INDEX

Explanations

mathematical or numerical terms, specifically focusing on rates

oai_token-act-pair · gpt-3.5-turbo

references to various rates, particularly in scientific or statistical contexts

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

emis

-0.86

bley

-0.84

quickShipAvailable

-0.83

iami

-0.81

paio

-0.79

arij

-0.78

IRT

-0.77

 Aires

-0.77

ĪĴ

-0.76

agra

-0.75

POSITIVE LOGITS

payers

1.53

payer

1.29

(%)

0.86

Beer

0.82

hare

0.78

 variability

0.77

Upload

0.76

Rate

0.72

limit

0.72

finder

0.71

Activations Density 0.032%

mathematical or numerical terms, specifically focusing on rates

references to various rates, particularly in scientific or statistical contexts

No Comments

No Known Activations