INDEX

Explanations

mentions or references of caregivers

oai_token-act-pair · gpt-3.5-turbo

references to caregivers or care-related roles 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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 panc

-0.67

 tranquil

-0.65

 Galile

-0.64

 filming

-0.63

 aside

-0.62

 forcefully

-0.61

 Temple

-0.59

 siege

-0.59

 sorcery

-0.57

 Mobile

-0.56

POSITIVE LOGITS

rs

4.72

1.64

RS

1.48

rers

1.38

rs

1.37

rb

1.36

hs

1.35

rt

1.34

rd

1.34

rf

1.31

Activations Density 0.003%

mentions or references of caregivers

references to caregivers or care-related roles in various contexts

No Comments

No Known Activations