INDEX

Explanations

positive descriptors related to performance or quality

oai_token-act-pair · gpt-3.5-turbo

words related to positive performance or quality assessments

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

hyde

-0.71

billion

-0.70

soDeliveryDate

-0.70

abel

-0.67

mega

-0.66

uthor

-0.65

gren

-0.64

ften

-0.63

Aaron

-0.63

":[

-0.63

POSITIVE LOGITS

 results

1.22

 odds

1.08

 grades

1.06

 reviews

1.06

 visibility

1.04

 intentions

1.03

 outcomes

1.02

 margins

1.02

 luck

0.98

 ratings

0.98

Activations Density 0.128%

positive descriptors related to performance or quality

words related to positive performance or quality assessments

No Comments

No Known Activations