INDEX

Explanations

positive affirmations or fortunate events

oai_token-act-pair · gpt-3.5-turbo

expressions of relief or fortunate circumstances

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

chairs

-0.70

design

-0.69

 kindred

-0.68

appro

-0.68

 newsletters

-0.67

vice

-0.65

abo

-0.64

 laundry

-0.63

dq

-0.63

raised

-0.62

POSITIVE LOGITS

 fortunately

0.89

Fortunately

0.84

Thankfully

0.83

 thankfully

0.77

 luckily

0.76

Luckily

0.76

ESA

0.75

wcs

0.74

 Thankfully

0.73

 mosqu

0.72

Activations Density 0.007%

positive affirmations or fortunate events

expressions of relief or fortunate circumstances

No Comments

No Known Activations