INDEX

Explanations

words or phrases that start with a hyphen, specifically focusing on negative sentiment or consequences

oai_token-act-pair · gpt-3.5-turbo

phrases or concepts related to negative or detrimental outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ÏĤ

-0.70

 Doodle

-0.69

estern

-0.67

 Loll

-0.67

 Vaughn

-0.66

 Warn

-0.64

 âĶľ

-0.64

 Cumm

-0.63

 Flores

-0.63

HCR

-0.62

POSITIVE LOGITS

based

1.14

of

1.10

heavy

0.96

operated

0.96

to

0.94

mounted

0.93

pain

0.92

time

0.91

level

0.91

tested

0.90

Activations Density 0.110%

words or phrases that start with a hyphen, specifically focusing on negative sentiment or consequences

phrases or concepts related to negative or detrimental outcomes

No Comments

No Known Activations