INDEX

Explanations

action words related to browsing or perusing online content

oai_token-act-pair · gpt-3.5-turbo

references to the act of browsing, particularly online

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ament

-0.72

 pronounced

-0.71

stood

-0.70

 Lauder

-0.69

adra

-0.67

oil

-0.64

onement

-0.63

cision

-0.61

orian

-0.60

 healed

-0.60

POSITIVE LOGITS

 browsing

0.92

oad

0.84

 browse

0.82

 Browse

0.81

Īè

0.79

halla

0.77

itors

0.77

âĸ¬

0.75

0.74

Ń·

0.74

Activations Density 0.039%

action words related to browsing or perusing online content

references to the act of browsing, particularly online

No Comments

No Known Activations