INDEX

Explanations

verbs related to taking action or making judgments

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

 Leban

-0.70

ashtra

-0.66

outhern

-0.64

 Neutral

-0.62

azeera

-0.62

ettlement

-0.62

 Mandatory

-0.61

 Nort

-0.60

ource

-0.60

 guiActiveUn

-0.60

POSITIVE LOGITS

ifies

1.51

izes

1.43

ulates

1.27

uates

1.25

tains

1.24

itates

1.22

ends

1.21

zes

1.20

ates

1.20

iates

1.19

Activations Density 0.274%

verbs related to taking action or making judgments

No Comments

No Known Activations