INDEX

Explanations

instructions or recommendations to take specific actions

oai_token-act-pair · gpt-3.5-turbo

the word "be" and its variations in imperative sentences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

--+

-0.77

viz

-0.75

regation

-0.67

 ILCS

-0.65

Osc

-0.63

regate

-0.62

 Combine

-0.60

 Junction

-0.57

 Scrib

-0.56

CU

-0.56

POSITIVE LOGITS

 careful

1.19

 sure

1.16

 wary

1.14

 warned

1.12

hold

1.12

gins

1.12

yond

1.09

 aware

1.08

 mindful

1.00

ijing

0.98

Activations Density 0.042%

instructions or recommendations to take specific actions

the word "be" and its variations in imperative sentences

No Comments

No Known Activations