Rotary Scaling Factor of 4 for 8k context (Do not merge)

#23
by nbroad HF staff - opened
Files changed (1) hide show
  1. config.json +1 -1
config.json CHANGED
@@ -38,7 +38,7 @@
38
  "rotary_emb_fraction": 1.0,
39
  "rotary_emb_interleaved": false,
40
  "rotary_emb_scale_base": null,
41
- "rotary_scaling_factor": null,
42
  "scale_attn_by_inverse_layer_idx": false,
43
  "scale_attn_weights": true,
44
  "summary_activation": null,
 
38
  "rotary_emb_fraction": 1.0,
39
  "rotary_emb_interleaved": false,
40
  "rotary_emb_scale_base": null,
41
+ "rotary_scaling_factor": 4.0,
42
  "scale_attn_by_inverse_layer_idx": false,
43
  "scale_attn_weights": true,
44
  "summary_activation": null,