INDEX

Explanations

information or explanations being provided

oai_token-act-pair · gpt-3.5-turbo

instances of the word "explain" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sabha

-0.73

luster

-0.71

oreal

-0.69

illet

-0.67

Ranked

-0.66

inates

-0.65

sembly

-0.64

etooth

-0.64

ches

-0.62

estial

-0.62

POSITIVE LOGITS

why

1.27

WHY

1.09

how

1.02

why

0.94

how

0.85

 concise

0.83

 succinct

0.82

 plainly

0.75

ļéĨĴ

0.75

HOW

0.75

Activations Density 0.047%

information or explanations being provided

instances of the word "explain" and its variations

No Comments

No Known Activations