INDEX

Explanations

references to the frequency or quantity of things/people in a comparison

oai_token-act-pair · gpt-3.5-turbo

references to the phrase "like many."

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

ornia

-0.73

UGE

-0.68

afety

-0.67

osterone

-0.66

hess

-0.64

ezvous

-0.64

isma

-0.62

steen

-0.61

ocamp

-0.61

ousy

-0.61

POSITIVE LOGITS

 embodiments

0.71

 predecessors

0.70

 predec

0.69

catentry

0.68

 commenters

0.67

ãĥīãĥ©ãĤ´ãĥ³

0.66

 types

0.64

 peers

0.62

 aspiring

0.62

 approaches

0.61

Activations Density 0.106%

references to the frequency or quantity of things/people in a comparison

references to the phrase "like many."

No Comments

No Known Activations