INDEX

Explanations

words related to approval or endorsement

oai_token-act-pair · gpt-3.5-turbo

instances of the word "approved" and its variants in relation to decisions or validations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

tal

-0.75

fall

-0.72

birds

-0.72

Kind

-0.72

venth

-0.72

lla

-0.69

ulative

-0.68

istics

-0.67

 Bundy

-0.67

sie

-0.65

POSITIVE LOGITS

 approve

1.03

 approved

0.99

 authorizing

0.99

 unanimously

0.97

 approvals

0.96

 approval

0.94

 approves

0.93

 approving

0.89

 Appro

0.87

 disapprove

0.76

Activations Density 0.026%

words related to approval or endorsement

instances of the word "approved" and its variants in relation to decisions or validations

No Comments

No Known Activations