INDEX

Explanations

phrases related to legal permissions and restrictions

oai_token-act-pair · gpt-3.5-turbo

modal verbs indicating ability or permission

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

 Cortex

-0.74

 Lauder

-0.67

Vision

-0.66

omen

-0.65

 Bomber

-0.62

 Dough

-0.62

 Darius

-0.62

 Mant

-0.61

 Niet

-0.61

 Brent

-0.61

POSITIVE LOGITS

 legally

1.18

 lawfully

1.18

sue

1.16

 revoke

1.14

 impose

1.12

 unilaterally

1.08

 freely

1.07

 choose

1.05

't

1.04

 invoke

1.04

Activations Density 0.254%

phrases related to legal permissions and restrictions

modal verbs indicating ability or permission

No Comments

No Known Activations