INDEX

Explanations

questions or phrases asking for definitions or explanations

oai_token-act-pair · gpt-3.5-turbo

the phrase "What is" followed by a question or inquiry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

iard

-0.78

lems

-0.76

cember

-0.68

lator

-0.67

legram

-0.67

owan

-0.66

lette

-0.65

gem

-0.64

lia

-0.64

lication

-0.64

POSITIVE LOGITS

 happening

1.10

 wrong

0.85

 YOUR

0.84

 your

0.83

 meant

0.81

 occurring

0.76

 going

0.74

 Wrong

0.71

 stopping

0.70

 causing

0.70

Activations Density 0.050%

questions or phrases asking for definitions or explanations

the phrase "What is" followed by a question or inquiry

No Comments

No Known Activations