You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
stablelm_model_quantize: loading model from './stablelm-base-alpha-3b/ggml-model-f16.bin'
stablelm_model_quantize: n_vocab = 50688
stablelm_model_quantize: n_ctx = 4096
stablelm_model_quantize: n_embd = 4096
stablelm_model_quantize: n_head = 32
stablelm_model_quantize: n_layer = 16
stablelm_model_quantize: ftype = 1
ggml_common_quantize_0: invalid quantization type 0 (f32)
stablelm_model_quantize: failed to quantize model './stablelm-base-alpha-3b/ggml-model-f16.bin'
main: failed to quantize model from './stablelm-base-alpha-3b/ggml-model-f16.bin'
I'm a newb, my guess was that means the stablelm-quantize expected 32 instead of f16, so I tried the same steps but used python3 ../examples/stablelm/convert-h5-to-ggml.py ./stablelm-base-alpha-3b/ 0 (0 instead of 1), but no cigar.
The text was updated successfully, but these errors were encountered:
I followed examples/stablelm step-by. At the 4-bit part, I get the following error:
I'm a newb, my guess was that means the stablelm-quantize expected 32 instead of f16, so I tried the same steps but used
python3 ../examples/stablelm/convert-h5-to-ggml.py ./stablelm-base-alpha-3b/ 0
(0
instead of1
), but no cigar.The text was updated successfully, but these errors were encountered: