INDEX

Explanations

phrases related to literary or entertainment agents

oai_token-act-pair · gpt-3.5-turbo

references to agents and agencies, particularly in the context of representation and management

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

shall

-0.72

SPONSORED

-0.69

 Poverty

-0.68

 Kali

-0.67

unin

-0.67

kai

-0.65

inki

-0.65

Freedom

-0.64

zees

-0.64

fed

-0.64

POSITIVE LOGITS

 buyers

0.97

 broker

0.97

 invoice

0.93

urers

0.91

 buyer

0.91

 bidder

0.90

 valuation

0.89

 brokerage

0.89

 auction

0.88

 seller

0.85

Activations Density 0.106%

phrases related to literary or entertainment agents

references to agents and agencies, particularly in the context of representation and management

No Comments

No Known Activations