INDEX

Explanations

information related to statistics, estimates, reports, and calculations from experts or authorities

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

ivas

-0.92

Interstitial

-0.90

ulkan

-0.68

 Leilan

-0.68

anti

-0.67

illard

-0.67

 Carbuncle

-0.66

ardless

-0.64

artney

-0.64

eem

-0.64

POSITIVE LOGITS

 estimating

1.04

 estimate

1.04

 forecasts

1.03

 estimates

1.01

 forecasting

0.98

 probabilities

0.97

 forecast

0.97

 predicting

0.97

 calculations

0.96

 averages

0.95

Activations Density 3.127%

information related to statistics, estimates, reports, and calculations from experts or authorities

No Comments

No Known Activations