INDEX

Explanations

terminology related to mining or drilling, specifically regarding the use of explosives and creating channels or tunnels

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

NetMessage

-0.88

JO

-0.77

Attach

-0.72

warts

-0.71

hij

-0.70

nda

-0.69

VPN

-0.69

JO

-0.69

andals

-0.68

andal

-0.67

POSITIVE LOGITS

 averages

1.46

 estimate

1.39

maximum

1.35

 cutoff

1.33

 average

1.31

 ratio

1.30

 averaged

1.29

 median

1.29

 ratios

1.27

 approximate

1.26

Activations Density 0.795%

terminology related to mining or drilling, specifically regarding the use of explosives and creating channels or tunnels

No Comments

No Known Activations