INDEX

Explanations

phrases where an opinion or input is provided

oai_token-act-pair · gpt-3.5-turbo

occurrences of the phrase "weigh in."

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

CLASSIFIED

-0.74

CVE

-0.70

=]

-0.68

STER

-0.66

soever

-0.65

lvl

-0.62

dar

-0.58

killed

-0.58

":"/

-0.58

ŀ

-0.57

POSITIVE LOGITS

ordinate

1.02

clusions

1.01

 unison

0.99

 accordance

0.95

 favor

0.91

 lieu

0.91

 conjunction

0.90

 vitro

0.89

 favour

0.88

between

0.88

Activations Density 0.165%

phrases where an opinion or input is provided

occurrences of the phrase "weigh in."

No Comments

No Known Activations