INDEX

Explanations

phrases or sentences mentioning positive attributes or features of something

oai_token-act-pair · gpt-3.5-turbo

statements that highlight the notable or defining features of something

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

iti

-0.67

hung

-0.64

en

-0.63

iet

-0.62

ut

-0.62

atche

-0.61

hol

-0.60

ievers

-0.60

ibus

-0.60

stad

-0.60

POSITIVE LOGITS

 undoubtedly

0.87

 longevity

0.80

 sheer

0.80

 indeed

0.76

 namely

0.74

how

0.73

 evidenced

0.73

 simplicity

0.72

 astounding

0.72

 undeniable

0.71

Activations Density 0.140%

phrases or sentences mentioning positive attributes or features of something

statements that highlight the notable or defining features of something

No Comments

No Known Activations