INDEX

Explanations

states within the United States

oai_token-act-pair · gpt-3.5-turbo

statements regarding specific states and their attributes or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ongyang

-0.95

 behaviours

-0.80

 behaviour

-0.79

atos

-0.72

 calib

-0.70

oyal

-0.68

 fins

-0.68

 analys

-0.67

aviour

-0.67

ariat

-0.66

POSITIVE LOGITS

 Medicaid

0.85

aucuses

0.81

 NAACP

0.78

edu

0.77

population

0.77

sylvania

0.76

gov

0.75

 Interstate

0.69

 District

0.68

bred

0.68

Activations Density 0.670%

states within the United States

statements regarding specific states and their attributes or actions

No Comments

No Known Activations