INDEX

Explanations

phrases related to priority or order

oai_token-act-pair · gpt-3.5-turbo

references to rules or systems of allocation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

enhagen

-0.77

endiary

-0.76

ciating

-0.76

ŃĶ

-0.70

FontSize

-0.68

cific

-0.67

REF

-0.66

 enthusi

-0.66

tions

-0.65

idia

-0.63

POSITIVE LOGITS

eat

0.77

 thou

0.76

 sheep

0.66

 rinse

0.65

 Lose

0.65

 flock

0.65

 lett

0.64

 faire

0.63

 reap

0.62

 Heal

0.62

Activations Density 0.308%

phrases related to priority or order

references to rules or systems of allocation

No Comments

No Known Activations