INDEX

Explanations

expressions related to being happy or pleased about something

oai_token-act-pair · gpt-3.5-turbo

expressions of happiness related to the state of being

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

ongyang

-0.75

idation

-0.67

ritz

-0.66

 rigs

-0.65

 strain

-0.63

risis

-0.63

 inference

-0.62

rones

-0.60

 fray

-0.60

ortmund

-0.59

POSITIVE LOGITS

 able

1.36

 reminded

1.06

 alive

0.98

 reunited

0.98

 honest

0.91

que

0.91

 judged

0.89

 treated

0.89

amed

0.88

 reborn

0.88

Activations Density 0.105%

expressions related to being happy or pleased about something

expressions of happiness related to the state of being

No Comments

No Known Activations