INDEX

Explanations

words related to specific names and locations

oai_token-act-pair · gpt-3.5-turbo

references to specific names and terms related to people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 willingness

-0.65

 shelf

-0.63

 shelves

-0.63

HOME

-0.62

sky

-0.61

 point

-0.60

prints

-0.60

 slick

-0.60

 rolling

-0.59

 beacon

-0.59

POSITIVE LOGITS

riv

1.36

ilege

1.09

icz

0.97

loo

0.86

ulously

0.85

abulary

0.85

atical

0.84

itives

0.83

anium

0.83

PLIC

0.83

Activations Density 0.007%

words related to specific names and locations

references to specific names and terms related to people and places

No Comments

No Known Activations