INDEX

Explanations

SO, which could stand for Significant Other

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "SO" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ãĥ¼ãĥĨãĤ£

-0.75

riott

-0.67

marks

-0.67

ury

-0.67

killers

-0.64

cause

-0.61

kamp

-0.61

uador

-0.60

heimer

-0.60

 ransom

-0.59

POSITIVE LOGITS

FTWARE

1.48

OTS

0.92

othe

0.89

BI

0.88

VERS

0.87

IL

0.86

CC

0.85

LY

0.85

ETH

0.84

arin

0.84

Activations Density 0.009%

SO, which could stand for Significant Other

occurrences of the term "SO" in various contexts

No Comments

No Known Activations