Instructions to use zeromodels/gpt-oss-120b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- KerasFormers
How to use zeromodels/gpt-oss-120b with KerasFormers:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Keras
How to use zeromodels/gpt-oss-120b with Keras:
# Available backend options are: "jax", "torch", "tensorflow". import os os.environ["KERAS_BACKEND"] = "jax" import keras model = keras.saving.load_model("hf://zeromodels/gpt-oss-120b") - Notebooks
- Google Colab
- Kaggle
| { | |
| "metadata": { | |
| "total_size": 65248815744.0 | |
| }, | |
| "weight_map": { | |
| "/layers/gpt_oss_decoder_layer_25/post_attention_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/input_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/mlp/experts/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/mlp/router/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/post_attention_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/self_attn/k_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/self_attn/o_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/self_attn/q_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/self_attn/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_26/self_attn/v_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/input_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/mlp/router/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/self_attn/k_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/self_attn/o_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/self_attn/q_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/self_attn/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/self_attn/v_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/mlp/experts/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_27/post_attention_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/input_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/mlp/experts/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/mlp/router/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/post_attention_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/self_attn/k_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/self_attn/o_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/self_attn/q_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/self_attn/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_28/self_attn/v_proj/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/input_layernorm/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/mlp/experts/vars": [ | |
| "model_00000.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/mlp/router/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/post_attention_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/self_attn/k_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/self_attn/o_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/self_attn/q_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/self_attn/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_29/self_attn/v_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/input_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/mlp/router/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/self_attn/k_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/self_attn/o_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/self_attn/q_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/self_attn/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/self_attn/v_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/post_attention_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/input_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/mlp/router/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/self_attn/k_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/self_attn/o_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/self_attn/q_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/self_attn/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/self_attn/v_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/mlp/experts/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_3/post_attention_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/input_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/mlp/router/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/self_attn/k_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/self_attn/o_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/self_attn/q_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/self_attn/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/self_attn/v_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/mlp/experts/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_20/post_attention_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/input_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/mlp/experts/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/mlp/router/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/post_attention_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/self_attn/k_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/self_attn/o_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/self_attn/q_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/self_attn/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_21/self_attn/v_proj/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/input_layernorm/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/mlp/experts/vars": [ | |
| "model_00001.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/mlp/router/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/post_attention_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/self_attn/k_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/self_attn/o_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/self_attn/q_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/self_attn/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_22/self_attn/v_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/input_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/mlp/router/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/self_attn/k_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/self_attn/o_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/self_attn/q_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/self_attn/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/self_attn/v_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/mlp/experts/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_23/post_attention_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/input_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/mlp/experts/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/mlp/router/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/post_attention_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/self_attn/k_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/self_attn/o_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/self_attn/q_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/self_attn/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_24/self_attn/v_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/input_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/mlp/experts/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/mlp/router/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/self_attn/k_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/self_attn/o_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/self_attn/q_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/self_attn/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_25/self_attn/v_proj/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/post_attention_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/input_layernorm/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/mlp/experts/vars": [ | |
| "model_00002.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/mlp/router/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/post_attention_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/self_attn/k_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/self_attn/o_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/self_attn/q_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/self_attn/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_5/self_attn/v_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/input_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/mlp/experts/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/mlp/router/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/post_attention_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/self_attn/k_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/self_attn/o_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/self_attn/q_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/self_attn/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_6/self_attn/v_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/input_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/mlp/router/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/self_attn/k_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/self_attn/o_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/self_attn/q_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/self_attn/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/self_attn/v_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/mlp/experts/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_7/post_attention_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/input_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/mlp/experts/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/mlp/router/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/post_attention_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/self_attn/k_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/self_attn/o_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/self_attn/q_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/self_attn/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_8/self_attn/v_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/input_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/mlp/router/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/self_attn/k_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/self_attn/o_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/self_attn/q_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/self_attn/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/self_attn/v_proj/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/post_attention_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/input_layernorm/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/mlp/experts/vars": [ | |
| "model_00003.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/mlp/router/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/post_attention_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/self_attn/k_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/self_attn/o_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/self_attn/q_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/self_attn/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_15/self_attn/v_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/input_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/mlp/router/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/self_attn/k_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/self_attn/o_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/self_attn/q_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/self_attn/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/self_attn/v_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/mlp/experts/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_9/post_attention_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/mlp/experts/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_16/post_attention_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/input_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/mlp/experts/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/mlp/router/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/post_attention_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/self_attn/k_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/self_attn/o_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/self_attn/q_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/self_attn/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_17/self_attn/v_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/input_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/mlp/router/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/self_attn/k_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/self_attn/o_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/self_attn/q_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/self_attn/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/self_attn/v_proj/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/input_layernorm/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/mlp/experts/vars": [ | |
| "model_00004.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/mlp/router/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/post_attention_layernorm/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/self_attn/k_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/self_attn/o_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/self_attn/q_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/self_attn/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer/self_attn/v_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/input_layernorm/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/mlp/router/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/self_attn/k_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/self_attn/o_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/self_attn/q_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/self_attn/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/self_attn/v_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/mlp/experts/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_18/post_attention_layernorm/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/input_layernorm/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/mlp/experts/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/mlp/router/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/post_attention_layernorm/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/self_attn/k_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/self_attn/o_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/self_attn/q_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/self_attn/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_19/self_attn/v_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/input_layernorm/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/mlp/experts/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/mlp/router/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/self_attn/k_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/self_attn/o_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/self_attn/q_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/self_attn/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_2/self_attn/v_proj/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/mlp/experts/vars": [ | |
| "model_00005.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_1/post_attention_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/input_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/mlp/experts/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/mlp/router/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/post_attention_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/self_attn/k_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/self_attn/o_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/self_attn/q_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/self_attn/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_10/self_attn/v_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/input_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/mlp/router/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/self_attn/k_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/self_attn/o_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/self_attn/q_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/self_attn/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/self_attn/v_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/mlp/experts/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_11/post_attention_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/input_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/mlp/experts/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/mlp/router/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/post_attention_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/self_attn/k_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/self_attn/o_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/self_attn/q_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/self_attn/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_12/self_attn/v_proj/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/input_layernorm/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/mlp/experts/vars": [ | |
| "model_00006.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/mlp/router/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/post_attention_layernorm/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/self_attn/k_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/self_attn/o_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/self_attn/q_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/self_attn/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_13/self_attn/v_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/input_layernorm/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/mlp/experts/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/mlp/router/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/self_attn/k_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/self_attn/o_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/self_attn/q_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/self_attn/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_14/self_attn/v_proj/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/dense/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/embedding/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/mlp/experts/vars": [ | |
| "model_00007.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_30/post_attention_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/input_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/mlp/router/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/self_attn/k_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/self_attn/o_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/self_attn/q_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/self_attn/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/self_attn/v_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_rms_norm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/mlp/experts/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_31/post_attention_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/input_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/mlp/experts/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/mlp/router/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/post_attention_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/self_attn/k_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/self_attn/o_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/self_attn/q_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/self_attn/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_32/self_attn/v_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/input_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/mlp/router/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/self_attn/k_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/self_attn/o_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/self_attn/q_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/self_attn/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/self_attn/v_proj/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/mlp/experts/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_33/post_attention_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/input_layernorm/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/mlp/experts/vars": [ | |
| "model_00008.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/mlp/router/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/post_attention_layernorm/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/self_attn/k_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/self_attn/o_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/self_attn/q_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/self_attn/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_34/self_attn/v_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/input_layernorm/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/mlp/experts/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/mlp/router/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/post_attention_layernorm/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/self_attn/k_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/self_attn/o_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/self_attn/q_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/self_attn/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_35/self_attn/v_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/input_layernorm/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/mlp/experts/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/mlp/router/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/self_attn/k_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/self_attn/o_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/self_attn/q_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/self_attn/vars": [ | |
| "model_00009.weights.h5" | |
| ], | |
| "/layers/gpt_oss_decoder_layer_4/self_attn/v_proj/vars": [ | |
| "model_00009.weights.h5" | |
| ] | |
| } | |
| } |