INDEX

Explanations

words related to conducting online searches, particularly using the Google search engine

oai_token-act-pair · gpt-3.5-turbo

terms related to search engines and the act of searching online

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

roach

-0.78

ework

-0.77

pered

-0.72

ameron

-0.72

Pry

-0.65

rone

-0.64

rones

-0.64

ufact

-0.63

iations

-0.62

iHUD

-0.61

POSITIVE LOGITS

 queries

1.03

 query

0.94

 keywords

0.94

 searches

0.89

 lookup

0.85

 warrants

0.83

 indexes

0.81

 engines

0.80

domain

0.80

 metab

0.80

Activations Density 0.035%

words related to conducting online searches, particularly using the Google search engine

terms related to search engines and the act of searching online

No Comments

No Known Activations