INDEX

Explanations

negative adjectives describing behavior or personality traits

oai_token-act-pair · gpt-3.5-turbo

descriptive phrases that convey negative or critical traits about individuals or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atform

-0.92

 GOODMAN

-0.86

nesday

-0.85

cade

-0.81

 Minutes

-0.79

verage

-0.76

Opening

-0.75

 reopened

-0.73

launch

-0.72

sequent

-0.71

POSITIVE LOGITS

 narcissistic

1.51

 arrogant

1.48

 greedy

1.48

 narciss

1.44

 sociop

1.44

 psychopath

1.41

 lazy

1.39

 cowardly

1.37

 incompetent

1.36

 selfish

1.36

Activations Density 0.471%

negative adjectives describing behavior or personality traits

descriptive phrases that convey negative or critical traits about individuals or groups

No Comments

No Known Activations