INDEX

Explanations

names of US Supreme Court Justices and legal terms related to court decisions and opinions

oai_token-act-pair · gpt-3.5-turbo

references to judicial opinions and dissenting judgments

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

 highs

-0.94

 realise

-0.89

icro

-0.84

 Seah

-0.80

 flares

-0.80

 rumours

-0.79

obyl

-0.78

 showers

-0.76

 organise

-0.73

 pulses

-0.73

POSITIVE LOGITS

Attorney

1.18

 appellate

1.15

 plaintiffs

1.11

 Scalia

1.11

 Plaintiff

1.11

 unconstitutional

1.11

 petitioner

1.10

 supra

1.09

 justices

1.08

 plaintiff

1.08

Activations Density 0.732%

names of US Supreme Court Justices and legal terms related to court decisions and opinions

references to judicial opinions and dissenting judgments

No Comments

No Known Activations