INDEX

Explanations

words related to fees or charges

oai_token-act-pair · gpt-3.5-turbo

references to various types of fees

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

ational

-0.80

ullivan

-0.79

 Cosponsors

-0.78

algia

-0.72

ãĤ¨ãĥ«

-0.70

ãĤ¤ãĥĪ

-0.70

dor

-0.69

 Alma

-0.68

ettings

-0.67

 Ange

-0.67

POSITIVE LOGITS

 waived

0.95

 waivers

0.90

fee

0.89

 refund

0.86

-$

0.83

 reimbursement

0.82

 levied

0.81

 waiver

0.80

 hike

0.78

 paid

0.78

Activations Density 0.028%

words related to fees or charges

references to various types of fees

No Comments

No Known Activations