INDEX

Explanations

mentions of specific names or usernames along with corresponding text

oai_token-act-pair · gpt-3.5-turbo

social media engagement and sharing prompts

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

etheless

-0.85

 thrust

-0.77

hement

-0.76

 displacement

-0.71

 apprehended

-0.70

 shun

-0.67

 submar

-0.67

 shove

-0.66

 penal

-0.66

 tackling

-0.66

POSITIVE LOGITS

CHAPTER

1.00

ÂłÂłÂłÂł

0.96

Anonymous

0.91

Follow

0.90

Original

0.90

QUEST

0.89

FAQ

0.88

robe

0.88

EGIN

0.87

EDIT

0.87

Activations Density 0.118%

mentions of specific names or usernames along with corresponding text

social media engagement and sharing prompts

No Comments

No Known Activations