INDEX

Explanations

occurrences of the word "Things."

oai_token-act-pair · gpt-3.5-turbo

references to "things," particularly in the context of devices or concepts related to the Internet of Things

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

 anonymity

-0.81

apego

-0.70

 solo

-0.63

lished

-0.63

ECD

-0.63

adel

-0.61

lishing

-0.59

 dissertation

-0.58

 rehab

-0.58

lain

-0.58

POSITIVE LOGITS

 Matters

0.84

 Things

0.82

iverse

0.81

Things

0.80

tons

0.71

Else

0.70

tery

0.70

 Happ

0.69

soever

0.66

 Resist

0.66

Activations Density 0.052%

occurrences of the word "Things."

references to "things," particularly in the context of devices or concepts related to the Internet of Things

No Comments

No Known Activations