INDEX

Explanations

phrases related to seeking assistance, collaboration, and support

oai_token-act-pair · gpt-3.5-turbo

expressions of willingness to participate or contribute

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

usterity

-0.81

othy

-0.72

terness

-0.71

 Kardash

-0.69

ecause

-0.68

always

-0.67

inventoryQuantity

-0.67

 apologized

-0.66

 apologizing

-0.66

never

-0.66

POSITIVE LOGITS

 theirs

0.94

 sponsoring

0.90

 participate

0.89

 expertise

0.86

 their

0.83

 sponsorship

0.82

 join

0.81

 themselves

0.80

 participating

0.80

 collaborate

0.80

Activations Density 0.464%

phrases related to seeking assistance, collaboration, and support

expressions of willingness to participate or contribute

No Comments

No Known Activations