INDEX

Explanations

terms related to government, policies, and investigations

oai_token-act-pair · gpt-3.5-turbo

references to political or social issues, particularly related to authority and governance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 anecd

-0.78

 skelet

-0.74

theless

-0.69

pse

-0.68

DragonMagazine

-0.66

âĹ¼

-0.65

 UNCLASSIFIED

-0.64

 dilig

-0.62

:=

-0.62

espie

-0.62

POSITIVE LOGITS

î

0.62

".

0.61

$.

0.61

 Donald

0.60

 âĢİ

0.60

 [âĢ¦]

0.59

ADVERTISEMENT

0.57

 frontman

0.57

 Belfast

0.56

âĢİ

0.55

Activations Density 2.800%

terms related to government, policies, and investigations

references to political or social issues, particularly related to authority and governance

No Comments

No Known Activations