INDEX

Explanations

the word "cane" with varying degrees of activation

oai_token-act-pair · gpt-3.5-turbo

references to agricultural themes and specific crops

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

aries

-0.92

ebus

-0.90

oleon

-0.89

engers

-0.83

eday

-0.81

edom

-0.80

ullivan

-0.78

arters

-0.77

estyles

-0.77

esters

-0.76

POSITIVE LOGITS

 cane

1.01

chnology

0.79

 Daisy

0.71

grass

0.71

 regenerate

0.68

LECT

0.67

 grazing

0.67

cipl

0.66

CLOSE

0.66

ciplinary

0.65

Activations Density 0.038%

the word "cane" with varying degrees of activation

references to agricultural themes and specific crops

No Comments

No Known Activations