INDEX

Explanations

words related to political actions and committees

oai_token-act-pair · gpt-3.5-turbo

terms related to political action and organizations

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

ensual

-0.76

ensive

-0.68

 seams

-0.67

sea

-0.66

ģĸ

-0.61

acterial

-0.61

umerable

-0.60

azy

-0.59

ousands

-0.59

rake

-0.59

POSITIVE LOGITS

 pund

0.76

 Franch

0.74

zai

0.72

 appoint

0.68

interstitial

0.67

 appointments

0.66

 PACs

0.65

 pamph

0.64

isans

0.64

olicy

0.64

Activations Density 0.216%

words related to political actions and committees

terms related to political action and organizations

No Comments

No Known Activations