INDEX

Explanations

phrases and words used in formatting text documents

oai_token-act-pair · gpt-3.5-turbo

instances of a specific character or symbol repeatedly occurring in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Crest

-0.80

 Ridley

-0.72

 Airways

-0.68

 Celest

-0.63

 Spice

-0.63

 Monk

-0.62

 Patt

-0.62

ffen

-0.62

 Mechdragon

-0.60

 Skull

-0.59

POSITIVE LOGITS

ship

0.86

ÂŃ

0.85

deck

0.83

tile

0.83

national

0.79

city

0.79

right

0.77

City

0.77

RELATED

0.76

tal

0.76

Activations Density 0.005%

phrases and words used in formatting text documents

instances of a specific character or symbol repeatedly occurring in the text

No Comments

No Known Activations