INDEX

Explanations

phrases related to governmental decisions and actions

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_fs768-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_768

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

768

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

eem

-0.68

etting

-0.65

 Apart

-0.64

Interested

-0.63

TG

-0.61

 behold

-0.58

 Pair

-0.56

icking

-0.54

dot

-0.52

Bridge

-0.52

POSITIVE LOGITS

 been

1.72

been

1.50

 undergone

1.33

 gotten

1.31

 begun

1.25

 Been

1.21

 become

1.18

 gone

1.16

 risen

1.11

 fallen

1.06

Activations Density 2.247%

phrases related to governmental decisions and actions

No Comments

No Known Activations