INDEX

Explanations

references to elementary schools, especially in the context of news stories

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

redo

-0.75

ulkan

-0.74

cific

-0.70

————————————————

-0.68

pard

-0.67

idity

-0.65

sten

-0.64

athering

-0.64

pez

-0.63

ALSE

-0.63

POSITIVE LOGITS

 school

1.22

 School

1.18

chool

1.08

 SCHOOL

1.06

 Teachers

1.06

 schools

1.05

School

1.04

school

1.04

 classroom

0.96

 kindergarten

0.94

Activations Density 0.047%

references to elementary schools, especially in the context of news stories

No Comments

No Known Activations