INDEX

Explanations

proper nouns or names preceded by the word "Said"

oai_token-act-pair · gpt-3.5-turbo

reported speech or quotations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ä¿

-0.78

RAY

-0.71

ç¥ŀ

-0.71

æł

-0.70

istration

-0.70

illation

-0.67

ç«

-0.67

ä½ľ

-0.65

Ð¾Ð

-0.65

andering

-0.65

POSITIVE LOGITS

 Ones

1.04

 Doesn

1.01

 Called

1.00

 Alive

0.97

 Them

0.96

 Cause

0.96

 Wrong

0.95

 Makes

0.95

 Difference

0.93

 Gets

0.90

Activations Density 0.067%

proper nouns or names preceded by the word "Said"

reported speech or quotations

No Comments

No Known Activations