INDEX

Explanations

terms related to classification or categorization

oai_token-act-pair · gpt-3.5-turbo

terms related to classification or organization into categories

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

RIS

-0.72

bats

-0.67

 Zimmer

-0.66

 Lann

-0.64

 ultras

-0.63

rique

-0.63

 heartbeat

-0.61

 Trojan

-0.59

 Manor

-0.59

 helic

-0.58

POSITIVE LOGITS

naire

0.93

 categories

0.85

category

0.83

omial

0.80

models

0.79

ahead

0.77

etter

0.77

wagon

0.76

gross

0.75

bars

0.75

Activations Density 0.008%

terms related to classification or categorization

terms related to classification or organization into categories

No Comments

No Known Activations