INDEX

Explanations

references to characters and elements from a sci-fi TV series

oai_token-act-pair · gpt-3.5-turbo

references to "Star Trek" characters and episodes, particularly the Borg and Captain Picard

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

udic

-0.90

 tremend

-0.88

eanor

-0.74

gow

-0.72

urgy

-0.71

anguage

-0.68

eger

-0.67

ideshow

-0.67

arettes

-0.66

opping

-0.66

POSITIVE LOGITS

 Yards

0.86

kies

0.83

brace

0.82

 Voyager

0.80

thal

0.75

wana

0.72

ajor

0.70

ble

0.69

 Kira

0.68

iotic

0.68

Activations Density 0.041%

references to characters and elements from a sci-fi TV series

references to "Star Trek" characters and episodes, particularly the Borg and Captain Picard

No Comments

No Known Activations