INDEX

Explanations

words related to apologies or expressing regret

oai_token-act-pair · gpt-3.5-turbo

terms related to apologies and the act of apologizing

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

VK

-0.67

 Huntington

-0.66

 Morning

-0.65

PRES

-0.65

Boh

-0.64

dream

-0.64

 Dare

-0.64

 Sage

-0.63

 Grail

-0.63

 FISA

-0.63

POSITIVE LOGITS

etics

1.46

etic

1.41

etically

1.29

acy

1.04

izes

1.04

ologies

1.02

ises

0.99

ogenic

0.97

opoulos

0.96

orie

0.96

Activations Density 0.008%

words related to apologies or expressing regret

terms related to apologies and the act of apologizing

No Comments

No Known Activations