INDEX

Explanations

people's names

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

cffff

-0.88

alion

-0.78

*/(

-0.77

ozy

-0.76

ypes

-0.76

querque

-0.76

technical

-0.73

ombat

-0.71

raltar

-0.69

ints

-0.69

POSITIVE LOGITS

 Marie

1.36

 Anne

1.28

Anne

1.27

 Louise

1.25

Mae

1.22

Marie

1.19

 Lynn

1.19

 Jane

1.15

 herself

1.14

 Nicole

1.12

Activations Density 0.681%

people's names

No Comments

No Known Activations