INDEX

Explanations

phrases related to official statements or reports

oai_token-act-pair · gpt-3.5-turbo

formal statements or announcements from government or official sources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 civilisation

-1.00

 favourites

-0.95

 licence

-0.85

 favourite

-0.83

 ageing

-0.83

 neighbour

-0.81

 colours

-0.81

 organise

-0.80

 programmes

-0.79

 labou

-0.78

POSITIVE LOGITS

DOJ

1.07

HHS

1.01

 Subcommittee

0.99

 Grassley

0.93

 ACLU

0.93

HRC

0.91

 FOIA

0.90

 summarizes

0.89

 Politico

0.88

 Congressional

0.88

Activations Density 1.316%

phrases related to official statements or reports

formal statements or announcements from government or official sources

No Comments

No Known Activations