INDEX

Explanations

terms related to light bulbs

oai_token-act-pair · gpt-3.5-turbo

references to light bulbs and their related concepts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

GAN

-0.76

REAM

-0.74

SEA

-0.74

cedented

-0.73

CB

-0.71

ht

-0.64

VO

-0.63

rolled

-0.63

arnaev

-0.63

Snake

-0.63

POSITIVE LOGITS

 bulb

1.31

 bulbs

1.29

ascus

0.90

 sockets

0.80

ophobia

0.80

orescent

0.78

orescence

0.76

oresc

0.76

 optic

0.73

 illum

0.73

Activations Density 0.021%

terms related to light bulbs

references to light bulbs and their related concepts

No Comments

No Known Activations