INDEX

Explanations

words related to responses, reactions, and attitudes towards events or announcements

oai_token-act-pair · gpt-3.5-turbo

instances of reactions or responses to various events or announcements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Grimoire

-0.69

agate

-0.66

estate

-0.66

 theorem

-0.62

 expires

-0.61

 fitted

-0.61

 unfolds

-0.60

 reusable

-0.60

variable

-0.59

 Norn

-0.59

POSITIVE LOGITS

 applause

1.53

 backlash

1.32

 outcry

1.29

 cheers

1.26

 criticism

1.25

 scorn

1.21

 ridicule

1.20

 skepticism

1.19

 mockery

1.19

 disapproval

1.16

Activations Density 0.246%

words related to responses, reactions, and attitudes towards events or announcements

instances of reactions or responses to various events or announcements

No Comments

No Known Activations