INDEX

Explanations

terms related to forensic investigations

oai_token-act-pair · gpt-3.5-turbo

references to forensic science and investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

unin

-0.87

self

-0.77

Reilly

-0.75

Mos

-0.70

Pac

-0.69

bid

-0.68

comes

-0.67

Hop

-0.66

eworld

-0.66

————————————————

-0.64

POSITIVE LOGITS

nsic

1.25

 forensic

0.97

nsics

0.96

abilia

0.84

 Forensic

0.83

 laboratories

0.82

apons

0.81

 examination

0.81

 technician

0.78

 pathology

0.74

Activations Density 0.008%

terms related to forensic investigations

references to forensic science and investigations

No Comments

No Known Activations