INDEX

Explanations

phrases related to well-being and health

oai_token-act-pair · gpt-3.5-turbo

references to well-being and health

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hatt

-0.70

ett

-0.70

SON

-0.68

£ı

-0.66

ãĤ¨ãĥ«

-0.64

ulla

-0.64

 teasp

-0.62

sonian

-0.62

 Cosponsors

-0.61

OTS

-0.61

POSITIVE LOGITS

 wellbeing

0.98

fulness

0.86

 affairs

0.81

 enjoyment

0.80

 expectancy

0.77

 awareness

0.77

ful

0.73

 advis

0.73

 preservation

0.73

 outcomes

0.71

Activations Density 0.029%

phrases related to well-being and health

references to well-being and health

No Comments

No Known Activations