metadata

license: llama2
language:
  - en
pipeline_tag: conversational

Goliath 120B

An auto-regressive causal LM created by combining 2x finetuned Llama-2 70B into one.

Prompting Format

Both Vicuna and Alpaca will work, but due the initial and final layers belonging primarily to Xwin, I expect Vicuna to work the best.

Merge process

The models used in the merge are Xwin and Euryale.

The layer ranges used are as follows:

- range 0, 16
  Xwin
- range 8, 24
  Euryale
- range 17, 32
  Xwin
- range 25, 40
  Euryale
- range 33, 48
  Xwin
- range 41, 56
  Euryale
- range 49, 64
  Xwin
- range 57, 72
  Euryale
- range 65, 80
  Xwin

Screenshots

Tested with SillyTavern using the Vicuana 1.1 prompt template. Preset: Miro Gold. (NOTE: I'm not an expert prompter and I don't RP, so take my attemtps with a grain of salt and try for yourself.)

Benchmarks

Coming soon.