INDEX

Explanations

times and years, specifically the year 2008

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

 ancest

-0.83

detail

-0.66

ymes

-0.65

ancy

-0.64

 Rooney

-0.62

lihood

-0.62

esan

-0.60

icular

-0.60

icles

-0.56

amins

-0.56

POSITIVE LOGITS

-'

0.86

å¹

0.79

 onwards

0.68

iyah

0.65

finals

0.65

 Airways

0.64

byss

0.64

 Reconstruction

0.64

edition

0.62

 Olympics

0.62

Activations Density 0.041%

times and years, specifically the year 2008

No Comments

No Known Activations