INDEX

Explanations

mentions of legal figures, specifically judges

oai_token-act-pair · gpt-3.5-turbo

references to judges and judicial decisions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Rivals

-0.73

ulus

-0.72

 Roads

-0.68

 Horde

-0.67

iliar

-0.67

ulations

-0.66

haar

-0.66

kefeller

-0.66

uld

-0.64

ohyd

-0.64

POSITIVE LOGITS

 presiding

1.03

 overseeing

0.99

rooms

0.93

 ruled

0.80

 Advocate

0.77

redd

0.76

room

0.75

jud

0.75

 sentenced

0.74

 rulings

0.72

Activations Density 0.023%

mentions of legal figures, specifically judges

references to judges and judicial decisions

No Comments

No Known Activations