INDEX

Explanations

mentions of electronic activities and services

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_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

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

 Kitchen

-0.71

 Anthem

-0.69

NEC

-0.67

 Kats

-0.63

 Vulkan

-0.61

 Ernst

-0.61

 Katz

-0.61

ãĤ¤ãĥĪ

-0.61

 Petty

-0.60

NESS

-0.59

POSITIVE LOGITS

tymology

1.36

gypt

1.25

agles

1.22

ighth

1.21

ternally

1.21

arthed

1.17

ugen

1.13

thereum

1.11

pee

1.05

commerce

1.03

Activations Density 0.025%

mentions of electronic activities and services

No Comments

No Known Activations