INDEX

Explanations

mentions of television broadcasting networks and schedules

oai_token-act-pair · gpt-3.5-turbo

references to televised events and broadcast details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ndra

-0.99

selage

-0.78

zsche

-0.76

ometric

-0.74

posal

-0.72

ebus

-0.71

ropy

-0.69

closure

-0.68

naire

-0.67

 bloom

-0.66

POSITIVE LOGITS

 broadcasts

1.21

 telev

1.16

levision

1.15

tv

1.13

 broadcast

1.11

TV

1.11

ITV

1.11

TVs

1.09

TV

1.07

 Television

1.07

Activations Density 0.169%

mentions of television broadcasting networks and schedules

references to televised events and broadcast details

No Comments

No Known Activations