INDEX

Explanations

locations in North America

oai_token-act-pair · gpt-3.5-turbo

references to "North" in various contexts

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

rative

-0.66

ration

-0.65

TING

-0.63

NER

-0.62

 pinch

-0.60

tee

-0.60

xxxxxxxx

-0.59

 resc

-0.58

 confir

-0.57

 progressively

-0.57

POSITIVE LOGITS

ampton

1.42

 Carolina

1.21

umber

1.20

 Dakota

1.11

woods

1.05

umb

1.03

ridge

1.03

 Pole

1.03

western

1.01

 Korea

0.98

Activations Density 0.036%

locations in North America

references to "North" in various contexts

No Comments

No Known Activations