INDEX

Explanations

words related to criminal cases, investigations, and law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to investigative cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

erd

-0.78

atomic

-0.78

semb

-0.75

orks

-0.75

icc

-0.75

undai

-0.74

lav

-0.72

ipher

-0.72

assadors

-0.72

rongh

-0.72

POSITIVE LOGITS

cases

1.09

 involving

0.86

case

0.76

 against

0.76

 Case

0.74

ysis

0.71

 Scientology

0.70

 histories

0.70

 cases

0.69

 Grimoire

0.69

Activations Density 0.037%

words related to criminal cases, investigations, and law enforcement

references to investigative cases

No Comments

No Known Activations