INDEX

Explanations

words related to identity, especially in the context of organizations or legal terms

oai_token-act-pair · gpt-3.5-turbo

terms related to various forms of "ity," indicating qualities or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

accompan

-0.86

kef

-0.76

ask

-0.65

alogue

-0.64

 ellipt

-0.63

opa

-0.62

£ı

-0.62

soDeliveryDate

-0.61

iston

-0.61

/**

-0.61

POSITIVE LOGITS

Peb

0.88

ilty

0.77

leness

0.76

fulness

0.75

rics

0.74

phal

0.71

ity

0.70

 Rating

0.70

ality

0.69

 IMAGES

0.69

Activations Density 0.025%

words related to identity, especially in the context of organizations or legal terms

terms related to various forms of "ity," indicating qualities or states

No Comments

No Known Activations