INDEX

Explanations

adjectives depicting a passive-aggressive behavior

oai_token-act-pair · gpt-3.5-turbo

terms related to passive behavior or passivity

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ograp

-0.81

ĸļ

-0.76

AMI

-0.72

 Nightmares

-0.70

ographers

-0.70

ographics

-0.70

ittee

-0.69

ucks

-0.69

utra

-0.68

ologists

-0.67

POSITIVE LOGITS

comm

0.94

 Passive

0.92

 passive

0.85

aggressive

0.80

 minded

0.79

reet

0.78

wd

0.78

ives

0.78

 passively

0.77

 reactive

0.76

Activations Density 0.013%

adjectives depicting a passive-aggressive behavior

terms related to passive behavior or passivity

No Comments

No Known Activations