INDEX

Explanations

instances of the word "inspiration" at varying activation levels

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "inspiration."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ington

-0.74

ewater

-0.70

eb

-0.70

VA

-0.69

irth

-0.66

 Strait

-0.65

 Roses

-0.64

 Univ

-0.64

don

-0.63

POSITIVE LOGITS

 inspiration

1.41

iration

1.20

 inspir

1.14

 muse

1.07

 inspired

0.94

 curiosity

0.83

 inspiring

0.79

 inspire

0.79

 Insp

0.76

Insp

0.76

Activations Density 0.005%

instances of the word "inspiration" at varying activation levels

occurrences of the word "inspiration."

No Comments

No Known Activations