adapted model files for PR#2 'adjust qllama call to output hidden sta…

…tes'

README.md

@@ -1,4 +1,4 @@
-# GBA Model Toolkit for M LX
+# GBA Model Toolkit for MLX
 
 ## Introduction
 Welcome to the GreenBitAI (GBA) Model Toolkit for [MLX](https://github.com/ml-explore/mlx)! This comprehensive Python package not only facilitates the conversion of [GreenBitAI's Low-bit Language Models (LLMs)](https://huggingface.co/collections/GreenBitAI/greenbitai-mlx-llm-6614eb6ceb8da657c2b4ed58) to MLX framework compatible format but also supports generation, model loading, and other essential scripts tailored for GBA quantized models. Designed to enhance the integration and deployment of GBA models within the MLX ecosystem, this toolkit enables the efficient execution of GBA models on a variety of platforms, with special optimizations for Apple devices to enable local inference and natural language content generation. 
@@ -54,7 +54,7 @@ A high-performance HTTP API for text generation with GreenBitAI's mlx models. Im
 #### Quick Start
 1. Run:
    ```shell
-   python -m gbx_lm.fastapi_server.py --model GreenBitAI/Llama-3-8B-instruct-layer-mix-bpw-4.0-mlx
+   python -m gbx_lm.fastapi_server --model GreenBitAI/Llama-3-8B-instruct-layer-mix-bpw-4.0-mlx
    ```
 2. Use:
    ```shell

gbx_lm/models/qgemma.py

@@ -167,9 +167,10 @@ def __call__(
         self,
         inputs: mx.array,
         cache=None,
+        hidden_states=False
     ):
         out = self.model(inputs, cache)
-        out = self.model.embed_tokens.as_linear(out)
+        out = (self.model.embed_tokens.as_linear(out), out) if hidden_states else self.model.embed_tokens.as_linear(out)
         return out
 
     @property

gbx_lm/models/qllama.py

@@ -288,12 +288,14 @@ def __call__(
         self,
         inputs: mx.array,
         cache=None,
+        hidden_states=False
     ):
         out = self.model(inputs, cache)
         if self.args.tie_word_embeddings:
-            out = self.model.embed_tokens.as_linear(out)
+            out = (self.model.embed_tokens.as_linear(out), out) if hidden_states else self.model.embed_tokens.as_linear(
+                out)
         else:
-            out = self.lm_head(out)
+            out = (self.lm_head(out), out) if hidden_states else self.lm_head(out)
         return out
 
     def sanitize(self, weights):

gbx_lm/models/qmixtral.py

@@ -190,9 +190,11 @@ def __call__(
         self,
         inputs: mx.array,
         cache=None,
+        hidden_states=False
     ):
         out = self.model(inputs, cache)
-        return self.lm_head(out)
+        out = (self.lm_head(out), out) if hidden_states else self.lm_head(out)
+        return out
 
     def sanitize(self, weights):
         if "model.layers.0.block_sparse_moe.experts.0.w1.weight" not in weights:

gbx_lm/models/qphi3.py

@@ -197,9 +197,11 @@ def __call__(
         self,
         inputs: mx.array,
         cache=None,
+        hidden_states=False
     ):
         out = self.model(inputs, cache)
-        return self.lm_head(out)
+        out = (self.lm_head(out), out) if hidden_states else self.lm_head(out)
+        return out
 
     @property
     def layers(self):

gbx_lm/models/qqwen2.py

@@ -179,12 +179,14 @@ def __call__(
         self,
         inputs: mx.array,
         cache=None,
+        hidden_states=False
     ):
         out = self.model(inputs, cache)
         if self.args.tie_word_embeddings:
-            out = self.model.embed_tokens.as_linear(out)
+            out = (self.model.embed_tokens.as_linear(out), out) if hidden_states else self.model.embed_tokens.as_linear(
+                out)
         else:
-            out = self.lm_head(out)
+            out = (self.lm_head(out), out) if hidden_states else self.lm_head(out)
         return out
 
     def sanitize(self, weights):

gbx_lm/models/qstarcoder2.py

@@ -154,12 +154,14 @@ def __call__(
         self,
         inputs: mx.array,
         cache=None,
+        hidden_states=False
     ):
         out = self.model(inputs, cache)
         if self.args.tie_word_embeddings:
-            out = self.model.embed_tokens.as_linear(out)
+            out = (self.model.embed_tokens.as_linear(out), out) if hidden_states else self.model.embed_tokens.as_linear(
+                out)
         else:
-            out = self.lm_head(out)
+            out = (self.lm_head(out), out) if hidden_states else self.lm_head(out)
         return out
 
     @property