INDEX

Explanations

phrases related to clarification or direct statements

oai_token-act-pair · gpt-3.5-turbo

expressions related to accountability and transparency in communication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Created

-0.73

 Seen

-0.62

 Purch

-0.61

 Transactions

-0.61

conservancy

-0.61

soDeliveryDate

-0.59

 Located

-0.59

 Lumpur

-0.58

IDs

-0.57

effects

-0.56

POSITIVE LOGITS

 sarcast

1.07

 stating

1.05

 explaining

1.05

 mentioning

1.00

 rhet

1.00

 praising

0.98

 apologizing

0.98

 criticizing

0.97

 reiter

0.96

:"

0.96

Activations Density 0.348%

phrases related to clarification or direct statements

expressions related to accountability and transparency in communication

No Comments

No Known Activations