INDEX

Explanations

locations or places

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

lihood

-0.66

 eclipse

-0.58

stakes

-0.58

FREE

-0.57

 polarized

-0.57

 toget

-0.57

ucl

-0.57

Companies

-0.56

lyak

-0.56

pert

-0.56

POSITIVE LOGITS

uitous

1.39

itude

1.19

ieth

1.15

 Lauderdale

1.11

dinand

1.10

resses

1.05

itud

0.96

raits

0.91

folios

0.91

itudes

0.90

Activations Density 0.081%

locations or places

No Comments

No Known Activations