INDEX

Explanations

words related to legal punishment, specifically executions and trials

oai_token-act-pair · gpt-3.5-turbo

references to executions and the process surrounding them

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

thora

-0.74

æ©Ł

-0.73

CRE

-0.71

CE

-0.68

ECO

-0.68

UTH

-0.67

 explan

-0.67

 Develop

-0.66

WD

-0.64

£ı

-0.62

POSITIVE LOGITS

 inmate

1.03

icts

0.93

ifix

0.91

 prisoners

0.91

 inmates

0.90

ictions

0.88

olphin

0.87

 abolition

0.86

 murderer

0.84

death

0.83

Activations Density 0.076%

words related to legal punishment, specifically executions and trials

references to executions and the process surrounding them

No Comments

No Known Activations