INDEX

Explanations

names or terms related to specific individuals

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

names

-0.80

sheet

-0.76

icles

-0.73

giving

-0.71

ician

-0.71

ured

-0.70

taking

-0.70

ifully

-0.69

icians

-0.69

mother

-0.68

POSITIVE LOGITS

ÄŁ

1.00

uthor

0.96

BIP

0.90

qua

0.88

illac

0.88

$$

0.86

elta

0.81

qa

0.80

ibur

0.78

ivari

0.78

Activations Density 0.042%

names or terms related to specific individuals

No Comments

No Known Activations

names or terms related to specific individuals

No Comments

No Known Activations