INDEX

Explanations

words related to policy and government actions

oai_token-act-pair · gpt-3.5-turbo

sentences containing impactful statements or assertions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 disemb

-0.86

 caster

-0.78

 nude

-0.75

 unexplained

-0.74

 reluct

-0.74

 cryptic

-0.73

 inactive

-0.72

 girlfriend

-0.72

 poked

-0.72

 booked

-0.71

POSITIVE LOGITS

 Instead

1.37

 Therefore

1.36

 Moreover

1.35

 Otherwise

1.33

 Such

1.27

 Governments

1.26

 Especially

1.25

 Hence

1.24

 Unfortunately

1.24

 Indeed

1.23

Activations Density 0.424%

words related to policy and government actions

sentences containing impactful statements or assertions

No Comments

No Known Activations