INDEX

Explanations

adjectives describing characteristics or qualities that make someone or something easily recognizable or distinct

oai_token-act-pair · gpt-3.5-turbo

terms related to distinctiveness and recognition

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

Ö¼

-0.82

VR

-0.70

udeb

-0.68

ceans

-0.68

onz

-0.66

ttp

-0.66

erg

-0.65

×

-0.65

ration

-0.65

Bay

-0.64

POSITIVE LOGITS

 distinctive

1.09

eatures

0.97

ively

0.94

 colours

0.87

 brill

0.85

 warr

0.84

 likeness

0.83

 trademarks

0.83

 distinguishing

0.83

 individuality

0.80

Activations Density 0.011%

adjectives describing characteristics or qualities that make someone or something easily recognizable or distinct

terms related to distinctiveness and recognition

No Comments

No Known Activations