INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

mentions of specific individuals or names

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

ners

-0.85

Interstitial

-0.76

 Africans

-0.74

ularity

-0.72

ition

-0.68

stall

-0.68

ainment

-0.67

 Gamb

-0.66

uese

-0.65

 cook

-0.64

POSITIVE LOGITS

 Shea

0.86

iqueness

0.79

hops

0.75

jo

0.74

soDeliveryDate

0.71

ynthesis

0.70

ãĤ´ãĥ³

0.66

ept

0.65

cedented

0.65

Ĩ

0.65

Activations Density 0.014%

names of individuals

mentions of specific individuals or names

No Comments

No Known Activations