INDEX

Explanations

mentions of non-human entities or characteristics

oai_token-act-pair · gpt-3.5-turbo

references to humans and their conditions, often in a context of subhuman or inhumane treatment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 LINE

-0.72

ieri

-0.71

 Line

-0.67

REL

-0.67

 Sheet

-0.63

ENG

-0.63

 pane

-0.63

 Rolls

-0.62

 Channel

-0.61

NEY

-0.61

POSITIVE LOGITS

ity

1.57

itarian

1.32

ized

1.03

ization

1.00

ciating

1.00

izations

0.98

ities

0.98

theless

0.97

itive

0.95

zee

0.94

Activations Density 0.035%

mentions of non-human entities or characteristics

references to humans and their conditions, often in a context of subhuman or inhumane treatment

No Comments

No Known Activations