INDEX

Explanations

phrases related to striving or attempting, especially around exploration or perfection

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

eka

-0.87

cia

-0.70

lihood

-0.68

redo

-0.68

 appointing

-0.66

ensable

-0.66

cens

-0.65

anasia

-0.64

si

-0.64

ence

-0.62

POSITIVE LOGITS

 furiously

0.74

 assignments

0.73

 constituents

0.65

redients

0.65

 circuits

0.65

 frantically

0.65

 rats

0.64

bilt

0.63

 bugs

0.63

 plat

0.62

Activations Density 0.209%

phrases related to striving or attempting, especially around exploration or perfection

No Comments

No Known Activations