INDEX

Explanations

phrases related to decision-making

oai_token-act-pair · gpt-3.5-turbo

phrases related to making or breaking situations, particularly in contexts involving deceit or manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

":["

-0.64

hov

-0.63

eger

-0.62

 propensity

-0.60

 hemor

-0.59

EStreamFrame

-0.59

 coron

-0.58

 revolving

-0.56

Downloadha

-0.56

Constructed

-0.55

POSITIVE LOGITS

çİĭ

0.78

URE

0.74

estine

0.73

 noises

0.73

ends

0.70

ota

0.69

 Strongh

0.68

 Murd

0.68

asty

0.64

 mockery

0.64

Activations Density 0.076%

phrases related to decision-making

phrases related to making or breaking situations, particularly in contexts involving deceit or manipulation

No Comments

No Known Activations