Instructions to use keras/llama3.1_instruct_8b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- KerasHub
How to use keras/llama3.1_instruct_8b with KerasHub:
import keras_hub # Create a LlamaCausalLM model task = keras_hub.models.LlamaCausalLM.from_preset("hf://keras/llama3.1_instruct_8b")import keras_hub # Create a Backbone model unspecialized for any task backbone = keras_hub.models.Backbone.from_preset("hf://keras/llama3.1_instruct_8b") - Keras
How to use keras/llama3.1_instruct_8b with Keras:
# !pip install -U keras tensorflow huggingface_hub # Keras needs TensorFlow installed to read "hf://" paths, so the tensorflow backend is selected here; # "jax" and "torch" also work for computation once TensorFlow is installed. import os os.environ["KERAS_BACKEND"] = "tensorflow" import keras model = keras.saving.load_model("hf://keras/llama3.1_instruct_8b") - Notebooks
- Google Colab
- Kaggle
Download model.weights.json from keras/llama3.1_instruct_8b: direct link, hf CLI and curl.
- Browser
- Download file 31.5 kB
-
https://huggingface.co/keras/llama3.1_instruct_8b/resolve/main/model.weights.json
- Command line
-
hf download hf://keras/llama3.1_instruct_8b/model.weights.json
-
curl -L -o model.weights.json https://huggingface.co/keras/llama3.1_instruct_8b/resolve/main/model.weights.json
31.5 kB
| { | |
| "metadata": { | |
| "total_size": 16060522496.0 | |
| }, | |
| "weight_map": { | |
| "/layers/reversible_embedding/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_1/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_2/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_3/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_4/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_5/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_6/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_7/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_8/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_9/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_10/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_11/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_12/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_13/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_14/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_15/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_16/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_17/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_feedforward_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_self_attention_layer/_key_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_self_attention_layer/_output_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_self_attention_layer/_query_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_self_attention_layer/_value_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_18/_self_attention_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_feedforward_gate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_feedforward_intermediate_dense/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_feedforward_layernorm/vars": "model_00000.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_19/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_20/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_21/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_22/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_23/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_24/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_25/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_26/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_27/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_28/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_29/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_30/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_feedforward_gate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_feedforward_intermediate_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_feedforward_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_feedforward_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_self_attention_layer/_key_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_self_attention_layer/_output_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_self_attention_layer/_query_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_self_attention_layer/_value_dense/vars": "model_00001.weights.h5", | |
| "/layers/llama_transformer_decoder_31/_self_attention_layernorm/vars": "model_00001.weights.h5", | |
| "/layers/llama_layer_norm/vars": "model_00001.weights.h5" | |
| } | |
| } |