INDEX

Explanations

keyword prompts directing readers to additional content or related articles

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate related articles or recommendations

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

adra

-0.65

cffffcc

-0.65

 respectively

-0.62

 administr

-0.61

 accordingly

-0.60

 CLSID

-0.59

ensibly

-0.59

 damned

-0.59

 seams

-0.59

 dece

-0.59

POSITIVE LOGITS

<|endoftext|>

1.14

1.07

:'

0.90

:-

0.89

 âĸº

0.87

*:

0.82

 âĨĴ

0.81

!:

0.80

 Discuss

0.80

Activations Density 0.087%

keyword prompts directing readers to additional content or related articles

phrases that indicate related articles or recommendations

No Comments

No Known Activations