INDEX

Explanations

words related to achieving or accomplishing goals and objectives

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

pan

-0.66

gnu

-0.65

ny

-0.64

sche

-0.64

lua

-0.64

pool

-0.63

timer

-0.63

corn

-0.63

guy

-0.62

LINE

-0.62

POSITIVE LOGITS

ments

0.86

 independence

0.82

ances

0.81

 victories

0.80

 accol

0.80

 notoriety

0.80

 immortality

0.78

 degrees

0.77

 preferential

0.77

 dominion

0.75

Activations Density 2.774%

words related to achieving or accomplishing goals and objectives

No Comments

No Known Activations