 0.11943581983457419, | |
| "max_error": 1.547119140625, | |
| "sparsity": 0.39500434199968976, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.mlp.down_proj": { | |
| "mse": 0.0003877365670632571, | |
| "rmse": 0.019691027577636903, | |
| "relative_error": 0.1375730755535001, | |
| "max_error": 1.1070480346679688, | |
| "sparsity": 0.38813774784406024, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.q_proj": { | |
| "mse": 0.00043195407488383353, | |
| "rmse": 0.020783504874872127, | |
| "relative_error": 0.14014456795182906, | |
| "max_error": 1.1500954627990723, | |
| "sparsity": 0.3933456341425578, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.k_proj": { | |
| "mse": 0.00047492279554717243, | |
| "rmse": 0.021792723454106704, | |
| "relative_error": 0.14076720736364193, | |
| "max_error": 1.0417613983154297, | |
| "sparsity": 0.39103733499844867, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.v_proj": { | |
| "mse": 0.0002991932269651443, | |
| "rmse": 0.017297202865352083, | |
| "relative_error": 0.13732420381908977, | |
| "max_error": 0.3966026306152344, | |
| "sparsity": 0.39792752265930176, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.o_proj": { | |
| "mse": 0.0002707390522118658, | |
| "rmse": 0.016454149999676853, | |
| "relative_error": 0.13354189124164606, | |
| "max_error": 0.79656982421875, | |
| "sparsity": 0.3795540432135264, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.mlp.gate_proj": { | |
| "mse": 0.00024663598742336035, | |
| "rmse": 0.015704648592800807, | |
| "relative_error": 0.12067143447230794, | |
| "max_error": 0.9877128601074219, | |
| "sparsity": 0.3910504877567291, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.mlp.up_proj": { | |
| "mse": 0.00029432709561660886, | |
| "rmse": 0.017155963849828107, | |
| "relative_error": 0.11803363898383314, | |
| "max_error": 0.685436487197876, | |
| "sparsity": 0.39433077971140545, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.mlp.down_proj": { | |
| "mse": 0.000363750325050205, | |
| "rmse": 0.019072239644315637, | |
| "relative_error": 0.13519176110611156, | |
| "max_error": 0.9110107421875, | |
| "sparsity": 0.3868025739987691, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.q_proj": { | |
| "mse": 0.0003545563668012619, | |
| "rmse": 0.0188296671983671, | |
| "relative_error": 0.1252711574598276, | |
| "max_error": 0.7074196338653564, | |
| "sparsity": 0.3935478130976359, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.k_proj": { | |
| "mse": 0.00045099898125045, | |
| "rmse": 0.021236736596060374, | |
| "relative_error": 0.13586665545865728, | |
| "max_error": 0.6388540267944336, | |
| "sparsity": 0.39120660225550336, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.v_proj": { | |
| "mse": 0.0002450101019348949, | |
| "rmse": 0.015652798533645505, | |
| "relative_error": 0.12490046996554008, | |
| "max_error": 0.3263397216796875, | |
| "sparsity": 0.3997873266537984, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.o_proj": { | |
| "mse": 0.00031266873702406883, | |
| "rmse": 0.017682441489343854, | |
| "relative_error": 0.14236252818242942, | |
| "max_error": 0.533966064453125, | |
| "sparsity": 0.37480757633845013, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.mlp.gate_proj": { | |
| "mse": 0.00029599471599794924, | |
| "rmse": 0.01720449697020954, | |
| "relative_error": 0.12983011117592555, | |
| "max_error": 1.7310435771942139, | |
| "sparsity": 0.39267564813296, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.mlp.up_proj": { | |
| "mse": 0.00030257849721238017, | |
| "rmse": 0.01739478362074045, | |
| "relative_error": 0.12050215488502299, | |
| "max_error": 0.97015380859375, | |
| "sparsity": 0.3948414425055186, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.mlp.down_proj": { | |
| "mse": 0.00045694861910305917, | |
| "rmse": 0.021376356544160167, | |
| "relative_error": 0.1531174021900089, | |
| "max_error": 1.0404205322265625, | |
| "sparsity": 0.3818763593832652, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.q_proj": { | |
| "mse": 0.00035391293931752443, | |
| "rmse": 0.01881257396842666, | |
| "relative_error": 0.1285467129658473, | |
| "max_error": 0.6924819946289062, | |
| "sparsity": 0.3974920411904653, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.k_proj": { | |
| "mse": 0.0004143030964769423, | |
| "rmse": 0.02035443677621521, | |
| "relative_error": 0.13067844683847013, | |
| "max_error": 0.5817642211914062, | |
| "sparsity": 0.39565104246139526, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.v_proj": { | |
| "mse": 0.00038161498378030956, | |
| "rmse": 0.019534968230849763, | |
| "relative_error": 0.147332272604324, | |
| "max_error": 0.9012093544006348, | |
| "sparsity": 0.4072374105453491, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.o_proj": { | |
| "mse": 0.00033256871392950416, | |
| "rmse": 0.018236466596616357, | |
| "relative_error": 0.14477406632677517, | |
| "max_error": 0.4289398193359375, | |
| "sparsity": 0.37938331564267475, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.mlp.gate_proj": { | |
| "mse": 0.00028728891629725695, | |
| "rmse": 0.01694959929606765, | |
| "relative_error": 0.12529640050529323, | |
| "max_error": 2.8129539489746094, | |
| "sparsity": 0.39487291375796, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.mlp.up_proj": { | |
| "mse": 0.00029650842770934105, | |
| "rmse": 0.017219420074710445, | |
| "relative_error": 0.11983224446764382, | |
| "max_error": 1.1838493347167969, | |
| "sparsity": 0.39692315459251404, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.mlp.down_proj": { | |
| "mse": 0.00039985671173781157, | |
| "rmse": 0.019996417472582722, | |
| "relative_error": 0.1455710859387371, | |
| "max_error": 0.802032470703125, | |
| "sparsity": 0.38471150398254395, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.q_proj": { | |
| "mse": 0.0004023751534987241, | |
| "rmse": 0.020059290952043247, | |
| "relative_error": 0.14578801331360433, | |
| "max_error": 0.7612266540527344, | |
| "sparsity": 0.40085141857465106, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.k_proj": { | |
| "mse": 0.00043891152017749846, | |
| "rmse": 0.020950215277593175, | |
| "relative_error": 0.14059755161229245, | |
| "max_error": 0.74713134765625, | |
| "sparsity": 0.4005891978740692, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.v_proj": { | |
| "mse": 0.00029954846831969917, | |
| "rmse": 0.01730746857052467, | |
| "relative_error": 0.1481067317555512, | |
| "max_error": 0.6542212963104248, | |
| "sparsity": 0.40333550175031024, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.o_proj": { | |
| "mse": 0.0002567254123277962, | |
| "rmse": 0.01602265309890332, | |
| "relative_error": 0.13831738673252697, | |
| "max_error": 0.8655905723571777, | |
| "sparsity": 0.3795807957649231, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.mlp.gate_proj": { | |
| "mse": 0.0002522421127650887, | |
| "rmse": 0.015882131870913575, | |
| "relative_error": 0.12078378065085432, | |
| "max_error": 1.383749008178711, | |
| "sparsity": 0.39456597963968915, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.mlp.up_proj": { | |
| "mse": 0.0002984075399581343, | |
| "rmse": 0.017274476546573974, | |
| "relative_error": 0.12008383861653409, | |
| "max_error": 1.294891357421875, | |
| "sparsity": 0.39678724606831867, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.mlp.down_proj": { | |
| "mse": 0.0003980969777330756, | |
| "rmse": 0.01995236772248035, | |
| "relative_error": 0.1442788715270542, | |
| "max_error": 1.263031005859375, | |
| "sparsity": 0.3845464487870534, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.q_proj": { | |
| "mse": 0.0004637680249288678, | |
| "rmse": 0.021535273969208468, | |
| "relative_error": 0.1589079629753252, | |
| "max_error": 1.0300445556640625, | |
| "sparsity": 0.3949236770470937, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.k_proj": { | |
| "mse": 0.00039793920586816967, | |
| "rmse": 0.019948413617833616, | |
| "relative_error": 0.14562156625501185, | |
| "max_error": 0.6796073913574219, | |
| "sparsity": 0.39361003041267395, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.v_proj": { | |
| "mse": 0.00041931832674890757, | |
| "rmse": 0.020477263653840755, | |
| "relative_error": 0.1635639593646707, | |
| "max_error": 1.3378448486328125, | |
| "sparsity": 0.4030240873495738, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.o_proj": { | |
| "mse": 0.00032342210761271417, | |
| "rmse": 0.017983940269382406, | |
| "relative_error": 0.1426007133866918, | |
| "max_error": 0.7264251708984375, | |
| "sparsity": 0.38150570789972943, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.mlp.gate_proj": { | |
| "mse": 0.00028976169414818287, | |
| "rmse": 0.017022388027188865, | |
| "relative_error": 0.12964864613739466, | |
| "max_error": 1.3582611083984375, | |
| "sparsity": 0.3942805926005046, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.mlp.up_proj": { | |
| "mse": 0.00032685688347555697, | |
| "rmse": 0.01807918370600722, | |
| "relative_error": 0.1251529485843237, | |
| "max_error": 0.8334503173828125, | |
| "sparsity": 0.3965672453244527, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.mlp.down_proj": { | |
| "mse": 0.0004322672029957175, | |
| "rmse": 0.020791036602240822, | |
| "relative_error": 0.1497711394289894, | |
| "max_error": 0.8078460693359375, | |
| "sparsity": 0.3850051462650299, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.q_proj": { | |
| "mse": 0.0003896941489074379, | |
| "rmse": 0.01974067245327367, | |
| "relative_error": 0.13325108200686736, | |
| "max_error": 1.254664421081543, | |
| "sparsity": 0.39376699924468994, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.k_proj": { | |
| "mse": 0.00041156832594424486, | |
| "rmse": 0.020287146816254, | |
| "relative_error": 0.13210728904408894, | |
| "max_error": 0.5255508422851562, | |
| "sparsity": 0.3895898461341858, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.v_proj": { | |
| "mse": 0.00038736805436201394, | |
| "rmse": 0.019681667977130748, | |
| "relative_error": 0.1433697685089609, | |
| "max_error": 0.5119838714599609, | |
| "sparsity": 0.4001801212628682, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.o_proj": { | |
| "mse": 0.00028125158860348165, | |
| "rmse": 0.016770557194186533, | |
| "relative_error": 0.13204851607001525, | |
| "max_error": 0.4458160400390625, | |
| "sparsity": 0.37967918316523236, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.mlp.gate_proj": { | |
| "mse": 0.00032663048477843404, | |
| "rmse": 0.018072921312793734, | |
| "relative_error": 0.13361007385773416, | |
| "max_error": 2.034881591796875, | |
| "sparsity": 0.39373833934466046, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.mlp.up_proj": { | |
| "mse": 0.0003490446542855352, | |
| "rmse": 0.018682736798593915, | |
| "relative_error": 0.12953514226570328, | |
| "max_error": 1.2417755126953125, | |
| "sparsity": 0.3947412073612213, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.mlp.down_proj": { | |
| "mse": 0.0005468087620101869, | |
| "rmse": 0.023383942396657302, | |
| "relative_error": 0.16905879133109025, | |
| "max_error": 1.0428466796875, | |
| "sparsity": 0.3783057928085327, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.q_proj": { | |
| "mse": 0.00032731646206229925, | |
| "rmse": 0.018091889400012905, | |
| "relative_error": 0.13296779800436376, | |
| "max_error": 0.6055755615234375, | |
| "sparsity": 0.3961111108462016, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.k_proj": { | |
| "mse": 0.00031906779622659087, | |
| "rmse": 0.017862468928638922, | |
| "relative_error": 0.13306408632610922, | |
| "max_error": 0.5009899139404297, | |
| "sparsity": 0.39445637663205463, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.v_proj": { | |
| "mse": 0.0010864576324820518, | |
| "rmse": 0.03296145677123588, | |
| "relative_error": 0.233601473782399, | |
| "max_error": 1.5577328205108643, | |
| "sparsity": 0.4020301600297292, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.o_proj": { | |
| "mse": 0.00035141559783369303, | |
| "rmse": 0.018746082199587545, | |
| "relative_error": 0.14249920315906311, | |
| "max_error": 0.6178430318832397, | |
| "sparsity": 0.37696723143259686, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.mlp.gate_proj": { | |
| "mse": 0.0002806248376145959, | |
| "rmse": 0.01675186072096458, | |
| "relative_error": 0.1269324185889845, | |
| "max_error": 1.7291717529296875, | |
| "sparsity": 0.3946213126182556, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.mlp.up_proj": { | |
| "mse": 0.0003319134411867708, | |
| "rmse": 0.018218491737429055, | |
| "relative_error": 0.12558151840247786, | |
| "max_error": 0.9978065490722656, | |
| "sparsity": 0.3965361813704173, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.mlp.down_proj": { | |
| "mse": 0.0005792526644654572, | |
| "rmse": 0.02406766844680758, | |
| "relative_error": 0.17307819842319885, | |
| "max_error": 0.8018798828125, | |
| "sparsity": 0.3727709949016571, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.q_proj": { | |
| "mse": 0.00046296603977680206, | |
| "rmse": 0.021516645644170516, | |
| "relative_error": 0.1563077917107896, | |
| "max_error": 1.8148994445800781, | |
| "sparsity": 0.40086010098457336, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.k_proj": { | |
| "mse": 0.0003960970207117498, | |
| "rmse": 0.019902186329942492, | |
| "relative_error": 0.1465407128124648, | |
| "max_error": 0.8343238830566406, | |
| "sparsity": 0.3987782100836436, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.v_proj": { | |
| "mse": 0.0004818201996386051, | |
| "rmse": 0.021950403177131057, | |
| "relative_error": 0.1446332083614048, | |
| "max_error": 0.6265144348144531, | |
| "sparsity": 0.3999891479810079, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.o_proj": { | |
| "mse": 0.0003606123791541904, | |
| "rmse": 0.018989796711765778, | |
| "relative_error": 0.13786364482520144, | |
| "max_error": 0.6202011108398438, | |
| "sparsity": 0.38008535901705426, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.mlp.gate_proj": { | |
| "mse": 0.0002804403775371611, | |
| "rmse": 0.016746354156566768, | |
| "relative_error": 0.12733057276550633, | |
| "max_error": 1.127471923828125, | |
| "sparsity": 0.39562689264615375, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.mlp.up_proj": { | |
| "mse": 0.0003492810938041657, | |
| "rmse": 0.01868906348119578, | |
| "relative_error": 0.12867766251674148, | |
| "max_error": 0.7787200212478638, | |
| "sparsity": 0.3956194420655568, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.mlp.down_proj": { | |
| "mse": 0.0004999992088414729, | |
| "rmse": 0.02236066208414842, | |
| "relative_error": 0.15978571822457543, | |
| "max_error": 0.71771240234375, | |
| "sparsity": 0.37760470310846966, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.q_proj": { | |
| "mse": 0.0004021856584586203, | |
| "rmse": 0.020054567022466985, | |
| "relative_error": 0.14437445525936182, | |
| "max_error": 0.6470222473144531, | |
| "sparsity": 0.39068830013275146, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.k_proj": { | |
| "mse": 0.00033365990384481847, | |
| "rmse": 0.01826635989585277, | |
| "relative_error": 0.14008225933028182, | |
| "max_error": 0.477325439453125, | |
| "sparsity": 0.3892371952533722, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.v_proj": { | |
| "mse": 0.0006478288560174406, | |
| "rmse": 0.025452482315433216, | |
| "relative_error": 0.17123315471774958, | |
| "max_error": 0.71356201171875, | |
| "sparsity": 0.3960329790910085, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.o_proj": { | |
| "mse": 0.0003620412026066333, | |
| "rmse": 0.01902738034009499, | |
| "relative_error": 0.13769422973477458, | |
| "max_error": 0.6836128234863281, | |
| "sparsity": 0.3778938849767049, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.mlp.gate_proj": { | |
| "mse": 0.0003168500552419573, | |
| "rmse": 0.017800282448375847, | |
| "relative_error": 0.13268715293682903, | |
| "max_error": 1.0362014770507812, | |
| "sparsity": 0.3935291866461436, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.mlp.up_proj": { | |
| "mse": 0.0003891361120622605, | |
| "rmse": 0.0197265331992791, | |
| "relative_error": 0.13520297794484687, | |
| "max_error": 0.615497350692749, | |
| "sparsity": 0.3948841691017151, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.mlp.down_proj": { | |
| "mse": 0.0008043116540648043, | |
| "rmse": 0.02836038882076204, | |
| "relative_error": 0.20311435488158017, | |
| "max_error": 1.1024436950683594, | |
| "sparsity": 0.3633190989494324, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.q_proj": { | |
| "mse": 0.0004803207120858133, | |
| "rmse": 0.021916220296524975, | |
| "relative_error": 0.16486716854740027, | |
| "max_error": 1.235321044921875, | |
| "sparsity": 0.3935517966747284, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.k_proj": { | |
| "mse": 0.0003249625733587891, | |
| "rmse": 0.01802671831917249, | |
| "relative_error": 0.14422419397315825, | |
| "max_error": 0.5352425575256348, | |
| "sparsity": 0.386841356754303, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.v_proj": { | |
| "mse": 0.0005751471617259085, | |
| "rmse": 0.023982225954358543, | |
| "relative_error": 0.1580459159536353, | |
| "max_error": 0.8300552368164062, | |
| "sparsity": 0.4017306864261627, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.o_proj": { | |
| "mse": 0.0003667864657472819, | |
| "rmse": 0.01915167005112823, | |
| "relative_error": 0.13689448182689637, | |
| "max_error": 0.561619758605957, | |
| "sparsity": 0.38343966007232666, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.mlp.gate_proj": { | |
| "mse": 0.0003047164937015623, | |
| "rmse": 0.017456130547792152, | |
| "relative_error": 0.13115302478655436, | |
| "max_error": 0.812896728515625, | |
| "sparsity": 0.39314127961794537, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.mlp.up_proj": { | |
| "mse": 0.0003970471443608403, | |
| "rmse": 0.019926041863873527, | |
| "relative_error": 0.13375999423388735, | |
| "max_error": 0.5598907470703125, | |
| "sparsity": 0.39349907636642456, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.mlp.down_proj": { | |
| "mse": 0.0007750659133307636, | |
| "rmse": 0.027840005627347914, | |
| "relative_error": 0.19518316136884764, | |
| "max_error": 1.1442718505859375, | |
| "sparsity": 0.3629276553789775, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.q_proj": { | |
| "mse": 0.0006292385514825583, | |
| "rmse": 0.02508462779238628, | |
| "relative_error": 0.18952891182131756, | |
| "max_error": 0.9436798095703125, | |
| "sparsity": 0.3831785321235657, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.k_proj": { | |
| "mse": 0.0004591235483530909, | |
| "rmse": 0.021427168463263896, | |
| "relative_error": 0.17446175340619755, | |
| "max_error": 0.546600341796875, | |
| "sparsity": 0.3804145057996114, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.v_proj": { | |
| "mse": 0.0004375155840534717, | |
| "rmse": 0.020916873190165677, | |
| "relative_error": 0.13928792059497572, | |
| "max_error": 0.628143310546875, | |
| "sparsity": 0.40235785643259686, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.o_proj": { | |
| "mse": 0.0003790340560954064, | |
| "rmse": 0.01946879698634218, | |
| "relative_error": 0.13653305972852475, | |
| "max_error": 0.5649566650390625, | |
| "sparsity": 0.3839749793211619, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.mlp.gate_proj": { | |
| "mse": 0.000287585542537272, | |
| "rmse": 0.016958347282010473, | |
| "relative_error": 0.1283479041198418, | |
| "max_error": 0.6532325744628906, | |
| "sparsity": 0.3902471164862315, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.mlp.up_proj": { | |
| "mse": 0.00037556461757048965, | |
| "rmse": 0.01937948961068092, | |
| "relative_error": 0.13156661044997323, | |
| "max_error": 0.4593353271484375, | |
| "sparsity": 0.39123860994974774, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.mlp.down_proj": { | |
| "mse": 0.000621433777268976, | |
| "rmse": 0.024928573510511507, | |
| "relative_error": 0.17474916127539988, | |
| "max_error": 1.1068572998046875, | |
| "sparsity": 0.37065619230270386, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.q_proj": { | |
| "mse": 0.0005015458445996046, | |
| "rmse": 0.022395219235354778, | |
| "relative_error": 0.17169572521600163, | |
| "max_error": 0.5677337646484375, | |
| "sparsity": 0.3853060007095337, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.k_proj": { | |
| "mse": 0.00029890250880271196, | |
| "rmse": 0.017288797205205222, | |
| "relative_error": 0.14959696274730824, | |
| "max_error": 0.3920612335205078, | |
| "sparsity": 0.3819878399372101, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.v_proj": { | |
| "mse": 0.000724277866538614, | |
| "rmse": 0.026912411013110923, | |
| "relative_error": 0.14962019306647226, | |
| "max_error": 0.6297454833984375, | |
| "sparsity": 0.39883246024449664, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.o_proj": { | |
| "mse": 0.0004041783686261624, | |
| "rmse": 0.020104187838014306, | |
| "relative_error": 0.1295416984528413, | |
| "max_error": 0.5848617553710938, | |
| "sparsity": 0.38157660762468976, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.mlp.gate_proj": { | |
| "mse": 0.00027247756952419877, | |
| "rmse": 0.01650689460571548, | |
| "relative_error": 0.12545018602522295, | |
| "max_error": 0.8299407958984375, | |
| "sparsity": 0.3903500735759735, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.mlp.up_proj": { | |
| "mse": 0.0003646870609372854, | |
| "rmse": 0.019096781428745668, | |
| "relative_error": 0.1283754918157126, | |
| "max_error": 0.613494873046875, | |
| "sparsity": 0.39058443903923035, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.mlp.down_proj": { | |
| "mse": 0.0004927993286401033, | |
| "rmse": 0.022199083959481378, | |
| "relative_error": 0.1529945194550293, | |
| "max_error": 1.103057861328125, | |
| "sparsity": 0.37695950269699097, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.q_proj": { | |
| "mse": 0.0007047529215924442, | |
| "rmse": 0.02654718293138547, | |
| "relative_error": 0.21038494364638508, | |
| "max_error": 2.104689598083496, | |
| "sparsity": 0.3775596817334493, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.k_proj": { | |
| "mse": 0.00041854113806039095, | |
| "rmse": 0.020458277983750023, | |
| "relative_error": 0.18190260334084488, | |
| "max_error": 0.5651397705078125, | |
| "sparsity": 0.3764225244522095, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.v_proj": { | |
| "mse": 0.001006178674288094, | |
| "rmse": 0.03172031958048491, | |
| "relative_error": 0.16094822854012167, | |
| "max_error": 0.5909767150878906, | |
| "sparsity": 0.3886295557022095, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.o_proj": { | |
| "mse": 0.00039930478669703007, | |
| "rmse": 0.01998261210895688, | |
| "relative_error": 0.12292562984103378, | |
| "max_error": 0.5795326232910156, | |
| "sparsity": 0.3832038442293803, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.mlp.gate_proj": { | |
| "mse": 0.0002544289454817772, | |
| "rmse": 0.01595082899042483, | |
| "relative_error": 0.1222619364364751, | |
| "max_error": 0.472900390625, | |
| "sparsity": 0.3903925319512685, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.mlp.up_proj": { | |
| "mse": 0.00035120753454975784, | |
| "rmse": 0.01874053186411095, | |
| "relative_error": 0.12499276106744055, | |
| "max_error": 0.47731781005859375, | |
| "sparsity": 0.3901938199996948, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.mlp.down_proj": { | |
| "mse": 0.00044792864355258644, | |
| "rmse": 0.021164324783762567, | |
| "relative_error": 0.14414071967240039, | |
| "max_error": 0.8178272247314453, | |
| "sparsity": 0.3800166845321655, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.q_proj": { | |
| "mse": 0.0008198362193070352, | |
| "rmse": 0.028632782248797187, | |
| "relative_error": 0.23429249311394743, | |
| "max_error": 2.0684211254119873, | |
| "sparsity": 0.3688664734363556, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.k_proj": { | |
| "mse": 0.000433493813034147, | |
| "rmse": 0.020820514235583783, | |
| "relative_error": 0.19564769009440797, | |
| "max_error": 0.6053543090820312, | |
| "sparsity": 0.36629557609558105, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.v_proj": { | |
| "mse": 0.000945365522056818, | |
| "rmse": 0.030746796939792247, | |
| "relative_error": 0.15454764338401902, | |
| "max_error": 0.622802734375, | |
| "sparsity": 0.3821777403354645, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.o_proj": { | |
| "mse": 0.0003894663241226226, | |
| "rmse": 0.019734901168301365, | |
| "relative_error": 0.11860803840521572, | |
| "max_error": 0.6717376708984375, | |
| "sparsity": 0.38870225350062054, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.mlp.gate_proj": { | |
| "mse": 0.0002371702139498666, | |
| "rmse": 0.01540033161817844, | |
| "relative_error": 0.11839049007172214, | |
| "max_error": 0.4968757629394531, | |
| "sparsity": 0.3898722231388092, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.mlp.up_proj": { | |
| "mse": 0.00033398266532458365, | |
| "rmse": 0.018275192620724513, | |
| "relative_error": 0.12133559382817463, | |
| "max_error": 0.43823719024658203, | |
| "sparsity": 0.3903917074203491, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.mlp.down_proj": { | |
| "mse": 0.0004961593658663332, | |
| "rmse": 0.022274635033291416, | |
| "relative_error": 0.1508845522317275, | |
| "max_error": 0.6260223388671875, | |
| "sparsity": 0.37552665670712787, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.q_proj": { | |
| "mse": 0.0007220531697385013, | |
| "rmse": 0.026871047053259784, | |
| "relative_error": 0.22073010922767622, | |
| "max_error": 2.3776988983154297, | |
| "sparsity": 0.3688009977340698, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.k_proj": { | |
| "mse": 0.00037996683386154473, | |
| "rmse": 0.019492737977553198, | |
| "relative_error": 0.18836281532377985, | |
| "max_error": 0.4958610534667969, | |
| "sparsity": 0.36698784430821735, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.v_proj": { | |
| "mse": 0.00116857944522053, | |
| "rmse": 0.0341844912967932, | |
| "relative_error": 0.17169464285196595, | |
| "max_error": 0.6185035705566406, | |
| "sparsity": 0.37959418694178265, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.o_proj": { | |
| "mse": 0.0006720547680743039, | |
| "rmse": 0.025924019134275916, | |
| "relative_error": 0.1581579338945812, | |
| "max_error": 0.8412322998046875, | |
| "sparsity": 0.3735810915629069, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.mlp.gate_proj": { | |
| "mse": 0.00022489803086500615, | |
| "rmse": 0.014996600643646085, | |
| "relative_error": 0.115714184097836, | |
| "max_error": 0.4696512222290039, | |
| "sparsity": 0.39038098851839703, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.mlp.up_proj": { | |
| "mse": 0.00032189959893003106, | |
| "rmse": 0.017941560660378213, | |
| "relative_error": 0.11866134286342558, | |
| "max_error": 0.476654052734375, | |
| "sparsity": 0.39087199171384174, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.mlp.down_proj": { | |
| "mse": 0.0004895612364634871, | |
| "rmse": 0.022126030743526665, | |
| "relative_error": 0.14925474911149889, | |
| "max_error": 1.036773681640625, | |
| "sparsity": 0.37520698706309, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.q_proj": { | |
| "mse": 0.0006163844955153763, | |
| "rmse": 0.024827091966546873, | |
| "relative_error": 0.20982990187825332, | |
| "max_error": 2.61297607421875, | |
| "sparsity": 0.3778902590274811, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.k_proj": { | |
| "mse": 0.0002822788374032825, | |
| "rmse": 0.01680115583533712, | |
| "relative_error": 0.16959531387746635, | |
| "max_error": 0.38373374938964844, | |
| "sparsity": 0.3764355480670929, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.v_proj": { | |
| "mse": 0.0010366367641836405, | |
| "rmse": 0.03219684400967959, | |
| "relative_error": 0.15896200144834857, | |
| "max_error": 0.8222465515136719, | |
| "sparsity": 0.3907877604166667, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.o_proj": { | |
| "mse": 0.00035855130408890545, | |
| "rmse": 0.018935450987206653, | |
| "relative_error": 0.11633226927799124, | |
| "max_error": 0.4291839599609375, | |
| "sparsity": 0.38507161537806195, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.mlp.gate_proj": { | |
| "mse": 0.00022159340733196586, | |
| "rmse": 0.01488601381606123, | |
| "relative_error": 0.11488584610343797, | |
| "max_error": 0.4425048828125, | |
| "sparsity": 0.39049506187438965, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.mlp.up_proj": { | |
| "mse": 0.0003204459208063781, | |
| "rmse": 0.017901003346359613, | |
| "relative_error": 0.11815594409499737, | |
| "max_error": 0.458160400390625, | |
| "sparsity": 0.39040162165959674, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.mlp.down_proj": { | |
| "mse": 0.0005074003711342812, | |
| "rmse": 0.02252554929706002, | |
| "relative_error": 0.15142420969607912, | |
| "max_error": 1.1690216064453125, | |
| "sparsity": 0.37409966190656024, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.self_attn.q_proj": { | |
| "mse": 0.000330743525410071, | |
| "rmse": 0.0181863554735431, | |
| "relative_error": 0.16296419603354362, | |
| "max_error": 2.6734619140625, | |
| "sparsity": 0.3882693250974019, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.self_attn.k_proj": { | |
| "mse": 0.00015552040713373572, | |
| "rmse": 0.01247078213801106, | |
| "relative_error": 0.1298306243992016, | |
| "max_error": 0.3675403594970703, | |
| "sparsity": 0.38860785961151123, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.self_attn.v_proj": { | |
| "mse": 0.0009869136847555637, | |
| "rmse": 0.03141518239252422, | |
| "relative_error": 0.1478885440817368, | |
| "max_error": 0.8889617919921875, | |
| "sparsity": 0.40032769242922467, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.self_attn.o_proj": { | |
| "mse": 0.00041504771797917783, | |
| "rmse": 0.020372719945534464, | |
| "relative_error": 0.12272351921304352, | |
| "max_error": 0.7488861083984375, | |
| "sparsity": 0.3843702971935272, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.mlp.gate_proj": { | |
| "mse": 0.00024942512391135097, | |
| "rmse": 0.015793198659908986, | |
| "relative_error": 0.12099965617749729, | |
| "max_error": 0.4992237091064453, | |
| "sparsity": 0.3896881739298503, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.mlp.up_proj": { | |
| "mse": 0.00035625233431346714, | |
| "rmse": 0.018874647925549954, | |
| "relative_error": 0.1237330520571607, | |
| "max_error": 0.8887786865234375, | |
| "sparsity": 0.38925836483637494, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.30.mlp.down_proj": { | |
| "mse": 0.0006168729159981012, | |
| "rmse": 0.02483692646037551, | |
| "relative_error": 0.16833931163403912, | |
| "max_error": 0.8421630859375, | |
| "sparsity": 0.37400227785110474, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.self_attn.q_proj": { | |
| "mse": 0.00023798629990778863, | |
| "rmse": 0.015426804591612244, | |
| "relative_error": 0.13627539548587017, | |
| "max_error": 1.79534912109375, | |
| "sparsity": 0.3895214796066284, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.self_attn.k_proj": { | |
| "mse": 0.000154533758177422, | |
| "rmse": 0.012431160773532856, | |
| "relative_error": 0.11974400881879725, | |
| "max_error": 0.38648033142089844, | |
| "sparsity": 0.38752278685569763, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.self_attn.v_proj": { | |
| "mse": 0.0006401444552466273, | |
| "rmse": 0.025301076167756724, | |
| "relative_error": 0.13013701519197032, | |
| "max_error": 0.5755758285522461, | |
| "sparsity": 0.39263129234313965, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.self_attn.o_proj": { | |
| "mse": 0.00041197784594260156, | |
| "rmse": 0.02029723739681343, | |
| "relative_error": 0.12611778971549734, | |
| "max_error": 0.92120361328125, | |
| "sparsity": 0.3814876278241475, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.mlp.gate_proj": { | |
| "mse": 0.00030560250161215663, | |
| "rmse": 0.017481490257187933, | |
| "relative_error": 0.13104694248333834, | |
| "max_error": 0.8943405151367188, | |
| "sparsity": 0.38756319880485535, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.mlp.up_proj": { | |
| "mse": 0.0004041235661134124, | |
| "rmse": 0.02010282482919782, | |
| "relative_error": 0.1322530721920411, | |
| "max_error": 0.696624755859375, | |
| "sparsity": 0.3878062665462494, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.31.mlp.down_proj": { | |
| "mse": 0.0013301592553034425, | |
| "rmse": 0.03647134841630403, | |
| "relative_error": 0.26829023337261737, | |
| "max_error": 1.2521438598632812, | |
| "sparsity": 0.3534819881121318, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| } | |
| }, | |
| "summary": { | |
| "method_name": "Broad-tritplane3", | |
| "model_family": "image_text_to_text", | |
| "selected_components": [ | |
| "text_backbone", | |
| "multimodal_connector" | |
| ], | |
| "quantized_modules": 225, | |
| "quantized_params": 326369280, | |
| "quantized_fraction": 0.6431146020807851, | |
| "avg_relative_error": 0.141314631960763, | |
| "avg_effective_bits": 9.0, | |
| "full_model_effective_bits": 11.498197785434504, | |
| "compression_ratio": 1.3915224193019373 | |
| }, | |
| "method_name": "Broad-tritplane3", | |
| "model_family": "image_text_to_text", | |
| "format_family": "tritplane_small", | |
| "format_version": "1.0", | |
| "total_packed_bytes": 367165440, | |
| "total_fp16_bytes": 652738560, | |
| "compression_ratio": 1.7777777777777777 | |
| } |