Upload 6 files
Browse files
    	
        text_encoder_2/config.json
    ADDED
    
    | @@ -0,0 +1,32 @@ | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | 
|  | |
| 1 | 
            +
            {
         | 
| 2 | 
            +
              "_name_or_path": "/mnt/DataSSD/AI/models/sd-webui/Diffusers/models--black-forest-labs--FLUX.1-dev/snapshots/01aa605f2c300568dd6515476f04565a954fcb59/text_encoder_2",
         | 
| 3 | 
            +
              "architectures": [
         | 
| 4 | 
            +
                "T5EncoderModel"
         | 
| 5 | 
            +
              ],
         | 
| 6 | 
            +
              "classifier_dropout": 0.0,
         | 
| 7 | 
            +
              "d_ff": 10240,
         | 
| 8 | 
            +
              "d_kv": 64,
         | 
| 9 | 
            +
              "d_model": 4096,
         | 
| 10 | 
            +
              "decoder_start_token_id": 0,
         | 
| 11 | 
            +
              "dense_act_fn": "gelu_new",
         | 
| 12 | 
            +
              "dropout_rate": 0.1,
         | 
| 13 | 
            +
              "eos_token_id": 1,
         | 
| 14 | 
            +
              "feed_forward_proj": "gated-gelu",
         | 
| 15 | 
            +
              "initializer_factor": 1.0,
         | 
| 16 | 
            +
              "is_encoder_decoder": true,
         | 
| 17 | 
            +
              "is_gated_act": true,
         | 
| 18 | 
            +
              "layer_norm_epsilon": 1e-06,
         | 
| 19 | 
            +
              "model_type": "t5",
         | 
| 20 | 
            +
              "num_decoder_layers": 24,
         | 
| 21 | 
            +
              "num_heads": 64,
         | 
| 22 | 
            +
              "num_layers": 24,
         | 
| 23 | 
            +
              "output_past": true,
         | 
| 24 | 
            +
              "pad_token_id": 0,
         | 
| 25 | 
            +
              "relative_attention_max_distance": 128,
         | 
| 26 | 
            +
              "relative_attention_num_buckets": 32,
         | 
| 27 | 
            +
              "tie_word_embeddings": false,
         | 
| 28 | 
            +
              "torch_dtype": "bfloat16",
         | 
| 29 | 
            +
              "transformers_version": "4.41.2",
         | 
| 30 | 
            +
              "use_cache": true,
         | 
| 31 | 
            +
              "vocab_size": 32128
         | 
| 32 | 
            +
            }
         | 
    	
        text_encoder_2/model.safetensors
    ADDED
    
    | @@ -0,0 +1,3 @@ | |
|  | |
|  | |
|  | 
|  | |
| 1 | 
            +
            version https://git-lfs.github.com/spec/v1
         | 
| 2 | 
            +
            oid sha256:2328bbaf8115a861edeb4b8c6954149634d198aa82a062c72b315e5b89c70e42
         | 
| 3 | 
            +
            size 4896162200
         | 
    	
        text_encoder_2/quantization_map.json
    ADDED
    
    | @@ -0,0 +1 @@ | |
|  | 
|  | |
| 1 | 
            +
            {"encoder.block.0.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.0.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.0.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.0.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.0.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.0.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.0.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.1.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.2.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.3.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.4.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.5.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.6.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.7.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.8.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.9.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.10.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.11.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.12.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.13.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.14.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.15.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.16.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.17.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.18.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.19.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.20.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.21.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.22.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.0.SelfAttention.q": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.0.SelfAttention.k": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.0.SelfAttention.v": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.0.SelfAttention.o": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.1.DenseReluDense.wi_0": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.1.DenseReluDense.wi_1": {"weights": "qint8", "activations": "none"}, "encoder.block.23.layer.1.DenseReluDense.wo": {"weights": "qint8", "activations": "none"}}
         | 
    	
        transformer/config.json
    ADDED
    
    | @@ -0,0 +1,19 @@ | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | |
|  | 
|  | |
| 1 | 
            +
            {
         | 
| 2 | 
            +
              "_class_name": "FluxTransformer2DModel",
         | 
| 3 | 
            +
              "_diffusers_version": "0.30.0.dev0",
         | 
| 4 | 
            +
              "_name_or_path": "/mnt/DataSSD/AI/models/sd-webui/Diffusers/models--black-forest-labs--FLUX.1-dev/snapshots/01aa605f2c300568dd6515476f04565a954fcb59/transformer",
         | 
| 5 | 
            +
              "attention_head_dim": 128,
         | 
| 6 | 
            +
              "axes_dims_rope": [
         | 
| 7 | 
            +
                16,
         | 
| 8 | 
            +
                56,
         | 
| 9 | 
            +
                56
         | 
| 10 | 
            +
              ],
         | 
| 11 | 
            +
              "guidance_embeds": true,
         | 
| 12 | 
            +
              "in_channels": 64,
         | 
| 13 | 
            +
              "joint_attention_dim": 4096,
         | 
| 14 | 
            +
              "num_attention_heads": 24,
         | 
| 15 | 
            +
              "num_layers": 19,
         | 
| 16 | 
            +
              "num_single_layers": 38,
         | 
| 17 | 
            +
              "patch_size": 1,
         | 
| 18 | 
            +
              "pooled_projection_dim": 768
         | 
| 19 | 
            +
            }
         | 
    	
        transformer/diffusion_pytorch_model.safetensors
    ADDED
    
    | @@ -0,0 +1,3 @@ | |
|  | |
|  | |
|  | 
|  | |
| 1 | 
            +
            version https://git-lfs.github.com/spec/v1
         | 
| 2 | 
            +
            oid sha256:3301dec5f0c4017cc3a8d72dd6904d0e09b36c2f8c4ad8685851636d4fbe4024
         | 
| 3 | 
            +
            size 11910932232
         | 
    	
        transformer/quantization_map.json
    ADDED
    
    | @@ -0,0 +1 @@ | |
|  | 
|  | |
| 1 | 
            +
            {"time_text_embed.timestep_embedder.linear_1": {"weights": "qint8", "activations": "none"}, "time_text_embed.timestep_embedder.linear_2": {"weights": "qint8", "activations": "none"}, "time_text_embed.guidance_embedder.linear_1": {"weights": "qint8", "activations": "none"}, "time_text_embed.guidance_embedder.linear_2": {"weights": "qint8", "activations": "none"}, "time_text_embed.text_embedder.linear_1": {"weights": "qint8", "activations": "none"}, "time_text_embed.text_embedder.linear_2": {"weights": "qint8", "activations": "none"}, "context_embedder": {"weights": "qint8", "activations": "none"}, "x_embedder": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.0.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.1.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.2.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.3.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.4.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.5.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.6.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.7.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.8.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.9.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.10.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.11.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.12.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.13.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.14.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.15.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.16.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.17.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.norm1.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.norm1_context.linear": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.to_q": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.to_k": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.to_v": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.add_k_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.add_v_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.add_q_proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.to_out.0": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.attn.to_add_out": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.ff.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.ff.net.2": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.ff_context.net.0.proj": {"weights": "qint8", "activations": "none"}, "transformer_blocks.18.ff_context.net.2": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.0.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.0.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.0.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.0.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.0.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.0.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.1.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.1.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.1.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.1.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.1.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.1.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.2.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.2.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.2.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.2.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.2.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.2.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.3.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.3.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.3.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.3.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.3.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.3.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.4.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.4.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.4.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.4.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.4.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.4.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.5.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.5.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.5.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.5.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.5.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.5.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.6.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.6.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.6.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.6.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.6.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.6.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.7.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.7.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.7.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.7.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.7.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.7.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.8.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.8.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.8.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.8.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.8.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.8.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.9.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.9.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.9.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.9.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.9.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.9.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.10.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.10.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.10.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.10.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.10.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.10.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.11.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.11.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.11.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.11.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.11.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.11.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.12.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.12.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.12.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.12.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.12.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.12.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.13.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.13.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.13.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.13.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.13.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.13.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.14.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.14.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.14.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.14.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.14.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.14.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.15.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.15.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.15.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.15.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.15.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.15.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.16.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.16.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.16.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.16.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.16.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.16.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.17.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.17.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.17.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.17.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.17.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.17.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.18.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.18.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.18.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.18.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.18.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.18.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.19.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.19.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.19.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.19.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.19.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.19.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.20.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.20.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.20.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.20.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.20.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.20.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.21.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.21.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.21.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.21.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.21.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.21.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.22.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.22.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.22.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.22.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.22.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.22.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.23.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.23.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.23.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.23.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.23.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.23.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.24.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.24.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.24.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.24.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.24.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.24.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.25.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.25.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.25.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.25.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.25.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.25.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.26.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.26.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.26.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.26.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.26.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.26.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.27.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.27.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.27.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.27.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.27.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.27.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.28.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.28.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.28.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.28.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.28.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.28.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.29.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.29.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.29.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.29.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.29.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.29.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.30.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.30.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.30.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.30.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.30.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.30.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.31.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.31.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.31.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.31.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.31.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.31.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.32.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.32.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.32.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.32.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.32.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.32.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.33.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.33.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.33.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.33.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.33.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.33.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.34.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.34.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.34.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.34.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.34.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.34.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.35.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.35.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.35.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.35.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.35.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.35.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.36.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.36.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.36.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.36.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.36.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.36.attn.to_v": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.37.norm.linear": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.37.proj_mlp": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.37.proj_out": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.37.attn.to_q": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.37.attn.to_k": {"weights": "qint8", "activations": "none"}, "single_transformer_blocks.37.attn.to_v": {"weights": "qint8", "activations": "none"}, "norm_out.linear": {"weights": "qint8", "activations": "none"}, "proj_out": {"weights": "qint8", "activations": "none"}}
         | 
