INDEX

Explanations

mentions of well-known figures in various contexts

oai_token-act-pair · gpt-3.5-turbo

positive attributes and characteristics of individuals, especially in the context of their reputation

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

olutions

-0.69

Solution

-0.67

unicip

-0.66

 dispers

-0.64

phrine

-0.63

ļéĨĴ

-0.63

 downstream

-0.63

 incent

-0.63

olving

-0.62

 Implementation

-0.62

POSITIVE LOGITS

his

1.35

he

1.26

His

1.17

his

1.13

Born

1.06

His

1.04

him

1.03

 Born

1.02

 biography

0.95

HIS

0.92

Activations Density 0.753%

mentions of well-known figures in various contexts

positive attributes and characteristics of individuals, especially in the context of their reputation

No Comments

No Known Activations