INDEX

Explanations

phrases related to well-being and mental health

oai_token-act-pair · gpt-3.5-turbo

the term "being" and related concepts of existence and well-being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Brave

-0.64

My

-0.64

 Cour

-0.63

 Select

-0.62

lur

-0.61

 Central

-0.61

 straw

-0.61

Rec

-0.60

 Court

-0.59

 Guide

-0.59

POSITIVE LOGITS

being

3.53

having

1.49

doing

1.23

still

1.11

yip

1.10

taking

1.07

tons

1.06

together

1.05

moving

1.01

violence

1.01

Activations Density 0.018%

phrases related to well-being and mental health

the term "being" and related concepts of existence and well-being

No Comments

No Known Activations