INDEX

Explanations

mentions of medical conditions, procedures, or pharmaceuticals

oai_token-act-pair · gpt-3.5-turbo

words and phrases related to end users and their interactions with services or products

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

-0.72

ilver

-0.70

Ans

-0.67

itsch

-0.67

ï¸ı

-0.66

pload

-0.65

ombat

-0.65

 anecd

-0.64

 autos

-0.63

 obser

-0.62

POSITIVE LOGITS

BuyableInstoreAndOnline

0.74

enment

0.74

ciating

0.68

itures

0.67

ctive

0.67

ruction

0.65

 civilisation

0.64

 cliff

0.62

itory

0.61

 Sole

0.61

Activations Density 0.102%

mentions of medical conditions, procedures, or pharmaceuticals

words and phrases related to end users and their interactions with services or products

No Comments

No Known Activations