INDEX

Explanations

references to living or life

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "living" or lifestyle-related themes

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ussian

-0.73

 throttle

-0.72

oute

-0.71

acca

-0.68

ags

-0.66

 clicks

-0.66

 preced

-0.65

 unse

-0.64

 tighten

-0.64

asu

-0.64

POSITIVE LOGITS

 Living

3.77

Living

2.92

living

2.86

 living

2.34

 Livingston

1.59

liv

1.31

 Growing

1.23

 Live

1.17

 Dying

1.14

 Dwell

1.14

Activations Density 0.014%

references to living or life

references to the concept of "living" or lifestyle-related themes

No Comments

No Known Activations