Image-Text-to-Text
Transformers
English
ternary-quant
quantization
ternary
vlm
multimodal
video
edge
smolvlm
Instructions to use AsadIsmail/SmolVLM2-256M-Video-Instruct-ternary with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use AsadIsmail/SmolVLM2-256M-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/SmolVLM2-256M-Video-Instruct-ternary")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("AsadIsmail/SmolVLM2-256M-Video-Instruct-ternary", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use AsadIsmail/SmolVLM2-256M-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/SmolVLM2-256M-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/SmolVLM2-256M-Video-Instruct-ternary", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/AsadIsmail/SmolVLM2-256M-Video-Instruct-ternary
- SGLang
How to use AsadIsmail/SmolVLM2-256M-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/SmolVLM2-256M-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/SmolVLM2-256M-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/SmolVLM2-256M-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/SmolVLM2-256M-Video-Instruct-ternary", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use AsadIsmail/SmolVLM2-256M-Video-Instruct-ternary with Docker Model Runner:
docker model run hf.co/AsadIsmail/SmolVLM2-256M-Video-Instruct-ternary
| { | |
| "model_name": "HuggingFaceTB/SmolVLM2-256M-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": true, | |
| "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": 576, | |
| "intermediate_size": 1536, | |
| "num_hidden_layers": 30, | |
| "num_attention_heads": 9, | |
| "num_key_value_heads": 3, | |
| "hidden_act": "silu", | |
| "max_position_embeddings": 8192, | |
| "initializer_range": 0.041666666666666664, | |
| "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/SmolVLM2-256M-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": [ | |
| 576, | |
| 12288 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 7077888, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 7962624, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.0.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.1.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.2.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.3.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.4.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.5.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.6.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.7.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.8.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.9.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.10.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.11.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.12.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.13.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.14.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.15.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.16.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.17.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.18.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.19.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.20.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.21.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.22.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.23.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.24.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.25.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.26.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.27.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.28.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.q_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.k_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.v_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 192, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 110592, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 124416, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.self_attn.o_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 331776, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 373248, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.mlp.gate_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.mlp.up_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 1536, | |
| 576 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| }, | |
| "model.text_model.layers.29.mlp.down_proj": { | |
| "scheme": "tritplane_small_v1", | |
| "shape": [ | |
| 576, | |
| 1536 | |
| ], | |
| "dtype": "torch.float16", | |
| "num_elements": 884736, | |
| "n_planes": 3, | |
| "group_sizes": [ | |
| 32, | |
| 32, | |
| 32 | |
| ], | |
| "rescued_rows": 0, | |
| "stored_bytes": 995328, | |
| "effective_bits": 9.0 | |
| } | |
| }, | |
| "stats": { | |
| "model.connector.modality_projection.proj": { | |
| "mse": 7.197992545115994e-06, | |
| "rmse": 0.002682907479790534, | |
| "relative_error": 0.11205706606793851, | |
| "max_error": 0.08349037170410156, | |
| "sparsity": 0.39360107978185016, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.q_proj": { | |
| "mse": 0.0013483315706253052, | |
| "rmse": 0.03671963467445319, | |
| "relative_error": 0.16815540680186142, | |
| "max_error": 1.08331298828125, | |
| "sparsity": 0.3855794270833333, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.k_proj": { | |
| "mse": 0.0028410570230334997, | |
| "rmse": 0.053301566797173044, | |
| "relative_error": 0.17472703191922495, | |
| "max_error": 1.3830413818359375, | |
| "sparsity": 0.38705934087435406, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.v_proj": { | |
| "mse": 7.697434193687513e-05, | |
| "rmse": 0.008773502261746738, | |
| "relative_error": 0.26405431733549006, | |
| "max_error": 0.1550607681274414, | |
| "sparsity": 0.36772099137306213, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.self_attn.o_proj": { | |
| "mse": 8.906672155717388e-05, | |
| "rmse": 0.00943751670500105, | |
| "relative_error": 0.19060343236824306, | |
| "max_error": 0.45730066299438477, | |
| "sparsity": 0.37106863657633465, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.mlp.gate_proj": { | |
| "mse": 0.00040929930401034653, | |
| "rmse": 0.020231146878275253, | |
| "relative_error": 0.1419062065958032, | |
| "max_error": 1.097259521484375, | |
| "sparsity": 0.37886329491933185, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.mlp.up_proj": { | |
| "mse": 0.0004508332640398294, | |
| "rmse": 0.02123283457383468, | |
| "relative_error": 0.14234672531956097, | |
| "max_error": 0.53997802734375, | |
| "sparsity": 0.37992726763089496, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.0.mlp.down_proj": { | |
| "mse": 0.0008368903072550893, | |
| "rmse": 0.028929056452900243, | |
| "relative_error": 0.19196299090759175, | |
| "max_error": 1.2700939178466797, | |
| "sparsity": 0.3618815839290619, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.q_proj": { | |
| "mse": 0.0007013276917859912, | |
| "rmse": 0.02648259224067748, | |
| "relative_error": 0.13958020708504135, | |
| "max_error": 0.6333541870117188, | |
| "sparsity": 0.382051944732666, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.k_proj": { | |
| "mse": 0.000988653744570911, | |
| "rmse": 0.03144286476405912, | |
| "relative_error": 0.1455112403778434, | |
| "max_error": 0.6535797119140625, | |
| "sparsity": 0.3812180459499359, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.v_proj": { | |
| "mse": 0.0002527590550016612, | |
| "rmse": 0.01589839787530999, | |
| "relative_error": 0.1651288154639537, | |
| "max_error": 0.32366466522216797, | |
| "sparsity": 0.38062427441279095, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.self_attn.o_proj": { | |
| "mse": 0.0003450903750490397, | |
| "rmse": 0.01857660827624461, | |
| "relative_error": 0.16996100250599222, | |
| "max_error": 0.47043609619140625, | |
| "sparsity": 0.3712092836697896, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.mlp.gate_proj": { | |
| "mse": 0.000484843913000077, | |
| "rmse": 0.0220191714875941, | |
| "relative_error": 0.13968309802033246, | |
| "max_error": 1.3267822265625, | |
| "sparsity": 0.3746183514595032, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.mlp.up_proj": { | |
| "mse": 0.00040211790474131703, | |
| "rmse": 0.020052877717208498, | |
| "relative_error": 0.14426614649991462, | |
| "max_error": 0.9427471160888672, | |
| "sparsity": 0.3751269777615865, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.1.mlp.down_proj": { | |
| "mse": 0.00048813666217029095, | |
| "rmse": 0.022093815020731277, | |
| "relative_error": 0.15637205852884573, | |
| "max_error": 0.6649723052978516, | |
| "sparsity": 0.37116722265879315, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.q_proj": { | |
| "mse": 0.0006890347576700151, | |
| "rmse": 0.026249471569348117, | |
| "relative_error": 0.14443112460408408, | |
| "max_error": 0.7698745727539062, | |
| "sparsity": 0.38810723026593524, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.k_proj": { | |
| "mse": 0.0009982096962630749, | |
| "rmse": 0.03159445673315297, | |
| "relative_error": 0.14936239745733543, | |
| "max_error": 0.704331636428833, | |
| "sparsity": 0.3819384078184764, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.v_proj": { | |
| "mse": 0.0001580154785187915, | |
| "rmse": 0.012570420777316545, | |
| "relative_error": 0.1459083603552263, | |
| "max_error": 0.2442169189453125, | |
| "sparsity": 0.3867247800032298, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.self_attn.o_proj": { | |
| "mse": 0.00020851312729064375, | |
| "rmse": 0.014439983631938222, | |
| "relative_error": 0.1452175528692902, | |
| "max_error": 0.4612579345703125, | |
| "sparsity": 0.3761614263057709, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.mlp.gate_proj": { | |
| "mse": 0.00047677115071564913, | |
| "rmse": 0.02183508989483783, | |
| "relative_error": 0.14153918833106555, | |
| "max_error": 0.917572021484375, | |
| "sparsity": 0.37640417615572613, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.mlp.up_proj": { | |
| "mse": 0.00039637379813939333, | |
| "rmse": 0.0199091385584458, | |
| "relative_error": 0.14161712479816688, | |
| "max_error": 0.8540101051330566, | |
| "sparsity": 0.37754125396410626, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.2.mlp.down_proj": { | |
| "mse": 0.00046519990428350866, | |
| "rmse": 0.021568493324372676, | |
| "relative_error": 0.15361370060839757, | |
| "max_error": 1.094594955444336, | |
| "sparsity": 0.37401439746220905, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.q_proj": { | |
| "mse": 0.0005331539432518184, | |
| "rmse": 0.02309012653174119, | |
| "relative_error": 0.1348162689302894, | |
| "max_error": 0.6665191650390625, | |
| "sparsity": 0.38799169659614563, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.k_proj": { | |
| "mse": 0.0007470968994311988, | |
| "rmse": 0.02733307336234253, | |
| "relative_error": 0.1413098154511647, | |
| "max_error": 0.6169281005859375, | |
| "sparsity": 0.38139588634173077, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.v_proj": { | |
| "mse": 0.0004237517132423818, | |
| "rmse": 0.020585230463669377, | |
| "relative_error": 0.18180216432257135, | |
| "max_error": 0.5510983467102051, | |
| "sparsity": 0.3888135353724162, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.self_attn.o_proj": { | |
| "mse": 0.0002635819837450981, | |
| "rmse": 0.016235208152194973, | |
| "relative_error": 0.13647064133434378, | |
| "max_error": 0.424593448638916, | |
| "sparsity": 0.37899164358774823, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.mlp.gate_proj": { | |
| "mse": 0.0004195571527816355, | |
| "rmse": 0.020483094316573254, | |
| "relative_error": 0.13320362273749767, | |
| "max_error": 0.984832763671875, | |
| "sparsity": 0.3788798749446869, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.mlp.up_proj": { | |
| "mse": 0.0003752692136913538, | |
| "rmse": 0.01937186655155754, | |
| "relative_error": 0.13594513376034312, | |
| "max_error": 0.7312850952148438, | |
| "sparsity": 0.37912137309710187, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.3.mlp.down_proj": { | |
| "mse": 0.00043255381751805544, | |
| "rmse": 0.020797928202541123, | |
| "relative_error": 0.14610587095414906, | |
| "max_error": 0.5510711669921875, | |
| "sparsity": 0.37412817279497784, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.q_proj": { | |
| "mse": 0.0005180431180633605, | |
| "rmse": 0.022760560583240486, | |
| "relative_error": 0.13324238521975287, | |
| "max_error": 0.5334064960479736, | |
| "sparsity": 0.3855663637320201, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.k_proj": { | |
| "mse": 0.0006387970643118024, | |
| "rmse": 0.02527443499490745, | |
| "relative_error": 0.1322609457318361, | |
| "max_error": 0.5743579864501953, | |
| "sparsity": 0.38137779633204144, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.v_proj": { | |
| "mse": 0.0003402388247195631, | |
| "rmse": 0.018445563822219235, | |
| "relative_error": 0.15925160330699087, | |
| "max_error": 0.6718473434448242, | |
| "sparsity": 0.3893982768058777, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.self_attn.o_proj": { | |
| "mse": 0.0002361372025916353, | |
| "rmse": 0.015366756410890208, | |
| "relative_error": 0.12824887974136698, | |
| "max_error": 0.3770596981048584, | |
| "sparsity": 0.3795040448506673, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.mlp.gate_proj": { | |
| "mse": 0.00039061365532688797, | |
| "rmse": 0.0197639483739178, | |
| "relative_error": 0.12951246837987818, | |
| "max_error": 0.7590484619140625, | |
| "sparsity": 0.38080134987831116, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.mlp.up_proj": { | |
| "mse": 0.00035471480805426836, | |
| "rmse": 0.018833873952383465, | |
| "relative_error": 0.13288753791762106, | |
| "max_error": 0.8072547912597656, | |
| "sparsity": 0.38195424278577167, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.4.mlp.down_proj": { | |
| "mse": 0.00036673006252385676, | |
| "rmse": 0.019150197453913022, | |
| "relative_error": 0.13669866370563186, | |
| "max_error": 0.74365234375, | |
| "sparsity": 0.3782966434955597, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.q_proj": { | |
| "mse": 0.0005537824472412467, | |
| "rmse": 0.02353258267256798, | |
| "relative_error": 0.14450434182422822, | |
| "max_error": 0.7357746362686157, | |
| "sparsity": 0.3894193768501282, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.k_proj": { | |
| "mse": 0.0007001789635978639, | |
| "rmse": 0.026460894988602785, | |
| "relative_error": 0.14708869697686042, | |
| "max_error": 0.8047637939453125, | |
| "sparsity": 0.3886658449967702, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.v_proj": { | |
| "mse": 0.0002934595395345241, | |
| "rmse": 0.017130660802622998, | |
| "relative_error": 0.14227328603825723, | |
| "max_error": 0.3230438232421875, | |
| "sparsity": 0.3922887643178304, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.self_attn.o_proj": { | |
| "mse": 0.00024279598437715322, | |
| "rmse": 0.015581912089892987, | |
| "relative_error": 0.13013907191116078, | |
| "max_error": 0.72021484375, | |
| "sparsity": 0.38283059000968933, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.mlp.gate_proj": { | |
| "mse": 0.00037037619040347636, | |
| "rmse": 0.019245160181289123, | |
| "relative_error": 0.12710817255642473, | |
| "max_error": 0.94329833984375, | |
| "sparsity": 0.38237884640693665, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.mlp.up_proj": { | |
| "mse": 0.000334016076521948, | |
| "rmse": 0.01827610671127601, | |
| "relative_error": 0.12802060244515473, | |
| "max_error": 0.547760009765625, | |
| "sparsity": 0.3841914435227712, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.5.mlp.down_proj": { | |
| "mse": 0.0003303704143036157, | |
| "rmse": 0.018176094583370096, | |
| "relative_error": 0.1295633858602185, | |
| "max_error": 0.38228416442871094, | |
| "sparsity": 0.3808529774347941, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.q_proj": { | |
| "mse": 0.0005166067276149988, | |
| "rmse": 0.022728984306717246, | |
| "relative_error": 0.138632836397485, | |
| "max_error": 0.6797466278076172, | |
| "sparsity": 0.3860255181789398, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.k_proj": { | |
| "mse": 0.000593587988987565, | |
| "rmse": 0.02436366123938611, | |
| "relative_error": 0.1322367030222788, | |
| "max_error": 0.7423858642578125, | |
| "sparsity": 0.38642336924870807, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.v_proj": { | |
| "mse": 0.000291938689770177, | |
| "rmse": 0.017086213441549213, | |
| "relative_error": 0.13881721107153586, | |
| "max_error": 0.37963199615478516, | |
| "sparsity": 0.39365415771802265, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.self_attn.o_proj": { | |
| "mse": 0.0002378478238824755, | |
| "rmse": 0.01542231577560502, | |
| "relative_error": 0.12703174147183285, | |
| "max_error": 0.4151611328125, | |
| "sparsity": 0.38085536162058514, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.mlp.gate_proj": { | |
| "mse": 0.0003241289232391864, | |
| "rmse": 0.01800358084490934, | |
| "relative_error": 0.1198097989827277, | |
| "max_error": 0.8785018920898438, | |
| "sparsity": 0.3865718146165212, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.mlp.up_proj": { | |
| "mse": 0.00032126219593919814, | |
| "rmse": 0.017923788548719215, | |
| "relative_error": 0.12279689065912855, | |
| "max_error": 0.7001953125, | |
| "sparsity": 0.38781700531641644, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.6.mlp.down_proj": { | |
| "mse": 0.0004000381741207093, | |
| "rmse": 0.020000954330249076, | |
| "relative_error": 0.13876845110530803, | |
| "max_error": 0.60394287109375, | |
| "sparsity": 0.37938209374745685, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.q_proj": { | |
| "mse": 0.0005055120564065874, | |
| "rmse": 0.02248359527314498, | |
| "relative_error": 0.1499289102751937, | |
| "max_error": 0.6329377293586731, | |
| "sparsity": 0.3862636188666026, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.k_proj": { | |
| "mse": 0.0005039324169047177, | |
| "rmse": 0.02244843907501628, | |
| "relative_error": 0.14378671588156453, | |
| "max_error": 0.4136314392089844, | |
| "sparsity": 0.38603455821673077, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.v_proj": { | |
| "mse": 0.0003961301117669791, | |
| "rmse": 0.019903017654792427, | |
| "relative_error": 0.15188102000396358, | |
| "max_error": 0.3514251708984375, | |
| "sparsity": 0.39087216059366864, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.self_attn.o_proj": { | |
| "mse": 0.0002564229944255203, | |
| "rmse": 0.016013213119968155, | |
| "relative_error": 0.12790553872780777, | |
| "max_error": 0.44970273971557617, | |
| "sparsity": 0.38609283169110614, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.mlp.gate_proj": { | |
| "mse": 0.00032391323475167155, | |
| "rmse": 0.01799758969283586, | |
| "relative_error": 0.12125376696601475, | |
| "max_error": 0.6748809814453125, | |
| "sparsity": 0.38591625293095905, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.mlp.up_proj": { | |
| "mse": 0.0003243618120905012, | |
| "rmse": 0.018010047531600275, | |
| "relative_error": 0.12185872575431715, | |
| "max_error": 0.5790252685546875, | |
| "sparsity": 0.3880777259667714, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.7.mlp.down_proj": { | |
| "mse": 0.0003642227384261787, | |
| "rmse": 0.019084620468486625, | |
| "relative_error": 0.13154414043382356, | |
| "max_error": 0.715606689453125, | |
| "sparsity": 0.38265689214070636, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.q_proj": { | |
| "mse": 0.0005204624030739069, | |
| "rmse": 0.02281364510712628, | |
| "relative_error": 0.14934448046875148, | |
| "max_error": 0.6000328063964844, | |
| "sparsity": 0.3891400694847107, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.k_proj": { | |
| "mse": 0.0005498177488334477, | |
| "rmse": 0.023448192869247892, | |
| "relative_error": 0.14287133343382633, | |
| "max_error": 0.8906059265136719, | |
| "sparsity": 0.3896635075410207, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.v_proj": { | |
| "mse": 0.00037112177233211696, | |
| "rmse": 0.01926452107715416, | |
| "relative_error": 0.15247994251188193, | |
| "max_error": 0.3671379089355469, | |
| "sparsity": 0.39051347970962524, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.self_attn.o_proj": { | |
| "mse": 0.00028121424838900566, | |
| "rmse": 0.01676944389027274, | |
| "relative_error": 0.1365649522147827, | |
| "max_error": 0.3606128692626953, | |
| "sparsity": 0.3792518675327301, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.mlp.gate_proj": { | |
| "mse": 0.0003200193459633738, | |
| "rmse": 0.017889084547940786, | |
| "relative_error": 0.12279411165092348, | |
| "max_error": 0.75714111328125, | |
| "sparsity": 0.3854057391484578, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.mlp.up_proj": { | |
| "mse": 0.0003422408190090209, | |
| "rmse": 0.01849975186344457, | |
| "relative_error": 0.12242036334954902, | |
| "max_error": 0.69757080078125, | |
| "sparsity": 0.38790403803189594, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.8.mlp.down_proj": { | |
| "mse": 0.0004655543598346412, | |
| "rmse": 0.0215767087349911, | |
| "relative_error": 0.14664865020455325, | |
| "max_error": 0.8413848876953125, | |
| "sparsity": 0.3801288306713104, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.q_proj": { | |
| "mse": 0.000637753342743963, | |
| "rmse": 0.025253778781480662, | |
| "relative_error": 0.1518784089192349, | |
| "max_error": 1.01800537109375, | |
| "sparsity": 0.385803480943044, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.k_proj": { | |
| "mse": 0.0005865234998054802, | |
| "rmse": 0.024218247248830378, | |
| "relative_error": 0.14176735612251312, | |
| "max_error": 0.5834999084472656, | |
| "sparsity": 0.38840965429941815, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.v_proj": { | |
| "mse": 0.00033242779318243265, | |
| "rmse": 0.01823260247969095, | |
| "relative_error": 0.15765383117505966, | |
| "max_error": 0.29273271560668945, | |
| "sparsity": 0.3878158728281657, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.self_attn.o_proj": { | |
| "mse": 0.00028635465423576534, | |
| "rmse": 0.016922016848938702, | |
| "relative_error": 0.1454966294321818, | |
| "max_error": 0.3713645935058594, | |
| "sparsity": 0.37526623407999676, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.mlp.gate_proj": { | |
| "mse": 0.000311644485918805, | |
| "rmse": 0.017653455353522297, | |
| "relative_error": 0.1249146031586121, | |
| "max_error": 0.9822998046875, | |
| "sparsity": 0.38692521055539447, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.mlp.up_proj": { | |
| "mse": 0.00036663212813436985, | |
| "rmse": 0.01914764027587655, | |
| "relative_error": 0.12337701910061448, | |
| "max_error": 0.94927978515625, | |
| "sparsity": 0.3882254163424174, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.9.mlp.down_proj": { | |
| "mse": 0.0005080121918581426, | |
| "rmse": 0.022539125800663667, | |
| "relative_error": 0.14893565799269487, | |
| "max_error": 0.66717529296875, | |
| "sparsity": 0.3807440896828969, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.q_proj": { | |
| "mse": 0.0005783926462754607, | |
| "rmse": 0.02404979513998946, | |
| "relative_error": 0.14321199153596983, | |
| "max_error": 0.5550918579101562, | |
| "sparsity": 0.3850449323654175, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.k_proj": { | |
| "mse": 0.0006032341625541449, | |
| "rmse": 0.024560825771014802, | |
| "relative_error": 0.13983058068683443, | |
| "max_error": 0.5511245727539062, | |
| "sparsity": 0.3812873760859172, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.v_proj": { | |
| "mse": 0.00036479264963418245, | |
| "rmse": 0.019099545796541405, | |
| "relative_error": 0.15200630471135587, | |
| "max_error": 0.3874664306640625, | |
| "sparsity": 0.38743308186531067, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.self_attn.o_proj": { | |
| "mse": 0.0002817007771227509, | |
| "rmse": 0.016783944027633994, | |
| "relative_error": 0.1327482788234791, | |
| "max_error": 0.3631315231323242, | |
| "sparsity": 0.37979842225710553, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.mlp.gate_proj": { | |
| "mse": 0.0003068649675697088, | |
| "rmse": 0.017517561690192755, | |
| "relative_error": 0.1253477859427753, | |
| "max_error": 0.7594871520996094, | |
| "sparsity": 0.38676999012629193, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.mlp.up_proj": { | |
| "mse": 0.0003818606201093644, | |
| "rmse": 0.019541254312591204, | |
| "relative_error": 0.12517816125892817, | |
| "max_error": 0.77325439453125, | |
| "sparsity": 0.3887253701686859, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.10.mlp.down_proj": { | |
| "mse": 0.00045870940084569156, | |
| "rmse": 0.02141750220837368, | |
| "relative_error": 0.14044486522492353, | |
| "max_error": 0.635406494140625, | |
| "sparsity": 0.3837864299615224, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.q_proj": { | |
| "mse": 0.0007442517671734095, | |
| "rmse": 0.027280978119807387, | |
| "relative_error": 0.16183725129926163, | |
| "max_error": 1.0256156921386719, | |
| "sparsity": 0.3862244387467702, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.k_proj": { | |
| "mse": 0.000621186918579042, | |
| "rmse": 0.024923621698682598, | |
| "relative_error": 0.13973611133744107, | |
| "max_error": 0.509178638458252, | |
| "sparsity": 0.383620282014211, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.v_proj": { | |
| "mse": 0.0004583809641189873, | |
| "rmse": 0.021409833351032588, | |
| "relative_error": 0.1670124344223718, | |
| "max_error": 0.5122928619384766, | |
| "sparsity": 0.39319299658139545, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.self_attn.o_proj": { | |
| "mse": 0.0002903812564909458, | |
| "rmse": 0.01704057676520797, | |
| "relative_error": 0.13252044090501391, | |
| "max_error": 0.4071311950683594, | |
| "sparsity": 0.38034797708193463, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.mlp.gate_proj": { | |
| "mse": 0.0003105386858806014, | |
| "rmse": 0.01762210787280005, | |
| "relative_error": 0.12933114225218184, | |
| "max_error": 0.8038406372070312, | |
| "sparsity": 0.3841499984264374, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.mlp.up_proj": { | |
| "mse": 0.00040370377246290445, | |
| "rmse": 0.020092380955548907, | |
| "relative_error": 0.12837796720995137, | |
| "max_error": 0.685943603515625, | |
| "sparsity": 0.3878592054049174, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.11.mlp.down_proj": { | |
| "mse": 0.0004240164998918772, | |
| "rmse": 0.020591660930869008, | |
| "relative_error": 0.13469262611398652, | |
| "max_error": 0.9059791564941406, | |
| "sparsity": 0.3883192241191864, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.q_proj": { | |
| "mse": 0.0006758665549568832, | |
| "rmse": 0.02599743362251134, | |
| "relative_error": 0.15761044754162273, | |
| "max_error": 0.8436298370361328, | |
| "sparsity": 0.3802786668141683, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.k_proj": { | |
| "mse": 0.0007090824656188488, | |
| "rmse": 0.02662860239702506, | |
| "relative_error": 0.15551523386893096, | |
| "max_error": 0.559050977230072, | |
| "sparsity": 0.3804313739140828, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.v_proj": { | |
| "mse": 0.0004230560443829745, | |
| "rmse": 0.020568326241650645, | |
| "relative_error": 0.15647129868254786, | |
| "max_error": 0.3714733123779297, | |
| "sparsity": 0.3853955666224162, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.self_attn.o_proj": { | |
| "mse": 0.00032992413616739213, | |
| "rmse": 0.018163813921293955, | |
| "relative_error": 0.14528737615043072, | |
| "max_error": 0.6679039001464844, | |
| "sparsity": 0.3783516585826874, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.mlp.gate_proj": { | |
| "mse": 0.00033747541601769626, | |
| "rmse": 0.018370503967439115, | |
| "relative_error": 0.13214178641522772, | |
| "max_error": 1.1501893997192383, | |
| "sparsity": 0.3830558856328328, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.mlp.up_proj": { | |
| "mse": 0.00040743034332990646, | |
| "rmse": 0.020184903847427823, | |
| "relative_error": 0.13063388391145142, | |
| "max_error": 0.5685052871704102, | |
| "sparsity": 0.3861743410428365, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.12.mlp.down_proj": { | |
| "mse": 0.0005002933321520686, | |
| "rmse": 0.02236723791960171, | |
| "relative_error": 0.14873200405833786, | |
| "max_error": 0.9012985229492188, | |
| "sparsity": 0.38256346186002094, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.q_proj": { | |
| "mse": 0.0006164388614706695, | |
| "rmse": 0.024828186834134094, | |
| "relative_error": 0.14328696942268673, | |
| "max_error": 0.6360483169555664, | |
| "sparsity": 0.38452449440956116, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.k_proj": { | |
| "mse": 0.0007497998885810375, | |
| "rmse": 0.027382474113582897, | |
| "relative_error": 0.1531189776561995, | |
| "max_error": 0.9166738986968994, | |
| "sparsity": 0.3774353762467702, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.v_proj": { | |
| "mse": 0.0004211587365716696, | |
| "rmse": 0.02052215233769766, | |
| "relative_error": 0.16202526075703877, | |
| "max_error": 0.437835693359375, | |
| "sparsity": 0.3824477990468343, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.self_attn.o_proj": { | |
| "mse": 0.0002900735125876963, | |
| "rmse": 0.017031544633053584, | |
| "relative_error": 0.13733989618133918, | |
| "max_error": 0.37995147705078125, | |
| "sparsity": 0.3769260048866272, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.mlp.gate_proj": { | |
| "mse": 0.00037667236756533384, | |
| "rmse": 0.019408049040677267, | |
| "relative_error": 0.13379875792840404, | |
| "max_error": 1.042694091796875, | |
| "sparsity": 0.3838045100371043, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.mlp.up_proj": { | |
| "mse": 0.0003902724711224437, | |
| "rmse": 0.019755315009446033, | |
| "relative_error": 0.1317081900823692, | |
| "max_error": 0.5592594146728516, | |
| "sparsity": 0.38581980268160504, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.13.mlp.down_proj": { | |
| "mse": 0.00047372948029078543, | |
| "rmse": 0.02176532747952085, | |
| "relative_error": 0.15069972530670295, | |
| "max_error": 0.7411956787109375, | |
| "sparsity": 0.38301331798235577, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.q_proj": { | |
| "mse": 0.0006042954628355801, | |
| "rmse": 0.024582421826085, | |
| "relative_error": 0.16320079081404096, | |
| "max_error": 1.1795806884765625, | |
| "sparsity": 0.38293707370758057, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.k_proj": { | |
| "mse": 0.0006902359891682863, | |
| "rmse": 0.026272342666162952, | |
| "relative_error": 0.1706266573671637, | |
| "max_error": 0.91925048828125, | |
| "sparsity": 0.38038013378779095, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.v_proj": { | |
| "mse": 0.0005118001718074083, | |
| "rmse": 0.022623000946103686, | |
| "relative_error": 0.16328002565875313, | |
| "max_error": 0.455291748046875, | |
| "sparsity": 0.39064308007558185, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.self_attn.o_proj": { | |
| "mse": 0.00037746495218016207, | |
| "rmse": 0.019428457277410424, | |
| "relative_error": 0.1458681735320647, | |
| "max_error": 0.425567626953125, | |
| "sparsity": 0.37572739521662396, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.mlp.gate_proj": { | |
| "mse": 0.0003888887877110392, | |
| "rmse": 0.019720263378338515, | |
| "relative_error": 0.13598917960076568, | |
| "max_error": 1.51251220703125, | |
| "sparsity": 0.38527123133341473, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.mlp.up_proj": { | |
| "mse": 0.00041537353536114097, | |
| "rmse": 0.020380714790240822, | |
| "relative_error": 0.13581233566034792, | |
| "max_error": 1.3936614990234375, | |
| "sparsity": 0.38643692930539447, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.14.mlp.down_proj": { | |
| "mse": 0.0005030913162045181, | |
| "rmse": 0.02242969719377678, | |
| "relative_error": 0.1555272505212418, | |
| "max_error": 0.78131103515625, | |
| "sparsity": 0.3817029396692912, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.q_proj": { | |
| "mse": 0.0004953333409503102, | |
| "rmse": 0.022256085481286016, | |
| "relative_error": 0.1476775336378965, | |
| "max_error": 0.9005780220031738, | |
| "sparsity": 0.3890375892321269, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.k_proj": { | |
| "mse": 0.000503656396176666, | |
| "rmse": 0.0224422903505116, | |
| "relative_error": 0.14472695233278485, | |
| "max_error": 0.5825338363647461, | |
| "sparsity": 0.38626662890116376, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.v_proj": { | |
| "mse": 0.0004316933045629412, | |
| "rmse": 0.02077723043533332, | |
| "relative_error": 0.15378818724264243, | |
| "max_error": 0.43953514099121094, | |
| "sparsity": 0.3967345456282298, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.self_attn.o_proj": { | |
| "mse": 0.0006721029058098793, | |
| "rmse": 0.025924947556550223, | |
| "relative_error": 0.19645459071561105, | |
| "max_error": 0.67523193359375, | |
| "sparsity": 0.36476921041806537, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.mlp.gate_proj": { | |
| "mse": 0.0004530326114036143, | |
| "rmse": 0.021284562748706264, | |
| "relative_error": 0.14303214494613653, | |
| "max_error": 1.4091482162475586, | |
| "sparsity": 0.3865902821222941, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.mlp.up_proj": { | |
| "mse": 0.0004576505452860147, | |
| "rmse": 0.02139276852784638, | |
| "relative_error": 0.14229226999420433, | |
| "max_error": 1.36956787109375, | |
| "sparsity": 0.38725224137306213, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.15.mlp.down_proj": { | |
| "mse": 0.0006192649598233402, | |
| "rmse": 0.02488503485678371, | |
| "relative_error": 0.17374001759345364, | |
| "max_error": 0.9531974792480469, | |
| "sparsity": 0.3778705398241679, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.q_proj": { | |
| "mse": 0.0006575874285772443, | |
| "rmse": 0.02564346756149106, | |
| "relative_error": 0.15608954401553143, | |
| "max_error": 0.6125068664550781, | |
| "sparsity": 0.3841396967569987, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.k_proj": { | |
| "mse": 0.0007115787593647838, | |
| "rmse": 0.026675433630304564, | |
| "relative_error": 0.15354155216739387, | |
| "max_error": 0.572357177734375, | |
| "sparsity": 0.3814230064551036, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.v_proj": { | |
| "mse": 0.0006480485899373889, | |
| "rmse": 0.025456798501331406, | |
| "relative_error": 0.18891827773205863, | |
| "max_error": 0.7852058410644531, | |
| "sparsity": 0.3937717080116272, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.self_attn.o_proj": { | |
| "mse": 0.00033904294832609594, | |
| "rmse": 0.01841311891902336, | |
| "relative_error": 0.13880119776337932, | |
| "max_error": 0.642242431640625, | |
| "sparsity": 0.3791282872358958, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.mlp.gate_proj": { | |
| "mse": 0.00041361607145518064, | |
| "rmse": 0.020337553231772516, | |
| "relative_error": 0.13730181885927414, | |
| "max_error": 1.1731719970703125, | |
| "sparsity": 0.3867873152097066, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.mlp.up_proj": { | |
| "mse": 0.000469113263534382, | |
| "rmse": 0.02165902268188438, | |
| "relative_error": 0.14207421352305882, | |
| "max_error": 1.0403289794921875, | |
| "sparsity": 0.38714109857877094, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.16.mlp.down_proj": { | |
| "mse": 0.0005275580333545804, | |
| "rmse": 0.02296863150809339, | |
| "relative_error": 0.1583542605408581, | |
| "max_error": 0.8113040924072266, | |
| "sparsity": 0.3811118006706238, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.q_proj": { | |
| "mse": 0.0008716285228729248, | |
| "rmse": 0.02952335554900433, | |
| "relative_error": 0.20022424635526176, | |
| "max_error": 1.641489028930664, | |
| "sparsity": 0.38736175497372943, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.k_proj": { | |
| "mse": 0.0005483600543811917, | |
| "rmse": 0.02341708893908873, | |
| "relative_error": 0.1601606538375164, | |
| "max_error": 0.8247528076171875, | |
| "sparsity": 0.38466013471285504, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.v_proj": { | |
| "mse": 0.000659817480482161, | |
| "rmse": 0.02568691263040697, | |
| "relative_error": 0.1790594356512437, | |
| "max_error": 0.4578428268432617, | |
| "sparsity": 0.3840060730775197, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.self_attn.o_proj": { | |
| "mse": 0.0003701736277434975, | |
| "rmse": 0.01923989677060398, | |
| "relative_error": 0.14388707335289797, | |
| "max_error": 0.4621315002441406, | |
| "sparsity": 0.373514066139857, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.mlp.gate_proj": { | |
| "mse": 0.0004113586328458041, | |
| "rmse": 0.020281978030897384, | |
| "relative_error": 0.1389533819092617, | |
| "max_error": 0.9721755981445312, | |
| "sparsity": 0.3881768087546031, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.mlp.up_proj": { | |
| "mse": 0.0004939618520438671, | |
| "rmse": 0.022225252575479703, | |
| "relative_error": 0.14402551159901172, | |
| "max_error": 0.9300651550292969, | |
| "sparsity": 0.38775522510210675, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.17.mlp.down_proj": { | |
| "mse": 0.0005498765385709703, | |
| "rmse": 0.02344944644487307, | |
| "relative_error": 0.15951960838798954, | |
| "max_error": 0.6857528686523438, | |
| "sparsity": 0.3802957435448964, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.q_proj": { | |
| "mse": 0.0005783605156466365, | |
| "rmse": 0.024049127128580705, | |
| "relative_error": 0.17173222027156138, | |
| "max_error": 0.7184696197509766, | |
| "sparsity": 0.3897790511449178, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.k_proj": { | |
| "mse": 0.00047414624714292586, | |
| "rmse": 0.021774899474921252, | |
| "relative_error": 0.15233903987588057, | |
| "max_error": 0.7012252807617188, | |
| "sparsity": 0.3888014753659566, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.v_proj": { | |
| "mse": 0.000987939303740859, | |
| "rmse": 0.031431501773552896, | |
| "relative_error": 0.22352992716553835, | |
| "max_error": 0.8485660552978516, | |
| "sparsity": 0.3943232794602712, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.self_attn.o_proj": { | |
| "mse": 0.0003785373701248318, | |
| "rmse": 0.019456036855558016, | |
| "relative_error": 0.14437396293336985, | |
| "max_error": 0.5561981201171875, | |
| "sparsity": 0.3811718324820201, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.mlp.gate_proj": { | |
| "mse": 0.0004155401838943362, | |
| "rmse": 0.02038480276809997, | |
| "relative_error": 0.14142635788829938, | |
| "max_error": 1.0764923095703125, | |
| "sparsity": 0.38669275244077045, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.mlp.up_proj": { | |
| "mse": 0.0005035310750827193, | |
| "rmse": 0.022439498102290956, | |
| "relative_error": 0.1477698899567096, | |
| "max_error": 1.2791112661361694, | |
| "sparsity": 0.38763804237047833, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.18.mlp.down_proj": { | |
| "mse": 0.0007883885409682989, | |
| "rmse": 0.028078257441805376, | |
| "relative_error": 0.19378469242506685, | |
| "max_error": 0.7373738288879395, | |
| "sparsity": 0.3672477801640828, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.q_proj": { | |
| "mse": 0.00085911073256284, | |
| "rmse": 0.029310590791774224, | |
| "relative_error": 0.2099378337547514, | |
| "max_error": 1.4614830017089844, | |
| "sparsity": 0.3847314715385437, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.k_proj": { | |
| "mse": 0.0005041738622821867, | |
| "rmse": 0.02245381620754447, | |
| "relative_error": 0.16828826446926123, | |
| "max_error": 0.4177018404006958, | |
| "sparsity": 0.38169729709625244, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.v_proj": { | |
| "mse": 0.0009285964188165963, | |
| "rmse": 0.030472880054510702, | |
| "relative_error": 0.17457744249872342, | |
| "max_error": 0.595149040222168, | |
| "sparsity": 0.3823392987251282, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.self_attn.o_proj": { | |
| "mse": 0.0005175211699679494, | |
| "rmse": 0.02274909162951236, | |
| "relative_error": 0.14462214893241224, | |
| "max_error": 0.8733596801757812, | |
| "sparsity": 0.37877262632052106, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.mlp.gate_proj": { | |
| "mse": 0.0004557396750897169, | |
| "rmse": 0.021348060218430078, | |
| "relative_error": 0.1448868135770436, | |
| "max_error": 1.4225969314575195, | |
| "sparsity": 0.38894954323768616, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.mlp.up_proj": { | |
| "mse": 0.0005315069574862719, | |
| "rmse": 0.02305443465987123, | |
| "relative_error": 0.15028917582618634, | |
| "max_error": 0.7982025146484375, | |
| "sparsity": 0.38891713817914325, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.19.mlp.down_proj": { | |
| "mse": 0.000709193293005228, | |
| "rmse": 0.026630683299630673, | |
| "relative_error": 0.18214488847515778, | |
| "max_error": 1.5100746154785156, | |
| "sparsity": 0.3719211121400197, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.q_proj": { | |
| "mse": 0.000490160018671304, | |
| "rmse": 0.02213955777948837, | |
| "relative_error": 0.15512064902503814, | |
| "max_error": 0.6261138916015625, | |
| "sparsity": 0.3857632875442505, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.k_proj": { | |
| "mse": 0.0004383391060400754, | |
| "rmse": 0.020936549525651917, | |
| "relative_error": 0.15853934508516257, | |
| "max_error": 0.7386703491210938, | |
| "sparsity": 0.3873758216698964, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.v_proj": { | |
| "mse": 0.0013027676614001393, | |
| "rmse": 0.0360938729066325, | |
| "relative_error": 0.21833712499698749, | |
| "max_error": 1.1059064865112305, | |
| "sparsity": 0.3972831070423126, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.self_attn.o_proj": { | |
| "mse": 0.00040057982550933957, | |
| "rmse": 0.020014490388449553, | |
| "relative_error": 0.13684108508923026, | |
| "max_error": 0.46060943603515625, | |
| "sparsity": 0.3773680627346039, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.mlp.gate_proj": { | |
| "mse": 0.00039384985575452447, | |
| "rmse": 0.019845650801989953, | |
| "relative_error": 0.1389769276223436, | |
| "max_error": 0.7752799987792969, | |
| "sparsity": 0.389324426651001, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.mlp.up_proj": { | |
| "mse": 0.0005102139548398554, | |
| "rmse": 0.022587916124331954, | |
| "relative_error": 0.1450008967710386, | |
| "max_error": 0.770294189453125, | |
| "sparsity": 0.3897652328014374, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.20.mlp.down_proj": { | |
| "mse": 0.0010023219510912895, | |
| "rmse": 0.03165946858510562, | |
| "relative_error": 0.21241143531363935, | |
| "max_error": 1.1079025268554688, | |
| "sparsity": 0.35979583859443665, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.q_proj": { | |
| "mse": 0.0005578782875090837, | |
| "rmse": 0.02361944723123477, | |
| "relative_error": 0.16805025021029119, | |
| "max_error": 0.6688270568847656, | |
| "sparsity": 0.38523581624031067, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.k_proj": { | |
| "mse": 0.00043896384886465967, | |
| "rmse": 0.02095146412221971, | |
| "relative_error": 0.15661089496783753, | |
| "max_error": 0.5441436767578125, | |
| "sparsity": 0.37966881195704144, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.v_proj": { | |
| "mse": 0.0008802193915471435, | |
| "rmse": 0.029668491561708078, | |
| "relative_error": 0.1836733334029918, | |
| "max_error": 0.6075363159179688, | |
| "sparsity": 0.3894856870174408, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.self_attn.o_proj": { | |
| "mse": 0.0007023724028840661, | |
| "rmse": 0.026502309387750837, | |
| "relative_error": 0.1819082329396076, | |
| "max_error": 0.8114643096923828, | |
| "sparsity": 0.36885730425516766, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.mlp.gate_proj": { | |
| "mse": 0.0003605367091950029, | |
| "rmse": 0.01898780422257937, | |
| "relative_error": 0.13205976034712932, | |
| "max_error": 0.982025146484375, | |
| "sparsity": 0.3887227276961009, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.mlp.up_proj": { | |
| "mse": 0.0004587272123899311, | |
| "rmse": 0.021417918021832353, | |
| "relative_error": 0.13750663410174493, | |
| "max_error": 0.5613021850585938, | |
| "sparsity": 0.3892076313495636, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.21.mlp.down_proj": { | |
| "mse": 0.0006418560515157878, | |
| "rmse": 0.0253348781626395, | |
| "relative_error": 0.16878117065872833, | |
| "max_error": 0.8363590240478516, | |
| "sparsity": 0.37388818462689716, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.q_proj": { | |
| "mse": 0.0005556874093599617, | |
| "rmse": 0.023573022915187643, | |
| "relative_error": 0.16564819605910688, | |
| "max_error": 0.8030319213867188, | |
| "sparsity": 0.3844240407148997, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.k_proj": { | |
| "mse": 0.0004220264090690762, | |
| "rmse": 0.02054328136080203, | |
| "relative_error": 0.15166877254433841, | |
| "max_error": 0.5363616943359375, | |
| "sparsity": 0.38093774517377216, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.v_proj": { | |
| "mse": 0.0009258572245016694, | |
| "rmse": 0.03042790207197449, | |
| "relative_error": 0.17516537989649306, | |
| "max_error": 0.5467529296875, | |
| "sparsity": 0.3875174820423126, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.self_attn.o_proj": { | |
| "mse": 0.00047359790187329054, | |
| "rmse": 0.021762304608503452, | |
| "relative_error": 0.14042370182939093, | |
| "max_error": 0.5455656051635742, | |
| "sparsity": 0.3834725817044576, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.mlp.gate_proj": { | |
| "mse": 0.00036050137714482844, | |
| "rmse": 0.018986873811789776, | |
| "relative_error": 0.13319807744372072, | |
| "max_error": 0.7116546630859375, | |
| "sparsity": 0.38889604806900024, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.mlp.up_proj": { | |
| "mse": 0.00046704590204171836, | |
| "rmse": 0.021611244805464547, | |
| "relative_error": 0.13599457121265152, | |
| "max_error": 0.8751449584960938, | |
| "sparsity": 0.38965070247650146, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.22.mlp.down_proj": { | |
| "mse": 0.0008220822783187032, | |
| "rmse": 0.02867197723071611, | |
| "relative_error": 0.18627136659083046, | |
| "max_error": 0.8769989013671875, | |
| "sparsity": 0.3663081427415212, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.q_proj": { | |
| "mse": 0.0007962422678247094, | |
| "rmse": 0.028217765110382313, | |
| "relative_error": 0.2095703072026533, | |
| "max_error": 0.6340522766113281, | |
| "sparsity": 0.3846842447916667, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.k_proj": { | |
| "mse": 0.0006293078185990453, | |
| "rmse": 0.02508600842300435, | |
| "relative_error": 0.1930527722567741, | |
| "max_error": 0.5530929565429688, | |
| "sparsity": 0.37853853901227313, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.v_proj": { | |
| "mse": 0.0005694902502000332, | |
| "rmse": 0.02386399484998338, | |
| "relative_error": 0.11862559606961656, | |
| "max_error": 0.44977378845214844, | |
| "sparsity": 0.4073410928249359, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.self_attn.o_proj": { | |
| "mse": 0.00047892300062812865, | |
| "rmse": 0.021884309461989625, | |
| "relative_error": 0.11686789647284099, | |
| "max_error": 0.7507171630859375, | |
| "sparsity": 0.3899056414763133, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.mlp.gate_proj": { | |
| "mse": 0.0003371600469108671, | |
| "rmse": 0.018361918388634317, | |
| "relative_error": 0.1304621429587916, | |
| "max_error": 0.684112548828125, | |
| "sparsity": 0.3855345944563548, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.mlp.up_proj": { | |
| "mse": 0.0004635937511920929, | |
| "rmse": 0.021531227349877037, | |
| "relative_error": 0.1341410726411575, | |
| "max_error": 0.530548095703125, | |
| "sparsity": 0.38697569568951923, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.23.mlp.down_proj": { | |
| "mse": 0.0006501522730104625, | |
| "rmse": 0.02549808371251578, | |
| "relative_error": 0.16533883590116874, | |
| "max_error": 0.6298317909240723, | |
| "sparsity": 0.3731357951958974, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.q_proj": { | |
| "mse": 0.0007914926391094923, | |
| "rmse": 0.02813347897273802, | |
| "relative_error": 0.21282701567599138, | |
| "max_error": 0.6182622909545898, | |
| "sparsity": 0.3789725601673126, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.k_proj": { | |
| "mse": 0.0004922481020912528, | |
| "rmse": 0.02218666496099071, | |
| "relative_error": 0.19124297990888794, | |
| "max_error": 0.48739147186279297, | |
| "sparsity": 0.3734628160794576, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.v_proj": { | |
| "mse": 0.0007634047651663423, | |
| "rmse": 0.02762978040387477, | |
| "relative_error": 0.13271880790172866, | |
| "max_error": 0.7925567626953125, | |
| "sparsity": 0.39220739404360455, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.self_attn.o_proj": { | |
| "mse": 0.0007033598376438022, | |
| "rmse": 0.026520932065894707, | |
| "relative_error": 0.16131433471100196, | |
| "max_error": 0.6284704208374023, | |
| "sparsity": 0.38041027386983234, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.mlp.gate_proj": { | |
| "mse": 0.00032479356741532683, | |
| "rmse": 0.018022030058107406, | |
| "relative_error": 0.12703308340892464, | |
| "max_error": 0.6070327758789062, | |
| "sparsity": 0.3842460811138153, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.mlp.up_proj": { | |
| "mse": 0.00045559569844044745, | |
| "rmse": 0.021344687827195962, | |
| "relative_error": 0.13184735868206654, | |
| "max_error": 0.5149078369140625, | |
| "sparsity": 0.38552404443422955, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.24.mlp.down_proj": { | |
| "mse": 0.0006079337326809764, | |
| "rmse": 0.024656312227926067, | |
| "relative_error": 0.1573348187876713, | |
| "max_error": 0.6859645843505859, | |
| "sparsity": 0.3744344810644786, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.q_proj": { | |
| "mse": 0.0004431141715031117, | |
| "rmse": 0.021050277231027426, | |
| "relative_error": 0.1508082002071661, | |
| "max_error": 0.4593334197998047, | |
| "sparsity": 0.3852458695570628, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.k_proj": { | |
| "mse": 0.00032679151627235115, | |
| "rmse": 0.01807737581266571, | |
| "relative_error": 0.13846183201116602, | |
| "max_error": 0.37456512451171875, | |
| "sparsity": 0.38373180230458576, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.v_proj": { | |
| "mse": 0.0007395596476271749, | |
| "rmse": 0.02719484597542657, | |
| "relative_error": 0.14064360022779163, | |
| "max_error": 0.5509586334228516, | |
| "sparsity": 0.38878942529360455, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.self_attn.o_proj": { | |
| "mse": 0.0010019183391705155, | |
| "rmse": 0.03165309367456071, | |
| "relative_error": 0.19080784602830844, | |
| "max_error": 0.6958675384521484, | |
| "sparsity": 0.3678787350654602, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.mlp.gate_proj": { | |
| "mse": 0.0002824858238454908, | |
| "rmse": 0.016807314593518228, | |
| "relative_error": 0.11826062574595378, | |
| "max_error": 0.5374832153320312, | |
| "sparsity": 0.38684646288553876, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.mlp.up_proj": { | |
| "mse": 0.00038697797572240233, | |
| "rmse": 0.019671755786467113, | |
| "relative_error": 0.12197996063452735, | |
| "max_error": 0.4824066162109375, | |
| "sparsity": 0.38694029053052265, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.25.mlp.down_proj": { | |
| "mse": 0.0005231230752542615, | |
| "rmse": 0.022871883946327234, | |
| "relative_error": 0.14492867635958706, | |
| "max_error": 0.9400634765625, | |
| "sparsity": 0.3764494061470032, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.q_proj": { | |
| "mse": 0.00035285300691612065, | |
| "rmse": 0.018784381994521957, | |
| "relative_error": 0.14592313439797372, | |
| "max_error": 0.4511146545410156, | |
| "sparsity": 0.3857472042242686, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.k_proj": { | |
| "mse": 0.0002649167727213353, | |
| "rmse": 0.01627626408981297, | |
| "relative_error": 0.13852624184120496, | |
| "max_error": 0.4108009338378906, | |
| "sparsity": 0.3841206133365631, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.v_proj": { | |
| "mse": 0.0008469641907140613, | |
| "rmse": 0.0291026492043948, | |
| "relative_error": 0.13792327436458263, | |
| "max_error": 0.5840492248535156, | |
| "sparsity": 0.39561934272448224, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.self_attn.o_proj": { | |
| "mse": 0.00047108199214562774, | |
| "rmse": 0.021704423331331053, | |
| "relative_error": 0.12231351329083083, | |
| "max_error": 0.4640541076660156, | |
| "sparsity": 0.38412563999493915, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.mlp.gate_proj": { | |
| "mse": 0.00027475401293486357, | |
| "rmse": 0.01657570550338246, | |
| "relative_error": 0.11745689266006713, | |
| "max_error": 0.6743698120117188, | |
| "sparsity": 0.3876640399297078, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.mlp.up_proj": { | |
| "mse": 0.0003821777645498514, | |
| "rmse": 0.01954936736955576, | |
| "relative_error": 0.12075877151242563, | |
| "max_error": 0.472869873046875, | |
| "sparsity": 0.3872906764348348, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.26.mlp.down_proj": { | |
| "mse": 0.0005594809772446752, | |
| "rmse": 0.023653350232993954, | |
| "relative_error": 0.1500328444037981, | |
| "max_error": 0.8460102081298828, | |
| "sparsity": 0.3750576476256053, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.q_proj": { | |
| "mse": 0.0003002102894242853, | |
| "rmse": 0.017326577545040026, | |
| "relative_error": 0.13426799982484489, | |
| "max_error": 0.3865089416503906, | |
| "sparsity": 0.39018192887306213, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.k_proj": { | |
| "mse": 0.00022559559147339314, | |
| "rmse": 0.015019839928354534, | |
| "relative_error": 0.13020156777110098, | |
| "max_error": 0.3020808696746826, | |
| "sparsity": 0.3891661862532298, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.v_proj": { | |
| "mse": 0.0007057865732349455, | |
| "rmse": 0.02656664399646567, | |
| "relative_error": 0.13515331971447941, | |
| "max_error": 0.7693214416503906, | |
| "sparsity": 0.39161362250645954, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.self_attn.o_proj": { | |
| "mse": 0.00045776620390824974, | |
| "rmse": 0.02139547157480409, | |
| "relative_error": 0.13572113165922312, | |
| "max_error": 0.78155517578125, | |
| "sparsity": 0.38155863682429, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.mlp.gate_proj": { | |
| "mse": 0.00027885756571777165, | |
| "rmse": 0.016699028885470306, | |
| "relative_error": 0.11946844800261402, | |
| "max_error": 0.822509765625, | |
| "sparsity": 0.38759133219718933, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.mlp.up_proj": { | |
| "mse": 0.0004284470924176276, | |
| "rmse": 0.02069896355901975, | |
| "relative_error": 0.1247453787481905, | |
| "max_error": 2.0409393310546875, | |
| "sparsity": 0.3873121440410614, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.27.mlp.down_proj": { | |
| "mse": 0.0006952874828130007, | |
| "rmse": 0.02636830451153431, | |
| "relative_error": 0.16718845457586556, | |
| "max_error": 0.791473388671875, | |
| "sparsity": 0.36890024940172833, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.q_proj": { | |
| "mse": 0.00021549634402617812, | |
| "rmse": 0.014679793732412528, | |
| "relative_error": 0.12561023288607884, | |
| "max_error": 0.3762359619140625, | |
| "sparsity": 0.3898433446884155, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.k_proj": { | |
| "mse": 0.0001749363000271842, | |
| "rmse": 0.013226348703522987, | |
| "relative_error": 0.12099917059147738, | |
| "max_error": 0.2971935272216797, | |
| "sparsity": 0.3881022135416667, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.v_proj": { | |
| "mse": 0.0008044744608923793, | |
| "rmse": 0.028363258996320913, | |
| "relative_error": 0.12933421533858702, | |
| "max_error": 0.5359439849853516, | |
| "sparsity": 0.3934883773326874, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.self_attn.o_proj": { | |
| "mse": 0.0006236206972971559, | |
| "rmse": 0.02497239870931817, | |
| "relative_error": 0.14459681208460812, | |
| "max_error": 0.5996036529541016, | |
| "sparsity": 0.3753486176331838, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.mlp.gate_proj": { | |
| "mse": 0.0002845228591468185, | |
| "rmse": 0.01686780540398835, | |
| "relative_error": 0.12138743948161114, | |
| "max_error": 0.8363265991210938, | |
| "sparsity": 0.3879963457584381, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.mlp.up_proj": { | |
| "mse": 0.0004395138821564615, | |
| "rmse": 0.020964586381716704, | |
| "relative_error": 0.12384208036720651, | |
| "max_error": 1.02935791015625, | |
| "sparsity": 0.3881142735481262, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.28.mlp.down_proj": { | |
| "mse": 0.0006876399274915457, | |
| "rmse": 0.026222889381064506, | |
| "relative_error": 0.16633863256359221, | |
| "max_error": 0.6464881896972656, | |
| "sparsity": 0.37210987011591595, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.q_proj": { | |
| "mse": 0.00020917913934681565, | |
| "rmse": 0.014463026631615377, | |
| "relative_error": 0.12511935357864523, | |
| "max_error": 0.39916908740997314, | |
| "sparsity": 0.3913885752360026, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.k_proj": { | |
| "mse": 0.000153984350617975, | |
| "rmse": 0.012409043098401061, | |
| "relative_error": 0.11980498352536048, | |
| "max_error": 0.3233306407928467, | |
| "sparsity": 0.3890305459499359, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.v_proj": { | |
| "mse": 0.0009475867846049368, | |
| "rmse": 0.03078289759923417, | |
| "relative_error": 0.13662062360874777, | |
| "max_error": 0.6898956298828125, | |
| "sparsity": 0.3957308530807495, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.self_attn.o_proj": { | |
| "mse": 0.0005539952544495463, | |
| "rmse": 0.023537103782104253, | |
| "relative_error": 0.13449006927473472, | |
| "max_error": 1.0374298095703125, | |
| "sparsity": 0.37956833839416504, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.mlp.gate_proj": { | |
| "mse": 0.00031886447686702013, | |
| "rmse": 0.017856776777095584, | |
| "relative_error": 0.13113133117491768, | |
| "max_error": 0.5156631469726562, | |
| "sparsity": 0.38739465673764545, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.mlp.up_proj": { | |
| "mse": 0.0005141778383404016, | |
| "rmse": 0.022675489814784632, | |
| "relative_error": 0.13509925356520125, | |
| "max_error": 0.6751518249511719, | |
| "sparsity": 0.38900267084439594, | |
| "effective_bits": 9.0, | |
| "sparse_nnz": 0, | |
| "n_planes": 3, | |
| "rescued_rows": 0 | |
| }, | |
| "model.text_model.layers.29.mlp.down_proj": { | |
| "mse": 0.0019380510784685612, | |
| "rmse": 0.04402330153985002, | |
| "relative_error": 0.30400376003737134, | |
| "max_error": 2.2239151000976562, | |
| "sparsity": 0.3337251643339793, | |
| "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": 211, | |
| "quantized_params": 113246208, | |
| "quantized_fraction": 0.4415316287123117, | |
| "avg_relative_error": 0.1487850632142753, | |
| "avg_effective_bits": 9.0, | |
| "full_model_effective_bits": 12.909278599013819, | |
| "compression_ratio": 1.2394185993648237 | |
| }, | |
| "method_name": "Broad-tritplane3", | |
| "model_family": "image_text_to_text", | |
| "format_family": "tritplane_small", | |
| "format_version": "1.0", | |
| "total_packed_bytes": 127401984, | |
| "total_fp16_bytes": 226492416, | |
| "compression_ratio": 1.7777777777777777 | |
| } |