INDEX

Explanations

phrases instructing or encouraging actions

oai_token-act-pair · gpt-3.5-turbo

commands or instructions that encourage action or engagement

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

emale

-0.71

edge

-0.70

album

-0.69

Reply

-0.61

ungle

-0.61

iege

-0.59

hell

-0.58

eller

-0.58

 burden

-0.57

apo

-0.57

POSITIVE LOGITS

 yourselves

1.05

 yourself

1.02

ings

0.85

 Yourself

0.83

ardless

0.81

 thou

0.79

ye

0.75

ya

0.71

 Tata

0.71

able

0.68

Activations Density 0.209%

phrases instructing or encouraging actions

commands or instructions that encourage action or engagement

No Comments

No Known Activations