INDEX

Explanations

numerical values indicating earthquake magnitudes

oai_token-act-pair · gpt-3.5-turbo

references to earthquake magnitudes

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

inelli

-0.78

bones

-0.77

nee

-0.76

tek

-0.73

riad

-0.71

ohydrate

-0.71

foundation

-0.70

bies

-0.70

 Byrne

-0.69

rose

-0.69

POSITIVE LOGITS

 magnitude

1.13

 scale

0.86

 anomaly

0.81

 sensitivity

0.80

 equival

0.80

emonic

0.78

 earthquake

0.77

 iceberg

0.77

 submer

0.76

 extent

0.75

Activations Density 0.011%

numerical values indicating earthquake magnitudes

references to earthquake magnitudes

No Comments

No Known Activations