INDEX

Explanations

phrases indicating comparison, specifically emphasizing something surpassing a certain quality level

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

NCT

-0.76

ahime

-0.75

enser

-0.74

ode

-0.73

anta

-0.72

oris

-0.70

rote

-0.63

iom

-0.60

idon

-0.59

ãĥī

-0.59

POSITIVE LOGITS

 rudimentary

0.70

 mediocre

0.63

 superficial

0.63

anke

0.63

 guesses

0.62

 cosmetic

0.60

ascript

0.59

 mere

0.59

DERR

0.58

 suffice

0.57

Activations Density 0.039%

phrases indicating comparison, specifically emphasizing something surpassing a certain quality level

No Comments

No Known Activations