INDEX

Explanations

terms and phrases related to comparisons of statistics or physical specifications

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

resent

-0.68

****************

-0.68

oaded

-0.66

 Rothschild

-0.63

yden

-0.62

 Staten

-0.61

 tatt

-0.61

 confid

-0.59

estern

-0.59

avorite

-0.58

POSITIVE LOGITS

pace

0.87

eme

0.85

pring

0.77

cape

0.75

peed

0.74

mith

0.72

ames

0.71

uman

0.70

paces

0.69

iculty

0.69

Activations Density 0.074%

terms and phrases related to comparisons of statistics or physical specifications

No Comments

No Known Activations