INDEX

Explanations

occurrences of the name "Xavier" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

references to the name "Xavier" and related names 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 @ 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

patrick

-0.89

wives

-0.88

lear

-0.87

tery

-0.86

fully

-0.81

thritis

-0.80

py

-0.79

vati

-0.78

Seattle

-0.78

bang

-0.77

POSITIVE LOGITS

 tremend

0.83

Bec

0.81

 Gord

0.77

 Byrne

0.76

 notor

0.76

 Olivier

0.75

CLASSIFIED

0.75

 Xavier

0.74

 Maduro

0.73

Agu

0.72

Activations Density 0.022%

occurrences of the name "Xavier" with varying levels of activation

references to the name "Xavier" and related names in various contexts

No Comments

No Known Activations