INDEX

Explanations

recommendations or suggestions within a text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "recommendations."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

plane

-0.72

jer

-0.69

atan

-0.68

ker

-0.67

Lago

-0.67

san

-0.65

ipel

-0.64

sen

-0.63

CV

-0.63

ophon

-0.61

POSITIVE LOGITS

 recommendations

1.30

 recommendation

1.21

Recommend

1.19

 recommending

1.00

uggest

0.96

recomm

0.95

 suggestions

0.89

 Recommend

0.88

 recommended

0.87

 guideline

0.85

Activations Density 0.009%

recommendations or suggestions within a text

instances of the word "recommendations."

No Comments

No Known Activations