INDEX

Explanations

strong negative emotions or hostility, particularly related to hatred

oai_token-act-pair · gpt-3.5-turbo

references to hatred and its various expressions and implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

UNCH

-0.83

 helicop

-0.82

ODE

-0.80

umm

-0.73

USE

-0.69

å¸

-0.67

aqu

-0.66

glas

-0.64

change

-0.61

AMA

-0.61

POSITIVE LOGITS

 hatred

0.95

 towards

0.90

 prejudice

0.83

yip

0.83

 toward

0.82

 vengeance

0.82

ãĥĨ

0.78

lessly

0.78

wart

0.77

 rage

0.74

Activations Density 0.029%

strong negative emotions or hostility, particularly related to hatred

references to hatred and its various expressions and implications

No Comments

No Known Activations