INDEX

Explanations

instances where someone or something is being 'left' in a particular state or situation

oai_token-act-pair · gpt-3.5-turbo

instances of the word "left" indicating abandonment or neglect

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

issance

-0.67

andise

-0.65

 srfAttach

-0.60

wcsstore

-0.60

andan

-0.60

NT

-0.58

antz

-0.58

ALLY

-0.57

 GENERAL

-0.56

CoC

-0.56

POSITIVE LOGITS

overs

1.33

 undone

1.06

 untreated

1.04

 behind

0.99

behind

0.91

wing

0.90

 unanswered

0.87

 intact

0.86

 untouched

0.84

 unfinished

0.84

Activations Density 0.033%

instances where someone or something is being 'left' in a particular state or situation

instances of the word "left" indicating abandonment or neglect

No Comments

No Known Activations