INDEX

Explanations

names of political figures and historical events

oai_token-act-pair · gpt-3.5-turbo

references to notable events or figures in history

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

defense

-1.07

Additionally

-1.02

Defense

-1.01

colored

-1.01

caliber

-0.96

avored

-0.95

rollment

-0.94

 honored

-0.93

 utilize

-0.93

 utilizes

-0.92

POSITIVE LOGITS

 Â£

1.57

 Ukip

1.46

 organising

1.45

 Notting

1.45

 realise

1.41

 organise

1.41

 organised

1.38

 paed

1.38

 Yorkshire

1.37

 realised

1.37

Activations Density 1.383%

names of political figures and historical events

references to notable events or figures in history

No Comments

No Known Activations