INDEX

Explanations

references to "Stone" structures or locations

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

oresc

-0.85

orescence

-0.85

merce

-0.84

ornia

-0.83

olulu

-0.75

ership

-0.74

unct

-0.71

achine

-0.71

mercial

-0.71

olitan

-0.70

POSITIVE LOGITS

hill

1.05

falls

0.97

works

0.90

castle

0.90

fish

0.87

wright

0.86

lake

0.85

Stone

0.84

bats

0.84

 stones

0.83

Activations Density 0.028%

references to "Stone" structures or locations

No Comments

No Known Activations