INDEX

Explanations

technical terms related to electronics and engineering

oai_token-act-pair · gpt-3.5-turbo

technical specifications and features of devices

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Syri

-0.77

 Feminist

-0.75

 Yemeni

-0.75

Jews

-0.74

 ethn

-0.73

 Social

-0.73

 Moroccan

-0.71

 Haitian

-0.70

 Bris

-0.70

 Anthropology

-0.70

POSITIVE LOGITS

 voltage

1.30

 airflow

1.13

 vibration

1.12

 braking

1.10

 sensor

1.07

 torque

1.06

 capacitor

1.05

 calibration

1.05

 ignition

1.04

 circuitry

1.04

Activations Density 0.977%

technical terms related to electronics and engineering

technical specifications and features of devices

No Comments

No Known Activations