INDEX

Explanations

comparative phrases indicating a higher degree or frequency

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "more" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

pton

-0.64

 Difference

-0.62

 overrun

-0.62

pex

-0.61

 Released

-0.60

-0.59

adium

-0.58

 Ward

-0.58

IPS

-0.58

ĺħ

-0.57

POSITIVE LOGITS

 than

1.51

than

1.10

 Than

1.00

 efficiently

0.99

 importantly

0.95

 accurately

0.92

 closely

0.92

 intimately

0.86

 prominently

0.86

 broadly

0.83

Activations Density 0.090%

comparative phrases indicating a higher degree or frequency

references to the concept of "more" in various contexts

No Comments

No Known Activations