INDEX

Explanations

phrases related to government departments and services

oai_token-act-pair · gpt-3.5-turbo

terms related to health and human services departments

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

 pacif

-0.76

ãĥª

-0.71

stick

-0.70

arding

-0.68

ÅŁ

-0.67

 Schne

-0.67

ning

-0.64

manent

-0.62

ãĥĥãĥī

-0.62

NING

-0.60

POSITIVE LOGITS

 Services

1.28

 Provider

0.98

ettlement

0.93

 Assistance

0.87

 Administration

0.86

 Advis

0.85

Services

0.85

Pv

0.84

 Employees

0.84

 Administrator

0.84

Activations Density 0.009%

phrases related to government departments and services

terms related to health and human services departments

No Comments

No Known Activations