INDEX

Explanations

mentions of incidents involving fatalities or medical conditions related to them

oai_token-act-pair · gpt-3.5-turbo

terms related to concepts of morality or ethical standards

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

scr

-0.65

 press

-0.64

 splendid

-0.60

new

-0.59

 Brave

-0.59

 feather

-0.57

 latest

-0.57

mil

-0.57

 magnificent

-0.56

tee

-0.54

POSITIVE LOGITS

ality

4.94

alities

3.13

ally

2.15

als

1.58

alties

1.41

nel

1.40

ALLY

1.35

ativity

1.31

arity

1.28

ities

1.27

Activations Density 0.013%

mentions of incidents involving fatalities or medical conditions related to them

terms related to concepts of morality or ethical standards

No Comments

No Known Activations