INDEX

Explanations

technical or specialized terms related to programs, projects, or initiatives

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_fs12288-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

12,288

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

 divergence

-0.75

 impulse

-0.66

wom

-0.65

 ability

-0.64

 nutshell

-0.64

 Sierra

-0.63

NESS

-0.63

 sincerity

-0.63

 synopsis

-0.63

 propensity

-0.62

POSITIVE LOGITS

alysed

1.32

rolled

1.24

ivated

1.24

urated

1.21

itored

1.21

rified

1.20

ased

1.19

ocked

1.19

opped

1.19

idered

1.17

Activations Density 0.551%

technical or specialized terms related to programs, projects, or initiatives

No Comments

No Known Activations