INDEX

Explanations

phrases related to accusations

oai_token-act-pair · gpt-3.5-turbo

phrases involving accusations

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

PsyNetMessage

-0.76

edin

-0.75

 partName

-0.70

minus

-0.70

std

-0.69

GROUP

-0.68

OTAL

-0.68

thing

-0.67

english

-0.66

atl

-0.63

POSITIVE LOGITS

 continuing

0.72

 bung

0.69

 being

0.69

 extremism

0.69

 complicity

0.69

 prof

0.68

 hypocrisy

0.68

 sabot

0.67

 perpet

0.66

 maintaining

0.66

Activations Density 0.084%

phrases related to accusations

phrases involving accusations

No Comments

No Known Activations