INDEX

Explanations

references to large and influential companies or organizations

oai_token-act-pair · gpt-3.5-turbo

references to large corporations or influential entities

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

iggins

-0.81

tein

-0.80

jured

-0.78

edi

-0.73

cells

-0.70

agan

-0.70

 Loft

-0.70

ãĤ¯

-0.69

Vari

-0.68

odox

-0.67

POSITIVE LOGITS

 headquartered

0.94

 corporation

0.85

 titan

0.83

 advertis

0.80

 retailer

0.80

 conglomerate

0.76

orporated

0.76

 subsidiaries

0.76

 executives

0.74

 regulator

0.73

Activations Density 0.086%

references to large and influential companies or organizations

references to large corporations or influential entities

No Comments

No Known Activations