INDEX

Explanations

mentions of government or government-related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of government and its various roles and implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Carnival

-0.76

hirt

-0.74

uberty

-0.69

vre

-0.68

issan

-0.67

ussen

-0.67

wark

-0.66

CV

-0.66

east

-0.65

jen

-0.65

POSITIVE LOGITS

 agencies

1.20

 shutdown

1.15

 bureaucracy

1.07

 Accountability

1.04

 Shutdown

0.95

 bureaucrats

0.94

 officials

0.94

 intervention

0.94

 watchdog

0.93

 coffers

0.92

Activations Density 0.077%

mentions of government or government-related terms

mentions of government and its various roles and implications

No Comments

No Known Activations