INDEX

Explanations

references to IT-related terms and processes

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

itives

-0.71

ktop

-0.71

lets

-0.64

terday

-0.61

 unrestricted

-0.60

————————————————

-0.58

 triglycer

-0.58

ãĥ£

-0.57

á

-0.57

":""},{"

-0.57

POSITIVE LOGITS

IME

1.13

EMS

1.06

unes

0.98

MIN

0.97

ILA

0.92

TO

0.89

ECH

0.87

rehensive

0.87

LA

0.85

AMI

0.85

Activations Density 0.048%

references to IT-related terms and processes

No Comments

No Known Activations