INDEX

Explanations

phrases related to penalty or consequence

oai_token-act-pair · gpt-3.5-turbo

references to increasing quantities or allowances

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

Maker

-0.75

 Fidel

-0.65

 Abyss

-0.63

 Mate

-0.62

 Kubrick

-0.62

 mash

-0.62

 Marcos

-0.61

[/

-0.61

 Mori

-0.61

 Ferdinand

-0.59

POSITIVE LOGITS

graded

1.02

grades

0.98

grading

0.85

dates

0.83

icum

0.81

adesh

0.80

rights

0.79

erd

0.78

essions

0.77

olicy

0.76

Activations Density 0.057%

phrases related to penalty or consequence

references to increasing quantities or allowances

No Comments

No Known Activations