INDEX

Explanations

mentions of things being hard-wired or wired up in a literal or figurative sense

oai_token-act-pair · gpt-3.5-turbo

terms related to being "wired" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

gur

-0.71

mol

-0.69

vic

-0.68

anmar

-0.66

alos

-0.66

Brave

-0.66

birth

-0.65

illon

-0.65

anches

-0.63

VK

-0.63

POSITIVE LOGITS

 wired

1.24

wired

1.18

 wiring

1.05

 circuitry

0.90

 connectivity

0.88

 Ethernet

0.86

oreal

0.83

ulla

0.77

 circuits

0.76

 arrang

0.76

Activations Density 0.008%

mentions of things being hard-wired or wired up in a literal or figurative sense

terms related to being "wired" in various contexts

No Comments

No Known Activations