INDEX

Explanations

terms related to rates and numbers with significance mentioned

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

emis

-0.85

quickShipAvailable

-0.80

bley

-0.80

iami

-0.80

 Aires

-0.77

paio

-0.76

 Jakarta

-0.74

IRT

-0.73

ĪĴ

-0.72

enegger

-0.71

POSITIVE LOGITS

payers

1.56

payer

1.36

Beer

0.83

(%)

0.83

 variability

0.77

uation

0.76

limit

0.76

Rate

0.75

finder

0.74

 hike

0.73

Activations Density 0.027%

terms related to rates and numbers with significance mentioned

No Comments

No Known Activations