INDEX

Explanations

tweets mentioning a specific Twitter user

oai_token-act-pair · gpt-3.5-turbo

mentions of social media handles

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Takeru

-0.74

 ambul

-0.73

 nasal

-0.71

 noses

-0.68

 Orig

-0.68

 relegation

-0.67

Tac

-0.67

cel

-0.66

 indemn

-0.65

 Penal

-0.65

POSITIVE LOGITS

realDonaldTrump

1.56

GOP

1.01

truth

0.87

Real

0.86

White

0.85

abby

0.85

raw

0.84

ny

0.84

ky

0.84

Sen

0.84

Activations Density 0.022%

tweets mentioning a specific Twitter user

mentions of social media handles

No Comments

No Known Activations