INDEX

Explanations

phrases related to differences or changes in various contexts

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate differences or changes

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

ortium

-0.59

iasco

-0.58

umar

-0.58

(>

-0.56

ighed

-0.55

 showc

-0.55

anium

-0.54

pointer

-0.54

Rap

-0.54

 Reward

-0.53

POSITIVE LOGITS

 differently

1.67

 different

1.66

different

1.43

 worse

1.37

 opposite

1.30

 simpler

1.25

 radically

1.22

 vastly

1.21

 harsher

1.17

 similar

1.17

Activations Density 0.649%

phrases related to differences or changes in various contexts

phrases that indicate differences or changes

No Comments

No Known Activations