INDEX

Explanations

words related to protection or authority figures

oai_token-act-pair · gpt-3.5-turbo

references to guardianship or protector 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

urses

-0.76

mers

-0.72

mia

-0.71

bers

-0.69

hist

-0.68

urred

-0.67

cloth

-0.66

RESULTS

-0.66

reme

-0.65

skirts

-0.65

POSITIVE LOGITS

hip

1.01

arium

0.93

hips

0.93

 guardians

0.93

 Guardians

0.88

ians

0.86

arians

0.85

iola

0.81

 angel

0.80

Guard

0.79

Activations Density 0.026%

words related to protection or authority figures

references to guardianship or protector roles in various contexts

No Comments

No Known Activations