INDEX

Explanations

words related to fictional and technical terms, such as DNA, oxygen, and spacecraft

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

OWS

-0.77

å°Ĩ

-0.75

gag

-0.74

orously

-0.72

ailand

-0.67

ows

-0.66

eston

-0.64

iesta

-0.63

owed

-0.62

ested

-0.61

POSITIVE LOGITS

ndum

1.18

tta

0.99

nda

0.96

nder

0.96

mma

0.94

bral

0.92

tto

0.92

cht

0.91

ttes

0.91

mony

0.91

Activations Density 0.019%

words related to fictional and technical terms, such as DNA, oxygen, and spacecraft

No Comments

No Known Activations