INDEX

Explanations

special characters, numbers, and symbols

oai_token-act-pair · gpt-3.5-turbo

terms related to visual imagery and social media engagement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 promul

-0.73

nown

-0.67

 examinations

-0.66

-0.64

ocry

-0.62

geries

-0.61

 contemplated

-0.60

 athlet

-0.60

gow

-0.59

SPONSORED

-0.59

POSITIVE LOGITS

3.38

2.14

+#

1.91

++

1.68

 plus

1.58

+=

1.55

+++

1.52

"+

1.50

+.

1.49

)+

1.48

Activations Density 0.012%

special characters, numbers, and symbols

terms related to visual imagery and social media engagement

No Comments

No Known Activations