INDEX

Explanations

keywords related to films and companies involved in different projects

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_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

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

ozy

-0.80

":[

-0.79

hari

-0.77

sson

-0.76

anges

-0.74

stad

-0.74

rette

-0.71

tch

-0.70

olin

-0.70

lund

-0.70

POSITIVE LOGITS

own

1.84

 predecessor

1.37

 predecessors

1.29

 namesake

1.25

 respective

1.17

 newest

1.16

ELF

1.14

 flagship

1.08

 inability

1.05

 successor

1.04

Activations Density 1.255%

keywords related to films and companies involved in different projects

No Comments

No Known Activations