INDEX

Explanations

references to internet search engines, particularly Google

oai_token-act-pair · gpt-3.5-turbo

references to search engines and the search results they provide

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ework

-0.78

roach

-0.75

ameron

-0.72

pered

-0.66

 Leilan

-0.66

iHUD

-0.65

rone

-0.65

ivari

-0.65

EStreamFrame

-0.65

cil

-0.64

POSITIVE LOGITS

 queries

0.95

 engines

0.88

 query

0.86

 keywords

0.83

 algorithm

0.83

 searches

0.82

 warrants

0.81

 warrant

0.79

 lookup

0.79

 metab

0.79

Activations Density 0.020%

references to internet search engines, particularly Google

references to search engines and the search results they provide

No Comments

No Known Activations