INDEX

Explanations

references to large numerical values or quantities

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

anwhile

-0.71

Discuss

-0.66

æĦ

-0.65

 Dating

-0.64

alys

-0.62

haps

-0.62

agon

-0.60

 compar

-0.60

odox

-0.60

emia

-0.60

POSITIVE LOGITS

 dollars

1.26

 pounds

1.04

aires

1.03

 hectares

1.02

 euros

0.99

 gallons

0.97

lbs

0.96

 acres

0.95

 annually

0.92

 dollar

0.92

Activations Density 0.905%

references to large numerical values or quantities

No Comments

No Known Activations