INDEX

Explanations

news reports with details of events or developments

oai_token-act-pair · gpt-3.5-turbo

references to confirmation of news or reports

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

alone

-0.69

opian

-0.68

itant

-0.67

 advant

-0.63

 naÃ¯ve

-0.62

oeuv

-0.61

pet

-0.58

ilitating

-0.58

 disadvantaged

-0.58

ividually

-0.58

POSITIVE LOGITS

IDF

0.79

 Xperia

0.78

 Lumia

0.77

 Manufacturer

0.74

 Xiaomi

0.74

 Samsung

0.73

 leaked

0.73

 LAPD

0.72

 Houth

0.71

 SpaceX

0.71

Activations Density 1.031%

news reports with details of events or developments

references to confirmation of news or reports

No Comments

No Known Activations