INDEX

Explanations

references to something being of high quality or significance

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "prime" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Canaver

-0.89

anwhile

-0.86

 Polo

-0.71

oldemort

-0.69

urrencies

-0.69

erves

-0.68

okia

-0.66

ravings

-0.66

uum

-0.66

sbm

-0.65

POSITIVE LOGITS

knit

1.07

val

1.02

 ministerial

0.93

 minister

0.90

iary

0.86

etime

0.84

ises

0.82

essential

0.79

iaries

0.76

IME

0.76

Activations Density 0.019%

references to something being of high quality or significance

references to the concept of "prime" in various contexts

No Comments

No Known Activations