INDEX

Explanations

terms related to capabilities or capacity

oai_token-act-pair · gpt-3.5-turbo

phrases related to various forms of "capability" or abilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

inos

-0.71

cus

-0.69

birth

-0.68

paying

-0.66

aternity

-0.65

eree

-0.65

ca

-0.65

edi

-0.64

wind

-0.63

oval

-0.63

POSITIVE LOGITS

 capabilities

1.06

 capability

1.00

afety

0.87

auga

0.83

indal

0.82

xes

0.82

 guiActiveUn

0.80

yip

0.80

Ability

0.79

 capacities

0.78

Activations Density 0.014%

terms related to capabilities or capacity

phrases related to various forms of "capability" or abilities

No Comments

No Known Activations