INDEX

Explanations

prepositional phrases indicating a contrast or a discrepancy between perception and reality

oai_token-act-pair · gpt-3.5-turbo

expressions indicating conditions related to situational context or dependency

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

 catchy

-0.63

 Staples

-0.58

phabet

-0.57

 unfocusedRange

-0.55

 glim

-0.55

ById

-0.54

 Telegram

-0.54

CTR

-0.54

 unsolved

-0.54

 Emblem

-0.53

POSITIVE LOGITS

é¾įå

0.71

fil

0.68

advert

0.65

enture

0.65

 Tours

0.63

pires

0.62

haps

0.61

ean

0.61

cho

0.61

ased

0.60

Activations Density 0.105%

prepositional phrases indicating a contrast or a discrepancy between perception and reality

expressions indicating conditions related to situational context or dependency

No Comments

No Known Activations