INDEX

Explanations

phrases related to specific terms or entities, such as movie titles or technical names

oai_token-act-pair · gpt-3.5-turbo

references to titles, names, or entities associated with popular culture

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

meric

-0.74

dated

-0.69

 parity

-0.67

duty

-0.66

ĻĤ

-0.65

 reconc

-0.64

NetMessage

-0.64

compliance

-0.64

=]

-0.63

ãģĤ

-0.63

POSITIVE LOGITS

 Trilogy

1.10

 Collection

1.08

 Podcast

1.03

 Collective

0.99

 Awards

0.99

 Museum

0.97

 Massacre

0.97

 Club

0.96

âĦ¢

0.94

 Hotel

0.94

Activations Density 0.510%

phrases related to specific terms or entities, such as movie titles or technical names

references to titles, names, or entities associated with popular culture

No Comments

No Known Activations