INDEX

Explanations

code-related phrases

oai_token-act-pair · gpt-3.5-turbo

the use of the pronoun "we" indicating collaboration or shared experience

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.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

 AIDS

-0.72

 Publication

-0.69

Times

-0.63

 Deaths

-0.62

 Crimes

-0.61

 coronary

-0.61

 JUSTICE

-0.61

 starvation

-0.60

 Amendments

-0.60

 Commissioner

-0.60

POSITIVE LOGITS

've

1.13

're

1.07

'll

1.03

ighed

1.02

ird

1.01

athered

1.00

'd

0.97

 automate

0.94

arers

0.91

 assume

0.88

Activations Density 0.138%

code-related phrases

the use of the pronoun "we" indicating collaboration or shared experience

No Comments

No Known Activations