INDEX

Explanations

references to rain in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to rain and its effects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ernel

-0.87

ovo

-0.84

orea

-0.78

versions

-0.75

VICE

-0.73

entric

-0.72

umbn

-0.71

nesota

-0.70

 Cosponsors

-0.70

netic

-0.70

POSITIVE LOGITS

bows

1.50

forest

1.34

drops

1.20

storms

1.10

storm

1.01

drop

0.97

water

0.92

 rains

0.92

 showers

0.91

coat

0.91

Activations Density 0.014%

references to rain in various contexts

references to rain and its effects

No Comments

No Known Activations