INDEX

Explanations

words related to legal and administrative terms

oai_token-act-pair · gpt-3.5-turbo

terms related to qualifications and legal or official processes

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

etheless

-0.81

 carbohyd

-0.76

 suburbs

-0.74

 nursing

-0.74

 ferment

-0.73

 whisk

-0.71

imeters

-0.71

 turtles

-0.70

alyses

-0.69

 farmland

-0.69

POSITIVE LOGITS

 Prevention

1.05

 Clause

1.04

 Assistance

0.99

 Ones

0.99

 Integrity

0.98

 Direction

0.96

Era

0.96

 Index

0.96

Gate

0.95

 Therapy

0.95

Activations Density 0.522%

words related to legal and administrative terms

terms related to qualifications and legal or official processes

No Known Activations