INDEX

Explanations

phrases related to data management and ethical practices

oai_token-act-pair · gpt-3.5-turbo

terms related to various practices and methodologies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

otos

-0.78

ilee

-0.77

amen

-0.72

lein

-0.70

bid

-0.69

amaz

-0.68

DonaldTrump

-0.67

zie

-0.67

gin

-0.66

itialized

-0.65

POSITIVE LOGITS

 Practices

0.99

 practices

0.91

hops

0.90

onduct

0.86

afety

0.79

uits

0.74

 governing

0.72

 violations

0.72

iblings

0.70

olicy

0.70

Activations Density 0.019%

phrases related to data management and ethical practices

terms related to various practices and methodologies

No Comments

No Known Activations