INDEX

Explanations

words related to the action of canceling something

oai_token-act-pair · gpt-3.5-turbo

terms related to canceling events or reservations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

rity

-0.82

adata

-0.73

istics

-0.67

berger

-0.66

ixel

-0.65

olds

-0.64

erers

-0.64

vern

-0.64

ixels

-0.63

ea

-0.62

POSITIVE LOGITS

 plans

0.85

 invitations

0.79

 orders

0.79

 pregnancies

0.79

 reservations

0.78

 cancel

0.77

 warranties

0.75

 registrations

0.74

 canceled

0.73

 cance

0.73

Activations Density 0.062%

words related to the action of canceling something

terms related to canceling events or reservations

No Comments

No Known Activations