Training checkpoint 124000 - 1.3B tokens

README.md

@@ -18,9 +18,9 @@ pipeline_tag: text-generation
 
 ## Current Training Status
 
-- **Training Step**: 122,000
-- **Tokens Processed**: 1.25B tokens
-- **Current Loss**: 4.5518
+- **Training Step**: 124,000
+- **Tokens Processed**: 1.27B tokens
+- **Current Loss**: 4.5927
 - **Spike Rate**: 0.0503
 - **Learning Rate**: 1.98e-04
 
@@ -54,7 +54,7 @@ This represents ongoing training of the first large-scale spiking neural network
 from huggingface_hub import hf_hub_download
 checkpoint = hf_hub_download(
     repo_id="rootxhacker/piking-llm-5b-3epochs-exp",
-    filename="checkpoint_122000.pt"
+    filename="checkpoint_124000.pt"
 )
 
 # Load with custom spiking model code
@@ -65,4 +65,4 @@ checkpoint = hf_hub_download(
 
 **🔬 This is live research in progress! Check back for updates as training continues.**
 
-**Training Progress**: 8.3% complete towards 15B tokens
+**Training Progress**: 8.5% complete towards 15B tokens

checkpoint_124000.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a48258d9cf7e744b0205f957f235b75306c272993a93c48a493dac0ed605658c
+size 766783466

config.json

@@ -4,11 +4,11 @@
   "hidden_size": 768,
   "num_layers": 12,
   "max_seq_length": 1024,
-  "training_step": 122000,
-  "tokens_processed": 1249280000,
-  "loss": 4.551811754550579,
-  "spike_rate": 0.050348762208124606,
-  "learning_rate": 0.00019795068984869974,
-  "epoch": 0.249856,
-  "progress_percent": 8.328554654433248
+  "training_step": 124000,
+  "tokens_processed": 1269760000,
+  "loss": 4.592672362989102,
+  "spike_rate": 0.05032278543870527,
+  "learning_rate": 0.0001978616335133679,
+  "epoch": 0.253952,
+  "progress_percent": 8.46508833729281
 }