INDEX

Explanations

phrases related to geographical locations

oai_token-act-pair · gpt-3.5-turbo

geographical features and locations

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

 unethical

-0.72

 bullying

-0.71

 impuls

-0.70

 accuser

-0.68

 Deadpool

-0.68

 coercion

-0.68

 dismissing

-0.68

 inappropriately

-0.68

 Seth

-0.67

 torture

-0.66

POSITIVE LOGITS

Located

1.13

 limestone

1.02

 inland

0.95

 tallest

0.94

 tourist

0.93

 asphalt

0.93

 Located

0.93

 southwestern

0.92

 kilomet

0.92

 mountainous

0.87

Activations Density 1.488%

phrases related to geographical locations

geographical features and locations

No Comments

No Known Activations