INDEX

Explanations

words related to television networks and programs

oai_token-act-pair · gpt-3.5-turbo

references to television networks and their programming

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

orget

-0.76

orio

-0.75

vironment

-0.73

ravel

-0.73

umatic

-0.71

heon

-0.71

celona

-0.69

istry

-0.69

 Flask

-0.69

ertodd

-0.69

POSITIVE LOGITS

 television

1.15

TV

1.15

 airing

1.15

tv

1.13

 aired

1.09

 airs

1.07

 commercials

1.06

TVs

1.06

 broadcasts

1.06

 Television

1.06

Activations Density 0.104%

words related to television networks and programs

references to television networks and their programming

No Comments

No Known Activations