INDEX

Explanations

mentions of electronic devices or technological terms, particularly focusing on terms including 'DR'

oai_token-act-pair · gpt-3.5-turbo

references to a specific product or product type indicated by "DR" and its variations

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

eers

-0.90

 Corpus

-0.86

 Pose

-0.80

eering

-0.79

quo

-0.72

eer

-0.71

ropolitan

-0.71

ERN

-0.68

 ç¥ŀ

-0.65

lace

-0.64

POSITIVE LOGITS

adeon

1.04

ESS

1.02

inks

1.01

inking

1.00

INK

0.96

IVE

0.89

ink

0.87

abbit

0.84

IVERS

0.84

OME

0.83

Activations Density 0.022%

mentions of electronic devices or technological terms, particularly focusing on terms including 'DR'

references to a specific product or product type indicated by "DR" and its variations

No Comments

No Known Activations