INDEX

Explanations

contractions of "is" or "has" followed by a descriptive term

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate existence or presence

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

roit

-0.81

eal

-0.79

ares

-0.71

soever

-0.68

enth

-0.67

 approves

-0.64

 WARN

-0.62

eals

-0.62

|--

-0.62

ear

-0.59

POSITIVE LOGITS

 been

1.12

 gotta

1.08

 plenty

0.99

 always

0.93

 nothing

0.92

 gonna

0.89

no

0.85

 still

0.84

 ample

0.83

 lots

0.82

Activations Density 0.035%

contractions of "is" or "has" followed by a descriptive term

phrases that indicate existence or presence

No Comments

No Known Activations