INDEX

Explanations

quotes and spoilers in the text

oai_token-act-pair · gpt-3.5-turbo

instances of quoted text or notable mentions within the document

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

aucas

-0.77

heastern

-0.75

bodied

-0.74

ruciating

-0.74

ink

-0.69

miss

-0.68

 Palest

-0.65

kinson

-0.65

swick

-0.64

obs

-0.63

POSITIVE LOGITS

Quote

1.05

 Investigator

0.94

 Quote

0.90

 Preferences

0.78

Hero

0.78

Maker

0.76

Posts

0.75

Code

0.74

âĸ¬âĸ¬

0.71

Artist

0.70

Activations Density 0.017%

quotes and spoilers in the text

instances of quoted text or notable mentions within the document

No Comments

No Known Activations